Вы находитесь на странице: 1из 261

Bee Products

Properties, Applications,
and Apitherapy
Bee Products
Properties, Applications,
and Apitherapy

Edited by
Avshalom Mizrahi
The Israeli College of Complementary Medicine
Tel Aviv, Israel

Yaacov Lensky
Triwaks Bee Research Center
The Hebrew University
Rehovot, Israel

Springer Science+Business Media, LLC

Library of Congress Catalog1ng-1n-PublIcatlon Data

Bee products : properties, applications, and apitherapy / edited by

Avshalom Mizrahl and Y a a c o v Lensky.
p. cm.
"Proceedings of an International Conference on Bee Products:
Properties, Applications, and Apitherapy, held May 26-30, 1996, in
Tel Aviv, Israel"—T.p. verso.
Includes bibliographical references and index.

1. B e e products—Therapeutic use—Congresses. 2. Bee products-

-Physiological effect—Congresses. I. M l z r a h i , A . II. Lensky,
Yaacov. III. International Conference on B e e Products: Properties,
Applications, and Apitherapy (1996 : Tel Aviv, Israel)
[DNLM: 1. H o n e y — c o n g r e s s e s . 2. Bee Venoms—therapeutic use-
-congresses. 3. Bee Venoms—pharmacology—congresses. 4. Propolis-
-congresses. QV 785 B414 1996]
RM666.B378B44 1996
615' .36~dc21
for Library of Congress 96-51895

Proceedings of an International Conference on Bee Products: Properties, Applications, and Apitherapy,

held May 2 6 - 3 0 , 1996, in Tel Aviv, Israel

ISBN 978-1-4757-9373-4 ISBN 978-1-4757-9371-0 (eBook)

DOI 10.1007/978-1-4757-9371-0

© Springer Science+Business Media New York 1997

Originally published by Plenum Press, New York in 1997
Softcover reprint of the hardcover 1st edition 1997

All rights reserved
109 8 7 6 5 4 3 2 1
No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any
means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, without written
permission from the Publisher

The nature .and diversity of presentations at the conference on: "Bee Products: Prop-
erties, Applications and Apitherapy" held at Tel-Aviv on May 26--30, 1996, emphasize the
increasing interest of physicians, practitioners, scientists, herbalists, dieticians, cosmeti-
cians, microbiologists, and beekeepers in different facets of bee products.
This volume consists of a selection of 31 contributions presented at the conference
and which provide information on the present status of our knowledge in this area. In spite
of their diversity, they reflect the mainstream of the conference, namely: "Imported" Prod-
ucts (honey, pollen and propolis), Exocrine Secretions of Workers (venom, royal jelly).
Toxicity and Contaminants, Quality Control, Marketing, Apitherapy, Cosmetics, etc.
Since antiquity, honey as well as other bee products were used as food, as a cure for
ailments of humans and animals, and as cosmetics.
We hope that this volume will contribute to interdisciplinary studies on chemical
composition, pharmacological effects, nutrition, and other aspects of bee products. Critical
and unbiased experimental research may unravel the yet unknown composition and mode
of action of bee products and elucidate many unanswered questions.
The noteworthy features of this conference were the participants from all parts of
the world and of different cultural backgrounds, who shared their keen interest and curios-
ity regarding honey bees and their products. We thank all of them for their personal con-
tribution to the success of this conference.

Avshalom Mizrahi
Yaacov Lensky


The Conference was organized by:

The Israeli Honey Production and Marketing Board
The Israeli Beekeepers' Associations
and in informal alliances with:
• American Apitherapy Society
• Apimondia - The International Federation of Beekeeping Association
• Asian Apicultural Association
• International Bee Research Association
• Israeli Dietetic Association
• Ministry of Agriculture, State ofIsrael
• Ministry of Tourism, State of Israel
Local Organizing Committee
Avshalom Mizrahi, Ph.D. (Chairman)
Yaacov Lensky, Ph.D. (Vice Chairman)
Moshe Almaliah, M.Sc.
Tsila Dvir, M.Sc.
Abraham Hefez, Ph.D.
Anatol Karakowsky, M.D.
Yanay Sachs
David Sadeh
Yeshayahu Stem, M.Sc.
Boris Yakobson, D.V.M.
International Advisory Committee
Stefan Bogdanov, Ph.D. (Switzerland)
Raymond Borneck, President, Apimondia (France)
Kate Chatot (U.S.A.)
Theodore Cherbuliez, M.D., President AAS (U.S.A.)
Zhibin Lin, M.D. (China)
Charles Mraz (U.S.A.)
Tetsuo Sakai, Ph.D., President, AAA (Japan)
Mira Spitzer-Adir (Croatia)
Artur Stojko, Ph.D. (Poland)
Bradford S. Weeks, M.D. (U.S.A.)
Siriwat Wangsiri, Ph.D. (Thailand)


1. The Past and Present Importance of Bee Products to Man

Eya Crane

2. Bee Products: Chemical Composition and Application . . . . . . . . . . . . . . . . . . . . . 15

Justin o. Schmidt

3. Honey as an Antimicrobial Agent. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

P. C. Molan

4. Non-Peroxide Antibacterial Activity of Honey 39

Stefan Bogdanov

5. Antioxidant Properties of Honey Produced by Bees Fed with Medical Plant

Extracts ..................................................... 49
Gennady Rosenblat, Stephane Angonnet, Alexandr Goroshit, Mina Tabak,
and Ishak Neeman

6. Speeding Up the Healing of Burns with Honey: An Experimental Study with

Histological Assessment of Wound Biopsies ........................ 57
Th. J. Postmes, M. M. C. Bosch, R. Dutrieux, J. van Baare, and
M. J. Hoekstra

7. The Effect of Honey on Human Tooth Enamel and Oral Bacteria 65

S. R. Grobler and N. 1. Basson

8. Honey Contact with Teeth in Situ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

I. Gedalia, S. R. Grobler, I. Grizim D. Steinberg, L. Shapira, I. Lewinstein,
and Mo. Sela

9. Medicinal Herbs as a Potential Source of High-Quality Honeys. . . . . . . . . . . . . . 77

Zohara Yaniv and Michal Rudich

10. The Unique Properties of Honey as Related to Its Application in Food Processing 83
Tsila Dvir

11. Honey as a Clarifying and Anti-browning Agent in Food Processing and a New
Method of Mead Production. .. .. . . .. .. . . .. . . . . . . . .. . .. . . . . . . . . . . 89
ChangY. Lee

x Contents

12. Bee-Pollen: Composition, Properties, and Applications 93

M. G. Campos, A. Cunha, and K. R. Markham

13. Clinical Evaluation ofa New HypoaUergic Formula of Pro polis in Dressings. . . 101
W. Fierro Morales and 1. Lopez Garbarino

14. Present State of Basic Studies on Propolis in Japan. . . . . . . . . . . . . . . . . . . . . . . . 107

Tsuguo Yamamoto

15. The Usage and Composition of Propolis Added Cosmetics in Korea 121
Park Jong-Sung and Woo Kun-Suk

16. Eucalyptus Propolis Beverages with Their Composition and Effects 125
Woo Kun-Suk and Park Jong-Sung

17. An Inhibitory Effect of Propolis on Germination and Cell Division in the Root
Tips of Wheat Seedlings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
K. Sorkun, S. Bozcuk, A. N. Gomiirgen, and F. Tekin

18. The Exocrine Glands of the Honey Bees: Their Structure and Secretory Products 137
Pierre Cassier and Yaacov Lensky

19. Alarm Pheromones of the Queen and Worker Honey Bees (Apis mellifera L.) 151
Yaacov Lensky and Pierre Cassier

20. Protein Traffic between Body Compartments of the Female Honey Bee (Apis
melli/era L.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Yoseph Rakover and Yaacov Lensky

21. Effects of Feeding, Age of the Larvae, and Queenlessness on the Production of
Royal Jelly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Nuray Sahinler and Osman Kaftanoglu

22. The Use of Royal Jelly during Treatment of Childhood Malignancies 179
Osman Kaftanoglu and Atilla Tanyeli

23. The Role of Hymenopterous Venoms in Nature 185

Eli Zlotkin .

24. Effect of Apamin and Melittin on Ion Channels and Intracellular Calcium of
Heart Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 203
G. Bkaily, M. Simaan, D. Jaalouk, and P. Pothier

25. Bee Venom in Treatment of Chronic Diseases. . . . . . . . . . . . . . . . . . . . . . . . . . .. 213

Th. Cherbuliez

26. Apitherapy in Orthopaedic Diseases 221

Franco Feraboli

27. The Monitoring of Possible Biological and Chemical Contaminants in Bee

Products. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 227
Boris A. Yakobson
Contents xi

28. Heavy Metals in Propolis: Practical and Simple Procedures to Reduce the Lead
Level in the Brazilian Propolis ................................... 231
Nivia Macedo Freire Alcici

29. Acaricide Residues in Beeswax and Honey. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239

S. Bogdanov, V. Ki1chenmann, and A. Imdorf

30. Judging the Quality of Honey by Sensory Analysis. . . . . . . . . . . . . . . . . . . . . . .. 247

Michel Gonnet

31. Methods for the Characterization of the Botanical and Geographical Origin of
Some Bee Products and for Their Quality Control . . . . . . . . . . . . . . . . . . .. 253
Giancarlo Ricciardelli D'Albore

Index 263
Bee Products
Properties, Applications,
and Apitherapy



Eva Crane

International Bee Research Association

Woodside House, Woodside Hill, Gerrards Cross
Bucks SL9 9TE
United Kingdom


At this Conference we are considering the products of social bees, which beekeepers
harvest from them. Candidate bees (Table I, Figure I) are: first, all the honey bees: Apis
melli/era from Europe, eastern Mediterranean lands and Africa; Apis cerana the hive bee
in Asia, and Apis dorsata, Apis jlorea and related species in the tropics of Asia. Second, in
the tropics of all continents there are stingless bees (Meliponinae), some 500 species in
all. In addition, honey-but not wax-is produced by colonies of honey wasps (Vespidae)
and honey ants (Formicinae) and is harvested from them. The wasps live in parts oftropi-
cal South America, and the ants in some dry areas of Australia and North America.
What we now think of as bee products were essential to the bees for their survival
and development during and after the evolutionary period: this was and is their function.
Stingless bees and honey bees evolved roughly 100 million and 50 million years ago, re-
spectively, whereas man has existed to use the products for only I or 2 million years-a
tiny fraction as long as social bees.
The earliest records of man's harvesting from bees' nests are in the Mesolithic rock
art of Europe and Asia, painted not more than 8000 years ago (Figures 2 and 3). There are
also rock paintings in Australia showing stingless bee nests.
Man used bee products in many ways: beeswax in various technologies, and honey
as food and also in medicine and as offerings to the gods he worshipped. Man also had
ideas about the origins of the various bee products, and attributed certain properties to
them. But their true origins were not known until a few centuries ago, and their detailed
chemical compositions were determined only in the late 1900s.
I shall say most about honey, and then deal with other products: beeswax, propolis,
rollen, bee brood, bee venom and royal jelly. Finally, I shall discuss changes in the impor-
tance of the various bee products during the period when man has been harvesting them.
2 E. Crane

Table 1. Substances collected or produced by certain social insects

Insect Where native Honey Wax Prop. Pollen Brood Venom Rj
Honey bees (Apis) Old World
A. melli/era Europe & E.
Mediterranean; Africa xx xx xx xx xx xx xx
A. cerana Asia xx xx x x x x
A. dorsata Asia, tropics xx xx x x x x x
A.jlorea Asia, tropics xx xx x x x x x
Stingless bees (Meliponinae) tropics xx xx x x x
Honey wasps S. America, tropics x x
Honey ants Parts of Australia & N. x x
x Collected or produced by the insects.
xx Known to be commercially harvested and marketed by man.


The earliest known written records' of the use of honey by man relate to religious
sacrifices in various regions; indeed honey may well have been one of the earliest non-
animal sacrifices. It was sometimes offered together with milk, or butter or ghee, oil, or
incense. According to inscriptions on clay cylinders from Sumer in Mesopotamia, when
the foundations of a new temple for the god Ningirsu were laid about 2500 Be, Gudea the
ruler of Lagash made offerings of honey and butter. Then, when the image of the god was
finally erected, he offered honey with other foods. The use of honey as an offering prob-
ably had a still older origin, because other inscriptions show that it was already customary
by Gudea's time.
In Ancient Egypt much honey was sacrificed in religious ceremonies, and when Is-
raelites later presented the first harvest of their produce to God, this included honey. For
instance in Jerusalem at the time of Hezekiah in the late 700s Be, 'they gave generously
from the first fruits of their corn and new wine, oil and honey, .. .' (II Chronicles 31.5).
Honey had, however, been forbidden as a burnt offering around 1300 Be: 'You shall not
burn any leaven or any honey as a food-offering to the Lord' (Leviticus 2.11).
What may be the earliest recorded medical prescription that includes honey is also
from Sumer, dated to about 2000 Be. Oil was to be spread over a preparation of river dust
kneaded with honey, water, and other ingredients. This was presumably for external appli-
cation, and many Ancient peoples used honey in this way2. In the Ebers papyrus compiled
in Egypt about 1550 Be, I found honey in 147 prescriptions for external use, and in 102
for internal use, both out of a total of several hundred. For internal use, honey was some-
times included because of its own properties, sometimes as a binder, and often to disguise
the taste of other, unpalatable ingredients. The Roman poet Lucretius (c. 99-55 BC) re-
ferred to this use of honey:

Physician-like, who when a bitter draught

Of wormwood is disgusted by a child
To cheat his taste, he brims the nauseous cup
With the sweet lure of honey.

• Historial records cited will be detailed in a forthcoming book on the history of man's use of bees, to be published
by Duckworth in London.
23}· ...

, a
0° 0°
..~./ -':.\:./J.
.. '-" ~......},>' 0
>") A


Figure 1. Natural distribution of spec ies of bees kept in hives' , - - - - Apis melli/era; - , - , -Apis cera lla; Shaded areas: Meliponinae,

4 E. Crane


Figure 2. Mesolithic rock painting from around 6000

Be, showing honey collection from Apis mellifera, La
Arana shelter, Bicorp, eastern Spain (copy by E. Her-
nandez-Pacheco who found the painting in 1924).
Past and Present Importance of Bee Products to Man 5

Figure 3. Post-Mesolithic rock painting showing honey collection from Api" dorsa/a, Rajat Prapat, Central India
(drawing: Y. Mathpal, 1984).

Honey was also used as a preservative, notably for the body of a king or general killed
in battle, so that this could be taken home for burial. The custom is known for instance from
Babylon, and from Ancient Greece. But according to Plutarch, when Agesilaus King of Sparta
died in 360 BC, his body was preserved in beeswax 'since they had no honey'.
The Hebrew scriptures refer many times to honey as plentiful in Canaan, and four pas-
sages indicate the source of the honey or how it was obtained. The first (Deuteronomy 32.13)
referred to Jacob, in a period about 1700 BC: the Lord ' satisfied him with honey from the
crags' . This was referred to again in Psalms 81.16, written around 1000 BC: the God of Jacob
'satisfied him with honey from the rocks'. So honey was obtained from bees nesting in rocks,
as is usual in dry country; there is no mention of nests in trees. The other two passages are in-
termediate in date. In Judges 14.8, Samson 'turned aside to look at the carcass of the lion, and
he saw a swarm of bees in it, and honey. He scraped the honey into his hands and went on,
eating as he went'. Then in I Samuel 14.25- 27: 'There was honey comb in the countryside ',
and Jonathan ' stretched out the stick that was in his hand, dipped the end of it in the honey
comb. put it to his mouth and was refreshed.' These passages describe common methods of
harvesting honey from bees' nests or from traditional hives, used all over the world.
Of course the fact that man used honey or other bee products does not show that he
was a beekeeper. By keeping bees in hives, however, man could harvest honey in larger
amounts, and more easily. Hives are known to have been used in Egypt from about 2500
BC, and Figure 4 shows a more complete later pictorial record . In Mesopotamia hives are
known from the 700s BC. But the earliest references I found relating to Israel are in the
Babylonian Talmud, compiled about AD 500.
Already in Ancient civilizations, honeys from different plants, and in different re-
gions, were differentiated. In dry Mediterranean regions much honey came from aromatic


Past and Present Importance of Bee Products to Man 7

plants such as thyme, and Roman writers praised this honey, for instance from Mount
Hymettus in Greece and Mount Hybla in Sicily. They disliked honey from Spanish broom
(Spartium junceum). Export-import trade in honey also started early. Although Ancient
Egypt produced much honey, some was imported from several places in Asia Minor, and
from Syria, Rhodes and Greece.
Until about 200 years ago it was believed that honey had its origin in the heavens, and
this idea contributed to its status as a sacred substance in pre-religious beliefs and in religions.
In the 300s BC, Aristotle said that 'honey falls from the air, principally at the rising of the
stars, and when the rainbow rests upon the earth'. Pliny (AD 23- 79) questioned 'whether it is
that this liquid is the sweet of the heavens, or whether a saliva emanating from the stars, or a
juice exuding from the air while purifying itself ... it comes to us pure, limpid and genuine' .
As late as 1609 in England the Reverend Charles Butler, who was very knowledgeable about
beekeeping, in England wrote: 'The greatest plenty of purest nectar cometh from above;
which Almighty God doth miraculously distil out of the air ... which thence doth descend into
the earth in a dew or small drizzling rain.' Vaillant in France has been quoted as the first to
state-in 1717-that the nectar the bees collect is produced by nectaries in the flowers. But
the belief that it fell from heaven persisted into the 1800s.
We now know that honey is made mainly from the nectar of flowers by bees and a
few other insects-which evaporate water from the nectar and, while doing so, secrete
into it enzymes invertase and glucose oxidase. The bees thereby produce a highly super-
saturated solution of certain sugars which has antimicrobial properties essential for its
storage in the nest safe from spoilage. Some of the tropical social bees--especially the
cavity-nesting stingless bees--evolved in an environment where maintenance of nest hy-
giene was very difficult, with no cold winters, and often a hot and humid atmosphere
which favoured the growth of pathogenic micro-organisms. And the entrance to a nest of
stingless bees must be very small (Figure 5), as a protection against the many enemies, so
the nest is poorly ventilated. Perhaps as a result, stingless bee honeys differ from Apis

Figure 5. Nest of Melipona interrupta grandis'.

8 E. Crane

mellifera honeys in several ways: for instance, they contain more acids, their enzyme
composition is somewhat different, and they have greater antimicrobial activity. Although
their water content is higher, they are stored safe from spoilage 4 , although we still do not
fully understand the reasons.


Beeswax is a very inert substance, which is not digested by mammals (including hu-
mans), and any eaten is excreted. Until the 1700s it was believed that beeswax was col-
lected by worker bees from flowers, but various observers then found wax scales on the
underside of the bee's abdomen, and in 1793 Fran"ois Huber in Switzerland finally estab-
lished that the wax is secreted by the bees themselves. The importance of beeswax to
bees-and to man-is due to its inertness and its physical properties. The wax is plastic at
temperatures from 32° upwards, so the bees can manipulate it in the hive. The wax from
anyone species of bee has a composition which is remarkably uniform, although it is very
Beeswax was used by early civilizations to cast copper-and later other met-
als-from an original wax model into superb objects of art, by the lost-wax process. The
earliest such casting known, dated to between 3500 and 3000 BC, was found in a cave in
the Judean desert. Other notable centres of early lost-wax casting were in the Yellow
River basin of China, the Red River basin in Vietnam, and Benin in West Africa. In
Mesoamerica wax from stingless bees was used to cast gold into splendid ornaments and
jewellery, but most of them were looted by the Spanish in the 1500s, and many destroyed.
Beeswax candles were used for lighting in Ancient Egypt, Crete, Greece and Rome,
and more widely in later centuries. By the AD 400s, beeswax was the mandatory material
for lights in the Christian churches. Saint Augustine said that 'the wax of the candle pro-
duced by the virgin bees from the flowers of the earth is a symbol of the Redeemer born of
a Virgin Mother'. This created a huge demand for beeswax by churches and monasteries
in Christian countries, until after the Reformation of the Roman Catholic Church in the
1500s, when reformed churches prohibited the use of candles.
Beeswax was an important component of ointments and cosmetics during Ancient
times and later. It was also used in making incendiary weapons. For instance during the
siege of Jerusalem in 1099, the Muslim defenders threw at the advancing Christians incen-
diary devices which contained pitch, beeswax, sulphur and tow. They used rather similar
mixtures in 1097 and 1147.


Propolis is the name given to various sticky substances which bees collect; some of
these are plant secretions, and others are plant exudates from wounds. It is quite hard
when cold, but becomes viscous when warmed, and Apis mellifera uses it in various build-
ing operations in the nest or hive. The Asian hive bee Apis cerana does not collect or use
propolis, but stingless bees use it extensively.
The composition of all propolis is very complex, and varies according to the plant of
origin. Anyone sample may contain a hundred different substances, including about forty
flavonoids which are the main source of its antimicrobial action. Bees may mix wax with
propolis, but I do not know that they otherwise change propolis in any way.
Past and Present Importance of Bee Products to Man 9

Propolis has been used by man since early times, for various purposes: as an adhe-
sive and to seal cracks; to protect wooden and other surfaces; and especially in medicine
because of its antimicrobial properties. We do not know what methods were used for har-
vesting it in the Ancient World, although writers in Greece and Rome were familiar with
it. The Greek Historia animalium referred to a substance mitys which was probably propo-
lis, as 'a cure for bruises and suppurating sores'. According to Varro in Rome, propolis
was used 'by physicians in making poultices, and for this reason it brings even a higher
price than honey on the Via Sacra'.
A number of early records mention substances which mayor may not have been
propolis: Asis 6 believed that 'black wax' referred to in the Egyptian Ebers papyrus (c.
1550 BC) may have been propolis. He also considered that Hebrew tzori was an early
word for propolis. This occurs six times in the Hebrew scriptures, and was usually trans-
lated as balm or balsam. In Genesis (c. 1700 BC), tzori was taken to Egypt, once with
honey; in Ezekiel it is mentioned together with honey, and its healing properties are noted
three times in Jeremiah. Twice, tzori came from Gilead, but it was not balm of Gilead
which is produced from a tree, Commiphora opobalsamum.
As far as I know, propolis was first produced commercially in the 1950s. It was har-
vested by fitting-at the periphery of a hive-a grid or grids, with holes about 2 mm; the
bees closed these up with propolis, which could be removed by shattering, after cooling
the grid in a deep-freeze. In 1984 exports included 55 tonnes from China and smaller
quantities from Argentina, Canada, Chile and Uruguay, with unknown amounts from at
least eleven other countries.


Bees evolved during the same period as the flowering plants from which they collect
nectar and pollen, and they pollinate plants by transferring pollen from one flower to an-
other, although this was not understood until Arthur Dobbs in Ireland established it in
1750. Bees also collect pollen and store it in their combs; it supplies them with protein
and essential minerals and vitamins. Hunter-gatherer peoples doubtless ate pollen along
with the honey combs they harvested from bees' nests. But if early man deliberately col-
lected pollen, it would probably have been shaken or blown from wind-pollinated flowers.
There are few past references to uses of pollen. The earliest T know to its application
in medicine are in books by Arab and Jewish physicians in Islamic Spain. Maimonides
(1135-1204), a Jew in Cordoba who was a physician to the Sultan of Egypt, recom-
mended it as an astringent and sedative tonic. In the early 1200s Ibn el-Beithar described
pollen as an aphrodisiac, also beneficial for the stomach, bowels and heart; it reduced the
'fervour' of the blood, and cured swellings produced by eating certain foods 7 •
The device beekeepers use today for harvesting pollen removes it from the bees'
legs as they enter the hive. It dates only from 1941, when the first pollen traps were pro-
duced in both Germany and the USA; a hive pollen dispenser, to coat outgoing bees with
hard-collected pollen, had been used 9 years earlier. The possibility of marketing bee-col-
lected pollen for dietary purposes was explored in the 1950s, after some of the difficulties
of commercial royal jelly production were realized. Pollen collection was best done in dry
regions of the world where handling and storing were much easier than in humid areas. By
the late 1980s, bee-collected pollen was produced commercially in at least 18 countries,
and Western Australia alone produced between 60 and 130 tonnes a year.
10 E. Crane

Harvested pollen is sold as a dietary supplement and for treating certain diseases,
and it is essential for some types of work on crop pollination and plant breeding.


Aristotle said that bees collected their young from flowers, but in 1586 Luiz Mendez
de Torres in Spain established that new bees are produced from 'seed' placed by the fe-
male queen in cells of the comb, and that she is the mother of all the other bees. Bee brood
was the main protein food that man harvested from bees' nests or hives. To many peoples
in the tropics, insects--especially immature stages-were a main source of protein; in
fact, bee brood was their food from bees' nests, and honey was a seasonal treat and a
medicine. On the other hand peoples of European origin did not eat insects; honey was
their food from bees, and in the western world brood was rarely listed among bee prod-
ucts. In 1951 Bodenheimer8 said that 'the aversion to insect food in Western civilization is
... not based on hereditary instinct. It is established by custom and prejudice.' He pointed
out that animal rearing and crop growing, which developed earlier in Mediterranean re-
gions and Europe than in the tropics, provided people with an adequate diet without the
need to hunt such tiny game as insects, so Europeans came to despise food insects, and the
peoples who ate them.
The eating of certain insects was forbidden by religions of eastern Mediterranean
Laws attributed to Moses in about 1300 Be (Leviticus 11.21) allowed insects such
as locusts to be eaten, whereas others-which would include bees-seem to have been
proscribed as unclean. According to Allegro 9 , eating bee brood was forbidden in Zadokite
fragments of the Dead Sea scrolls: 'Let no man defile his soul with any living being or
creeping thing by eating of them, from the larvae of bees [in honey] to all the living things
that creep in water'.
Muslims have explained to me that they do not eat the digestive system of an animal
(it is unclean), and since this cannot be removed from the bee, bees were not to be eaten.
Buddhism proscribed the killing or eating of any animals.
In some Asian countries bee brood is a well known and important product, but I do
not know amounts produced, exported or imported.


It was known in Ancient Greece that, when a bee stings, she cannot retract her sting
from human skin, and dies as a result. According to Broadman 1o , bee venom was referred
to as early as the 400s Be by the physician Hippocrates, and by other writers in Antiquity.
But the composition of bee venom was not established until the late 1900s.
I found more documentation from past centuries on the use of stinging bees for mili-
tary purposes than in medicine. In the Ancient World, besieged people sometimes released
bees (among other animals) into tunnels which had been excavated and occupied by the
enemy below their defended position. Tacitus described such action in about 357 Be, and
Appian's Roman history recorded how in 72 Be the Roman army under Lucullus suffered
a reverse in this way in Pontus, south of the Black Sea.
In the Middle Ages the military tactics were different: hives of bees were dropped,
thrown or projected at the enemy. There are various unsubstantiated records, but two of
Past and Present Importance of Bee Products to Man 11

the more reliable relate to incidents in 908 in England, and in 1191 at Akko (Acre) 100
km north of Tel-Aviv. An English manuscript from 1326 now in Christ Church, Oxford,
included a design for a machine to hurl skeps of bees at a besieged castle. Several reports
of the military use of skeps survive from the Thirty Years' War in Europe (1618-1648).
During the fighting in tropical Africa in the 1914-1918 World War, trip-wires were tied to
hives hidden in trees, where passing enemy troops would activate them and cause the bees
to sting.
For use in medicine, an injectable solution of bee venom is prepared, so that doses
can be quantified. As far as I know this was not attempted until the late 1880s, and the
first person to succeed was J. Langer at the University of Prague, in 1897/99. From 1930
the firm Mack at Illertissen in south Germany produced bee venom solution commer-
cially!!. Dr Bodog Beck in the USA was one of the pioneers in the use of bee venom, and
his 1935 book!2 gives much information. A method developed later, which I saw as a com-
mercial operation in Czechoslovakia in 1960, used a framework covered with a very thin
membrane, fixed in front of a hive entrance. Bare wires were stretched across the mem-
brane so that bees leaving the hive received an electric shock and stung into the mem-
brane, which was so thin that a bee could retract her sting, and she could sting again.
Drops of venom crystallized on the underside of the membrane, and were scraped off.
Since 1973, bee venom is known to have been used in medicine in at least 12 coun-
tries in Europe, 3 in Asia, and 3 in the Americas.


Young worker bees provide food for larvae (from glands in the head), and the food
is richer for larvae in queen cells (royal jelly) than for larvae in worker or drone cells.
Huber in Switzerland, in 1793, was the first to distinguish between worker brood food
(gelee) and queen brood food (gelee royale), and the prestigious name royal jelly has been
used for the queen food ever since. The first person to get a chemical analysis of royal
jelly done was probably Langstroth in the USA, in 1852, but an effective analysis was not
possible until the 1940s.
The only reference I have found to a specific traditional use of royal jelly is Ray-
mond Borneck's 1976 observation in certain areas of Ivory Coast in West Africa l3. If
honey hunters there found royal jelly in queen cells, it was given to the old people.
The harvesting of royal jelly involves much labour-intensive work, done to a strict
timetable. Colonies of bees are organized so that they rear a large number of female larvae
as queens, providing them with royal jelly. When the larvae are 3 days old the royal jelly
is removed from the cell with a vacuum pump, and the larvae are discarded.
Royal jelly was sold as a commercial product in the early 1950s in France, which
produced l.5 tonnes of it in 1958. In 1984 world production included 400 tonnes in China
and 234 tonnes in Taiwan, much of which was exported. Most royal jelly is sold for me-
dicinal and dietary purposes.


I believe that man ate the contents of bees' nests from his first existence as a spe-
cies. Early peoples used the entire contents of a nest for one or more purposes: as food,
12 E. Crane

medicine, preservative, adhesive, or military weapon; for modelling or casting metals; for
purposes of magic or as a religious symbol. Bees, honey and beeswax were held in espe-
cially high regard, and in some religions they were sacred.
Until about 1600, there was very little alteration in beliefs about bee products, or in
uses of them, but every century since then has brought one or more fundamental change.
In the 1600s honey bees (Apis mellifera) native to Europe were introduced by settlers to
North America in the New World, and flourished and spread there. During the 1700s the
true origins of most of the bee products were established for the first time.
Around 1800 the Industrial Revolution started, and many new materials were then
manufactured. Some of these competed with bee products, and could be sold much more
cheaply. For instance in England in 1400 sugar cost 20 times as much as honey. Soon after
1800 the two cost about the same, and by 1900 sugar cost only a fifth as much as honey.
The place of honey in diets of various peoples up to the 1800s has been discussed in a re-
cent paperl4. Paraffin wax was manufactured from the 1850s, and later competed success-
fully with beeswax for many purposes. In the 1800s also, European honey bees were
introduced to Australia and New Zealand, and to some parts of non-tropical Asia where
they became much more cost-effective producers than the native Apis cerana. And in
1853 Langstroth in the USA described his movable-frame hive, which has become the ba-
sis of present world beekeeping.
Throughout the 1900s the world's trade in the main bee products, honey and bees-
wax, has been dominated by production from temperate-zone Apis mellifera in movable-
frame hives, and latterly from parts of the world where this bee is not native. Also, after
low-cost alternatives to honey and beeswax became available, most purchasers of them
were in affluent societies.
By 1950, honey-the main product of movable-frame beekeeping-was difficult to
sell, and I well remember how interested and excited beekeepers became at the idea of ob-
taining income by harvesting other bee products. These new products were more expen-
sive to produce than honey or beeswax, but they commanded much higher prices.
Advanced analytical methods were developed, and these provided a more detailed knowl-
edge of their composition, and it became possible to explore new uses of them, especially
in medicine. In addition, it was established that honeys from certain plants contained spe-
cific substances with potentially useful properties.
Each bee product varies somewhat according to the species of bee from which it is
harvested. Honey, wax, brood, venom and royal jelly vary because of the physiology of
the bees themselves; honey, pollen and propolis vary because of the different plants in the
regions where the different bee species live. Almost all the papers at forthcoming Sessions
relate to products harvested from one bee, European Apis mellifera native to the temperate
zone. It is less easy to harvest the products from other bee species, but products of tropical
bees are likely to have somewhat different compositions, properties and uses, and I think
that more knowledge about them might well lead to further possibilities for diversifica-

I. Crane E. Bees and Beekeeping: Science, Practice and World Resources. Heinemann Newnes, Oxford, 1990
2. Manjo G. The Healing Hand: Man and Wound in the Ancient World. Harvard University Press, Cambridge,
MA, USA, 1975
3. Davies N. de. The Tomb of Rekhmire at Thebes. Ayer Co., Salem, NH, USA, 1944
Past and Present Importance of Bee Products to Man 13

4. Bruijn L. L. M. de (1996) Composition and Properties of Honeys of Stingless Bees (Apidae, Meliponinae).
In press
5. Camargo J. M. F. de (1970) Ninhos e Biologia de algumas Especies de Meliponideos (Hymenoptera: Api-
dae) de Regiiio de Porto Velho, Territ6rio de Rondonia, Brasil. Rev. BioI. trop. 16(2),207-239
6. Asis M. Propoleo: el Oro purpura de las Abejas. CIDA. Havana, 1989
7. Monferrer J. P. (1991) La Miel en la Espana musulmana (al-Andalus). Vida apic. (46), 64-68; (47), 24-28
8. Bodenheimer, F. S. Insects as Human Food: A Chapter in the Ecology of Man. W. Junk, The Hague, 1951
9. Allegro J. M. (1956) Personal communication
10. Broadman J. Bee Venom: The Natural Curative for Arthritis and Rheumatism. G. P. Putnam's Sons, New
II. Forster H. (1985) Personal communication
12. Beck B. F. Bee Venom Therapy. Appleton Century. New York, 1935
13. Bomeck R. (1976) L'Apiculture en Cote d'Ivoire. Rev. fro Apic. (344), 334-335, 338-339
14. Allsop K. A., Miller J. B. (1996) Honey revisited: A Reappraisal of Honey in pre-industrial Diets. Brit. J.
Nutrition 75(4),513-520

Chemical Composition and Application

Justin O. Schmidt'

Carl Hayden Bee Research Center

2000 E. Allen Road, Tucson, Arizona 85719

Honey bees are master chemists and chemical engineers. Their success in the animal
kingdom is largely because of the chemistry and the application of their products: honey,
beeswax, venom, propolis, pollen, and royal jelly. Three of these products, beeswax,
venom, and royal jelly, are chemically synthesized by the bees themselves. The other
three are derived from plants and are modified and engineered by the bees for their own
use. The use of these products explains the amazing honey bee success: honey is used as a
stable, reliable food source that serves during times of shortages, enables the bees to warm
their nest during cold weather, and has allowed them to become perennial species that can
exploit virtually all habitats in the world; beeswax is used as a pliable, stable and mois-
ture-proof material with which to construct their nest, to store honey safely, and to rear
their brood; venom gives honey bees the advantage of a formidable defense that is capable
of stopping or deterring all but the most determined and capable of predators; propolis is
an outstandingly good caulking for use in sealing the nest cavity and is also one of the
best antimicrobial agents known; pollen is a nutrient-rich food that, like honey, can be
stored in the hive indefinitely to seJ;ve as a reserve during times or seasons of shortages;
and royal jelly is a balanced food source that does not spoil readily and is used to feed bee
larvae. Without these unique products honey bees likely would have evolved to be !ittle
different from their ancestors-solitary bees in which each female bee during a brief sea-
son provisions a few cells with pollen and nectar for the next generation.
The usefulness of honey bee products for mankind is based on the same properties
that make these products useful for the bees themselves. In the case of propolis these prop-
erties extend back beyond the bees to the plants themselves which produce the original
resins that bees collected to become propolis. Quite simply, honey is an excellent, stable
sweetener and energy source for humans, just as it is for bees; beeswax is a malleable

• Address correspondence to: Dr. Justin O. Schmidt, Carl Hayden Bee Research Ctr., 2000 E. Allen Road, Tucson,
AZ 85719. Tel: 602 1l70--6380; e 109 Fax: 602 670--6493.

16 J. O. Schmidt

plastic material, that in addition be being an excellent material for molding, bums cleanly;
venom is useful because it causes pain and possesses a host of pharmacological activities;
propolis is anti-microbial toward bacteria, viruses, fungi, molds, and possesses a multitude
of other pharmacological activities; pollen is a phenomenonally nutritious and well-bal-
anced food that can be consumed by people and domestic animals; and royal jelly has a
variety of moisturizing, emulsifying and stabilizing properties that make it useful to peo-
ple. The goal of this chapter is to examine the chemistry of honey bee products, to use this
information to explain the application of the products, and to predict their usefulness.


Pollen as trapped by beekeepers from honey bee colonies is a product collected from
many, often dozens, of species of plants visited by the bees. This feature enhances the nu-
tritional balance of the pollen, but also means that bee pollen is not a uniform product,
rather it varies somewhat from sample to sample. This variability complicates the analysis
of pollen chemistry and requires that statements vis-a-vis pollen be given as averages or as
values for a specific species of pollen. All chemical and nutritional analyses here will be
given as means derived from large numbers of literature reports that appear reliable. Table
I is a listing of the chemical composition of pollen and a comparison of pollen nutrient
density with that established for Recommended Daily Allowance (RDA) or Estimated
Daily Intake (EDI = Estimated Safe and Adequate Daily Dietary Intakes) for human die-
tary needs (I). In general, compared to many standard human foods, pollen is rich in pro-
tein, low in fat, and possesses a wealth of minerals and vitamins. No obvious human
nutritional deficiencies are present in pollen with the possible exceptions of vitamin BI2
and the fat soluble vitaminsD and K. In the case ofB l2 , the vitamin is not usually in short-
age because the body usually retains a multi-year reserve. Shortage only occurs in cases of
defective body recycling (pernicious anemia) and is particularly needed for pregnant
women who have metabolic deficiencies, or are strict vegetarians. Vitamin D is somewhat
of a misnomer, as it is not truly a vitamin. Humans can synthesize the vitamin from 7-de-
hydrocholesterol if they are exposed to sunlight. Vitamin K is a minor vitamin whose sole
role is to aid in blood clotting and which is produced naturally by intestinal bacteria. Evi-
dence of the digestibility of pollen is provided by Bell et al. (2) and Schmidt et al. (3) and
a testament to its overall balance is demonstrated by mice that survived well for over a
year on a diet containing only pollen (4). Pollen has not been analyzed in detail for some
of the trace elements such as boron, chromium, molybdenum, iodine, fluoride, and sele-
nium, but it would not be surprising if it also contained adequate quantities of these ele-
One means to evaluate the nutritional content of pollen is to compare the levels of
dietary nutrients in good wholesome food to those in pollen. In Table 2 the quantities of
11 well established and measured nutrients for two vegetables, one fruit, two meats, and
two staples are compared to pollen. Pollen ranks number I in quantity for four of the nu-
trients, number 2 for another four, and ranked lower only for vitamin C, sodium, and fat.
Overall, pollen has a higher ranking than any of the compared foods, even tomatoes and
cabbage which are considered to be classic examples of the most nutritious foods avail-
able. In terms of protein, pollen ranked number 2, and above beef. The overall conclusion
is that pollen is a food source par excellence that is probably not exceeded by any other
food. The one caveat is that pollen is much too expensive to be considered a primary food
and, indeed, consumption of large quantities can cause adverse effects (4). However, this
Bee Products: Chemical Composition and Application 17

Table 1. Average chemical composition and nutritional

value of bee pollen

Chemical Composition' % of RDA/EDI b

Energy 2.46 kcal/g'
Protein 23.7 % 420
Carbohydrates 27 % 83
Lipids 4.8 % 59 d
Cholesterol -0 _Od

Phosphorus .53 % 590

Potassium .58 % 190
Calcium .225 % 250
Magnesium .148% 470
Sodium .044% 27d
Iron 140 ppm 830
Manganese 100 ppm 2500
Zinc 78 ppm 580
Copper 14 ppm 560
Nickel 4.5 ppm
Boron trace
Chromium ? ?
Molybdenum ? ?
Iodine ? ?
Fluoride ? ?
Selenium ? ?
Thiamin 9.4 ppm 760
Niacin 157 ppm 940
Riboflavin 18.6 ppm 1300
Pyridoxine 9 ppm 500
Pantothenate 28 ppm 450
Folic acid 5.2 ppm 2600
Biotin .32 ppm 440
Vit.B 11 0 0
VitaminC 350 ppm 520
Vitamin A 0 Of

Carotenes 95 ppm -900 f

Vitamin D 0 0
Vitamin E 14 ppm 160
Vitamin K 0 0
'Data for pollen from (II).
bCalculations based on RDA (Recommended Daily Allowance) and
ED! (Estimated Daily Intake = Estimated Safe and Adequate Daily
Dietary Intakes) using U.S. women aged 25-50 with a 2200 kcal in-
take (\).
'Calculated on the basis of 4 kcal/g for protein and carbohydrates.
and 9 kcal/g for fat (28).
dThese values are recommended maximum intakes for nutrients gen-
erally accepted as harmful to health in excess (I).
eNutritional requirements in the human diet not established.
fPollen contains no preformed ViI. A. but carotenes can be can-verted
to ViI. A equivalents based on 6 Ilg ~-carotene = I ViI. A equivalent
and assuming half pollen carotenes are ~-caratene.

Table 2. Nutritional comparison of pollen and typical nutritious foods a

Potassium Calcium Sodium ViI. A Thiamin Riboflavin Niacin Vil.C

Food' Protein (g) Fat (g) (g) (mg) (mg) Iron (mg) (Inl. U) (mg) (mg) (mg) (mg) Rank score'
Pollen 96.3 19.5 2.4 915 179 57 14500 3.82 7.56 63.8 142 62
Tomato 50.0 8.8 11.0 588 138 22 41000 2.75 1.88 31.2 1050 60
Cabbage 54.1 8.3 2.4 2037 835 16 5410 2.11 2.75 12.8 1950 58
Chicken 152.8 35.9 2.0 60 484 8.9 484 .28 1.29 57.7 0 31
Beans 40.1 6.5 1.7 443 3800 15 1070 .65 .25 4.9 16 30
Apple 3.4 10.3 1.9 122 19 5.3 1560 .53 .34 1.9 68 27
Bread 43.2 12.3 1.1 407 2200 12 trace 1.06 .49 11.5 trace 23
Beef 59.4 82.7 .7 26 145 7.5 143 .17 .46 12.2 0 17
'All values are based on amount in the quantity of food that provides 1000 kcal of energy; data for the foods from (29) and for pollen calculated based on the values in Table I.
'Fresh raw tomato, cabbage, apple; fried chicken leg and breast; baked beans; whole wheat bread; broiled sirloin beer steak.
'Each food is ranked from 7 (highest nutrient content) to 0 (lowest nutrient content) for each of the II nutrient categories. Because increased dietary levels of fat and sodium are considered detrimental to
health, the rankings are reversed (7=lowest; O=highest) for these two nutrients. Rank score is the total of the II rankings for each food (lowest possible score=O; highest=77)

Bee Products: Chemical Composition and Application 19

does not preclude pollen from being an excellent food supplement which can enhance the
health and well-being of individuals, especially those who otherwise might have an unbal-
anced diet.
Pollen or pollen products have been shown to have several beneficial applications
for human use. Pollen has been successfully used for treatment of some cases of benign
prostatitis (5,6,7,8,9) and for oral desensitization of children who have pollen allergy (10).
Pollen has been shown to be an excellent dietary component in diets for specialty or valu-
able animals (see (11) for more discussion).


Honey is a supersaturated solution of sugars, mainly fructose, glucose, and maltose-

like sugars, with traces of sucrose, glucose oxidase, hydrogen peroxide, phenolics, fla-
vonoids, terpenes, etc. (12). The sugars make honey hygroscopic (moisture absorbing) and
viscous, and the sugar concentration plus other factors including low pH, hydrogen perox-
ide, and the flavonoids, phenolics and terpenes make honey antimicrobial or prevent mi-
crobial growth (13).
The main use of honey is as a flavorful sweetener and energy source which is eaten
with and as a component of a wide variety of foods. The sweetness is from the sugars, par-
ticularly fructose, and flavor is created by a wide variety of trace essences derived from
plant esters, alcohols, aldehydes, and other compounds (12). Secondary, but important,
uses of honey are for the promotion of health and well being. Some of these uses include
aiding in the healing of wounds, healing of serious skin burns, and healing gastric ulcers.
The basis for the wound and burn healing properties of honey is its antimicrobial, mois-
turizinglfluid removal, and oxygen barrier properties. By keeping a wound or burn clean
and moist, and free from bacteria and the damaging effects of oxygen, the wound can heal
much more quickly than if left unaided. Modern creams and antibiotics may help heal
these types of wounds, but they often have the disadvantages of killing tissue and causing
heavy scabs and scars. The healing properties of honey were clearly demonstrated in a
study comparing honey treatment to that of silver sulfadiazine, the standard treatment, for
burn victims. The results of the study (Table 3) clearly showed that honey treatments re-
sulted in a much greater sterility of the wounds, a faster rate of healing, and a faster onset
of healing (14). Similar results have been shown by T. Postmes in tests with burned pigs.
In these experiments, honey was shown not only to be better than standard treatments, but
also better than artificial honey made from the sugars, but omitting the glucose oxidase,
hydrogen peroxide, flavonoids, and other minor components of honey (T. Postmes, per-
sonal communication).
For many years, advocates have claimed that honey can help treat gastric ulcers.
With recent discoveries, an understanding of how this can occur has emerged. Until re-
cently the bacterial origin of gastric ulcers was unknown. Now, the culprit is known to be
the bacterium Helicobacter pylori. Some honey has been shown to inhibit H. pylori (P.C.

Table 3. Honey and silver sulfadiazine (Std) treatments for burns a

% Sterile in 7 Mean days for Healing in 15

Treatment (n=52) days granulation days (%)
Honey 91 7.4 87
Silver sulfadiazine 7 13.4 10
"Data from (14)
20 J. O. Schmidt

Molan, pers. communication) and the flavonoid content and low pH of honey likely aid in
stimulating growth and healing.

Propolis is plant resin collected by bees for use in and around the hive. In plants it is
usually the sticky coating around buds that serves to protect them from the elements of
weather plus from attack by bacteria, fungi, molds and viruses. These are properties that
are useful to the bees and are enhanced by the sticky properties of the propolis. Like pol-
len, propolis is a bee product that cannot be clearly defined and varies from sample to
sample. This is a natural outcome of the collection process---propolis collecting bees will
use resins from a large variety of tree and other plant species, and these naturally will dif-
fer in their qualitative and quantitative chemical composition. Nevertheless, different pro-
polis samples do share considerable similarity in their physical and overall general
chemical nature, thereby enabling a general discussion ofthe properties of pro polis.
Much work has been conducted on the chemistry and properties of propolis. Hun-
dreds of chemical compounds have been identified from propolis. The main chemical
classes present in propolis are flavonoids, phenolics, and various aromatic compounds
(Figure 1). These compounds are poorly soluble in water, usually are soluble in alcohols,

o OH


H 0 - ) 0 > - CH=CHCOOH



Figure 1. Typical flavonoids and phenolics present in propolis.
Bee Products: Chemical Composition and Application 21

Table 4. Compounds in propolis that possess known pharmacological activities

Chemical Activities References
Quercetin Anti-viral 30
Ulcer healing
Capillary strengthening 31
Pinocembrin Anti-bacterial 32
Anti-fungal 33
Anti-mold 34
Local anesthesia 35
Caffeic acid Anti-bacterial 32
Anti-fungal 33
Anti-viral 30
Anti-inflammatory 36
Caffeic acid phenethyl ester Tumor cytotoxicity 37
Acacetin Anti-inflammatory 36
Pinostrobin Local anesthesia 35

and are often poorly soluble in hydrocarbon solvents. Propolis also contains some volatile
oils, terpenes, and beeswax, but these compounds are not believed to contribute as signifi-
cantly to the chemical properties and effects of propolis.
Flavonoids are well-known plant compounds that have antioxidant, anti-bacterial,
anti-fungal, anti-viral, and anti-inflammatory properties. Other properties of propolis in-
clude acting as a local anesthetic, reducing spasms, healing gastric ulcers, and strengthen-
ing capillaries. Compounds responsible for these activities are listed in Table 4.


Venom is synthesized by honey bees for only one purpose: as a defensive agent
against predators, primarily large mammalian and other vertebrate predators. In order to
be of defensive value the venom must induce pain, cause damage, or have some other
pharmacological or sensory activity in the potential predator (15). Bee venom, unlike
many other insect allomones, or chemical defenses. is water soluble, not fat soluble, and
must be injected or applied to moist tissues to be active. This water solubility is an advan-
tage as it allows a whole new suite of highly active defensive compounds to be used. Bee
venom is composed of a diversity of proteins, peptides, active amines, and other com-
pounds which possess a variety of activities (16). The major chemical components and
their primary activities are listed in Table 5. The main pain-inducing and lethal component
appears to be melittin (17) and this component might be responsible for much of the activ-
ity of bee venom in apitherapy use.
Mankind has used bee venom primarily for apitherapy to treat a variety of autoim-
mune diseases, with recent usage for immunotherapy of bee sting allergic patients. The
immunotherapy use will not be considered further (see (18) for further discussion). Api-
therapy has been particularly successful with individuals suffering from rheumatoid arthri-
tis, gout, and multiple sclerosis, but a variety of other immune disorders including
scleroderma and asthma have been treated (T. Cherbuliez, pers. communication). The
benefit of apitherapy for treatment of arthritis has received some research attention by the
22 J. O. Schmidt

Table 5. Components in honey bee venom and some of their activities a

Component (% venom) Chemical nature Activity/pharmacology
Melittin (3G-50) Small, highly basic 26 amino acid Pain, cardiotoxin, hemolysin,
polypeptide of2840 Mol. Wt. membrane activity
Phospholipase A2 (10-20) Basic, stable protein of 15,800 Mol. Membrane and phospholipid
Wt. disruptant, toxic, lungs are target
Apamin (3) Highly basic 18 amino acid Neurotoxin, causes tremors
Hyaluronidase (2) Protein of 35,000 Mol. WI. Promotes spreading of other
components, no other activities
Mast cell degranulating peptide (2) Highly basic 22 amino acid Releases histamine, etc. from mast
polypeptide cells, pain, anti-inflammatory
Histamine «1) Small, unstable biogenic amine of Bum-itch, redness, immediate local
III MoI.Wt. skin effects
'Data from references (16) and (38).

medical establishment. Cohen et al. (19) demonstrated in controlled experiments that bee
venom and local pain-inducing agents significantly improved the symptoms of rheumatoid
arthritis patients. Steigerwaldt et al. (20) reported moderate improvement in 66% of bee
venom treated patients versus only 27% improvement in the controls. Vick et al (21) using
severely arthritic dogs reported significant improvement in mobility and activity in their
cages of bee venom treated animals compared with controls.
Some of the problems in demonstrating efficacy of bee venom treatments for im-
mune diseases stem from the very nature of immune disorders. Immune disorders are char-
acterized by "flare ups" and remissions that occur unpredictably. In addition, immune
disorders are particularly susceptible to treatment placebo effects. These two factors com-
bine to make clinical research trials on immune diseases very difficult and often inconclu-
sive. These same problems also plague medical research concerned with evaluating
established treatments. In the cases of arthritis and multiple sclerosis, modern medicine
has no cures, it simply treats to suppress symptoms. The established treatments include
use of steroids, strong anti-inflammatory drugs, antibiotics, antimalarials, and gold
salts--drugs with serious side effects, and that often fail to deliver relief. This frustrating
situation led one researcher to comment "rheumatoid arthritis rarely kills the patient; corti-
costeroids often do" (22). These problems lead this writer to observe that apitherapy has
never killed anyone and has negligible side effects. Thus, what valid criticisms can be
raised against apitherapy for rheumatoid arthritis and multiple sclerosis?
The question arises: how does bee venom work? The answer is not clear, but we
have some hints. Bee venom has anti-inflammatory effects, it might well "shock" the im-
mune system which somehow might correct imbalances, it causes pain, and it might
stimulate the nervous system which, in turn, can exert influence on the immune system.
Bee venom possesses chemical components responsible for these activities: anti-inflam-
matory action - mast cell degranulating peptide, apamin; "shocks" immune sys-
tem-phospholipase A2 , hyaluronidase; pain-melittin; stimulates nervous
system-melittin, apamin, mast cell degranulating peptide. Overall, bee venom appears to
have the chemical properties to affect the immune system and immune disorders, and api-
therapy has been shown to work in many cases-so all that is needed is a clearer under-
standing of how apitherapy works and to convince mainstream practitioners to use
Bee Products: Chemical Composition and Application 23


Royal jelly is a creamy product secreted by young nurse worker bees for feeding to the
queen, queen larvae, and other young larvae. It is totally synthesized by the bees in the hypo-
pharyngeal and mandibular glands and is derived from the proteins and nutrients in the pollen
ingested by the secreting bees. Royal jelly consists of an emulsion of proteins, sugars, and lip-
ids in a water base (Table 6). The proteins have no particularly unusual properties (23) and
have the main presumed function of providing the growing larva or the queen a readily di-
gested source of protein. The remainder of the composition, except the lipids, also appears to
be oriented toward providing a balance of nutrients for the consuming individuals. The lipids
are unusual because they lack the normal triglycerides and diglycerides that are composed of
fatty acids having carbon chains of even numbers from 14 to 20 that are typical of insect fats.
Instead royal jelly lipids are composed mostly of short chained 8-10 carbon hydroxy fatty ac-
ids or diacids. These compounds have active functionalities at both ends of the molecule, are
more soluble in water than usual fatty acids, are highly acidic, and act as good detergents and
antimicrobial agents. It is this latter property, antimicrobial activity, that appears to be the
main function of the lipids in royal jelly.
For humans, royal jelly possesses the appealing properties of being a creamy emul-
sion that is strongly antibacterial (24,25). These make it an ideal component of cosmetics
and skin care products. Internal uses of royal jelly are less promising, as all the antibacte-
rial activities disappear when the pH is raised to above 6 by the natural buffering systems
in the body (which maintain a pH of about 7.4) (24). In fact, no clear evidence from con-
trolled experiments exists to support claims of internal usefulness of royal jelly; that in
conjunction with the lack of a theoretical chemical basis for activity leads to the conclu-
sion that there is little future promise for pharmaceutical use of royal jelly. Royal jelly is a

Table 6. Chemistry of royal jelly"

Component Quantity Comments

Water 67.0%
Proteins 12.5% 18+ major pep tides/proteins of 14-94 kd with
digestibility greater than beef
Sugars 11.0% Similar in composition to honey
Fructose 6.0%
Glucose 4.2%
Others .8%
Fats 5.0% Mostly hydroxy fatty acids, diacids, simple acids
(£)-10-Hydroxydec-2-enoic acid 31.8% of fat

10-Hydroxydecanoic acid 21.6% offat

Other hydroxy fatty acids 9.5% offat
Dicarboxylic acids 4.5% offat
Gluconic acid 24.0% offat
Others 8.6% of fat
Minerals <1.0% K, Mg, Na, Ca, Zn, Fe, Cu, Mn
Vitamins <0.1% Thiamin, riboflavin, pyridoxine, niacin, pantothenic acid,
inositol, biotin, folic acid, Vito C
Sterols <.01% 24-Methylene cholesterol, !3-stigmasterol, etc.
Undetermined 3.5%
pH 3.8
'See (11) for references and (23) for proteins.
24 J. O. Schmidt

highly nutritious material. However, its cost precludes its use for any but the most special-
ized food products for people or animals and its benefits are questionable. Recently, royal
jelly has been shown to cause serious reactions, including death, in some individuals who
ingest it (26). This indicates that both more research into the causes of the adverse reac-
tions, and caution in ingesting royal jelly are needed.


Beeswax is synthesized de novo by honey bees in four pairs of glands located on the
ventral side of the abdomen. Bees use the wax as their primary building material for mak-
ing combs for rearing their brood and for storage of honey and pollen. Beeswax is com-
posed of a variety of monoesters, diesters, hydroxylated esters, hydrocarbons, and free
fatty acids (Table 7). This composition distinguishes the material as a wax rather than a fat
because it is composed mostly of esters and long chained hydrocarbons, classic wax com-
ponents. Triglycerides and diglycerides, typical of fats, are missing. The chemistry of the
beeswax components is ideal for the uses of it by both bees and man. These components,
and the wax itself, are not soluble in water (or honey), repel water soluble materials, re-
main strong to temperatures of 50° C, are reasonably flexible, and are not readily degraded
or decomposed by moisture or microorganisms. The strength, flexibility, and waterproof-
ing qualities of beeswax have made it an excellent material for polishes, finishes, and
waxes that preserve, add shine, and generally enhance products coated with it. Beeswax
stability also makes it an excellent wax for addition to cosmetics and skin products. His-
torically, beeswax was an excellent material for making molds for castings; indeed, even
today we have artifacts over 3000 years old that were produced by the lost-wax process
(27). Beeswax also bums with a clean flame and produces a pleasant odor. This, plus the
resistance of beeswax to degradation, has made it ideal for use as candles. The stable,
flexible, and preserving properties of beeswax are good for use as waxes for musical in-
strument strings, skis, archery, and a variety of other specialty uses. Finally, the flexibil-
ity, safety, stability, and ability to accept colors of beeswax has made it a prime material
for modem crafts and hobbies for both children and adults.


Although honey bees and humans are dramatically different, they share two funda-
mental features-both are social animals, and both live in highly complex societies. These

Table 7. Gross chemical composition of beeswax a

Chemical class Quantity (%)

Monoesters 35
Diesters 14
Triesters 3
Hydroxy esters & polyesters 12
Acid esters & polyesters 3
Long-chained hydrocarbons 14
Long-chained fatty acids 12
'Data from (39)
Bee Products: Chemical Composition and Application 25

features cause both species to maintain more or less permanent residences, to have devel-
oped specialized behaviors, to engage in a diversity of activities, and to need for a multi-
tude of materials. Material properties and uses are governed by their chemistry and vice
versa. Honey bees need a stable food supply for long-term energy and growth; people
likewise need a stable food supply. Honey bees need structural materials such beeswax
and propolis to construct their nest; people likewise have housing needs. Honey bees need
materials such as propolis and venom to defend against diseases and predators; people
have similar needs. Is it any wonder then, that since antiquity, human beings have gone to
honey bees as a chemical warehouse of materials and foods. Honey and pollen are the
foods that promote health and well being in honey bees. They have served the same func-
tion for people. Bees use wax to build their combs and people have taken advantage of the
wonderful chemical properties of beeswax to make objects for their homes and daily lives
and to coat and preserve materials. Bees use propolis and venom to defend against micro-
organisms and enemies. People also use propolis, sometimes in conjunction with honey,
for its antimicrobial properties. People use the same properties in bee venom that drive off
predators of bees to enhance human health by fighting off some of their bodies' own inter-
nal enemies that cause autoimmune diseases. Overall, much of the human application of
bee products can be explained on the basis of the chemistry of the bee products. This is
not to say that bee products should not be used for purposes for which we have no chemi-
cal understanding; indeed, the process has usually operated in reverse-first, people dis-
covered uses for bee products, then later came the chemical understanding of how and
why the bee products were useful. Perhaps the message from this is that we should look to
traditional uses of bee products to guide us in our investigations and to use research to dis-
cover how best to use bee products and their components to improve human life. But for
this process to operate, individuals concerned with bee products must be fair and honest in
representing the legitimate uses and benefits of the bee products.

I. Recommended Dietary Allowances, 10th Edition. National Academy Press, Washington, DC. 1989.
2. Bell, R.R., E.J. Thornber, J.L.L. Seet, M.T. Groves, N.R. Ho and D.T. Bell. (1983) Composition and Pro-
tein Quality of Honeybee-Collected Pollen of Eucalyptus calophylla. J. Nutr. 113,2479--2484.
3. Schmidt, PJ., J.O. Schmidt and c.w. Weber. (1984) Mesquite Pollen as a Dietary Protein Source for Mice.
Nutr. Reports IntI. 30, 513-22.
4. Liebelt, R.A., D. Lyle and J.Walker. (1994) Effects of a Bee Pollen Diet on Survival and Growth of Inbred
Strains of Mice. Am. Bee J. 134,615-620.
5. Denis, L.J. (1966). Chronic Prostatitis. Acta Urol. Belg. 34,49-55.
6. Ask-Upmark, E. (1967). Prostatitis and its Treatment. Acta Med. Scand, 181,355-57.
7. Hayashi, A.U., J. Mitsui. H. Yamakawa et al. (1986). Clinical Evaluation ofCemilton in Benign Prostatic
Hypertrophy. Hinyokika Kiyo 32, 135-41.
8. Samochowiec, L., T. Dutkiewicz, J. Wojcicki and J. Gieldanowski. (1992) The Influence of Pollen Extracts
(Cemitin GBX and Cernitin T60) on Allergic Reactions. Phytother. Res. 6, 314--317.
9. Rugendorff, E.W., W. Weidner, L. Ebeling and A.C. Buck. (1993) Results of Treatment with Pollen Extract
(Cernilton N) in Chronic Prostatitis and Prostatodynia. Brit. J. Urology 7l, 433-438.
10. Wortmann, F. Oral Immunotherapy. in: Clinical Immunology and Allergology. (Steffen, C. and H. Ludwig
Editors) ElsevierlNorth-Holland, Amsterdam. 1981. pp. 389-398.
II. Schmidt, J.O. and S.L. Buchmann. Other Products of the Hive. in: The Hive and the Honey Bee (Graham,
J.M., Editor) Dadant & Sons, Hamilton, IL. 1992. pp. 927-988.
12. White, J.W.Jr. Composition of Honey. in: Honey A Comprehensive Survey (Crane, E. Editor). Heinemann,
London. 1975. pp. 157-206.
13. Molan, P.c. (1992) The Antibacterial Activity of Honey I. The Nature of the Antibacterial Activity. Bee
World 73, 5-28.
26 J. O. Schmidt

14. Subrachmanyam, M. (1991) Topical Application of Honey in Treatment of Burns. Brit. J. Surg. 78,
15. Schmidt, J.~. Hymenopteran Venoms: Striving Toward the Ultimate Defense Against Vertebrates. in: In-
sect Defenses Adaptive Mechanisms and Strategies of Prey and Predators (Evans, D.L. and J.~., Schmidt
Editors), State Univ. New York Press, Albany, NY. 1990. pp. 387-419.
16. Banks, B.E.C. and R.A. Shipolini. Chemistry and Pharmacology of Honey-bee Venom. in: Venoms of the
Hymenoptera (Piek T., Editor). Academic Press, London. 1986 pp. 330-416.
17. Schmidt, J.O. (1995) Toxinology of Venoms from the Honeybee Genus Apis. Toxicon 33, 917-927.
18. Schmidt, J.O. Allergy to Venomous Insects. in: The Hive and the Honey Bee (Graham, J.M., Editor)
Dadant & Sons, Hamilton, IL. 1992. pp. 1209-1269.
19. Cohen, A., J.B. Pearah, A.w. Dubbs and CJ. Best. (1942) Bee Venom in the Treatment of Chronic Arthri-
tis: A Comparative Study. Trans Med. Soc. State Pennsylvania 45, 957-959.
20. Steigerwaldt, F., H. Mathies and F. Damrau. (1966) Standardized Bee Venom (SBV) Therapy of Arthritis.
Indust. Med. Surg. 35, 1045--1049.
21. Vick, J.A., G.B. Warren and R.B. Brooks. (1975) The Effect of Treatment with Whole Bee Venom on Daily
Cage Activity and Plasma Cortisol Levels in the Arthritic Dog. Am. Bee J. 115,52-53,58.
22. Calin, A. Diagnosis and Management of Rheumatoid Arthritis. Addison-Wesley, Menlo Park, CA. 1983.
23. Thien, F.C.K., R. Leung, B.A. Adldo, J.A. Weiner, R. Plomley and D. Czarny. (1996) Asthma and Anaphy-
laxis Induced by Royal Jelly. Clin. Exp. Allerg. 26, 216--222.
24. Blum, M.S., A. F. Novak and S. Taber, III. (1959) IO-Hydroxy-L\2decenoic Acid, an Antibiotic Found in
Royal Jelly. Science 130,452-453.
25. Yatsunami, K. and T. Echigo. (1985) Antibacterial Action of Royal Jelly. Bull. Fac. Agr. Tamagawa Univ.
No. 25, 13-22.
26. Bullock, RJ., A. Rohan and J-A. Straatmans. (1994) Fatal Royal Jelly-Induced Asthma. Med. J. Australia
27. Crane, E. The Archaeology of Beekeeping. Duckworth, London. 1983.
28. Atwater, W.O. (1910) Principles of Nutrition and Nutritive Values of Food. U.S. Dept. Agric. Bull. 142
(Second review).
29. Adams, C.F. (1975). Nutritive Value of American foods in Common Units. USDA Agric. Handb. No. 456.
Washington DC: Government Printing Office.
30. Konig, B. and J.H. Dustmann. (1985) Fortschritte der celler Untersuchungen zur Antivirotischen Aktivitiit
von Propolis. Apidologie 16, 228--230.
31. Budavari, S. (Editor). The Merck Index. Merck & Co, Rahway, NJ 1989.
32. Vilanueva, Y.R., M. Barbier, M. Gonnet and P. Lavie. (1970) Les Flavonoides de la Propolis Isolement
d'une Nouvelle Substance Bacteriostatique: la Pinocembrine {dihydroxy-5,7-flavone} Ann. Inst. Pasteur.
Paris llB, 84--87.
33. Metzner, J., E.-M.Schneidewind and E. Friedrich. (I 977} Zur Wirkung von Propolis und Pinocembrin auf
Sprosspilze. Pharmazie 32, 730.
34. Miyakado, M., T. Kato, N. Ohno and TJ Mabry. (1976) Pinocembrin and (+)-I3-endesmol from Hymeno-
clea monog)Jra and Baccharis glutinosa. Phytochemistry 15, 846.
35. Paintz, M. and J. Metzner. (1979) Zur lokalaniisthetischen Wirkung von Propolis und einigen Inhaltsstof-
fen. Pharmazie 34, 839-841.
36. Bankova, Y.S, S.S. Popov and N.L. Marekov. (1983) A Study of Flavonoids of Propolis. J. Natural Prod.
37. Grunberger, D., R. Ganerjee, K. Eisinger, E.M. Oltz, L. Efros, M. Caldwell, V. Estevez and K. Nakanishi.
(1980) Preferential Cytotoxicity on Tumor Cells by Caffeic Acid Phenethyl Ester Isolated from Propolis.
Experientia 44, 230--232.
38. Schumacher, M.J., J.O. Schmidt, N.B. Egen and J.E. Lowry. (1990) Quantity, Analysis and Lethality of
European and Africanized Honey Bee Venoms. Amer. J. Trop. Med. Hyg. 43, 79-86.
39. Tulloch, A.P. {I 980) Beeswax--Composition and Analysis. Bee World 61, 47--62.


P. C. Molan

Honey Research Unit

Department of Biological Sciences
University ofWaikato
Hamilton, New Zealand

That honey has antibacterial properties has been known for more than a centuryl.
Although it has been used as a medicine since ancient times in many cultures 2 .3 , in its an-
cient usage there was no recognition of its antibacterial properties - it was just known to
be an effective remedy. This is not surprising considering that it is only since the latter
part of the last century that it has become known that many ailments are the result of in-
fection by microorganisms. Now it can be seen that the effectiveness of honey in many of
its medical uses is probably due to its antibacterial activity.
It is well established that honey inhibits a broad spectrum of bacterial species. There are
many reports of bactericidal as well as bacteriostatic activity. There have also been reports of
honey having antifungal activity. These numerous reports of the antimicrobial activity of
honey have been comprehensively reviewed4 ; the collation of data shows that honey is active
against a wide range of bacterial and fungal species, many of which cause infections. How-
ever, there are ailments which may be treated with honey which have not had the infectious
agents tested for their sensitivity to the antimicrobial activity of honey. Also, there has not
been much distinction made in the different types of antimicrobial activity in honey to which
the various microbial species are sensitive. For serious consideration to be given to the use of
honey as a therapeutic agent it is necessary that these aspects be further investigated.


The numerous reports of investigations which have established the nature of the an-
timicrobial factors in honey are cited in a comprehensive review of this subject 4 ,s. A brief
summary of what has been established is given here.

2.1. Explanation of Antibacterial Activity

2.1.1. Osmotic Effect. Honey is a saturated or super-saturated solution of sugars,

84% being a mixture of fructose and glucose. The water content is usually only l5~21%

28 P.c. Molan

by weight. The strong interaction of these sugar molecules with water molecules leaves
very few of the water molecules available for microorganisms. This "free" water is what is
measured as the water activity (a w ): mean values for honey have been reported from 0.562
to 0.62. Although some yeasts can live in honeys that have a high water content, causing
spoilage of the honey, the aw of ripened honey is too low to support the growth of any spe-
cies, no fermentation occurring if the water content is below 17.1 %.
Many species of bacteria have their growth completely inhibited if the a w is in the
range 0.94-0.99. These values correspond to solutions of a typical honey (a w of 0.6 undi-
luted) of concentrations from 12% down to 2% (v/v). On the other hand, some species
have their maximum rate of growth when the aw is 0.99, so inhibition by the osmotic
(water-withdrawing) effect of dilute solutions of honey obviously depends on the species
of bacteria.

2.1.2. Acidity. Honey is characteristically quite acidic, its pH being between 3.2 and
4.5, which is low enough to be inhibitory to many animal pathogens. The optimum pH for
growth of these species normally falls between 7.2 and 7.4. The minimum pH values for
growth of some common wound-infecting species is: Escherichia coli, 4.3; Salmonella
sp., 4.0; Pseudomonas aeruginosa, 4.4; Streptococcus pyogenes, 4.5. Thus in undiluted
honey the acidity is a significant antibacterial factor. But if honey is diluted, especially by
body fluids which are well buffered, the pH will not be so low and the acidity of honey
may not be an effective inhibitor of many species of bacteria.

2.1.3. Hydrogen Peroxide. The major antibacterial activity in honey has been found
to be due to hydrogen peroxide produced enzymically in the honey. The glucose oxidase
enzyme is secreted from the hypopharyngeal gland of the bee into the nectar to assist in
the formation of honey from the nectar. The hydrogen peroxide and acidity produced by
the reaction:

serve to preserve the honey. The hydrogen peroxide produced would be of effect as a ster-
ilising agent only during the ripening of honey. Full-strength honey has a negligible level
of hydrogen peroxide because this substance is short-lived in the presence of the transition
metal ions and ascorbic acid in honey which catalyse its decomposition to oxygen and
The enzyme has been found to be practically inactive in full-strength honey, it giv-
ing rise to hydrogen peroxide only when the honey is diluted. This is because the acidity
produced in the action of the enzyme drops the pH to a point which is too low for the en-
zyme to work any more. On dilution of honey the activity increases by a factor of 2 500 -
50 000, thus giving a "slow-release" antiseptic at a level which is antibacterial but not tis-

2.1.4. Phytochemical Factors. The evidence for the existence of other antibacterial
factors is mainly that the peroxide-generating system does not account for all of the ob-
served antibacterial activity, but there have also been some reports of isolation of antibac-
terial substances from honey that are not hydrogen peroxide. Furthermore, it has ben
found that heating honey, which inactivates the glucose oxidase, causes loss of activity
against some species whilst it is retained against others. Although the stability of the en-
zyme varies in different honeys, there have been reports of honeys with stability well in
Honey as an Antimicrobial Agent 29

excess of this variation, showing that there must be an additional antibacterial factor in-
volved. The most direct evidence for the existence of non-peroxide antibacterial factors in
honey is seen in the reports of activity persisting in honeys treated with catalase to remove
the hydrogen peroxide activity.
Several chemicals with antibacterial activity have been identified in honey by vari-
ous researchers: pinocembrin, terpenes, benzyl alcohol, 3,5-dimethoxy-4-hydroxybenzoic
acid (syringic acid), methyl 3,5-dimethoxy-4-hydroxybenzoate (methyl syringate), 3,4,5-
trimethoxybenzoic acid, 2-hydroxy-3-phenylpropionic acid, 2-hydroxybenzoic acid and
1,4-dihydroxybenzene. However, the quantities of these present were far too low to ac-
count for any significant amount of activity.

2.2. Variation in Antibacterial Activity

In almost all reports on the medical use of honey as an antibacterial agent no consid-
eration is given to the selection of type of honey for therapeutic use. Aristotle, c.350
B.C. 6, and Dioscorides, c.50 A.D. 7, recommended that honey collected in specific regions
and seasons (and therefore presumably from different floral sources) be used for the treat-
ment of particular ailments, but in modem medicine clinical practitioners have not heeded
these views nor the laboratory findings of large differences in the antibacterial potency of
different honeys.
It was recognised more than 40 years ago that there are differences in the antibacte-
rial activity of different honeys, and a method was devised to determine the "inhibine
number" of honeys as a measure of their antibacterial activity. The "inhibine number" is
the degree of dilution to which a honey will retain its antibacterial activity, representing
sequential dilutions of honey in steps of 5% from 25% to 5%. Studies measuring the "in-
hibine number" of honeys report activity to range over the five-fold difference in concen-
tration in the dilution series, and studies using a wider range of dilutions report the
minimum inhibitory concentrations of the honeys tested to range from 25 to 0.25%, >50 to
1.5%, 2O--D.6%, and 50-1.5%. The data showed activities to be fairly well spread over
these ranges. A study of 345 samples of New Zealand honeys8 found a large number with
low activity (36% of the samples had activity near or below the level of detection), the
rest having almost a Gaussian distribution over a twenty-fold range of activity.
The major variations seen in overall antibacterial activity are due to variation in the
level of hydrogen peroxide that arises in honey, and in some cases to the level of non-per-
oxide factors. Hydrogen peroxide can be destroyed by components of honey: it can be de-
graded by reaction with ascorbic acid and metal ions, and by the action of the enzyme
catalase which comes from the pollen and nectar of certain plants, more from the nectar.
Also, very large differences have been found between honeys from different floral sources
in the thermal stability of their glucose oxidase content, and in the sensitivity of this hy-
drogen peroxide-producing enzyme to denaturation by light because of a photosensitizing
component that comes from some floral sources.
Although it appears that the honey from certain plants has better antibacterial activ-
ity than that from others, there is not enough evidence for such definite conclusions to be
justified because the data are from small numbers of samples. However, honeys from
some sources have been been studied in large enough numbers or have been included in
enough different studies for some trends to be noted. Honeydew honey from the conifer
forests of the mountainous regions of central Europe has been found to have particularly
high antibacterial activity. Also honey from manuka (Leptospermum scoparium) in New
30 P.C. Molan

Zealand has been found to have a high activity, about half of this type of honey having an
exceptionally high level of non-peroxide activity 9.
Thus it is important that when honey is to be used as an antimicrobial agent it is se-
lected from honeys that have been assayed in the laboratory for antimicrobial activity. It is
also important that honey for use as an antimicrobial agent be stored at low temperature
and not exposed to light, so that none of the glucose oxidase activity is lost. Although all
honey will stop the growth of bacteria because of its high sugar content, when the sugars
are diluted by body fluids this antibacterial action is lost. The additional antibacterial com-
ponents then become important.


3.1. Limitations to Usage

The popular literature on health and self-treatment of ailments gives the impression
that honey can be taken to cure almost anything, but a rational consideration would sug-
gest that the antimicrobial activity would be insignificant when an oral dose of honey be-
comes diluted after absorption from the gut into the many litres of fluid in the circulation
and tissues of the body. Realistically, the potential for honey as an antimicrobial agent in
medicine is in topical application rather than as a systemic agent, although there are some
situations such as gastrointestinal infections or mastitis where the honey could remain lo-
calised and thus not become too dilute to be effectively antibacterial.

3.2. Honey as an Antiseptic Dressing

3.2.1 Established Usage of Honey as a Dressing. Honey has a well established usage
as a wound dressing in ancient and traditional medicine lo . In recent times this has been re-
discovered, and honey is in fairly widespread use as a topical antibacterial agent for the
treatment of wounds, burns and skin ulcers, there being many reports of its effective-
ness ll - 23 . The observations recorded are that inflammation, swelling and pain are quickly
reduced, unpleasant odours cease, sloughing of necrotic tissue occurs without the need for
debridement, dressings can be removed painlessly and without causing damage to re-
growing tissue, and healing occurs rapidly with minimal scarring, grafting being unneces-
sary. In many of the cases honey was used on infected lesions not responding to standard
antibiotic and antiseptic therapy. It was found in almost all of the cases to be very effec-
tive in rapidly clearing up infection and promoting healing.

3.2.2. Importance ofAntibacterial Activity. Much of the effectiveness of honey as a

dressing appears to be due to its antimicrobial properties. The healing process will not oc-
cur unless infection is cleared from a lesion: swabbing of wounds dressed with honey has
shown that the infecting bacteria are rapidly cleared 13,16,18,20.24. In this respect honey is su-
perior to the expensive modern hydrocolloid wound dressings as a moist dressing. Al-
though tissue re-growth in the healing process is enhanced by a moist environment, and
deformity is prevented if the re-growth is not forced down by a dry seab forming on the
surface, moist conditions favour the growth of infecting bacteria. Antibiotics are ineffec-
tive in this situation, and antiseptics cause tissue damage, so slow the healing process 25 .
Honey is reported to cause no tissue damage, and appears to actually promote the healing
Honey as an Antimicrobial Agent 31

There are also numerous reports of sugar being used as a wound dressing, this also be-
ing found to be effective 26-31. Antibacterial activity is attributed by several authors to the high
osmolarity of the sugar or honeyll. 17.22.27, it not being generally recognised that some honeys
can have additional antibacterial activity considerably greater than that due to the osmolarity.
This additional activity would be of particular significance in situations where the dressing
becomes diluted by body fluids, and in regions of a lesion that are not in direct contact with
the dressing. Staphylococcus aureus is exceptionally osmotolerant: for complete inhibition of
its growth the a w has to be lowered below 0.86, which would be a typical honey at 29% (v/v).
In the reports of sucrose syrup or paste being used as a wound dressing it is noted that infec-
tion with Staphylococcus aureus is hard to clear. Measurements that have been reported27 of
the dilution occurring from the uptake of water from surrounding tissues when an abdominal
wound was packed with sugar reveal that a saturated sucrose syrup containing undissolved
granules becomes diluted in 7.5 hours to a concentration that is 30% of that of a saturated so-
lution. Although the aw of this solution is low enough to prevent the growth of most human
pathogens, it is not low enough to seriously restrict the growth of Staphylococcus aureus, a
species which has developed resistance to many antibiotics and has become the predominant
agent of wound sepsis in hospitals 32 . But Staphylococcus aureus is one of the species most
sensitive to the antibacterial activity of honey. There have been many reports of complete in-
hibition of Staphylococcus aureus by honeys diluted to much lower concentrations 4 , showing
the importance of the other antibacterial factors in selected honeys.
To know for certain the clinical significance of the additional antibacterial activity
in honey, a clinical trial will need to be conducted to compare dressings of sugar and se-
lected honeys. The little comparative work reported to date indicates that more rapid heal-
ing is achieved with honey than with sugar I2 . IS • Since infection is one of the most common
impediments to wound healing 33 , then such results would be expected if the sugar dressing
were not able to fully suppress the growth of bacteria as the sugar became diluted. The ad-
ditional antibacterial activity of honey could be the reason for the remarkable rates of
healing reported when honey has been used as a dressing 1 I. 13. 14.

3.2.3. Effectiveness against Wound-Infecting Species of Bacteria. The seven species

of bacteria most commonly involved in wound infection have been tested for their sensi-
tivity to the antibacterial activity of honei 4 . The two major forms of antibacterial activity
were examined separately: a honey with an average level of activity due to hydrogen per-
oxide and no detectable non-peroxide activity was used; also a manuka honey with an av-
erage level of non-peroxide activity, with catalase added to remove any hydrogen
peroxide. The results of this study are summarised in Table 1. Overall there was little dif-
ference between the two types of antibacterial activity in their effectiveness, although

Table 1. The minimum concentration of honey (%, v/v) in the growth

medium needed to completely inhibit the growth of various specics
of wound-infecting bacteria

Species Manuka honey Other honey

Escherichia coli 3.7 7.1
Proteus mirabilis 7.3 3.3
Pseudomonas aeruginosa 10.8 6.8
Salmonella typhimurium 6.0 4.1
Serratia marcescens 6.3 4.7
Staphylococcus aureus 1.8 4.9
Streptococcus pyogenes 3.6 2.6
32 P. C. Molan

some species were more sensitive to the action of one type of honey than they were to the
other. The results thus showed that these honeys, with an average level of activity, could
be diluted nearly ten-fold yet still completely inhibit the growth of all the major wound-in-
fecting species of bacteria. It is notable that the manuka honey, with an average level of
activity, could be diluted with 54 times its volume of fluid yet still completely inhibit the
growth of Staphylococcus aureus, the major wound-infecting species, and a species noto-
rious for its development of resistance to antibiotics.
There are frequent reports of hospital wards being closed because of the presence of
strains of methicillin-resistant Staphylococcus aureus (MRSA). Because these strains are
resistant to all of the antibiotics in common use it is necessary to protect patients with im-
paired immunity from exposure to them in case they contract infections which will not re-
spond to treatment. The collection of strains of MRSA at Waikato Hospital have been
tested for sensitivity to the two honeys described above J5 . All of the strains were found to
be completely inhibited by both honeys at 10% (v/v) in the growth medium, and many of
the strains by the honeys at 5% (v/v).

3.2.4. Microbiological Safety. The use of honey as a wound dressing has been ar-
gued against because of the risk of it possibly causing wound botulism 36 . Clostridia are
widely distributed in nature, but there is a very low incidence of wound botulism. How-
ever, honey sometimes contains spores of Clostridium botulinum J7 , so there is a definite
risk of introducing the spores into wounds if honey is used as a dressing. Ifhoney could be
sterilized for use as a wound dressing this would remove the risk. The glucose oxidase ac-
tivity which generates the hydrogen peroxide is very labile and would not withstand auto-
claving 5. The non-peroxide activity of man uk a honey is stable to much more heating 38 , but
there is some loss on autoclaving J9, and any hydrogen peroxide activity present in addition
to the non-peroxide activity would be completely lost. Honey is too viscous for steriliza-
tion by filtration through microporous membranes, but sterilization by gamma-irradiation
is a possibility. However, there have been no reports on whether or not the antibacterial
factors in honey withstand this sterilizing treatment. Therefore a study was recently under-
taken to determine the effect of gamma-irradiation on the antibacterial activity of hone/D.
Honey samples were selected for their antibacterial activity, some manuka honeys with a
high level of non-peroxide activity and a low level of hydrogen peroxide activity, others
honeys with a high level of hydrogen peroxide activity only. They were put through a
commercial sterilising plant which subjected all items processed to the standard 25 kGy of
gamma-irradiation used for sterilising medical materials. The results of this study, summa-
rised in Table 2, showed that there was no significant loss of either type of antibacterial
activity when the honey samples were gamma-irradiated. A control honey very heavily
seeded with Clostridial spores had no viable spores present after the same irradiation treat-

3.3. Honey for the Treatment of Mastitis in Dairy Animals

One type of infection in which a localised high concentration of honey could be
achieved is mastitis in dairy cows and goats. This can be an expensive and difficult condi-
tion to treat. The standard treatment is the introduction of antibiotics into the teat canal of
the infected udder, but milk has to be withheld from use until clear of antibiotic residues.
Honey could possibly be suitable for the treatment of mastitis if inserted into the infected
udder via the teat canal as it is harmless to tissues and would leave no undesirable residues
in milk.
Honey as an Antimicrobial Agent 33

Table 2. Comparison of the antibacterial activity of various samples of honey before and
after sterilization by gamma-irradiation. The activity is shown as the diameter (mm), ± S.D.
(n=16), of the clear zone obtained in an agar well diffusion assay using plates seeded with
Staphylococcus aureus. Total activity (i.e. 25% w/v honey in water) and non-peroxide
activity (i.e. 25% w/v honey in catalase solution) are shown
Honey 1 Honey 2 Honey 3 Honey 4 Honey 5
Untreated, no catalase 20.8±1.1 16.0±0.8 12.7±0.5 13.6±0.5 15.4±0.6
Irradiated, no catalase 21.3±1.0 14.9±0.6 13.1±0.7 14.6±0.5 16.3±0.6
Untreated, with catalase O.O±O.O O.O±O.O 12.8±0.4 12.1±O.6 16.3±0.5
Irradiated, with catalase O.O±O.O O.O±O.O 13.2±0.4 13.5±0.5 16.5±0.6

As a first step in evaluating this possibility, the seven species of bacteria that most
commonly cause mastitis in dairy cattle were tested for their sensitivity to the antibacterial
activity of honey. Cultures of these were spread on nutrient agar plates containing various
concentrations of two types of natural honey and an artificial honey, and the growth of the
bacteria was assessed to find the concentration of honey that was necessary to prevent
growth of the bacteria. The natural honeys used were a rewarewa honey with an average
level of activity due to hydrogen peroxide and no detectable non-peroxide activity, and a
manuka honey with an average level of non-peroxide activity and no detectable non-per-
oxide activity. The artificial honey was used to assess the sensitivity of the bacteria to the
osmotic action and acidity of honey.
The results of this studl 1 are summarised in Table 3. It can be seen that the growth
of all seven species was completely inhibited by a I in 10 dilution of the natural honeys,
in some cases a I in 20 dilution being sufficient. The manuka honey was noticeably more
effective. Since only one species was inhibited by the artificial honey at a I in 10 dilution,
it can be seen that the other antibacterial factors in natural honeys are important, and hon-
eys should be selected for a high level of these if they are to be subjected to trial in the
treatment of clinical mastitis.

3.4. Honey for the Treatment of Peptic Ulcers

Honey is a traditional remedy for dyspepsia and peptic ulcers42, but there has been
no rational basis for its use. The finding that Helicohacter pylori is probably the causative
agent in many cases of dyspepsia and peptic ulcers raised the possibility that the antibacte-
rial properties may be responsible for its therapeutic action. Consequently, the sensitivity
of Helicobacter pylori to honey was tested42 , using isolates of Helicobacter pylori from bi-

Table 3. Minimum inhibitory concentration of honeys (% v/v in nutrient agar)

for cultures of various mastitis-causing bacteria streaked on the agar plates

Bacterial species Manuka honey Rewarewa honey Artificial honey

Actinomyces pyogenes 1-5% 1-5% 5--10%
Klebsiella pneumoniae 5--10% 5--10% >10%
Nocardia asteroides 1-5% 5--10% >10%
Staphylococcus aureus 1-5% 1-5% >10%
Streptococcus agalactiae 1-5% 5--10% >10%
Streptococcus dysgalactiae 1-5% 5--10% >10%
Streptococcus uberis 1-5% 5--10% >10%
34 P. C. Molan

opsies of gastric ulcers. All five isolates tested were found to be sensitive in an agar well
diffusion assay to a 20% (v/v) solution of a manuka honey with an average level of non-
peroxide activity, but none showed sensitivity to a 50% (v/v) solution of a honey in which
the antibacterial activity was due primarily to its content of hydrogen peroxide. Assess-
ment of the minimum inhibitory concentration by inclusion of manuka honey in the agar
showed that the growth of all of a further seven isolates tested was completely inhibited
over the incubation period of 72 h by the presence of 5% (v/v) honey.

3.5. Honey for the Treatment of Gastroenteritis

Honey has been found to be effective in treating bacterial gastroenteritis in infants 43 .
Used in place of glucose in an oral re-hydration fluid, it was found to be as effective as
glucose in achieving re-hydration, whilst the antibacterial activity cleared the infection in
bacterial diarrhoea. However, there is little information available on the sensitivity of the
gastroenteritis-causing species of bacteria to the antibacterial activity of honey, and on
which of the antibacterial factors in honey is most effective against them. Therefore honey
was tested for its relative antibacterial potency against all the bacterial species that com-
monly cause gastroenteritis, comparing manuka honey and a honey with the usual hydro-
gen peroxide activity, also an artificial honey to assess how much of the antibacterial
activity was due simply to the acidity and the osmotic effect of the sugar in hone/ 4 , With
some of the species of bacteria the assessment was repeated with additional strains ob-
tained from clinical isolates supplied by medical and animal health laboratories to see if
there was any variation in sensitivity between different strains of a species.
Cultures of the bacteria were streaked on nutrient agar plates containing various
concentrations of the honeys, and the growth of the bacteria was assessed to find the con-
centration of honey that was necessary to prevent growth of the bacteria. The honeys used
were a mixed pasture honey with an average level of activity due to hydrogen peroxide
and no detectable non-peroxide activity, and a man uk a honey with an average level of
non-peroxide activity. Honey concentrations were in a 5% (v/v) step dilution series in-
itially and then with I % dilution steps, the honey being diluted with either sterile distilled
water (for the pasture honey and artificial honey) or a sterile solution of 0.2% catalase (for
the manuka honey). Plates where inhibition of growth was observed were swabbed with a
loopful of sterile water and streaked onto freshly prepared nutrient agar plates which did
not contain honey. The plates were then incubated to find any surviving bacteria growing
into visible colonies if the initial inhibition had been due to prevention of growth (bate-
riostasis) rather than killing the bacteria (bactericidal activity).
The results, summarised in Table 4, showed that honey with an average level of hy-
drogen peroxide activity is bacteriostatic at 4-8% (v/v) and bactericidal at 5-10% (v/v).
The non-peroxide activity of an average manuka honey is bacteriostatic at 5-11 % (v/v)
and bactericidal at 8-15% (v/v). Activity (just bacteriostatic) was not seen with artificial
honey unless it was at 20-30% (v/v), clearly showing the importance of factors other than
sugar and acidity,

3.5. Honey for the Treatment of Tineas

Honey has been reported to have antifungal activity, but not many species of fungi
have been tested. An important group of fungi which regularly infect humans are the der-
matophytes (Deuteromycotina). Cutaneous or superficial mycoses, caused through host in-
fection by these fungi, are one of the most common diseases of humans. Only a small
Honey as an Antimicrobial Agent 35

Table 4. Minimum inhibitory concentration of honeys in nutrient agar plates (% v/v) giving partial
inhibition (PI), bacteriostatic activity (BS) and bactericidal activity(BC) against various strains of
bacteria which cause gastroenteritis

Manuka honey with catalase Pasture honey

Bacterial strain PI BS BC PI BS BC
Escherichia coli 916 6% 7% 10% 5% 6% 6%
Escherichia coli ex AHL 6% 7% 10% 6% 6%
Escherichia coli K88+ 6% 7% 10% 6% 6%
Salmonella enteritis 3484 7% 8% 10% 4% 5% 6%
Salmonella hadar 326 6% 7% 8% 6% 6%
Salmonella infantis 93 7% 8% 10% 6% 7% 10%
Salmonella typhimurium 298 6% 7% 8% 6% 8%
Salmonella typhimurium 1739 6% 7% 9% 6% 7%
Salmonella typhimurium ex WH 5% 10% 5% 10%
Shigella boydii 2616 6% 7% 10% 5% 6%
Shigella jlexneri 983 6% 7% 10% 6% 6%
Shigella sonnei 86 6% 7% 10% 5% 5%
Shigella sonnei ex WH 5% 6% 10% 6% 10%
Vibrio cholorae 5% 7% 10% 6% 7% 10%
Vibrio paraheamolyticus 5% 6% 10% 4% 6%
Yersinia enterocolitica 10% 11% 15% 7% 8% 9%

number of species of these, from the genera Epidermophyton , Microsporum and Tricho-
phyton, regularly infect humans 45 • Superficial fungal infections are amongst the most diffi-
cult diseases to successfully treat, antibiotics which successfully combat bacterial diseases
being largely ineffective against fungi. A common predisposition to some fungal infec-
tions is poor host immunity, thus bacterial infections may also be present quite often. So a
treatment which has both antifungal and antibacterial activities would be most beneficial.
Therefore the effectiveness of honey against the dermatophyte species which most fre-
quently cause superficial mycoses (tineas such as ringworm and athletes foot) was invest i-
gated46 •
For this investigation two sorts of natural honey were used: a mixed pasture honey
with an average level of antibacterial activity due to hydrogen peroxide production, and a
manuka honey with an average level of non-peroxide antibacterial activity. An artificial
honey was also used, to assess how much of the antibacterial activity was due simply to
the acidity and the osmotic effect. The honeys were tested against clinical isolates of
seven species of dermatophytes. An agar well diffusion assay was used, the contents of the
wells being replaced with freshly prepared honey solutions at 24 hour intervals over the 3
- 4 days of incubation. The honeys were diluted with either sterile distilled water or a ster-
ile solution of 0.2% catalase, a 5% (v/v) step dilution series being used for testing.
The results are summarised in Table 5. No inhibitory activity was detected with any
of the seven species with the pasture honey at any concentration up to the highest tested,
50% (v/v), when catalase was present, nor with the artificial honey even at 100%. This
showed that it was the the hydrogen peroxide in the pasture honey, and the non-peroxide
activity in the manuka honey, that were inhibiting the growth of the fungi. Although the
concentrations of honey needed to inhibit some of the dermatophytes are higher than
needed to inhibit bacteria, less dilution of the honey is likely with a tinea than with in-
fected wounds, bums and ulcers where there would be serum exudation. It could be that
36 P. C. Molan

Table 5. Minimum inhibitory concentration of honeys in agar wells (% v/v) giving a clear zone
around the wells in an agar well diffusion assay against seven species of fungi which cause tineas
Manuka honey with
Species Pasture honey Manuka honey catalase
Epidermophyton jloccosum 5-10% 5-10% 20-25%
Microsporum canis 10-15% 20-25% 20-25%
Microsporum gypseum 15-20% 45-50% 50-55%
Trichophyton rubrum 2.5-5% 5-10 15-20%
Trichophyton tonsurans 15-20% 20-25% 20-25%
T. mentagrophytes var. interdigitate 10-15% 20-25% 40-45%
T. mentagrophytes var. mentagrophytes 10-15% 15-20% 20-25%

manuka honey may be more effective, even though the dermatophytes are less sensitive to
its activity than they are to hydrogen peroxide, if there is insufficient dilution of honey on
tineas for the enzymic production of hydrogen peroxide to be activated. Which type of
honey is most effective, and the practical usefulness of honey as a topical antifungal salve,
will only be known if comparative clinical trials are conducted.

I. Dustmann J H. (I 979) Antibacterial Effect of Honey. Apiacta 14, 7-11.
2. Majno G: The Healing Hand. Man and Wound in the Ancient World. Harvard University Press Cambridge,
Massachusetts. 1975.
3. Ransome H M: The Sacred Bee in Ancient Times and Folklore. George Allen and Unwin London. 1937.
4. Molan PC. (1992) The Antibacterial Activity of Honey. I. The Nature of the Antibacterial Activity. Bee
World 73, 5-28.
5. Molan PC. (\992) The Antibacterial Activity of Honey. 2. Variation in the Potency of the Antibacterial Ac-
tivity. Bee World 73,59-76.
6. Aristotle (350 B.C.). Translated by Thompson D'A W. Historia Animalium in: The Works of Aristotle
(Smith J A, Ross W D editors) Oxford University Press Oxford 1910 Volume IV.
7. Gunther R T: The Greek Herbal of Dioscorides (Translated by Goodyear J, 1655). Hafner N. Y. 1934, re-
printed 1959.
8. Allen K L, Molan PC, Reid G M. (\ 991) A Survey of the Antibacterial Activity of Some New Zealand
Honeys. J. Pharm. Pharmacol. 43, 817-822.
9. Allen K L, Molan P C, Reid G M. (1991) The Variability of the Aantibacterial Activity of Honey. Apiacta
26, 114--121.
10. Zumla A, Lulat A. (\ 989) Honey - a Remedy Rediscovered. J. Royal Soc. Med. 82, 384--385.
II. Bulman M W. (\ 955) Honey as a Surgical Dressing. Middlesex Hosp. J. 55, 188-189.
12. Hutton D J. (1966) Treatment of Pressure Sores. Nurs. Times 62,1533-1534.
13. Cavanagh D, Beazley J, Ostapowicz F. (1970) Radical Operation for Carcinoma of the Vulva. A New Ap-
proach to Wound Healing. J. Obstet. Gynaecol. Br. Cmwlth. 77. 1037-1040.
14. BIomfield R. (1973) Honey for Decubitus Ulcers. J. Am. Med. Assoc. 224, 905.
15. Burlando F. (1978) Sull'azione Terapeutica del Miele nelle Ustioni. Minerva Dermat.1l3, 699-706.
16. Armon P J. (1980) The Use of Honey in the Treatment ofInfected Wounds. Trop. Doct. 10,91.
17. Bose B. (1982) Honey or Sugar in Treatment ofInfected Wounds? Lancet i, 963.
18. Dumronglert E. (1983) A Follow-up Study of Chronic Wound Healing Dressing with Pure Natural Honey.
J. Natl Res. Counc. Thail. 15,39-66.
19. Kandil A. Elbanby M, Abd-Elwahed K, Abou Sehly G, Ezzat N. (1987) Healing Effect of True Floral and
False Nonfloral Honey on Medical Wounds. J. Drug Res. (Cairo) 17, 71-75.
20. Effem SEE. (1988) Clinical Observations on the Wound Healing Properties of Honey. Br. J. Surg. 75,
21. Farouk A, Hassan T, KashifH, Khalid S A, Mutawali I, Wadi M. (1988) Studies on Sudanese Bee Honey:
Laboratory and Clinical Evaluation. Int. J. Crude Drug Res. 26, 161-168.
Honey as an Antimicrobial Agent 37

22. Green A E.(l988) Wound Healing Properties of Honey. Br. 1. Surg. 75, 1278.
23. Mcinerney R J F. (1990) Honey - a Remedy Rediscovered. 1. Royal Soc. Med. 83, 127.
24. Braniki F J. (1981) Surgery in Western Kenya. Ann. Royal CoIl. Surg. Engl. 63, 348-352.
25. Branemark P-I, Ekholm R, Albrektsson B, Lindstrom J, Lundborg G, Lundskog J. (1967) Tissue Injury
Caused by Wound Disinfectants. 1. Bone Joint Surg. Am. Vol. 49, 48-62.
26. Knutson R A, Merbit L A, Creekmore M A, Snipes H G. (1981) Use of Sugar and Povidone-iodine to Een-
hance Wound Healing: Five Years Experience. South. Med. J. 74. 1329-1335.
27. Chi rife J, Herszage L, Joseph A, Koh E S. (1983) In Vitro study of Bacterial Growth Inhibition in Concen-
trated Sugar Solutions: Microbiological Basis for the Use of Sugar in Treating Infected Wounds. Antimi-
crab. Agents Chemother. 23, 766--773.
28. Rahal F, Mimica I M, Pereira V, Athie E. (1984) Sugar in the Treatment aflnfected Surgical Wounds. Inter-
nal. Surg. 69, 308.
29. Middleton K, Seal 0 V. (1985) Sugar as an Aid to Wound Healing. Pharm. J. 235, 757-758.
30. Trouillet J L, Fagon J Y, Domart Y, Chastre J, Pierre J, Gibert C. (1985) Use of Granulated Sugar in Treat-
ment of Open Mediastinitis after Cardiac Surgery. Lancet ii, 180··184.
31. Shimamoto Y, Shimamoto H, Fujihata H, Nakamura H, Matsuura Y. (1986) Topical Application of Sugar
and Povidone-iodine in the Management of Decubitus Ulcers in Aged Patients. Hiroshima J. Med. Sci. 35,
32. Lowbury E J L, Ayliffe G A J: Drug Resistance in Antimicrobial Therapy. Thomas Springfield, Illinois.
33. Smith M, Enquist I F. (1967) A Quantitative Study of Impaired Healing Resulting from Infection. Surg.
Gynecol. Obstet. 125,965--973.
34. Willix 0 J, Molan P C, Harfoot C 1. (J 992) A Comparison of the Sensitivity of Wound-infecting Species of
Bacteria to the Antibacterial Activity of Manuka Honey and Other Honey. 1. Appl. Bacteriol. 73,388-394.
35. Hancock B M: Microbiology Dept., Waikato Hospital, Hamilton, New Zealand: unpublished findings.
36. Mossel 0 A A. (1980) Honey for Necrotic Breast Ulcers. Lancet ii, 1091.
37. Huhtanen C N, Knox 0, Shimanuki H. (1981) Incidence of Clostridium botulinum Spores in Honey. J.
Food Prot. 44, 812-814.
38. Molan PC, Russell K M. (1988) Non-peroxide Antibacterial Activity in Some New Zealand Honeys. 1.
Apic. Res. 27, 62-67.
39. Allen K L: University ofWaikato, Hamilton, New Zealand: unpublished findings.
40. Molan PC, Allen K L. (1996) The Effect of Gamma-irradiation on the Antibacterial Activity of Honey. J.
Pharm. Pharmacol. In press.
41. Allen K L, Molan pc. The Sensitivity of Mastitis-causing Bacteria to the Antibacterial Activity of Honey.
Submitted for publication.
42. Al Somai N, Coley K E, Molan PC, Hancock BM. (1994) Susceptibility of Helicobacter pylori to the An-
tibacterial Activity of Manuka Honey. J. Royal Soc. Med. 87, 9-12.
43. Haffejee I E, Moosa A. (1985) Honey in the Treatment of Infantile Gastroenteritis. Br. Med. 1. 290,
44. Brady N F, Molan P C. The Sensitivity of Enteropathogenic Bacteria to the Antibacterial Activity of
Honey. Paper in preparation.
45. Rademaker M. (1993). Superficial Dermatophyte Infections. N. Z. Med. J. 106, 14--16.
46. Brady N F, Molan P C, Harfoot C G. The sensitivity of dermatophytes to the antimicrobial activity of
honey. Submitted for publication.



Stefan Bogdanov

Federal Dairy Research Institute

Bee Department
3097 Liebefeld, Switzerland


Honey acts as an antibacterial agent against many bacteria (1). There are two sorts
of antibacterial agents or so called "inhibines." One of them is heat- and light-sensitive
and has its origin in the HP2 ' produced by honey glucose oxidase (2,3,4). Some workers
believe that hydrogen peroxide is the main antibacterial agent (2,5,6). Other authors find
that the non-peroxide activity is the more important one (7,8,9). The H 20 2 amount in
honey is very small and it can be produced only after aerobic incubation of diluted honey
solutions, which might mean that it is not very important for the antibacterial action of
honey (10). The argumentations of the pro and contra peroxide side are based on the re-
sults with the specific antibacterial test used. However, a certain antibacterial test might
be sensitive only to certain types of antibacterian substances. In a previous study from dur
laboratory it was found that while in an agar disc diffusion test only the peroxide activity
was measured, in a liquid medium test only the non-peroxide substances were active( 10).

1.1. Antibacterial Substances

The main honey substances are sugars, which by their osmotic effect exert an an-
tibacterial action (1). The antimicrobial tests used in different studies are mostly carried
out at concentrations, where the sugars are not osmotically active.
Many different antibacterial substances, which are more or less heat- and light-sta-
ble have been characterised. It has bee claimed, that honey contains lysozyme, a well
known antibacterial agent (8). However, in an other study it was found that no lysozyme
activity was present in honey (10). Honey contains a number of flavonoids (11,12), many
of which are known to have an antibacterial action (13). The profile and the concentration
of the different flavonoids have been correlated to the floral origin of honey (11,12) , but
not to its antibacterial activity (14). From New Zealand honeys, mainly manuka and vi-
per's bugloss honey, several aromatic acidic substances with antibacterial activity have

40 S.Bogdanov

been isolated (1). These substances were proved to have a floral origin. Only in viper's
bugloss honey the antibacterial activity measured could be explained by the amount of the
antibacterial substance present. Another investigation claimed, that the low honey pH was
responsible for the antibacterial activity (15). Some workers have isolated volatile sub-
stances with antibacterial activity (17,18), but their contribution to the antibacterial action
was not examined. Other antibacterial substances have also been chemically charac-
terized, mainly in New Zealand honey (1). Also, other workers found non-peroxide activ-
ity of honey, extractable by organic solvents, but were not able to identify the chemical
nature of the substances (18,19).

1.2. Origin of the Antibacterial Substances

Several clearly identified substances were shown to have a floral origin (1). In one
study it was found, that the non-peroxide activity in blossom and honeydew honeys was
not significantly different (20). Another study claimed, that the non-peroxide, partly vola-
tile antibacterial activity of honey, has bee origin (21).
Summarising the results of the previous studies, it seems that only in very rare cases
the antimicrobial activity was quantitatively correlated to the amounts of the antimicrobial
substances present. Also, in most studies it was not clear, whether the activity had a bee or
a plant origin. This is surely due to the heterogenous nature of non-peroxide activity.
In our study we looked for answers to following questions:
I. Which different honey substance groups are responsible for the antimicrobial
2. Does the activity have a plant or a bee origin?
3. What is the effect of heat and storage of honey on the non-peroxide activity?
Here we summarise the work done in our laboratory, dealing with the non-peroxide
honey activity, using an antibacterial test, which was shown to reflect only the non-perox-
ide part of antimicrobial activity (10)


2.1. Materials and Honey Samples

Turbidity Test. For the turbidity test we used the following liquid medium: I % pep-
ton, 1% Lab-Lemco (both Oxoid) and 0.1 % glucose
Test strains: Staphylococcus aureus and Sarcina lutea were both used for inoccula-
tion of bacteria growth tests as suspensions with 0.2 absorption units at 520 nm
The spectrometer for measuring the turbidity of the bacterial suspensions was a
Spectronic 20, where 20 mi test tubes could be read directly at 520 nm
Thermostatable shaking incubator
20 mi sterile test tubes
A "honey-sugar" standard was a solution of 40% fructose, 35% glucose, 7% mal-
tose, 0,2 gil 00 g KCI and mM lactic acid.

Honey Fractionation. The columns used for honey fractionation were:

• C-18 1000 mg SPE (Solid Phase Extraction) Baker 7020 disposable columns
Non-peroxide Antibacterial Activity of Honey 41

• 2 cm 3 /column 50 mesh (Dowex 50 W x 8) strong acidic cation exchanger

• 2 cm 3 /column 50 mesh (Dowex I x 8) strong basic anion exchanger
The SPE columns were mounted on Baker-IO SPE extraction manifold with vac-
uum. Biorad polypropylene disposable columns Nr.73l-l550 were used for the ion ex-
change fractionation, without the use of vacuum.

Reagents and Honey Samples. All reagents were of analytical purity grade. Destilled
water was used for all dilution steps.
The honey samples anlysed in this study were either market samples (of Swiss and
foreign origin) or they were taken at different parts of Switzerland for the purpose of the
present study.

2.2. Methods

2.2.1. Honey Analysis. The honey humidity, acidity, invertase- and diastase (Phade-
bas method) activities were all determined according to the Swiss Food Manual (22). The
free acidity is expressed in maeq/kg, the invertase in Hadorn units (invertase number) and
the diastase in Schade units (diastase number). HP2 production was determined as de-
scribed (10).

2.2.2. Antibacterial Growth Test.

• Mix 10 mlliquid growth medium with 10 ml20g/100g honey solution.
Add one drop of bacteria suspension and mix well.
• Incubate in the shaking water bath at a constant shake-speed for a maximal bacte-
rial grwoth: 12 hours for Staphylococcus and 36 hours for Sarcina. Whole honeys
were tsted only against Staph.aureus was tested, while honey fractions were
tested against both strains
• Read turbidity at 520 nm
Control incubation: 10 ml liquid growth medium were mixed with 10 ml 20gll00g
honey sugar standard. It was shown earlier that due to osmosis, this sugar concentration
has an inhibitory effect of about 10--20 %, compared to the growth in the medium without
the "honey sugar" standard.

2.2.3. Honey Fractions. 50 gllOO ml honey water solutions were used for all frac-
tionation steps. The initial pH of each honey solution was measured. The antimicrobial tests
were carried out with the initial honey solutions and with the honey solutions after the re-
moval of the fraction by each fractionation step. Before the antibacterial test the honey con-
centration of the solution after each fractionation step was adjusted to 20 gil 00 ml and the
pH of the solution was set at the pH ofthe honey solution before the fractionation. Removal of Volatile Substances. 50 gllOO ml honey solution was heated at

60° C under vacuum (15 mg Hg) for 2 hours in order to remove all volatile substances. Removal of Non-Polar Non-Volatile Substances. The Baker C-IS columns

were activated with one volume of ethanol, followed by one volume of water. The honey
solution was then passed through under constant vacuum. The honey filtrate was used for
the antimicrobial test.
42 S.Bogdanov


Honey Solution

Destilation ~
1. Removal of volatile substances

(2 h. 60°C under vacuum)

honey without

Removal of
2. Non-polar, non-volatiles (C-18)
3. Acids (anion exchange)
4. Bases (cation exchange)



Figure 1. Scheme of the fractionation and testing of the different antimicrobial fractions. Removal of Bases (Relatively Polar). The cation exchange column was
converted into the H-form by passing 2 ml of 2 M Hel. Then it was washed with water
until the eluate was neutral. The honey solution, where the bases at the acidic pH of
honey are in protonized form, is passed through. The flitrate is then set at the pH of
the initial honey solution. Removal of Acids (Relativly Polar). The anion exchange column was con-
verted into the OH-form by passing of2 ml of2 M NaOH. Then it was washed with water
until the eluate was neutral. The honey solution is then set at pH 11. The honey solution,
where the acids at this high pH are in their dissociated form, is passed through. The flitrate
is then set at the pH of the initial honey solution.

2.2.4. Expression ofAntimicrobial Inhibition. Whole Honey Solutions. Results were calculated by the turbidities of the in-
cubation mixtures at the end of the bacterial test. They were expressed in % inhibition
compared to the absorbance of the control (control = 0 % inhibition or 100 % growth). Honey Fractions. The increased bacterial growth after the removal of a cer-
tain honey fraction (see below) was attributed to the removal of antibacterial substances
by this fractionation step.
Non-peroxide Antibacterial Activity of Honey 43

Example: A honey has an initial bacterial inhibition of 90 %. The inhibition after re-
moval of: volatiles: 80%, non-polars: 65%, bases: 40% and acids: 20%. The initial inhibi-
tion of 90% was set as 100 parts to represent the whole antibacterial capacity of this
honey. The partial inhibition after the loss of the different fractions in relation to the an-
tibacterial capacitys of the whole honey was calculated as: 80/90x I 00, 65/90x I 00 etc.=
89, 72, 44, 22. The inhibition capacity of each absorbed fraction is then: 100-89; 100-72
etc = 11,28, 56 and 78. Then the sum of the inhibition capacities of the fractions was set
as 100 %, e.g. for the above example:

11+28+56+78 = 173 = 100 %.

In table 2 the % inhibition, attributed to each fraction, is a relative inhibition

number, e.g. the relative inhibition of the volatile fraction is 1lI173xl 00 = 6%.
The average sum of the inhibition parts of each fraction of the ten honeys in table 2
was 119 for Staph.aureus and 223 for Sarcina lutea. If all antibacterial substances were
fractionated by our procedure and if their action is additive a sum of 100 should be ex-
pected in the above example. The significantly higher percentage than 100 for the frac-
tional inhibition of Sarcina might be due to interactions of the different antibacterial
fractions, when they act as a whole.


3.1. Relative Distribution of Antimicrobial Activity among Different
Honey Fractions
We fractionated honey in 4 basic substance groups: volatile, non-volatile and non-
polar, acidic and basic substances.

Table 1. Relative distribution of antibacterial activity in different honey fractions The sum of the
antibacterial activities against Staph. aureus und Sarcina /utea, attributed to each fraction is set as
100 % (see Methods)

% antibacterial activity in different fractions

Acidic Basic Non-polar Volatile
Honey St. Say. 51. Say. St. Say. St. Sar.
Manuca N.Z. 100 75 0 10 0 5 0 10
Sunflower It 58 46 13 15 16 25 13 15
Rape CH 25 40 7 33 63 22 5
Lavender Fr 25 27 34 30 23 29 18 14
Mountain CH 24 25 60 25 8 25 8 24
Blossom S. America 62 73 13 20 9 7 16 0
Honeydew CH 45 46 26 15 26 15 2 24
Honeydew CH 32 31 37 31 19 31 12 6
Honeydew CH 43 26 22 26 19 26 15 23
Honeydew Europe 43 32 25 31 26 37 6 0
Average 46 42 24 24 21 22 10 12
Minimum 24 25 0 10 0 5 0 0
Maximum 100 75 60 33 63 37 18 24
44 S.Bogdanov

Table 2. Correlation between free acidity, diastase and invertase

activity and bacterial inhibition of Slaph.aureus. r-correlation
coefficient, p---probability level at 95 %. For units see Methods
Free acidity vs. Diastase number vs. Invertase number vs.
inhibition inhibiton inhibition
0.35 0.65 0.58
p 0.001 0.0005 0.001
n 81 37 37

In table I the relative distribution of antibacterial activity in these fractions against

Staph.aureus and Sarcina lutea is summarised. The acidic fraction has the greatest inhibi-
tory power, while the volatiles are the weakest bacterial inhibitors. The relative distribu-
tion of the antibacterial activity in the different fractions is the same against both bacterial
strains tested. On the average, for both strains the following relative distribution of an-
tibacterial activity was observed: acids: 44%, bases: 24%; non-polars: 21 % and 11: vola-
If the differences between the distribution of activity among the different groups
were tested by a t-test, only the difference between the volatile activity on one side and
the activity in the acidic (p=O.OOO) and the basic fraction (p=0.05) proved to be signifi-
cantly different. This is due to the variation of the distribution among the fractions of the
different honey types. In the manuca honey 90% of the activity was found in the acidic
fraction, in the rape honey the greatest part of the activity was in the non-polar fraction
and in one Swiss blossom honey the basic fraction had the highest activity.

3.2. Origin of the Non-peroxide Bacterial Inhibition

3.2.1. Plant Origin. In fig .2 the bacterial inhibition of 9 unifloral and 2 mixed (dif-
ferent blossom and different honeydew origins) are shown. Rhododendron honey had the


::cs 40
.- 30
o ro o ro
a::: o :c a:::
Figure 2. Non-peroxide activity of unifloral honey against Staph.aureus: The values (averages) are for: Rh Rho-
dodendron, n=3 ; Eucalyptus, n=4; Orange, n=3 ; Chestnut, n=7, Blossom, n=30; Acacia, n=7; Sunflower, n=4;
Lavender, Dandelion, n=2; Honeydew, n=IO; Rape, n=7.
Non-peroxide Antibacterial Activity of Honey 45

lowest, while rape honey had the highest inhibitory power. However, there is a consider-
able variation in each honey type. In order to prove for significant differences between the
honey types, a greater number of honeys should be analysed. Together with the fact, that
the relative distribution of the antibacterial activity is different in the various honey types,
the data shown here suggest, that some of the non-peroxide activity is of floral origin.

3.2.2 Bee Origin. Honey acidity, distase and and invertase are known to have a bee
There is a highly significant correlation between the free acidity, the diastase and
the invertase activity on one hand and the bacterial inhibition on the other (table 2). These
results show, that a part of the non-peroxide antibacterial activity has a bee origin.

3.3. Influence of Heat and Storage

The experiments were carried out with light blossom- and dark honeydew honeys.
Heating of both honey types at 70° C for 15 minutes had no or very little effect on
the non-peroxide actitivity (table 3.). Under the same conditions the peroxide accumula-
tion capacity of blossom honeys is severely damaged (10).
In a next experiment glass pots with blossom or honeydew honeys were stored in the
light (day-light) or in the dark at room temperature (about 20-25° C). The results are sum-
marised in fig. 3. After 5 months of storage the activity remains the same, while after 15
months there is a small drop of activity of about 20 %. The results were the same both for
both types of honeys stored in the light or in the dark. Under the same storage conditions
the peroxide accumulating capacity of honey is strongly reduced, especially when blossom
honeys are stored in the light (10).

1. There are four different honey fractions, which account for the non-peroxide an-
tibacterial activity. The antibacterial activity of these fractions is:

acids> bases"" non- polars > volatiles.

2. There is evidence for the floral- and for the botanical origin of the non-peroxide
antibacterial activity
3. The non-peroxide activity is only slightly affected by heat and by storage for 15
months in the light or in the dark.

Table 3. Effect of heat on non-peroxide activity Fresh honeys of floral

or honeydew origin were heated for 15 minutes at 70° C. Values are
mcans ± SEM and are expressed in % of the initial inhibition

Bacterial inhibiton
Honey n % of initial
Blossom (light) 3 86 ±4
Honeydew (dark) 4 94 ± I
46 S.Bogdanov



"'iii 60



blossom, blossom, honeydew, honeydew,
light dark light dark
3, 5 and 15 months

Figure 3. Effect of storage on the non-peroxide activity The non-peroxide activity of 7 mixed blossom honeys and
5 honeydew honeys, stored in glass pots in the light and in the dark, was tested. The values are averages for the
above honeys.


In honey there are two sorts of antibacterial agents or so called inhibines. One of
them is heat- and light sensitive and has its origin in HP2 . The other consists of thermo-
stable substances. In our study we tested the antibacterial activity by a turbidity test with
20% honey solutions with Staphylococcus aureus and Sarcina lutea strains. Under these
conditions destroying all the HP2 with catalase had no effect on the antibacterial activity.
Thus with this test only the non-peroxide antibacterial activity is measured. We used this
test to measure the non-peroxide antibacterial activity of whole honeys and of different
honey fractions. The results can be summarised as follows:

I. By fractionation in different substance classes the non-peroxid antibacterial ac-

tivity is distributed among 4 fractions with different chemical characteristics:

acidic; basic (both relatively polar); non-volatile and non-polar; volatile.

2. The antibacterial activity of the different fractions , tested in 10 different honeys


acids> bases = non- polar, non-volatiles> volatiles.

This order was the same for both Staph. au reus and Sarcina lutea as test strains
3. The distribution of activity was, however, dependent on the honey type : In
manuca honey almost the whole activity was found in the acidic fraction, in rape
honey the greatest activity was in the non-polar fraction and in one Swiss moun-
tain honey the basic fraction had the highest inhibitory power.
Non-peroxide Antibacterial Activity of Honey 47

4. There were differences between the antibacterial actIVIties of different honey

types: rhododendron, eucalyptus and orange honeys had a relatively low, laven-
der, dandelion, honeydew and rape honeys had a relatively higher activity. This
result, together with point 3. suggests, that some of the non-peroxide activity has
a plant orign.
5. There is also a significant correlation between the acidity, diastase- and inver-
tase activity, all of bee origin, on one hand, and the non-peroxide activity, on the
other. Thus, a substantial part of the non-peroxide activity has also a bee origin.
6. The non-peroxide activity is not or only slightly affected by heat (15 minutes
70 C and by storage for 15 months at room temperature.

I. Molan. P. (1992) The antimicrobial activity of honey I. The nature of antibacterial activity. Bee world, 73.
2. White, J.w., Subers, M.H. and Schepartz A.I., (1963), The identification ofinhibine, the antibacterial fac-
tor in honey, as hydrogen peroxide and its origin in honey glucose-oxidase system. Biochim. Biophys. Acta
n 57-70
3. White, 1. W. and Subers, M.H. (1964), Studies of honey inhibine, 3. The effect of heat. J.Apic.Res. 3
4. Dustmann, J.H. (1972) Ueber den Einfluss des Lichtes auf den Peroxid-Wert des Honigs. L.Lebensm.Un-
ters.Forsch. 148,263-268
5. Dustmann.J.H. (1979) Antibacterial effect of honey. Apiacta 14, 7-11
6. Morse, R.A. (1986) The antibiotic properties of honey. Pan-Pacific Entomologist 62, 370-371
7. Gonnet, M. and Lavie, P. (1960) Influence du chaufage sur Ie facteur antibiotic du miel. Annales de
I' Abeille (Paris) 3, 349-364
8. Mohrig, W. and Messner, R. (1968), Lysozym als antibacterielles Agens im Honig und Bienengift. Acta Bi-
ologica Medica Germanica 21, 85-95
9. Radwan S.S. EI-Essawy A.A. and Sarhan, M.M. (1984) Experimental evidence for thc occurrence in honey
of specific substances active against micro-organisms. Zentralblalt Mikrobio!. 139,249-255
10. Bogdanov, S. (1984) Characterisation of antibacterial substances in honey. Lebensm. Wiss. Techno!., 17,
II. Sabaticr, S., Amiot, M.1., Tachini, M and Aubert. S. (1992) Identification offlavonoids in sunflower honey.
I.Food Sci. 57, 773-774
12. Tomas Barberan, FA., Ferreres, F Garcia-Viguera, C. and Tomas-Lorente, F Flavonoids in honey of dif-
ferent geographical origin. (1993) Z.Lebensm.Untersuch.Forsch. 196,38-44
13. Metzner 1., Bekemeier, H .• Paintz, M and Scheidewand E. and (1975) Zur antimikrobiellen Wirksamkeit
von Propolis und Propolisinhaltsstoffen. Pharmazie, 34, 97-102
14. Bogdanov S. (1989) Determination of pinocembrin in honey using HPLC. J.Apic.Res. 28, 55-57
15. Yatsunami, K. and Echigo, T. (1984) Antibacterial activity of honey and royal jelly. Honeybee Science 5,
16. Lavie, P. (1963) Sur I'identification des substances antibacteriennes presentes dans Ie mie!.
C.R.Seanc.Acad.Sci. Paris. 256, 1858-1960
17. Toth, G., Lemberkovics, E. and Kutasi-Szabo (1987) The volatile components of some Hungarian honeys
and their antimicrobial effects. 127,496-497
18. Roth, L.A., Kwan, S. and Sporns P. (1986) Use ofa disc assay to detect oxytetracycline residues of honey.
I.Food Prot. 49, 436-441
19. Schuler, R and Vogel, R. (1956) Wirkstoffe des Bienenhonigs. Arzneimittel Forsch. 6, 661-663
20. Bogdanov, S. , Rieder, K. and Ruegg, M. Neue Qualitatskriterien bei Honiguntersuchungen. Apidologie,
21. Lavie, P. Proprietes antibactcriennes et action physiologique des produits de la ruche et des abeilles in:
Traite de Biologie de l'Abeille (R.Chauvin, editor) Masson & Cie pp.2-115
22. Swiss Food Manual, Chapter 23 A, Honey, Bern. Eidgen6ssische Druck und Materialzentralle, 1995



Gennady Rosenblat,1 Stephane Angonnet,2 Alexandr Goroshit,3 Mina Tabak, I

and Ishak Neeman l

IDepartment of Food Engineering and Biotechnology

Technion-Israel Institute of Technology
Haifa 32000, Israel
Paris 91305, France
3Tzuf Laboratori es Ltd.
P.O.B. 408, Kiryat Shmona, Israel


Honey is known to exert beneficial effect on many pathological conditions. The full
gamut of its biological activity has yet to be elucidated. In this study five type of honey
(regular commercial honey, Chinese honey and three other honeys, namely, Laryngomel,
Bronchomel, and Dermomel) were studied for their antioxidant effect. Unlike regular
honey, three last honey species were a product of bees which have been fed on a mixture
of several medico-herbal water extracts with regular honey. All the assessed honeys dem-
onstrated prevention of j3-carotene degradation in linoleic acid emulsion and obviation of
superoxide radical generation by xanthine/xanthine oxidase system. Laryngomel and
Bronchomel were particularly effective in reactive oxygen species scavenging at a concen-
tration range 7-50 J.lg/ml. The relationship between the antioxidant properties of honey
and its physiological activity is discussed.


One of the common properties of many plant natural products is the considerable de-
gree of protection afforded against oxidative attack. The spontaneous reaction of oxygen
or oxygen containing radicals with organic compounds leads to cell and tissue damage.
Radicals and the radical-generating process are normally neutralized by antioxidative de-
fense mechanisms [1], but in certain situation such as aging, inflammation, etc., radical

50 G. Rosenblat et al.

generation may increase with a consequent acceleration of accumulating damage [2,3]. In

such situation the use of antioxidants in medication or of plant extracts with antioxidant
properties in food has been advocated to minimize any oxidative degradation of the target
molecules [4,5].
The antioxidant potential of honey, which is a widely used natural product contain-
ing mostly floral compounds, has not been sufficiently evaluated. Most studies on the bio-
logical properties of honey have thus far focused on its antibacterial activity [6]. Yet, in "
folk medicine" honey also is believed to exert a beneficial effect on ulcers, the nervous
system, the heart, the liver and the digestion [7]. The full spectrum of honey biological ac-
tivities is still unc1arified but evidently much extend beyond were antibacterial activity.
The present investigation attempted to fill the gap by studying the antioxidant effect of
several types of honey.


All solvents were obtained either from Frutarom (Haifa, Israel) or Biolab (Jerusa-
lem, Israel). Specific compounds were all purchased from Sigma (St. Louis, USA) except-
ing linoleic acid (from Fluka Chemie, Buchs, Swizerland) and OPA (from Zymed, USA)

Honey Venue and Production

Commercial Chinese honey was received from Cho Yung International, Israel.
Regular commercial honey was produced by kibbutz Ayelet HASHAHAR (Israel). The
commercial honeys Bronchomel, Laryngomel and Dermomel were produced by Tzuf
Laboratories Ltd., Kiriat Shmona, Israel; unlike regular honey, these three honey species
were a product of bees which have been fed on a mixture of several medico-herbal water
extracts with regular honey. The taxonomical data on the medical herbs used in the men-
tioned extracts are given in Table 1. Glucose-fructose syrup, which was used in part of the
experiments as a model solution, was prepared by adding 35 g of each sugar to 100 ml of
hot bidistil1ed water.

~-carotene Degradation Test

Antioxidant activity was assessed via emulsion with p-carotene and linoleic acid,
whose degradation was determined by the method of Marco [8] with modifications.
Briefly, 0.4 mg of p-carotene, 100 !-II of linoleic acid and 200 !-II of Tween 40 were dis-
solved in 20 ml of chloroform, which was then evaporated by nitrogen. The model emul-
sion was now prepared by adding 30 ml of water. One gram of honey was next dissolved
in 2 ml of water (or of appropriate buffer) and mixed with 3 ml of the emulsion; 2 ml of
95 % ethanol was added to the mixture, which was finally incubated at 50°C for several
time. The antioxidant activity of I g of honey in 7 ml of reaction solution was compared
with that of 1 !-Ig/ml (5.5 10-3 mM) and 5 !-Ig/ml (2.8 10.2 mM) ofBHA.
The degradation of honey protein in the different tests was achieved by autoc1aving
for 30 min 2 ml of water (or buffer) containing 19 of honey at I atm. Citric buffer, 0.1 M,
pH=2, or citric-phosphate buffer, 0.1 M, pH=4.5 or phosphate buffer, 0.1 M, pH=7 were
used for the pH-change in the different tests.
Antioxidant Properties of Honey and Medical Plant Extracts 51

Table 1. Medicinal herbal are used in forming of honey Laryngomel , Bronchomel ,

and Dermomel and medical-biological characteristic of the honey
Type of honey Family of the herbal- Medical- biological characteristic
Laryngomel Lauraceae active against laringitis,tracheitis, glossitis
Bronchomel Mirtaceae active against inflamation of the upper respiratory tracts
Dermomel Cruciferaceae active against suppurative wounds and chronic ulcer
*All plants are used in conventional or alternative medicine

Inhibition of O 2- Production by Xanthine/Xanthine Oxidase System

The production of superoxide anion was evaluated by chemoluminescence. Reaction
was initiated at room temperature by adding 250 f.ll solution of xantine oxidase solution
(0.65 units/ml) to 2-ml of Hanks buffer (pH=8.3) containing 100 f.lM of xanthine and 60
f.lM of lucigenin. The mix was put in a luminometer (model 597B, Technion Physics De-
partment, Haifa, Israel). The light production remined constant throughout the 2 - 5 min-
utes following onset of the reaction, during which period 25 f.ll of the differently diluted
(in water) honey was added dropwise. The chemoluminiescence was now measured di-
rectly in triplicate samples.

Xanthine Oxidase Activity

Activity of xanthine oxidase was determined by uric acid production. Uric acid gen-
eration was monitored by absorption spectroscopy, at 290 nm, under the same condition as
in experiment with superoxide anion determination.


Effect of Honey on the Degradation of ~-Carotene in Emulsion with

Linoleic Acid
The oxidation of ~-carotene in emulsion with linoleic acid in the presence of various
honey species is shown on Fig. 1. A change in the absorbency at 470 nm (expressed in
52 G. Rosenblat et al.


,.Q -
rJl c:

" ..."
,.Q 0

'~ci 60
•.r:: •
.2 0
-~ 40
o w............ ' I ", I I I

o 50 100 150 200 250 300 350

Time [min]

Figure 1. Effect of honey on p-carotene degradation in emulsion with linoleic acid.

percent of initial absorbency) is characteristic of 0-carotene degradation in course of the

oxidation. As can be seen from the data the antioxidant activities of all the honey solutions
in distillated water (142,8 mg/ml) at normal for honey pH (about 4.0 ) are comparable
with that of BHA (which is a potent antioxidant employed in the food and chemical indus-
tries [9]) in a concentration of I /-!g/ml or 5 /-!g/ml, respectively . The antioxidant effect of
honey in emulsion with linoleic acid was found to be pH dependent. The antioxidant ef-
fect of all species of honey is maximal at physiological pH (7) but reduces with decrease
in the pH, so that at pH of human stomach (pH=2) the tested honey types do not show an-
tioxidant activity excepting Laryngomel and Bronchomel which demonstrate a negligible
antioxidant effect. The capacity of honey to protect against 0-carotene oxidation was not
altered by heating of the honey solution in the protein- cleaving state, thus supporting our
contention regarding the non-proteine nature of the honey antioxidant moiety (data not
shown). Nevertheless, the possibility in this case that thermal inactivation of antioxidant
protein is compensated for by products of Maillard reaction cannot be entirely excluded,
and it has been observed repeatedly that various Maillard reaction products formed
through the interaction of protein and carbohydrates during the heat-processing of food do
exert antioxidant activity [1OJ-

Inhibition of 02- Production by Xanthine / Xanthine Oxidase

Xanthine / xanthine oxidase system is one of the two natural system of superoxide
anion generation . Xanthine oxidase-derived superoxide anion has been linked to postis-
chemic tissue injury and the generation of neutrophil chemotaxis [II] . Hence curtailment
of 0 2' generation by this enzymatic pathway would be beneficial in the case of ischemia.
Honey demonstrates a significant inhibitory effect on radical generation by xanthine/xan-
thine oxidase system. Indeed extintion of the chemoluminescence was observed in the
presence of the regular honey (159-900 /-!g/ml), while an even greater effect was shown
Antioxidant Properties of Honey and Medical Plant Extracts 53

12° ~1


o 675 1450 2 25
Concentration, IJ.gfml

Figure 2. Effect of honey on superoxide anion production by a xanthine/xanthine oxidase enzymatic system.

by Dermomel and Chinese honey in the same concentration range. Strong effect on super-
oxide amount was found for Bronchomel and Laryngomel at a concentration of 7- 50
Jlg/ml (Fig 2. but data for Bronchomel is not shown). This effect was not abnogated by
heating the honey samples at 100 °C and I atm for 30 min , thus supporting the supporting
nonproteine nature of the inhibiting compounds. No significant inhibitory effect on the su-
peroxide anion generation by xanthine/xanthine oxidase was displayed by glucose-fruc-
tose syrup even in markedly concentrations.
To justify whether this activity was due to an inhibitory effect on the enzyme itself,
a control experiment was carried out, in which production of uric acid was measured by
UV spectroscopy at 290 nm. Under the same experimental condition it was not de-
mostrated an inhibition of uric acid production by xanthine oxidase even in presence of
markedly concentrations of honey.


Honey contains various minerals and organic compounds but is comprised mainly of
sugars (about 80 %) and water (about 20%) [12]. In addition, honey is known to contain a
number of enzymes such as diastase, invertase, saccharase, catalase, glucose oxidase. De-
pending on the source from which bees obtain the material for honey production, the
honey will acqure its specific aroma, color and test. Apparently , the medicinal and the nu-
tritional properties of honey depend in part, also from the chemical composition of the
flowers from which bees collect nectar. For this regard, there is growing interest in the
honey created by bees that are nourished on medical herbs for the properties of the final
product are enhanced during processing of the natural nectar in the bee's body. Using
various plant extract that are well known in alternative and conventional medicine, it have
been able to develope honey substitutes for easing cough and bronchitis (Bronchomel),
54 G. Rosenblat et al.

sore and inflamed throat (Laryngomel ), or for treatment of wounds (Dermomel). This and
natural honey studied by us have all demonstrated antioxidant activity, particularly an in-
hibitory effect on superoxide radical production by xanthine Ixanthine oxidase. The inhi-
bition of 02- production is manly due to scavenger properties since the tested honey are
not xanthine oxidase inhibitors. Regular honey was found to be a weaker than the other
samples tested by us. Presumably the protective properties of manufactured honey against
reactive oxygen species reside mainly in the substances extracted from the medicinal
herbs. Although butulated hydroxyanisole was used by us as the standard for assesing the
antioxidant properties of honey it would be incorrect to compare BHA and honey antioxi-
dant activity quantitatively, for obviously honey contains the biologically active ingradi-
ents in very low concentration. Nevertheless a simple calculation (using data from the
~-carotene degradation test) shows that active dose of natural honey which is recom-
mended for medicinal use (one teaspoonful, which is about 7 g ) contains an amount of
antioxidant which is equal to 50-250 Ilg of the potent antioxidant BHA. Turning now to
the mechanism undrelying honey biological activity, accumulating evidence supports the
assumption that the protective effect of honey against inflammation of whatever sourse is
associated with its antioxidant moiety. Indeed, it has been shown that synthesis of inflam-
mation mediators like leukotrienes and prostaglandins is involves the formation of arachi-
donic acid lipoperoxides [13]. Furthermore, in the wake of many types injury, including
trauma, cells are known to rupture and release their contents, which include transition
metal ions that can rapidly catalyze radical-mediated transformation and tissue injury [2].
Macrophage activation and the disruption of mitochondrial function may also result in the
formation of excess reactive oxygen species [14,15]. Human phagocytes destroy bacteria
or virus-infected cells throughout an oxidative burst of nitric acid NO", hypochlorite CIO'
and superoxide 02- , but unfortunately the damage produced by radicals liberated from the
phagocytic cells can sometime extend beyond the intended target and injure also surround-
ing tissue [16 ]. Uniquely vulnerable to such oxidative damage are epithelial cells lining
the respiratory airways [17]. A part from potential oxidant exposure owing to normal cel-
lular metabolism, the respiratory epithelium is exposed to relatively high oxygen tension
and often also exposed to air pollutants, phagocytes, catalase-negative bacteria, and reac-
tive xenobiotic-drug metabolites. From all the above, we can conjecture that the signifi-
cant protective effect of honey (particularly Bronchomel and Laryngomel) against
respiratory airway inflammation is explanable, in part, by its antioxidant properties. In
conclusion, the in vitro antioxidant effect of honey revealed by the present study seems to
comprise an is important attribute of honey, especially for such as was produced by bees
nourished on mediI' ina I herbs.

I. Reiter R.J. (1995). Oxidative processes and antioxidative defense mechanisms in the aging brain. FASEB J.
2. Kehrer J.P., Smith c.v. Free radicals in biology: sourse, reactivities, and roles in the etiology of human dis-
eases in: Natural antioxidants in human health and disease (Balz Fri Editor.), Academic Press, Foreword,
1994, pp.25--62,
3. Harman D. (1993). Free radical involvement in aging. Pathophysiology and therapeutic implication. Drags
Aging. 3, 60-80
4. Bermond P. Biological effects of food antioxidants in: Food antioxidant, (Hudson BJ.F. Editor), Elsevier
Science Publishing Co., Inc. N.Y. 1990, pp. 193-251
5. Sies H., (1993). Strategies of antioxidant defense. Eur. J. Biochemistry. 215, 213-219
6. Molan P.c. (1992). The antibacterial activity of honey. Bee World. 73,5- 26
Antioxidant Properties of Honey and Medical Plant Extracts 55

7. Loyrish N: Bees and people. Mir Publishers, Moskow. 1977

8. Marco, G.J. ( 1968). A rapid method for evaluation of antioxidants. JAm Oil Chem.Soc .. 45, 594-598
9. Kukagawa K., Kunugi A., Kurechi T. Chemistry and implications of degradation of phenolic antioxidants
in Food antioxidant. (Hudson B.1.F, Editor), Elsevier Science Publishing Co. Inc. N.V. 1990
10. Eichner, K. (1981). Antioxidative effects of Maillard reaction intermediates. Prog.Food Nut/:Sci .. 5,
11. Cotelle, N., bernier, J.L., Henichart, J.P., Catteau, J.P., Gaydou, E.. Wallet, 1. C. (1992). Scavenger and an-
tioxidant properties of ten synthetic flavones. Free Rad.Biol.Med .. 13,211-219
12. White, I.w., Jr. (1978). Honey. Adv. Food Res., 24, 288--374
13. Borgeat,P., Samuelson, B. (1979). Proc. Natl.A cad. Sci. USA. 76.3213-3217
14. Kehrer, J.P.(l993). Free radicals as mediators of tissue injury and disease. Crit.Rev. Toxicol .. 23,21-48
15. Rosen, G.M., Pou S., Ramos c.L., Cohen M.S., Britigan B.E.(1995). Free radicals and phagocytic cells,
FASEB J. 9. 200--209
16. Wright D.T., Cohn L.A., Li H., Fisher B., Li C.M., Adler K.B. (1994). Interaction of oxygen radicals with
airway epithelium, Environ. Health Penpect. 102 (Suppl 10),85--90
17. Meyer A.S., Isaksen A. (1995). Application of enzymes as food antioxidants. Trend. FoodSci. Technol. 6,


An Experimental Study with Histological Assessment of
Wound Biopsies

Th. 1. Postmes 1: M. M. C. Bosch 2 , R. Dutrieux 3, 1. van Baare 4 , and

M. 1. Hoekstra 4

IDepartment ofInternal Medicine

Academic Hospital Maastricht
2Leiden Cytology and Pathology Laboratory
3Dcpartment of histopathology
Academic Hospital, Utrecht
4Burns Unit Red Cross Hospital
Beverwijk, The Netherlands.


In a pilot-study deep dermal burns, identical in depth and extent, were made at each
flank of Yorkshire pigs. Wound healing characteristics and measurements of dermal thick-
ness were investigated by comparing biopsies histologically in pairs. Wounds were treated
with either honey of a defined antibacterial activity, or sugar, or silver sulfadiazine (SSD).
Biopsies were taken on post burn days 7,14,21,28,35 and 42. Wounds treated with SSD
were fully epithelialized after 28-35 days, whereas those treated with honey and sugar
were closed within 21 days. In 5 out of 6 wounds the neodermis of the sugar treated burns
was thicker than the neodermis of those treated with honey. In all honey experiments, on
day 21, wounds were best microscopically characterized by (i) a quiet granulation tissue,
(ii) an inconspicuous inflammation and (iii) a decrease of actine staining of myofi-
broblasts. In contrast, sugar treated wounds appeared ditTerent especially on day 21 and
later. Furthermore, we observed in various sections signs of inflammation which were
linkcd to pcrivascular infiltrates and well-stained myofibroblasts. These results suggest a
difference between sugar and honey treatment. If antibacterial activity and anti-inflamma-
tory activity count in wound treatment, then honey has to be prefered above sugar.

* Correspondence should be addressed to: Dr. Th.Postmes, St.Servaasklooster 22; 6211 TE Maastrcht.

S8 Th. J. Postmes et al.


Epidermal regeneration of a wound is a complex process in which residual epi-

thelial cells proliferate in an integrated manner into intact epidermis. In deep second de-
gree burns re-epithelialisation primarily takes place from the wound margins; apparently
most epithelial cells from hair follicles in the thermal injured area are not viable any-
more. Granulation tissue is formed on the base of vessels and fibroblasts in the residual
dermis and from the fatty tissue septa. Since the choreography of wound healing factors
is still in the dark, an evaluation of the efficacy of current wound dressings remains
Topical application of honey to open wounds has been recognized for centuries to be
effective in controlling infection and producing a clean granulating wound bed. Both
honey and sugar have been highly praised and recommended as proper treatments for trau-
matic wounds, burns, decubitis and ulcera l - 5 •
Though honey is a complex medium it is safe and non-toxic 3- s. Moreover it is able
to disinfect infected burn wounds, but it can also prevent colonization of non-infected
burn wounds by bacteria until complete epithelization3 •
According to Condon (1993) honey acts mainly as a hyperosmolar medium that pre-
cludes bacterial growth and as such does not differ from simple sugar6. Thus the real ex-
planation of the antimicrobial effect of honey lies in its physical properties, not in its
chemical composition 6 • However, the antibacterial effect after dilution up to 4% proves
that honey must have intrinsic antibacterial properties quite distinct from sugar 7,R. More-
over the presence of e.g. traces of zinc, vitamins, amino acids and a large number of or-
ganic substances may contribute to and support the nutrition of the injured tissue.
In spite of the numerous publications on honey and sugar as topicals for wound
treatment, not a single clinical study comparing both substances has been published. In
this pilot experiment, which is part of a larger evaluation on topical treatments, we wish to
report some observations on the healing of deep second degree burns after treatment with
honey, sugar or silver sulfadiazine.


A standard burn wound model, published by Hoekstra et al. (1993)., was strictly fol-
lowed as described 9 • In brief:


Experiments were performed in accordance with the Dutch Law on Animal Experi-
mentation and the experimental protocol employed was approved by the Animal Welfare
Committee of the University of Amsterdam. Three Yorkshire pigs of approximately 14
weeks (with a body weight of 25-35 kg) were thermally injured with a brass block of 6.7
x 6.7 cm, weighing 450 g. Twelve areas of 7 x 7 cm ( six on each flank) were marked in a
symmetrical way, using the processus spinosi as an anatomical landmark. The block was
heated up to 170°C and applied during 20 s without exerting pressure. The epidermal rem-
nants of the burned skin were removed shortly after thermal injury. The total burned sur-
face area amounted to not more than 10 % of the total body surface area.
Speeding Up the Healing of Burns with Honey 59

The details of the control of the animal's wellbeing, housing, food, general anaesthe-
sia, daily wound treatment and covering were all similar to Hoekstra et aI., 1993.

Topical Agents
Experiment (a), pig no 120, Silver sulfadiazine 1% cream vs honey.
SSD ( Flammazine R , Duphar, Weesp, The Netherlands), as a topical agent and the
most commonly used ointment for treating burns, was compared to unprocessed honey
with a known antibacterial activity. The total inhibine (i) score of this particular lime
honey was 15 as described previouslylo.
Experiment (b), pig no 64. Honey vs honey. The honey was the same as in exp.(a).
Experiment (c), pig 137. Honey vs sugar (artificial honey). The honey was the same
as in expo (a). The osmolarity of the sugar paste containing 40% glucose, 40% fructose,
10% saccharose and 10% water was approximately that of honey.

Biopsy and Histology

Each week (on post burn days 7, 14, 21,28, 35 and 42) biopsies were taken symmet-
rically from wounds on the left and right flanks. Biopsies included the wound bed as well
as the healthy skin of the wound margins. Tissue fixing and staining were as described
previousl/. An additional staining was included to detect myofibroblasts by an anti-Alpha
smooth muscle actin stain.


In this pilot study the wound healing pattern of the silver sulfadiazine treated flank
of pig study no 120 was similar to the pattern reported earlier9 .
Macroscopically it is almost impossible to assess the grade of epithelialization due
to the presence of the crust. The time of a hundred percent epithelialization and the meas-
urements of the dermal thickness of experiments a, band c, as found microscopically, are
summarized in table I.
An uninterrupted epidermis was found on day 21 for both honey (experiments a,b
and c) and sugar (experiment c). Placed side by side the histology of the biopsies turned

Table 1. Some data related to deep second degree burns treated with silver sulfadiazine (SSD),
honey and sugar (artificial honey)

Thickness of the dermis I after day:

Exp. 100%
no Pig Treatment epithelia1ised 14 21 28 35 42
a. 120 honey 21-28 days' 1.7 3.3 2.7 2.9 3.0
vs SSD 28-35 days 1.0 1.1 2.7 3.3 2.g
b. 64 honey 21 days 3.3 1.5 2.0 2.0 1.7
vs honey 21 days 2.5 3.3 1.8 1.5 1.7
c. 137 sugar 21 days 1.8 3.5 2.7 2.3 2.2
vs honey 21 days 1.3 2.5 1.5 1.3 2.4
'The thickness of the dermis is measured in the middle of:he biopsy. All data are given as a ratio of thickness of the burn wound
versus that of normal skin. The pig's skin is normally 3 to 4 mm thick and in many ways it resembles the human skin.
'On day 21 the wound is almost fully epitheiialised, the next biopsy was on day 28, so the 100% point lies somewhere between
day 21 and 28.
60 Th. J. Postmes et a!.

.' ,


,< ,
'",; •

- ,

, ..

Figure I. On day 21 , in sugar treated burns, deeper layers still show perivascular infiltration (Papanicoulaou
staining, x25 ).

out to be quite different. All sections of honey treated burns were, more or less, similar.
On post burn day 21 the general picture was primarily a quiet granulation tissue and a
light degree of inflammation was seen. In contrast, the sugar treated burns showed more
inflammation as revealed by the many perivascular infiltrations.
According to the stained sections of the biopsies on day 28 the dermal thickness of
the honey treated pig wounds was 5.1 mm, whereas the neodermal layer sugar of the
treated wounds of the contralateral side was much thicker (7 .6 mm), see fig 2.
On days 14, 21, 28 and 35 the honey treated burns showed a variation in dermal
between 4.0 mm and 9.5 mm while healed bum wounds in the contralateral sugar
side varied between 5.5 and 10.5 mm. Part of the swelling of the derm is can be put down
to local oedema in the first week of wound healing.
In the honey treated wounds the typical actin staining in myofibroblasts appeared to
be weak or diminished whereas sugar treated burns still showed a positive actin stain in
many myofibroblasts. The honey treated burns were covered with a translucent layer of
possibly non absorbed honey. This layer, however, could not be observed in the sugar ex-
periments. In both honey and sugar treated wounds a few isolated bacterial colonies were
found. These were less frequently seen in the honey than in the sugar wounds.
A few micro-pustules were found in the neo-epidermis of the sugar treated wounds
and less in the contralateral control side treated with honey. Bacteria were seen in the
eschars of the sugar and the honey treated bums, though less in the latter and none in
those of the SSD treated burns.
Speeding Up the Healing of Burns with Honey 61

Figure 2. (a): Normal pig's skin with sub-

dermal fat and muscle tissue in the depth.
(Papanicoulaou staining, x 12.5). (b). Pig
number 137: honey treated side on day 28,
the epidermis is uninterrupted and thus the
wound is full y epilheliali sed. Clearly the
dermis is thicker than normal. Staining and
enlargement as in (a). (c): Pig number 137:
sugar treated contralateral side on day 28.
Staining and enlargement as in (a). Note:
the dermis is 2.5 times thicker than normal.


Honey and sugar paste, in comparable hyperosmolar concentrations, are non-toxic

substances. The outgrowth of epithelial cells of the hair follicles is most likely inhibited
by the uptake of either silver or SSDII. For the speeding up of wound healing by honey
and sugar there are at least two explanations. First, in deep second degree burns, there are
still a number of epithelial cells which might survive. Some belong to hair follicles , and
others to sweat glands. Second, simple sugars in honey and sugar create a moist environ-
ment, which, as we know today, is a conditio sine qua non for quicker wound healing.
Our study confirms the clinical study with honey vs SSD of Subrahmanyan (1991).
In his study, honey turned out to be much better than SSD, which was proved by a signifi-
cant reduction of the number of days spent in hospital 5 Today, in terms of re-epithelialisa-
tion, honey scores better than SSD (Flammazine R) and also seems to be better than sugar.
In fig. 2 the most striking difference between sugar and honey is the thickness of the der-
mis. At cellular level the latter coincides mainly with a quantitative difference in inflam-
matory reaction and myofibroblast expression, both being less in the honey treated
During the process of wound healing myofibroblasts are supposed to playa role in
wound contraction. However, when contraction stops and the wound is fully epitheliali-
zed, these myofibroblasts which contain alpha myofibroblast smooth muscle actin disap-
pear. The scar classically becomes less cellular probably as a result of apoptosis. It is then
composed of typical fibroblasts with well-developed rough endoplasmatic reticulum but
with no more actin filaments. In hypertrophic scars, on the other hand, the expression of
alpha-smooth muscle actin in myofibroblasts persists l2 . Actually, in our study a persisting
actin staining was seen in the sugar treated bums of pig 137 (table I).
62 Th. J. Postmes et al.

Hence, the "early" disappearance of myofibroblasts in the honey treated bums sug-
gests a more advanced phase of wound healing than that of the sugar treated wound on the
contralateral side. Granulation tissue, as required for the formation of the neodermis, is
clearly suppressed initially by SSD. The effect of sugar on dermal thickness, as illustrated
in fig. 2, is an interesting observation. Apparently, the large variation in dermal thickness
in experiment b, where honey was applied on both flanks, implies that more data are
needed to reach a final conclusion in the honey vs sugar inquiry.
Unprocessed honey, unlike sugar, generates hydrogen peroxide when it becomes di-
luted. Hydrogen peroxide has been used for decades as an antibacterial agent with great
success. Mainly due to the so-called Fenton reaction it can easily produce free hydroxyl
(OB) radicals which are very bactericidal indeed.
In vitro, hydrogen peroxide has also a biphasic effect on fibroblasts. Within a con-
centration of 10-8 and 10--{i molliiter it stimulates cell proliferation, whereas at higher con-
centrations an inhibiting effect becomes prominent I 3. In vivo, data are lacking.
Nevertheless there is no reason to assume that fibroblasts in situ would react otherwise.
The hydrogen peroxide concentration provided by honey in an open wound may
have a growth inhibiting effect on the fibroblast. According to White et al. (1963), a 14 %
honey solution may generate in one hour 0.0-2.12 mmollliter 7• Of the 90 samples investi-
gated the one hour mean value was 0.47 mmolll ± 0.55 (s.d.Y4.
In bleeding wounds or in the presence of traces of catalase, hydrogen peroxide
breaks down almost instantly. This and other micro-environmental conditions make it al-
most impossible to predict the concentration of hydrogen peroxide at the interface of
honey and wound bed.


Honey is an ideal topical therapeutic agent, for it does not adhere to the wound sur-
face. In comparison to silver sulfadiazine it is definitely superior because of its quick re-
epithelialization and its absence of a sustained inflammatory reaction, which in SSD is
seen, even long after a complete epithilialization. Honey is also better than sugar paste, for
it has a natural antibacterial activity, which is lacking in sugar paste of similar osmolarity.
In addition it provides and maintains an environment in which healing can take place at an
optimal rate. Honey treatment of burns is certainly cost effective, because it shortens the
duration of treatment with ca. 25%, as shown in this study, and it certainly reduces, be-
yond all question, the duration of hospitalizations.

We wish to thank the Dutch Burns Foundation for its financial and personal support
of this study.

I. Knutson, R.A., Merbitz, L.A., Creekmore, M.A. and Snipes HG. (1981). Use of sugar and povidone -io-
dine to enhance wound healing:five years' experience. Southern Medical Journal 74, 1329--1335.
Speeding Up the Healing of Burns with Honey 63

2. Orouet, N. (1983) Utilisation du sucre et du miel dans Ie traitement des plaies infectees. Nouv Press Med
3. Efem, S.E. (1988). Clinical observations on the wound healing properties of honey. British Journal of Sur-
gery 75, 679--{i81.
4. Efem, S.E. (1993) Recent advances in the management of Fournier's gangrene: Preliminary observations.
Surgery 113, 200 -204
5. Subrahmanyam, M. (1991). Topical application of honey in treatment of bums. British Journal of Surgery
6. Condon, R.E. (1993). Curious interactions of bugs and bees. Surgery 113, 234-235.
7. White). W., Subers, M.H. and Shepartz AI. (1963). The identification of inhibine, the antibacterial factor in
honey,as hydrogen peroxide and its origin in a honey glucose-oxidase system. Biochemica Biophysica Acta
8. Molan, P.c. and Russel, K.M. (1988). Non-peroxide antibacterial activity in some New Zealand honeys.
Journal of Apicultural Research 27, 62 -67.
9. Hoekstra, MJ, Hupkens, P., Outrieux, R.P., Bosch, M.M.C and Kreis, R.W. (1993). A comparative burn
wound model in the New Yorkshire pig for the histo- pathological evaluation of local therapeutic regimens:
silver sulfadiazine cream as a standard. British Journal of Plastic Surgery 46,585--589.
10. Postmes, L Bogaard van den A.E. and Hazen, M. (1993). Honey for wounds, ulcers, and skin graft preser-
vation. Lancet 341,756--757.
II. Teepe, R.G.C., Koebrugge, E.l, L6wik, C.W.G.M., Petit, P.L.c.P., Bosboom, R.W., Twiss, I.M., Boxma,
H., Vermeer, BJ. and Ponec, M. (1993). Cytotoxic effects of topical antimicrobial and antseptic agents of
human kerationocytes in vitro. The Journal of Trauma 35, 8-19.
12. Schmitt-Graff., Oesmouliere A. and Gabbiani G. (1994) Heterogeneity of myofibro blastic cell plasticity.
Virchows Archive 425, 3--24.
13. Schmidt RJ., Chung, L.Y., Andrews, A.M. and Turner, T.O. (1992). Hydrogen is a murine (L 929) fi-
broblast cell proliferant at micro- and nanomolar concentrations. In:Second European Conference in ad-
vances in wound management. Proc. Int. Conf.Center Harrogate Oct. 20th-23th p. 117-121.
14. White, lW. and Subers, M.H. (1963). Studies on honey inhibine. 2. A chemical assay. Journal of Apicultu-
ral Research 2, 93-100.



S. R. Grobler' and N. J. Basson

Oral and Dental Research Institute

Faculty of Dentistry
University of Stellenbosch
Private Bag Xl
7505 Tygerberg, South Africa.

Various fruit juices with relatively low pH values are known to have erosive effects
on human tooth enamel in a reasonably short time (Grobler et al.1989! Clin Prev Dent,
11:23-28; 12:5--8). Honey, however, with a relatively low pH, could do the same. The
honey sample consisted mainly of nectar gathered from the blossoms of Eucalyptus trees.
The honey used did not contain any artificial preservatives or dilutents, neither had it been
heated by any artificial method. Sixteen human incisor crowns were ultimately ground wet
using 1200-grade silicon carbide paper. Each surface was divided into five segments and
each segment exposed to pure honey and diluted honey as well as to artificial honey for
different periods of time. The enamel segments were then investigated for their hardness
as well as for any etch pattern, by scanning electron microscopy. Scanning electron mi-
croscopy showed no erosion of enamel by natural honey over a period of thirty minutes
neither did Knoop microhardness tests show any deterioration of the enamel structure
even in a 4 times diluted honey solution. However, the theoretical solubility and ion prod-
uct values can be linked to the results obtained by the SEM study for the undiluted as well
as for the four times diluted artificial honey sample. The absence of any effect by pure
honey could only be partially attributed to the normal building blocks of enamel, namely
calcium, phosphorus and fluoride levels.
Seven different oral Streptococcus species, a Candida albicans strain and a Staphy-
lococcus aureus strain were tested for antibacterial sensitivity towards the honey. Minimal
inhibitory concentrations (MIC) were determined with a broth dilution method. The MIC
was the lowest concentration of the honey which yielded no growth.

* Correspondence address: S.R. Grobler, Oral and Dental Research Institute, University of Stellenbosch, Private
Bag XI, Tygerberg, 7505, South Africa.

66 S. R. Grobler and N. J. Basson

The oral streptococci as well as the C. albicans strain were relatively resistant to
Bluegum honey. However, the two species Streptococcus anginosus and Streptococcus or-
atis were inhibited at 17% and 12% respectively.


Honey has been used by man for many centuries. Written records date back to 3000
BC, which describe the practice of loading beehives onto small boats and moving them up
and down the Nile River, in Egypt, depending on the season, to enable the bees to collect
nectar where it was most abundant. Honey was also used by the Assyrians about 4000
years ago to embalm the bodies of their dead. Today honey is used extensively as a sweet-
ening agent and is also considered as a health food, for example, it is used as a sweetener
in cough mixtures. In a recent study in the British Journal of Surgery (1991)2, it was re-
ported that honey not only renders burn wounds sterile within 7 days. but that 87% of burn
wounds also healed within 15 days, against 10% in the control group.
Large expansion of fruit farming practices took place in such a way that additional
pollination became a necessity for flowering crops and orchards. Here hived honey-bees
are employed to help with the pollination process. These small hard worker-bees are per-
fect pollinators which beekeepers can manipulate to supply in large quantities at appointed
Two sub-species of the true honey-bee are found in South Africa. namely, the
striped African honey-bee and the Cape honey-bee l . The pH of honey produced by these
bees is slightly low, with an average value of about 3.9 for South Africa. It is interesting
to note that the average value reported is the same as that for the USA 4.
The demineralizing effects of excessive intake of fruit drinks, especially of citrus
fruits and other acidic beverages, are well documented because of their low pH values s- 7 •
Furthermore, it was reported that the highest dental caries incidence was found in farm
workers in a citrus producing group in comparison to a control group. Published results 7
on different soft drinks showed the following degree of enamel demineralization: Pepsi
Cola = orange juice> apple juice> Diet Pepsi Cola.
Therefore, owing to the low pH values reported for honey one could presume that it
might have a potential to cause erosion when it comes into contact with enamel. At the
same time acid produced from dietary sugars through the metabolism of oral bacteria are
also responsible for enamel erosion and dental caries 8 •
Several workers 9- 1J investigated the antibacterial properties of honey and found it ef-
fective against organisms such as Staphylococcus aureus, Salmonella typhi, Shigella spe-
cies, Escherichia coli and other pathogenic species. However, little information is
available with regard to the antimicrobial effect of honey on the oral bacteria.
Therefore, the purpose of this study was to determine the erosive effect of honey on
human tooth enamel and to evaluate honey for its antimicrobial activity on certain oral


The honey samples used in this experiment were harvested in the Western Cape of
South Africa and consisted mainly of nectar gathered from the blossoms of Bluegum
(Eucalyptus) trees. In order to be 100% certain that the honey sample collected were not
Human Tooth Enamel, Honey, and Bacteria 67

diameter ~ 3 mm

Figure 1. The division of the enamel surface into four segments as used in subsequent phases of the study.

contaminated it was harvested by the researchers and was not heated by any artificial
Sixteen human incisor crowns were polished up to 1200 grit fineness with silicon
carbide paper before being exposed to the different honey concentrations for different pe-
riods of time. The surfaces were ground until an area of enamel approximately 3 m in di-
ameter at the mid-central region had been smoothed. This was necessary to obtain a very
smooth enamel surface which would highlight the slightest signs of enamel erosion,
should any take place. The polished areas were washed and covered with a circular plastic
adhesive tape (Fig. 1) and the tape then cut with a surgical scalpel into four segments.

Phase 1
Seven crowns were used in the first phase. The first specimen was prepared by re-
moving the tape from one of the four segments from a crown. Honey was poured into a
Petri dish and the crown with the one exposed segment immersed in the honey and con-
tinuously agitated for 10 minutes. A second outer segment of tape was then removed and
the specimen again agitated in the honey for 10 minutes. Similarly, the third segment was
removed and agitated for 10 minutes. The fourth segment was not exposed to honey and
served as the control. In this way the four segments were exposed for 30, 20, 10 and 0
minutes. The effect of the honey treatments of these specimens were evaluated by scan-
ning electron microscopy.

Phase 2
In this phase three crowns were exposed to a solution of 50% honey to 50% distilled
water by volume, while three were exposed to a solution of 25% honey to 75% water.
Again the exposure periods were 0, 10, 20, and 30 minutes per segment.

Phase 3
Six crowns were also divided into segments as explained above. In this investigation
the surfaces were exposed to artificial undiluted honey (3 crowns) and to a solution con-
taining 25% artificial honey and 75% distilled water (3 crowns). In this phase the immer-
sion times were 0, 0, 30 and 30 minutes per segment. The artificial honey contained: 38%
68 S. R. Grobler and N. J. Hasson

Table 1. The minimal inhibitory concentration of honey and ofa carbohydrate control for different
oral Streptococcus species (% vol/vol)
50% 30% 25% 21% 17% 12% 6% 3%
Organism (NCTC) H C H C H C H C H C H C H C H C
Streptococcus mutans (10449) + + + + + + + + + +
Streptococcus salivarius (8618) + + + + + + + + + +
Streptococcus sanguis (7864) + + + + + + + + + +
Streptococcus anginosus (10708) - + - + + + + + + +
Streptococcus gordonii (3165) + + + + + + + + + +
Streptococcus oralis (11427) - + - + - + + + + +
Streptococcus sobrinus (10921) + + + + + + + + + +
Candida albicans (NCPF 3118) + + + + + + + + + + + + + +
Staphylococcus aureus (8530) + + + + + + + + + +
H-honey, C~ontrol, + growth, - no growth
NCPF-National Collection of Pathogenic Fungi (London)

D-fructose, 31 % D-glucose, 7% maltose, 1.5% sucrose, 100 mg Call, 308 mg PII, 0.7mg
FII and the pH adjusted to that of the natural honey sample with HCI 4,14.
All the segments in all the phases were evaluated for erosion by SEM.
The honey was tested for its antimicrobial activity against a control solution with a
sugar content similar to that of natural honey namely the abovementioned artificial honey.
Eight different oral microbial species, obtained from the National Collection of Type Cul-
tures (NCTC) (London), were used in the study (Table I). A strain of Staphylococcus
aureus was included as a reference organism. The cultures were maintained on Brain
Heart Infusion (BRI) agar (Oxoid) and subcultured weekly. The Minimal Inhibitory Con-
centration (MIC) of the honey (expressed as a vollvol percentage) were determined with
the broth dilution method described by Ericsson and Sherris 15 • In short, two-fold dilutions
of natural honey and the control solution were aseptically prepared in double strength ster-
ile BHI broth (Oxoid) to give final volumes of 5 ml in each tube. The inocula used were
prepared from overnight growth cultures in BHI broth. One drop of a 1I 100th dilution of a
culture was used to inoculate 5 ml of the test solution. The tubes were incubated at 37°C
for 24 hours and examined for growth. The MIC was noted as the lowest concentration of
honey which yielded no growth.

Fig I gives a diagrammatical outlay of the shape of the circular plastic adhesive tape
which covered the smoothed enamel surface. The circular area was divided into 4 differ-
ent segments by cutting it with a scalpel.
All the segments in phase 1 and 2 showed no sign of enamel erosion under the scan-
ning electron microscope (5l00x magnification) as a result of exposure to the different
honey concentration for 0, 10, 20 and 30 minutes. Figure 2 gives a scanning electron pho-
tomicrograph (5100x magnification) of a typical polished enamel surface which was not
exposed to any honey solution.
In phase 3, Figure 3 (5100x) shows a high degree of enamel erosion when the
enamel was exposed for 30 minutes to the 4 times diluted honey sample (25% artificial
honey and 75% water). A lower degree of enamel erosion was also observed when the
enamel was exposed for 30 minutes to undiluted artificial honey.
Human Tooth Enamel, Honey, and Bacteria 69

Figure 2. Scanning electron micrograph of the control enamel section, which was not exposed to honey. x5100

Figure 3. Scanning electron micrograph of enamel section exposed to 4 times diluted artificial honey for 30 min.
x5100 magnification.
70 S. R. Grobler and N. J. Basson

The MIC of the natural honey and of the control solution is shown in Table I. Ex-
cept for the yeast species, all the organisms tested failed to grow at honey and sugar con-
centrations higher than 21 %. The two oral strains Streptococcus anginosus and
Streptococcus oralis were more sensitive to the antimicrobial activity of honey than the S.
aureus strain and failed to grow at concentrations of 17% and 12% respectively. The oral
yeast Candida albicans was more resistant to honey than the bacteria and was able to
grow at concentrations up to 30%.


The natural honey was analysed l4 and the artificial honey prepared to contain the
same amounts of different carbohydrates (namely fructose, glucose, maltose, sucrose), cal-
cium, phoshorous and fluoride. This was done to investigate the effect of these species
(which form the building blocks of enamel) on the solubility of enamel in honey, BE-
CAUSE we soon observed that enamel does not dissolve in pure honey even when diluted
up to 4 times. However, a low degree of erosion was already observed for the undiluted
artificial honey in contrast to pure honey. while the 4x diluted artificial honey illustrated a
higher degree of enamel erosion, again in contrast to 4 x diluted pure honey (30 min.
5100x enlargement).
The fact that enamel dissolves in artificial honey and not in pure honey can be at-
tributed to factors other than the concentrations of the elements that are normally associ-
ated with the solubility of enamel (calcium, phosphorous and fluoride). In order to
establish why enamel dissolves more in diluted artificial honey than in undiluted artificial
honey we have to consider the composition of pure honey and artificial honey at different
We looked mainly at the concentrations of the elements which are also found in the
apatite crystal because these elements will have an affect on the solubility and the solubil-
ity product of enamel. What was very interesting to note is that honey also contains F (this
sample 0.7 ppm). The question arose as to whether any correlation exists between concen-
trations of fluoride, calcium and phosphorous in the honey and the water used in its proc-
essing. Bees use water extensively, in the making of honey and to cool their hives,
especially in very hot weather4 • The rapid movement of their wings produces draughts
which cause evaporation, thus cooling the hives as well as condensing the honey. How-
ever, in an extensive study by Grobler et aI 16 ., on the analysis of honey samples from vari-
ous areas and of the water collected from each source used by the bees for their water
supply, it was concluded that the elemental composition of water does not contribute sub-
stantially towards the levels of calcium, phosphorous or fluoride of honey. On the other
hand, it is possible that the relatively high concentrations of the above mentioned elements
in honey relative to that in water could mask such a possible relationship.
The pH of the different honey solutions is similar, namely 4.24 regardless of the di-
lution. Under a normal situation this pH is low enough to dissolve enamel quite rapidlyl7.
This is due to the fact that the buffering capacity of honey is quite high-4x higher than
that of saliva. This means that a small amount of saliva will not decrease the acidicity of
honey to a harmless level of about 6.8.
The joint effect can be summarised by calculating the ionic products of the honey
solutions for hydroxyapatite (HAP) and fluoroapatite (F AP) as well as the solubility prod-
ucts of hydroxy-apatite and fluoro-apatite I4 . From these values it can be seen that the ionic
product value for artificial honey with respect to HAP (63.2) is higher than the solubility
Human Tooth Enamel, Honey, and Bacteria 71

product value for HAP (54.6), which means that the 100% honey solution is under satu-
rated with respect to HAP and HAP will dissolve but not FAP. On the other hand the ionic
product of honey with respect to FAP (57.9) is lower than the solubility product (59.6)
with respect to F AP and the honey solution is supersaturated with respect to F AP and FAP
will not dissolve in the 100% honey. This explains why enamel which consists of both
HAP and FAP will dissolve only slightly in 100% artificial honey, but will dissolve more
readily in the 4x diluted honey solution as was found in this study.
It has been shown convincingly that honey contain antibacterial activity and that this
activity can be ascribed to much more than just the high sugar content of honei 8 • How-
ever, it is obvious from our results that the high sugar content can mask the real antibacte-
rial activity of honey at honey concentrations of 25% and higher. Our results also show
that except for S. anginosus and S. oratis, the oral streptococci are relatively resistant to
the true antimicrobial activity of honey.

I. Grobler S.R .• Senekal P.J.c. and Van Wyk Kotze T.1. (1989) The degree of enamel erosion by five different
kinds of fruit. c/in. Prev. Dent. 11,23-28.
2. Subrahmanyam M. (1991) Topical application of honey in treatment of burns. Br. 1. Surg. 78,497-498.
3. Anderson R.H., Buys B. and Johannsmeier M.F. (1983) Beekeeping in South AJrica. 2nd edn, Bulletin No.
394. p. 144. Department of Agriculture.
4. Root AI. (1983) ABC and XYZ oj Bee Culture. A.1. Root Pub!. Co. Medina, Ohio.
5. Grobler S.R. and Van der Horst G. (1982) Biochemical analysis of various cool drinks with regard to
enamel erosion, de- and remineralization. 1. Dent. Assoc. South Afi-ica 37, 681-684.
6. Grobler S.R., Jenkins G.N. and Kotze D. (1985) The effects of the composition and method of drinking of
soft drinks on plaque pH. & Dent. 1. 158,293-296.
7. Grabler S.R., Senekal P.J.c. and Laubser J.A. (1990) In vitro demineralization by orange juice, apple juice,
Pepsi Cola and Diet Pepsi Cola. Clin. Prevo Dent. 12. 5--8.
8. Silverstone L.M., Johnson N.W., Hardie 1.M. and Williams R.A.D. (1981) Dental caries: Aetiology, pathol-
ogy and prevention. MacMillan Press Ltd. London.
9. Cavanagh, D, Beazley, J & Ostapowicz. F (I 970) Radical operation for carcinoma of the vulva: a new ap-
proach to wound healing. Journal o{ Ohstetrics and Gynaecologv o{ the British Common Wealth, 77,
10. Dold, H, Du, DH & Dziao, ST (1937) Nachweis antibakterieller, hitzc-und lichtempfindlicher hem-
mungsstoffe (inhibine) im naturhonig (Bliitenhonig). Zeitschrifi (iir Hygiene und Infektions-Krankheiten,
11. Ibrahim A.S.( 1981). Antibacterial action of honey. Proceedings of the First International Conference on Is-
lamic Medicine. Kuwait: Minister of Health, 363- 365.
12. Jeddcr A.,Kharsany A.,Ramsaroop U.G., Bhamjee A., Haffejee I.E. and Moosa A. (1985). The antibacterial
action of honey. An in vitro study. S. Aji: Med. 1. 67. 257-258.
13. Zumla A. and Lulat A. (1989). Honey a remedy rediscovered. 1. Roy. Soc. Med. 82,384-385.
14. Grobler S.R., Du Toit 1.1. and Basson N.J. (J 994) The etlect of honey on human tooth enamel in vitro ob-
served by electron microscopy and microhardness measurements. Arch. Oral Bioi. 39. 147-153.
15. Ericsson H.M. and Sherris J.e. (1971) Antibiotic sensitivity testing. Report of an international collabora-
tive study. Acta Path Microbial Scand 79B, Suppl 217, 1-82.
16. Du Toit 1.1., Grobler S.R., Van Wyk Kotze TJ. and Basson N.J. (1995) Fluoride, calcium and phosphorus
levels in bee honey and water. S. AJr. 1. Sci. 91, 391-392.
17. Driessens F.e.M. (1982) Mineral aspects oj dentistry (Edited by Meyers H.M.) Vol. 10, p. 117. S. Karger,
18. Molan P.c. (1992) The antibacterial activity of honey. 1. The nature of the antibacterial activity. Bee Wold


1. Gedalia: S. R. Grobler, 1. Grizim D. Steinberg, L. Shapira, I. Lewinstein,

and Mo. Sela

Hebrew University-Hadassah School of Dental Medicine

Jerusalem, Israel; and
Faculty of Dentistry
University of Stellenbosch
Tygerberg, South Africa.


Honey is a sweetening agent affecting dental caries like sucrose. It contains also a
solubility-reducing agent, an organic phosphorus ester that is degradable by salivary
emzymes. In the experimental design the changes of microhardness in prepared enamel
surfaces from extracted human teeth were monitored by measurements of the tooth enamel
microhardness at baseline and after intra-oral exposure, during a certain time period, to
honey. Normal and salivary' flow deficient subjects volunteered for the study. pH measure-
ments of saliva were carried out at baseline, during and after exposure of the enamel
specimens in the mouth to the honey. The pH of the saliva (close to 7.0 at start) mixed up
with that of the honey (3.9), decreased from about 6 to 4 in the saliva-honey mixture. Af-
ter swallowing the mixture the pH returned to the baseline value. The microhardness of
the surface enamel did not change in subjects with almost complete lack of saliva flow
(dry-mouth subjects), as opposed in the subjects with a regular flow of saliva.


It was observed in rats that tooth destruction increased at a high rate with the addi-
tion of honey to their basal bread did. In vitro tests showed that pure honey with a rela-
tively low pH (3.9) does not exert an erosive effect on human tooth enamel 2 • An organic
phosphorus ester was suggested as the solubility reducing agent in honey, that is degrad-
able by salivary enzymes 3

* Correspondence to: Prof. I. Gedalia, Oral Biology-Dental Research, Hebrew University-Hadassah School of
Dental Medicine, Jerusalem, Israel.

74 I. Gedalia et al.

The effect of honey consumption on human dental enamel was investigated in dry-
mouth subjects.


Honey samples harvested in the southern part of Israel (Yad Mordechai) during
autumn, consisting of nectar from the blossoms of wild flowers, were used. Ca, P and F
were analyzed according to previous examinations in honey 4-7 .
Changes of microhardness in prepared tooth enamel surfaces were monitored by
measurements of the tooth enamel micro hardness at baseline and after intraoral exposure
to honey, during a similar time period. Normal and salivary flow deficient subjects (xeros-
tomic) volunteered for the study g.9.
pH measurements were taken of the saliva at baseline, during and after exposure of the
enamel specimens in the mouth to the honey-saliva mixture by means of indicator paper.
Significance of the differences in microhardness between the baselines of the tested
groups was determined by the Student t-test. Paired t-test was used to determine the sig-
nificance of changes in microhardness before and after honey consumption within each
subject group .


The results of the Ca, P and F composition of the honey are: 166 ppm Ca ± SD 5,
410 ppm P ± SD 15 and 0.05 ppm F ± SD 0.01.
The salivary pH levels and the microhardness levels are presented in Figs. 1 and 2.
The pH of the honey-saliva mixture decreased from about 6 to 4 in the normal and the
salivary-flow deficient subject groups (p < 0.05), returning to the baseline pH after the
mixture was swallowed (p < 0.05), Fig. 1. The initial microhardenss of the surface of the
enamel specimens decreased significantly (p < 0.000 I) after the honey consumption in the
subjects with a regular flow of saliva, whereas in the dry-mouth subjects no enamel micro-
hardness decrease took place (Fig.2).

o Normal
• Irradiated

iij 5

Figure I. Mean saliva pH values (±SE) at baseline,
in the mouth, during consumption of the honey and at
completion of swallowing the honey-saliva mixture.
Honey Contact with Teeth in Situ 7S

Z 300
= Before = After

> ,--!-
-; 250
C r--
"E 200
Qi 150
UJ 100
Figure 2. Mean microhardness (±SE) expressed in
Vickers hardness numbers before and after expo-
sure to honey in the mouth of human enamel speci-


A pH below 5.5 at the tooth enamel surface leads to mineral loss 10. 11 • The results of
the present study indicate that tooth enamel decalcification occurred during consumption
of the acidic honey in subjects with normal saliva secretion (Fig. 2). It is known that even
trace amounts, 0.01 ppm of F in solution, decrease the rate of enamel solubility 12 • The F
concentration was very low in the honey used in our study , 0.05 ppm, which probably ex-
plains why it did not suppress bacterial fermentation or exert a remineralizing effect (re-
turn oflost Ca and P to the tooth enamel surface). No erosive defects on in vitro exposed
enamel surface to pure honey at a pH 4.24 was observed 2• The difference between the in
vitro2 and the present in situ results regarding erosive effects on tooth enamel from honey,
may be in part explained by the suggested solubility-reducing agent in honey, the phos-
phate ester, that is degradable by salivary enzymes 3 . In the dry-mouth subjects, the ex-
pected solubility-reducing agent in the honey was probably active protecting the tooth
enamel when the honey-saliva mixture was below the pH level of 5.5 10• 11 , Fig. I.


The authors are indebted to Prof. I Roman, Dept. of Applied Science and Applied
Physics, Hebrew University, Jerusalem, for his first-rate cooperation in this study.
Thanks to Mrs. Shula Konig for her secretarial assistance.

I. Koenig K.G. ( 1967) Caries Induced in Laboratory Rats. Br. Dent. J. 123,585- 589.
2. Grobler R.S .. Du Toit I.J., Basson N.J. (1994) The Effect of Honey on Human Tooth Enamel In Vitro Ob-
served by Electron Microscopy and Microhardness Measurements. Arch. Oral BioI. 39. 147- 153.
3. Edgar W.M .• Jenkins G.N. (1974) Solubility-Reducing Agents in Honey and Partly-Refined Crystalline
Sugar. Br. Dent. J. 136. 7- 14.
4. Chakrabarti C .. L. ( 198 1) Progress in Analytical Atomic Spectroscopy. 2, 207.
5. Havezov I., Russeva E., Jordanov N . (1979) Fl ameless Atomic Absorption Determination of Phosphorus
Using ZrC Coated Graphite Atomizer Tubes. Fresenius Z. Anal. Chern. 296, 125- 127.
6. Nicholson K. , DuffE.J. (1981) Fluoride Determination in Water. Anal. Lett. 14, 493- 517.
76 I. Gedalia et al.

7. du Toit I., Grobler S.S., v Kotze W.TJ., Basson, NJ. (1996) Fluoride, Calcium and Phosphorus Levels in
Bee Honey and Water. S.Afr. J. of Science 91,391-392.
8. Gedalia I., Davidov I., Lewinstein I., Shapira L. (1992) Effect of Hard Cheese Exposure, With and Without
Fluoride Prerinse, on the Rehardening of Softened Human Enamel. Caries Res. 26,290-292.
9. Sela Mo., Gedalia I., Shah L., Skobe Z., Kashket S., Lewinstein I. (1994) Enamel Rehardening with
Cheese in Irradiated Patients. Amer. 1. Dent. 7, 134-136.
10. Grobler R.S., Jenkins G.N., Kotze D. (1985) The Effect of the Composition and Method of Drinking of
Soft Drinks on Plaque pH. Br. Dent. J. 158,293--296.
II. Gedalia I., Dakuar A., Shapira L., Lewinstein I., Goultschin J., Rahamim E. (1991) Enamel Softening with
Coca-Cola & Rehardening with Milk or Saliva. Am. J. Dent. 4, 120-122.
12. White OJ. (1987) Reactivity of Fluoride Dentifrices with Artificial Caries. I. Effects of Early Lesions: F
Uptake, Surface Hardening and Remineralization. Caries Res. 21, 126--140.



Zohara Yaniv and Michal Rudich

The Volcani Center
Bet Dagan, Israel

The use of plants for therapeutic purposes dates back to the beginning of time. An-
cient civilizations, such as Egypt, China, India and the Incas of the Americas knew well
the medicinal properties of herbs and used them in the prevention and treatment of dis-
eases. Through the writings of great physicians, such as Hippocrates and Dioscorides,
much of this ancient art had survived to our times. Nowadays we are witnessing the re-
vival of the herbal tradition. There is a great interest in using this tradition in order to de-
velop and define new, natural drugs for human use.
Honey, too, has its important place in human nutrition since prehistoric time. It was
considered not only as a sweetener but also as a high-quality food, with curative proper-
ties. The appreciation of honey is expressed in old manuscripts and documents, legends
and mythologies.
Honeybees feed on plants nectar, and honey is produced. Its qualities reflect the
source of nectar which was used,so that the properties of the nectar and its chemical con-
tent are of great importance. In fact, the quality of the honey is directly related to the
source of nectar. Therefore, there is a very great potential for using medicinal plants and
nectars with very specific chemical compositions to produce honeys in the future with
curative properties above and beyond those known hitherto.
The taste, aroma and color of the honey is directly affected by the source of the
plant nectar. (Table I). Examples include eucalyptus honey, which is dark, aromatic and
spicy; thyme honey, light and scented; carrot honey, reddish and spicy, and so on. These
examples indicate a direct transfer of components from the nectar to the honey.
The use of honeys produced from diverse sources of nectar for specific therapeutical
purposes is widely practised in France. Table 2 illustrates some examples of honeys, their
source of nectar and their special medicinal indications. It is clear that the honeys and
their source of nectar are used for identical medicinal treatments. Examples include laven-
der honey for diseases of the respiratory system, tilia honey, for spasms, insomnia and as a
tranquilizer; acacia honey as an intestinal regulator, etc. (Table 2). It should be noted that
this practice falls within the framework of folk medicine.

78 Z. Yaniv and M. Rudich

Table 1. Colors and aromas of honeys

Source of nectar Color Aroma and taste

Eucalyptus dark spicy
Thistle dark green tone spicy
Carrot reddish spicy
Onion brown onion flavor
Red Clover bright yellow delicate flavor
Cotton light non·scented
Sage light delicate flavor
Rosmary light delicate flavor
Thyme light delicate scent
Savory light delicate flavor

Scientific evidence for a transfer of active metabolites from the nectar to the honey
produced is provided by studies performed in several laboratories.
Some important topics are summarized as follows:



The antibacterial properties of honey have been known for a long time (Molan et ai,
1988). Scientists in various parts of the world have noticed that the intensity of these an-
tibacterial properties depends on the plant source of the honey. In Brazil, Cortopassi-Laur-
ino and Gelli showed that in Apis-bee honeys, the strongest antibacterial properties were
found in honeys produced from mimosa and eucalyptus pollen. Mimosa was found to be
also the best source of antibacterial honey produced by Melipona subnitida and Plebeiasp
bees. (Cortopassi-Laurino and Gelli, 1991). Indeed, the use of mimosa and eucalyptus is
known in phytotherapy and aromotherapy as antiseptic and antiflammatory agents.
A study was performed in Poland recently on the antibacterial properties of honeys
againsts everal strains of bacteria. In this study, too, it was found that various types of

Table 2. Honeys: sources and curative properties (folk medicine in France)

Source of nectar Curative properties Particular indications

Acacia Intestinal regulator. Good for intestinal stasis in infants.
Erica Antiseptic (urinary tract), Diuretic. Urinary tract infections and kidney
Chestnut, forest flowers, sunflower Stimulates blood circulation. Improves blood circulation in
general and varicose veins in
Lavender Antiseptic, Anti-inflammatory Diseases of the respiratory
(respiratory system), system. Diseases of the arteries.
Oak sap, fir tree Anti-anemic,Antiseptic Certain types of anemia
Anti-inflammatory (respiratory
Tilia Anti spasmodic, Tranquilizer Spasms of diverse origin,
restlessness, insomnia and
Medicinal Herbs as a Potential Source of High-Quality Honeys 79

honey differ substantially in their antibacterial activity, depending on the plant source.
The most active ones were honeydew and lime honey (Leszczynska Fik and Fik, 1993).
This difference could be due to the compositions or amounts of essential oils and
other components which were transferred from the plant to the honey.
The direct effect of feeding medicinal plant extracts to honeybee colonies, on the an-
timicrobial activity of the honey produced, was studied in Egypt (Mishref et ai, 1989). Ex-
tracts from geranium, chamomile and majoram were fed to honeybee colonies once during
an 8-week period. At the end of this period, honey in the combs was extracted and its an-
timicrobial activity was determined. Results showed that the honeys from colonies fed
with medicinal plant extracts showed greater antibacterial activities than honey from con-
trol colonies, in the order chamomile> geranium> majoram.
These findings correlate with the actual use of geranium, majoram and chamomile,
as antiseptic and antimicrobial agents in folk medicine. The essential oils of geranium and
majoram are widely used as antiseptic components of aromatherapic mixtures.



The following examples demonstrate the possibility that diverse secondary metabo-
lites, besides essential oils, can be transferred by the honeybee, from the nectar to the
honey, thus creating a honey with a specific chemical composition:
1. Sunflower honey was shown to be a very rich source of flavonoids. (Sabatier et
ai, 1992; Sabatier et ai, 1988).The flavonoids were identified and their struc-
tures could provide an index of floral origin.
2. Analysis of many experimental and commercial citrus honey samples revealed
the presence of the flavanone hesperetin in all samples. This flavanone was not
detected in any non-citrus honey samples. The analysis of the flavonoids present
in orange nectar revealed that the flavanone hesperidin was the major flavonoid
detected. Hesperetin is probably produced by hydrolysis of hesperidin by the
bee enzymes present in honey (Ferreres et ai, 1993).
3. Of a similar nature is the study of Czeczuga (Czeczuga, 1985), showing the
presence of identical carotenoids in working bees and in the flowers being vis-
ited by the bees. The author suggested that some of the flower carotenoids are
converted to more highly oxygenated forms in the honeybee. The total content
and composition of carotenoids in foraging bees was related to the flower spe-
cies visited.
4. The presence of the glucoside, arbutin in both bitter honey and in the nectar of
the strawberry tree, Arbustus unedo, confirmed this tree to be the main source of
the honey. (Floris and Prota, 1989). In Sardinia, this bitter honey is considered
as having therapeutic value.


We should be aware of the possibility of the presence of plant toxins in honey. Nec-
tars from plants such as azalea, andromeda and rhododendrom (Ericaceae) are known to
80 Z. Yaniv and M. Rudich

contain gayanotoxins. Cases of intoxication, following the ingestion of such honey have
been reported. Symptoms are similar to aconitine intoxication: progressive paralysis from
the extremities to the diaphragm. (Alcaraz and Rios, 1991).
The more recent report of the presence ofpyrrolizidine alkaloids in honey, may indi-
cate a new health hazard (Deinzer et ai, 1977): these authors found that the hepatotoxic al-
kaloids known to occur in tansy ragwort (Senecio jacobaea L.) are also present in honey
produced from the nectar of this species. These alkaloids (six were identified), are poten-
tially carcinogenic, mutagenic and teratogenic, and may pose health hazards to the human
consumer. However, due to the fact that ragwort honey is very bitter in taste and off-color,
it is not likely that the individual consumer would consume enough honey to suffer acute



Since the source of plant nectar can affect the composition and properties of the
honey, future production of therapeutical honeys can be envisaged. Nectar-rich medicinal
plants with desired therapeutical properties could be used as the source of nectar.
Table 3 contains a list of some medicinal plants which could be used as a source of
active compounds. Some of these plants have nectar-rich flowers, visited by the honeybee.
Examples include aromatic plants from the Labiatae family, known for their high content
of essential oils. Salvia ojJicinalis and S. Fruticosa; Coridothymus capitatus and Ma-
jorana syriaca. All three species are known for the treatment of colds, indigestion and ex-
ternal inflammations.
Crataegus oxycantha is prescribed as a cardiac depressant and as a hypotensive. Re-
lama raelam, a desert plant, is used for rheumatic pain and external wounds and has a po-
tential application in chemotherapy against cancer. Echinacea angustifolia stimulates and
strengthens the immune system and Cassia senna is a very popular natural laxative.
Plants could also be used by feeding the bees with a sweetened extract of plant parts,
such as leaves. roots, stems, on flowers. These plant parts should be selected on the basis
of a high content of active compounds. The sweet extracts would then serve as a rich
source for the production of medicinal honey.
We hope that in the future the list of curative and medicinal honeys will be unlim-
ited. This direction will offer a new avenue into natural medicine.

Table 3. Potential medicinal qualities of honeys

Plant origin Medicinal potential

Salvia officinalis (Sage) Stimulates bile secretion. wound and acne treatment, regulates
Coridothymus capitatus (thyme) Antiseptic, expectorant and anti-spasmodic
Majorana syriaca (Majoram) Expectorant, toothache and gum infection, indigestion, heart problems.
Crataegus oxycantha (Hawthorn) Vasodilator, regulates blood pressure.
Retama raetam (White Broom) Relieves abdominal and rheumatic pain, and wound remedial.
Anti-carcinogenic activity.
Echinacea angustifolia Anti-inflammatory, strengthens and stimulates the immune system.
Cassia senna Laxative.
Medicinal Herbs as a Potential Source of High-Quality Honeys 81

Alcaraz, M.J. and Rios, J.L (1991). Pharmacology ofDiterpenoids. in: Ecological Chemistry and Biochemistry of
Plant Terpenoids. (Harborne, 1.8. and Tomas-Barberan, F.A. eds.) p 230-263. Clarendon Press Oxford.
Cortopassi-Laurino, M. and Gelli, D.S. (1991). Pollen analysis, physico-chemical properties and antibacterial ac-
tion of Brazilian honeys from Africanized honeybees (Apis mellifera ) and stingless bees. Apidologie. 22.
Czeczuga, B. 1985. Investigations on carotenoids in insects. VII. Contents of carotenoids in worker bees feeding
on flowers of different plants. Zoologica Poloniae 32. 183--190.
Deinzer, M.L. Thomson, P.A Burgett, D.M. and Isaacson, D.L. 1977. Pyrro1izidine alkaloids: their occurrence in
honey from tansy ragwort (Seneciojacobaea L.) Science, 195497-499.
Ferreres, F. Garcia-Viguera, C. Tomas-Lorents, F. and Tomas-Barberan, F.A. 1993. Hesperetin: a marker of the
floral origin of citrus honey. J. Sci. Food. Agric. Sussex. 61. 121-123.
Floris I. and"Prota, R. 1989. The bitter honey of Sardinia. Apicoitore-Moderno. 80. 55--{)7.
Leszczynska Fik, A. and Fik, M. (1993). Antibacterial properties of various types of honey and the effect of honey
heating on antibacterial activity. Medycyna- Weterynaryjna. 49.415-419.
Mishref, A. Magda, S.A. and Ghazi, I.M. 1989. The effect of feeding medicinal plant extracts to honeybee colo-
nies on the antibacterial activity of the honey produced. Proceedings o[the Fourth International Confer-
ence on Apiculture in Tropical Climates cairo, Egypt. 80-87.
Molan, P.C Smith, I.M. and Reid, G.M. 1988 . A comparison of the antibacterial activities of some New Zealand
honeys. J. of Apicultural Research. 27. 252-256.
Sabatier, S. Amiot, M.J.Aubert, S.Tacchini, M.and Gonnet, M. 1988. Importance of flavonoids in sunflower hon-
eys. BuUetin-Technique-Apicole. 15. 171-178.
Sabatier, S.Amiot. M.J. Tacchini, M. and Aubert, S. 1992. Identification of flavonoids in sunflower honey. J.
Food. Sci. off. Publ. Ins!. Food. Techno!. Chicago. ILL. The Institute. 57. 773--774.



Tsila Dvir

Human Nutrition Unit

Ministry of Agriculture
12, Aranya St., Tel Aviv, Israel


Honey has a wide range of characteristics which differentiate it from sugar and other
sweeteners. Some of the unique features of honey are the following:
• Honey is a natural ingredient. Nowadays, when natural foods are so valued, this is
very significant.
• As opposed to the common belief, honey is not particularly calorie rich.
• Honey contains many important nutritional values, like minerals and vitamins.
• Honey has an important role in natural health care. It has potential in prevention
and treatment of certain diseases.
• Honey can serve many functions in cooking like browning (especially with mi-
crowave). It is an excellent bonding material and an ideal base for sauces.
A strategy aimed at increasing the consumption of honey by the private consumers
must emphasis these qualities and highlight honey's advantages over "competing" ingredi-
ents. Furthermore, these features should be attractive to food manufacturers because of
their appeal to the consumer..
Such strategy can include the following components:
• Bold and detailed labeling of products that contain honey. The label should list
the nutritional values of the honey.
• Presentation of honey and honey products in separate sections and on many
shelves that are devoted to specific categories, like health food, honey products,
spreads, sauces, cake decorations etc.
• Marketing honey in new forms which are more convenient to use, like spread in-
stead of liquid honey.
• Unique and unusual containers that emphasis the natural aspects of honey.

84 T. Dvir


In ancient times honey was the common and unquestionable sweetener, and also a
symbol to good life and wealth.
All this was completely changed with the industrial revolution, the massive growth
of sugar cane in the Americas and the knowledge to cheaply extract sugar from cane and
sugar beets.
This was an unavoidable development; there was no way by which honey could pos-
sibly supply the rapidly growing population which coincided with the rise of the standard
of living; among other things this meant an enormous demand for sweeteners.
The unfortunate thing that happened alongside this development was that gradually
the wonderful food called bee-honey lost its glory: people regarded it as not much differ-
ent from jam or cane sugar.
In the last few decades the honey suffered two additional blows.
First, the western world has become more calorie aware; and with the massive cam-
paign against overweight all the natural sweeteners were among the first to be blamed as
"the enemy", "the danger". Needless to say, the consumption of honey has suffered
As a "natural" result, people started to shift to artificial sweeteners.
This forum is not the place to analyze the consequences of this shift to artificial
sweeteners, or to put it more boldly-the health problems which directly result from the
exaggerated use of these non-natural substitutes. However, I feel that this matter is of
critical importance to the honey future and therefore I will take the liberty to give at least
a few clues:
• Aspartame sugar substitutes cause symptoms from memory loss to brain tumors.
It is sought to be one of the most dangerous substance [1].
• Saccharin, found in many "diet" drinks is considered a possible cancer hazard [2].
• The real amount of sugar substitute consumed in out diet exceeds by far the daily
amounts approved as "safe" by USA FDA [1].
• It appears that many of the health problems that the US soldiers contracted in the
Golf war are a result of the huge quantities of diet drink that they have consumed
As if the damage caused by artificial sweeteners substitutes was not enough, another
problem evolved. the public, and many professionals, got the idea the actually honey is al-
most the same as sugar. Consequently, housewives, cooks, institute mangers and industri-
alists said: if so, why bother with honey since sugar is cheaper and readily available?
I am not sure how much of this basic misconception was due of somebody claiming
that "honey equals sugar" or simply that there was nobody to clearly emphasize the differ-
ences between them. Whatever the reason-the damage is very real.


To show the full scope of the difference we should compare honey with sugar, item
by item.
To sum it up: honey is very different-and clearly superior-to sugar, when com-
pared by almost all criteria.
Honey and Food Processing 85

Table 1. The differences between honey and sugar [4,5]

Property Honey Refined Sugar
Source A natural, unprocessed product Industrially processed
Health Qualities Mild antibiotic None
Curing sore throat
Avoiding constipation
Curing diarrhea in babies
Nutritional Properties: Water Content 20% 1%
Calories 304 cal per 100 gr. 375 cal per 100 gr
Sugar Kinds 13 kinds of sugars (glucose 33%, fructose only sucrose
40%, other II kinds 5%)
Vitamins (mg per 100 gr.) None
Thaine B1: 0.004-0.006
B6 Pyridoxine 0.008--0.032
Niacin 0.1l..{J.36
VitaminC 2.2-2.4
Minerals mg per 100 gr. None
Calcium 0.4-3
Iron 0.1-3.4
Potassium 1.0-47.0
Also: Iodine, Copper,
Magnesium, and traces of other
Gourmet Qualities Rich taste None
Rich aroma
Rich color
Smooth texture
Syrupy quality
Used "as is"
Cooking and Baking Qualities Adds color None
Preserves freshness & moisture
Adds market value
Special Qualities Symbolized tradition and "roots" None
Relates to the "back to nature" trends

These two misconceptions--that honey, like sugar, is Calorie rich, and that sugar is
an equivalent substitute to honey, have discouraged many potential honey consumers. The
potential lose is estimated by perhaps hundreds of millions of consumers that would have
used considerable amounts of honey in there daily diet.
So we must find ways to reverse the trend-that is, to re-introduce wide sectors of
the public and food industry with honey and its properties.
Most of the data listed above is probably known to you. Furthermore, I will tell you
a secret. Just a few days ago I went over note of a lecture I have delivered some 8 years
ago. Surprise--it was almost all there.
Now, the fact that we all knew the benefits of honey and the misconceptions about it
is not the important issue.
The real problem is that the users--I mean the public and the food industry as well
as the professionals-are either ignorant of the facts or, as in the industry case, simply
prefer to ignore them because of economical and other practical reasons.
86 T.Dvir


The honey industry has to form a strategy to face this problem. I would like to men-
tion some of my thoughts on the subject.

4.1. General Finding

Let me start with some general findings relating to the issue:
• A "Back to Nature" fad is growing in Israel
• The amount of health groups has doubled since 1987
• Confusion exists as to the difference between health foods and dietetic/low Calo-
rie foods.
• Women pay attention more than men to the food they eat.
• Men think that health food/dietetic food is for sick people

4.2. Honey Drawbacks

I then ask myself why honey is consumed less than other sweet spreads:
• Availability-not always available
• Price--more expensive
• Variety-less then other
• Authenticity-how to tell real honey apart from imitations?

4.3. Honey vs. Sugar

We should start a campaign to show that honey is not sugar-All facts listed in Ta-
ble 1.
We should emphasize nutritional value of honey-Again, see table 1.

4.4. Honey Is Much Safer

We must convey the fact that honey is much safer then the artificial sweeteners. To
do so we may have to point out what is reported about the dangers:

Aspartame sugar substitutes cause worrying symptoms from memory loss to brain tumours.
But despite USA FDA approval as a "safe" food additive, aspartame is one of the most dan-
gerous substance ever to be foisted upon an unsuspecting public (Nexus Magazine, ref. 1).

4.5. Honey Uses

Then we should talk about honey uses
• as a natural alternative to refined sugar
• barbecue sauce and dips
• in hot beverages, such as tea
• in cold beverages, such as ice tea, ice coffee, ice water+lemon and honey
• in manufactured foods: breads, cereals, sauces, bottled and canned beverages,
fruit and carbonated drinks, milk products and candies
Honey and Food Processing 87

4.6. Honey in Cooking and Baking

Honey can and should be inserted into many home cooking and baking recipes.
Honey can be used as food softener, for browning (good idea for micro-oven users),
as a binding element, and as sweetener. It would serve well in many recipes, like:
• salads & vegetables
• spreads & butter
• breads, muffins and rolls
• marmalades and sauces
• cakes (traditional cakes, cheese cakes, oriental cakes, fruit cakes)
• cookies
• deserts and pies
• candies & snacks
• main entree

4.7. Honey Promotion

And finally, promotion of honey consumption in all sectors is important.

For example, honey should be placed in different sectors throughout supermarkets.
Such sectors are spreads, cake decorating, sauces & salad dressings, Chinese food, natural
and health foods [6,7].
Marketing honey in new forms which are more convenient to use, like spread in-
stead ofliquid honey.
Unique and unusual containers that emphasis the natural aspects of honey.


Ifwe take these actions, I believe that 4 years from today, in 2000 Honey congress,
the chairman or chairwoman will probably rise and point to a graph titled "Honey con-
sumption 199Cr2000". It will show a dramatic increase.

I. Gold M. D. (95), The Bitter Truth about Artificial Sweeteners, Nexus magazine, Vol. 2 # 28
2. Ames and Gold (I 987),Ranking Possibly Carcinogenic Hazards, Science pp. 236-271
3. Woodrow C M, Aspartame, Methanol & Public Health in: Journal of Applied Nutrition, 36 (I), pp. 42-53
4. Crane E., Honey, Heinemann London 1976, p. 264
5. Gebhards S. and Mathews R. Nutritive Values Foods, USA department of Agriculture, 1991
6. Hayes G. W. (1985), Lets Promote Honey, American Bee Journal, Feb 1985
7. Thomas and Payne (1988), A Leak at Japanese Market, American Bee Journal, June 1988



ChangY Lee

Department of Food Science and Technology

Cornell University
Geneva, New York 14456

Today's consumers demand foods more natural with less food additives. We found
honey to have very useful characteristics in food processing such as a fruit juice clarifying
agent, an anti-browning agent among others. Honey can replace some of the conventional
food additives in fruit juice or wine production and it can also be processed into a wide
range of new beverage products. The following describes some of our research on honey
that has been carried out for over ten years in our laboratory.
Honey consists primarily of carbohydrates with an average concentration of about
80%. Water is the second major component at around 17%. The other important compo-
nent, although its concentration is relatively low, is protein. The protein content was re-
ported to have a wide range of 58-786 mgllOOg of honey with a mean value of 169
mgll OOg (1). In spite of low concentrations, honey protein is an important constituent be-
cause it influences many properties of honey and honey products (2).
One of the unique properties of honey protein is that it readily interacts with pheno-
lic polymers, including tannins, and forms macromolecule complexes (3). It is known that
the reaction between proteins and tannins produces hazes in most natural fruit juices, and
that sediments of fruit juices consist largely of phenolic materials mixed with protein (4).
When a protein such as gelatin is added to a hazy juice, it entraps the particles and coagu-
lates them by hydrogen bonds between the hydroxy groups of polyphenolic tannins and
the carbonyl groups of the proteins (5,6). In a model solution of honey protein with tannic
acid, it was found that the maximum rate of interaction occurred at the ratio of 1:2-3, tan-
nic acid:honey protein (3). We employed this characteristic of honey protein in apple juice
processing and found that the honey protein acts as gelatin-like in that it entraps the hazy
particles and clarifies the juice. The optimum clarification conditions were found to be
4-5% honey concentration by weight at pH 3--4 at room temperature. Clarification oc-

90 ChangY. Lee

curred over a wide temperature range with the rate increasing rapidly as the temperature
was raised (7). This process has been used commercially for several years.
In order to find the derivation of this unique honey protein, honey samples from
various floral sources, including citrus, dandelion, locust, basswood, clover, goldenrod,
and honey from sucrose-fed bees were collected from various regions and compared for
their juice clarification characteristics and electrophoretic patterns. There are reports that
the honey protein could be originated in either the plant nectarines or the bees and that
honey protein consisted of 4-7 components (8,9). We found that all honey samples we
studied, regardless of their origin, contained the specific protein fraction that is responsi-
ble for clarifying apple juice and that is originated from honey bees (Apis melli/era) (10).
In order to compare the protein component among honeys from different bee species.
honey from A. melli/era bees from Cornell University, honey from A. laboriosa bees from
Chhomrong, Nepal at the altitude of about 2,000 m and honey from A. cerana, common
Indian honey bees from Kapre Chhap, Nepal at about 1,200 m from sea level were ana-
lyzed for their protein component. The electrophoretic pattern showed that protein frac-
tions from the three diverse bee species were different and only Apis meli!fera and A.
cerana bees produced the protein fraction with the apple juice clarifying activity, but A.
laboriosa bees did not (11).
Honey also has an inhibitory activity on polyphenol oxidase which is the major
cause of enzymatic browning in fruit and vegetable products. In model solutions of
polyphenols and polyphenol oxidase, added honey prevented browning reactions (12). We
found this inhibitory effect of honey on Drowning in preparation of several fruit juices and
wines and dehydrated fruit products. Grape juice prepared with added honey exhibited a
similar color to commercial juice that was treated with ascorbic acid and enzyme. Table 1
shows effect of honey added to Niagara grape juice on color. Juice prepared with no treat-
ment (control) was dark-brown with a higher absorbency at 420 nm and lower Hunter "L"
value, but juices treated with honey showed no brown color with a low absorbency and
higher Hunter "L" value.
As early as in 1935, honey was added to fruit juices to make fermented products as a
supplementary sweetening agent (13). We added honey to apple cider to bring soluble sol-
ids to 20° Brix and fermentation was carried out in the normal procedure at room tempera-
ture, 18-22°C for 30 days. The final products were analyzed for sensory quality and we
found that apple wine treated with mild honey (orange blossom or clover-locust) was
much better than apple wine made with added sugar (14). We also used honey in white
grape wine production with no added sulfur dioxide. Several white grape cultivars were
made into wines by conventional and new honey-treated (ameliorated with 22% honey so-
lution) methods. We observed that wines made by the honey treatment with no S02 were
very close to the conventionally prepared wines (commercial) in overall sensory quality. It
appeared that honey acted as an anti-browning agent similar to S02.

Table 1. Application of honey in Niagara grape juice production

Juice preparation pH (420 nm) Hunter"L"
Control (no treatment) 3.78 0.71 47.30
Commercial" 3.11 0.07 58.41
Honey treated# 3.11 0.07 58.36
*Treated with ascorbic acid and enzyme.
#Added honey (3% by weight).
Honey as a Clarifying and Anti-browning Agent 91

In order to apply honey as an anti-browning agent in minimal processing of fruits,

apple slices were treated by dipping them into a 3-5 % honey solution, packaged under
the modified atmosphere conditions and then stored for 2-3 week at 3-5°C. We found that
apple slices treated with honey were lighter in color and firmer in texture compared to
those of apple slices prepared with commercial anti-browning acidulants. We also applied
honey to various fruit slices for dehydrated products. Sensory quality of the dehydrated
fruits (apple and pear slices, raisin) treated with honey was much better than those pre-
pared by a conventional method with added S02'
Although mead (honey wine) has been produced for many years, it has never been
as popular as grape wines in the U. S. One of several reasons is that conventional mead
lacks the overall quality of a typical alcoholic beverage. Due to the hazing problem in
mead caused by honey protein, traditional mead-making requires boiling a honey solution
for 30-{:)0 minutes before fermentation. This heat process is so excessive that it develops
undesirable flavors, such as harshness and bitterness. To mask these undesirable flavors, a
high residual sugar content and a long aging process are required. In order to overcome
these problems and improve the mead quality, honey was passed through an ultra-filtra-
tion (UF) membrane with various molecular weight cut-offs (l0-50K) to remove haze
forming proteins and then fermented in the usual manner. A complete fermentation and
bottling process were accomplished within 3 weeks so there was no need for a long clarifi-
cation and stabilization period. Sensory quality and stability of the UF-treated mead are
superior to mead made by the conventional method. Several commercial firms are produc-
ing products by using this new method. This UF -treatment of honey opened the door for a
wide range of honey application in food processing and there are already many new prod-
ucts being produced commercially using UF-treated honey.

1. White, J. w.; Rudyj, O. N. (1978) The protein content of honey. J. Apic. Res. 17.234-238.
2. Paine, H. S.; Gertler, S. I.; Lothrop. R. E. (1934) Colloidal constituents of honey. Ind. Engng. Chem. Ana-
lyt. Edn. 26, 73-81.
3. Lee, C. Y. (1984) Interaction of honey protein and tannic acid. J. Apic. Res. 23, 106-109.
4. Johnson, G.; Donelly, B. J.; Johnson, D. K. (1968) The chemical nature and precursors of clarified apple
juice sediment. 1. Food Sci. 33, 254-257.
5. Calderon, P.; Van Buren, J.; Robinson, W. B. (1968) Factors influencing the formation of precipitates and
hazes by gelatine and condensed and hydrolyzable tannins. J. Agric. Food Chem. 16,479-482.
6. Gustavson, K. H. (1954) Interaction of vegetable tannins with polyamides as proof of the dominant func-
tion of the peptide bond of collagen for its binding of tannins. J. Polymer Sci. 12,317-324.
7. Lee, C. Y.; Kime, R. W. (1984) The use of honey for clarifying apple juice. J. Apic. Res. 23, 45-49.
8. White, J. W.; Kushnir, I. (1967) Composition of honey. VII. Proteins. 1. Apic. Res. 6, 163-178.
9. Bergner, K. G.; Diemair, S. (1975) Protein in honey. II. Gel-chromatography, enzymatic activity and origin
of honey proteins. Z. Lebensm. Forsch. 157.7-13.
10. Lee, C. Y.;Smith, N. L.; Kime, R. w.; Morse, R. A. (1985) Source ofthe honey protein responsible for ap-
ple juice clarification. J. Apic. Res. 24, 190-194.
11. Lee, C. Y.; Smith, N. L.; Underwood, B. A.; Morse, R. A. (1990) Honey protein from different bee species
in relation to apple juice clarification activity. Amer. Bee 1. 130,478-479.
12. Oszmianski, J.; Lee, C. Y. (1990) Inhibition of polyphenol oxidase activity and browning by honey. J.
Agric. Food Chem. 38, 1892-1895.
13. Fabian, F. W. (1935) The use of honey in making fermented drinks. The Fruit Prod. J.I4, 363-365.
14. Kime, R. W.; Lee, C. Y. (1987) The use of honey in apple wine making. Amer. Bee J. 127,270-271.

Composition, Properties, and Applications

M. G. Campos,] A. Cunha,l and K. R. Markham 2

ILaborat6rio de Farmacognosia da Faculdade de Farmacia

Universidade de Coimbra
3000 Coimbra, Portugal
2New Zealand Institute for Industrial Research and Development
POBox 31310, Lower Hutt, New Zealand.


During ancient times, people throughout the world commonly used pollen, praising
it for its goodness and medicinal properties. Some of the reasons the ancients used bee-
pollen are why wc use it today. To date no scientific evidence has been cited to disprove
the claimed properties of bee-pollen.
One claim, attributes to bee-pollen the ability to reduce the rate of ageing. It is said
that this product has a special factor that can improve health, and increase vital energy.
The basis for this claim is the fact that the queen bee, and only the queen bee, can live five
or six ycars on a diet of royal jelly, that the bees do with bee-pollen. Other bees only eat
this diet for the first two days of their life and after that, they eat honey. They live for only
a few weeks.
An undeniable fact is that, this product can't be synthesised in the laboratory, can't
be easy adulterated and has been taken by people over thousands of years, without ma-
nipulation or recorded side effects. It is fortunate that a number of scientists, world wide
are earring out research in many different areas, out of interest, to test and prove the nutri-
tive and curative properties claimed for the bee-pollen that we "steal" from the hive.


Concerning the value of bee pollen as a natural health food for people, it is well
known that bees collect pollen because of its high content of protein (average 35%). Ap-
proximately half of the "protein" fraction is in the form of free amino acids which can be
assimilated immediately by the body.

94 M. G. Campos et al.

It has been established by Peris (1984)1, that in a daily dose of 15 g (about a soup
spoon) of bee-pollen provides the minimum amount of amino acids that the human body
needs. Further, at least 40% of the content is made up of various forms of sugar. Huidobro
et al. (1986,1987)2,3 reported a value of 61%, but they included in this fraction not only
carbohydrates like reducing sugars (fructose, glucose), maltose, sucrose and polysaccha-
rides, but also starch and other polysaccharides that can't be absorbed such as cellulose,
hemicellulose and lignin, esporopolenin, etc" We have also verified that about 50% (Cam-
pos et ai, 1994)4 of the weight of the bee-pollen corresponds to a material that can't be ex-
tracted with organic or aqueous solvents.
Lipids and minerals (carriers of calcium, phosphorus, magnesium, iron, copper and
manganese, etc.) represent 5 and 3% respectively. Most (60%) of the fatty acids aTe in the
free form. Bound fatty acids, which reflect the compositional profile of pollen, were char-
acterised by a high level of a-linolenic acid (70%) and by small amounts of linoleic and
oleic acid. Palmitic acid is the most abundant saturated fatty acid (Seppanen et al.,l989f
Vitamins in bee-pollen include not only vitamin B complex and ascorbic acid, but
also vitamins A, D and E. The levels vary between the pollen species and with season. For
example, Herbert et aI., (1987)6 established that the seasonal thiamine (vitamin B 1) levels
in bee-collected pollen varied greatly depending on the floral source and the time of year.
It is important not to forget that in biological samples, vitamin B 1 activity is due not only
to thiamine but also to the mono, di and triphosphate derivatives of thiamine. Thiamine
and its mono and diphosphate forms are normally the most prevalent. Pollen apparently
supplies the honey bees requirement for this vitamin since there is no real evidence that it
can be synthesised by insects. Thiamine is relatively stable in acid pH, and pollen pro-
vides an acid environment. In stored pollen the vitamin has been shown to exist for up to 4
years (Hagedorn and Burger, 1968l
Xie et al. (1994), in China, have studied the effect of bee pollen on maternal nutri-
tion and foetal growth. They state that plant pollen collected by the honeybee is a natural
nutrient. They studied the effects of the bee pollen from Brassica campestres L. on mater-
nal nutrition and foetal growth using pregnant Sprague-Dawley rats. Pollen-fed dams had
greater body weight and higher levels of haemoglobin, total protein, serum iron and albu-
min while the foetuses of pollen-fed dams had greater body weight and lower death rate.
No gross external, visceral or skeletal malformation was observed in the foetuses. These
results were interpreted as indicating that bee-pollen could improve maternal nutrition
without affecting normal foetal development. Bee pollen was therefore considered a prac-
tical and effective nutrient during pregnancy.s


Gillian et al., 1989 isolated one-hundred and forty-eight moulds from the following
samples of almond, Prunus du/cis. pollen: floral pollen collected by hand; corbicular pol-
len from pollen traps placed in colonies of honeybees, Apis mellifera, in the almond or-
chard; and bee bread stored in comb cells for one, three and six weeks. They verified that
corbicular pollen (pelletized bee pollen) contained 99,8% almond pollen when collected
from pollen traps placed in bee colonies in an almond orchard. If micro-organisms are re-
sponsible for the fermentation and accompanying chemical changes in pollen stored in
comb cells by honeybees, the moulds maybe a component of a required microbial content.
It was considered, for example that they could produce antibiotics, organic acids and en-
zymes, products for which there are uses in industry. These compounds may limit the
Bee-Pollen 95

growth of deleterious micro-organisms in stored pollen and provide enzymes for utilisa-
tion of nutrients. If so, pollen moulds must produce enzymes involved in protein, lipid,
and carbohydrate metabolism. Indeed, most moulds from all pollen sources were shown to
produce caprylate esterase-lipase, leucine aminopeptidase, acid phosphatase, phosphoami-
dase, ~-glucosidase and N-acetyl-~-glucosamidase. A high percent of the isolates (50 %)
from all sources also gave positive reactions for alkaline phosphatase. The majority of
moulds identified were Aspergilli (17%), Mucorales (21%) and Penicillium (32%). We
have also found a predominance of Penicillium in samples of Eucalyptus corbicular pol-
len collected in Portugal9 • In general, the number of isolates decreased in pollen following
collection and storage by the bees. Moulds that may have been introduced by bees during
collection and storage of pollen include Aureobasidium pullulans, Penicillium corylophi-
lum, Penicillium crustosum and Rhizopus nigricans.. Mucor sp., the dominant moulds in
floral pollen, were not found in corbicular pollen and bee bread. Thus, as with yeast and
Bacillus ssp, the mould flora found in corbicular pollen and bee bread may be the result of
microbial inoculations by bees. Chemical changes in pollen therefore may result both
from additions by bees of secretions of glands during regurgitation of honey sac contents
and from microbial fermentation. Such modifications may allow some species to survive
but not others. Even though moulds were more numerous than yeast or Bacillus ssp in the
samples, pollen was rarely overgrown by moulds. Potential microbial spoilage of stored
pollen thus may be controlled by antibiotic substances produced by the normal microflora
of bees or by those naturally present in pollen and/or honey.
These results may assist in the understanding of the antibiotic power of bee pollen.
We must emphasis that this is our interpretation of the research work referred to. The anti-
biotic power of flavonoids that has already been proved for propolis activity (Houghton et
aI., 1995)\0 must also be considered a factor. 39
Chauvin and Lavie (1956)11 in a study of the antibiotic activity of bee-pollen, found
a "factor" with activity against Spullorum. S. gallinarum. S. Dublin, E. coli, Proteus vul-
garis, S. subtilis Caron and B. pyocyaneus. We have followed the extraction procedure
that they used to extract this factor and it seems that this factor could be the flavonoids.
We are currently checking the activity of flavonoids from extracts of bee-pollens from dif-
ferent floral origins. Chauvin and Lavie also verified that different floral sources of bee
pollen gave different levels of activity. In our research we have found that flavonoid type
varies according to the family of pollen plant source (Campos et aI., 1996).12


Another property accredited to well known species of bee-pollen when consumed as

a medicine is the lipid-lowering effect on serum.
Literature relating to bee pollen however was not found in our literature search. We
therefore quote below, research carried out on pollen generally which may help to provide
an understanding of this activity.
Some groups have revealed that extracts of pollen have beneficial properties, lower-
ing serum lipid levels (Samochowiec et Wojcicki, 1981; Wojcicki et Samochowiec,
1984)13., reducing atherosclerosis plaque intensity (Wojcicki et aI., 1986)1\ and decreas-
ing platelet aggregation both in vitro (Kosmider et al. 1983)15 and in vivo (Wojcicki et
Samochowiec, 1984)16. These findings have been confirmed in humans (Wojcicki et aI.,
1983)17. Additionally, studies in humans suggest that a diet supplemented with polyun-
saturated fatty acids (as found in pollen extracts) decreases whole blood viscosity and re-
96 M. G. Campos et al.

duces triglyceride and cholesterol levels in patients with cardiovascular disease (Saynor et
al. 1984)18. With these studies in mind Seppanen et al. (1989),19 analysed the fatty acid
composition of the fat-soluble pollen extract (Cernitin GBX) by gas chromatography in
order to account for the anti-atherosclerotic activity. The analyses revealed that most
(more than 60%) of the fatty acids were in the free form, characterised by a high content
oflinolenic acid (18:3n-3, a-LLA) (70%)
If fatty acids are involved in the beneficial effects referred to, the role of a-linolenic
acid as a precursor of eicosapentenoic acid (20: 5n-3, EPA) is significant, since EPA is
considered to be responsible for reduced platelet aggregation. EPA in vivo is incorporated
into platelet phospholipids, to some extent replacing arachidonic acid and exerting an an-
tithrombotic effect either by competing with remaining arachidinic acid for cyclo-oxy-
genase and lip oxygenase or by being converted to less proagreggatory PGH 3 and TXA 3
(Moncada et Vane, 1984).20


The same extract of pollen from AB Cernelle, Vegeholm, Sweden (Cernitin GBX),
was analysed by Zhang et aI., 1995 21 , who isolated and characterised a cyclic hydroxamic
acid [2,4-dihydroxy-2H-I,4-Benzoxazin-3(4H)-one] from a pollen extract, which inhibits
cancerous cell growth in vitro. This hydroxamic acid is the active compound in the pollen
extract which might be responsible for the symtomatic relief in patients with benign pros-
tate hyperplasia. Seventy-nine patients with this disease, aged from 62 to 89 years, were
treated with pollen extract and showed a mild beneficial effect on prostate volume and on
urination (Habib et aI, 1995; Yasumoto et aI., 1995). Interestingly, one of the major bene-
ficial qualities attributed to bee-pollen by the ancients was its usefulness in the treatment
of prostatitis.



As evident from the above, many chemical, biochemical and microbiological studies
have been carried out with a wide variety of compounds from pollen, but only recently
have scientists focused on a special group, the phenolic compounds. In fact, these exist in
low quantities in bee pollen and other plant parts (Campos et aI., 1990)22 (Markham,
1982)23, (Markham and Campos, 1996)24, as is common for non-nutrient compounds,
which are important for life.
Because in pollen we can find that many groups of compounds differ during the year
or from one species to another, we have developed a method that can identify the corbicu-
lar pollen species with precision using the phenolic compounds. This method offers repro-
ducible conditions that produce a consistent profile (Campos et aI., 1996)13. We have
verified that in bee pollen, and in pollen in general these compounds are reliable species
indicators when. mature pollen is used.
A special type of flavonoid glycoside was commonly found in bee-pollen. After
comparison with the literature we verified that these glycosides often had the same gly-
cosidic linkage. Because these pollens attract insects these glycosides may have a special
taste which honeybees recognize as indicating a suitable food, as been observed with other
Bee-Pollen 97

insects (Harborne, 1994)25. Certain types of flavonoids are known to have a distinctive
taste that insects are attracted to.
The profile of phenolic compounds in pollen is different from one species to another
thus providing a species-specific character for pollen identification. In our studies we have
verified that in the bee-pollen collected from the Central Coast of Portugal, the floral spe-
cies were Eucalyptus globulus, Ranunculus sardous, Salix atrocinerea, Taraxacum sp and
Ulex europaeus. Between the Northern and Southern regions we found the Cistus
ladanifer and in the Central Interior Erica australis was the major species collected for the
bees. Salix atrocinerea and Rhaphanus raphanistrum were minor species in the samples
examined by us.
New Zealand samples were found to contain the same species as those from the
Central Portuguese Coast and the Central Interior. This may be due to the climate and be-
cause the samples were also collected close to the coast. Only the Eucalyptus was not
found in these samples. In contrast the Eucalyptus bee-pollen was commonly found in
samples of bee pollen from Australia. The results of Godinho et Nansen, 1995 26 suggest,
as do ours, that weeds are preferred by the bee, effectively competing with the cultivated
After analysis of the free-radical and anti-oxidant activity of bee-pollens, it was evi-
dent that only the phenolic compounds are active. The derivatives of cinnamic acid are the
most effective phenolics, as evidenced by the loss of much activity when they are absent.
This activity is relevant to the claim that bee-pollen has regenerative properties for the
body and long lives are often attained by bee-pollen users.
The literature records much research work that proves that flavonoids, and in gen-
eral, phenolic compounds, have an antioxidant / radical scavenging effect in the human
body and that they can be given to prevent and cure some diseases (Pathak et aI., 1991)27.
It is probable that active free radicals, together with other factors are responsible for cellu-
lar ageing and can cause biological imbalance that in extreme conditions can be responsi-
ble for premature death. Ageing is a concern in all stories of human life and the story of
the search for the elixir o/youth, continues to this day.
There are many theories presented in literature to explain the ageing process. These
include the "Genetic Theory"-that ageing and long life are genetically programmed
("Programmed Ageing") (Hayflick, 1965 28 ; Warner et aI., 1987 29 ; Holliday et al. 1985 3°).
The Stochastic Theory-this theory claims that ageing is the result of a series of destruc-
tive events that affect all levels cellular organisation ("Random ageing"), e. g. Error ca-
tastrophe theory (Orgel, 1963)31 and Free radical theory (Harman, 1956)32. All these
events cumulatively induce ageing and cellular death.
A free radical is a neutral or charged structure that possesses an unpaired electron
and is represented by the symbol R·. Our body produces active forms of oxygen and oxy-
gen free radicals in the course of normal metabolism. These radical species are very reac-
tive, produce secondary free radicals, inter and intramolecular bridges, oxidation,
halogenations and molecular fragmentation. Cell membrane lipids are vulnerable to free
radicals. "Peroxidation" of polyunsaturated fatty acids gives many derivatives that are in-
dicative of the intensity of the phenomena of cellular oxidation: for example, lipid hydrop-
eroxides, aldehydes (malonylaldehyde-MDA and 4-0H nonenal), conjugated dialdehydes,
hydrocarbons and fluorescent conjugates (lipofuscins). Some of these derivatives have
biological activities (chemostatic action, cellular division effect). Other deleterious effects
of free radicals result from their action on polysaccharides (hialuronic acid de-polymeriza-
tion), proteins (chemical modification of the crucial aminoacids for the enzymatic func-
tions, fragmentation of the peptide chain), and nucleic acids (chromosome bridges).
98 M. G. Campos et aI.

The free radical theory of ageing became more credible after the discovery that ac-
tive free radicals are involved in cellular degradation process, such as cardiovascular dis-
eases, arthritis, cancer, diabetes, etc .. They have also been implicated in Parkinson disease
and Alzheimer disease (Feher et al., 1986)33,34.
Bee-pollen has been used for centuries to protect the body from diseases and espe-
cially to slow the process of ageing. In studying this, Dudov et Starodub (1994)35, fed rats
with bee pollen for one month and studied the resulting state of the erythrocyte redox sys-
tem. It was established that the content of glutathione, total SH-groups, as well as activi-
ties of glutathione peroxidase and glutathione reductase in these animals were increased in
comparision with the control group. Simultaneously a decrease of malondialdehyde and
dienic conjugates in erythrocytes was demonstrated. It was concluded that the antioxida-
tive system is non-specifically activated and oxidative processes blocked in erythrocytes
of rats fed on bee pollen. In another study, primary and secondary humoral immune re-
sponse (the level of specific IgM and IgG) as well as the intensity of delayed-type hy-
persensitivity of sheep erythrocytes were investigated in rabbits fed with bee pollen for
one month. It was shown that bee pollen, acts as an immunomodulator in that it stimulated
humoral immune response and changed the reaction of delayed-type hypersensitivity
(Dudov et al.,1994).36
The free radical scavenging activity of different floral species of bee pollens were
recently studied by us. The results show a big difference between the species containing
derivatives of phenolic acids (which seem to be the more effective scavengers) and those
containing only flavonoids (Campos et aI, 1994)37 (Campos et al., 1996 b)38.
Returning to the story of life and ageing, lipid oxidation in the presence of oxygen,
causes rancidity, a problem recognised since ancient times in relation to the preservation
of oils and fats. Oxidation of vegetable oils and animal fats, produces organoleptic
changes (colour, smell and taste), as well as changes in density, viscosity and solubility.
Surprisingly, it was not until 1950 that scientists started to recognize the significance of
lipid oxidation in Biology and Medicine (Dinis, 1995).39
Other evidence of the benefits of bee-pollen comes from Olympic athletes that eat
this product as a supplement. The coach concluded that bee pollen increased the athletes
crucial recovery time after stressed performance and enabled them to actually improve
their performance the second time around (Fischer, 1986)40. This may be explained by the
anti-oxidant theory. Physical exercise with aerobic characteristics Clln impose on the or-
ganism a supplementary consumption of oxygen that in humans can amount to ten times
the level of basal metabolism. The concept of "oxidative stress" implies a disruption of the
precarious equilibrium between the production and quenching of oxygen free radicals.
With physical exercise this "stress" is brought about by excessive production of such radi-
cals. Paradoxically, people involved in sport regularly live in a situation of "permanent
oxidative stress" with the consequent biological cost.
Today scientists know that fitness training, like an adaptive mechanism, increases
the anti-oxidant defences. Thus more oxygen is consumed without negative consequences.
Nevertheless, in studies carried out at "Faculty of Sport Sciences and at the Centre of Ex-
perimental Cytology of University of Porto", by Silva (1993)41, it has been verified that a
fit athlete when exercised to exhaustion, suffers dramatically high oxidative injury. These
results suggest that anti-oxidant defences could be insufficient in an extreme situation like
this one. It was also shown in this study that those athletes with a high level of oxidative
injury, who recovered quickest (within one hour after the exercise) produced high plasma
levels of vit.E. This suggests that there is a systemic anti-oxidant response to the physical
Bee-Pollen 99

It is tempting to speculate about the anti-oxidant protection provided by bee pollen

in these cases, because of the amount of phenolic compounds in the product given to the
Olympic athletes. It is known today, for example, that some flavonoids can protect vit C,
because when taken together the flavonoids are oxidized first. On the basis of this and
other factors discussed above we can rationalize the benefits of bee pollen in the diet
(Bors et aI., 1995)42.
Diet has long been linked to Medicine and good health, and the use of bee products,
including bee pollen, is frequently recorded. For example, about 487-380 b.c. (before
Christ), Herodicos de Selymbrie, master of gymnastics, inaugurated a diet method, the
"Grand Arte". This formed the basis of the Medicine, that Hippocrate de Cos (460-377
a.c.) established in his Diet Treatise, The Food and the Man Nature (Debry, 1991).43 Even
then, in an empirical way, they knew that the utilisation of a correct diet could have a pro-
phylactic effect in delaying or preventing the onset of some diseases.
Research of the biological activity of compounds isolated from natural sources and
used in folk medicine has established a relationship between chemical composition and
pharmacological activity of many such natural drugs. With bee pollen, the results ofinves-
tigations of its therapeutic activity, add rationale to the "magic" properties attributed to it
in the past. Bee pollen it seems, can be more than a prophylatic and has a place in treat-
ment alongside other natural or synthetic products. More scientists are now beginning to
give credence to the potential of folk remedies, and bee-pollen is one of them.

"We need to ponder the present, to survive in the future, while learning from the past."

I. Peris J. (1984), "Produccion y comercio de los produtos apicolas en Espana". El Campo del Banco de Bil-
bao. Apicultura 93. Bilbao.
2. Huidobro J.F., Simal J., Muniategui S. (1986) "El polen: Determinacion del contenido en agua" Offarm 5
(3), 73-77.
3. Huidobro J.F., Simal J., Muniategui S. (1987) "El polen apicola: Determinacion del contenido em gluci-
dos" Offarm 6 (5) 57-71.
4. Campos M.G., Cunha A. Rauter A. (1994) "Portuguese bee-pollen as a source of flavonoids" Acta Hor-
5. Sepp"nen T., Laakso 1., Wojcicki J., Samochowiec L., (1989). An Analytical study on fatty acids in pollen
extract. Phytotheraphy Research, 3 (3) 115-116.
6. Herbert E. W. Jr., Vanderslice J.T., Meis-Hsia Huang, Higgs D. J. (1987) Levels of thiamine and its esters
in bee collected pollen using liquid chromatography and robotics. Apidologie 18 (2), 129-136.
7. Hagedorn H. H. and Burger M. (1967) Effect of the age of pollen used in pollen supplements on their nutri-
tive value for the honeybee. II. Effect of vitamin content on pollens. J. Apic. Res .• 7,97-101.
8. Xie Y., Wan B., Li W. (1994) Effect of bee-pollen on maternal nutrition and foetal growth. Hua-Hsi-I-Ko-
Ta-Hsueh-Hsueh-Pao. 25 (4) 434-437.
9. Results not published
10. Houghton P., Woldemariam T., Basar A., Lau C. (1995) Quantification of pinocembrin content of propolis
by densitometry and high performance liquid chromatography. Phytochemical Analysis, 6, 207-2\0.
II. Chauvin R., Lavie P. (1956) Recherches sur la substance antibiotique du pollen. Annales de I'Institute Pas-
teur 90 (4) 523-527.
12. Campos M.G., Markham K., Mitchell K., Cunha A. (1996). An approach to the characterisation of bee pol-
lens via their flavonoid/phenolic profiles. Phytochemical Analysis to be submited
13. Samochowiec L., Wojcicki J., (l981) Effect of pollen on serum and liver lipids in rats fed on a high-lipid
diet. Herba Polon. 27, 33
Wojcicki J., Samochowiec L., (1984) Further studies in Cernitins: screening of the hypolipidemic activity
in rats. Herba Pol on. 30, 115.
100 M. G. Campos et al.

14. W6jcicki J., Samochowiec L., Bartlomowicz B., Hinek A., laworska M., Gawronska-Szklarz B., (1986).
Effect of atherosclerosis in rabbits. Atherosclerosis 62, 39.
IS. Kosmider K., W6jcicki 1., Samochowiec L.. Woyke M., G6rnik w., (1983). Effect ofCernilton on platelet
aggregation in vivo. Herba Polon. 29, 237.
16. W6jcicki 1., Samochowiec L (1984). Further studies on Cernitins: screening of the hypolipidemic activity
in rats. Herba Polon. 30, lIS.
17. Wojcicki J., Kosmider K., Samochowiec L., Woyke M. (1983) Clinical evaluation of Cernilton as lipid-
lowering agent. Herba Polon. 29, 55.
19. Sepp"nen T., Laakso I., Wojcicki J., Samochowiec L., (1989). An Analytical study on fatty acids in pollen
extract. Phytotheraphy Research, 3 (3) 115--116.
20. Moncada S., Vane J. R., (1984) Prostacyclin and its clinical applications. An. Clin. Res. 16,241.
21. Zhang-X., Habib F.K., Ross M., Burger U., Lewenstein A., Rose K.. Jaton 1.e. (1995) Isolation and charac-
terisation of a cyclic hydroxamic acid from a pollen extract, which inhibits cancerous cell growth in vitro.
1. Med. Chern. 38 (4) 735--738.
22. Campos M.G., Sabatier S., Amiot M., Aubert S. (1990) Characterisation of flavonoids in three hive prod-
ucts: Bee-pollen, propolis and honey. Planta Medica 56 (7) 580--581.
23. Markham K. R. Techniques of Flavonoid Identification. London, Academic Press. 1982,
24. Markham K., Campos (1996) 7- and 8-0-methylherbacetin-3-0-sophorosides from bee-pollens and some
structure/activity observations. Phytochemistry in press
25. Harborne J. in The Flavonoids: advances in research since 1986. (1. Harborn Editor) Chapman & Hall.
26. Godinho J. et Nansen e. (1995) Estrategia alimentar da abelha domestica (Apis melliferaj como polini-
zador da cultura da meloa (Cucumis melo) em estufa. 0 Apicultor 3 (8) 25--29.
27. Pathak D., Pathak K., Singla A. (1991). Flavonoids as medicinal agents--recent advances. Fitoterapia
28. Hayflick (1965) The limit in vitro lifetime of human diploid strains. Exp. Cell Res. 37, 614-636.
29. Warner H.T., Butler R.N., Sprott R.L., Schneider E.L.. Modem theories of ageing. Ageing vol. 31 (Raven
Press N. Y.) 1987
30. Holliday R.. Kirkwood T.B.L., Cuzin F. (l985}--EMBO Workshop on "Oncogenes, immortalization and
cellular Ageing" Grignon, 3-7 Setembro.
31. Orgel L.E .. (1963) The maintenance of the accuracy of protein synthesis and its relevance 5' to ageing.
Prod. Natl. Acad. Sci., 49, 517-521.
32. Harman D. (I 956}--Thefree radical theory of ageing in "Free Radicals in Biology, (A. Pryor Editor) Aca-
demic Press, 1984,5,255--275.
33. Feher J., Csmos, Vereckci A. (1986) Clinical importance of free radical reactions and their role in the
pathogenesis of various human diseases. in Free Radical Reactions in Medicine, 48-147.
34. Various Comunications--IQ Congresso de Radicais Livres em Quimica, Biologia e Medicina. Instituto Su-
perior Tecnico, Lisboa. Portugal 21-23 de Junho de 1993.
35. Dudov LA., Starodub N.F., (1994). Antioxidant system of rat erythrocytes under conditions of prolonged
intake honeybee{lower pollen load. Ukr. Biokhim. Zh.66 (6) 94-96.
36. Dudov LA., Morenets A.A., Artiukh v.P.. Starodub N.F., (1994). 1mmunomodulatory effect of honeybee
flower pol/en load. Ukr. Biokhim. Zh. 66 (6) 91-93.
37. Campos M.G., Cunha A., Navarro M.e., Utrilla M.P. (1994). Free radical scavenging activity of bee pol-
len. Bull. Group Polyphenols. 17,415-416.
38. Campos M. Markham K., Mitchell K., Veiga J. Cunha A., Paredes F., Frazao L. (1996 b) Therapeutic activ-
ity of bee pollen-Preliminary essays. International Conference on: Bee-products: properties. applications
and apitherapy. May 26--30. Tel-Aviv, Israel.
39. Dinis, T. (1995). Peroxida,iio /ipidica membranar. Actividade antioxidante de farmacos fen6licos
(acetaminofeno, salicilato e 5-aminosalicilato). Disserta9ao de doutoramento apresentada it Faculdade de
Farmacia da Universidade de Coimbra. pp 13-14.
40. Fischer W. L. How to fight cancer & win. 1986
41. Silva P. C. 1993 Radicais livres de oxigenio em Medicina Desportiva, I Q Congresso de Radicais Iivres em
Quimica, Biologia e Medicina
42. Bors w., Michel e., Schikora S., (1995) interaction offlavonoids with ascorbate and determination of their
univalent redox potentials: A pulse radiolysis study. Free Radic. BioI. Med. 19 (I) 45--52.
43. Debry G. (1991). Evolution des concepts en nutrition humain. Cah. Nutr. Diet. 26 (6),435-442.



w. Fierro Morales and 1. Lopez Garbarino

Department of Surgery--Outpatient Service
Hospital "G. Saint Bois"
Montevideo, Uruguay


A new hypoallergic formula of propolis in dressings was evaluated against the

standard formula. Patients (229) with wounds of diferent types that required ambulatory
care were included. The new formula presented the same therapeutic action than the stand-
ard one with a notorious disminution of signs oflocal intolerance (l ,8% vs.18%)
The therapeutics properties of pro polis were demostrated : anti-intlamatory, anti-micro-
bian and stimulant of wound healing by a faster grow and initiation of granulation tissue.
Other properties comprobated were the analgesic effect mainly in bums, the easier handling
of the ambulatory patient and the posibility of avoid hurting the granulation tissue.
The wounds and bum were completed healed in an average of I I days, the septic
wounds en 17,5 days, and a complete reparation in the 67% of the ulcers in 36 days.
In this paper a methodology for the use of this dressing is described, pointing the
importance of acomplishing it by the medical personnel.


The complexity of the composition of propolis and the sinergic action among its dif-
erent componcnts was well established by researchers of the Oxford University in 1990
At the local level a group of actions are important: anti-microbian, anti-inflamatory
(by inhibition of the hydrofolate reductase(3), minimizing the prostaglandin production)
and anti-oxidant(4) (neutralizing the nocive effects of the free radicals at the local level).
These properties complemented each other, explain the therapeutic benefits obtained.
Other benefit obtained is the analgesic effect at the locallevel(5).

102 W. Fierro Morales and J. Lopez Garbarino

The utilization of propolis in the treatment of cutaneous lesions of diferent nature

such as burns, wounds and ulcers is positive in the reparation process, shortening the heal-
ing of wounds and reducing the risk ofinfections(6), but in some cases it is observed aller-
gic dermititis by contact.
In 1990 it was published en Acta Chir. Plast.(7) a paper studying the beneficial ef-
fects of propolis in the treatment of burns against other products.
The treatment of burns is an old issue not exent of contradictions(8). The utilization
of topic anti-microbians it has become generalizated in spite of the lack of alleatory re-
search. In the cutaneous burns it is increased the local production of prostaglandins and
free radicals. The liberation of proteolitic enzimes and the aditional production of oxigen
radicals contributes to the production of aedema, controlling these factors limits the con-
vertion of the burns of partial thickness in complete burns(9).
These phenomenae also are presents in the phisiophatology of the wouds. wich in
the case of a terrain with vascular insufficiency or metabolic anomalies. are lessening or
stopping the repairing process.
Carefully pathologic studies were done in rats and explain the anti-inflamatory and
healing properties of propolis : minimizing the acute inflamatory exudate (by inhibition of
the degranulation of basofiles), estimulating the macrophagic activity, promoting the co-
lagen production and stimulating the epitelization ( incremente of the number of mitosis of
the basal layer and favoring the queratinization) Diaz, P.Prof of Histology, School of
Medical Sciences of La Habana (oral comunication); (10,11,12).


To evaluate the tolerance and efficacy of the new hypo allergic formula of propolis
in dressings against the standard formula.


The present research was done in the period from Sept./91 to Aug.l93 at the Emer-
gency - Dept.of Surgery- Hosp. "G.Saint Bois" Hospital (Montevideo) with ambulatory
patients. In the initial months were progressively evaluated 4 formulae named A, B, C,
and S, and the best results were obtained with the type C (or NF ); 229 patients were in-

Table 1. Population of study and days of treatment.

Propo Ii s type
No.of cases Age average Days oftreatm.
NF(C), B' NF(C) B NF(C) B
Bums 61 15 25 35 II 16,5
Wounds 29 10 41 29 11 11,2
Infected Wounds 53 30 48 46 17,5 16
Ulcers 22 9 63 70 36
'Propolis C=Ncw Fonnula. 2% of pro polis in hydrosoluble cream.
'Propolis B=Standard formula. 8% of propolis in hydrosoluble cream.
New Hypoallergic Formula of Propolis in Dressings 103

c1uded, 115 females and 114 males, from I year old to 92 years old, with varied pathology
classified as follow: a) bums, b) wounds, c) infected wounds, and d) ulcers.

• Wounds. Washing with SF an Benzalconiun c1orure, applying the propolis dress-
ing and closing over it with simple cotton dressing (oclusive treatment)(l3,14).
Each other 48 hs.- 72 maximum - washing and replacement of the propolis dress-
ing (burns excepted).
• Burns. Flictenae resection, washing with SF, drying with warm air, applying the
propolis dressing that remains there until complete granulation when fall by itself.
Clinical controls each 48-72 hs., allowing moisture the dressing with propolis so-
• Infected wounds, ahcesses and ulcers. After bacteriological study it was made a
surgical cleaning and/or drainage of the abcess, washing with SF and benzal-
coniun c1orure. Applying the propolis dressing and simple cotton dressing. In
some cases it was used propolis lotion in order to get a better penetration. In 13
cases the area was drained with dressing. Clinical control and replacement each
48-72 hs. or whenever is indicated. In patients with ulcers were indicated to re-
main in bed with an elevated leg to favour the venous pressure.


1. Among the 165 patients treated with propolis NF it were observed 3 cases
(1,8%) presenting local signs of intolerance and occurring in an average of 16
days after the treatment was initiated. The signs were prurite, eriteme and exu-
dation and they promtly disappeared after the treatment was suspended. Unlike-
ness, the group treated with the propolis B registered a 18% of local intolerance
and appearing at an average of the 9th. day of the initiated the treatment.
2. Only 4 patients had antecedents of local intolerance against the propolis B but
they did not registered any intolerance with treatment of propolis NF.
3. After this initial period it was confirmed the good tolerance of propolis NF, so
from that moment the study continued only with propolis NF.
4. There were included 61 patients with burns according the following age catego-
rization: less of 10 years 19 cases; from II to 40 years 25 cases; from 40 and
more years 17 cases.
It was obtained good reparation and analgesic in an average of 11 days in the 3
groups, but we had the clinical impression that the younger group got a faster evolution. In
36% of the cases it was indicated ATB. The bums included were only those that did not
required an inpatient treatment, most of them of 2nd. degree, with a few hours of evolu-
tion, and produced by agents as water, foods, bitumen and some cases by a direct contact
with fire or hot metals.
Those bums with several hours of evolution and signs of infection were included in
the group of infected wounds and treated as such.
It was observed, the same as other authors( 15), the propolis has the following bene-
ficial effects: Easy handling of the ambulatory patient; less nursing care by replacing
104 W. Fierro Morales and J. Lopez Garbarino

fewer times the dressing, analgesic effect, minimized pain with no replacing and only
moisture the dressing, it has a revitalizing power, because the non replacement it has no
trauma effect in the granulation tissue.
5. A group of 83 patients with infected wounds and abcesses was formed(ages
from 3 to 84 years). In 18 of them it was found the following germs :
estaf.aureus (5 cases) estrepto B hemoliticcus (2 cases), B.piocianic (2 cases)
and only a case of nuemococcus, proteus and serratia. In 6 patients no patho-
genic germ was isolated. In 85% of cases it was indicated a simultaneous treat-
ment of ATB and propolis. In all but one case was observed good tolerance to
propolis. A cure average of 17,5 days was obtained.
6. Non infected wounds with a few hours of evolution formed a group of 39 pa-
tients. Only in 41 % of the cases were indicated ATB. In 3 cases the wounds were
drained with a dressing. In 28 cases the treatement with propolis NF was effec-
tive and it was obtained the cure in II days average.
7. The repairing process also was evaluated in 31 patients with ulcers, 22 were
treated with propolis NF (age X = 65 r=I7-92 y.), 12 cases older than 62 years.
Localization all of them in legs: 8 maleol.intern., 7 maleol.extern., 2 bi-maleol.;
16 had a terrain of veinstasis, 8 had arteriolar compromise, 5 post-trauma ulcers,
and 2 diabetic patients.
In 10 cases a pathogen germ was isolated: estafil.aureus in 4 cases, estreptoc. in 2,
piocianic en 3, and proteus in I. ATB was indicated in 62% of the cases. All the skin lay-
ers were compromised by the wound and in 2 cases reached the aponeurotic level. The
size of the ulcers was somewhere among 2 and 5 cms. A patient 17 years old with a post-
thrauma ulcer of 10 cms.of 45 days of evolution and already including the aponeurotic
layer was healed in 28 days. The cicatrization was obtained in 68% of the cases in an av-
erage of 36 days. Similar results were obtained by Rodriguez in Cuba(16) testing the pro-
polis in 80 patients. In those cases treated with propolis B, the treatment was suspended
due to allergic contact dermatitis. The venous estasis subjacent was responsible for the
slow evolution and the lack of cicatrizal response in most of them (17,18).
The anti-inflamatory and regenerative effect of propolis was shown in the cicatriza-
tion of those wounds with a previous history of non tendency to the cicatrization.
8. In 16 cases of wounds drained with the propolis dressing, it was shown the fast
regression of the inflamatory process and the supuration (anti-bacterian action).
We must stress that the dressing was putted after a surgical cleaning and that the
same was performed as many times as necessary (19). In all cases the cicatriza-
tion was obtained in an average of 22 days.

I) 229 patient with wounds of varied nature were evaluated and it was shown the
good tolerance to the new formula of propolis dressing with only a 1,8% of the cases with
signs of local intolerance. There is no quantitative research in this field in Uruguay, ac-
cording to our previous experiences the results obtained with the treatment of propolis NF
in the present research, as a measure of the allergic reaction to the propolis, were signifi-
cantly lower than those observed with the propolis standard formula. 2) A very satisfac-
tory evolution and cicatrization was obtained in c~ses of wounds with and without
New Hypoallergic Formula of Propolis in Dressings 105

infection and burns. A fast cure, shorter treatment period and less septic complications
were obtained. In 67% of the cases with ulcers (trophics and non trophics) the cicatriza-
tion was obtained. The anti-inflamatory property was demostrated by the fast reduction of
aedema of the limits of the wound. The cicatrizing action was evident by the early forma-
tion of the granulation tissue(between 4th. and 5th.day) being more notorius in those with a
torpid evolution. The anti-microbian capacity was evident by a fast regression of the sep-
tic component of the supurated wounds. 3) Other positive elements to be quoted are: the
importance of carefully follow the methodology, to be a natural product, easy application,
cheap, and produced in this country by standard pharmaceutical procedures.

1. Villanueva V. Barbier N. Gonnet M, Lavie P. Les flavonoids de la propolis isolement de une nouvelle sub-
stance bacteriostatique: Ie pinocembrine(dihidroxy-5,7flavonona). Annales del, Institute Pasteur
1970; 118( 1):84--7. 2) Greenaway W, Scaysbrook T and Whatley FR. The composition and plant origins of
propolis: A report of work at Oxford Bee World 1990;71 (3): I 07-18.
3. Strehl E. Vol pert R, Eistner E, Biochemical Activities of propolis extracts. Inhibition of Dihydrofolate Re-
ductase. Zeitschrift Fur Naturforschung C-A Journal of Biosciences 1994; 49 (1-2):39--43.
4. Pascual C, Torricella RG, Gonzalez R. Scavengig action of propolis extract against oxigen radicals. J. Eth-
nopharmacology 1994;41 (1-2):9-13.
5. Silvestre N, Stranieri G. y Bezerque P. Anestesia troncular de propolcos comparado con lidocaina. Interna-
tional Dental Research 1984.
6. Ponce de Leon R. y Benitez P. Estudio morfologico comparativo del efecto de la propolina, el alcohol y el
balsamo de Shostakovski como agentes cicatrizantes. Investigaciones Cubanas sobre el propoleo. I Sim-
posio sobre Propoleos. Varadero-Cuba 1988:269-71.
7. Troshev K. and others. Regulation of traumatic skin wound healing by influencing the process in the regn-
bouring bone of wound. Acta Chir. Plast.1990;32(3): 152-163.
8. Glastrup H, Knudsen L. y Mazanti C. Tratamiento de las quemaduras, nuevos enfoques de accidentes y
catastrofes. Hezagono 1980;7(8): 1-10
9. Muller MJ, y Hernodon ON. Cirugia. EI reto de las quemaduras. The Lancet 1994;24(6):360-364.
10. Magro F. Application of propolis to dental sockets and skin wounds. Univ. Sch. Dent. 1990;32( I): 4--13
11. Bunta S. Efecto anti-inflamatorio de las pomadas con propoleos I1ISimposio Intemacional de Apiterapia.
APIMONDIA 1978:94-7.
12. Neychev H. and others. Inmunomodulatory action of pro polis. Acta Microbiol. Bulg. 1988;23:58--62
13. Boswick J. Quemados. Clin. Quinirg. Norteamerica. 1987; I.
14. Lepore G. y Giuria H. Quemaduras menores, tratamiento de urgencia y seguimiento ambulatorio. Catedra
de Cirugia Plastica y Quemados. ProfJ. Hornblas.
15. Ramirez M.Propoleos en el tratamiento de quemados. I' Jornadas Nacionales de asistencia integral del
nino quemado. 1989.
16. Rodriguez A. y col. Resultados de la aplicacion del propoleos en ulceras y gangrenas de las extremidades
inferiores. Investigaciones Cubanas sobre el propoleos. I Simposio sobre Propoleos. Varader-Cuba
17. Mescon H.y col. Dermatitis por estasis 0 eczema varicoso. In: Robbins. Patologia estructural y funciona!.
Interamericana Mexico. 1975: 1331-2.
18. Apositos para las ulceras de las piemas. Drug and Therapeutics Bulletin. 1094;24(3): 172-5.
19. Temesio P. Nuevo metodo del tratamiento local con prop6leos. Investigacion clinica. Policlinica de diabe-
tes. Hosp.Maciel 1983.



Tsuguo Yamamoto

Nihon Natural Foods Co., Ltd.

6-26-12 Nishishinjuku, Shinjuku-ku, Tokyo, Japan


In October 1985, the 30th International Apicultural Congress held in Nagoya first
introduced propolis into Japan, when the latest reports on basic studies on propolis and the
clinical application of propolis to the intractable diseases were presented by the re-
searchers and the medical doctors from various countries abroad. Together with the intro-
duction of a crude propolis by the Brasilian beekeepers, the name of 'propolis' became
known instantly with the possibility of future promising substance. However during first
five years since 1985 propolis was regarded only as health foods and folk remedies and
was not paid much attention to by scientists. Propolis studies made public in 1987 related
only to general findings, introduction of overseas literatures, and distribution of flavonoid
components I).
In 1991 at the 50th Japan Cancer Society Meeting, Matsuno of the National Institute
of Health reported three compounds with tumor killing activities in propolis from his ex-
perience curing terminal-stage uterine cancer with propolis2) With this as momentum,
many of pharmaceutical companies and research institutes have started studying propolis
with the prospect of its use more promising. Among them Hayashibara Biochemical Labo-
ratories Inc. in the course of his interferon research, took an interest in the BRM (biologi-
cal response modifiers)-like effects in propolis, such as the antivirus effect 3 ) and immune
activating effect 4 ) and has been studying propolis these ten years, and published a fairly
number of remarkable research results.
This paper describes representative studies on the antimicrobial effect and cytotoxic
effect of propolis from 1985 up to now in Japan.


The strong antimicrobial activity of Propolis is often compared to a "natural antibi-

otic". Its antimicrobial characteristic against various microbes has recently been studied
throughout the world.

108 T. Yamamoto

In Japan, Aga et a1. 5), 6) Matsun0 71, ltoh et a1. 8), and Nakano et a1. 9 ) have studied and
reported the effects of propolis and antimicrobial substances isolated from propolis on
various types of fungi, yeast's and bacteria, as well as on specific pathogenic microbes,
such as Helicobacter pylori, and Methicillin-resistant Staphylococcus aureus (MRS A) be-
tween 1992 and 1995.

1. Antimicrobial Activity of Brazilian Propolis 5)

In 1992, Aga et al. of Hayashibara Biochemical Laboratories Inc. published their re-
search results on the antimicrobial activity of an ethanol aqueous solution extract of Bra-
zilian propolis against 8 strains of fungi, 4 strains of yeast, and 42 strains of bacteria,
including Enterobacter and Actinomyces.
In their study, the propolis extract showed strong antimicrobial activity against Mi-
crococcus lysodeikticus (MIC: 15.6 /-tg/ml), Bacillus cereus, Enterobacter aerogenes
(MIC: each 31.3 /-tg/ml), Corynebacterium equi, Mycobacterium phlei, Thermoactinomy-
ces intermedius, Arthroderma benhamiae & Microsporum gypseum (MIC: each 62.5
/-tg/ml), but showed very weak antimicrobial activity against Enterobacter such as Bifido-
bacterium, Lactobacillus, Eubacterium, and Bacteroides.
This antimicrobial spectrum agreed with the results of German propolis studied by 1.
Metzner et al. 10) and of American propolis studied by L. A. Lindenfelser lll in which an-
timicrobial activity was recognized against 7 strains of bacteria such as Micrococcus
lysodeikticus & Bacillus cereus, and Arthroderma benhamiae.
The MIC values of Brazilian propolis show about 3 times more antimicrobial activ-
ity against Bacillus subtilis, Staphylococcus aureus and Arthroderma benhamiae than that
shown by German propolis. This result suggests a difference in antimicrobial substance
contents and the presence of stronger antimicrobial substances. Accordingly, it is expected
that isolation and identification of these substances and further elucidation of the mecha-
nism of the antimicrobial action will be conducted.

2. Isolation and Identification of Antimicrobial Compounds in

Propolis 6)
In a later study, Aga et a1. isolated and identified three antimicrobial compounds
from Brazilian propolis. As shown in Fig.l, they identified these compounds as 3,5-
diprenyl-4-hydroxy cinnamic acid (as Compound I), 3-prenyl-4-dihydro cinnamoloxy cin-
namic acid (as Compound 2), and 2,2-dimethyl-6-carboxy ethenyl-2H-l-benzopyran (as
Compound 3). The results of this study were published in 1994.
Table I gives the antimicrobial activity (MIC values) of these compounds against
three types of microbes selected as specimens to be studied.
As shown in Table 2, compared with the original propolis extract, 3,5-diprenyl-4-hy-
droxycinnamic acid (Artepillin C) had stronger antimicrobial activity against cutaneous
fungi (ex. Microsporum, Arthroderma), putrefying bacteria (ex. Bacillus), Corynebac-
terium, and pyogenic bacteria (ex. Pseudomonas).
It has long been said that propolis is effective for the treatment of dermal disorder
and burns. This characteristic of propolis suggests that Artepillin C plays a leading role in
the antimicrobial and anti-inflammatory activities of pro polis.
Later, Kimoto et a1. disclosed that Artepillin C played a principal role not only an-
timicrobial activity, but also in "anticancer action" of Brazilian propolis to be introduced
below. 121
Present State of Basic Studies on Pro polis in Japan 109


Compound 1 R'=H, R 2 =CH 2CH=C(CH 3 )2

(3,5-diprenyl-4-hydroxycinnamic acid) (ArLepillin C)

Compound 2 R'=CO(CH 2 )2 Ph , R2=H

(3-prenyl-4-dihydrocinnamoloxycinnamic acid)


Compound 3

Figure 1.

3. Anti-Helicobacter pylori Substances in Propolis 8)

Itoh et a1. of the Zenyaku Kogyo Co., Research Institute examined the antimicrobial
activity of Chinese, Argentine and Brazilian propolis against Heliocobacter pylori, whose
connection to gastritis and gastric ulcer was suspected. Their research results were pub-
lished in 1994.
According to these results, Argentine propolis showed the highest antimicrobial ac-
tivity with an MIC value of 50 f..\g/ml, followed by Chinese propolis at 100 Ilg/ml, and
Brazilian propolis at 200 Ilg/m1. They reported that each propolis showed an anti-Helioco-
bacter pylori effect.

Table 1. Antimicrobial activity of isolated compounds 1_3 6 )

Compo- MIC ( '-' g/ml)
silion' B.cereus E.aerogenes A.benhamiae
Compound 5.296 15.6 31. 3 15.6
compound 2 2.396 31. 3 62.5 ,250
compound 3 0.896 125 125 62.5
Crude propolis 31. 3 31. 3 125
, as a percenlage relalive lo lhe dry solid crude propolis
110 T. Yamamoto

Table 2. Antimicrobial activity of artepillin C6)

MlC (/I. g/ml)

Slrains Arlepillin C Propolis

Microsporum gypseum (lFO 8231) 7.8 62. S

Arlhroderma benhamiae (JCM 1885) 15.6 62. S
Bacillus cereus (lFO 3466) 15.6 31. 3
Bacillus sublilis (ATCC 6633) 31. 3 31. 3
Corynebaclerium equi (lFO 3730) 31. 3 62. S
Micrococcus lysodeiklicus (lFO 3333) 31. 3 lS.6
Pseudomonas aeruginosa (lFO 3453) 31. 3 12S
Enlerobacler aerogenes (lFO 3321) 31. 3 31. 3
Mycobaclerium smegmalis (JCM 5866 T ) 31. 3 SOO
Mycobaclerium phlei (JCM 5865 T ) 62.5 62. S
Slaphylococcus aureus (ATCC 6538P) 62.5 2S0
Slaphylococcus epidermidis (ATCC 12228) 62.5 SOO
Thermoaclinomyces inlermedius (JCM 3312 T ) 62.S 62.5
Micrococcus luleus (lAM 1099) 125 2S0
Propionibaclerium acnes (JCM 6425 T ) 12S SOO
Flavobaclerium meningoseplicum (lFO 12535) 2S0 2S0
Kloeckera apiculala (JCM 5947) SOO SOO
Saccharomyces cerevisiae (lFO 0214) SOO 2S0


OH 0 OH 0
pinocembrin galangin (R=OH), chrysin (R=H)

compound MIC ( /.1 g/ml)

pinocembrin 12.5
galangin 25
chrysin 25

Figure 2.
Present State of Basic Studies on Propolis in Japan 111

They further isolated fractions that had anti-Heliocobacter pylori activity from pro-
polis by column chromatography, and identified these fractions as pinocembrin (MIC:
12.5 Ilg/mi), galangin and chrysin (MIC: each 25 Ilg/ml). The anti-Heliocobacter pylori
activity of pinocembrin showed an antimicrobial activity equal to that of Lansoprazole,
which was used as a control (Fig.2).
However, pinocembrin showed a lower antimicrobial activity against another mi-
crobes other than Heliocobacter pylori (MIC: 50 ->200 Ilg/ml). This suggested that one of
the factors contributing to propolis's anti-ulcer effect was specific anti-Heliocobacter py-
lori activity due to flavonoids such as pinocembrin.

4. Anti-MRSA Compound Isolated from Brazilian Propolis9 )

In 1995, Nakano et al. of Hayashibara Biochemical Laboratories Inc. studied active

substances in Brazilian propolis to determine whether Brazilian propolis exhibits anti-
MRSA activity.
This substance was 3-prenyl-4-dihydrocinnamoloxycinnamic acid having a chemical
structure shown in Fig.3. As illustrated in Fig.3, only 2 Ilg/ml of this substance evaluated
by the MIC method had an anti-MRSA activity 100 to 400 times higher than that of each
component of the propolis ethanol extract and propolis.
In 1994, this substance had been isolated from Brazilian propolis by Aga et a1. 6) at
Hayashibara Biochemical Laboratories as an antimicrobial substance together with
Artepillin C. At the time, however, no study was made on anti-MRSA activity.
Regarding studies of anti-MRS A activity of propolis made so far, Grange et al. 13 ) re-
ported on the relatively strong anti-MRS A activity of the extract of French propolis in 1990.
Activity of the same degree as that in this report was recognized in Brazilian propolis.
The anti-MRSA activity of propolis is considered due to the synergetic effect of
complex components in propolis. During their work of isolation, the fraction containing
this 3-prenyl-4-dihydro-cinnamoloxycinnamic acid was the most active, so Nakano et al.

o o

3-prenyl-4-dihydrocinnamoloxycinnamic acid

co.pound IIIC( JL9.m1) compound IIIC( JL 9.111)

propolis exLracL 200 9a1an9in )aoo
caffeic acid )800 Kaemferol >800
coumarie acid )800 ArLepi11in C 200
cinnallic acid )800 anLi-KRSA compound 2

Figure 3.
112 T. Yamamoto

estimated that this substance was the main component of the anti-MRSA activity of Bra-
zilian propolis.
Findings on the anti-microbial activity of propolis and isolated substances against
specific microbes in Japan have been described here in. It is interesting that, in addition to
the known effect of flavonoids, both Artepillin C reported by Aga et al. and New Clero-
dane Diterpenoid founded by Matsuno as an anti-tumor substance have strong anti-micro-
bial activities.
We expect that new substances with antimicrobial activity against specific patho-
genic microbes will be discovered from propolis, and through basic research, application
studies for clinical use will be soon to follow.



Commercially available books and European papers l4) describe that propolis and its
components are effective in inhibiting human malignant tumors and cancer cells. How-
ever, there had been no papers objectively evaluating the mechanism or degree of their an-
titumor effects before 1990.
In Japan, Matsun0 2 )?) of the National Institute of Health, Arai et al. 15 ). Kimoto et
al. 12 ) of Hayashibara Biochemical Laboratories Inc., and Suzuki et al. 16 ) of the Suzuka Col-
lege of Technology published results of studies on propolis, and on some of its isolated or
fractionated active components between 1991 and 1996.

1. Cytotoxic Effect of New Clerodane Diterpenoid Isolated from

Brazilian Pro polis 7)

In 1990, Dr. Matsuno of the National Institute of Health found that the ethanol ex-
tract of Brazilian propolis transformed human hepatic carcinoma cells and uterine carci-
noma cells cultured in vitro, and that it inhibited their growth.
Afterwards, he orally administered a large quantity of a propolis drink to one of his
relatives, who had cancer of uterine cervix and was unable to undergo surgery because of
her poor physical strength. He also continuously applied the propolis ethanol extract di-
rectly to the affected part. As a result, the lesion became a burn-like scar several weeks
later, and her cancer of uterine cervix disappeared2 ).
Since then, he started to study the purification and isolation of the antitumor active
substances contained in propolis, and observed the cytotoxic effect of fractionated sub-
stances on human hepatocellular carcinoma, HuHI3. As a result, he found this effect in
the following three substances that is, the known substances of quercetin, and caffeic acid
phenethylester, and a new compound belonging to clerodane diterpenoid. He reported his
findings at the 50th Japanese Cancer Association Congress in 1991.
Clerodane diterpenoid, in particular, was very active in destroying tumor cells, espe-
cially human cervical carcinoma cells (HeLa cells) and Burkitt's Lymphoma cells in addi-
tion to HuHI3.
His findings suggested that this substance showed selective toxicity to tumor cells,
stopped the cell growth cycle in the gene synthesis phase (phase S), changed the proper-
ties of the cell membranes, and killed cells by disturbing their ionic permeability.
Present State of Basic Studies on Propolis in Japan 113

Rl 100

R2 ;e
(.) 50
> • primary rabbi t. kidney
... !luman diploid f1brcb1ast.
:::l • 1IIIH13 (t.umor)

1 100 1000
clerodane diterpenoid (~g I ml)

Figure 4.

Although the details of the mechanism are still being analyzed, this substance acts
on tumor cells in phase S, when tumor cells are growing more actively than normal cells
and synthesizing genes rapidly. Therefore, Dr. Matsuno assumes that tumor cells are ulti-
mately destroyed because they are growing at a different speed.
The effects of clerodane diterpenoid on HuH 13 and two normal cells ( untrans-
formed primary rabbit kidney cells and human diploid fibroblast cells) were examined. As
a result, as is shown in FigA, a large difference in the effective concentrations of this sub-
stance was found, and it was proved that an appropriate concentration of this substance
could kill tumor cells without affecting normal cells.
These results emphasized the possibility that a new treatment, which could kill tu-
mor cells exclusively without damaging normal cells, could be developed by determining
an appropriate concentration and administration method for this substance. Dr. Matsuno
confirmed its treatment effects on cancer patients, and made public the details of his expe-
rience at the 51 st Japanese Cancer Association Congress.

2. Biological Effects of Propolis on Macrophage Function and Tumor

Metastasis 15) 17)
Dr. Arai et al. of Hayashibara Biochemical Laboratories Inc. has begun to study the
biological activity of Brazilian propolis from various perspectives since 1990. In order to
make the propolis easy to use, they powdered the propolis extract using anhydrous mal-
tose, and tried to confirm the effectiveness and mechanism of a BRM-like substance, aim-
ing at isolating and identifying this substance.
This propolis powder was dispersive in water, was free of endotoxin that macro-
phage activates, and contained 13.8% propolis-derived solids, so that concentrations dur-
ing the experiment were all expressed in terms of propolis powder.
They discovered a macrophage activation phenomenon related to the immune func-
tion of living organisms, then in 1993 published the results of their minute studies on the
effects of propolis on macrophage spreading, phagocytosis, motility, and cytokine produc-
tion!7). In 1994, they made public its inhibitory effect on lung metastasis in mice!5).
After adding a culture medium containing propolis powder solution to the abdomi-
nal macrophage obtained from the BALB/c mouse, stretching as shown in Fig.5 was ob-
served. This phenomenon correlated to propolis concentration and time, and a dose
reaction was recognized.
114 T. Yamamoto

Figure 5.

Similarly, the effects of this propolis on fowl phagocytosis and motility, TNF pro-
duction by LPS (lipopolysaccharide; pyrogen) coexistence. cytotoxic factor NO (nitrogen
oxides) production inhibition were studied. As a result, it was revealed in vitro that its ef-
fects depended on concentration and time, or the presence or absence of an LPS stimulus.
Moreover, in vivo, though TNF production in mouse blood was increased by an LPS
stimulus, by administering 0.2 mg and 2.0 mg propolis Imouse 3 hours before an LPS
stimulus, ten times more TNF could be produced.
Propolis as one of BRM (biological response modifiers)-like substance activated
macrophages by, Propolis by itself did not produce cytokine in vivo; however cytokine
production was sharply accelerated by an LPS stimulus. These results suggest the activa-
tion effect of immune cells which produce cytokines.
Prior to a test on the inhibition of tumor metastasis by this propolis, its effects on the
growth of mouse colon carcinoma cell Colon 26 were studied. As shown in Fig.6, this pro-
polis inhibited the growth of the cells in a concentration- dependent way, and had a direct
Next, 40 f.lg of this propolis was administered to BALB/c mice (7-week-old fe-
males), on to which Colon 26 cells were then transplanted. Various concentrations of this
propolis were continuously administered for 6 days, and the metastatic lesions in the lung
were counted 14 days later. As shown in Table 3, the dose of propolis was optimum. Lung
metastases were reduced to 80% in the group receiving 0.1 mg propolis powder, 57% in
the 0.2 mg group, and 70% in the 0.4 mg group.
Based on these results, they assumed the following . The dose of propolis adminis-
tered to the mice was too small to directly inhibit cell growth in vivo. Administration of
propolis activated immune cells, mainly macrophages, and inhibited and removed the im-
plantation of metastatic tumor cells as foreign matter to lung tissues, leading to a reduc-
tion in the number of metastatic lesions.
Present State of Basic Studies on Propolis in Japan 115


• Maltose (Base)
• Propolis POwder


~ 60

'" 40


o 0.2 0.4 0.6 0.8 1.0

concentrat ion (mg/ml)

Figure 6.

3. Cytotoxic Effect of Artepillin C Isolated from Brazilian Propolis t2 )

Three substances isolated and identified from Brazilian propolis by Aga et aI., espe-
cially Artepillin C, have been shown to have a stronger antimicrobial activity than the
crude propolis61.
The above-mentioned study on the antitumor effect of propolis made by Dr. Arai et
al. suggested the presence of a substance in propolis that was effective in killing tumor
cells. This study also suggested that c1erodane diterpenoid, an antitumor substance in

Table 3. Inhibition lung metastasis of mouse colon carcinoma (colon 26) by propolis l51

dose Colonies
Group 1. v. /mous numbers Average ± SE

Propolis powder O. 1mg/O. 2ml 13 94.7 ± 13.7

O.2mg/O.2ml 13 66. P± 9.6
O.4mg/0.2ml 13 81. 0 ± 13.3
Anhydrous maltose O.4mg/O.2ml 13 134.0 ± 6.1
Physiological saline O.2ml 13 115.9 ± 15.4

grafted cells: 5xlO'/mous, . ; p<O.05

116 T. Yamamoto

10 T 100
control Artepillin C
HGC -b- ....... • G-361

HLC .0- Itt.. mARA

Colon 26
-<>- ...... 80
• 8-16

QI lI- 60

> ~

Arlepillin C ; 100~9/m1 malignant melanoma cells

10' 0
0 2 4 5 6 1 5 10

days concentration (U g 1m I)

Figure 7

propolis discovered by Dr. Matsuno, also had relatively strong antimicrobial activity 7).
Stimulated by these facts, Kimoto et al. examined the antitumor effect of these three an-
timicrobial substances, and found that only Artepillin C had a strong cytotoxic effect on
various cultured tumor cells and transplanted carcinoma cells. They published their re-
sults in 1995 12).
Artepillin C showed a noticeable inhibitory effect on the growth of the 18 types of
cultured tumor cells tested, with a dose of only 1{}-100 I!g/ml. Its cytotoxic effect was es-
pecially eminent in rapidly growing cells.
Fig.7 shows the effects of Artepillin C on human gastric carcinoma cells (HOC), hu-
man lung cancer cells (BLC), and mouse colon carcinoma cells (Colon 26), when 100
I!g/ml each of Artepillin C was added and cultured. Fig.7 also presents the effects of
Artepillin C at different concentrations on malignant melanoma cultured cells (0-361,
IHARA and B-16). Artepillin C demonstrated a conspicuous effect in inhibiting the
growth of these cells.
Fig.8 sets out the effects of 100 I!g/ml of Artepillin C on HLC (Fig.8-a and -b) and
HTSA (Fig.8-c and -d). Fig.8-a and-c are microscopic photos showing cells cultured with-
out treatment. Fig.8-b and-d show cells given Artepillin C and cultured for 24 - 48 hours.
Severe cell injury (Fig.8-b) and cell necrosis (Fig.8-d) can be observed.
Though Artepillin C is insoluble in water, newly developed water-soluble [Artepillin
C]-Na was used to examine its antitumor effect. The results of this examination are shown
in Fig.9 and Fig.IO.
[Artepillin C]-Na was more effective than Artepillin C in its anticancer effect, re-
sulting from DNA synthesis inhibition of human leukemia cells (HL-60 and THP-I) and
malignant lymphoma (U937) (Fig.9). Also [Artepillin C]-Na was more effective than
Artepillin C in antitumor effects causing cell death and growth inhibition of hepatocellular
Present State of Basic Studies on Propolis in Japan 117

.. •
. . , .. ., •• \

.J. •
It • • •
. J
• •
.*' .. • . #I

" •

Figure 8.


•• IIL-60

A U937







20 40 60 80 100

concentrat i on (~g/m l )

Figure 9.
118 T. Yamamoto






'" 40


0.001 ,000

concentratIon Lug/mIl

Figure 10.

carcinoma cells (rat-derived; HTSA and RL-34, human-derived; PLC /PRFIS), and human
larynx carcinoma cells (KB) (Fig .! 0).
Next, in vivo, 500 f.!g of Artepillin C were intravenously administered for 4 weeks
every other day to nude mice transplanted with HLC, HOC, and PLC /PRFIS as xenografts,
and Colon 26 and HTSA as allografts. Fig.!! shows the growth inhibition of the trans-
planted carcinoma cells, and the histopathologic findings.
As a result, nuclear degeneration (caryolysis and pycnosis) and apotosis-Iike popula-
tion death were observed in these carcinoma cells (Fig.ll-c and -d). Small-population
physalis turned to large-population degeneration and necrosis, which inhibited cell growth
(Fig. I!-e and -f).
Afterwards, collagen, macrophages, and helper cells increased around carcinoma
cells which resembled solitary islands (Fig. I I-f).
Conversely, in Fig.ll-a and -b, HLC in the mouse given Artepillin C separated into
two small tumors, and no further swelling was observed.
These results suggest that the cytotoxicity of Artepillin C to many carcinoma cells
inhibits cancer cell growth by inhibiting the DNA synthesis of carcinoma cells (Fig.9) and
by damaging the respiratory enzyme system of intracytoplasm glomerular intima (Fig. I 0).
In particular, Artepillin C has a strong antitumor effect on leukemia cells, and its fu-
ture application as an adjuvant drug for intravenous injection chemotherapy is expected.
Several research reports in Japan on the antitumor effects of propolis have been de-
scribed here. However, there is no complete agreement between propolis and the compo-
nents isolated from propolis in terms of the effect on cancer cells and the mechanism. This
fact suggests that new antitumor substances and cytotoxic substances (antimicrobial sub-
stances) may be discovered in propolis in the future.
Present State of Basic Studies on Propolis in Japan 119

Figure 11.

In concluding this paper, we would like to say that we expect Japanese researchers
to discover new substances which are more effective in dealing with tumor cells without
damaging normal cells in another phases than the gene synthesis period (S period); for in-
stance, in the mitotic period (M period).


I wish to thank Prof. Avshalon Mizrahi of the Israel College of Complementary

Medicine and Prof. Mitsuo Matsuka of Tamagawa Univ. for this opportunity to introduce
"the present status of propolis in Japan", and Mr. Satoshi Iritani of Hayashibara for his
valuable comments and support to complete this paper.
120 T. Yamamoto

I. Characterization and Actions of Propolis; Fragrance Journal, 83, 1987
2. Matsuno T.; Hammingbird, 1, 14, 1991
3. Tatefuji T. et a1.; Shoyakugaku Zasshi. 47(1), 60, 1993
4. Moriyasu 1. et a1.; Biotherapy, 7(3), 364. 1993
5. Aga H. et a1.; Medicine and Biology, 124 (5), 205, 1992
6. Aga H. et a1.; Biosci. Biotech. Biochem., 58 (5), 945, 1994
7. Matsuno T.; Honeybee Science, 13 (2), 49, 1992
8. Itoh K. et a1.; Honeybee Science. 15(4), 171, 1994
9. Nakano M. et a1.; Honeybee Science, 16 (4),175,1995
10. Metzner 1. et a1.; Pharmazie, 34, 97, 1979
11. Lindenfelser L.A.: Amer Bee 1., 107 (3), 90, 1967
12. Kimoto T. et a1.: Nihon Iji Shinpou, 3726. 43. 1995
13. Grange 1. M. et a!.; J. Royal Soc. Med., 83, 159, 1990.
14. Grunberger D. et a1.: Experientia, 44, 23, 1988
15. Arai S. and Kurimoto M.; Honeybee Science. 15 (4),155.1994
16. Suzuki l. et a1.; Honeybee Science, 17 (1), I, 1996
17. Moriyasu J. et a1.: Biotherapy 8(3). 346, 1993



Park Jong-Sung and Woo Kun-Suk

lNHB International Co. Ltd.

Janghakhoegwan Bldg. 602-ho, Oaechi-dong 945-15, Kangnam-gu,
135-280, Seoul, Korea
20epartment of Agricultural Biology, College of Agriculture and Life Sciences
Seoul National University
44 1-744, Suwon, Korea


Propolis cosmetics commercialized in Korea are the liquid foundation (GAC 4927),
Propoleo cream (GAC 4914), and the eye cream (GAC 4922). Liquid foundation has 3%
of propolis with 20 other ingredients. The propoleo cream includes 2% of propolis. The
cream is specially beneficial for oily or pimply face. The eye cream with I % of propolis
helps to absorb nutrients and to prevent excessive moisturizing. PropoJis in cosmetic
shows a great deal of advantages in antibacterial activity, moisturization, revitalization,
and elasticity of skin. Additionally, Amiga Lotion (G 153- 1122) and Amiga Skin (G 153-
1123) are also products with propolis. To see the consumer's satisfaction, questionnaires
were distributed, and the results are shown 8 I % on the liquid foundation and Propoleo
cream, and 83% on the eye cream as table 1,2,3.


The demand of consumers for natural ingredient in the cosmetology is increasing re-
cent decade. To meet the desire, Taepyengyang trade Co. founded in 1988 and imported
cosmetic raw materials at the beginning. With changing name to Korea Antuco in 1990
and Amiga Co. in 1994, company developed 5 natural cosmetics in 1990. Founding the
marketing company, NHB International Co. Ltd. the company has developed 12 propolis
added cosmetics in 1990. Propolis cosmetics initially commercialized and attracted many
people's attention in Korea. These products are foundation (GAC 4927), Propoleo cream
(GAC 4914), and eye cream (GAC 4922). Additionally, Amiga Lotion, cleansing water,

122 Park Jong-Sung and Woo Kun-Suk

Brice massage cream, revital pack, royal jelly cream, and essence lotion are also products
with propolis.
Propolis imported from a satellite town in Chile, because Korean propolis may have
an allergy problem on the skin due to a lacquer tree. Propolis, what is called bee glue, is
resinous materials collected by honey bees from the bark and buds of various trees (1).
The composition generally consists of terpenes, resins, volatile oils, flavonoids and other
compounds. Though about 130 chemicals identified from propolis, all propolis do not
have same composition (1). The physical characters is 65.5°C melting point and brittle in
cold room temperature (3). The solubility are partial in water, but rather completely in
ether or chloroform. The color ranges from yellow to dark reddish brown (3). ,
The new brand name Brice is combined word from English brilliant and italian Dolce,
which means bright and beautiful grace. All the ingredients of products came from the natural
source. The marketing system is multi-level marketing and organized by members. The cur-
rent members are about 20 thousands and the number of member is increasing rapidly since
consumers have begun to realize the benefits of natural ingredients and functional cosmetics.
Because of extraordinary variability of propolis in the area and origin of plant, it is
not reasonable to declare that propolis shows a great effect on skin and in health of every-
body, but there is no doubt it is effective on inflammations and other skin troubles, ac-
cording to the literatures (2,4) and empirical data showed in this study.


To evaluate the benefits of Brice, questionnaires were distributed among 1,600

members during marketing seminar and the empirical data were collected in 1996. The
members are divided into 4 groups after questioned in according to their general impres-
sion on the cosmetics, so A group is the most favorable persons and D group is the least.
Questionnaires were asked on the effects of individual cosmetics.


4.1. Liquid Foundation

Liquid foundation has 3% of pro polis with 20 other ingredients. It helps balance pH of
your skin, blocks the aging UV rays as sunscreen, and affinity enables to stay longer on the
skin. Propolis works as antibacterial agent on the skin. Moisturizing effect is excellent and
long-lasting. The satisfaction of consumers varies as shown table 1, and the average is 81 %.

4.2. Propoleo cream

The Propoleo Cream includes 2% of propolis. The cream is specially beneficial for
oily or acne. It also has characteristic in soothing your vulnerable skin and improving
troubles. Propoleo cream also shows 81 % in consumer's satisfaction as in table 2.

4.3. Eye cream

Eye cream with 1% of propolis helps to absorb nutrients and to prevent excessive
moisturizing. This also increases the firmness and elasticity of skin. Lines and wrinkles
Usage and Composition of Propolis Added Cosmetics in Korea 123

Table 1. Satisfaction rates on an liquid foundation (unit: percent)

Characters A B C D Average
pH balance 90 90 85 80 86
UV blocking 85 85 80 70 80
Bacteriostasis 95 80 70 60 76
Moisturizing 90 90 85 70 84
Affinity 90 85 80 60 79
Elasticity 90 85 80 75 82

are softened and smoothed. With the help of eye cream, blood circulation around eyes im-
proved to velvety and smoothy skin. Users massage the cream around eyes in the morning
and evening. The average effect is 83% as shown in table 3.

4.4. Other Cosmetics

Moisture shampoo prevents scurf, acne, and depilation. Approximately 48 kinds
natural ingredient exist in the shampoo. The essence lotion makes healthy hairs by provid-
ing nutrients and coating the hair for protection. This lotion including propolis has 50 dif-
ferent nutrients. Brice massage cream is able to cleans dust, exudate, and sebum on the
skin, even in the skin pore. The major compositions of massage cream are propolis and
rose hip oil, which makes silky skin. Brice royal-jelly cream inhibits the reduction of col-
lagen, which is cause of wrinkle and fine lines. Plus it protects skin from UV and oxidant
with balancing moisture and nutrients.

As a result of questionnaires, the liquid foundation needs improvement on bacte-
riostasis and affinity, which shows lower satisfaction on the characteristics. Propoleo
cream generally shows very good characters except skin pore care and revitalization of
skin. Eye cream also proved to have excellent effects on the skin care, but satisfaction on
wrinkle care need some improvement.

Table 2. Satisfaction on a Propoleo cream (unit: percent)

Characters A B C D Average
pH balance 90 90 85 80 86
Bacteriostasis 95 85 80 70 82
Moisturizing 90 90 80 80 85
Affinity 90 85 75 75 81
Elasticity 95 90 75 70 82
Acne and scar care 98 80 70 65 78
Skin pore care 85 80 70 70 76
Revitalization 85 80 75 65 76
124 Park Jong-Sung and Woo Kun-Suk

Table 3. Satisfaction on an eye cream (unit: percent)

Characters A B C D Average
Puffiness reducing 95 90 85 80 88
Blood circulation 90 90 80 75 84
Wrinkle care 85 80 75 70 78
Elasticity 90 85 85 70 83

At this moment only the propolis from Chile is manufactured in the process, but
natural propolis in Korea can be purchased and used commercially as long as the restric-
tion is made on the origin of plant like the green house environment.
Propolis in cosmetic shows a great deal of advantages in antibacterial activity, mois-
turizing, revitalization, and elasticity on skin. The products can be used for everybody re-
gardless gender and sex and age. The great benefits proved on skin trouble, scar, and acne.
Body cares, soaps, two-way cakes, and lipsticks will be developed.

I. Bonney, R. (1995) Propolis. Bee Culture. II, 630-634.
2. Maeda G. The miracle of propolis (translated into Korean). Young-woo Publisher, Seoul. 1994.115pp.
3. Root, A. I. ABC and XYZ of Bee Culture. A.1. Root Company, Medina, Ohio, U.S.A. 1954. pp 538-541.
4. Toth, G. (1988) Cosmetic use of hive products: facts and prospect. American Bee Journal. 6, 431---434.



Woo Kun-Suk and Park Jong-Sung

lDepartment of Agricultural Biology, College of Agriculture and Life Sciences

Seoul National University
441-744, Suwon, Korea
2Amiga Food Co. Ltd.
Janghakhoegwan Bldg. 602-ho, Daechi-dong 945-15
Kangnam-gu, 135-280, Seoul, Korea


The Combi Propol and the Pro Rapa are propolis beverages made from Eucalyptus.
The brand-new Combi Propol is a foaming agent in two types, foaming tablets and foam-
ing granules. Both have 20% of propolis. The Pro Rapa is concentrated beverage and 85%
of the component without water is propolis. Water is added to 1-5 ml of Pro Rapa for
drinking. The procedure for making the Pro Rapa is the following: wax extracts matured
for 50-60 days in ethanol, and then the extract mixed and filtered at 25-30°C. After 24
hours, filtrate was mixed with honey, wax extract, and peppermint.


Bees use propolis to build hives with wax, to polish hive walls (1, 5). They also pol-
ish combs before larva rearing, which results in black and smaller combs by successive
use. They also embalm dead animals which are too big to take out from hive. They stuff
the cracks with propolis for water-proof (2). The propolis has the bactericide effect in the
hive. Propolis is able to inhibit the germination of some seeds and a half of hemp seed in-
hibited with 111»20 dilution. The lack of air-borne flora around apiary may have something
to do with the antibiotic effect of propolis (4).
Propolis is hard and pliable at 15°C. Melting point is 60-69 DC. The color of propo-
lis is variable from yellow to dark brown and it is thermostable and partially hydrosoluble.
Alcohol or chloroform readily resolves propolis.
Main components of propolis is flavonoids (3, 9). Flavonoids works in the body in
enhancing the exchange between blood and tissues, which provide nutrients and remove

126 Woo Kun-Suk and Park Jong-Sung

residues (2). Flavonoids components contain glucides, protides, and lipids, which could
explain the possibility of its integration with physiological metabolisms (2). Flavonoids
undergo several cleavages under the intestinal flora environment and some flavonoids are
removed by urine and some by feces, and only I % of the administrated dose is absorbed
with the rest being removed unchanged (2). Vitamin P or citrine, some called this as
bioflavonoids or flavonoid derivates, works synergically with vitamin C (2).
Propolis has curative bactericidal properties, but they are very variable depending on
the bacteria studied (4). Propolis with different origins do not always have a constant anti-
biotic value. Bacteriostatic effect of propolis in vitro was tested on cultures of E. coli, ba-
cillus, entrococci, acidolactic bacteria, which are found in the large intestine or animals as
well as on staphylococcus cultures and acidophilous bacteria (6).
Propolis beverages made from Eucalyptus were produced in Korea. The Combi Pro-
pol and Pro Rapa is the beverages having propolis as a major component. The Combi Pro-
pol is tablet-type and the Pro Rapa is liquid-type, which needs to blend with water before


Questionnaires were distributed among 1,600 members during marketing seminar in

1996. Most of users have been administrated the Pro Rapa for 6-7 years. To see consum-
ers' satisfaction rates on effects of products, we distributed questions on characteristics of
the Combi Propol and the Pro Rapa.


4.1. The Combi Propol

Two kinds of the Combi Propol is selling, one is for children with chocolate flavor and
the other is for adults with propolis natural flavor. The types ofthe Combi Propol are foaming
agents, foaming tablets and foaming granules. Propolis is about 20 percent and other ingredi-
ents are cellulose, foaming agent, ketonic acid, and ~-carotene. Questionnaires were asked
about the effects on antibiotics and cell revitalization, and so forth (table 1).

4.2. The Pro Rapa

The composition of the Pro Rapa is 85% of propolis, 0.5% of peppermint, and 10%
of honey. The procedure for making the Pro Rapa is extracting and maturing propolis for

Table 1. Effects of the Combi Propoi (unit: percent)

Characteristics Excellent Very good Good Not bad

Antibiotics 97.8
Immune and anticancer activity 98.1
Cell revitalization and restoration 98.3
Strengthening veins 96.9
Propolis flavor 87.8
Consumers' satisfaction 93.8
Eucalyptus Propolis Beverages with Their Composition and Effects 127

Table 2. Effects of the Pro Rapa (unit: percent)

Characteristics Excellent Very good Good Not bad

Circulatory disease 90.6
Gastrointestinal disease 91.2
Respiratory disease 94.4
Arthritis 86.6
Dermatitis 96.1
Anti-oxidant 84.6
Anti-intlammation 99.0

5(}-{)0 days in ethanol, and then the extract mixed and filtered at 25-30 0c. After 24
hours, filtrate was mixed with honey, wax extract, and peppermint. The products were
filled in the bottles and were inspected for distribution. The Pro Rapa is liquid product
with propolis natural flavor and is administrated with water. The effects of the Pro Rapa
were arranged in table 2.


Propolis has used as a mean of anaesthetizing the pains in the operations on the ab-
domen in domestic animals without side effects (7). It also used as a antioxidant in stabi-
lizing the quality of frozen fish (8). According to Laszt (2), a successful treatment of
circulatory disease, it is necessary that the troubles should be detected in their incipient
Propolis shows a great deal of advantages in curing and preventing disease (5), but
because of propolis variability and not knowing exact way of metabolism of propolis, the
benefits need to have scientific background in further studies.

I. Bonney, R. (1995) Propolis. Bee Culture. II, 630-634.
2. Derevici, A. Contribution to the study of pro polis in: A remarkable hive product propolis. Apimondia Pub-
lishing House. Bucharest. 1978. pp.78-95
3. Derevici, A and A. Popescu and N. Popescu. Considerations on the characteristics of alcohol propolis ex-
tract in: A remarkable hive product propolis. Apimondia Publishing House. Bucharest. 1978. pp.74-78.
4. Lavie, P. The antibiotic from propolis in: A remarkable hive product propolis. Apimondia Publishing
House. Bucharest. 1978. pp.41-48.
5. Maeda G. The miracle ofpropolis( translated into Korean). Young-woo Publisher, Seoul. 1994. 115pp.
6. Palmbakha, S. E. Study of the antimicrobial effects of propolis on the gastro-intestinal microtlora in: A re-
markable hive product propolis. Apimondia Publishing House. Bucharest. 1978. pp.49-51.
7. Tsakoff, T. Study of the local anaesthetic characteristics of propolis and their effect in operations on sheep
and dogs in: A remarkable hive product propolis. Apimondia Publishing House. Bucharest. 1978. pp.67-71.
8. Ushkalova, V. N. Antioxidant value of pro polis in: A remarkable hive product propolis. Apimondia Publish-
ing House. Bucharest. 1978. pp.71-73.
9. Graham, J. E. The hive and the honeybee. Dadant & Sons Inc. 1992. pp.943-952.



K. Sorkun, S. Bozcuk, A. N. Gi:imiirgen, and F. Tekin

Hacettepe University
Science Faculty
Department of Biology
Beytepe-Ankara Tiirkiye


In the present work, the effect of five different propolis samples collected from dif-
ferent regions of Turkey «:ankiri, Aksaray, Milas-Kalemli, Giimii~hane-Kaleta~) in 1994
and from Giimii~hane-Kaleta~ in 1995 on germination percentage and cell division in the
root tips of wheat seedlings (Triticum durum Desf. cv.Kunduru) were studied. For this rea-
son, the ethanolic extracts of propolis (EEP) were prepared in four different concentra-
tions. Our findings show that, propolis solutions (EEP) depending on the concentration,
significantly inhibited the percentage of germination in comparison with distilled water
and their self alcohol controls. In the root tip cells of EEP-treated seedlings, the mitotic
cell division was also inhibited significantly with respect to distilled water controL Propo-
lis samples from Milas-Kalemli and (:ankiri regions inhibited the mitotic index signifi-
cantly as compared with their alcohol controls. The maximum inhibitions in both
germination and mitotic index were obtained with the propolis collected from (:ankiri.
There was a significant difference between the 1994-1995 Giimii~hane-Kaleta~ propolis
collections, on germination.


Propolis is a kind of bee product, collected by bees from the buds and leaves of vari-
ous plants and trees 1. Its chemical composition is very complex and has therefore at-
tracted the attention of various researchers 2.3,4. It was shown that the chemical
composition of propolis changed depending on climate, locations and years 5.6. In order to
determine the plant sources and the composition of propolis samples another research was
done 7. There are also several works showing that propolis can be used for medical pur-

130 K. Sorkun et al.

poses 8,9. Keskin et al. (1995) studied the antibacterial effect and Arkan et al. (1995) re-
ported the antifungal effect of propolis 10, 11, Besides the use of propolis in the medical and
cosmetic industries 12, there are also various studies that show the effect of propolis on
plant diseases 13, 14,15,
In the literature, it is reported that propolis has an inhibitory effect on seed germina-
tion as well as on plant growth and development 16, 17,18 , Abdou and Omar (1988) investi-
gated the effect of propolis on mitotic cell division in the root tip cells of broad beans,
barley, onion and garlic 19 , They have shown that propolis has an inhibitory effect on mi-
totic division in the root tips of the plants mentioned, Recently Sorkun and Bozcuk (1994)
showed that propolis delayed and/or inhibited the rate of germination in the seeds of some
culture plants 20 , In another research, the inhibitive effect of propolis on cell division of
human lymphocyte cells was also shown 21 ,
The aim of this study is to investigate the effects of propolis collected from different
regions of Turkey on germination and mitotic cell division in the root tip cells of wheat


2.1. Research Materials

Propolis samples obtained from hives in four different regions in Turkey were used ,
The propolis samples were collected from the <;ankiri, Aksaray, Milas-Kalemli and
Gumu~hane-Kaleta~ regions (Figure 1) in 1994 and only from the Gumu~hane-Kaleta~ re-




w cP" .. o
----'"V """"-.t....
- -'" -
<!I .' .... .".,.,
' 0
w ~
... L
- .....
<t '
•or? _/ I'



Figure t. Various locations which propolis samples were collected from Turkey,
Inhibitory Effect of Pro polis on Germination 131

gion in 1995. The samples were obtained by scraping the walls, frames and other hive
On the other hand the wheat seeds (Triticum durum Desf. cv. Kunduru) were sup-
plied by the Ministry of Agriculture'S Research Center in Ankara.

2.2. Preparation of Ethanolic Extracts of Propolis (EEP)

Propolis samples were kept in a deep freezer at -20"C for a few days. Then the hard-
ened propolis was ground by a grinder 22 and 30 g of ground propolis was dissolved in 50
ml ethanol (96%). This mixture was preserved for a couple of days in a bottle corked
tightly and kept in the incubator at 30"C. After dissolving, it was filtered twice with What-
man No:4 and No: I filter papers. The amount of alcohol which had evaporated during the
extraction process was completed up to 50 ml by adding ethanol. This solution was called
the ethanolic extracts of propolis (EEP). From this solution 30 ml was taken and com-
pleted up to I It with distilled water. This solution was then named the 30 EEP stock so-
lution 23 and kept in a refrigerator at 4"C until needed. Later on other propolis solutions at
4 different concentrations (3.75, 7.5, 15 and 20 EEP) were prepared from the 30 EEP
stock solution by adding the proper amounts of distilled water.
For the control group, 30 ml ethanol (96%) was taken and completed up to I It with
distilled water. This solution was named the alcohol stock solution and it was diluted
properly with distilled water for alcohol controls.

2.3. Germination Procedure

Several groups each having 100 wheat seeds of a medium size were selected. After
taking the dry weights of the seeds each group was soaked in one of the test solutions
mentioned earlier. At the end of 24 hours soaking time, the test solutions were drained and
surface of the seeds were gently dried with filter paper and then reweighed. The difference
between the two weights was accepted as the amount of imbibition water. The experiment
was performed using three groups.
After the imbibition period the seeds were taken to the plastic germinating boxes,
24xl4xl6 cm in size. Beforehand, a thin layer of cotton and two sheets of filter paper
were placed at the bottom of the boxes and each box was moisturized with 25 ml of the
proper test solutions . Then the imbibed seeds were placed between the two sheets of the
filter paper. The germinating boxes with lids closed were kept in a growth chamber at
25"C, in the dark for 5 days. The total percentage of germination was determined for the
five day period.
For mitotic index studies, approximately 1-1.5 cm pieces were cut from the root tips
of two day-old seedlings and fixed in the Farmer's fixative (alcohol and glacial acetic
acid, 3: 1)) and kept in a refrigerator until needed. From the root tips, the squash prepara-
tions were made in 1% acetocarmine. For each test solution 10 root tips were prepared us-
ing the squash method and in 10 different grid areas, dividing and non dividing cells were
counted for each preparation. The mitotic index was determined as a percentage of the
number of mitotic cells in relation to the number of counted cells.
The means, standard errors, ArcSinv'P transformations and analysis of variance were
calculated with an IBM computer using the Systat 5 package program. The differences
among the means were compared with LSD at a 5% level 24.
132 K. Sorkun et al.

90 o o water (ControQ
80 C Alcohol (Control)
EJ Aksaray
• Mlas -Kal
60 o Cank~ 1

50 Gum. -Kan 94
Gum-Ka~ 95
O+-----"---'----+-l-L...-......- -;-I-.L..
o water 375 75 15 20

Figure 2. The total gennination percentage of Triticum durum cv. Kunduru seeds with different propolis samples
and at different concentrations.


Effects on seed germination of 5 propolis samples collected from different locations

and each of four different concentrations are shown in Fig. 2. As can be seen clearly all
propolis samples studied, inhibited germination significantly in comparison to both dis-
tilled water and their self alcohol controls (P<0.05). Among the propolis samples; <;ankiri
was the most effective one, reducing the germination percentage most at 7.5, 15 and 20
EEP. Indeed the inhibitory effects of 15 and 20 EEP were so great that no germination had
taken place. On the other hand, Giimih;;hane-Kaleta~ 95 and Milas-Kalemli samples had
their maximum inhibitory effects at 3.75 and 20 EEP respectively.
Figure 3 represents the amount of water imbibed by the seeds during the 24 hour-
soaking period. It was shown that all propolis samples induced the entrance of imbibition

4 DOw ater ( ControQ

§ Alcohol(ControQ
o Aksaray
;: • Mlas -Kal
c:: 2.5
g a Cank~ 1
:0 2 • Gum -Kah 94
§ 1.5 • Gum-Kah 95
« 0 lSDr
o water 375 75 15 20

Figure 3. Amount of imbibition water of Triticum durum cv. Kunduru seeds soaked for 24 hours in different pro-
polis samples and at different concentrations.
Inhibitory Effect of Propolis on Germination 133

Alcohol (Control)
20 Aksaray
~ • Mlas-Kal
c 15 o <;:ankrl
:!!: • Gum·Ka~ 94
i= • Gum·Kal. 95
0 10


o waler 375 75 15 20

Figure 4. The effect of different propolis samples at different concentrations on mitotic index in the root tip cells
of Triticum durum cv. Kunduru seedlings.

water significantly at 3.75 EEP in relation to the self alcohol control (P<O.OS). Similarly at
7.5 and IS EEP, all propolis samples, except Aksaray, increased the amount of water sig-
nificantly as compared to their self alcohol controls (P<O .OS). At 20 EEP the amount of
water imbibed was still higher with <;ankiri, Giimii~hane-Kaleta~ 94 and 95 samples.
Results of mitotic index show that all propolis samples studied, reduced the mitotic
activity significantly in relation to both distilled water and their self alcohol controls (Fig-
ure 4). Among the propolis samples <;ankiri was the most effective at all concentrations
and it inhibited the mitotic index to a great extent depending on the EEP concentration.
Even at IS and 20 EEP concentrations no mitotic activity could be found with <;ankiri.
There was also no mitotic activity with the Milas-Kalemli sample at only 20 EEP concen-


In the present study, it was shown that propolis had a phytoinhibitory effect on seed
germination depending on the concentration used, location and the time collected. The in-
hibitory effect of propolis on seed germination was also shown earlier by other workers.
For example, ethanol extracts of propolis, was shown to be highly inhibitory on seed ger-
mination of Cannabis sativa (Derevici et al. 1964), Triticum vulgare and Zea mays
(Sorkun and Bozcuk, 1994) and this effect was highly dependent on propolis concentration
16.20. According to Gonnet (1968) the seedling growth of rice and lettuce was also inhibited

by ethanol extracts of propolis 18 . Furthermore, Albore (1979), Barberan et al. (1993) re-
ported that propolis collected from different locations at different times would have differ-
ent chemical compositions and therefore different phytoinhibitory effects 7. 5. All these
findings support our observations.
On the other hand, Zimonjic et al. (1989) have found that propolis at low concentra-
tion (0.3 mg/ml) inhibited the mitotic activity of human lymphocyte cells and reduced the
mitotic index by approximately 40% 2 1. Similarly Abdou and Omar (1988) have shown in
134 K. Sorkun et al.

the root tips of Vicia faba, Hordeum vulgare, Allium cepa and Allium sativum that the mi-
totic cell division is adversely affected with propolis concentrations 19. All these studies
again support our findings to a great extent.
H.owever, our results in Figure 2 and 4 indicate that there is a great similarity be-
tween the propolis effects upon seed germination and the mitotic index. This could mean
that, propolis inhibits the seed germination through the mitotic division of the embryonic
cells. On the other hand, according to our findings, propolis does not inhibit, on the con-
trary it can even increase the entrance of imbibition water (Figure 3) which is crucial at
the beginning of germination. So, the scarcity of water within the seed can not be the rea-
son for the inhibited germination. Therefore, it is possible that the propolis molecules that
enter into the seeds at the beginning of germination either playa role as an inhibitor or
cause the synthesis of some inhibitory substances within the seeds so that mitotic activity
and also germination can not take place. Further work is needed to elucidate these possi-

I. Root, A.I., "The ABC and XYZ of Bee Culture", Root Company. Medina, Ohio, U.S.A., 1972, pp.538 -539.
2. Garcia, c., Viguera, F., Ferreres, F., Barberan, T., Study of Canadian propolis by GC- MS and HPLC Pro-
polis Rcsearch, North Yorkshire YO 139 HT., 1995, pp. 12-16.
3. Bankova, v., Christov, R., Stoev, G., Popov, S., (1992) Determination of phenolics from propolis by capil-
lary gas Chromotography, Journal of Chromatography, 607: I, 150-153
4. Christov, R., Bankova, R., (1992) Gas chromatographic analysis using an electron capture detector, Journal
of Chromatography, 623: I, 182-185
5. Barberan, T., Garcia, c., Olivier, P., Ferreres F., (1993) Phytochemical Evidence for the Botanical Origin of
Tropical Propolis from Venezuela., Phytochemistry, 34--1, 191-196.
6. Omar, M. (1989) Some characteristics of pro polis from Upper Egypt. Proceedings of the Forth Int. Confer-
ence on Apiculture in Tropical Climates, Cairo, Egypt., 6-10 Nov. 1988, 1989, pp 8&--92.
7. Albore, G.R., (1979) I 'origine Geographique de la Propolis. Apidologie, I 0(3),241-267.
8. Grunberger, D., Banerjee, R., Eisinger, E.M., Oltz, L., Efros, M., Coldwell, v., Nakanishi, K., Preferential
Cytoxicity on Tumor Cells by Caffeic Asid Phenethyl Ester Isolated from Propolis. Propolis Research,
North YorkshireYO 138 HT., 1995, pp.79-86.
9. Gafar, M., Sacalus, A. M., Treatment of Common Chronic Recurring Aphthae with Propolis, A Remarkable
hive product propolis, Apimondia Publishing House-Bucharest- Romcnia, 1978, pp. 133-138.
10. Keskin, N., Hazir, S., Sorkun K., Dogan, c., (1995) Antibacterial activity of Propolis extracts, 34. Con-
gress Apimondia Lausenne- Switzerland.
II. Arkan 0., Sorkun, K., Dogan, c., GuIer, P., (1995) Mycelial forms of Morchella conica Pers. on the Nutri-
tion with pollen and propolis, 34. Congress Apimondia Lausenne-Switzerland.
12. Moise, A., Hooper, T., Encyclopedia of Beekeeping, Butler and Tenner Ltd. U.K., 1985, pp. 308--309.
13. Saeed, F.A., Mohamed, S.H., (1992) The Influence of Propolis Extracts on Soy-bean and Sunflower Wild
Disease, Assiut-lournal of Agricultured Science, 23-2, 190
14. Fahmy, F.G., Omar, M.O.M., (1989) Potential use of Pro polis to control white rot disease of onion, Assiut-
Journal of Agricultural Sciences, 20: 1,265-275.
15. Fahmy, F.G., Omar, M.O.M., (1988) Effect of Propolis extracts on certain potato viruses., Fourth Interna-
tional Congress on Agriculture in Tropical Climates, Cairo, Egypt, 6-10, November.
16. Derevici, A., Popesco, A., Popesco, N., (1964) Research on Certain Biological Properties of Pro polis., An-
nals Abeille, 7 (3), 191-200.
17. Ghisalberti, E.L., (1979) Propolis, A Review International Bee Research Assosiation, Thirtieth Annual
General Meeting, U.K. pp.75.
18. Gonnet, M., (1968) Proprietes phytoinhibitrices de quelques substances extraites de la colonie d' abeilles
(Apis mellifica L.) I. Action sur la crosissance de Lactuca sativa. AnnIs Abeille 11(1):41--47.
19. Abdou, R.F., Omar, M., (1988) Effect of the bee product propolis on the mitotic cells of Vicia faba, Hor-
deum vulgare, Allium cepa and Allium sativum, Assiut-Journal of Agricultural Sciences. 19:2, 159-170.
Inhibitory Effect of Pro polis on Germination 135

20. Sorkun, K., Bozcuk, S., Investigation of the effect of Propolis on Seed Germination of some Culture
Plants., XII. Ulusal Biyoloji Kongresi 6--8 Temmuz, Edirne- TURK lYE, 1994, pp.54--59.
21. Zimonjic, D., Giga, D., Popeskovi<;: D., (1989) Propolis Induces Decrease of Mitotic Activity but Does not
Effect SCE Frequency in Cultured Human Lymphocytes, Acta- Veterinaria, 39 (1),11-18.
22. Bianchi, E.M., (1995) The Preparation of the Tincture, The soft Extract, The Oinment, The Soap and Other
Propolis-Based products, Apiacta 3--4, 121-127.
23. Krol, w., Scheller, S., Shabi, 1., Pietsz, G., Czuba, Z., (1993) Synergistic Effect of Ethanolic Extract of
Propolis and Antibiotics on the Growth of Staphylococcus aureus. Arzneim.-Forsch.l Drug Res. 43 (I),
24. Snedecor, G.W. and Cochran, w.G. (1979) Statistical Methods, Iowa State Univ" Ames Iowa, U.S.A.


Their Structure and Secretory Products

Pierre Cassier' and Yaacov Lensky2

'Universite Pierre et Marie Curie

Laboratoire d'Evolution des Etres organises
105 boulevard Raspail, 75006 Paris, France
2The Hebrew University of Jerusalem, Faculty of Agriculture
Triwaks Bee Research Center
76100 Rehovot, Israel

The insect societies are stable structures submitted to various qualitative and quanti-
tative regulations funded on the perception of mechanical, acoustic, visual and mainly
chemical stimuli. The chemical informations provided by the biotic environment are
named allomones if they trigger a defensive behaviour, kairomone if they allow an attrac-
tive one to the emitor. The pheromones secreted by conspecific individuals, belonging or
not to the same group or society, act immediatiy on the behaviour (primer) or after a lag
time on the physiology (releaser) of the receptive insect. The blend or pheromonal bou-
quet is truly the identity card of each insect.
The honey bee (Apis mellifera L.) possess numerous exocrine glands (Fig. 1) at the
level of the buccal apparatus (mandibular, hypopharyngeal, genal, and labial glands), of the
sting apparatus (Koschewnikow glands, sheaths of the sting, setaceous membrane), of the legs
(tarsal glands), of the abdominal tergites (Nassanof or Renner glands) or of the abdominal
stemites (wax glands). Their pheromonal and volatile secretions are always mixed with non-
volatile, proteinic or glycoproteinic ones. They act separatly, or synergistically according to
multiple schemes, in the relations between the members of the society, in the regulation of
various functions : reproduction, breeding, wax building, thermoregulation, territorial de-
fence, swarming, alarm, foraging activities, caste differentiation, recruitment, ... and in the lo-
cation and the optimal working of each nutritive source. So, the society of the honey bee is
frequently compared with an homeostatic ally regulated superorganism.
The exocrine glands of the queen, drones and workers of the Honey bee belong to
types I and II as defined by Noirot and Quennedey (1). They have already been described
with some details (2, 3, 4). The structural, cytological parameters as well as the ontoge-
netic and analytical ones allow to suggest the level of the activity, the interactions and, to
a certain extend, the nature of the secretory products.

138 P. Cassier and Y. Lensky

Figure 1. The exocrine g lands of the honey bee. External and internal
views. I... VIII : tergites I to VII ; Abd : abdomen ; Ant: antenna; DG :
Dufour gland; H : head; HG : h ypopharyngeal gland; Md : mandible ;
MG : m andibular gland; NG: Nassanofgland; 0: ocelli ; Prb: probos-
ci s; SG : salivary gland: St : sting; TG : tergal glands ; Th : thorax :
WI , W2 : wings I and 2; WG . wax gland .


In the type I (Fig. 2), the epidermal and glandular cells are directly apposed to a po-
rous cuticle; enlarged pore canals and epicuticular pores allow secretory products to move
to the exterior throught the cuticular barrier. There is no intermediate cell. The secretory
product is mainly a non-proteinic one.
1. 1. The tarsal glands or glands of Arnhart (5) are located into the fifth tarsomere of each
leg of the queen, the drones and the workers (2) but the level of activity varies according
to the sex and the physiological state of the bee (Table I). They are lined by a thick pali-
sade-like epithelium (50--80 J..lm) and by a thin cuticular intima of which only the inner en-
docuticular layers (2-5 J..lm) are associated with the apical part of the glandular cells. The
outer layers edge the central reservoir. The cytoplasm contains numerous smooth and
coated vesicles, mitochondria , lipid droplets, a well developed smooth endoplasmic reticu-
lum, some cisternae of rough endoplasmic reticulum (6 , 7). The oily, colourless, alcohol
soluble secretory product is extruded through an articular slit located between the 5th tar-
somere and the arolium and forms a trail (foot-print) behind moving bees.
The study of the foot-print substance using gas chromatography and mass spec-
trometer (GC-MS) procedures allows to detect at least 120 fractions (alkanes, alkenes, al-
cohols, organic acids, esters, aldehydes, ... ) and to identified 27 compounds (C S-C 3S ; MW :
86 - 492). Eleven fractions are specific of queens, 10 of workers and 2 of drones . The oth-
ers compounds are also detected in the secretion of various exocrine glands of the honey
bee (8, 9).
The role of the foot-print secretions is only partly known. In the queen bees, the se-
cretions of the tarsal and of the mandibular glands act synergistically and inhibit the con-
struction of swarming cups and cells (10). In the workers bees, the tarsal secretions are not
directly involved in the orientation of the foragers to the hive entrance or to the food
sources; their frequently emphazised attractive power is linked to contamination by com-
ponents of the flowers, by others bees, or by nectar regurgitation. Also, they are not in-
volved in the attraction of flying drones, in the acceptance of grafted female larvae in
Exocrine Glands of the Honey Bees 139

Ep __1_'( ___ L ____ _

- ,- - - - - --
- -


Figure 2_ Type I exocrine gland. BL : basal laminae; D : desmo-

some; Ep : epicuticle; M : mitochondria; Mv : mic rovilli ; N :nucle-
us; Pc : procuticle; SD : septate desmosome.

queen cups, etc ... (6) . Likely, the «adhesive» tarsal secretion acts synergistically as «slow
release substance» trapping pheromones produced by various exocrine glands (mandibular
glands, Nassanof glands, Koschewnikow glands, tergal glands).
1. 2. The proximal parts of the sting sheaths of the honey bee workers have all the charac-
teristics of very primitive exocrine glands: porous cuticle, enlarged pore canals and epicu-
ticular pores, hypertrophied epithelial cells secreting an electron dense material. The
non-volatile part of the secretory product embedded the setae of the sheaths.The intensity
of alarm behaviour triggered by the volatile components of the Koschewnikow glands was
much lower than that elicited by sting sheaths. The volalile compounds released by both
the sting sheaths and Koschewnikow glands launch a recruiting activity and hence may
have a synergistic effect. The chemical identity of the volatile components released at the
level of the sting sheaths is always unknown (11,12 , 13).
1. 3. Of all the organs tested the setaceous membrane (tergum IX) volatiles (isoamyl ace-
tate, isoamyl alcohol, hexyl acetate, nonanol, benzyl acetate, benzyl alcohol) elicited the
strongest reactions of the guards (II, 14).
Attraction : setaceous membrane > sting sheaths > quadrate plates and Kos-
chewnikow glands > venom glands and venom sac > Koschewnikow glands > Dufour
gland and pure venom.

Table I. Secretory activity of the tarsal glands. Quantitative

data. (Lensky et coil., 1984; Hyams, 1988)

Mg / hour / bee
Categories Number n of insects x ± SD
3-day-old queens 5 0.2180 ± 0.009
6-month-old queens 8 0.6500 ± 0.051
18-24-month-old queens 8 0.4660 ± 0.038
Workers 70 0.0718 ± 0.020
Drones 70 0.0630 ± 0.015
140 P. Cassier and Y. Lensky

Alarm: sting sheaths> setaceous membrane> venom gland and venom sac.
Stinging: setaceous membrane> sting sheaths> quadrate plates and Koschewnikow
glands> venom sacs.
The setaceous membrane is directly connected to the sting sheaths and is disposed
all around the base of the sting. The fine structure of the epithelium of the setaceous mem-
brane of workers and queens is similar and does not show any characteristic of an exo-
crine gland. The flexible, untanned cuticle (4--5 !lm) rests on a flattened, inactive
epithelium (3.5-4.5 !lm). It is abundantly covered with cuticular infoldings and multi-
forked bristles which, in the vicinity of the sting sheaths are embedded in an electron
densc substance as on the sting sheaths. These cuticular expansions increase the surface of
the evaporation area; so, the outer surface of the setaceous membrane is ideally suited for
the discharge of pheromonal secretions from the sting sheaths and the Koschewnikow
glands. The function of the setaceous membrane is to serve as a platform for the discharge
of alarm pheromones that are secreted elsewhere and accumulate on its surface.
I. 4. The wax gland complex (15, 16) of the honey bee workers consists of three cell
types: epithelial cel1s, oenocytes and adipocytes, which act synergistical1y to secrete wax,
a complex mixture of hydrocarbons, fatty acids and proteins (lipophorins). The hydrocar-
bons coming from the oenocytes and the proteins (17) from the haemolymph transit across
the epithelium via the large smooth endoplasmic reticulum cisternae connected to ex-
tracellular spaces and through the mirror cuticular plates along a wel1 developed extracel-
lular and pore canal filamentous system linked to wax canal filaments of the epicuticle.
There is no excretory ducts or intermediate cel1s. The height of the epithelial cells is age-
dependent; in freshly emerged workers « one week old) the epidermis is a flat epithelium.
At the height of wax secretion (2 week-old) the lengthened cells have a fibrillated appear-
ance. In fact, the secretion of wax is a constant process even in overwintering workers.
During the ageing of the workers, the fat body cells and the oenocytes show the volumet-
ric variations similar to the epithelial cells. The wax may act as a slow release substance
for pheromones secreted by other exocrine glands. The volatile and odoriferous part of the
wax secretion may also act as a pheromonal component implies in the regulation of the
wax secretion or building?


In the type III (Fig. 3) exocrine glands (2, 3), each glandular cell shows a central
reservoir lined by the invaginated plasma membrane; this membrane forms a10 () numerous
cristae and microvilli; the content of the secretory vesicles is poured between them. The
glandular cell is caped by a tubular cell named the duct cell because it secrete, in the cen-
tral cavity, a thin duct edged by an epicuticular wall. In the reservoir of the glandular cell,
the proximal part of the duct (end-apparatus) has a porous structure permeable to the se-
cretory products. So, each glandular cell and its associated duct cell form a glandular unit.
The ducts of the glandular units are connected either directly to the external cuticle (ex. :
Nassanov, Renner or Koschewnikow glands), or, in the most developed conditions, to a
general excretory canal (ex. : mandibular glands). Glandular cells involved in the secre-
tion of non-proteinic pheromones contain numerous mitochondria and a well developed
endoplasmic. During the secretory process mitochondria undergo maturation (swelling,
disappearance of cristae and inner membrane, sticking to endoplasmic membranes, accu-
mulation of macromolecular substances), and generate secretory vesicles. Rough en-
Exocrine Glands of the Honey Bees 141

Figure 3. Type III exocrine gland. BL : basal laminae; EC : epi-

thelial cell; Ep : epicuticle; D :desmosome: DC : duct cell : GC :
glandular cell ; Mt : microtubule; MVB : multi vesicular body; N :nu-
cleus; Pc : procuticle; RER : rough endoplasmic reticulum: SD :
spetate desmosome ; Tr : tracheole.

doplasmic reticulum and Golgi apparatus are implied in the synthesis of the proteinic part
of the secretory products.
II. I. The Nassanov gland or abdominal scent gland of the workers of the honey bee (18,
19) is a complex structure where the glandular units are associated with oenocytes and fat
body cells; the respective part of these cells to the secretory activity of the Nassanof gland
is to be elucidated using labelled precursors of the pheromonal or protein components
(20). The Nassanov scent is used for orientation, particularly at the nest entrance, in
swarm clustering, at water collection sites and possibly at flowers. The proteins may act as
pheromone carriers, enzymes for pheromone degradation, slow release substances, caste
or sex specific modulators of pheromone activity; they enhance the attractiveness of the
volatile fraction. They may also form part or all of the surface cuticular proteinic reper-
toire . The pheromones produced by the Nassanov gland have been chemically identified;
they are composed of the following terpenoids : geraniol, nerolic acid, geranic acid, (E)-
citral, (Z)-citral, (E-E)-famesol, nero!. If the geraniol is the major compound, the most at-
tractive and efficient components are (E)-citral and geranic acid (18).
II. 2. The Renner's glands or tergal glands (Figs. 4-10) of the queen located near the rear
of tergites III, IV and V, (2, 21 , 22, 23, 24). The secretion of the tergal glands attracted
drones to queens on their mating flights from distances of less than 30 cm and increased
their mating activities. Queens with paraffin-coated tergites were about 16 % less attrac-
tive than queens with no coating (25). As the mandibular gland secretion, the tergal gland
products are also attractive to workers at distances of several centimeters; so, in the
queen's retinue the honey bee workers lick her with their tongues or antennates her abdo-
men. This behaviour stabilizes the court once it has been formed and likely participate in
the inhibition of the ovaries of the workers. The general organization of the Renner gland
of the queen bee is similar to the Nassanov glands of the workers although they are lo-
cated in different segments. So, the type III glandular units are intimately associated with
fat body cells and oenocytes. The coexistence of rough and smooth endoplasmic reticulum
in the glandular cells suggests that they are able to synthesize two kinds of products: the
non-volatile macromolecular protein components and the volatile, small molecular, phero-
142 P. Cassier and Y. Lensky

Figure 4. Tergal glands of the

queen bee. Scanning electron micro-
scopy. Dorsal view of a tergite (T)
and of an articular membrane (AM)
showing glandular pores (frame).

monal components. Although, Renner and Baumann (1964) and Sanford (1977) suggested
that the gland is active throughout a queen's lifetime, there is no convincing evidence con-
cerning the age-dependence of tergal gland secretion.
The various analysis of the tergal gland secretions are only preliminary ones. De Hazan
et al. (26), after wiping of the tergal gland areas, have identified by GC-MS 14 compounds
(Table II). Six compounds are specific for tergal gland secretion: benzoic acid, 9-hexade-

Figure 5. Tergal glands of the queen

bee. Scanning electron microscopy.
Detail of the articular membrane
showing cuticular scales (Sc) and
ovoid or ci rcul ar glandular pores (ar-
Exocrine Glands of the Honey Bees 143

Figure 6. Tergal glands of the queen

bee. Scanning electron microscopy.
Inner view of the cuticular mem-
brane with glandular cell (GIC),
ducts (d) connected to cuticular
holes (arrow). Oe : oenocytes.

cenoic acid, 9, 12-octadecadienoic acid, 2-dodecanol, 5-methyl-cyclo-hexanal, 2-decanal. The

others compounds are also present in other exocrine glands as follows: 1- mandibular glands
: diethyl-I ,2-benzene dicarboxylate, 3-methyl-2,6-dioxo-4-hexenoic acid, hexadecanoic acid,
tetradecanoic acid, tetratetracontane and 1,2-dodecadiene (27, 28); 2- tarsal glands: hexade-
canoic acid and 3-methyl-2,6-dioxo-4-hexenoic acid (8); 3- Koschewnikow glands of 4-day-
old queens: tetradecanoic acid and hexadecanoic acid (29). These results are not consistent
with those of Espelie et al.. (30) who prepared the extracts for GC-MS analysis from dis-
sected tergites of 1- to 4-day-old italian queen bees.

Figure 7. Tergal glands of the queen

bee. Scanning electron microscopy.
Detail of a group of glandular cells
(GlC) with their duct (d). Tergal
glands of the queen bee. The glandu-
lar unit.


Figure 8. Tergal gland of Ihe queen bee. Eleclron microscopy. General view of a glandular cell (GlC) under Ihe cUlicular inlima (CUI) of Ihe arlicular membrane. BL : basal laminae; :<
M : milochondria; N :nucleus; n : nucleolus; R : reservoir; SV : secrelory vesicle . t-
Exocrine Glands of the Honey Bees 145

Figure 9. Tergal gland of the queen

bee. Electron microscopy. Detail of
the central part of the reservoir with
the tenninal part or end apparatus of
the duct showing the perforated in-
ner epicuticle and the reticular outer
epicuticle (arrow). m : microvilli.

II . 3.The Koschewnikow glands are associated with the sting apparatus of the majority
of Apidae, Vespidae and primitive Formicidae (29, 31 , 32). In the workers and queen of
the honey bee they show the same localization and structure but glandular products and
ontogeny related to aging are different. The Koschewnikow glands are present in the 7th
abdominal segment and appear as a wide cellular mass (workers: 450 x 200 x 60 ~m;
queens: 600 x 330 x 60 ~m) organized along the intersegmental membrane joining the
quadrate plate and the spiracular plate. They are composed of about 600 type III secre-
tory units each including a large glandular cell and a narrow duct cell. The secretions ex-
truded by thin ducts, emitted on the intersegmental membrane can flow into the sting
chamber and reach the setaceous membrane. In an old, egg-laying queen the thickness of
the gland decreases to 30 ~m or less as a result of degeneration (33). The secretory cells
contain dense proteinic and lipidic granules in workers, glycoproteinic granules in
queens .
II. 3. 1. In the workers, the Koschewnikow glands secrete an alarm pheromone.
When guard worker bee are disturbed at the entrance of the hive, they extrude their stings,
expose their setaceous membrane while fanning and release alarm pheromone (34, 35)
146 P. Cassier and Y. Lensky

Figure 10. Tergal gland of the

queen bee. Electron microscopy.
Peripheral part of a glandular cell
(GIC) with numerous mitochondria
(M) and well developed smooth and
rough endoplasmic reticulum (ar-
rows). BL : basal laminae.

Table II. Chemical composition of the tergal glands

secretions of 1- to 4-day-old queens

I. Benzoic acid C7H,O,

2. 2-decanal C,oH 18 O
3. 3-methyl-2.6-dioxo-4-hexenoic acid C 7H,04
4. 1,2-dodecadiene C"H 2O
5. Cyclododecane C"H24
6. 2-dodecanol C"H 2O
7. S-methyl-cyclohexanal C,oH,oO
8. Diethyl-I ,2-benzene dicarboxylate C"H'404
9. Tetradecanoic acid C14 Hn02
10. 9-hexadecenoic acid C'6 H,OO,
II. Hexadecanoic acid C"H 32 O,
12. 9, 12-octadecadienoic acid C'8 HJ2 0 ,
13. 9-(Z)-octadecenoic acid C 18 H'40,
14. Tetratetracontane C 44 H9
Exocrine Glands of the Honey Bees 147

rich in isoamyl acetate,but containing also isoamyl alcohol, hexyl acetate, nonanol, benzyl
acetate, benzyl alcohol, etc ... (36).
II. 3. 2. The queens secrete a « stress pheromone» (Lensky et coll., 1991) which
may elicit: a.- an aggressive behaviour between two queens, following mutual detection,
which in general results in the death of one of them;b.- The balling behaviour of worker
bees towards either an introduced or one of multiple queens in the nest, except during
swarming season. Topical treatment of worker bees (<< pseudoqueens) with ethanolic ex-
tracts of queen Koschewnikow glands induced typical queen balling behaviour in workers
of a bee colony. It seems assume the maintenance of a monogynous status of each bee col-
ony. Twenty-eight compounds including acids, alcohols, alkanes and alkenes (C SHI6 -
C43 HRS ) were characterized by GC-MS in queen extract. None of them is present in worker
alarm pheromone which is secreted from worker Koschewnikow gland (29).
II. 4.The mandibular glands of the queen (25, 37) and workers (3S) of the honey bee are
paired organs connected to the basis of the mandibles where the secretory products can
flow along a grove to the spatular end. In the two castes the mandibular glands show the
same general organization: they consist of an axial cavity (2S0 x 90 11m) lined by a thin
cuticular intima (4 11m) elaborated by a flat epithelium and of a secretory epithelium ex-
clusively composed of type III glandular units. Endoplasmic reticulum and mitochondria
are preponderent organelles. The maturation of these last organelles generates an oily se-
cretory product (3S). In the drones, the reduced mandibular glands are exclusively com-
posed by the reservoir lined by a flat epithelum; there is no glandular units (39).
n. 4. I. The pheromonal activity of the mandibular glands secretion of workers has
been partially explored. A « queen substance-like» compound, the I 0-hydroxy-trans-2-
decenoic acid has been isolated (40), as well as a ketone, 2-heptanone which, according to
several authors, acts synergistically with isopentyl acetate from Koschewnikow glands as
an alarm pheromone (3). During the ageing of the workers, the size of the gland and the
amount of 2-heptanone per headspace sample progressively increase from 0.1 III at emer-
gence to 7 III in foraging bees. This increasing quantity of 2-heptanone si correlated to the
increase of hemolymphatic juvenile hormone ITT (41). Similar age-dependent changes of
2-heptanone were found in mandibular glands of workers from docile and aggressive colo-
nies and the amount of 2-heptanone in the two groups was alike. Neither crushed glands
nor 2-heptanone showed any direct effect as an alarm pheromone on guardian bees at hive
entrance. 2-heptanone had either an attractive or a repUlsive effect on gards, according to
the season. It showed a repulsive effect when added to sucrose solution which was visited
by foragers. 2-heptanone had a temporary repulsive effect on the visitation of flowers by
foraging bees, hence it seems to act as a « foraging-marking pheromone» (3S).
II. 4. 2. The pheromonal secretions of the honey bee queen's mandibular glands play
an important role in the control of social activities and in communication among the col-
ony members. The gland secretion is a complex mixture of different compounds some of
which remain to be chemically identified (2, 3, IS). The problem is a very complex one
because the compounds are progressively modified during the ageing of the queen bee
(25, 2S, 37). The most important compound, the queen substance, the 9-oxo-trans-2-de-
cenoic acid, is known since 1960. The pheromonal secretions concern: attraction of work-
ers to the queen, attraction of the drones during the nuptial flights, court formation,
inhibition of ovaries, level of foraging, inhibition of the construction of swarming cups
and cells, cohesion of the swarming group (IS). So, the queen substance is a sexual and
social pheromone and an anti-gonadotropic hormone. It acts as a releaser or as a primer.
The highest activity occurs in the 6- to IS-month-old queens; in IS- to 24-month-old
148 P. Cassier and Y. Lensky

queens the reduction of the secretory and pheromonal activities are associated with the re-
duction of the mitochondrial population as well as with the accumulation of lipid droplets
and various lytic structures.
II. 4. 3. The nature of the pheromonal secretion of the mandibular glands of the
drones is unknown but its acts synergistically with the queen substance as recruitment
pheromone during the mating flights. The secretory product appears in the mandibular
glands of 3-, 4-day-old drones. In the 7 -, 9-day-old drones the involution of the mandibu-
lar glands begins; in the Il-day-old drones the glands are reduced to the cuticular intima


Owing to the diversity of the exocrine glands and of their pheromonal blend, the
honey bees are provided with a tremendous chemical repertoire adapted to each sex, caste,
period or physiological state, function. The chemical diversity contrasts with the relative
uniformity of the glandular structures suggesting endogenous and exogenous regulations;
so, in spite of the same general organisation the secretions of the mandibular glands of the
workers and of the queens are quite different. Moreover, if it is difficult to appreciate the
exact function of each compound of the pheromonal blends, it is necessary to take also ac-
count of the temporarily variable chemical messages emitted by the flowers, the stored
products (nectar, honey, wax, propolis). These informations have according to time and
physiological state of the bees synergistic or antagonistic effects. The pheromones are
only a part of the social communication of bees; the language or dances of the bees is an
achieved mean of communication involving mechanical, chemical, chemical stimuli.

1. Noirot Ch. and Quennedey A. (1974) Fine structure of insect epidermal glands. Ann. Rev. Entomol.. 19,
2. Cassier P. and Lensky Y. (1992) Structure et role social de quelques glandes exocrines it secretion phero-
monale chez I' Abeille domestique Apis melli/era L. (Hymenoptera: Apidae). 1.- Glandes tarsales, glandes
tergales et glandes de Koschewnikow. Annee BioI., 31, 61-78.
3. Cassier P. and Lcnsky Y. (1992) Structure et role social de quelques glandes exocrines it secretion phero-
monale chez I' Abeille domestique Apis melli/era L. (Hymenoptera: Apidae). 11.- Glandes mandibulaires.
Annee BioI., 31, 78-95.
4. Lensky Y. and Cassier P. (1995) The alarm pheromones of queen and worker honey bees. Bee World, 73, 3,
5. Arnhart L. (1923) Das Krallenglied der Honigbiene. Arch. Bienenk., 5, 37-86.
6. Lenky Y., Cassier P., Finkel A., Teeshbee A., Schlesinger R., Delorme-loulie C. and and Levinsohn M.
(1984) Les glandes tarsales de l'Abeille mellifique (A pis melli/era L.) reines, ouvrieres et faux-bourdons
(Hymenoptera, Apidae). 11.- Role biologique. Ann. Sci. Nat., Zool., 18eme serie, 6, 167-175.
7. Lensky Y., Cassier P., Finkel A., Delorme-joulie C. and Levinsohn M. (l985a) The fine structure of the tar-
sal gland of the honeybee Apis melli/era L. (Hymenoptera). Cell Tissue Res., 240,153-158.
8. Finkel A. (1983) Biological effects and chemical characterization of the footprint substance of the honey-
bee (Apis melli/era). M. Sc. Thesis, Hebrew University of Jerusalem, 46 pp.
9. Lensky Y., Finkel A., Cassier P., Teeshbee A. and Schlesinger R. (1887) The tarsal glands of the honeybee
(Apis melli/era L.) queens, workers and drones. Chemical characterisation of foot-print secretions. Honey-
bee Science, 8, 97-102.
10. Lensky Y. and Siabezki Y (1981) The inhibiting effect of the queen bee (Apis mellifera L.) foot-print
pheromone on the construction of swarming queen cups. 1. Insect Physiol., 27, 313-323.
Exocrine Glands of the Honey Bees 149

II. Cassier P., Tel-Zur D. and Lensky, A. (1994) The sting sheaths of honey bee workers (Apis mellifera L.) :
structure and alarm pheromone secretion. 1. Insect Physiol., 40, 23-32.
12. Tel-Zur D. (1993). Alarm pheromones of the sting apparatus of the honey bee (Apis mellifera L. var ligus-
tica Spin.). Ph. D. Thesis, Hebrew University of Jerusalem, Israel, 1\0 p.
13. Tel-Zur D. and Lensky Y. (1995) Bioassay and apparatus for measuring the stinging response of an isolated
worker honey bee (Apis mellifera L. var /igustica Spin.). Compo Biochem. Physiol. A, in press.
14. Lenky Y., Cassier P. and Tel-Zur (1995) The setaceous membrane of the honey bee (Apis mell!fera L.)
worker's sting apparatus: structure and alarm pheromone distribution. J. Insect Physiol., 41, 7, 589-595.
15. Hepburn H. R. (1986) Honey bee and wax. Springer, Berlin, Germany.
16. Cassier P. and Lensky Y. (1995) Ultrastructure of the wax gland complex and secretion of beewax in the
worker honey bee Apis mellifera L. Apidologie, 26, 17-26.
17. Kurstjens S. P., McClain E. and Hepburn H. R. (1990) The proteins of the beewax. Naturwissenchaften, 77.
18. Winston M. (1991) The biology of the honey bee. Belknap Press, Harvard University Pres, Cambridge,
MA, 281 p.
19. Cassier P. and Lensky Y. (1994) The Nassanov gland of the workers of the honey bee (Apis mellifera L.).
Ultrastructure and behavioural function of the terpenoid and protein components. J. Insect Physiol., 40,
20. Zupko K., Sklan D. and Lensky Y. (1993) Proteins of the honey bee (Apis mellifera L.) body surface and
exocrine glands. J. Insect Physiol., 39, 41-46.
21. Renner M. and Baumann M. (1964) Uber Komplexe von subepidermalen Driisenzellen (Duftdriisen ?) der
Bienenkonigin. Naturwissenchaften, 51, 68-69.
22. Sanford M. T. (1977) An ultrastructural study of the subepidermal glands of the queen honey bee (Apis
mellifera L.). Ph. D. Thesis, University of Georgia, Athens, GA, 110 p.
23. Sanford M. T. and Dietz A. (1981) The fine structure of the subepidermal glands of the queen honey bee
(Apis mellifera L.). 28th Congo Apimondia, Acapulco, 287-292.
24. Billen J. P., Dumortier K. T. M. and Velthuis H. H. W. (1986) Plasticity of honey bee castes: occurrence of
tergal glands in workers. Naturwissenchaften, 73, 332-333.
25. De Hazan M., Lensky Y. and Cassier P. (1989). Effects of queen honeybee (Apis mellifera L.) ageing on her
attractiveness to workers. Compo Biochem. Physiol., 93A, 777-783.
26. De Hazan M., Lensky Y. and Cassier P. (1991) The chemical composition of the pheromonal secretion
from the tergal glands of the queen bee Apis mellifera L.
27. De Hazan (1986) The attractiveness, composition and structure of mandibular glands of the queen bee
(Apis mellifera L.). M. Sc. Thesis, Hebrew University of Jerusalem, 34 p.
28. Hyams Y. (1988) Characterisation of pheromonal components in the tarsal and mandibular glands of the
queen honeybee (Apis mellifera L.). M. Sc. Thesis, Hebrew University of Jerusalem, 42p.
29. Lensky Y., Cassier P., Rosa S. and Grandperrin D. (1991) Induction of balling in worker honeybees (Apis
mellifera L.) by « stress» pheromone from Koschewnikow glands of queen bees: behavioural, structural
and chemical study. Compo Biochem. Physiol.. Ser A. 100. 3, 585-594.
30. Espelie K. E., Butz V. M. and Dietz A. (1990) Dodecyl decanoate: a major component of the tergite glands
of honey bee queens (Apis mellifera L.). J. Apicult. Res., 29,15-19.
31. Grandperrin D. and Cassier P. (1983) Anatomy and ultrastructure of the Koschewnikow's glands of the
honey bee, Apis mellifera L. (Hymenoptera, Apidae). Int. 1. Insect Morphol. Embryol. 12,25-42.
32. Cassier P., Lensky Y., Rosa S., Grandperrin D., Teeshbee A. and Schlesinger R. (1990) Declenchement du
comportement d'emballage des ouvrieres de l'Abeille domestique (Apis mell!fera L.) par la pheromone de
« stress » des glandes de Koschewnikow de la Reine. Etude ethologique. structurale et chimique. Xth. Coli.
Physiol. Insectes, Toulouse (France), 19-21 septembre.
33. Grandperrin D. (1981) La glande de Koschewnikow. Origine de la pheromone du dard chez I'ouvriere de
l' Abeille (Apis mellifera L.). These 3eme cycle, Universite P. et M. Curie, Paris, 51 p.
34. Huber F. (1814) Nouvelles observations sur les Abeilles. Paschoud edit., Paris, Vol. I (362 p) et Vol. II
(479 pl.
35. Maschwitz U. W. (1964) Alarm substances and alarm behaviour in social Hymenoptera. Nature, London,
204, 324--327.
36. Mauchamp B. and Grandperrin D. (1982) Chromatographie en phase gazeuse des composes volatils des
glandes 11 pheromones des Abeilles : methode d'analyse directe. Apidologie, 13,29-37.
37. De Hazan M., Hyams Y., Lensky Y. and Cassier P. (l989b) Ultrastructure and ontogeny of the manibular
glands of the queen honeybees, Apis mellifera L. (Hymenoptera, Apidae). Int. J. Insect Morphol. Embryol.,
18, 311-320.
150 P. Cassier and Y. Lensky

38. Vallet A., Cassier P., Lensky Y. (1991) Ontogeny ofthe fine structure of the mandibular glands ofthe hon-
eybee (Apis mellifera L.) workers and the pheromonal activity of the 2-heptanone. J. Insect Physiol., 37,
II, 789--804.
39. Lensky Y., Cassier P., Notkin M., Delorme-Joulie C. and Levinsohn M. (\985) Pheromonal activity and
fine structure of the mandibular glands of honeybee drones (Apis melli/era L.) (Insecta, Hymenoptera, Api-
dae). J. Insect Phusiol., 31, 4, 265- 276.
40. Butler C. G. (1966) Mandibular gland pheromone of worker honeybees. Nature, London, 212, 530.
41. Cassier P. and Lensky Y. (\991) Evolution du titre de I'hormone juvenile III. des ecdysterordes et d'une
pheromone, la 2-heptanone, en relation avec Ie polyethisme des ouvrieres de I' Abeille domestique (Apis
melli/era L. var. ligustica ) (Hymenoptera; Apidae). C. R. Acad. Sci. Paris, 312, serie III, 343-348.



Yaacov Lenskyl and Pierre Cassicr2

ITriwaks Bee Research Center

Hebrew University, Faculty of Agriculture
76100 Rehovot, Israel
2Universite Pierre et Marie Curie
Laboratoire d Evolution des Etres organises
105 boulevard Raspail, 75006 Paris, France

Both queen and worker honey bees secrete specific alarm pheromones which trigger
different behavioural patterns in each female caste.


The society of the honeybee is basically a monogynous one. Attempts to achieve co-
habitation of multiple queens of identical or different ages in the same brood-nest, without
being confined to a cage or a compartment, were unsuccessful due to mutual aggressive
behaviour. Polygynous societies headed by multiple queens can be established when
queens are separated by a mechanical barrier, such as a queen excluder (I, 2, 3, 4, 5).
However, in the absence of a physical barrier in the brood nest, queen bees launch fatal
battles until only one survives. Virgin queens are more aggressive than mated ones (6). In-
dependently of the mutual aggressiveness of queens, worker bees also playa major role in
the elimination of supernumerary queens in experimental or feral colonies. Multiple
queens are tolerated in a bee colony only during the swarming period (7, 8,). Even follow-
ing the severing of mandibular tips and stingers of the queens to prevent mutual aggres-
sion (9) and their successful introduction into a single brood chamber of a bee colony, all
but one were balled by workers and eliminated 4 to 6 weeks later (8).
It is generally believed that pheromonal secretions from queen mandibular glands
affect the aggressive behaviour of workers towards queens (10, II). However, volatile
compounds originating from the sting apparatus or its vicinity are likely to be involved in
the mutual recognition of queens, as well as in evoking aggressive behaviour of workers
towards queens ("balling").

152 Y. Lensky and P. Cassier

Worker Koschewnikow glands (WKG) secrete alarm pheromone (12, 13), but
Koschewnikow glands of queens (QKG) may have a different function. Although Butler
and Simpson (14) attributed to the secretion of QKG an olfactory non-specific attractive
effect on workers, their function has not yet been established. The aggressive behaviour
of workers towards queens is welI documented: balled queens frequently opened their
sting chambers, protruding their stingers and sting sheath. Balling workers were attracted
to the sting sheath and licked it (15). Also, a queen's abdomen is more attractive to work-
ers than the head, presumably due to the secretion of tergal and Koschewnikow glands
It seems that some pheromonal secretions of queen bees may elicit: (a) aggressive
behaviour between two queens, following mutual detection, which in general results in the
death of one of them, and; (b) balling behaviour of worker bees towards either an intro-
duced, or one of multiple queens in the nest, except during swarming season.
The balling behaviour of workers has been explained by a "stress pheromone" hy-
pothesis : a disturbed queen produces a "stress pheromone" that stimulates an attack on
her (balling) and her own death (17, IS). However, the glandular origin and chemical
composition of the "stress pheromone" have not yet been established.
Studies were carried out to document the reaction of workers to extracts from QKG
and WKG and to induce in workers balling behaviour of a "pseudoqueen" (PQ) treated
with QKG. To avoid the effect of possible pheromonal secretions from queen mandibular,
tarsal (19) and/or tergal glands (16), we used worker bees as (PQ), following their treat-
ment with QKG extracts. To analyse the composition of QKG we used gas liquid chroma-
tography coupled with mass spectrometry. The observations of the balling behaviour of
the PQ by the sister workers were carried out in glass observation hives according to the
procedure previously described (20).

1.1. Reaction of Workers to PQ Covered with either Ethanol Extracts

of QKG or WKG Extracts

1.1.1. PQ Covered with QKG Extract. Each introduced worker pseudoqueen was im-
mediately surrounded by hive bees, who formed a dense ball around her, consisting of
about 15-35 workers. The "ball" around the PQ persisted for 5 to 10 min, after which
about two workers remained. Approximately 4 min later, they abandoned the PQ. Workers
that participated in balling of the PQ displayed aggressive behaviour, such as grasping and
biting of the wings and hind legs, as well as pulling. We also observed non-aggressive be-
haviour of workers participating in the ball, such as antennating and licking. Of 10 PQ,
seven were balled; two were aggressively attacked following their introduction, but there
was not any dense ball formation of workers; and one PQ succeeded in escaping from the
hive and flew off after being attacked by workers. None of the nine PQ that were balled
died. When the balling was terminated, the PQ could move around, but in most cases their
wings were damaged.

1.1.2. PQ Covered with WKG Extract. Immediately after introduction, about 6--S
workers approached the PQ, pushed her and arched their abdomen as during stinging,
while others only antennated her and then abandoned her. The hive workers did not dis-
play their aggressive behaviour continuously: they would approach the PQ and then re-
treat. This specific behaviour lasted for about 1.0--1.5 min.
Alarm Pheromones of the Queen and Worker Honey Bees 153

1.2. Reaction of Workers to Control-PQ

1.2.1. PQ Marked with White Paint Only. PQ marked with white paint only did not
arouse any attention inside the hive.

1.2.2. PQ Marked with White Paint and Covered with 2 III of Ethanol. PQ marked
with white paint and covered with 2 III of ethanol were inspected and antennated by about
five workers for about 30 sec.

1.3. Gas Liquid Chromatography-Mass Spectrometry (GC-MS)

Twenty-eight compounds were characterized according to their molecular weights,

which ranged from 112 to 604. The compounds (C8HI6-C43H88) included: acids, alcohols,
alkanes and alkenes. None of these compounds is present in workers' alarm pheromone,
which is produced in WKG (13).
We demonstrated by topically treating worker bees with QKG a specific balling be-
haviour toward "pseudo queen" workers similar to that displayed towards queens. Since
the workers used as pseudoqueens were removed from and introduced into their own colo-
nies, we eliminated the effect of a foreign colony odour, which could have been responsi-
ble for aggressive behaviour towards an introduced worker or queen.
The aggressive reaction of hive workers toward a worker covered with WKG extract
was similar to that of guardian bees towards an intruding foreign worker during which
alarm pheromone is released. However, this reaction di~ not resemble the behavioral pat-
tern of "balling" of a queen bee.
We observed several violent and non-violent activities towards the pseudoqueens
treated with QKG extracts, as described in the case of balled queens (18, 15). The onset of
balling of a pseudoqueen covered with QKG extract would occur almost immediately fol-
lowing her introduction into the observation hive, whereas it took about 8.5 min for the
behaviour to begin toward a foreign queen, as reported by Robinson (15). Moreover, the
relatively short balling duration of pseudo queens (7 min), as compared to that of an intact
queen (58 min, 16) may be due to the rapid evaporation rate of ethanol extracts from the
body surface of a worker versus continued secretion from the gland and its release from
the setaceous membrane in the queen bee.
One may presume that QKG secretions play an important role in the mutual detec-
tion of two queens and the subsequent duels between them. When their sting chambers
were sealed off and/or antennae were covered with wax, queens were unable to detect
each other and did not fight (8). In addition, the attraction of workers to queens can be
modified by sealing off their tergal glands and sting chamber with paraffin (16).
The pheromonal secretion of the QKG which is released on the setaceous membrane
did not elicit any behavioural pattern that is characteristic of worker alarm pheromone. It
assumes the function of a "suicide" pheromone and thus seems to be instrumental in the
maintenance of monogynous status of each bee colony: a supernumerary queen or a
pseudoqueen covered with the QKG extract is balled and may be eliminated. The phero-
monal secretion of the QKG released on the setaceous membrane did not elicit any behav-
ioural pattern that is characteristic of worker alarm pheromone. It assumes the function of
a 'suicide' pheromone and, thus, seems to be instrumental in maintaining the single-queen
status of a colony: a supernumerary queen or a pseudoqueen covered with QKG extract is
balled and may be eliminated.
154 Y. Lensky and P. Cassier

The chemical composition of pheromonal components of the WKG and QKG is en-
tirely different. We have not detected any of the alarm pheromone components of worker
bees (21, 22, 23) in QKG extracts.
We do not preclude the possibility that other queen bee pheromones, such as those
originating from the mandibular (10, II) and tergal glands act synergistically with QKG
secretions (unpublished observations).


Honey bee workers defend their colonies against many intruders that may attack
their nests to rob nectar, honey, pollen, brood or adult bees. The defence systems are es-
sential for colony survival (27,28). Guards at hive entrances intercept all bees entering the
nest and inspect them with their antennae. They are able to distinguish intruders and rob-
bing honey bees from different colonies using odour and other cues. The colony-specific
odour that adheres to the body of a bee is partly under genetic control and partly depend-
ent on the nature of food collected (26, 29).
When disturbed by an intruder, guards display a characteristic behaviour, which has
been classified into the following four stages: alerting, activating, attracting and culminat-
ing (30). During the alerting and culminating stages the guard releases alarm pheromones
either by opening the sting chamber and protruding the sting shaft, or as a result of sting-
ing. The alarm pheromones produced by glands associated with the sting apparatus tag the
intruder and attract other workers to continue stinging the attacker. The stinging reaction
can be enhanced by dark colours, rough textures and movement (1).
After a worker has thrust her stinger into an intruder's body and detached herself
from the sting apparatus, the sting continues to penetrate deeper while injecting venom
and, at the same time, releasing alarm pheromones to alert and recruit other workers.
The alarm pheromones of a worker are believed to consist of 2-heptanone secreted
by the mandibular gland and other components secreted from glands associated with the
sting apparatus.
The intensity of colony defensive behaviour is affected by the age of the workers
and colony strength, as well as by genetic and meteorological factors (31). The intensity
of the defensive reaction can be evaluated by several tests that examine the behaviour of
individual workers or groups of workers at hive entrances (31, 32). We have recently de-
veloped an apparatus and bioassay that applies only the volatile components of alarm
pheromones to stimulate either individual workers in the laboratory or groups of workers
inside observation hives or at hive entrances (19, 33, 34).
Guards and foragers produce 2-heptanone (2-H) in their mandibular glands, which
has two pheromonal functions: (a) Alarm releasing behaviour with a lower efficacy than
that of the sting apparatus (35, 36, 37, 38) ; (b) Releasing stinging behaviour as effectively
as iso-pentyl acetate (31, 39) mainly secreted by the Koschewnikow glands (13).
However, 2-H is from 20 to 70 times less potent than the alarm pheromone derived
from the stinger (36) 2-H also possesses some repellent properties affecting foraging bees
(35, 40, 41). Indeed, when 2-H was applied to alfalfa flowers, a short term repellence was
observed (42).
We studied the role of2-H (34, 43) as an alarm pheromone.
Alarm Pheromones ofthe Queen and Worker Honey Bees 155

2.1. The Changes of 2-H Levels According to the Age and Function of

2-H was detected (43) in the headspace of mandibular glands of emerging workers
(0.1 1l1/headspace); its level increased progressively with age, as did the level of juvenile
hormone III (44), and peaked in guards and foragers (7 1l1/headspace). According to pre-
vious reports, the secretion of 2-H begins only between two and three weeks after emer-
gence (35, 45, 46, 47). This discrepancy may be explained as follows: in these reports the
researchers had extracted 2-H with a solvent from crushed head capsules of workers,
whereas we used the headspace (43) of volatile substances, which were released from dis-
sected and homogenized mandibular glands prepared for GLC analysis without using any
A positive correlation between the level of 2-H and the aggressiveness of Italian,
Africanized bees and of their hybrids has been reported (47). Our analysis did not show
any significant difference between the level of 2-H of workers obtained from 'docile' Ital-
ian colonies and those from hybrid (Apis mellifera ligustica x A. m. syriaca) 'aggressive'
colonies. Moreover, the amount of 2-H produced by A m. adansonii in South Africa was
similar to that of A. m. ligustica from Canada (46). It seems that the intensity of defensive
behaviour of different bee races cannot be attributed to the amount of 2-H produced by
mandibular glands of workers.

2.2. The Role of 2-H as an Alarm Pheromone and as a Forage-Marking


Although 2-H is generally considered to be an alarm pheromone, our results did not
support this view. Even the synergistic effect of the compound on the alarm pheromones
originating from the sting apparatus is negligible (34). On the other hand, guards and fora-
gers are repelled by 2-H; for instance, it has a short-term repellent effect on foragers visit-
ing flowers of wax-flower shrubs which have been treated with 2-H. Hence, it seems to
act as a 'forage-marking pheromone'. Our results confirm previous and recent reports (48,

2.3. The Sting Apparatus and Associated Glands

The sting apparatus which serves efficiently as a mechanical and chemical defence
organ, is contained within the sting chamber between the tergal and sternal plates of the
seventh segment. Several glands and organs are associated with the sting apparatus of the
worker, which produce blends of odours used for alarm and colony defence (28) The
venom gland, venom sac and Dufour's gland have not been reported to produce phero-
mones (32). The Koschewnikow glands located on the upper part of the quadrate plates
are involved in the secretion of an alarm pheromone (32, 53) that accumulates on the
setaceous membrane (13, 37,54). Although this membrane is the source of an important
blend of pheromones released when the sting is protruded, it has not yet been established
that the membrane constitutes the glandular source of this chemical signal so charac-
teristic of disturbed bees (32). Other sting glands can also participate in the production of
sting alarm pheromones.
More than 40 compounds have been identified in extracts of the worker sting appa-
ratus and six major compounds are releasers of alarm behaviour (30, 32, 42, 55). How-
156 Y. Lensky and P. Cassier

ever, only Koschewnikow glands have been demonstrated to produce at least some of the
alarm pheromone components (12, 13). The intensity of defence reaction (alarm, attraction
and stinging) of guards stimulated with different parts of the sting apparatus decreases ac-
cording to the following order: whole sting > setaceous membrane > sting sheaths >
venom gland and venom sac> venom> Koschewnikow glands> Dufour's gland (43).

2.4. The Glands and Organs Involved in the Secretion and Distribution
of Alarm Pheromones

2.4.1. Koschewnikow Glands. The chemical composition of pheromonal components

of the WKG and QKG is entirely different. Mauchamp and Grandperrin (13) detected
isoamyl acetate, isoamyl alcohol, hexyl acetate, nonanol, benzyl acetate, benzyl alcohol
following the injection of the volatiles from WKG into a gas-liquid chromatograph with-
out using solvents.

2.4.2. Sting Sheaths. The intensity of alarm behaviour elicited by the volatile com-
ponents of the Koschewnikow glands was much lower than that elicited by sting sheaths.
The volatile compounds released by both the sting sheaths and Koschewnikow glands trig-
ger a recruiting activity and hence may have a synergistic effect. The chemical identity of
the volatile components secreted by the sting sheaths and those that accumulate on the
surface of the setaceous membrane is now being analysed by us.
Stinging reaction of individual workers in the laboratory has been recorded only fol-
lowing their stimulation with setaceous membrane or with sting sheaths, but not with other
glands or organs. Similar results were obtained with guards at hive entrances that were
'stimulated with sting glands and organs (33, 34).

2.4.3. Setaceous Membrane.

• Guards at hive entrances were exposed to the volatiles of setaceous membranes
and their attraction, alarm and stinging reactions were recorded. The results are
shown in descending order, as follows:
• Attraction: setaceous membrane > sting sheaths > quadrate plates and
Koschewnikow glands> venom glands and venom sac> Koschewnikow glands>
Dufour's gland and pure venoms.
• Alarm: sting sheaths> setaceous membrane> venom gland and venom sac.
• Stinging: setaceous membrane > sting sheaths > quadrate plates and
Koschewnikow glands> venom sacs.
Of all the organs tested, the setaceous-membrane volatiles elicited the strongest de-
fence reaction of the guards (20).
The secretions present on the surface of the setaceous membrane contained isoamyl
acetate, isoamyl alcohol, hexyl acetate, nonanol, benzyl acetate, benzyl alcohol (13). The
setaceous membrane (tergum IX) is directly connected to the sting sheaths and is disposed
all around the base of the sting. It is abundantly covered with cuticular infoldings and
multiforked bristles, which considerably increase the surface of the evaporation area.The
function of the setaceous membrane is to serve as a platform for the discharge of alarm
pheromones that are secreted elsewhere and accumulate on its surface (56).
Alarm Pheromones of the Queen and Worker Honey Bees 157


This review was prepared within the framework of the Agreement for Scientific Co-
operation between the Universite Pierre et Marie Curie (Paris VI) and the Hebrew Univer-
sity of Jerusalem.

1. Farrar e.L. (1953) Two queen colony management. Bee World 34, 189--194.
2. Loubel de I'Hoste E. (1959) La biruche. Laubet de l'Hoste, Toulouse.
3. Holzberstein 1.w. Jr (1955) Some whys and hows at two queen management. Glean. Bee Cult. 83,
4. Wafa A. K. (1956) Two queen colonies for a plentiful yield of honey. safe wintering. means of propagation
and swarming control. Bull. Fac. Agr. Cairo Univ. 98, 22.
5. Wallrebenstein W. (1955) Mein Beitrag zum Mehrmutterverfahren. 17th Int. Beekeep. Congr.
6. Darchen R. and Lensky Y. (1963) Etude preliminaire des facteurs favorisant la creation des societes poly-
gynes d' Apis mellifica var. ligustica. Ann. Abeille 6,69-73.
7. Darchen R. and Lensky Y. (1962) Les societes polygynes de Reines (Apis melli/iea var. ligustica). e. R.
Acad. Sci .. Paris, 255, 2300-2302.
8. Lensky Y.. Darchen R. and Levy R. (1970) L'aggressivite des reines entre e11es et des ouvrieres vis-a-vis
de la creation des societes polygynes d'Apis mellifica L. Rev. Compo Animal. 4, 50-62.
9. Darchen R. (1960) L'ablation du dard des reines et des ouvrieres d'Apis mellifiea. e. R. Acad. Sci., Paris,
250, 934-936.
10. Gary N.E. (1962) Antagonistic reactions of workers honeybees to mandibular gland contents of the queen
bee. Bee World 42, 14-17.
11. Yadava R.R.S. and Smith M. Y. (1971) Aggressive behavior of Apis melli/era L. workers towards intro-
duced queens II. Role of mandibular gland content of the queen and releasing aggressive behaviour. Can. 1.
Zool. 49, 1179-1183.
12. Grandperrin D. and Cassier P. (1983) Anatomy and ultrastructure of the Koschewnikow's gland of the hon-
eybee, Apis melli/era L. (Hymenoptera: Apidae). Int. 1. Insect Morphol. & Embryol. 12,25-42.
13. Mauchamp B. and Grandperrin D. (1982) Chromatographie en phase ga7euse des composes volatils des
glandes a pheromones des Abeilles: methode d'analyse directe. Apidologie 13, 29-37.
14. Butler e.G. and Simpson 1. (1965) Pheromones of the honeybee (Apis melli/era L.). An olfactory phero-
mone from the Koschewnikow gland of the queen. Ved. Pr. Wyz. Us. Vcel. Dole 4, 33-36.
15. Robinson G.E. (1984) Worker and queen honeybee behaviour during foreign queen introduction. Insectes
Soc. 31, 254-263.
16. De-Hazan M., Lensky Y. and Cassier P. (191\9) Effects of quecn honeybee (Apis melli/era L.) ageing on her
attractiveness towards workers. Compo Biochem. Physiol. 93A, 777-783.
17. Yadava R.R.S. and Smith M.Y.(1971b) Aggressive behaviour of Apis melli/era L. workers towards intro-
duced queens. 111. Relationship between the attractiveness of the queen and worker aggression. Can. J.
Zool. 49, 1359--1362.
18. Yadava R.R.S. and Smith M.Y. (J 971 c) Aggressive behavior of Apis melli/era L. workers towards intro-
duced queens I. Behavioural mechanisms involved in the release of worker aggression. Behaviour 39,
19. Cassier P., Finkel A. and Lcnsky Y (199 I) The chemical composition of tarsal gland secretions of honey-
bee, Apis melli/e.·a (L.) queens, workers and drones. (personal communication).
20. Lensky Y, Cassier P., Rosa S. and Grandperrin D. (1991) Induction of balling in worker honeybees (Apis
melli/era L.) by stress pheromone from Koschewnikow glands of queen bees: behavioural, structural and
chemical study. Compo Biochem. Physiol. 100A, 3. 585-594.
21. Boch R., Shearer D.A. and Stone B.C. (1962) Identification of isoamyl acetate as an active component in
the sting pheromone of the honeybee. Nature 195, 1018-1020.
22. Pickett J.A., Williams I.H. and Martin A.P. (1982) (2)-II-Eicoson-I-ol, an important new pheromonal com-
ponent from the sting of the honeybee, Apis melli/era L. (Hymenoptera, Apidae). J. Chern. Ecol. 8,
23. Free J.B., Ferguson A.W. and Simpkins J.R. (1988) Honeybee responses to chemical components from the
worker sting apparatus and mandibular glands in field tests. 1. Apic. Res. 28, 7-2 I.
158 Y. Lensky and P. Cassier

24. De-Hazan M. (1986) The attractiveness, composition and structure of mandibular glands of the queen hon-
eybee (Apis melli/era L.). M.Sc. Thesis. Hebrew University of Jerusalem.
25. Finkel A. (1983) Biological effects and chemical characterization of the foot-print substance in the honey-
bee (Apis melli/era L.). M.Sc. Thesis. Hebrew University of Jerusalem.
26. Hyams Y. (1988) Characterization of pheromonal components of the tarsal and mandibular glands of the
queen honeybee (Apis melli/era L.). M.Sc. Thesis. Hebrew University of Jerusalem.
27. Lensky Y, Cassier P. and Tel-Zur (1995) The setaceous membrane of the honey bee (Apis mellifera L.)
workers sting apparatus and alarm pheromone distribution. J. Insect Physiol. 21, 589-595.
28. Winston M.L. (1987) The biology of the honey bee. Harvard University Press, Cambridge, 281 pp.
29. Crane E. (1990) Bees and beekeeping: science, practice and world resources. Heinemann Newnes, Oxford,
30. Collins A.M. and Blum M.S. (1983) Alarm responses caused by newly identified compounds derived from
the honeybee sting. J. Chern. Ecol., 9, I, 57~5.
31. De Hazan M., Hyams J., Lensky Y. and Cassier P. (1989) Ultrastructure and ontogeny of the mandibular
glands of the queen honey bee, Apis mellifera L. (Hymenoptera, Apidae). Int. J. Morph. Embryol. 18, ~,
32. Blum M.S. (1992) Honey bee pheromones. In Graham, I. M.(ed.) The hive and the honey bee. Dadant &
Sons; Hamilton, pp 373-400.
33. Cassier P., Tel Zur, D. and Lensky Y. (1994) The sting sheaths of honey bee workers (Apis melli{era L.) :
structure and alarm pheromone secretion. J. Insect Physiol. 40, 233.
34. Tel-Zur D. (1993) Alarm pheromones of the sting apparatus of the honey bee (Apis mellifera L. var. ligus-
fica) PhD Thesis, Hebrew University of Jerusalem. Israel, II Op.
35. Boch R. and Shearer D.A. (1967) 2-heptanone and 10 hydroxy frans-dec-2-enoic acid in the mandibular
glands of worker honeybees of different ages. Zeitsch. Vergl. Physiol. 54, I-II
36. Boch R., Shearer D.A. and Petrasovits A. (1970) Efficacies of two alarm substances of the honey bee. J. In-
sect Physiol. 16, 17-24.
37. Maschwitz U.W (1964) Gefarhrenalarmstoffe und Gefaharenalarmierung bei sozialen Hymenopteren. Z.
Vergl. Physiol. 47, 59~55.
38. Shearer D.A. and Boch R. (1965) 2-heptanone in the mandibular gland secretion of the honey-bee. Nature
39. Free lB., Simpson l (1968) The alerting pheromones of the honey bees. Z. Vergl. Physio. 61, 361-365.
40. Melksham K.J., Jacobsen N. and Rhodes J. (1988) Compounds which affect the behaviour of the honeybee,
Apis melli{era L.: a review. Bee World 69,104-124.
41. Simpson J. (1966) Repellency of the mandibular gland scent of worker honey bees. Nature 209, 531-532.
42. Rieth J.P., Wilson WI. and Levin M.D. (1986) Repelling honeybees from insecticide-treated flowers with
2-heptanone. l Apicult. Res. 25, 78-84.
43. Vallet A., Cassier P. and Lensky Y. (1991) Ontogeny of the fine structure of the mandibular glands of the
honeybee (Apis mellifera L.) workers and the pheromonal activity of 2-heptanone. l Insect Physiol. 37,
44. Cassier P. and Lensky Y. (1991) Evolution du titre de I'hormone juvenile III, des ecdysterojdes et d'une
pheromone, la 2-heptanone, en relation avec Ie polyethisme des ouvrieres de I'abeille domestique, Apis
mellifera L. (Hymenoptera, Apididae). e. R. Acad. Sci. , Paris, III , 312, 343--348.
45. Crewe R.M. (1976) Aggressiveness of honey bees and their pheromone production. South Afr. J. Sc. 72, 7,
46. Crewe R.M. and Hastings H. (1976) Production of pheromones by workers of Apis mellifera adansonii. J.
Apicult. Res. 15, 149-154.
47. Kalmus H., Ribbands e.R. (1952) The origin of the odours by which honeybees distinguish their compan-
ions. Proc. Roy. Soc. B 140, 5Q.-59.
48. Butler e.G. (1966) Mandibular gland pheromone of worker honeybees. Nature 212, 530.
49. Free J.B., Pickett J.A., Ferguson A.W., Simpkins J.R. and Smith M. (1985) Repelling foraging honeybees
with alarm pheromones. J. Agr. Sc. 105,255--260.
50. Robertson G.E. (1968) A morphological and functional study of the venom apparatus in representatives of
some major groups of Hymenoptera. Austral. J. Zool. 16, 133--166.
51. Snodgrass R.E. (1956) Anatomy of the honey bee. Cornell University Press, Comstock Publishing Associ-
ates, Ithaca, NY, 334 pp.
52. Giufra M. and Nunez J.A. (1992) Honeybees mark with scent and reject recently visited flowers. Oecolo-
gia 89, 113-117.
53. Cassier P. and Lensky Y. (1992) Structure et role social de quelques glandes exocrines a secretion phero-
monale chez I'abeille domestique, Apis mellifera L. (Hymenoptera, Apididae). Annee BioI. 31, 61-95.
Alarm Pheromones of the Queen and Worker Honey Bees 159

54. Ghent R.L. and Gary N.E. (1962) A chemical alarm releaser in honey bee stings (Apis mellifera L.). Psy-
che, 69, I, 1--6.
55. Collins A.M. and Blum M. S. (1982) Bioassay of compounds derived from the honeybee sting. J. Chern.
Ecol. 8, 463-470.
56. Lensky Y. and Cassier P. (1995) The alarm pheromones of queen and worker honeybees. Bee World 76, 3,



Yoseph Rakover 1 and Yaacov Lensky 2

IOtorhinolaryngology Department
Central Emek Hospital
Afula, Israel
2Triwaks Bee Research Center
Hebrew University of Jerusalem, Faculty of Agriculture
Rehovot, Israel

Several "anatomical compartments" of the female honey bee body have bee studied: I. :
Exocrine compartments (the venom gland and the head glands). 2. : Internal organs (ovaries
and fat body). 3. : The haemolymph that bathes the above mentioned compartments. To ex-
plore the protein traffic from the haemolymph to the exocrine or external body compartments
immunological and electrophoretic methods were used. The studies did not show any immu-
nological identity between the proteins of larval food, venom and haemolymph but most of
the haemolymph antigens were common with those of ovaries and fat body. A tentative model
of the protein traffic between honey bee body compartments is proposed:
a. Some compartments are enveloped by a cellular layer (venom and head glands)
which prevents the macromolecular traffic between these compartments and the
b. Some other compartments are lined by a cellular layer (fat body and ovaries), al-
lowing macromolecular traffic between organs through the haemolymph.
The traffic of macromolecules between the body compartments of the female honey
bee may serve as a biological model and help in understanding the process in medical and
pharmacological disciplines.

Protein traffic between body compartments is one of the most important process in
animals to maintain the biological balance or the homeostatic status and to trigger defense

162 Y. Rakover and Y. Lensky

Postcerebral Fat
gland body

gland Oocyte

Mid-gut : Sting
Thoracic Venom Venom
gland gland sac

Figure I. Longitudinal, schematic section of a worker honey bee, showing the exocrine glands and the internal or-
gans (modified after Ribbands, 1953)(2).

Figure 2. The collection of haemolymph from the dorsal vessel of a worker with a fine glass capillary.
Protein Traffic between Body Compartments of the Female Honey Bee 163

mechanisms. So, a comprehensive study of this process is important particularly in physi-

ological, pathophysiological and medical disciplines.
In mammals, numerous investigations have been done about protein traffic between
body compartments, for example traffic between the cerebrospinal fluid and the blood (1).
Whereas in mammals the blood flows in a close system, in insects the "blood"
(haemolymph) flows in an open system, baths tissues and organs which suggests a differ-
ent model to study the protein traffic. The female honey bee, a new biological model with
a clear anatomical compartmentalization, allows to study the protein traffic (Fig. 1). So,
the protein components of three body compartments of the female honey bee are com-
pared by immunological and electrophoretic methods:
1. The exocrine glands: the venom gland, the venom sac and the sting apparatus
that secrete and drain venom; the hypopharyngeal and mandibular glands that
drain their secretions through ducts and groves via the mouthparts to the exterior
as a larval food.
2. The internal organs: ovaries and fat bodies. The vitellogenins are synthesized in
the fat body, released to the haemolymph, selectively taken up by the oocyte
through several cellular and acellular envelopes (tunica propria, external
sheaths), then deposited as vitellin platelets. The oocyte membranes are secreted
by the follicle cells (3).
3. The haemolymph that bathes the above mentioned compartments.
We examined the existence of separate body compartments with specific proteins
and the possible uptake of haemolymph proteins by the body compartments. This biologi-
cal model may help in understanding more about the protein traffic or its blocking be-
tween body compartments.


2.1 Preparation of Samples for Analyses

Italian honey bee (Apis mellifera L. vaL ligustica) adult workers were obtained from
the apiary of the Triwaks Bee Research Center, Rehovot. Haemolymph was collected
from the dorsal vessel of workers with a fine glass capillary (Fig 2.). Mandibular and hy-
popharyngeal glands were dissected from head capsules of nurse workers. Ovaries and fat
bodies were removed from laying workers. Guardians were captured at hive entrances.
The sting apparatus, venom gland and venom sac were removed fro IT their bodies. The
venom sac membrane was punctured at several sites and the venom oozed out. Royal jelly
was collected from queen cells after the removal of 3 to 4 day-old larvae.

2.2 Preparation of Antisera

Antisera were prepared in rabbits against haemolymph and venom of adult workers,
royal jelly, and ovaries oflaying queens (4, 5).

2.3 Immunological Analyses

We used double diffusion and immunoelectrophoresis methods for the immunologi-
cal analysis (5).
164 Y. Rakover and Y. Lensky

2.4 Electrophoretic Analyses

Polyacrylamide gradient slab gel electrophoresis with SDS (SDS-PAGE) and
isoelectric focusing were used for the electrophoretic analysis (4).


3.1 Are Haemolymph Proteins Incorporated into the Secretory

Products of Exocrine Glands?
To establish whether haemolymph proteins participate in the composition of protei-
naceous secretions of the exocrine glands we analyzed them using electrophoretic and im-
munological methods.

SDS-PAGE. The electrophoretic separations of the peptide components of the royal

jelly, venom and haemolymph on SDS-PAGE slabs are recorded in Fig. 3 (samples Nos. 5,
4, 3, respectively). Royal jelly was separated into 33 bands, venom into 10 bands and
haemolymph into 36 bands. Although several minor bands in the three samples shared
similar Rjs , no major haemolymph proteins were present either in royal jelly or in venom.

Isoelectric Focusing. Samples of haemolymph, royal jelly and venom were sepa-
rated on PAG-plates (Fig. 4 samples Nos 5, 4, 3, respectively). Haemolymph proteins were
separated into 29 bands within pI range of 4.0-7.2; royal jelly into 26 bands within pI
range of 4.7-9.2 and venom into 3 bands within pI above 11.0. pI values of venom frac-
tions corresponded neither to haemolymph nor to royal jelly. Several fractions of
haemolymph and royal jelly have similar pI's.
The results of electrophoretic analyses indicate that most of the protein constituents
of the three samples were not shared in common. A small number of bands had similar R,
or pI values. Which of these bands represented identical proteins is considered in the next
section, using immunological methods and specific antisera.

Immunoelectrophoresis. Samples of royal jelly, venom and haemolymph were ana-

lyzed by immunoelectrophoresis using antisera against royal jelly, venom and
haemolymph (Fig. 5). Royal jelly (well 2) formed 16 precipitation lines following absorp-
tion with antiserum against royal jelly (A) and one diffuse line with antiserum against
haemolymph (B), which seems to show a reaction of non-identity, as demonstrated below
by double-diffusion analysis. Venom (well 3) formed 6 precipitation lines with antiserum
against venom (C) and 2 lines (Nos 7 and 8) with antiserum against haemolymph (b). The
position of these two lines neither corresponded to venom nor to haemolymph precipita-
tion lines. This may indicate a reaction of partial- or non-identity. Haemolymph (well 1)
formed 24 precipitation lines with antiserum against haemolymph (B). Haemolymph pro-
teins were not precipitated by antiserum neither against royal jelly (A) nor against
venom (C).

Double Diffusion. Samples of haemolymph, royal jelly and venom were absorbed
with antiserum against royal jelly, haemolymph and venom. With the antiserum against
royal jelly (central well A) five precipitation lines were formed with royal jelly (well
No.2) and one line of non-identity, forming a spur precipitated with haemolymph (well
Protein Traffic between Body Compartments of the Female Honey Bee 165

- - -- -- - -
-- -
- -- - -- - -- •-- -
- - - -

• - -
- -


..• -•- - -
- -
- - - --

- -
Figure 3. SDS-PAGE (5- 15% gradient) protein separation of ovaries (I), fat body (2), haemolymph (3), venom
(4), royal jelly (5), ovalbumin (6), Rnase (7) and phosphorylase b (8).

No. I) (Fig. 6). With the antiserum against haemolymph (central well B) 15 precipitation
lines were formed with haemolymph (wells No. I). This antiserum precipitated neither
the antigens of royal jelly (wells No.2), nor the antigens of venom (wells No.3) (Fig.
7).Then, with the antiserum against venom (central well C) three precipitation lines were
formed with venom (well No.3), but none with haemolymph (well No.1) (Fig. 8).
The immunoelectrophoresis and double diffusion results demonstrated that no iden-
tical antigens were shared by royal jelly, venom and haemolymph. These data confirm and
166 Y. Rakover and Y. Lensky

Band 1 (-) pH





- 6.5




55 4.0
Figure 4. Isoelectric focusing of fat
body (I), ovaries (2), venom (3), royal
jelly (4) and haemolymph (5). PAG-
0 2 3 5cm
plates, pH range 3.5-9.5.

extend the results of the electrophoretic analyses. It is suggested that the head glands, the
venom glands and the haemolymph are separate compartments with regard to macro-
molecular traffic.

3.2 Are Proteins Shared in Common by Haemolymph, Fat Body and

To determine the presence of haemolymph proteins in the fat body and ovaries of
laying workers, electrophoretic and immunological methods were used.

SDS-PAGE (Fig. 3, Samples 1, 2 and 3 Respectively). Ovaries were separated into

22 bands, fat body into 21 bands and haemolymph into 36 bands. Both ovaries and fat
body shared, each, 17 bands of similar Rf s in common with haemolymph. Thirteen of
these bands were common to all three samples. Ovaries and fat body shared in common 20

Isoelectric Focusing (Fig. 4, Samples 2, 1 and 5, Respectively). Ovarian proteins

were separated into 18 bands within pI range of 4.5-7.8; fat body into 19 bands within pI
range of 4.5-7.2 and haemolymph into 29 bands within pI range 4.0-7.2.
Protein Traffic between Body Compartments of the Female Honey Bee 167

. ,.

Figure 5. Immunoelectrophoretic analysis of haemolymph (l), royal jelly (2), and venom (3) . Absorbed with an-
tisera against royal jelly (A), haemolymph (8) and venom (C).
Figure 6. Double-diffusion analysis of
haemolymph (I) and royal jelly (2) pre-
pared in the surrounding wells. Ab-
sorbed with antisera against royal jelly
(Al in the center well.

Figure 7. Double-diffusion analysis of

haemolymph (I), royal jelly (2) and
venom (3) prepared in the surrounding
wells. Absorbed with antisera against
haemolmph (B) in the center well.

Figure 8. Double-diffusion analysis

of haemolymph (I 1 and venom (3)
prepared in the surrounding wells.
Absorbed with antisera against
venom (e) in the center well.
Protein Traffic between Body Compartments of the Female Honey Bee 169

In general, the results confirmed those of SDS-PAGE, with small differences in the
number of common bands between the three samples.
The results of electrophoretic separations show that most (about 14) of the proteins of
ovaries, fat body and haemolymph are shared in common. The identity of the proteins of the
three samples was further analyzed by immunological methods using specific antisera.

Immunoelectrophoresis. The identity of haemolymph, fat body and ovaries proteins

was examined by Immunoelectrophoresis using antisera against haemolmph (B) and ova-
ries (A) (Fig. 9). Haemolymph formed 26 precipitation lines with antiserum against
haemolymph and 12 lines with antiserum against ovaries. Ovaries formed 16 lines with
antiserum against haemolymph and 7 lines with antiserum against ovaries. Fat body
formed 14 lines with antiserum against haemolymph and 7 lines with antiserum against
ovaries. It also emerges from the results that: haemolymph and fat body shared 14 lines
precipitated by anti haemolymph serum and 5 lines in common when precipitated with an-
tiserum against ovaries. Haemolymph and ovaries formed 5 common lines with anti
haemolymph serum and 6 common lines with antiserum against ovaries.

Double-Diffusion. Samples of ovaries, fat body and haemolymph were absorbed

with antisera against ovaries or haemolymph. With the antiserum against ovaries (central
well A) most of the lines formed by ovaries (well 2) and haemolymph (well I) merged, ex-
cept for line No.5 which formed a spur. The major line (Nos 2 and 3) was not formed by
the fat body (well 3) (Fig. 10). With the antiserum against haemolymph (central well B)
16 lines were formed by haemolymph (well I) which had in common with ovaries (well 2)
7 lines and 6 lines with fat body (Fig. 11).
The results of immunoelectrophoretic and double diffusion analyses confirm the
identity of several protein components of haemolymph, fat body and ovaries, which was
indicated by the electrophoretic analysis. The data suggest that contrary to the above de-
scribed separate compartments of exocrine glands, the ovaries and fat body cannot be con-
sidered as such, because of their association with haemolymph proteins.

Since neither antiserum against venom precipitated.haemolymph proteins nor anti-
haemolymph serum formed precipitates with venom, no bands with similar Rf or pI values
could be detected, it is concluded that the venom gland and the venom sac are separate com-
partments. The venom contains pharmacologically active components (6), which are lethal
whenever a honey bee is stung. When a drop of venom is removed with a glass capillary from
the sting of a donor bee and immediately injected into its body cavity, the bee dies instantly
(Rakover, unpublished observation). It seems that the venom gland and sac walls serve as a
macromolecular barrier at least with regards to the traffic of venom macromolecules. The fact
that no identical antigens were shared by royal jelly, venom and haemol-ymph indicates that
the head and venom glands are separate protein compartments from one another and from
haemolymph proteins. It seems therefore that the glandular excretory proteins are not taken
up from the haemolymph , except for their precursors, but that they are synthesized in situ.
Contrary to the exocrine gland' proteins, the immunochemical and the electro-
phoretic separations of haemolymph, fat body and ovaries revealed that most of their pro-
teins were identical. As a tentative model of protein traffic, the body compartments of the
honey bee seem composed of two main parts:
170 Y. Rakover and Y. Lensky

Figure 9. Immunoelectrophoretic anal-

ysis of haemolymph (I), ovaries (2), and
fat body (3). Absorbed with antiserum
against ovaries (A) and haemolymph

a. Compartments enveloped by a cellular layer which acts as a protein barrier and

prevents macromolecular traffic between the venom glands or the head glands
and the haemolymph.
b. Compartments lined with a cellular layer which allows a macromolecular traffic
between one organ (the fat body or the ovaries) and the haemolymph.
Separate fluid compartments have been described in the honey bee pupal molting
space ( 7) and in the queen bee spermatheca (8).
Protein Traffic between Body Compartments ofthe Female Honey Bee 171

Figure 10. Double-diffusion analysis ofhaemolymph (I) , ovaries (2) and fat body (3) charged in the surrounding
wells. Absorbed with antiserum against ovaries (A) in the center well.

Figure 11. Double-diffusion analysis of haemolymph (l), ovaries (2) and fat body (3) charged in the surrounding
wells. Absorbed with antiserum against haemolymph (B) in the center well.
172 Y. Rakover and Y. Lensky

The barrier to protein traffic in some insect tissues is related with the presence of
separate and scalariform junctions (9). In mammals, the protein traffic between compart-
ments is also composed of two main parts:
a. A barrier and prevention of macromolecule traffic by tight junction. For exam-
ple the blood brain barrier (10) or the pancreatic duct epithelium (11, 12).
b. Compartment which are lined with a cellular layer, permitting macromolecular
traffic between one organ and another such as between liver epithelial cells (13)
or in the vestibular labyrinth (14). The macromolecular\ traffic can be done via
gap junction or by receptor mediated protein transport (15).

I. Van-Deures B., Moller M., Amtorp O. (1978) Uptake of horseradish peroxidase from C.S.F into the cho-
roid plexus of the rat with special reference to transepithelial transport. Cell. Tiss. Res. 233 , 215--234
2. Ribbands R. (1953) The Behaviour and Social Life of Honey Bees. pp. 55-63; Bee Res. Assoc., London.
3. Engelmann F. (1980) Insect vitellogenin: identification, biosynthesis and role in oogenesis. [n : Advances
in [nsecs Physiology. Ed : Trehence JE., Berridge MJ. and Wigglesworth VB. Academic Press, London, 14,
4. Lensky Y, Skolnik H. (1980) [mmunochemical and electrophoretic identification of the vitellogenin pro-
teins ofthe queen bee (Apis mellifera L.). Compo Biochem. Physiol. 66B, 185-193.
5. Lensky Y., Rakover Y (1983) Separate protein body compartments of the worker honeybee (Apis mellifera
L.). Compo Biochem. Physiol. 75B: 607-615.
6. O'Connor R., Peck ML. (1978) Venom of Apidae. In: Handbook of Experimental Pharmacology. Ed: Bet-
tini S. Springer, Berlin. 48, pp. 613-615.
7. Lensky Y., Rakover Y (1972) Resorption of moulting fluid during the ecdysis of the honeybee. Compo Bio-
chern. Physiol. 41B, 521-531.
8. Lensky Y, Alumot E. (1969) Proteins in the spermathecae and haemolymph of the queen Bee. Compo Bio-
chern. Physiol. 30, 569-575.
9. Noirot-Timothee c., Noirot C. (1980) Separate and scalariformjunctions in arthropods. [nt. Rev. Cytol. 63,
10. Lane NJ. (1991) Morphology of glial blood-brain barriers. Ann. N.Y Acad. Sci. 633, 348-362.
II. Arendt T. (1991) Penetration of lanthanum through the main pancreatic duct epithelium in cats following
exposure to infected human bile. Dig. Dis. Sci. 36, 75-81.
12. Satir P., Gilula NB. (1973) The fine structure of membranes and intercellular communication in insects. In
: Ann. Rev. Entomol. Ed: Smith RF., MittlerTE.VoI18, p. 143-166.
13. Mesnil M., Asamoto M., Piccoli C., Yamasaki H. (1994) Possible molecular mechanism ofloss ofhomolo-
gous and heterologous gap junctional intercellular communication in rat liver epithelial cell lines. Cell Ad-
hes. Commun. 2, 377-384.
14. Kikuchi T., Adams JC., Paul DL.. Kimura RS. (1994) Gap junction systems in the rat vestibular labyrinth:
immunohistochemical and ultrastructural analysis. Acta Otolaryngol. (Stockh.) 114, 520-528.
15. Gitlin JD., Gitlin D. (1974) Protein binding by specific receptors on human placenta, murine placenta and
suckling murine intestine in relation to protein transport across these tissues. J. Clinic. Invest. 54,



Nuray Sahinler 1 and Osman Kaftanoglu 2

lMustafa Kemal University

Faculty of Agriculture
2Cukurova University
Faculty of Agriculture
01330 Adana-Turkey


The effects of feeding, the age of the larvae and queenlessness on the acceptance rates
and royal jelly production were studied. The average acceptance rates were 65.0±0.82 % in
queenright cell builders and 87.1±1.08 % in queenless cell builders. Feeding colonies with
pollen substitute increased the acceptance rates significantly (P<O.O 1) in queenless cell build-
ers but not in queenright cell builders (P>0.05). The age of the larvae was also important on
the acceptance of the cells. The acceptance rates of I or 2 days old larvae were higher than
that of3 days old larvae in both queenless and queenright colonies.
In queenright cell builders the average royal jelly yields were 153.7±4.27 mg per cell
when they were fed with sugar syrup and 185.3±5.68 mg when pollen substitute was given
besides sucrose syrup. In the queenless cell builders the average yields were 189.3±9.11 mg in
the sugar syrup fed and 225.6±14.52 mg in the pollen substitute fed colonies.
In general royal jelly yield was much higher in queenless cell builders than that of
qucenright. Feeding colonies with pollen substitutes in addition to sucrose syrup increased
the royal jelly yield by 36 % in queenright colonies and 40 % in queenless colonies. The
best result were obtained by grafting one day old larvae in queenless cell builders that
were fed with pollen substitute and sucrose syrup.


Royal jelly is the whitish cream like secretion from the food glands of the young
worker honey bees. It is used to feed the queen bee and young larvae in the colony. It

174 N. Sahinler and O. Kaftanoglu

shortens the developmental period of the larvae and causes the differentiation of the
worker larvae into queen. It prolongs the life of the queen, stimulates cell divisions and
promotes tissue regeneration.
There is no genetical difference between the worker bees and queen bees raised
from the same colony. The only difference comes from nutrition of the larvae during the
development. The queen larvae are fed with royal jelly throughout the larval stage, while
the older worker larvae are fed with the mixture of nectar and pollen. As a result queen
bees develop within 16 days, they are bigger, their ovaries and the spermathecae are fully
developed. They can live up to 2-3 years and can lay 1500-2000 eggs per day. On the
other hand worker bees develop within 21 days, live 35-40 days, their ovaries are not de-
veloped and their spermathecae are not functional.
Royal jelly has become a very popular bee product for the last 5--6 years. There is a
big demand for fresh and good quality royal jelly in the country. It has been used for the
healthy growth of children and for many maladies as folk medicine by elderly people, by
impotent couples, by patients suffering from cancer and other diseases in order to increase
appetite, stimulate the immune system and keep the body stronger. Most of royal jelly is
imported from China and other countries.
There are over 3.5 million colonies and 40.000 beekeepers in Turkey. We have a
great potential in producing good quality bee products such as honey, royal jelly, pollen,
propolis, bee venom etc., having rich flora, suitable climate and genetic richness of bees in
the country. However most of the beekeepers produce only honey and do not know how to
produce royal jelly. Therefore we have started a program to train beekeepers to produce
quality bee products.
Since the value of royal jelly is much higher than honey and other bee products, it is
an excellent source of income for the beekeepers. We decided to work on the factors af-
fecting the royal jelly production in order to initiate and spread the royal jelly production
in the country. We have studied the effects of feeding and the age of the larvae on the
grafting rates and royal jelly yield in queenless and queenright colonies. We are currently
studying the effects of different genotypes and season on the production of royal jelly in
subtropical climate.


This study was conducted at the Cukurova Region in Turkey, during july-September
of 1995. Total of8 cell builders were used during the experiment. One half of the colonies
were queenless and the other half were queenright. They were also divided into 2 groups
and they were either fed with syrup or with syrup and pollen substitute.
Pollen substitute was prepared by mixing 4 parts of soybean flour, I part dried skim
milk and making a cake with sugar syrup. About 250 grams of pollen substitute was
placed on top of the frames and they were replaced with the fresh one every week.
Queenless starter colonies were prepared by dequeening the colonies and rearrang-
ing the frames in the brood chamber as; honey, sealed brood, open brood, open space for
larvae transfer, open brood, sealed brood, honey and feeder. The supers and extra frames
were removed from the hives and the bees were shaken to the brood chambers in order to
have strong one-story free flying starter colonies. All the queenless cell builders were in-
spected regularly and the natural queen cells were removed. In order to strengthen the
colonies adult worker bees and/or frames of sealed brood were added to the queenless cell
Production of Royal Jelly 175

builders or new queenless colonies were prepared every 15 days. Grafting was repeated
every 2 days, and royal jelly was harvested with 48 hour intervals.
Queenright starter colonies were prepared by placing a queen excluder above the
brood chamber confining the queen and rearranging the frames as in the queenless cell
builder. Empty frames were exchanged with the brood frames between the brood cham-
bers and supers as the brood emerged.
Queen cell cups were made according to Laidlaw (I). Pure beeswax was melted in a
double-jacketed tray. The wax was dipped first into a soap-solution, and the excess water
was shaken off. The stick was then dipped into the melted wax 3 or 4 times to a depth of
about 8-10 mm. After the last dip, the formed cups were fixed to a grafting bar and sub-
merged in clean and cold water, where the cell cups were separated from the mold. By this
way 15 cell cups were prepared on a grafting bar.
Grafting was done in a tent or a room near the apiary. One drop of diluted royal jelly
was placed to the bottom of the queen cell cups when the new cell cups were used (2). The
larvae were lifted gently with some royal jelly from the cells and transferred to the bottom
of the cell cups by using a grafting needle. A clean wet towel was placed over the cell
cups in order to prevent the larvae from drying.
One frame with three bars of cells was put into the space in the cell builders as soon
as the grafting was finished. One day old larvae were grafted to the top bar, 2 days old lar-
vae to the middle bar and 3 days old larvae to the lowest bar. All the frames were removed
from the cell builders 2 days after grafting. The accepted cells were counted and the ac-
ceptance rates were determined. Later all the larvae in queen cells were removed by using
a pair of fine forceps. Royal jelly was harvested by using a special plastic royal jelly col-
lecting spoon and they were placed in a separate vial, weighed on an electronic balance
and the average royal jelly yield was determined.


3.1. Acceptance Rates

3.1.1. Queenright Cell Builders. The acceptance rates In queenright cell builders
were summarized in Table 1.
The acceptance rates in the queenright colonies were rather low. The average accep-
tance rates were 65.1±1.23 % in the syrup fed and 64.9±1.I0 % in the pollen substitutes
fed colonies. Addition of pollen substitute did not increase the acceptance rates in queen-

Table 1. Acceptancc ratcs (%) in quecnright colonies

Fed with syrup Fed with syrup and pollen substitute
Age of larvae N x± S; Max.- Min. N x±S, Max.- Min. x±S,
I Day old 150 68.7 ± 1.96 67~80 150 68.0 ± 1.61 60--73 68.4 ± 1.24 a*
2 Days old 150 66.7 ± 1.68 67~73 150 66.7 ± 1.68 60--73 66.7 ± 1.16 a
3 Days old 150 60.0 ± 1.81 53--67 150 60.0 ± 1.48 53--67 60.0 ± 1.14 b
Total/Average 450 65.1 ± 1.23 53~80 450 64.9 ± 1.10 53~73 65.0 ± 0.82
'Different letters indicate significant differences among the means (P<O.OS).
176 N. Sahinler and O. Kaftanoglu

Table 2. Acceptance rates (%) in queenless colonies

Fed with syrup Fed with syrup and pollen substitute
Age of larvae N x±S" Max.-min. N x±Sx Max.-min. x±S,
I Day old 150 88.7 ± 1.96 80--100 150 94.6 ± 1.64 87-100 91.7 ± 1.42a*
2 Days old 150 85.4 ± 2.45 73--100 150 93.9 ± 1.53 87-100 89.7± 1.42 a
3 Days old 150 77.9 ± 1.82 73--S7 150 82.1 ± 1.82 73-87 80.0 ± 1.34 b
Total/Average 450 84.0 ± 1.43**a 73--100 450 90.2 ± 1.41 **b 73--100 87.1 ± 1.08
'Different letters indicate significant differences among the means (P<O.05), "(P<O.OI).

right colonies (P>O.OS). These colonies probably collected enough fresh pollen for the de-
velopment of the larvae.
The acceptance rates oflarvae grafted at 1,2, 3 days were found to be 68.4±1.24 %
66.7±1.l6% and 60.0±1.l4 % respectively in queenright colonies. There was no signifi-
cant difference between the acceptance rates of I or 2 days old larvae (P>O.OS); however,
the acceptance rates of 3 days old larvae were lower than that of young ( I or 2 days old)
larvae (P<O.OOS).

3.1.2. Queenless Cell Builders. In general the acceptance rates in queenless colonies
were better than the queenright colonies (P<O.OS). In other words the presence of queen
lowered the acceptance rates of the grafted larvae in queen right cell builders. The accep-
tance rates in queenless cell builders were summarized in Table 2.
The average acceptance rates in queenless colonies that were fed with sugar syrup
and pollen substitutes were 84.0±1.43 % and 90.2±1.41 % respectively. Feeding the colo-
nies with pollen substitute increased the acceptance rates significantly (P<O.O I) in queen-
less cell builders.
Similarly the acceptance rates of 1 or 2 days old larvae were higher than the 3 days
old larvae in queenless cell builders. Grafting 1 or 2 days old larvae in queenless cell
builders increased the success in grafting. Chineese researchers(3) also reported that the ac-
ceptance rates were higher when younger larvae were grafted.
The acceptance rate may change with the season, condition of the colonies, avail-
ability of fresh pollen, feeding, queenlessness, genotype of the bees and other factors. Chi-
nese researchers (4) found that the acceptance rates of Italian bees, the royal jelly producing
Zau-A line and the Carpathian bees were 7S.I % , 77.2 % and 48.S %. They also indicated
that the acceptance rates changed between 62.4-94.0 % in 1989,42-89 % in 1990, and
49-96 % in 1991. The overall acceptance rate of the larvae in Anatolian bees (Apis mellif-
era anatoliaca) was similar or even higher than the Italian bees in queenless cell builders
but lower in the queenright cell builders. The acceptance rates and the royal jelly produc-
tion will probably be much higher in the spring season when there is an abundance of nec-
tar and pollen producing plants in the vicinity.

3.2. Royal Jelly Production

3.2.1. Queenright Cell Builders. The royal jelly production in queenright colonies
was summarized in Table 3.
The average royal jelly yield per cell in queenright cell builders that were fed with
sugar syrup and pollen substitute were found to be IS3.7 ± 4.27 mg. and 185.3 ± 5.68 mg.
Production of Royal Jelly 177

Table 3. Royal jelly yield per cell (mg) in queenright colonies

Royal jelly yield (mg)
Syrup Syrup + pollen substitute Average
Age of the larvae x±S, x± S, % Increase x±S;
I Day old 147.2 ± 7.53 c* 200.4 ± 11.21 a 36 173.8 ± 8.96
2 Days old 159.9 ± 7.91 b 187.7±8.91 a 17 170.9 ± 7.04
3 Days old 160.9 ± 6.85 b 167.8 ± 6.88 b 4 164.4 ± 4.79
Total/average 153.7 ± 4.27 a 185.3 ± 5.68 b 20 169.5 ± 4.08
*Different letters indicate significant differences among the means (P<0.05).

respectively. There was a 20 % increase in royal jelly production in the pollen substitute
fed colonies compared to syrup fed colonies. Feeding colonies with sugar syrup was not
enough for commercial royal jelly production. Addition of pollen substitute increased the
royal jelly yield significantly (P<0.05). Proteins in pollen substitute increased the activi-
ties of the mandibular and hypopharyngeal glands of the young worker bees to produce
more royal jelly.
The average royal jelly yield increased from 147.2 mg to 200.4 mg when I day old
larvae were grafted and the cell builders were fed with pollen substitutes in addition to
sugar syrup. The age of the larvae also affected the royal jelly yield significantly (P<0.05).
There was an average of 36 % increase in royal jelly yield when 24 hours old larvae were
grafted. Similarly, royal jelly yield increased by 17 % when 2 days old larvae were grafted
and 4 % when 3 days old larvae were grafted provided that they were fed with pollen sub-

3.2.2. Queenless Cell Builders. The royal jelly production in queenless colonies was
summarized in Table 4.
The average royal jelly yield was 189.3 ± 9.11 mg in sucrose syrup fed colonies and
225.6 ± 14.52 mg in pollen substitute fed colonies. As seen in queenright colonies addi-
tion of pollen substitute significantly (P<0.05) increased the royal jelly yield in queenless
cell builders too. The addition of pollen substitute increased royal jelly yield by 19 % in
queenless colonies. This difference was even higher (46 %) when 1 day old larvae were
grafted. More royal jelly was harvested when I day old larvae were grafted than that of 2
or 3 days old.
In general royal jelly yield depends on the genotype of the bees, condition of the cell
builders, availability of the food, season and duration of the cells in the cell builders etc.
The effects of genotype on the production of royal jelly was investigated in China and an

Table 4. Royal jelly yield per cell (mg) in queenless colonies

Royal jelly yield (mg)

Syrup Syrup + pollen substitute Average
Age of the larvae x±Si x±S; % Increase x±S;
1 Day old 194.9 ± 16.15 b 284.8 ± 27.33 a 46 239.8 ± 18.57
2 Days old 206.2 ± 20.12 b 200.0 ± 19.93 b 0 203.1 ± 13.80
3 Days old 166.8 ± 6.77 c 194.9 ± 16.56 b 17 180.8 ± 9.29
Total/average 189.3±9.11* a 225.6 ± 14.52** b 19 180.8 ± 9.29
*Different letters indicate significant differences among the means (P<0.05), **(P<O.OI).
178 N. Sahinler and O. Kaftanoglu

average of 0.375 ±0.03 gr. royal jelly was harvested from Zau A line, 0.232 ±0.03 gr. from
Carpathian bees and 0.347 ±0.06 gr. from US Italian bees during 15/5-311811992. The av-
erage royal jelly yield changed from 137.9 mg to 200.4 mg in different years when 24
hours old larvae were grafted and royal jelly was harvested at 48 hours intervals(3). The
royal jelly content increased to 327.5 mg in 1989,469.5 mg in 1980, and 346 mg in 1991
when it was harvested at 72 hours intervals. In our study royal jelly yield was less than the
results obtained from the royal jelly producing lines in China. However, there are several
races and ecotypes of honeybees in Turkey and they are different from each other both
morphologically and physiologically(S-7). There are also great variations among the eco-
types of Anatolian honeybees in terms of colony development, honey yield, wintering
ability, gentleness and disease resistance(8). The effects of different races and ecotypes on
the production of royal jelly is also under investigation.
As a result the acceptance rates and royal jelly production were much higher in
queenless colonies than queenright colonies. Feeding cell builders with pollen substitute
in addition to sugar syrup increased the royal jelly yield significantly. The best results
were obtained by grafting one day old larvae in queenless cell builders that were fed with
pollen substitute and sugar syrup.

I. Laidlaw, H.H., (1979). Contemporary Queen Rearing. Dadant and Sons Hamilton,IIlinois.
2. Giil, M.A., Kafianoglu, D. (1990) Effects of grafting techniques on the quality of queen bees (Apis mel/if-
era L.) raised under <;:ukurova Region conditions. <;:ukurova Univ. J. of Science and Engineering
4(2):41-53 (In Turkish)
3. Shibi, c., Fuhai , L., Shengming, H., Puxiu, L., Study on Relationship Between the Yield of Royal Jelly
and the age of grafted Larvae. Bee honey. Royal jelly. Environment. China 1993. P: 67-81.
4. Shibi, c., Shengming , H., Fuhai , L., Puxiu , L., Studies on the Relationship Between the Bee Races and
the Yield of Royal Jelly. Bee honey. Royal jelly. Environment.China. 1993. P:4{}-53.
5. Ruttner, F. (1988). Biogeography and taxonomy of honey bees. Springer, Verlag, Berlin. 293 pp
6. Dogaroglu, M.(l981). Tiirkiye'de yeti~tirilen onemli ari irk ve tiplerinin Cukurova Bolgesi kosullarinda
performanslarinin karsilastirilmasi. Ph.D. Thesis, C.U. Ziraat Fakultesi, Adana, Turkey. Unpublished.
7. Giiler, A., Kafianoglu, D., Bek, Y., Yeninar, H. (1996 ) Discrimination of some Anatolian honeybee (Apis
melli/era L.) races and ecotypes by using morphological characteristics. Submitted for publication.
8. Giiler, A., Kaftanoglu, D. (I 996)The performance of Anatolian honeybee (Apis melli/era L.) races and eco-
types in migratory beekeeping. Submitted for publication.



Osman Kaftanoglu' and Atilla Tanyeli 2

'Faculty of Agriculture
2Faculty of Medicine
University of Cukurova
01330 Adana, Turkey


Eight children who have malign diseases such as acute leukemia, lymphoma and
hepatoblastoma were included in this preliminary study. All the patients had I gram of
royal jelly (RJ) before breakfast once a day for one month. During this period complete
blood counts, general condition and weights of the patients were recorded. The values of
the same patients before and after the use of RJ were examined as controls. The patients
did not have GM-CSF and G-CSF while they were having RJ.
The average hematocrit values did not differ before and after the use of RJ. The av-
erage white blood cells increased from 2,857±388 to 3,732±366; neutrophils from
1,489±367 to 2,647±620; and lymphocytes from 1,137±180 to 1,665±255 after using RJ.
The general conditions (having an appetite and feeling better) increased and weight gain
were observed after RJ administrations. The patients who had RJ gained an average of
1.737±0.725 kg weight after one month.


The occurrence of different types of childhood malignancies such as leukemia, lym-

phoma and solid tumors are increasing every year due to genetic factors, viral infection,
environmental pollution and other factors. Patients have to go to the doctors and gct che-
motherapy and radiotherapy for the proper treatment.
There are mainly 3 types of blood cells in human body, namely the red blood cells,
white blood cells and thrombocytes. The majority of the blood cells are red blood cells,
and their major function is to transport oxygen from the lungs to the tissues. The percent-
age of the red blood cells is called the hematocrit and the normal hematocrit is 45 per cent
in healthy children.

180 O. Kaftanoglu and A. TanyeH

The second type of blood cells is a white blood cell or leukocyte. It is called white
because it is not colored by hemoglobin. White blood cells have several different func-
tions but the most important of these is to protect the body against foreign organisms.
Most of the white blood cells such as the neutrophils, the eosinophiles and the basophiles
are formed in the bone marrow along with the red blood cells.
The neutrophils are the most important white blood cells for protecting the body
against acute invasion by bacteria. These cells are about 12 microns in diameter and can
pass rapidly through the pores of the capillaries, enter the tissue spaces and attack almost
any agent that may be causing tissue damage. Neutrophils have the ability to ingest or
phagocytize particles that are foreign to the tissues including bacteria and tissue particles.
They contain digestive enzymes that are capable of digesting most ingested bodies and
proteins of bacterial bodies. A neutrophil can phagocytize and digest 5 to 25 bacteria be-
fore it becomes exhausted and dies.
Lymphocytes are produced from the lymph nodes and other lymph tissues instead of
bone marrow and released into the blood. They have different functions. A large propor-
tion of the lymphocytes are specially sensitized cells which are the products of the immu-
nity system of the body. These cells attack specific foreign invaders of the body such as
parasites and some cancer cells, etc.
Leukopenia is the decreased number of white blood cells in the circulation. This
usually results from damage to the bone marrow by toxic reaction to drugs, or by ionizing
radiation from x-ray or nuclear bomb exposures. When the number of white blood cells
falls extremely low the body becomes unprotected against bacterial invasion. If the person
is not treated with antibiotics or other bacteria-resisting drugs he usually will die of fulmi-
nating infection within several days.
The acute leukemia is the results of accumulation of early myeloid or lymphoid pre-
cursors in bone marrow, blood and other tissues. Acute leukemia presents with features of
bone marrow failure (anemia, infections due to leukopenia and hemorrhage due to throm-
bocytopenia) and mayor may not include features of organ infiltration by leukemic cells
(blasts). Infections are often bacterial in early stages and fungal infections are particularly
common in patients with prolonged periods of neutropenia. The disease is divided into
two main subgroups, acute myeloblastic leukemia (AML) and acute lymphoblastic leuke-
mia (ALL).
The acute lymphoid leukemia (ALL-Ll) is the most common malignancy among
children. The patients' parents usually try different folk medicines, herbs or plants and ex-
pect help from them besides medicinal treatment such as chemotherapy and radio-therapy.
Most of the patients or their parents insist on using royal jelly in order to feel better, in-
crease appetite, gain strength against bacterial and viral infections. Therefore we decided
to have a preliminary study on the effects of RJ on childhood malignancies.


Eight children between the age of 4--7 years and have malign diseases such as acute
leukemia (ALL), lymphoma (NHL) and hepatoblastoma were included in the study. They
had 1 gr. of RJ once a day before breakfast for one month. The average weight and the
general condition of the patients were observed and the hematocrit counts, average white
blood cells, neutrophils lymphocytes and thrombocytes were determined before and after
the use of RJ. The patients did not have granulocyte-macrophage colony stimulating factor
(GM-CSF) and granulocyte colony stimulating factor (G-CSF) while they were having RJ.
Royal Jelly and Treatment of Childhood Malignancies 181

Table 1. The age, sex, disease of the patients and the effects ofR] on weight and appetite

Weight Appetite
Patients Age Sex Disease Before RJ After RJ Before RJ After RJ
5.5 F ALL-LI 18.2 18.7 -,+
2 7.0 M NHL 19.6 21.0 +
3 4.0 M Hepatoblastoma 17.0 18.0 +
4 5.0 M ALL-LJ 13.7 14.9 -,+ +
5 4.0 F ALL-LJ 13.0 14.3 +
6 6.0 M NHL 19.0 23.0 -,+ +
7 7.0 M NHL 22.0 27.5 -,+ +
8 6.0 F ALL-L3 20.0 19.0 + +
Mean 5.6 17.81±1.1 19.55:!: 1.5

Royal jelly was produced in queenless cell builders which were fed with 50 % sugar
syrup and pollen substitute throughout the season. One day old larvae were grafted every
3rd day and RJ was harvested 48 hours after grafting. The larvae were lifted gently with
some RJ by using a grafting needle and transferred to the queen cell cups. A clean and wet
towel was put on the cells in order to prevent larvae from drying. As soon as the grafting
was done the bars with the cell cups were put on a frame and placed into the cell builders.
Two days after grafting they were taken from the cell builders. The larvae were re-
moved from the queen cell cups by using a pair of fine forceps and RJ was collected with
a special plastic spoon. It was either used fresh or stored in deep-freeze at _18°C.


The age, sex and the disease of the patients and the effects of RJ on appetite and
general conditions were summarized in Table 1.
As seen from the table, RJ increased the appetite in all the patients. Seven of them
gained and only one lost weight during treatment. They gained an average of 1.737±0.725
kg. weight after one month. An the parents of the patients' also indicated that RJ in-
creased the general condition of the children and they become more resistant to the viral
and bacterial infections. Consequently they have continued giving RJ to the children. The
average hematocrit values and trombocytes are summarized in Table 2.
The average hematocrit values did not differ before and after the use of RJ. How-
ever, the average trombocytes increased by 23.4 % from 188,000 to 232,000 cells per cu-
bic millimeters after using royal jelly.
The average white blood cells, neutrophils, and lymphocytes before and after RJ ap-
plications were summarized in Table 3.
The average white blood cells increased from 2,857±388 to 3,732±366; neutrophils
from 1,489±367 to 2,647±620; and lymphocytes from 1,13 7± 180 to 1,665±255 after using

Table 2. The average hematocrit values and trombocytes of the patients

Hematocrit Trombocytes
Before RJ After RJ % Increase Before RJ After RJ % Increase

X±S, 34.9 ± 1.48 35.2::1: 1.07 0.8 188,000 232,000 23.4

182 O. Kaftanoglu and A. Tanyeli

Table 3. The average white blood cells, neutrophils, and lymphocytes

x ± S,
Before RJ After RJ % Increase
White blood cells 2,857 ± 388 3,732 ± 366 30.6
Neutrophils 1,489 ± 367 2,647± 620 77.8
Leukocytes 1,137 ± 180 1,665 ± 255 46.4

RJ. In other words the white blood cells, neutrophils and leukocytes increased 30.6 %,
77.8 % and 46.4 % respectively.


Royal jelly is a very rich and nutritious bee product. There are many carbohydrates,
vitamins, proteins, lipids, sterols, essential amino acids, minerals and trace elements such
as zinc, selenium, etc. which may increase the appetite and regulate metabolism. It also
stimulates cell divisions and propagation resulting the growth of the organisms. Almost all
the patients' appetite increased and they gained weight after RJ application
Royal jelly also has a similar effects on silkworm (Bambyx mari) and other animal
species. Dietary addition of royal jelly increased and quickened the development of silk-
worms Bambyx mari (I). Afifi et al (2) indicated that royal jelly increased the growth of
guinea-pigs. The guinea-pigs were injected subcutaneously each day with RJ solution at
doses of 100, 200 or 300 mg/kg body weight and the average gain in body weight of the
animals was 136.2, 144.7, and 150.5 gr. respectively; whereas weight gain was only 119.5
gr. in water injected control group animals. The authors indicated that differences between
the control and treatment groups were highly significant.
Royal jelly and its component 10-Hydroxy deconoic acid have also strong antibacte-
rial effects(3) . The bactericidal effects of 10-Hydroxy deconoic acid was also confirmed by
other researchers (4). Fujiwara et al (5) found a potent antibacterial protein in RJ and they
named it as royalisin. It was stated that in bioassays royalisin had a potent antibacterial ac-
tivity against Gram-positive bacteria at low concentrations, but not against Gram-negative
bacteria. The antibacterial and antiviral properties of RJ may have prevented the patients
from getting bacterial and viral infections. All the patients did not have any cold or other
infections during the observations.
Royal jelly has anti-tumor effects(6) and it is effective against slowly growing tumors
(e.g. solid tumors) but not against rapidly growing ones (e.g. leukemia L 1210 or P 388).
It also stimulates the immunoglobulin production by the lymphocytes and increases the
IgM and IgG in patients with breast cancer(7) .
Because of all these positive properties RJ has been widely used by the patients who
are suffering from childhood malignancies, any type of cancer and other disorders. It is a
rich, supplementary bee product that increases the appetite, keeps the organism stronger,
stimulates the production of white blood cells and prevents the patients from getting bac-
terial, fungal or viral infections during chemotherapy and radio-therapy. It stores the liver
and kidney functions and lessens the adverse effects of the chemicals on the organs during
chemical treatments. However it should be kept in mind that RJ by itself is not a drug or
medicine to cure any type of malignancies therefore the patients should not only rely on
RJ. It may also cause allergic reactions especially on the asthmatic patients. They should
Royal Jelly and Treatment of Childhood Malignancies 183

have allergy test before taking RJ. As a result of this preliminary study we plan to have
more patients to investigate the RJ effect for patients who have cancer and conduct double
blind experiments to eliminate the placebo effect.

I. Saikatsu, S., Ikeno, K., Hanada, Y, Ikeno, T. (1989) Physiologically active substances in oral secretions
produced by the honey bee- effects of royal jelly on the silkworm. Ohu Univ. Dental J. 61(3):1-4 (In Japa-
2. Afifi, E. A., Khattab, M.M., EI-Berry, A.A., Abdel-Gawaad, A.A. (1989) Effect of royal jelly on guinea-pig
growth. Proceedings of the Fourth International Conference on Apiculture in Tropical Climates, Cairo,
Egypt, ~IO Nov. 1988.
3. Blum. M.S., Novak, A.F.. Taber. S. (1959). 10-hydroxy-2-decenoic acid. an antibiotic found in royal jelly.
Science 130(3373):452-453
4. Serra Bonvehi, J., Escola Jorda, R. (1991) Studie iiber die mikrobiologischc qualitat und bakteriostatische
aktivitat des weiselfuttersaftes (Gelee Royale):Beeinflussung durch organische sauren. Deutsche Le-
bcnsmittel-Rundschau 87(8):25~259
5. Fujiwara, S .. Imai, 1.. Fujiwara. M .• Yaeshima. T.. Kawashima. T.. Kobayashi. K. (1990) A potent antibac-
terial protein in royal jelly. Purification and determination of the primary structure of royalisin. J. BioI.
Chern. 265( 19): 11333-11337
6. Tamura. T., Fujii, A., Kuboyama, N. (1987) Anti-tumor effects of royal jelly. Nippon Yakurigaku Zasshi
89(2)73-80 (In Japanese)
7. Yamada, K., Ikede. I., Maeda. M .• Shirahata. S .• Murakami. H. (1990) Effect of immunoglobulin produc-
tion stimulating factors in foodstuffs on immunoglobulin production of human lymphocytes. Agricultural
and Biological Chemistry 54(4): I 087-1089.



Eli Zlotkin

Department of Cell and Animal Biology

Institute of Life Sciences
The Hebrew University of Jerusalem
Jerusalem 91904, Israel

Venom is defined as a mixture of substances which are produced in specialized
glandular tissues in the body of the venomous animal and introduced by the aid of a sting-
ing-piercing apparatus into the body of its prey or opponent in order to paralyze and/or
kill it. The academic interests in the topic of venoms (Toxinology) is substantially based
on considerations concerning (a) public health, dealing with clinical pathological prob-
lems of human envenomation and (b) pharmacology-neuropharmacology, which views
venoms as a potential source of useful substances for medicine, industry and biological re-
search [1].
Our approach to venoms is directed by eco-zoological and ecochemical considera-
tions which view a venom mechanism as an unique specialization and adaptation of the
venomous animal in order to solve a certain vital problem. In the present article hymenop-
terous venoms are compared to those of other organisms on the basis of the role that they
fulfill in nature.
With this background and from the ecological ecochemical point of view the vast
majority of the venomous animals (such as snakes, many spiders, venomous snails, and
various coelenterates) are slow, and even static, predators which feed on freshly killed
prey comprised of mobile and relatively vigorous animals. Such a drastic difference in the
locomotory capacity of the venomous predator when compared to its potential prey de-
mands special adaptations for prey encounter. These include, firstly, behavioural adapta-
tions in the form of ambush hunting tactics and, secondly, the development of the
venomous apparatus for the quick immobilization of the prey at the earliest moment after
encounter and with relatively low doses. The latter is expressed in the nanomolar or lower
range of concentrations of the active substances in the body of the affected animal.
The goal of fast action at low doses is achieved with the aid of two devices. The first
is a physical device, the stinging instrument, enabling the direct introduction of the venom

186 E. Zlotkin

into the circulation or in close proximity to the critical excitable nervous and neuro-mus-
cular target tissues which control and operate the animals locomotory system. The second
is a chemical device in the form of the neuroactive constituents of venom, the so-called
In this presentation a neurotoxin is defined as a substance which affects the function
of the excitable tissues due to a specific recognition and binding affinity to given sites in
these tissues [2]. The various neurotoxins so far obtained from various animal venoms are
commonly classified according to their effects and sites of action in nervous systems.
They are thus defined as ion channel toxins which modify ion conductance, presynaptic
toxins which affect neurotransmitter release, and postsynaptic toxins which interfere with
the binding and the resulting expression of neurotransmitters [1 ,3]. A very common char-
acteristic of the various biologically active venom constituents, including neurotoxins, is
their protein-polypeptide nature. This chemical characteristic has a double significance.
Firstly, polypeptides, through their high diversity of covalent structures and the resulting
spatial arrangements, may reveal a highly diverse array of functional specificities such as
site-directed and selective neurotoxicity. Secondly, polypeptide is the most readily avail-
able structure for adaptive modifications through the genetic machinery.
To summarize, the venomous animal appears as a slow obligatory predator which
feeds on fast animals. The hymenopterous insects indeed possess venoms which satisfy
the above definition but they are neither slow nor predatory.


Fig. I presents the systematic definition and allocation of the hymenopterous in-
sects. The Hymenoptera comprise a major order of holometabolous insects. More than
100,000 species of this order are know [4] the majority of which can be easily identified
by the characteristic "wasp waist" and the two pairs of well developed transparent wings.
As shown, (Fig. I), the sting possessing hymonopterans (Apocryta) are subdivided
into the terebrants (parasitic wasps) and aculates which are further divided into solitary
and social hymenoptera.
In the different groups of Hymenoptera the venomous apparatus is a modified ovi-
positor and exhibits a fundamental similarity of structure [5~8, (Fig.2)]. The sting is hol-
low and composed of a dorsally located stylet and two barbed lancets which form the
ventrolateral boundaries of the ventral venom canal. The two lancets possess grooves
along their entire length which fit to two guide-rails of the stylet. The sting is protruded
and set in motion by a complex arrangement of levers consisting of three pairs of plates
and their specific musculature. During the action of stinging the lancets slide along their
track-and-groove connection to the stylet and drive thc sting shaft deeper into the wound.
The liquid injected through the sting originates basically from two main secretory entities:
the branched venom gland and the "dufour" gland - the so-called "acid" and "alkaline"
glands respectively in bees and social wasps. The venom gland is drained into the venom
sac connected by a duct to the stylet bulb successively leading to the ventral venom canal.
With this background several peculiarities related to the different division of Hymenoptera
may be mentioned. In Aculeata, the ovipositor is converted into a sting and the egg pas-
sage opens to the exterior at its base. In Terebrantia the ovipositor retains its egg-laying
function [5,9]. In honeybees the venom sac is devoid of any musculature. Venom flow is
attributed to the pumping function of the back-and-forth movement of the lancets coupled
to a successive valve action of lobes at the lancet bases [7]. The sting loss of honeybee
ubphylum 1\1 'DIB LATA
Class ECT

ubclass Ii LOMET ABO LA

uborder ~
(stmgles )
uborder APO RYT
(wuh a ling)


paraslllc wasps
(e\ample Braconldae

olitar wasp
(example phecldae

oel3l Hymenoptera

Fam ~I------
(ants) <"a psI (bee,)

Figure I. The systematic allocation of the hymenopterous insects.





~ \/ conal

Figure 2. Structure of a sting apparatus of a worker Apis mellifera. Taken from [46].
188 E. Zlotkin

workers when encountering vertebrates is attributed to specific and purposeful mecha-

nisms absent in other groups of social Hymenoptera. The venom apparatus of the terebrant
parasitic wasps is characterized by a multiplicity of glands, the exact function of which
has not been clarified so far [11]. In ants (Formicidae) several subfamilies have deviated
morphologically from the above general pattern. The subfamilies Formicinae and
Dolichorinae have lost the major portion of the sting shaft as well as associated sclerites.
In Dolichorinae the function of the poison gland reservoir was replaced by anal glands
which synthesize a variety of low molecular weight organic compounds and are devoid of
protein components [12-13].


In both the aculeate and the terebrant groups of solitary wasps the venom system is
primarily employed for securing food for their offspring. This basic function, however, is
performed with different patterns of prey encounter and stinging behaviour. The relatively
large aculeate forms hunt prey which roughly equals their own size; the terebrants encoun-
ter a prey which is in the range of about 3 orders of magnitude larger than their own size
such as fifth instar larvae of the wax moth encountered by the Bracon parasitic wasps
[14]. For details on the biology of, and prey predator interactions of, solitary wasps the
reader is directed to the references [8,9] and [38].
The Sphecidae are the most intensively studied among the aculeate solitary wasps
[16,15]. In most cases the prey, composed of insects and spiders, is more or less deeply
paralysed by the injection of the venom. Once the wasp has grasped the prey with its legs
it immediately tries to insert the sting. Wasps which prey on Hymenoptera or adult
holometabolous insects typically apply only one sting, through either the thin interseg-
mental membranes on the base of the legs or the central integument between head and
prothorax. Sphecids preying on Orthoptera or larval insects particularly caterpillars, typi-
cally sting several times. The insertion of the sting is guided by tactile stimuli [17] sensed
by mechanoreceptors located on the stings sheaths.
In the group of terebrant parasitic solitary wasps the venom system of Braconidae
has received the greatest attention [11]. Parasitism by Braconidae may be either internal or
external. The former occurs in free-living hosts (such as beetles, caterpillars, aphids) and
the latter predominates in hosts that live in confined quarters (such as digging caterpillars
and wood-boring coleoptera). In contrast to the external parasitoids the internal ones sel-
dom, even temporarily, paralyse their free-living hosts. The external parasitoids may be
subdivided into those which induce a temporary paralysis and others which induce a per-
manent one. The latter were studied in detail through the venomous system of Braconidae
(Bracon hebe tor, Bracon brevicornis and Bracon gelechiae [11,14]. Bracon wasps require
for their development lepidopterous larvae such as Ephestia, Plodia or Galleria. The fe-
male wasp paralyses its host by injecting venom through its stinging apparatus and feeds
upon the haemolymph released from the wound made in the host integument. The male
wasp does not feed on the host's haemolymph. Oviposition is independent of the stinging
and feeding behaviour, the eggs being deposited on, or very near, the paralysed larvae. Af-
ter leaving the egg, the wasp larvae cling to the host, and they, too, feed upon the
haemolymph by puncturing the integument [14].
The venoms of social Hymenoptera, mainly those of honeybees and social wasps,
have attracted attention due to their medical significnce, either harmful [18,19] or benefi-
cial [20--22]. It is commonly accepted that the venoms of social Hymenoptera are primar-
Role of Hymenopterous Venoms in Nature 189

ily employed for defensive purposes [11,23,24,12]. This concept is supported by the fol-
lowing considerations:
1. According to their feeding habits the Apidae are herbivorous (nourished by nec-
tar and pollen) wasps and ants are omnivorous and do not consume living prey.
2. Bees and social wasps are exclusively diurnal and are warningly coloured, pos-
sessing the yellow and black stripes typical for other aposematic animals.
3. The nests of social Hymenoptera are static, immobile stores of food (including
their progeny) which has to be defended. The venom is supposed to serve as the
most effective device for defence [12].
4. The stinging behaviour of social Hymenoptera is often performed in a commu-
nal, organized manner associated with volatile low molecular alarm, recruitment
pheromones [25-27] such as isoamyl acetate in honeybees [28,29] or aliphatic
ketones in Vespa orientalis [30]. Such an organized and mobilized defence was
shown to occur in honeybees against mammals [18] as well as insects (Vespa
mandarinia). Adapted patterns of a mobilized active defensive behaviour were
also demonstrated to exist in social wasps [26,33]. While stinging, the wasps,
like bees, grip the victim tightly with their legs and mandibles. Wasps and hor-
nets try to sting the enemy several times in succession. Unlike bees, there is usu-
ally no sting autonomy in Vespidae.
5. The latter in bees was suggested to serve as a specific adaptation developed by
selection which is directed against the largest enemies of the honeybee, the
honey- and brood- mammals and birds. This is expressed in a preformed break-
ing point, strong barbs in the lancets, and the inclusion of the last ganglion in
the ripped-out sting enabling further ejection of venom. Several modes of attack
behaviour also imply a defensive specialization against mammals [10,26,27].
6. A convincing argument in favour of the defensive role of hymenopterous ven-
oms is supplied by their composition and pharmacology. The most characteristic
property of a defensive venom was suggested to be its pain-producing (algo-
genic) capacity. In this aspect, hymenopterous venoms possess an efficient array
of pain-producing agents [33,24,12].
Since this presentation deals with the ecological ecochemical implications of
hymenopterous venoms rather than the, more conventional, pharmacological clinical as-
pects the interested reader can find complementary information in references [38-47).


Social wasps, however, also employ their stings for offensive purposes related to
food-collecting. They are optional, occasional food hunters. Insects serve as the main tar-
get of this activity. Caterpillars, mantids, bees, bugs, grasshoppers, beetles, butterflies and
moths were observed to be taken into the nests of wasps and hornets as food [34-37,19].
Systematic, organized invasions of Vespa mandarinia on to Apis mellifera bees in Japan
were observed and described [31], ending with the seizure of pupae and brook as food.
Some of the Vespa species in Japan are highly predatory on Polistes wasps and attack
their nets, mainly after emergence of the young males and queens. V. mandarinia attacks
Vespula and Vespa species usually in the same manner as it attacks bees [37,35].
The chemistry of vespid venoms, generally, favours the concept of a defensive role.
This is expressed in the presence of pain-producing biogenic amines (such as serotonine,
190 E. Zlotkin

histamine, dopamine, etc.) and wasp kinins [40]. Internal release of histamine in the body
of the envenomated animal is induced by mast cell degranulation performed by a series of
short, strongly basic, peptides such as the MCD-peptide in bees [49], "Mastoparan" in
wasp venoms [50,56,57,58,60], basic kinins [51]] and finally by lytic peptides such as the
amphipathic (detergent-like) melittin of bees [52,53] and similar lytic peptides from ven-
oms of wasps [52] and ants [54]. The effects of these agents are strengthened by phos-
pholipase A2 which interacts synergically with lytic factors, and is known to occur in
hymenopterous venoms [52,55]. Non proteinaceous algogenic factors have been shown to
occur in formicid ant venoms in the form of formic and piperidine alkaloids [24].
With this background and the above-mentioned information on the predaceous activ-
ity of social wasps, certain data concerning the neurotoxicity of wasp venoms is of inter-
est. It was shown [59] that the venoms of Vespa mandarinia, Vespa xanthoptera and
Vespa analis insularis, when assayed in a crustacean neuromuscular preparation, sup-
pressed both the excitatory and inhibitory postsynaptic potential (EPSPs and IPSPs). A
component of the venom separated by Sephadex G-50 chromatography caused the depres-
sion of EPSPs while IPSPs were unaltered. Since no appreciable changes were found in
the potential, conductance and sensitivity of the postsynaptic membrane, the above block-
age of the excitatory transmission may be attributed to a presynaptic effect [59].
The venom of Vespa mandarinia has served as the source for the isolation of a
neurotoxin which blocked (1000{, M) the action potential in a lobster nerve through the re-
duction of sodium current. This tetrodotoxin-like action was induced by a polypeptide of
molecular weight about 20 kD with a pI of 9.1 and devoid of enzymatic or haemolytic ac-
tivity [60]. As shown in Fig. 3, the purification of the neurotoxin was performed by two
steps of gel filtration and cation exchange column chromatography. As shown in Fig. 4,
the suppressory presynaptic origin of the toxicity was suggested by the ability of the toxin
to block the excitatory junction potential in lobsters skeletal muscle without affecting the
muscles resting potential [60]. An induction of presynaptic block of synaptic transmission
in the insect CNS was recently [61] demonstrated by a wasp glycosylated peptide, vespu-
lakin in 1 and its synthetic analogues. All analogues have different effects on synaptic
transmission in the cockroach sixth abdominal ganglion namely: at first, a direct and re-
versible block of excitatory cholinergic nicotinic transmission with a concurrent activation
of the inhibitory GABA-ergic system and, secondly, a delayed irreversible block of the
transmission [61].
To summarize, the venoms of the omnivorous Vespidae possess neurotoxic polypep-
tides which seem to fulfill a role in prey paralysis, similar to the "classical" venom sys-
tems (see above).


In the solitary aculeate wasps the venoms of Sphecidae have received much atten-
tion, mainly through the study of the venom of Philanthus triangulum. Field observations
and prey records for many species of sphecid wasps suggest that their venoms affect a
large number of insect species from different orders, as well as spiders. In spite of the fact
that P. triangulum preys exclusively upon honeybees it may paralyse, in practice, all other
insects so far tested by forced stinging, and these include 37 genera from 15 families from
seven orders, as well as spiders [8,16]. Philanthus venom induces a flaccid paralysis
which is substantially due to a blockage of neuromuscular transmission and does not af-
Role of Hymenopterous Venoms in Nature 191

os A

c o

Of. --'
o 03
~ 02

f.0 80 120 .1&0

Fract i on No

004 OB
.0 ,,
Gi , -
"vc ,,
0 , " '0
G, , , , 04~
0 02 , H
JJ a b U
<l Z


00 0
Froe tlon No

Figure 3. Purification of a neuTtoxin from hornet venom. Upper: Column chromatography on Sephadex G-50 of
Vespa mandarinia venom supernatant. 69 ml of hornet venom supernatant was applied to a column (i.d. 4.1 X 98
cm) of Sephadex G-50 equilibrated with 50 mM acetate buffer. pH 5.6. Each fraction collected was at 15mLltube,
and the flow rate was 189 mLih. Toxic activity appeared in peak B (fractions 43-60). The fractions were pooled
and concentrated for further purification. (a) is an eluting position of chymotrypsinogen Aon the same column.
lower: CM-Sephadex column chromatography of the peak B obtained after the Sephadex G-50 chromatography.
The concentrated sample (ca. 10 ml) was applied 10 a column (i.d. 2 X 33 cm) of M-Sephadex equilibrated with
50 mM acetate buffer, pH 5.6. Elution was performed with a rigid line~r gradient of NaCI from 0 to 0.8 M. Frac-
tions of 4 ml were collected at a flow rate of 30 mLih. The toxic activity was found on peak H, and the active
fractions (fractions 50-55) were combined and then concentrated for further studies. The inset is polyacrylam ide
disc gel electrophoresis of MDT X in J3-alanine-glycine buffer at pH 4.3(a) and in NaDodS0 4 with reduced condi-
tions (b).

fect the visceral musculature [66,9]. In addition, the venom of Philanthus is able to eluci-
date Post-synaptic as well as central [67,68] neurotoxic effects.
The post synaptic effects of Philanthus venom have attracted attention. A toxic frac-
tion [62,63 ] was examined on the methathoracic extensor tibiae locust muscles accompa-
nied by an iontophoretic application of L-glutamate, the excitatory neurotransmitter of
insect motor nerves [64]. The study of Clark et a\. [69] has revealed that the toxic fraction
192 E. Zlotkin

conI. 4 9 59 66
(0 I
J'- ~'- _.,-- --''-- ____ [ 5mV
Ep'p lOOms
o ..
4 00 ('J 0
o 0

3 o
noon 0
0 nO 0
00 0
o 0

2 o oonC'o °00 cPo co l'J.o"'Q."Io">fhco O:,!fIiO (.J

Em mV

[ -50
..... .,,. ..... -.-....--.......... _• ... ~_ 111._"'· -.... _111il10 l1li _ ••• ..- ....

o 10 20 30 40 50 60 70 min

Figure 4. Effect of Vespa mandarinia toxin (MDTX) on neuromuscular junctions of lobster walking leg. Ordinate:
Peak amplitudes of excitatory postsynaptic potentials (0) and the resting membrane potential (C.) in the stretcher
muscle. Abscissa: Time after applying 0.1 fig of MDTX at 0 min. Taken from [60).

blocks the locust muscle glumate receptors following their activation. The fraction was
shown to suppress both the iontophoretic glutamate potential (Fig. 5) and the excitatory
junction potentials in a glutamate receptor activation dependent manner. The rate of re-
covery from the blocking effects was reduced by activating the muscle either by ionto-
phoretic application of the neurotransmitter or by electrical stimulation of the motor
nerve. Complementary assays by single channels by patch clamp analyses have supported
the conclusion that the Philanthus fraction blocks open channels gated by both junctional
and extrajunctional glutamate receptors on locust muscle.

Figure S. Effect of Philanthus toxic fraction on L-glumate potential evoked by glutamate iontophoresis at exci-
tatory neuromuscular junctions of locust extensor tibiae muscle . i. saline containing 0.1 V im I of Philanthus
toxic fraction was introduced (i) during repetitive applications of 1.5 nC iontophoric glutamate doses. Note the
reduction in response amplitude. After 1.5 min
exposure to Philanthus toxic fraction toxin-
free saline was introduced (,(.) ; ii , following 10
min wash with toxin-free saline. the response
was restored to control amplitude. Saline con-
taining 0.1 V /ml was reintroduced, but ionto-
phoretic application of glutamate was discon-
tinued for I min and then resumed. The
amplitude of the first response at the end of
(ii) 205
the I min period was nearly identical to the
control amplitude (cf.I). Note the decline in
amplitude of response to subsequent glutamate
doses. Measurements were made at the resting
potential of the muscle fibre , -60 m V . Calibra-
tion bars: potential (vertical), 4 mY; times
(horizontal), 20s.
Role of Hymenopterous Venoms in Nature 193

l .U



ii H

U. ~

S 10 IS 20 2S

/ PTX-433


E 1. 0

D .•
0 .'

0 .4


• , • 10 12 14

Figure 6. Fractionation of Philanthus venom by reverse-phase HPLe. (A) Frationation of lyophilized venom
glands, extracted with 50% CH 3-CN in water; 450 ~l (representing extracts of 225 wasps) was chromatographed
on a YMC-ODS 20 X 280 mM column and developed with a linear aqueous gradient of 5% CH 3CN/0.l %
CF 3COOH to 95% CH 3CN/0.l % CFjCOOH for 30 min at a flow rate of 8 mllmin. UV absorption was monitored
at 215 nm. (8) Fractionation of the main toxic fraction (hatched peak in A). Taken from [70).

The chemical identification of the factor responsible for the above open channel
blocking was achieved by Eldefrawi et al. [70] by two steps of chromatography on a re-
versed phase column by HPLC system (Fig. 6). The resulting toxin (PTX) was shown to
serve as a potent antagonist of transmission at quisqualate-sensitive glutamate synapses of
locust leg muscle . This philanthotoxin (PTX-433) has been purified, chemically charac-
terized, and subsequently synthesized along with two closely related analogues. It has a
butyrylltyrosyllspermine sequence and a molecular weight of 433 (Fig. 7). Its two ana-
194 E. Zlotkin


1 PTX-433 ~~N~N~N~NH2 H 0: H H


2 PTX-334 BuTyr·N~N~N~NH2

3 PTX-343 BuTyr·N~N~N~NH2

Figure 7. The chemical structures of the natural philanthotoxin, PTX-433, and two isomers PTX-334 and PTX-
343. Taken from [70].

logues, PTX-343 and PTX-334 (the numerals denoting the number of methylenes between
the amino groups of the spermine moiety), are also active on the glutamate synapse of the
locust leg muscle; PTX-334 was more potent and PTX-343 was less potent than the natu-
ral toxin. Such chemicals are useful for studying, labeling, and purifying glutamate recep-
tors and may become models for an additional class of therapeutic drugs and possibly



Parasitism by hymenopteran species is a complex phenomenon, affecting the behav-

ior, physiology and development of its host insect. The adult female parasitoid has the ca-
pability of regulating the host for the advantage of her offspring in order to provide a
suitable source of nutrition and dwelling. Ectoparasitoids often rely on the venomous ma-
terial to regulate their hosts. Venomous substances of ectoparasitoid often suppress the
hosts host cellular defense reaction [810, inhibit its growth [82] or arrest its larval-larval
molting [83].
The venom of Braconidae served as the major object of study [11]. Braconid and
other terebrant parasitic wasps appear to be incapable of stinging vertebrate animals. Their
selective toxicity to insects was demonstrated by the following series of evidence:
(1) Host preference and host sensitivity assays indicated a unique selectivity of bra-
conid venoms to lepidopterous insects [71]. The unusu!!l selectivity of Bracon (=Micro-
bracon=Habrobracon) hebetor venom to insects was demonstrated by its ability to
distinguish not only between related groups of insects but also between species belonging
to the same order [71].
Role of Hymenopterous Venoms in Nature 195

(2) A second, however indirect, indication of specificity follows from the high toxic-
ity of the authentic venom to the susceptible species. Beard [14] reported that one part of
Bracon hebetor venom in 2 x 10 8 parts of Galleria mellon ella haemolymph was sufficient
to cause permanent paralysis. Bracon brevicornis venom in a dose of 26.8 x 10~10 ml also
caused 100% paralysis in Galleria, Angasta and Plodia larvae [72].
(3) An additional indication of specificity is supplied by the unique pharmacology of
the Bracon venom. It was shown [73-76] that the flaccid paralysis induced by this venom
is a consequence of (a) a presynaptic blockage of the excitatory glutaminergic transmis-
sion in the insect skeletal neuromuscular junction; (b) the normal persistence of the inhibi-
tory (GAB Anergic) neuromuscular transmission, which is not affected by the venom, and
contributes to the former effect an additional, so-called, autopharmacological intoxication.
The heart and visceral musculature are not affected [14,11], thus enabling the continuous
prolonged survival of the paralysed prey.
(4) The final, and most convincing, proof of the insect selectivity of the braconid
venoms is supplied by a series of neurophysiological studies with non-insect preparations.
Rathmayer and Walther [77] report that they could not find any activity of Bracon hebetor
venom on neuromuscular transmission of a spider, crab and crayfish. The uniqueness of
the pharmacological selectivity of Bracon venom is further emphasized in light of the fact
that it has been almost established that L-glutamate is the excitatory neurotransmitter in
crustacean as well as in insect neuromuscular junctions. No effect of the Bracon venom
has been found on the cholinergic neuromuscular transmission of vertebrates such as in
frog nerve muscle [78] and rat diaphragm [76] preparations.
The presynaptic blocking action of the Bracon venom is exemplified by the
neuromuscular examination of the effect of two active fractions A and B isolated by col-
umn chromatography from the crude venom (Fig. 8). The elementary effect comprises the
suppression of the evoked junction potentials which is accompanied by the reduction in
the frequency of the miniature excitatory post synaptic potentials (m.e.p.s.p) without af-
fecting their amplitude and the muscle resting potential. These data suggest a presynaptic
suppression in the release of the excitatory neurotransmitter.
The purification and characterization of insecticide toxins from venom glands of the
parasitic wasp Bracon hebetor was recently reported by Quistad et al [80]. The potency of
venom from Bracon hebetor against lepidopterous larvae has been known for over 40
years, but previous attempts to purify and characterize individual protein toxins have been
largely unsuccessful. Three protein toxins were purified from venom of this small para-
sitic wasp and the amino acid sequences of 22-31 consecutive residues at the amino-ter-
minus were determined. These relatively large toxins (apparent molecular mass 73 kDa)
were labile under many isolation techniques, but anion-exchange chromatography allowed
purification with retention of biological activity. Two purified toxins were quite insectici-
dal (LD50 < 0.3 microgram/g) when injected into six species of lepidopterous larvae. On a
molar basis, one toxin (Brh-l) has the highest known biocidal activity against Heliothis
virescens larvae(LD50 = 2 pmol/g).


The clinical and therapeutic aspects of the honey bee venom are extensively studied
[38,46]. However, when dealt from the ecological point of view, the Apinae are strictly
herbivorous animals and the venom serves exclusively for defensive purposes [44]. The
196 E. Zlotkin



preparation ® ct"


2q fr .gllen~y
.----... m.e.p.s p.-
0 .... .. 0 amplitude of evoked pot .
~ e - - e m ,e.p.s.p. - ilmp l itude ~
'> ~-- _ .... rest,ng membrane pot.
u Figure 8. Neuromuscular effects of the two
'0 100 paralysing fractions (A and) obtained from a

OJ homogenate of the wasp Microbracon hebetor.
Fractions A and B possessed 50 and 80 paralys-
c 80 ing units per mI . respectively, and their effects
on resting membrane potential, the amplitude
OJ of evoked junction potentials and the frequency
60 and amplitude of the spontaneous miniature ex-
citatory postsynaptic potential (MEPSP) in a
flight muscle fibre of Pieris brassicae was
D monitored. The parameters are plotted as a per-
centage of the control values. The mode of ac-
20 tion of the two fractions is identical. Both
fractions depress the amplitude of the evoked
junction potentials as well as the frequency of
0 the MEPSPs without affecting the amplitude of
the MEPSPs, indicating a presynaptic blocking
effect. Taken from [79] .

latter aspect is strongly supported by the algogenic (pain producing) capacity of bee
The composition of bee venom is presented in Table I. Pain production by bee and
wasp venom was attributed substantially to biogenic amines and the histamine releasing
peptides [44 and above]. However, the study of Prince et al. [86] reveals that the pain pro-
ducing factor of bee venom is Melittin and offers an interesting explanation of its algo-
genic mode of action.
On the average a single bee sting contains 0.5 /-II of volume and 50 /-Ig of dry matter.
Subcutaneous injections of 2 /-II volume of the separate bee venom components, in physi-
ological saline at their respective doses as in a sting, have shown that only melittin yields
the sharp pain associated with a honey bee sting.
As shown in Fig. 9, melittin is an amphipathic peptide of 26 amino acids composed
mainly of hydrophobic amino acid residues with a positively charged polar C-terminal
segment (Fig. 9). The amphipathicity of melittin is supplied not only by its primary struc-
ture but also by its spatial arrangement. In the latter the hydrophobic residues are oriented
to one side through which melittin interacts with the membranal phosopholipids [85] .
Role of Hymenopterous Venoms in Nature 197

Table 1. The major components of honeybee (Apis mellifera) venom

Class of molecule Component % of venom Molecular weight
Proteins Hyaluronidase 10-12 41000
Phospholipase A, 20000
Peptides Melittin 50 12000 (as tetramer)
Secapin 0.5--2.0 3000
MCD peptide 1-2 2500
Tertiapin 0.1 2500
Apamin 1-3 2000
Procamine 1-2 600
Small peptides (less than 5 a.a) 13-15 <600
Physiologically active amines Histamine 0.5--2.0 150
Dopamine 0.2-1.0 150
Noradrenaline 0.1...{J.5 150
a-aminobutyric acid 0.5 150
Sugars glucoseFructose 2 180
Phospholipids 5 700
a-amino acids I 150
Volatile compounds (pheromones) 4--8 200
Taken from [46J.

When interacting with membranes in the absence of a membrane potential melittin ap-
pears in its monomeric form. However, the presence of a membrane potential causes the
conductance of the melittin-incorporated bilayer to increase. At a fixed voltage the con-
ductance increases with a fourth power of the melittin concentrations suggesting that four
melittin monomers associate to form a tetrameric anion selective pore.
The speculation of Prince et al [84] attributes the pain producing capacity of mel itt in
to its voltage dependent ionophoric properties. The mammalian skin is supplied nocicep-
tors which are afferent sensory nerve fibers which are excited by various mechanical,
physical and chemical stimuli. The nociceptor maintains a resting potential (+70 mV out),

5 10
(A) Gly-lie-Gly-Ala -Val-Leu-Lys-Val-Leu- Thr- Thr-Gly-Leu-

(B) lie Ala

(C) lie Ser

15 20 25
(A) - Pro-Ala-Leu-lIe-Ser- Trp-lie-Ly s-Arg-Ly s-Arg-Gln-Gln -NH
(B) Thr Asn Lys

(C) Glu

Figure 9. Amino acid sequence of melittin (A) from Apis melli/era and A. cerana. Variant residues in melittins
from A. florea and A. dorsata shown in (B) and (C), respecitvely. Taken from [86].
198 E.Zlotkin

so when melittin molecules encounter nociceptors they form their tetrameric pores, col-
lapse the resting potential and thus trigger a wave of depolarization down the nerve. The
transient depolarization results in the melittin tetramer dissociation to the inactive mono-
mers. When the nerve cell membrane repolarizes the melittin reaggregates into its iono-
phoric tetramer which in turn depolarizes the membrane again. The above oscilatory mode
of repolarization and depolarization induces a chronic stimulation of the nerve cell result-
ing in pain. If this model is correct the relatively short period of pain following a bee sting
may be caused by the diffu~ion and escape of the individual melittin molecules.

1. Shier, WT. and Mebs, D. (Eds.) Handbook of Toxinology, Marcel Dekker, N.Y., 1990.
2. Zlotkin. E. In: Comprehensive Insect Physiology, BiochemistlY and Pharmacology (Kerkut. E.A. and Gil-
bert, L.I., Eds.) Vol. 10, Pergamon, Oxford. 1985, pp. 499-546.
3. Hucho, Fand Ovchinnikov, Y.A. (Eds.) Toxins as Tools in Neurochemistry, de Gruyter, Berlin, 1983.
4. Berland, L.B. Superordre de Hymenopteroides. In: Traite de Zoologie, Vol. X(2). Masson, Paris, 1951.
5. Imms, A.D. A General Textbook of Entomology, Methuen, London, 1951.
6. Snodgrass, R.S. Principles olInsect Morphology. McGraw-Hill, New York and London, 1935.
7. O'Connor, R. and Peck, M.L. Venoms of apidae. In: Arthropod Venoms, Edited by S. Bettini. Springer, Ber-
lin and New York, 1978, pp. 613--{559.
8. Rathmayer, W Venoms of Sphecidae, Pompilidae, Multillidae and Bethylide. In: Anthropod Venoms, Ed-
ited by S. Bettini, Springer, Berlin and New York, 1978, pp. 661--690.
9. Piek, T. and Simon-Thomas, R.T. (1969) Parlysing venoms of solitary wasps. Camp. Biochem. Physiol. 30,
10. Maschwitz, V.Wl. and Kloft, W (1971) Morphology and function of venom apparatus of insects: bees,
wasps, ants and caterpillars. In: Venomous Animals and the Venoms, Vol. 3, Venomous Invertebrates. Edited
by W Bucherl and E. Buckley. Academic Press, New York, pp. I-56.
II. Beard, R.L. (1978) Venoms of Braconidae. In: Arthropod Venoms, Edited by S. Bettini, Springer, Berlin
and New York, pp. 778-800.
12. Schmidt, 1.0. (1982) Biochemistry of insect venoms. Ann. Rev. Ent. 27, 339-368.
13. Blum, M.S. and Hermann, H.R. (1978) Venoms and venom apparatuses of the Formicidae. In: Arthropod
Venoms, Edited by S. Bettini, Springer, Berlin and New York, pp. 801-899.
14. Beard, R.L. (1952) The toxicology of Habrobracon venom: a study of natural insecticid. Conn. Agric. Exp.
Stn. Bull. 562.
15. Rathmayer, W (1962) Das Paralysierungsproblem beim Bienewolf Philanthus triangulum F. (Hym.
Sphec.) Z. Vergi. Physiol, 45, 413-462.
16. Rathmayer, W (1962) Paralysis caused by the digger wasp Philanthus. Nature (Land.), 196, 1148--1151.
17. Steiner, A.L. (1962) Etude du comportement predateur d'un hymenoptere sphegien: Uris nigra Vd.L. (=
Notogonia pomplijormis pz), Ann. Sci. Nat. Zool. (Ser. 12) 4. 1-126.
18. Murray, 1.A. (1964) A case of multiple bee stings. Cent. Afr. J. Med., 10, 249--251.
19. Edery, H., Ishay, 1. Lass, I. and Gitter, S. (1972) Pharmacological activity of oriental hornet (Vespa orien-
talis) venom. Toxicon 10, 13-23.
20. Beck, B. Bee Venom Therapy, Appleton-Century, New York, 1935.
21. Banks, B.E.C., Rumjanek, FD., Sinclair, N.M. and Vernon, C.A. (1976) Possible therapeutic use ofa pep-
tide from bee venom. Bull. Inst. Pasteur 74,137-144.
22. Schultz, .R., Loos, M., Bub, F and Arnold, P.I. (1980) Differentiation of hemolytically active-fluid-phase
and cell-bound human Clq by an ant venom-derived polysacharide. J. Immunol. 124, 1251-1257.
23. Bucherl, W. Venomous chilopods or centipedes. In: Venomous Animals and their Venoms. Vol. 3 Venomous
Invertehrates, Edited by W Bucherl and E. Buckley, Academic Press, New York, 1971, pp. 169--195.
24. Blum, M.S. Chemical Defenses of Arthropods, Academic Press, New York and London, 1981. .
25. Mashwitz, V. (1964) Gefahrenalarmstoffe und Gefahrenalrmierung bei sozialen Hymenopteren. Z. Vergl.
Physiol. 47, 596--{)55.
26. Maschwitz, U. (1966) Alarm substances and alarm behavior insocial insects. Vitam. Horm. 24, 267-290.
27. Blum, M.S. and Brand, 1.M. (1972) Social insect pheromones: their chemistry and function. Am. Zool. 12,
Role of Hymenopterous Venoms in Nature 199

28. Boch, R. and Shearer, D.A. (1966) Iso-pentyl acetate in stings of honeybees of different ages. J. Apic. Res.
29. Morse, R.A., Shearer, D.A., Boch, R. and Benton, A.W. (1967) Observations on alarm substances in the ge-
nus. Apis. J. Apic. Res. 6,113-118.
30. Saslavsky, H., Ishay, J. and Ikan, R. (1973) Alarm substances as toxicants in the oriental hornet colony, Life
Sci. 12, 135-144.
31. Matsura, M. and Sakagami, Sh. F. (1973) A bionomic sketch of the giant hornet Vespa mandarinia, aserious
pest of Japaneses epiculture. 1. Fac. Sci. Hakkaido Univ. (Ser. VI) 19. 125-162.
32. Gaul, A.T. (1953) Additions to vespine biology XI defence flight. Bull. Brooklyn Ent. Soc., 48, 35-37.
33. Edery, H., Ishay, J., Gitter, S. and Joshua, H. Venoms ofvespidae. In: Arthropod Venoms, Edited by S. Bet-
tini, Springer, Berlin and New York, 1978, pp. 691-771.
34. Rivnay, E. and Bytinski-Salz, H. (1949) The oriental hornet (Vespa orientalis L.) its biology in Israel. Bull.
Agric. Res. Sta. Rechovot, 51. 36-42.
35. Kuhlhom, F. (1961) Uber das Verhalten sozialer Falten wespen (Hymenoptera: Vespidae) beim Stallein-
flug, innerhalb von Viehstallen und eim Fliegenfang Z. Angew. Zool. 48, 405-422.
36. Ishay, J. Bytinski-Salz, H. and Shulov, A. (1967) Contributions to the bionomics of the oriental hornet
Vespa orientalis. Isr. J. Ent. 2, 45-\06.
37. Sakagmi, Sh. and Fukushima, K. (1957) Some biological oservations on a hornet Vespa tropica var pul-
chra, with special referene to its dependence on Polites wasps. Treubia 24, 73-82.
38. Piek, T. (Ed.) Venoms of Hymenoptera. Academic Press, New York, 1986.
39. Akre, R.D. and Reed, H.C. Biology and distribution of social Hymenoptera. In: Handbook of Natural Tox-
ins, Volume 2 (Anthony T. Tu Ed.), Marcel Dekker, New York, 1984, pp. 3--48.
40. Shipolini, R.A. Biochemistry of bee venom. In: Handbook of Natural Toxins, Volume 2 (Anthony, T. Tu,
Ed.) Marel Dekker, New York, 1984, pp. 49--86.
41. Levin, I.w. Vibrational studies of model membrane - melittin interactions. In: Handbook ofNatural Toxins,
volume 2 (Anthony, T. Tu, Ed.) Marcel Dekker, New York, 1984, pp. 87-108.
42. Nakajima, T. Biochemistry ofvespid venoms. In: Handbook of Natural Toxins, volume 2 (Anthony, T. Tu,
Ed.) Marcel Dekker, New York, 1984, pp. \09--1 34.
43. Piek, T. Pharmacology of hymenoptera venoms. In: In: Handbook of Natural Toxins, volume 2 (Anthony,
T. Tu, Ed.) Marcel Dekker, New York, 1984, pp. 135-186.
44. Zlotkin, E. Toxins derived from arthropod venoms specifically affecting isnects. In: Comprehensive Insect
Physiology Biochemistry and Pharmacology, volume 10 (G.A. Kerkut and L.I. Gilbert, Eds.) Pergamon
Press, Oxford, 1985, pp. 499--541.
45. Piek, T. Insect venoms and toxins. In: Comprehensive Insect Physiology Biochemistry and Pharmacology,
volume \0 (G.A. Kerkut and L.l. Gilbert, Eds.) Pergamon Press, Oxford, 1985, pp. 595-634.
46. Hider, R.C. (1988) Honey bee venom: a rich source of pharmacologically active peptides. Endevour, 12,
47. Schmidt, J.O. (1995) toxinology of venoms from the honeybee genus Apis. Toxicon, 33, 917-927.
48. Nakajima, T. Pharmacological Biochemistry ofvespid venoms. In: Venoms of the Hymenoptera (Tom Piek,
Ed.) Academic Press, London, 1986, pp. 309--324.
49. Breithaupt, H. and Habermann, E. (1968) Mastzelldegranulierendes Peptid (MDCD-Peptid) aus Bi-
enengift: Isolierung biochimische und pharmako10gische Eigenschaffen. Naunyn-Schmiedebergs Arch.
Pharmakol. Exp. Pathol., 261, 252-270.
50. Hirai, Y., Kuwada, M., Yasuhara, T., Yoshida, H. and Nakajima, T. (1979) A new mast cell degranulating
peptidehomologous to mastoparan in the venom of the Japanese hornet (Vespa Xanthoptera). Chem.
Pharm. Bull., 27,1945-1946.
51. Yasuhara, T. Hoshida, H. and Nakajima, T. (1977) Chemical investigation of hornet (Vespa xanthoptera
Cameron) venom. The structure of a new bradykinin analogue "vespakinin". Chem. Pharm. Bull., 25,
52. Haermann, E. and Jentsch, J. (1967) Sequenzanalyse des Melittins aus den tryptischen und peptischen
spaltstucken. Hoppe-Seylers Z. Physiol. Chem" 348, 37-50.
53. Hegner, D., Schummer, U and Schnepel, G.H. (1973) The interaction ofa lytic peptide, melittin with spin-
labeled membranes. Biochim. Biophys. Acta, 291, 15-22.
54. Bemheimer, A.W., Avigad, L.S. and Schmidt, J.O. (1980) A hemolytic polypeptide from the venom of the
red harvester ant Pogomyrmex arbatus. Toxicon, 18,271-278.
55. Hoffman, D.R. (1994) Alergens in hymenoptora venoms. XXVI: The complete amino acid sequences of
two vespid venom phospholipases.lnt. Arch Allergy Immonol .. 104, 184--190.
56. Nakahata, W., Ishimoto, H., Mizuno, K., Ohizumi, Y. and Nakanishi, H. (1994) Dual effects of Mastoparan
on intracellular free Ca+2 concentrations in human astrocytoma cells. Br. J. Pharmacol. tt2, 299--303.
200 E. Zlotkin

57. Radermecker, M., Louis, R., Leclecq, M., Weber, T., Corhay, J.L. and Bury, T. (1994) Cytokine modulation
of basophil histamine release in wasp venom allergy. Allergy, 49, 641~44.
58. Dohtsu, K., Okumura, K., Higiwara, K., Palma, M.S. and Nakajima, T. (1993) Isolation and sequence
analysis of pep tides from the venom of Protonectarina sylveirae (Hymenoptera, Vespide). Nat. Toxins, I,
59. Kawai, N. and Hori, S. Effect of hornet venom on crustacean neuromuscular junctions. In: Animal. Plant
and Microbial Toxins. Vol. 2 (A. Ohsaka, K. Hayashi and Y. Sawai, Eds.), Plenum, New York, 1976, pp.
60. Abe, T., Kawai, N. and Niwa, A. (1982) Purification and properties of a presynaptically acting neurotoxin,
Mandaratoxin, from hornet ( Vespa mandarinia). Biochemistry, 21, 1693-1697.
61. Piek, T., Hue, B., Le Corronc, H., Mantel, P., Gobbo, M. and Rocchi, R. (1993) Presynaptic block of trans-
mission in the insect CNS by mono and di galactosyl analogues ofvespulakinin I, a wasp (Paravespula
maculifrons) venom neurotoxin. Compo Biochem. Physiol. C, lOS, 189-196.
62. Piek, T., May, T.E. and Spanjer, W Paralysis of insect skeletal muscle by the venom of the digger wasp
Philanthus triangulum F. In: Insect Neurobiology and pesticide action, Soc. Chern. Indust., London, 1980,
pp. 219-226.
63. Piek, T. and Spanjer, W. Effects and chemical characterization of some paralysing venoms of solitary
wasps. In: Pesticide and venom neurotoxicity (D.L. Shankland, R.M. Hollingworth and T. Smyth, Eds.),
Plenum, New York, 1978, pp. 211-226.
64. Usherwood, P.N.R. and Machili, P. (1968) Phrmacological properties of excitatory neuromuscular synapses
in the locust, J. Exp. BioI., 49, 341-361.
65. May, T.E. and Piek, T. (1979) Neuromuscular block in locust skeletal muscle caused by a venom prepara-
tion made from the digger wasp Philanthus triangulum F. from Egypt. J. Insect Physiol., 25, 68~91.
66. Rathmayer, W (1966) The effect of the poison of spider and digger wasps on the prey. Mem. Inst. Butantan
Simp. Int., 33, 651~57.
67. Piek, T., Mantel, P. and Jas, H. (l980a) Ion-channel block in insect muscle fibre membrane by the venom
of the digger wasp Philanthus triangulum F. J. Insect Physiol., 26, 345-349.
68. Piek, T., May, T.E. and Spanjer, W. (I 980b) Paralysis oflocomotion in insects by the venom of the digger
was Philanthus triangulum. In: Insect Neurobiology and Pesticide Action (Neurotox 79). Society of
Chemical Industry, London, pp. 219-226.
69. Clark, R.B., Donaldson, P.L. Oration, K.A.F., Lambert, J.J., Piek, T., Ramsey, W, Spanjer, Wand Usher-
wood, P.N.R. (1982) Block oflocust muscle glutamate receptors by 8-Philanthotoxin occurs after receptor
activation. Brain Res., 241, 105-114.
70. Eldefrawi, A.T., Eldefrawi, M.E., Katsuhiro, K., Mansur, N.A., Nakanishi, K., Oltz, E. and Usherwood,
P.N.R. (1988) Sstructure and synthesis of a potent glutamate receptor antagonist in wasp venom. Proc.
Nat!. Acad. Sci. USA, 85, 4910-4913.
71. Drenth, D. (1974b) Susceptibility of different species of insects to an extract of the venom gland of the
wasp Microbracon hebetor (Say). Toxicon, 12, 189-192.
72. Tamashiro, M. (1971) A biological study of venoms of two species of Bracon. Tech. Bull. Hawaii Agric.
Exp. Stn., 70.
73. Piek, T. and Engels, E. (1969) Action of the venom of the wasp Microbracon hebetor Say on larvae and
adults of the moth Philosamia cynthia Hubn. Compo Biochem. Physiol,. 28, 60~ 18.
74. Piek, T., Mantel, P. and Engels, E. (1971) Neuromuscular block in insects caused by the venom of the dig-
ger wasp Philanthus triangulum L. Compo Gen. Pharmacol., 2, 317-331.
75. Walther, C. and Rathmayer, W (1974) The effect of Habrobracon venom on excitatory neuromuscular
transmission in insects. J. Comp. Physio., 89, 23-38.
76. Piek, T., Veenendaal, R.L. and Mantel, P. (I 982b) The pharmacology of Microbracon venom. Compo Bio-
chern. Physiol,. 72C, 303-309.
77. Rathmayer, W. and Walther, C. Mode of action and specificity of Habrobracon venom. In: Animal. Plant
and Microbial Toxins (A. Ohsaka, K. Hayashi and Y. Sawai, Eds.) Vol. 2, Plenum Press, New York, 1976,
78. Deitmer, J.W (1973) Die Wirkung des Giftes der Schlupfwespe Habrobracon hebetor (Say) auf die
neuromuskulare Ubertragung am Sartoriusmuskel des Frosches. Diplomarbeit, Uniersitat Bonn.
79. Spanier, W., Grosu, L. and Piek, T. (1977) two different paralyzing preparations obtained from a homogen-
ate of the wasp Microbracon hebetor (Say) Toxicon, 15,413-421.
80. Quistad, O.B., Nguyen, Q., Bernasconi, P. and Leisy, D.J. (1994) Purification and characterization of insec-
ticidal toxins from venom glands of the parasitic wasp, Bracon hebetor. Insect Biochem. Mol. Bioi, 24,
Role of Hymenopterous Venoms in Nature 201

81. Hayakawa, Y. (1994) Cellular immunosupressive protein in the plasm ofa parsitized insect larvae. J. BioI.
Chern .. 269, 1453Cr14540.
82. Soldevila, A.!, and Jones, D. (1994) Characterization of a novel protein associated with the parazitation of
lepidopeteran hosts by an endoparasitic wasp. Insect Biochem. Mol. Bioi., 24, 29-38.
83. Coudron, T.A. (1996) Endocrinologically active venom proteins of ectoparasitic wasps. Toxicon, 34, p.
84. Prince, R.C., Gunson, D.E. and Scarpa, A. (1985) Sting like a bee! The ionophoric properties of mel ittin.
Trends in Biochemical Sciences, 10,99.
85. Banks, B.E.C. and Shipolini, R.A. Chemistry and pharmacology of honeybee venom. In: Venoms of
Hymenoptera (Tom Piek, Ed.) Academic Press, London, 1986, pp. 330-403.
86. Tu, A.T. (Ed.) Handbook of Natural Toxins Vol. 2, Marcel Dekker, Inc., New York, 1984.



G. Bkaily, M. Simaan, D. Jaalouk, and P. Pothier

MRCC Group in immuno-cardiovascular interactions

Department of Anatomy and Cell Biology
Faculty of Medicine
Universite de Sherbrooke
Sherbrooke, Quebec
Canada, Jl H 5N4


Several toxins have been reported to be highly specific blockers of a single type.of
ionic channel. A good example of this is tetrodotoxin (TTX), a highly specific fast Na+-
channel blocker l . Like TTX, scorpion toxins have become important tools for the study of
Na+ channels 2 However, different scorpion venoms have different types of action on this
channeI 2-4. Saxotoxin (STX) was also reported to be a specific fast Na+ channel
blocker l,56. Some natural toxins do not inhibit the fast Na+ channel but rather activate or
open this type of channel. These include gonioporatoxin (GPT)7, batrachotoxin (BXT)8
and grayanotoxin (GTX)9. Several other types of toxins were found to be specific for dif-
ferent types of K+ channels. Examples are charybdotoxin (ChTX), apamin, dendrotoxin,
noxiustoxin and gaboon viper venom 10.11. Our laboratory, as well as others, reported that
some toxins may specifically affect the L-type Ca 2 + channels 12 as well as the early embry-
onic fast transient (ft) slow Na+ channels lJ - 15 in heart muscle. Another toxin, co-conotoxin
(co-CgTx), was reported to block the L-type and N-type Ca 2 + channels but with only tran-
sient inhibitory effects on T-type Ca 2+ channels in neurons and not in heart muscle l6 . Mai-
totoxin (MTX) was found to activate a new class of voltage-independent Ca 2+ channel or
an entirely modified form of voltage-gated Ca 2+ channel in heart cells l7 .


Recently, apamin was reported to specifically block the L-type lea in old embryonic
chick heare 2 ,18,19. This toxin, at a very low concentration (lO-loM), decreased the over-

204 G. Bkaily et al.

A CONTROL c 12m'"

Figure I. Apamin decreased the
action potential (AP) overshoot
(OS) and duration in intact heart
and blocked the naturally occur-
ring slow Ca'· A Ps in reaggregated
single heart cells of old embryonic
chick. (A) Control slow Ca' · APs.
_ _ _ _ _ _---:/-_ _ _ __ _ _ _ _ 1100 VIs (8) Superfusion with Tyrode solu-
~ tion containing lO- t OM apamin for
0 .2 <ce 5 min. increased the OS and Vm,,'
(C) After 12 min , there was a com-
E plete block of the slow Ca'· APs.
Washout of apamin did not restore
1.2 the slow APs, and recovery was
w only possible when quinidine was
::> (n-52) (n - 3) added to the superfusion medium.
-...J 1.0 (D) Normal fast AP recorded in the
~ absence (upper trace) and presence
(lower AP trace) of lO- t OM
~ 0.8 apamin. Upper solid line is the
~ 0. 7 zero potential, lower trace is
::> dV (dt, the maximum excursion of
u 0.6
which gives +V m" A and B cali-
~0 . 5 bration is 0.8 s and in C i s 20 s.
(E) Apamin decreased the L-type
w 0.4
Ca'· current amplitude in a dose-
;:: 0.3 dependent manner in 10-day-old
a: chick embryo heart single cells.
a:: The ED,o for apamin was lO- t OM.
0. 1 Data presented are the means ±
0.0 l..-_--'-_ _ s.e.m. and n is the number of cul-
tured single cells tested. (Modified
[APAH I N 10 (H)
from refs. 12 and 19).

shoot and the duration of the action potentials recorded from isolated 19-day-old embry-
onic chick hearts (Fig.lD) and blocked completely the slow Ca 2+ action potential of old
embryonic single heart cell reaggregates (Figures I A-C). This bee venom toxin decreased
the slow Ca 2+ action potentials l2 and the L-type Ca 2+ current l9 in a dose-dependent manner
(Fig. IE). Very low concentrations of this toxin (l0- 12 M) decreased L-type lea amplitude
by 10% while a 50% blockade of this current was achieved at to- 10M apamin (Fig. I E). At
this low concentration (I O- loM), apamin also decreased the L-type lea by 50% in 20-week-
old single human foetal heart cells l9 . Apamin decreased the L-type and the tail current
within 5 min . This toxin had no effect on the TTX-sensitive IN a and the T -type l ea of heart
muscleI2.1 5.19.
Effect of Apamin and Melittin on Ion Channels 205


Using Fura-2 Ca2+ measurement techniques, spontaneous increase of intracellular Ca2+
and contraction of heart cells were found to be blocked by the highly potent Ca2+ channel an-
tagonist PN200-11O (Bkaily et aI., unpublished results). Using Fura-2 Ca2+ measurement
technique20 , apamin (l0-7M) did not affect the resting total intracellular free Ca2+ concentra-
tion ([Cal) of quiescent ventricular heart cells from human and chick origin. However, in
spontaneously contracting ventricular heart cells, apamin was found to decrease in a dose-de-
pendent manner the amplitude of the spontaneous increase of [Ca], and contractility of heart
cells without affecting the frequency of spontaneous beating. Using Fluo-3 confocal micros-
copy in order to monitor both spontaneous increase of cytosolic ([Ca]J and intranuclear
([Ca]n) free Ca2+ 20.21 during spontaneous contraction, 1O-9M apamin was seen to decrease the
amplitude of cytosolic and nuclear spontaneous Ca2+wave without affecting the frequency of
the increase of [Cal wave and the spontaneous contraction of heart cells. Figure 2 illustrates
an example of the effect of apamin on two attached isolated ventricular heart cells (Figs. 2A
and B) of 10-day-old chick embryo. As can be seen, superfusion with lO-9M of apamin de-
creased within 3 min. the amplitude ofthe spontaneous Ca2+wave through the cytosol and the
nucleus by 50% and slightly increased the basal level of the spontaneous increase of [Cal in
both attached cells (Figs. 2A and B). Increasing the concentration of apamin up to 1O-7M
largely decreased the spontaneous amplitude of total [Cal (Figs. 2A and B) and further in-
creased the basal level of this amplitude without affecting the frequency of the spontaneous
increase of[Cal (Figs 2A' and B'). Superfusion (in presence ofapamin) with the Ca2+ chela-
tor, EGTA, completely blocked the remaining small spontaneous increase of [Cal and de-
creased basal [Cal to near control levels (Figs. 2A' and B').
Figure 3 demonstrates an individual measurement of a spontaneous increase of [Ca]c
and [Ca]n' As seen in this figure, the resting level of cytosolic [Ca]c is lower than that of
the nucleus20 •21 • Spontaneous entry of Ca 2+ into the cytosol is rapidly buffered by the nu-
cleus and the peak amplitude of the spontaneous increase during the Ca2+ wave is higher
in the nucleus than that in the cytosol (Fig. 3A). The amplitude of the Ca2+ wave in both
the cytosol and the nucleus follow the same pattern of increase and decay (Fig. 3A). Su-
perfusion with 1O-7M apamin largely blocked the amplitude of the spontaneous increase of
both [Ca]c and [Ca]n and increased the basal level of [Ca]n without affecting the basal
level of [Cal (Fig. 3A and B).


In order to determine the state of membrane differentiation of cells, comparison
should be made of the properties of intact hearts at different stages of development in situ.
The electrical properties of the heart undergo sequential changes during developmenez-24 .
Myocardial cells in young chick hearts (2-3 days in vivo) possess slowly rising (10-30
Vis) action potentials (APs) preceded by pacemaker potentials. The upstroke is generated
by Na+ influx through slow Na+ channels which are insensitive to TTX and Mn2+ (Figs. 4A
and 4B). Kinetically fast Na+ channels which are sensitive to TTX make their initial ap-
pearance at about day 5, and increase in density until about day 18. The maximal rate of
rise of the AP (+Vrna) increases progressively from day 3 to day 18, where the adult V max
of approximately 150 Vis is attained. From day 5 to day 7, fast Na+ channels coexist with
206 G. Bkaily et al.


c.2 •
Intens,ty 150


1---' .....----1 + +


0 i I
0 OS 1.0 3.S Time (min) 40



, t·,· \11, 1\ ~ l' 1\ J~. .

a !



., '. ",' f.
,I 'f
I ~ .,. } '\ "
,,,"'' , oJ'.... I' , l'
50 \

a i if
i I I
4S 7.0 7.5 8.0 Time (min)

Figure 2. Time-lapse rapid scans (a,a',b,b') of spontaneously contracting chick myocytes loaded with 13.6~M
Flu03-AM as described in reference 18. Cells were scanned continuously every 320 msec for 50 sec at a resolution
of 512 x 512 pixels and 0.18 ~m pixel size. Increases in free calcium fluorescence are uniform and attest to the pe-
riodicity of the contraction. Calcium waves are initiated every 2 sec with a 1.5 sec duration. Time-span: 3 sec. A,
A' Band B': graphic representation of the amplitude ofCa'+ fluorescent waves during spontaneous contraction in
absence and presence of apamin. The increase in free Ca2+ fluorescence intensity levels is particularly intense in
the nuclear region which remain elevated even after cytosolic Ca ' + has returned to basal levels. [mages are shown
as pseudocolored representations according to an intensity scale from 0-255 illustrated on the left. Cell number is
cp278-2. (Bkaily et aI., unpublished results).
A 250 t"l
Nuclear Ca 2+ ~
?: ~
Cylosohc Ca 2 +
~ 200 //~ Control ....,
S >
tl.> l 3
u 150 I \
C 5'
/f, \
~ .,
u \
- -- Q,
'"~ 100
0 ::
~ ~
+ 50 5'
t 3.3 sec =
a =
a 1.5 :::.0 Time (sec) =
B =
~ 250 J - - Nuclear Ca 2+
- - - - Cytosohc Ca 2+ Apamin 1O- 7 M
a; 200
g 150
.... .... ... _-
'"~ 100-1
+ 50
0 15 3 .0 Time (sec)
I cell no. CP278- 1 1

Figure 3. Time-lapse rapid scan of a spontaneously contracting chick myocyte loaded with 13.6 ~M Flu03-AM. The cell was scanned continuously every 320 msec at a 512 x 512
resolution and O.IS ~m pixel size. In this instance, the Ca" wave had a 3 sec duration. Graphs on the left represe nt the spontaneous Ca'+ waves through the cytosolic and intranu-
clear spaces. In the absence of apamin, the increase in free Ca 2+ fluorescence intensity levels is particularly higher in the nuclear region which remains elevated even after cytosolic
Ca'+ has returned to basal levels. Cytosolic and nuclear Ca 2+ waves follow the same profile of a fast increase followed by a smaller component. 8) In prese nce of apamin, the ampli-
tude of the Ca" waves of both cytosolic and nucl ear regions largely decreased and is associated with an increase in basal nuclear Ca'+ level. Images are shown as pseudocolored rep-
resentations according to an intensity scalc from 0-255 illustrated on Figure I. Time span: 3.3 sec. The scalebar on the lower right of each time frame represents 5 microns. Cell
number is cp27S-I. (Bkaily et aI. , unpublished result s). c:>
208 G. Bkaily et al.


+10-'" TTX
+ 2. mw Nn 2 + MELITTIN: Isi(st)


® @

HP:-60 to HP: - 80 to
VS : - 32 "'v
CONTROL 120 P VS: -30 mV
+10-'" TTX 500P4L
+2. rna. Mn z " 10 tn S
5 min

Figure 4. Characteristics of the acti on potentials (APs) and the inward currents in single 3-day-old embryonic
chick hearts. (A-B) Action potentials and inward currents (B) recorded from single cells by switching between
current clamp and voltage clamp modes. (A,B) The APs and the inward slow current of single 3-day-old embry-
onic chick ventricular cells in culture were insensitive to 10-5M TTX and 2 mM Mn 2" . (C,D) Blockade of APs in-
ward currents by melittin (C) and apamin (D) in single heart cells from 3-day-old embryonic chick. The frequency
of stimulation was 0.02 Hz. HP ~ holding potential and VS ~ voltage step. Current traces in panels (C) and (D)
were taken from two different single cells. (C) A high concentration of melittin (I 0-4M) completely separated I"
from I" (st) in a typical 3-day-old embryonic chick heart single cell that showed overlapping of slow currents.
Digital subtraction of I" (s) (in the presence of 10-4 M melittin) from the I" (sts) component (in the presence of
10-sM melittin) shows that I" (st) is blocked by 10-4M melittin. Al so. digital subtraction of I" (sts) (in the presence
of 10-sM melittin) from the total control inward current shows the presence of a slow inward current that inacti-
vates rapidly (/" (ft)). (Modified from refs. 12 and 15).

a large complement of slow Na+ channels. TTX reduces Vmax to the value observed in 2-
day hearts, i.e. 10--20 VIs, but the APs persist. After day 8, the APs are completely abol-
ished by TTX and depolarization to less than -50mV now abolishes excitability. This
indicates that the AP-generating channels consist predominantly of fast Na+ channels,
most of the slow Na+ channels having been lost (functionally) so that the remaining num-
bers are insufficient to support regenerative excitation.
Single heart cells from 3-day-old embryonic chick heart cells exhibit three types of
slow Na+ inward currents l 5 and can be easily separated by melittin (Fig. 4C). The first
type, a fast transient (ft) slow Na+ inward current (/si(ft)' is activated from a holding poten-
tial (HP) of -80 mV and shows fast activation and inactivation (Figs. 4B-D). The maximal
mean ft l ea activated from a HP of -80 mV with a voltage step (VS) to -20 mV averaged
93.5 ± 14.5 /lAI /lF I5 • This value is lower than that reported for the TTX-sensitive fast lNa
in heart muscle. The low current density of the TTX-insensitive ft slow I Na may explain in
part the low Vmax of these single cells. The ft slow I Na is responsible for the rising phase of
the slow action potential in 3-day-old embryonic chick hearts (Figs. 4A and 4B). This type
Effect of Apamin and Melittin on Ion Channels 209

of ft slow I};a exists during the development of the foetal human heart and disappears com-
pletely at a foetal age of 20 weeks. In both 3-day-old chick embryonic heart cells and in
10-19-week-old foetal human heart cells, the decay phase of the ft slow INa fits well with
the sum of two exponentials 15 as in adult animal cardiac celIs 25 .26 . The kinetics of the ft
slow INa in young embryonic heart cells is similar to that Qfthe TTX-sensitive INa 26 or the
TTX-insensitive 27 and TTX-resistane S,29 Na+ currents lO •
The TTX- and Mn 2+-insensitive ft inward slow Na+ channel in young embryonic
heart seems to share a few characteristics with the L-type Ca2+ channel: (I) it is com-
pletely insensitive to the fast Na+ channel blocker, TTX; (2) it is highly sensitive to one
calcium blocker, apamin (Figs. I and 4); and (3) it is highly permeable to the divalent cat-
ion Ba2 + . However, these two types of channels do not seem to share many other proper-
ties such as: (1) threshold and reversal potential 11 ,14; (2) insensitivity of the ft slow INa to
Mn"+, Ne+, Cd"+, C0 2+, La l + and [Ca]o; (3) low sensitivity of the ft slow inward Na+ current
to verapamil, D-600, (-) D-888 and nifedipine; (4) high sensitivity of the ft slow INa to
melittin (Fig. 4C); (5) high permeability of the ft slow INa to Na+ and Lt; (6) stability of
the current in whole-cell voltage clamp conditions; the ft slow inward Na+ current is much
more stable that of I ca 11 ,14; (7) the time-course of the ft slow INa activation is different from
that reported for the T-type Ca 2+ channels in heart muscle 31 .l 2 . This TTX- and Mn 2+-insen-
sitive ft slow inward Na+ channel resembles the TTX-resistant fast Na+ channef 9 only in
that it is permeable to Na+ and Lt and impermeable to Ca 2+.
The coexistence of this channel with the fast Na+ channel in adult heart cells could
contribute to an increase in intracellular Na+, which may force the Na+-Ca 2+ exchanger to
pump Na+ ions out, thus creating a calcium influx through this exchanger. Such a phe-
nomenon would allow a Ca2 + overload in heart cells.
A close look at ventricular heart cells of the newborn normal hamster shows only
the presence of the fast Na+ current. However, ventricular cells of the newborn
cardiomyopathic hamster (one to two days old) show mainly the presence of the ft slow
Na+ current. In fetal human ventricular cells, as in heart cells from one-day-old
cardiomyopathic hamster, there is a TTX- and Mn 2 +-insensitive, ft slow inward Na+ cur-
rent. This ft slow Na+ current in both fetal human and newborn cardiomyopathic hamster
ventricular cells is sensitive to apamin and melittin. This slow Na+ channel, continues to
be functional during the development of the cardiomyopathic hamster.
The ft slow Na+ channel was also detected in older cardiomyopathic hamster ven-
tricular heart cells, However, a number of single cells show a decrease in this type of cur-
rent with increasing age of the cardiomyopathic hamster. This decrease in the number of
ventricular heart cells with functional ft slow Na+ channels may correlate with the devel-
opment of intracellular Ca 2 + overload and focal myocardial necrosis in this pathological
model. Apamin and melittin could be used to study the contribution of the slow Na + to the
pathology implicating this type of channel. Molecular cloning of TTX-resistant rat heart
Na+ channel isoform, as well as the primary structure, functional expression, and molecu-
lar cloning of human heart TTX-insensitive slow Na+ channel, was recently reported. Car-
diac cell line (MCM I), which originates from a transgenic mouse, shows a
TTX-insensitive Na+ current that seems to be similar in origin to the TTX-insensitive ft
slow sodium current in fetal heart cells and hamster cardiomyopathic heart cells.

Earlier studies showed that apamin blocked the slow Ca 2 + and slow Na+ action po-
tential by respectively inhibiting the L-type Ca 2+ channel and fast transient slow Na+ chan-
210 G. Bkaily et a!.

nel in heart cells. Using Fura-2 micro fluorometry and laser confocal microscopy, it was
demonstrated that apamin does effectively block Ca2+ influx through the L-type Ca2+ chan-
nels. Recent results in the literature reveal that the heart L-type Ca 2+ channel possesses an
apamin binding site 33 . These results strongly suggest that the main effect of apamin in
heart function is due to its high potency blockade of the L-type Ca 2+ channel. Thus,
apamin could be used as highly potent L-type Ca2+ channel antagonist with more specific-
ity to this type of Ca 2+channel than other well known Ca 2+ antagonists.
Like melittin, apamin was also found to block the ft slow Na+ channel in heart. This
type of channel seems to be functional in hereditary cardiomyopathy as well as in
Duchenne muscular dystrophy. Also, this type of channel appears to be functional in cer-
tain nerve cells. Recent cloning of the apamin binding protein in brain synaptosomes sug-
gested a non significant homology with any known ion channels or receptors. which may
suggest that the gene (Kcal 1.8)34 is likely to encode a protein associated not only with the
small conductance Ca2 +-activated potassium channel (which does not exist in heart cells)
but also with the slow Na+ channel.


Supported by MRCC grant MAll722 to Dr. G. Bkaily. Dr. Bkaily is a Merck Frosst-
FRSQ Professor. The authors thank Ms Mireille Dussault for her secretarial assistance.

1. Mille O. (1975) The receptor for tetrodotoxin and saxitoxin. A structural hypothesis. Biophys. J. 15,
2. Wheller KP., Watt DO., Lazdunski M. (1983) Classification ofNa+ channel receptors specific for various
scorpion toxins. Pflugers Archv - Eur. 1. Physiol. 397,164-165.
3. Carbone E., Prestipino G., Franciolini F., Dent MA. Possani LO. (1984) Selective modification of the squid
axon Na+ currents by Centruroides noxius toxin 11-10. J. Physiol. 79, 179-184.
4. Yatani A., Kunze DL. and Brown AM. (1988) Effects of dihydropyridine calcium channel modulators on
cardiac sodium channels. Am. 1. Physiol. 254, 140-147.
5. Kao CY. and Nishiyama A. (1965) Actions of saxitoxin on peripheral neuromuscular systems. 1. Physiol.
180, 50-{)6.
6. Strichartz G. (1984) Structural determinants of the affinity of saxitoxin for neuronal sodium channels.
Electraphysiological studies on frog peripheral nerve. J. Gen. Physial. 84, 281-305.
7. Nishio M., Muramatsu I., Kigashi S., Fujiwara M. (1988) Effects of goniopora toxin on the action potential
and membrane currents of guinea-pig single ventricular cells. Naunyn-Schmiedebergs Arch. Pharmacol.
8. Garbcr SS. (1988) Symmetry and asymmetry of permeation through toxin-modified Na+ channels. Bio-
phys. J. 54, 767-776.
9. Seyama 1., Yamada K., Kato R .. Masutani L Hamada M. (1988) Grayanotoxin opens Na+ channels from
inside the squid axonal membrane. Biophys. J. 53, 271-274.
10. Castle NA., Haylett OG., Jenkinson OH. (1989) Toxins in the characterization of potassium channels.
Trends in Neurosci. 12, 59--65.
II. Carbone E. and Swandulla D. (1989) Neuronal calcium channels: kinetics, blockade and modulation. Prog.
Biophys. Malec. BioI. 54, 31-58.
12. Bkaily G., Sperelakis N., Renaud J.-F., Payet MD. (1985) Apamin, a highly specific Ca'+ blocking agent in
heart muscle. Am. J. Physiol. 248, H961-965.
13. Bkaily G., Jacques D., Yamamoto T., Sculptoreanu A., Payet MD., Sperelakis N. (1988) Three types of
slow inward currents as distinguished by melittin in 3-day-old embryonic heart. Can. 1. Physiol. Pharma-
cal. 66, 1017-1022.
Effect of Apamin and Melittin on Ion Channels 211

14. Bkaily G., Peyrow M. Yamamoto T., Sculptoreanu A., Jacques D. and Sperelakis N. (1988) Macroscopic
Ca'+-Na+ and K+ currents in single heart and rabbit aortic cells. Mol. Cell. Biochem. 80.59-72.
15. Bkaily G., Jacques D., Sculptoreanu A., Yamamoto T., Carrier, D., Vigneault D. and Sperelakis N. (1991)
Apamin, a highly potent blocker of the TTX- and Mn'+- insensitive fast transient Na+ current in young em-
bryonic heart. J. Mol. Cell Cardiol. 23, 25-39.
16. McCleskey EW., Fox AP., Feldman DH., Cruz IJ., Olivera BM., Tsien RW., Yoshikami D.(l987) Omega-
conotoxin: direct and persistent blockade of specific types of calcium channels in neurons but not muscle.
Proc. Nat. Acad. Sciences of USA. 84, 4327-4331.
17. Kobayashi M., Ochi R., Ohizumi Y. (1987) Maitotoxin-activated single calcium channels in guinea-pig
cardiac cells. Br. J. Pharmacol. 92, 665-671.
18. Bkaily G. Single heart cells as models for studying cardiac toxicology in: In Vitro Methods in Toxicology
(Jolles G and Cordier A. Editors) Academic Press London 1992 pp.289-334.
19. Bkaily G., Sculptoreanu A., Jacques D., Economos D. and Menard D. (1992) Apamin, a highly potent fetal
L-type Ca'+ current blocker in single heart cells. Am. J. Physiol. 262, H463-471.
20. Bkaily G., Gros-Louis N., Naik R., Jaalouk D. and Pothier P. (1996) Implication of the nucleus in excita-
tion contraction coupling of heart cells. Mol. Cell. Biochem. 154, 113-121.
21. Bkaily G., Pothier P., D'Orleans-Juste P., Simaan M., Belzile E, Jaalouk D. and Neugebauer W (1996) The
use of confocal microscopy in the investigation of cell structure and function in heart, vascular endothe-
lium and smooth muscle cells. Mol. Cell. Biochem. In press.
22. Sperelakis N. Change in membrane electrical properties during development of the heart in: In vitro meth-
ods in toxicology. (Zipes DP., Bailey Je. and Elharrar V. Editors) Martinus Nijhoff Publisher The Hague
1980 pp.221-262.
23. Kojima M. and Sperelakis N. (1983) Calcium antagonistic drugs differ in ability to block the slow Na+
channels of young embryonic chick hearts. Eur. J. Pharmacol. 94, 9-18.
24. Bernard e. Establishment of ionic permeabilities of the myocardial membrane during embryonic develop-
ment of the rat in: Development and physiological correlates of cardiac muscle. (Lieberman M. and Sano
T. Editors) Raven Press N.Y. 1988 pp. I 69-1 84.
25. Brown AM., Lee KS., Powell T. (1981) Sodium current in single rat heart muscle cells. J. Physiol. 318,
26. Follmer CH. Ten Erck RE., Yeh JZ. (1987) Sodium current kinetics in cat atrial myocytes. J. Physiol.
(Lond) 384,169-197.
27. Anderson PAY. (1987) Properties and pharmacology of a TTX-insensitive Na+ current in neurones of the
jellyfish cyancea capillata. J. Exp. BioI. 133,231-248.
28. Bossu J-L., Fletz A. (1984) Patch-clamp study of the tetrodotoxin-resistant sodium current in group C sen-
sory neurones. Neurosci Let, 5 I, 241-246.
29. Ikeda S. Schofield GG. (1987) Tetrodotoxin-resistant sodium current ofrat nodose neurones: Monovalent
cation selectivity and divalent cation block. J. Physio!. (Lond) 389, 255-270.
30. Sperelakis N., Shigenobu K. and McLean M.J. Membrane cation channels: changes in developing hearts,
in cell culture, and in organ culture in: Developmental and physiological correlates of cardiac muscle. (Lie-
berman M. and Sano T. Editors) Raven Press N.Y. 1976 pp.209-234.
31. Bean BP. (1985) Two kinds of calcium channels in canine atrial cells. Differences in kinetics, selectivity,
and pharmacology. J. Physio!. (Lond). 86, 1-30.
32. Fox AP., Nowycky Me., Tsien RW. (1987) Kinetic and pharmacological properties distinguishing three
types of calcium current in chick sensory neurones. J. Physio!. (Lond) 394, 149-172.
33. Schetz JA. and Anderson PAY. (1995) Pharmacology of the high-affinity apamin receptor in rabbit heart.
Cardiovasc. Res. 30, 755-762.
34. Sokol PT., Hu W, Yi L., Toral J., Chandra M. and Ziai MR. (1994) Cloning of an apamin binding protein
of vascular smooth muscle. J. Protein Chern. 13, 117-128.



Th. Cherbuliez

1209 Post Road

Scarsdale, New York 10583

Apitherapy, the medicinal use of the products of the Hive of the Honey Bee has ap-
peared in the middle East and in China approximately at the same time, some two to three
thousand years ago. It currently includes the use of honey, Bees wax, propolis, royal jelly,
larvae, in addition to bee venom. The latter, Bee Venom Therapy (BVT), only, is the sub-
ject of this presentation. It should be understood that BVT applies as part of an overall ap-
proach which includes nutrition, vitamins supplements, minerals and exercise. These will
not be reviewed here.
If BVT has been known and used in parts of the world for more than twelve centu-
ries, its current use is limited to certain countries, such as China, Korea in the Far East,
Romania, with an extensive clinical program, Bulgaria and Russia. In the United States
the use of venom is officially limited to di-sensitization.


In the United States, Dr Christopher Kim in New Jersey, has a National Institute of
Health approved protocol for treatment with bee venom for chronic pain. He has treated
more than 2,000 patients with pain refractory to conventional therapies, and who carried
indications like arthritis, tendinitis, fibromyositis, neuritis, neuralgia, painful scars. Dr
John Santilli in Connecticut has a project, funded by the National MS Society for the in-
vestigation of Bee Venom therapy for MS. Dr Ph. Singer, in Michigan has a project treat-
ing eight MS patients. Dr Klinghart in New Mexico routinely includes BVT in his clinic's
treatment protocols. Typically researchers use venom in ampules. This venom has been
extracted from bees, having them sting on a material that allows the retrieval of the
venom. The venom then is dried, cleaned, weighted and dissolved in known amounts of
liquid. This technique permits the application of the double bind approach to research, as
one has created solutions that produce about the same subjective experience as bee venom
and therefore can be used as placebo. However the question of whether there is a real
equivalence between a given amount of venom from an ampule, where all the doses are

214 Th. Cherbuliez

identical to each other on the one hand and the same amount delivered by live bees has not
been answered. Two factors are at play here. First during the process the handling of bee
venom it is dried and then re-solved. The live bees injects a venom that has not been ma-
nipulated and has kept all of its constituents, including some possibly very volatile ones.
Further when one gets, say, ten bees, on gets ten slightly different "hits" as opposed to ten
identical ones. Indeed, we know that the venom's composition varies slightly with the
bee's age and with the bee's genetics, possibly also with the bee's feeding source. These
variables can affect bees from a same colony taken at the same time.

The literature on Bee Venom is very varied and quite rich. The American Api-
therapy Society (AAS) has a database containing some 630 titles of publications in english
from the last 30 years of which 450 were published in the last 5 years. This database is up-
dated yearly and is available to the membership
Apimondia publishes regularly a journal, Apiacta, that includes articles on api-
therapy. It is projected that an issue of this journal reserved for apitherapy will be publish-
ed annually.
The AAS, counting about 600 members has as its mission the collection and dis-
semination of information about the medicinal use of the products of the hive. It publishes
four times a year a Journal, runs workshops on Apitherapy. The Chair of the Standing
Committee on Apitherapy of Apimondia is currently held ad interim by an AAS member.

There are few clinics in the world devoted strictly to Apitherapy. On a trip organ-
ized by the AAS in 1993, our group visited one of them in Beijing, China .. Linked to the
Beijing Technological Institute, and operating since 1990 this Clinic accepts patients with
various therapies, but shifts them to BVT upon taking charge of the treatment. The range
of diagnoses treated in the Clinic is difficult to assess as the concepts used are expressed
in traditional chinese medicine terms. We did see arthritics, and one patient who gave the
history and the symptomatology of MS. The latter diagnosis does not exist in China. At
the time of our visit we heard that that there had been two anaphylactic reactions; both
were treated with epinephrine, with a favorable outcome. The diagnoses of these 2 pa-
tients could not be retrieved. No statistical studies were available. However the Director
of the Clinic stated that the treatments were mostly successful.
In Bucharest, Romania, I visited the Center for Apitherapy and research. The staff
includes 14 medical specialists, working part time and 2 full time researchers. The latter
have several tasks: to develop new combinations and formulas of bee products, mostly fol-
lowing a request from one of the specialists, to control the quality of their preparations.


Bee Venom includes some forty components, of which a dozen have been exten-
sively examined. 1 They include 11 peptides, 5 Enzymes, 3 physiologically active Amines,
Carbohydrates, Lipids and amino-acids. Of the peptides, the most represented are Melittin,
Apamin, Mast Cell Degranulating Peptide and Adolapin. All together, they have systemic
actions: anti-inflammatory, anti-fungal, anti-bacterial, anti-pyretic, stimulating ACTH,
Bee Venom in Treatment of Chronic Diseases 215

stimulating vascular permeability. The Enzymes, act on the cardio-vascular system and lo-
cally at the point of administration of the venom.
In the most summarized way one can say that Bee Venom acts on the immune sys-
tem, redirecting some of its faulty mechanism. Bee venom acts on all afflictions influ-
enced by cortisone, however without any of the side effects of the drug.


There are three frames of reference in our theoretical view of the action of the
venom. I) General, as the venom acts on the hypophyso-cortical axis to stimulate the adre-
nal cortex. This is more a construct than an established fact, as human experiments did not
support this view. However clinic has shown that conditions sensitive to cortisone do re-
spond, frequently to BVT.
2) Locating a sting on an acupuncture point has demonstrated the combined effect of
the local stimulation and the acupuncture action. 3) In joints and muscle diseases, the local
effect of the venom is the most powerfully demonstrated.


Chronic conditions can be defined as stable conditions, maintained over a significant

period of time. The intervention of bee venom challenges the elements of maintenance and
therefore shifts the status from chronic to acute. The acute condition gives the organism a
new assisted opportunity to find a more physiologic equilibrium, i. e. an opportunity to
heal itself. The action on the immune system is confirmed by the cluster of illnesses ame-
nable to BVT: Arthritis, of all types, lupus, endarteritis.
Throughout history, the chronic conditions the most frequently treated have been
and still are all forms of arthritis and conjunctive tissues pathology. In the last ten years an
important new indication has been considered, Multiple Sclerosis. It is estimated that
about 4,000 people suffering from MS in the United State have been or are treated with
In Brazil the main indication for BVT is asthma. The Chinese have an impressive
list of afflictions responding to BVT, covering a number of neurological conditions, joints
and connective tissues, benign tumors of the skin, functional conditions such as masculine
sexual impotence and neurosis. Some viral conditions such as shingles can respond favor-
ably. A promising avenue lies in the treatment of painful scars, and of treatment of scars
in condition of diminished functioning and of pain. Circumstances have shown that pre-
menstrual syndrome react often favorably to venom. So do many cases of chronic fatigue
syndrome (CFS). For the latter it is not known whether the long lasting type of CFS also
responds. Which ones of the different types of depression respond well to BVT is to be de-
termined. A number of successful treatment of carpal tunnel syndrome have been treated


As a general rule conditions that lead to instability of the person:

216 Th. Cherbuliez

• Uncompensated allergies
• Cardio-vascular conditions
• Open tuberculosis
• Syphilis?
• Gonorrhea
• unstable, insulin dependant diabetes
• use of B-blockers


There are many ways of administering the venom. The most frequently used is with
live bees, for reason of economy and practicality. This way of delivery will be implied
throughout this presentation. I have a The amount delivered with this technique varies be-
tween 0.1 and 0.3 milligrams of venom. The amount delivered can be approximately con-
trolled by the time one leaves the sting of the bee in the skin. Practical as this way is, it
does not allow an exact knowledge of how much venom a person receives, and for this
reason, researchers favor administering the venom from ampules. Bee venom can be ob-
tained in known concentration to be injected intradermally. The conventionally accepted
weight of a "dose" is 0.1 milligram. Venom can also be given through unbroken skin as a
liquid, which is forcefully rubbed into the skin (a chinese technique) and through electro-
phoresis or as a salve, often mixed with other active agents such as camphor. Finally it can
be applied as inhalations.


As a physician licensed in the state of New York, I am very careful to remind peo-
ple-they already know this!-that BVT is not an approved treatment modality. Hence it
falls under the rubric of research an no payment is due or accepted. I do not recommend
this therapy, I study it and teach it application
Most of my BVT practice consists of initiating people to BVT. That is getting them
started, and launched. This is the frame I will use in the representation that follows.
People accepting bee venom therapy have already, by this very choice indicated that
they espouse a certain philosophy of life. They [at least in the United State] do not belong
to the main stream. They often are desperate. They have suffered for years, spent great
amount of time and money, sometimes with little results. But, more important, they are
people agreeing to becoming responsible for their own therapy, accepting the fear [which
is significant] of the stings and the pain [to which one gets somewhat accustomed] that ac-
companies nearly every sting. They also accept to change significantly their way of ther-
apy. BVT takes place in a relationship. This is due in part only for technical reasons, many
stings are administered in places the person stung cannot reach easily, but more impor-
tantly, the element of pain is always present and requires emotional support. This aspect
of the therapy is so important that I strongly recommend that anyone starting BVT does it
with someone else. The first session begins with acquiring a certain sense about the person
seeking BVT. What their condition is, their general state of health, any relevant particu-
larities, their support system, their relation to pain, their level of determination, and their
resources: how will they procure the bees, whom or what will they have to guide and ad-
vise them in the course of their therapy. Then comes the teaching them about safety, fol-
Bee Venom in Treatment of Chronic Diseases 217

lowing the schema of the next paragraph and the application of the protocol fitting their
condition. It is essential that following the first stinging session, the person stung has a 48
hour available and knowledgeable contact to refer to with any question or problem


Any treatment with venom starts (or should start) with a test of allergy. One conven-
ient way to do so is to have a bee sting on the wrist, the sting being removed a fraction of
a second after the bee stung. In the following minutes the person stung is observed. It is
essential that the climate of the stinging session be comfortable and relaxed. I usually take
this time to explain the different reactions to the venom, to show a Epinephrine Kit give
principle of its use and if necessary write a prescription for such a kit. In addition, all peo-
ple involved in the therapy should be tested.
Bee Venom has an unusual therapeutic/toxic ratio. The amount of bee Venom ad-
ministered in one session varies from a fraction of one dose to 20 dependant on the indica-
tion. Treatments infrequently exceed 20 stings at one time. A healthy person can usually
tolerate 100 stings. (I am not here counting with the toxic components, that is the fact that
sometimes with relatively small amounts of venom", one can see intravascular hemolysis
and kidney failure*). This gives a ratio of 1 to 5. or better in most cases. When thinking of
safety, one must however include the element of allergy in general and anaphylaxis in par-
ticular. It is generally accepted in the United States that 3 % of the population is allergic
to Bee Venom. I believe that this figure includes a number of non-honey bee venom reac-
tions, as people often call "bee" any insect that flies and stings. The US statistics inform
us that about 40 people die each year in the States of anaphylactic reaction to stinging in-
sects. Mostly these accidents happen out doors. We do not know how many of these
events are due to honey bees.
Charles Mraz, the first one to treat an MS patient in the United States with bee
venom, indicatedt that in the 65 years of his experience with treating arthritics with bee
venom he has never encountered an anaphylactic reaction. This either speaks for a very
low proportion of strongly allergic people, and/or perhaps, for the proposition that arthrit-
ics do have a modified immune system, that does not create anaphylaxis.


Reactions can be classified in four categories, according to location and to latency

from the time of the sting to the onset of clinical manifestations.

Local-Immediate Reactions
For this purpose, "local" means that the reactions include the site of the sting, no
matter how widely spread they may be.
They follow by seconds the administration of venom. Well known, they include
Pain, Swelling, Redness, Warmth. They are harmless unless the swelling is located at such

• This is the case of a man stung hy 175 yellow jackets. not honey bees but not of a woman having receives 65
honey bee stings
t personal communication
218 Th. Cherbuliez

a place as to create mechanical difficulties. Intentional stinging includes attention to this


Local-Delayed Reactions
They occur hours later. Itching can be quite unpleasant, but responds usually to cold
or to Preparation H (an Anti-hemorrhoidal salve). Changing the technique of stinging by
early removal of the sting, say after one minute, often prevents itching.
Occasionally one sees a major swelling, mostly in the arms or legs, which can last
for days. They are easily confused with cellulitis, and are dues, when they last more then a
few hours, to trouble in the drainage of the area. A mechanical solution like elevation of
the affected limb or compression bandage is often all what is needed.
Rarely one see a reaction revive in an old site stung long ago, with stings applied

General-Immediate Reactions
"General" in this sense refers to reactions distant from and not necessarily including
the sting site.
This is the one that gives us concern. Onset is within minutes, usually, less than five.
The first complaint can be summarized by the statement "I don't feel right". It can then take
different forms such as nausea, dizziness, urge to defecate or urinate, urticaria, general oe-
dema, general weakness. It can be followed by itching of the palms and the soles, which show
nothing. Symptoms like itchy eyes, scratchy throat can follow. If the process continues,
breathing troubles come next, (and this is the moment to use the epinephrine kit). The process,
left unchecked, can lead to cardio-vascular collapse and death. This picture may be inter-
rupted at any time, and lead to progressive return to normal. Relaxation and reassurance are
important to observe as they tend to abort the process. Without being an absolute rule the fol-
lowing has frequently been observed: the longer the latency between the sting and the begin-
ning of a generalized reaction the more benign and slow the clinical process.

General Delayed Reactions

Frequently observed in ongoing treatment with BV. It typically takes place in the
second week. The most frequent picture is of flue-like manifestations, with gastro-intesti-
nal predominance, fever and general malaise. Often it is followed by a clear abatement of
symptoms and has therefore received the not quite earned name of "healing crisis".


There are no generally accepted protocols in BVT for the treatments of these condi-
tions, but, the majority of BVT practitioners adhere to certain principles. I will represent
some of them in the following descriptions of approaches to illnesses. For this purpose I
will use the application of bee venom with live bees.
As indicated previously treatment starts with general knowledge of the person,
his/her condition, and safety considerations. For the first session, my protocol for all con-
ditions accepts one test sting plus 2 complete stings, that is two stings where the sting is
left several minutes in place. For people wishing it, one can cool the place of the sting.
Bee Venom in Treatment of Chronic Diseases 219

This cuts considerably the initial, most acute pain. The most practical object to do this
with is a small all-metallic can, kept for this purpose in the freezer.
For subsequent stinging session, the protocol allows for three more stings than were
received on the previous time. So that if the person received, in addition to the test two
stings the first time, they may get 5 the second time. If they did get 5 they may have 8.
These numbers are only ceilings not to exceed, they never represent obligation. Sympto-
matic improvement and tolerance to the therapy is the guide to the number of stings and
their duration.

Arthritis. The selection of the places to sting is done near the affected joints, seeking
places sensitive to pressure, often called "trigger points". Most frequently, after painful or
disabling manifestations have abated, a lower "maintenance dose" is adequate.
Multiple Sclerosis. This is to be considered a systemic illness, even though, at least
sometimes, the symptom picture is localized. Hence the general notion that an MS patient
will have to be stung "everywhere". The selection of the points to sting will follow ana-
tomical, or acupunctural principles. This means that one will sting along the nerve paths,
with much stinging at the emergence of the nerves, along the spine as well as following
acupuncture ways. One of the most frequently presented symptom of MS is fatigue, and
this is one that generally responds quite well to the venom, as people speak of increased
energy. The timing of the stinging, morning or evening is also determined by individual
responses. Some people sleep better and are more rested when they sting in the evening,
others need their stings in the morning so as to have their strength to go to work. Some
.people will report that they stings until they feel relaxed!

AAS has collected 67 adverse reactions to Bee venom. Here, adverse refers to any reac-
tion that worries the person who has it. Although none of them were judged to be potentially
lethal, several received epinephrine, and 9 were admitted either to emergency rooms, or to in-
patient status in hospital. The major lesson taught by this series is the importance of informa-
tion. The majority of the hospitalization might have been avoided if the persons had been
informed of the course of reactions and their handling. The issue of what people should do
when they have experienced an adverse reaction has been insufficiently studied. The principle
is however that they should make certain that they are not at risk. The procedure to achieve
this goal is not well determined. It is usually left to immunologists, who often do not appreci-
ate the value of BVT and are convinced that the patient takes unnecessary risks. It is to be
noted however that the recommended dose of maintenance bee venom following de-sensitiza-
tion is 100 microgram), that is one dose! So, here may be the indication that there are ways of
reconciling the two opposites, official and alternative medicines.

There is, in the United States a general increase in the development of alternative
medicine as well as acceptance by the authorities of the poeple's right to choose their
treatment even if their choice is un-approved.
220 Th. Cherbuliez

AAS has developed a protocol allowing the rating of severity of symptoms in arthri-
tis and in MS. This protocol lends itself for following up people as they progress with
their treatment with bees, and of following a control group left unstung.
The need for communications amongst practitioners of all countries is more pressing
and more promising than ever.

1. Kim CM-H Bee venom and bee acupuncture therapy 1992 pp515
2. Bousquet-J. Huchard-G Michel-F-B Toxic reaction induced by hymenoptera venom Ann-AIlergy 1984
May 52(5). p 371--4
3. Bousquet-J Immunotherapy with Hymenoptera venoms. Position paper of the working Group on Immuno-
therapy of the European Academy of AIlergy and Clinical Immunology in Allergy 1987 Aug 42 (6)


Franco Feraboli

Orthopaedic and Traumatologic Department

Ospedale Civile di Cremona-Italy

For about the last seven years I have been using bee products (honey, pollen, propo-
lis and venom) for the treatment of orthopaedic and traumatological diseases.

I began employing honey as topical dressing in 8 cases of serious limb compound
fractures, with muscle and skin loss. I used sterile gauzes spread with acacia and lime
honey and applied on the areas concerned.
Honey was taken directly from unsterylized pots.
Before application some samples were taken in order to submit them to a bacterio-
logical test, and all proved negative.
In the first and more difficult case of tibial compound fracture I applied the gauzes
spread with lime honey immediately after the surgical operation on the soft tissues injury.
The dressing was changed every day: each time it was soaked with serum secreted from
the injured tissues.
There was no purulent secretion, and the healing process carried on regularly.
In order to stimulate further the cicatrization, after 10 honey dressings I applied re-
fined beet sugar, and then tiny small skin auto grafts (taken from the patient's thigh) ob-
taining complete re-epithelisation three months after the trauma.
Honey acid's pH, between 3.5 and 5.5, and its hygroscopic properties cause injury
essudation (1). This could be the reason for the excellent results obtained and for the lack
of infections for such serious injuries.

I often suggest pollen as the main element in a vegetarian diet which I use in order to re-
duce the overloading in arthritic patients, expecially those suffering in the lower limb joints.
I have analyzed pollen composition, and obtained these results: water 12.28 %, lip-
ids 5.59 %, protides 17.04 %, ashes 2.05 %, extracts unazotized 63,04 %.

222 F. Feraboli

From these results you could say that pollen composition is like that of an ham-
burger without the fats (2).
The only real problem is due to the special taste of pollen, essential for diet success,
but unpalatable to some. In fact a satisfying taste allows the patient to transfer to pollen
the psychological problems arising from an eccessive and unbalanced nutrition.
Where this occured the results were excellent: haematological tests performed six
months and one year after the beginning of the diet always showed completely normal val-
ues also in those patients who had hepatic, anemic and dislipidemical problems.


It has been used in 9 cases of extremely serious bone infections and 14 cases of soft
tissue infections. All the patients had previously had, without any success, a treatment of
surgical cleaning and antibiotic therapy.
The bacteria isolated were in 6 cases Staphylococcus Aureus, in 2 cases Pseudo-
monas Aeruginosa and in I case Escherichia Coli.
For their treatment I used only solid rough propolis and dissolved propolis.
The solid propolis was used when the lesion was wide or with undermined edges.
The liquid preparation was employed in case of deep septic lesion with thin fistolous
channels. Through these I have injected from 5 to 10 ml of diluted propolis each day.
Every 3-4 days I executed a careful flush with hydrogen-peroxide, in order to take out
from the septic focus the remains of pro polis, which tend to collect in the abcess cavity.
The results have always been excellent.
After a first phase of 2-3 days, in which the purulent secretion increased, there was
a progressive reduction of the secretion, a diminution of hyperpyrexia when present, and a
slow but progressive reduction of haematochemical values of inflammation (VES and
PCR). I have very much appreciated the cicatrising properties of propolis (3), expecially
the quick reduction of the abcess cavity volume and the consequent reduction of the
places where bacteria could hide.


My great interest for the bee world began studying the so-called "antirheumatic
properties" of bee venom, part of folk medicine. Therefore I read many publications on
this topic (4).
They all attribute the anti-inflammatory and antirheumatic qualities of the venom to
the stimulation of the axis hipophysis-adrenal glands, with consequent cortisol release (5).
Then I experimentally examined cortisol release in ten volunteer patients, who sub-
mitted to 4 consecutive days of apitherapy.
They were stung by 5 bees each on average.
I carried out four daily blood samples (at 9 a.m., 2 p.m., 6 p.m. and 10 p.m.) in order
to measure the cortisol level.
During the first day no venom treatment was carried out in order to measure the ba-
sal cortisol level.
My results for the following 3 days of treatment did not demonstrate any significant
increase of the cortisol.
Only 3 very emotional patients showed a slight cortisol increase.
Apitherapy in Orthopaedic Diseases 223

Table 1. Record of cases

Disease Number of cases Improved No improvement
Knee arthrosi s 26 17 9
Post traumatic ankle arthrosis 4 3
Back pain 34 27 7
Epicondylitis 15 8 7
Shoulder pain 31 20 11
Peripheral neuropathies 6 6
Metatarsalgia and Hallux valgus 23 19 4
Gout arthropathies 8 8
Psoriathic arthropathies 2 2
Undifferentiated arthritis 25 19 6
Rheumatoid arthritis 12 7 5
Achille tendinitis 3 2

Although this test could undermine the mainly accepted theory for the cause of
venom effectiveness I started to employ it for the treatment of some kinds of diseases as
can be seen in the Table 1.
The healing rate was 68 %.
The patients treated for arthritis and arthritic pathologies benefitted from a periodic
treatment (from 6 months to 2 years in length) consisting of 5---6 therapeutic sessions that
have always relieved the painful symptomatology. In these cases the follow up goes from
a minimum of 3 months to a maximum of 6 years (average 3.4 years).
It is worthing bearing in mind that 8 of the 17 patients with knee arthritis, which im-
proved with apitherapy, were then submitted to surgery to replace the joint, damaged by
the inflammatory arthritic process, with a prosthesis.
I did not consider 3 cases I treated 7 years ago for hip arthritis.
In this kind of pathology bee venom is completely ineffective, as several colleagues
who also use this therapy told me.
I employed in almost all the cases the sting of live bees applied on the trigger points,
with an average of 6 stings for each session (min. 1, max. 130).
This figure depends on not the kind of pathology but on the time employed for solv-
ing the painful symptomatology.
Recently I have utilized in some patients the lyophilized bee venom from Simic's
Apitronic Service. It seemed extremely useful and practical: using the syringe I can
choose the depth of the injection, allowing new therapeutic possibilities (6).
My first results concerning the effectiveness of this product are extremely interesting.
I have observed side effects only in 3 patients: one difficulty in breathing spontaneously
solved itself, and two cases of cutaneous rashes were treated by antihistamine tablets.
Generally (94 % of the cases) the biggest problem caused by apitherapy was itching.
I had good results in controlling this symptom by honey, spread on the treated area
and left there for 5-20 minutes.
I was very impressed by the quick and effective treatment of metatarsalgia in the
valgus great toe. The pain in this area is caused by the irritation of the great toe dorsal
digital nerve, crushed between the first metatarsal head and the shoe.
The venom injected near the first metatarsal head reaches quickly the subcutaneous
digital nerve. This action constitutes the effectiveness of the venom: in fact in my opinion
it has as its main target nervous tissue.
224 F. Feraboli

In order to verify the venom action on nervous tissues I tested some laboratory ani-
mals such as mice, fish and invertebrates.
The venom injection in the lumbar area of the mouse (2 mg of venom in a mouse of
20g weight) causes the temporary paralysis of the posterior limbs, while some venom
drops in a gold fish acquarium (2 mg of venom /1 It of water /25 g fish weight) causes to-
tal motor incoordination to the point of paralysis.
Finally venom administered to caterpillars induces immediate paralysis.
These simple tests are confirmed by Menez's experimental data (7), demonstrating
that hymenoptera toxins work at the synaptic and neuromuscolar level, stopping some
ionic channels, expecially those ofCa 2+ and K+ (8).
The specific action, therefore, on the nervous system with the stopping or slowing of
the pain is clearly evident in the treatment of the hallux valgus, and in the local anesthetic
effect following the first quick painful phase after the bee sting, referred to by almost all
the patients.
But this theory by itself cannot explain totally the good results obtained.
On examining the cases treated I noticed that most of the successes were obtained in
those cases where there were clearly the classical signs of inflammation: rubor, tumor,
calor, dolor and functio laesa.
It seems nonsensical to use an inflammatory substance like bee venom, to treat an
Nevertheless in these cases the apitherapy was very effective and the healing was
very quick.
Ifwe want to give credit to the homeopathic philosophy we can confirm that the indica-
tions for using bee venom are acute inflammations, when there are these symptoms: redness,
heat, swelling, pricking and burning pain, worsened by heat and improved by cold (9,10).
Acute arthritis is the archetypal rheumatological disease where the use of bee venom is
indicated. In this disease there is in fact redness and swollen joints with stabbing, burning and
very bad pain. Moreover the joints are stiff and the pain worsened by every small movement.
These symptoms are also present in the valgus great toe and in the gout arthropa-
thies, where apitherapy is strongly indicated.
I don't believe that the bee venom has special antiinflammatory properties; rather I
think that it has an important analgesic property.
The first and immediate effect of the apitherapy, as the patients told me, was always
the reduction or the vanishing of the pain, although in some cases the inflammation did
not disappear.
The latter progressively vanished during the treatment or after a few days from the
end of the therapy.
Interpreting this data in a psychoneuroimmunological perspective (the only one able
to bring together the many sides of apitherapy, the cultural, the psychological, the chemi-
cal for example) we could assume that the mitigation of the pain influences the central
nervous system, modulating favourably the peripheral inflammation.
Finally, I think that the bee products are an important support to the medical practice
and especially bee venom. It is, in fact, with precise indications and after a careful diagno-
sis, an effective alternative to the often indiscriminate use of cortisone.

1. Nahmias F. La miel cura y sana. Editorial De Vecchio Barccllona, 1987.
Apitherapy in Orthopaedic Diseases 225

2. Iannuzzi J. (I 993) Pollen: food for honey bee - and man? Am. Bee J. 133, 557-563
3. Kaal J. Natural medicine from honey bees (Apitherapy). Kaal's Printing House, Amsterdam, 1991.
4. Malone F. Bees don't get arthritis. Academy Books Rutland. 1979.
5. A.A.S. 1977-1985 Apitherapy Society proceedings. American Apitherapy Society, Inc.
6. Simics M. Bee Venom: exploring the healing power. Apitronic Publishing, Calgary, 1994.
7. Menez A. (1994) La struttura delle tossine degli animali velenosi. Le Scienze. 305, 56--62.
8. Habermehl G.G. Venomous Animals and their Toxins. Springer, Berlin 1981.
9. Hodiamont G. Remedes et venins du regne animal en homeopathie. Chez l' Auteur, Bruxelles, 1984.
10. Desmichelle G., Manson J., Droudard J.M. Homeopathie en Rhumatologie. Maloine Paris 1991 pp. 34--36.



Boris A. Yakobson

The Kimron Veterinary Institute

Bet Dagan, P.O.B. 12,50250, Israel

Recent significant increases in trade together with elevated awareness in health is-
sues, especially the interrelationship between environmental pollution and food, has re-
sulted in the development of international standards for contaminants in meat, milk and
other agricultural products. Consumption of honey and other bee products is especially
vulnerable to even a suspicion of hazardous contamination, as they are regarded as
"health" products and are not essential for human nutrition.
There are three main purposes for monitoring honey and bee products: consumer health
protection, international competition and better producfquality. In this review "contaminant"
refers to any chemical substance which should not occur in natural "clean" honey, beeswax,
propolis, and other bee products. The main classes of contaminants are classified as: inadver-
tent, microbial, drug and acaricides, and poisonous plants. An example of inadvertent con-
taminants may be pesticides such as monocrotophos on maize fields which can be found even
when the major source of nectar is an uncontaminated source. The most important microbial
contaminant of honey is Clostridium botulinum, a ubiquitous bacterium (4). Therapeutic
drugs and acaricides are frequently used to control and eliminate bee diseases. On rare occa-
sions bees collect nectar from plants which contain plant toxins, resulting in potentially toxic
honey (e.g. Rhododendron and romedotoxin or Nerium oleander) (12).
Depending on the source of contamination, an alternative classification of contami-
nants may be considered: environmental, disease and pest control related, storage, han-
dling and processing related (3).
It is increasingly accepted to make a careful monitoring of possible contaminants in
honey, wax, pollen, royal jelly, propolis and bee venom. Compared to the available litera-
ture concerning the analysis of pesticide residues in grains, fruits, vegetables, animal prod-
ucts, soils and water, very little has been published addressing residue analysis in bees,
honey and beeswax (1,5,10,21) .
. When pesticide residue laboratories are confronted with the analysis of bees and
hive products for a particular chemical, the normal response is to adapt a method devel-
oped for some other matrix. These adaptations, when successful, are rarely published.

228 B. A. Yakobson

Thus each laboratory, over time, has developed its own particular set of techniques for
analysis of bees and hive products with no standardization between laboratories (8).
The range of chemical residues found in bees and hive products is not limited to ag-
ricultural chemicals. Bees are notorious for collecting materials contaminated with chemi-
cals and bringing them back to the hive. These include environmental pollutants,
radioactive chemicals, heavy metals, inorganic compounds and industrial wastes. Honey
bees have been used to monitor local pollution.
Regarding the import/export regulations it should be noted that no accepted interna-
tional detailed specific code for product quality yet exists. A general requirement for
honey to be free of any foreign organic and inorganic matter, as well as free from mould,
insects, brood and grains of sand has been applied. It is also required that honey is to
originate from disease-free apiaries.
The analysis of chemical residues is a rapidly evolving field with new and improved
instrumentation being frequently introduced. Recently the major advance has been to in-
crease ability to separate compounds through the use of high resolution chromatography.
These techniques have contributed to the continuous lowering of detection limits for resi-
dues. Gas chromatography-mass spectrometry (GC-MS) is currently the method of choice
for a large number of compounds. On the horizon is the greater usage of the argon plasma
detector which can quantify a wide number of elements simultaneously.
Falsification of honey is an old problem. It will not be considered here, except to
mention that the laboratory methods to detect it are well documented (6).
Regarding possible microbial honey contamination, C. botulinum should be men-
tioned. C. botulinium is an ubiquitous bactearium and widespread as dust. Its spores are
present in and on a wide variety of agricultural products such as fruits and vegetables. Bo-
tulinum toxins are among the most highly toxic substances known. Diet changes after 6
months of age affect the flora of infant intestines so that Clostridium botulinum cannot
colonize. On July 5, 1978, the world's largest honey-producing group, the Sioux Honey
Association, warned that there may be a risk of infant botulism if honey is fed to infants
under one year of age. Contaminated honey has been proven to cause infant botulism very
rarely, but has not been shown conclusively to be a significant risk factor. Since honey is
not necessary for the nutrition of infants, it is recommended not to be given to infants up
to one year of age (4,7,11,16,18).
Pesticides remain the major contaminant of concern for honey originating from cul-
tivated plants (oranges, cotton, etc.). Recently, group-specific detection methods for resi-
dues have been developed, and the following groups are monitored: organochlorides,
organophosphates, carbamates.
Varroasis causes a major problem for beekeeping. A wide range ofacaricides is used
worldwide to control this pest (9,15,17). In Israel fluvalinate is mainly used and therefore
a screening program to examine the level of this acaricide is conducted at a national level
(23). Fluvalinate in honey has not been detected, however an early accumulation in wax
and propolis has been shown (levels range between 0.2 and 2.4 ppm). The established tol-
erance for fluvalinate by EPA (Environmental Protection Agency of the USA) is 0.05 ppm
for honey. Since beeswax is used in cosmetic products and fluvalinate is a known aller-
gen, a tight control and monitoring program is strongly recommended. A report by Liakos,
B. described contamination of Greek honey and wax with malathion and coumaphos (13).
Rare, but documented cases of malicious application of home insecticides and pet
parasitic ides for bee extermination have occurred. In such cases high residue levels were
detected in wax and honey. Various insecticides were found in the wax from wild bee
nests, when inappropriate chemical substances were used to remove bees.
Monitoring of Contaminants in Bee Products 229

Antibiotics such as tetracycline, oxytetracycline, chloramphenicol, sulfathiazole, fu-

magillin may also be found in bee products (19,20). Their origin stems from inappropriate
use of these drugs for therapy of bacterial and protozoal bee diseases, when the therapy
takes place within honey collection time or 30 days prior to it. A residue monitoring pro-
gram was initiated in Israel during the years 1995/96. Contamination of honey with strep-
tomycin, tetracycline, sulfonamides, fluvalinate and amitnaz was investigated. No residues
of any of these compounds were found. Residues of up to 5 ppm of fluvalilvate were
found in the recycled wax.
Chemical substances such as ethylene dibromide, phosphine and methyl bromide
used in wax moth control, can find their way into bee products. Our experience shows that
use of CO 2 in portable PVC built fumigation chambers eliminates the need for anti-wax
moth chemical treatment (22).
A number of contaminants can enter into bee products during storage, handling and
processing. A recent case of increased residue levels of zinc, was described in the Aus-
tralasian Beekeeper Journal. The zinc residues apparently came from drums, pipes or
tanks (2). Wood protectants used in hive manufacturing can also be a source of contami-
nation. Use of non-feed plastics for storage and processing may lead to honey contamina-
tion too.
A polluted environment, containing such radionuclides as Sr 132 and/or such heavy
metals as As, Pb, Cd, Hg, etc., was demonstrated on several occasions to be a source of
bee product contamination. For example, 2 peaks of increased level of radioactivity ap-
peared after the Chernobyl disaster in honey and pollen (14). The first peak was a result of
direct radioactive exposure. Several months later a secondary effect, through radiopolluted
soil and vegetation appeared. Long term monitoring of polluted areas is clearly required in
such cases.
Regarding pollen, one should pay attention to botanic composition, poisonous
plants, visible mould and fungi, mycotoxins, allergens and physical cleanliness.
Propolis accumulates environmental pollutants, drugs and in hive waste products to
a high degree. Therefore propolis for human consumption should be harvested only from
designated hives, which did not undergo any chemical treatment.
Royal jelly and bee venom are products with a high protein content. Therefore, spe-
cial measures such as aseptic collection, rapid refrigeration and strict hygiene of collectors
should be taken. The final product must meet bacteriological safety requirements deter-
mined as: total bacterial count less than 200 CF/g. Escherichia coli and Staphylococcus
aureus should not be detectable in I g, while Vibrio cholera and Salmonella spp. should
not be detectable in 100 g. Also no residues of antibacterial drugs should be detectable
from royal jelly products.
In summary, our experience and data from the literature suggests the following
measures to be taken to reduce bee product contamination:
I. Education of beekeepers regarding proper bee hive location.
2. Judicious use of pesticides, acaricides, and antibiotics (amount and timing).
3. Systematic monitoring at all level - producer, packer and national organizations.

I. Anon., (1989) Codex Alimentarius commission procedual manual FAO/WHO, FAO, Rome, 44-45.
2. Anon., (1992) Zinc residues found on exported honey. The Australasian Beeneefer II, 215.
3. Anon., (1993) Pesticide residues in Food Report FAO, Rome. 141-149.
230 B. A. Yakobson

4. Beetlestone E, (1994) Botulism spores and honey. Am. Bee 1. 134(7) 471-472.
5. Con cellon Martinez A. (1995) Public health and residues of Veterinary medicines in foods Alimentaria,
(265) 87-100.
6. Crane, E. (1975) Honey, W. Heinemann Ltd., London 314-325.
7. Fenicia L., Ferrini A.M., Aureli P., and Pocecco M. (1993) A case of infant botulism associated with honey
feeding in Italy. Eur. 1. Epidemiol. 9(6) 671-673.
8. Flamini, C., Robin, S. (1985) Methode de dosage des residus d'amitraze dans Ie miel. Bulletin d'Inforrna-
tion des Laboratories des Services Veterinaries 18; 47-54.
9. Infantidis, M.D. et al. (1988) Effectiveness of aqueous solution of malathion against Varroa mite applied in
field experiments. Prceedings of a meeting of the EC Experts 203-208.
10. Kastner B., Groschuer P. (1994) Examining bee honey with regard to quality and residues. Dtsh. Lelensm. -
Ruudsch. 90(5) 146--150.
II. Kothare S.Y., Kassuer E.C., (1995) Infant botulism: a rare cause of coloric ileus. Pediatr-Radio. 25 (I)
12. Ktochmal, C. (1994) Poison honeys. Am. Bee. 1.134(8); 549-550.
13. Liakos B. (1983) Studies on toxic residues of malathion in honey. Mellenike Kteniatrike 2611 0 308-313.
14. Molrahn, D., Assmann-Werthmuller, U. (1993) Caesium radioactivity in several selected species of honey.
Sci. Total Enrion. 130-131.
15. Nansen H. and Petersen 1.H., (1988) Residues in honey and wax after treatment of bee colonies with bom-
propyl ate. Tidsskr. Planteavl., 1-6.
16. Sakaguchi G., et al (1990) Distint characters of Clostridium botulinum type A strains and their associated
with infant botulism in lapan. Int. 1. Food Microbiol. 11(314) 231-241.
17. Slaveezki, Y., Gal H., and Lensky Y. (1991) The effect offluvalinate applicaiton in bee colonies on popula-
tion levels of Varroa jacobsoni and honey bees (Apis mellifera L.) and on residues in honey and wax. Bee
science 1:189--195.
18. Spika 1.S. et al (1989) Risk factors for infant botulism in the United States Archives of Disase in Child-
hood 64(6) 871-872.
19. Takeba K., Fujinuma K., Nakazawa H. (1995) Honeybee diseases and drug residues, Foods, Food Ingred.
1. lpn (165) 63-72.
20. Usleber E., et al (1995) Detection of Streptomycin in honey by enzyme immunoassay using solid phase ex-
traction and immunoaffinity chromatography. 46(4) 94-96.
21. Weitzman, R.1. (1994) Veterinary drug residues. Commission of the European Communities 1014.
22. Yakobson B.A. et al. (1990) Application of carbon dioxide in a portable fumigation chamber to control bee
wax moth. Proceedings ofthe International Symposium on Recent Research on Bee. Bee Pathology, Gent,
Belgium 192-193.
23. Yakobson B.A., and C. Efrat (1988) Recent developments in bee disease control in Israel. Proceedings of
International Bee Research Association, 4th International confearence of Agriculture in Tropical Climates,
Cairo, Egypt.


Practical and Simple Procedures to Reduce