Fitoterapia 78 (2007) 565 570

www.elsevier.com/locate/fitote

Short report

Antimicrobial synergism and antagonism of salicylaldehyde

in Filipendula vulgaris essential oil
N. Radulovi a,, M. Mii a , J. Aleksi a , D. okovi b , R. Pali a , G. Stojanovi a
a

Department of Chemistry, Faculty of Sciences and Mathematics, University of Ni, Viegradska 33, 18000 Ni, Serbia
b
Faculty of Chemistry, University of Belgrade, Studentski trg 16, 11001 Belgrade, Serbia
Received 6 July 2006; accepted 1 March 2007
Available online 24 May 2007

Abstract
The leaf essential oil of Filipendula vulgaris, consisting mainly of salicylaldehyde (68.6%), was screened for its antimicrobial
activity by the disk diffusion and microdilution broth assays. The essential oil remarkably inhibited the growth of all of the tested
bacteria and fungi. It seems that the antimicrobial nature of F. vulgaris essential oil can be attributed to the synergistic interactions of
the compounds constituting the oil rather than to the presence of a single inhibitory agent. A synergy in salicylaldehyde/linalool
mixtures was observed with a maximum interaction situated in the range between 60:40 and 80:20 (mol ratio). At this concentration
range (at a dose of 1.7 g/disk) no microbial growth was observed while the respective pure compounds, at the corresponding
quantities, are shown to be dramatically less active. The MIC value for the 60:40 mixture was determined to be less that 0.009 mg/ml.
In addition, an antagonistic relationship between salicylaldehyde and methyl salicylate was established. The maximum (negative)
interaction was shown to correspond approximately to the mixture at the 40:60 (methyl salicylate/salicylaldehyde) mol ratio resulting
in the complete loss of activity at the investigated dose.
2007 Elsevier B.V. All rights reserved.
Keywords: Filipendula vulgaris; Essential oil; Antimicrobial activity; Salicylaldehyde; Synergism; Antagonism

1. Plant
Filipendula vulgaris Moench leaves (syn. Filipendula hexapetala Gilib. ex Maxim.), (Rosaceae) [1] were collected
on 26 May 2005, from the Mountain Suva planina, at 1440 m, Devojaki grob, nearby Ni, Southern Serbia. A voucher
specimen has been deposited in the Herbarium of the Faculty of Sciences and Mathematics, University of Ni.
2. Uses in traditional medicine
F. vulgaris (dropwort) has a long history of use in folk medicine and phytotherapy in many countries such as Serbia
[2], Poland [3], Ukraine [4] and Bulgaria [5]. This herb has the same ethnopharmacological use as Filipendula ulmaria
Corresponding author. Tel.: +381 637582352; fax: +381 18533014.
E-mail address: vangelis0703@yahoo.com (N. Radulovi).
0367-326X/$ - see front matter 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.fitote.2007.03.022

566

N. Radulovi et al. / Fitoterapia 78 (2007) 565570

(meadowsweet). Powdered roots of F. vulgaris (tuberous roots containing starch are edible but bitter), or used in
decoction, were utilized to treat kidney problems, breathlessness, wheezing, sore throats and congestion. Due to the
higher content of tannins in the roots compared to F. ulmaria, F. vulgaris is frequently used to treat stomachache and
diarrhea. Many other properties including easing influenza and gout, and using a tea made from the leaves to cleanse
wounds and sore eyes, i.e. anti-inflammatory, antipyretic, analgesic and antirheumatic properties have been reported
[2]. F. vulgaris is one of the three most frequently used herbal remedies in the central part of the Northern Balkan
Mountains region [5].
3. Previously isolated classes of constituents
Tannins [68], phenolic acids [911] and flavonoids [12,13].
4. Newly-isolated constituents
No reports.
5. Tested material
Essential oil obtained by steam-distillation of dry finely ground leaves of F. vulgaris (yield: 0.05%, w/w of dry
plant material) was analyzed by GC and GC-MS. GC-MS: Varian 3400 GC equipped with split/splitless injector
(1:99) operated at 266 C; column JetW Scientific DB-5ms-ITD (30 m, 0.25 mm id, 0.25 m film); carrier gas:
hydrogen, rate flow l ml/min measured at 210 C; column temperature: linearly programmed from 40 C to 285 C at
4.3 C/min; transfer line at 280 C, coupled to Finnigan-MAT 8230 BE mass spectrometer; ion source temperature
170 C; EI, 70 eV, 0.1 mA. GC-FID conditions were identical as for GC-MS. Identification procedure: linear
retention indices for all of the compounds were determined by co-injection of the sample with a solution (2% in
hexane) containing the homologous series of C8C35 n-alkanes. The constituents were identified by comparison of
their MS to those from the Wiley 275.1 MS library and the MS library built in the Mass Finder 2.3 (Hochmuth, D.) as
well as by comparison of their retention indices [14] with literature values [15] and wherever possible, by co-injection
with an authentic sample ( Merck, Germany). Percentage area values were obtained electronically from the GC-FID
response without the use of an internal standard or correction factors. The identity of some of the constituting
compounds was corroborated by TLC co-chromatography of authentic substances and the essential oil on silica gel
plates (Kiesel 60 F254, Merck) in several solvent systems. The spots on TLC were visualized by UV light (254 nm)
and by spraying with 30% H2SO4 and Godin's reagent [16], followed by heating.
6. Studied activity
Disc diffusion method according to the NCCLS [17] was employed for the determination of antimicrobial activity
of the essential oil, corresponding pure constituents and their mixtures at a dose of 1.7 g / disk. The following
nutritive media have been used: Antibiotic Medium 1 (Difco Laboratories, Detroit Michigan USA) for growing
Gram(+) and Gram() bacteria and Tripton soy agar (TSA-Torlak, Belgrade) for Candida albicans and Aspergillus
niger. Nutritive media have been prepared according to the instructions of the manufacturer. All agar plates were
prepared in 90 mm Petri dishes with 22 ml of agar giving the final depth of 4 mm. 0.1 ml of a suspension of the tested
microorganisms (108 cells / ml) was spread on the solid media plates. Sterile filter paper disks (Antibiotica Test
Blattchen, Schleicher and Schuell, Dassel, Germany, 6 mm in diameter) were impregnated with 50 l of the samples
(all samples were tested in 1:30 dilution, w/v (mg/ml), in diethyl ether (Et2O) and were filter-sterilized using a
0.45 m membrane filter) and placed on inoculated plates. These plates, after standing at 4 C for 2 h, were incubated
at 37 C for 24 h for bacteria and at 30 C for 48 h for the fungi. Standard disks of Tetracycline, Gentamicin,
Neomicin and Nystatine (Institute of Immunology and Virology Torlak, 30 g, diameter 6 mm) as positive
controls, while the disks imbued with 50 l of Et2O as a negative control. The diameters of the inhibition
were measured in mm using a Fisher-Lilly Antibiotic Zone Reader (Fisher Scientific Co. USA). Each test was
performed in triplicate and repeated three times and results analyzed for statistical significance. Mean values were
selected.

N. Radulovi et al. / Fitoterapia 78 (2007) 565570

567

Broth microdilution susceptibility assay was used, as recommended by NCCLS, for the determination of MIC
values [18]. All tests were performed in Mueller Hinton broth (MHB; BBL) supplemented with Tween 80 detergent
(final concentration of 0.5% (v/v)), with the exception of the fungal organisms (Sabouraud dextrose broth-SDB +
Tween 80). Geometric dilutions of the oil, ranging from 0.009 to 18.00 mg/ml, were prepared in a 96-well microtiter
plate. Plates were incubated at 37 C for 24 h for bacteria and at 30 C for 48 h for yeasts. Each test was performed in
duplicate and repeated twice. Amikacin and Bifonazole were used as positive controls.
7. Used microorganisms
American Type Culture Collection Maryland, USA. Gram (+): Staphylococcus aureus (ATCC 6538), Gram():
Klebsiella pneumoniae (ATCC 10031), Pseudomonas aeruginosa (ATCC 9027), Salmonella enteritidis (ATCC
13076). Fungi: Aspergillus niger (ATCC 16404) and Candida albicans (ATCC 10231). Institute of Immunology and
Virology Torlak, Belgrade, Serbia and Montenegro. Gram() :Escherichia coli 95.
8. Results
Reported in Tables 2 and 3.
9. Conclusions
The essential oil from the leaves of F. vulgaris was a yellowish oily pungent smelling liquid with a similar density as
water obtained in a yield of 0.05% based on dry weight of plant material. Thirteen compounds identified, representing
more than 95.9% of the total oil, are listed in Table 1 in the order of elution on a DB-5ms column. The oil is
characterized by a high amount of phenylpropanoid derived compounds, PHPD, (salicylaldehyde 68.6%, -asarone
5.9%, methyl salicylate 2.4% and benzaldehyde 2.3%) and fatty acid derived compounds, FAD (green leaf volatiles,
formed by enzymatic degradation of unsaturated fatty acids [19], (E)-3-hexen-1-ol 6.0%, (E)-2-hexenal 4.2 %) while
the monoterpenoids (linalool 1.8%, nerol trace amount) constituted only a minor fraction.

Table 1
F. vulgaris essential oil composition
Number
1
2
3
4
5
6
7
8
9
10
11
12
13

Constituents
FAD

(E)-2-Hexenal
(E)-3-Hexen-1-ol FAD
Hexanol FAD
Propylbenzene PHPD
Benzaldehyde PHPD
Salicylaldehyde PHPD
Linalool TER
2-Phenylethanol PHPD
(E,Z)-2,6-Nonadienal FAD
Methyl salicylate PHPD
Nerol TER
-Ionone TER
-Asarone PHPD
Total identified

RI a

Identification method

857
861
874
943
959
1049
1098
1112
1154
1182
1220
1485
1703

4.2
6.0
1.2
tr b
2.3
68.6
1.8
1.0
0.7
2.4
tr
1.8
5.9
95.9

RI a, MS c
RI, MS
RI, MS
RI, MS
RI, MS, CI d, TLC e
RI, MS, CI, TLC
RI, MS, CI, TLC
RI, MS
RI, MS
RI, MS, CI, TLC
RI, MS
RI, MS
RI, MS

FAD, fatty acid derived compound.

PHPD, phenylpropanoid derived compound.
TER, terpenoid.
a
RI, experimentally determined retention indices relative to n-alkanes C8C35 on a DB-5 ms column, identification achieved by comparison with
literature values.
b
tr, trace, b0.1%.
c
MS, identified by mass spectrum comparison.
d
CI, identified by co-injection of an authentic sample.
e
TLC, identified by Si-gel co-CC with an authentic sample.

568

N. Radulovi et al. / Fitoterapia 78 (2007) 565570

The results obtained in the antimicrobial assays are presented in Tables 2 and 3. In the disk diffusion assay the
essential oil, 1.7 g/disk, was shown to possess a broad spectrum of a remarkable activity against all tested strains. The
diameters of growth inhibition zones ranged from 33 to 40 mm (including the diameter of the disk, 6 mm) with the
highest inhibition zone values observed against A. niger (39 mm) and C. albicans (40 mm). Significant reductions in
bacterial growth were obtained with E. coli (38 mm), K. pneumoniae (37 mm) and S. aureus (37 mm). The activity
was greater than that of the conventional antibiotics at a 30 g/disk with inhibition zones reaching in some cases more
than twice the value of the standard antibiotics used as positive controls. All of the microorganisms were completely
non-susceptible to control disks imbued with ether. P. aeruginosa was the most resistant of all the tested bacteria
although still the oil showed more then two times stronger activity then the antibiotics tested.
It seems the antimicrobial nature of F. vulgaris essential oil can be attributed to the synergistic interaction of the
compounds constituting the oil rather than to the presence of a single inhibitory agent. This becomes quite apparent
from the results presented in Table 1 showing a synergy in salicylaldehyde/linalool mixtures with a maximum
Table 2
Antimicrobial activity of the F. vulgaris essential oil, pure components and corresponding mixtures
Samples

E. coli

K. pneumoniae

P. aeruginosa

S. enteritidis

S. aureus

A. niger

C. albicans

Essential oil a
Salicylaldehyde a
Linalool a
Methyl salicylate a
Benzaldehyde a
Salicylaldehyde : linalool b
20:80
20:0
0:80
40:60
40:0
0:60
60:40
60:0
0:40
80:20
80:0
0:20
Salicylaldehyde:methyl salicylate b
20:80
20:0
0:80
40:60
40:0
0:60
60:40
60:0
0:40
80:20
80:0
0:20
Neomicin c
Gentamicin c
Nystatine c

38
27
28
26
26

37
20
17
12
25

33
13
15
11
18

35
18
19
19
20

37
22
26
21
12

39
37
18
23
20

40
40
26
23
23

36
11
23
55
19
20
N90 d
23
17
N90 d
26
13

20
9
13
46
11
9
N90 d
15
na
N90 d
17
na

18
na
14
39
7
11
N90 d
9
7
N90 d
12
na

25
13
16
49
14
12
N90 d
15
9
N90 d
17
7

22
16
19
10
17
18
N90 d
18
15
N90 d
20
10

27
15
15
27
17
13
N90 d
30
9
N90 d
35
na

27
18
23
35
20
18
N90 d
28
13
N90 d
33
9

12
11
24
10
19
21
na
23
15
11
26
11
13
15
nt

8
9
10
na
11
9
na
15
na
na
17
na
15
16
nt

7
na
9
na
7
na
na
9
na
na
12
na
14
15
nt

11
13
17
10
14
15
na
15
11
9
17
7
20
14
nt

13
16
19
11
17
16
na
18
12
10
20
8
14
17
nt

13
15
20
12
17
18
na
30
14
11
35
9
nt
nt
17

17
18
20
13
20
18
na
28
14
11
33
9
nt
nt
18

The growth inhibition zones (in mm), including the paper disk diameter of 6 mm; mean values of 9 independent experiments.
na, not active.
nt, not tested.
a
Tested at a dose of 1.7 g per disk (50 l per disk of 1:30, mg/ml, dilution in ether).
b
The constituting compounds were mixed in the above reported mol ratios in ether and the obtained mixtures tested at a dose of 1.7 g per disk.
Also the pure compounds in corresponding quantities found in the mixtures were tested separately.
c
Tested at a dose of 30 g per disk (Torlak).
d
No microbial growth at all was observed in a Petri dish (90 mm in diameter) containing a single filter disk imbued with the sample in question.

N. Radulovi et al. / Fitoterapia 78 (2007) 565570

569

Table 3
MIC a (mg/ml) of the F. vulgaris essential oil, pure components and selected mixtures
Samples

E. coli

K. pneumoniae

P. aeruginosa

S. enteritidis

S. aureus

A. niger

C. albicans

Essential oil
Salicylaldehyde
Linalool
Methyl salicylate
Benzaldehyde
Salicylaldehyde:linalool b 60:40
Salicylaldehyde:methyl salicylate b 60:40
Amikacin c
Bifonazole c

0.017
0.141
0.070
0.281
0.141
b0.009
N18.00
2.00
nt

0.017
0.070
0.141
0.281
0.070
b0.009
N18.00
2.00
nt

0.035
0.141
0.141
0.562
0.281
b0.009
N18.00
1.00
nt

0.035
0.070
0.070
0.281
0.281
b0.009
N18.00
4.50
nt

0.017
0.035
0.141
0.141
0.562
b0.009
N18.00
2.00
nt

0.009
0.141
0.281
0.141
0.281
b0.009
N18.00
nt
9.00

0.009
0.009
0.035
0.141
0.141
b0.009
N18.00
nt
32.00

nt, not tested.

a
Minimal inhibitory concentration.
b
The constituting compounds were mixed in mol ratios.
c
Values given as g/ml.

interaction situated in the range between 60:40 and 80:20 (mol ratio). At this concentration range (at a dose of 1.7 g/
disk) no microbial growth was observed in the Petri dishes (one disk per dish) while the respective pure compounds at
the corresponding quantities are shown to be dramatically less active. This synergism was not completely unexpected
[20]. The synergistic interaction does not seem to be microorganism selective and affects both fungi and bacteria. In
addition to the observed synergy of salicylaldehyde with linalool, the essential oil of F. vulgaris revealed an
antagonistic interaction of methyl salicylate with salicylaldehyde. The maximum (negative) interaction was shown to
correspond approximately to the mixture at the 40:60 (methyl salicylate/salicylaldehyde) mol ratio resulting in the
complete loss of activity at the investigated dose. Contrary to the observed synergistic interaction of salicylaldehyde
with linalool, the antagonism of salicylaldehyde and methyl salicylate seems to affect the fungi less then the bacteria. It
seems that these two interactions, if not a number of others, operate in the essential oil of F. vulgaris as antimicrobial
principles. Very little information has been reported on the synergistic role of chemical compounds in antimicrobial
activity. One study [21] states the significant role of minor constituents in biological activity where a mixture of all
compounds found in the essential oil of Thyme showed increased biological activity in comparison to the six major
compounds when investigated independently. Most studies focus on the antimicrobial activity of single compounds
[22]. With regard to the other components at least in part responsible for the strong broad spectrum of antimicrobial
activity shown, benzaldehyde and green leaf volatiles can be mentioned. The activity of benzaldehyde is presented in
Tables 2 and 3. It has been suggested for some of the FAD green leaf volatiles that these compounds might serve as
wound disinfectants (following mechanical damage, green plant tissue releases a characteristic odor known as green
leaf odor) [23].
Results obtained in a microdilution broth susceptibility assay (Table 3) confirmed the findings of the disk diffusion
technique (Table 2). The MIC values of the essential oil ranged from 0.009 to 0.035 mg/ml suggesting a strong
antimicrobial activity of the same order of magnitude or greater, especially in the case of the tested fungi, compared to
the positive controls used (Amikacin and Bifonazole). The mixture of salicylaldehyde and linalool (60:40, mol ratio),
as well as that of salicylaldehyde and methyl salicylate (40:60, mol ratio) had MIC values lower that 0.009 and greater
that 18.00 mg/ml, respectively, corroborating the proposed synergistic/antagonistic interaction between salicylaldehyde and the other two oil constituents.
Acknowledgements
The authors acknowledge the Ministry of Science and Environmental Protection of the Republic of Serbia for
financial support (Project 142054B).
References
[1] Lempiainen T. Ann Bot Fenn 1982;19:127.
[2] Tucakov J. Leenje biljem: fitoterapija. Beograd: Izdavako preduzee Rad; 1973. p. 624.

570
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]

N. Radulovi et al. / Fitoterapia 78 (2007) 565570

Smolarz HD, Sokolowska-Wozniak A, Skwarek T. Herba Pol 2001;47:269.
Vasiljuk MD, Bandaruk II, Kurtash YA, Gudivok VI. Farm Zh (Kiev) 1995;4:56.
Ploetz K, Blair O. Econ Bot 2004;58:231.
Musaelyan MS, Revazova LV, Manucharyan MA. Content of tannides in some plants of the Armenian flora. Deposited Doc; 1983. (VINITI
1935-83),19 pp.
Genig AY, Ladnaya LY, Kalashnikov ID. Mater S'ezda Farm B. SSR, 3rd, 162; 1977.
Kazarnovskii LS, Karavai NY, Lazareva KV. Tr Khar'kovsk Farmatsevt Inst 1962;2:23.
Smolarz HD, Sokolowska-Wozniak A. Chem Environ Res 2003;12:77.
Popescu ML, Istudor V, Dinu M, Gird CE, Vladescu A. Farmacia (Bucharest) 2000;48:85.
Genig AY, Ladnaya LY. Farm Zh (Kiev) 1980;1:50.
Popescu ML, Istudor V, Ristache MZ. Farmacia (Bucharest) 2003;51:92.
Smolarz HD, Dzido TH, Sokolowska-Wozniak A. Acta Pol Pharm 1999;56:169.
Van Den Dool H, Kratz PD. J Chromatogr 1963;11:463.
Adams RP. Identification of essential oil components by gas chromatography and mass spectroscopy. Carol Stream IL: Allured; 1995.
Godin P. Nature 1954;174:134.
NCCLS (National Committee for Clinical Laboratory Standards). Performance standards for antimicrobial disk susceptibility test. Approved
Standard. Wayne PA. M100-S9. 6th ed. ; 1997.
NCCLS (National Committee for Clinical Laboratory Standards). Performance standards for antimicrobial susceptibility test. Ninth International
Supplement.Wayne PA. M100-S9; 1999.
Hatanaka A, Kajiwara T, Matsui K. Z Naturforsch C 1995;50:467.
Kubo I, Muroi H, Kubo A. J Nat Prod 1994;57:9.
Nakatsu T, Lupo A, Chinn J, Kang R. Biological activity of essential oils and their constituents. In: Atta-ur-Rahman, editor. Studies in natural
products chemistry, vol. 21; 2000. p. 571.
Magiatis P, Skaltsounis A, Chinou I, Haroutounian S. Z Naturforsch C 2002;57:287.
Schulz S, Weissbecker B, Klein A, Hummel HE. Naturwissenschaften 1997;84:212.