Blackwell Science, LtdOxford, UKMMIMolecular Microbiology 0950-382X Blackwell Science, 200246Original ArticleR-plasmid transfer to Y. pestis in the fleaB. J.

Hinnebusch, M.-L. Rosso, T. G. Schwan and E. Carniel

Molecular Microbiology (2002) (2), 349354

High-frequency conjugative transfer of antibiotic

resistance genes to Yersinia pestis in the flea midgut

B. Joseph Hinnebusch,1* Marie-Laure Rosso,2

Tom G. Schwan1 and Elisabeth Carniel3
1
Laboratory of Human Bacterial Pathogenesis, Rocky
Mountain Laboratories, National Institute of Allergy and
Infectious Diseases, National Institutes of Health, 903 S
4th St., Hamilton, MT 59840, USA.
2
Institut Pasteur de Madagascar, BP 1274, Antananarivo
101, Madagascar.
3
Yersinia Laboratory, Institut Pasteur, 28 Rue du Docteur
Roux, 75724 Paris Cedex 15, France.
Summary
The acquisition of foreign DNA by horizontal transfer
from unrelated organisms is a major source of variation leading to new strains of bacterial pathogens.
The extent to which this occurs varies widely, due in
part to lifestyle factors that determine exposure to
potential donors. Yersinia pestis, the plague bacillus,
infects normally sterile sites in its mammalian host,
but forms dense aggregates in the non-sterile digestive tract of its flea vector to produce a transmissible
infection. Here we show that unrelated co-infecting
bacteria in the flea midgut are readily incorporated
into these aggregates, and that this close physical
contact leads to high-frequency conjugative genetic
exchange. Transfer of an antibiotic resistance plasmid
from an Escherichia coli donor to Y. pestis occurred
in the flea midgut at a frequency of 10-3 after only 3
days of co-infection, and after 4 weeks 95% of coinfected fleas contained an average of 103 antibioticresistant Y. pestis transconjugants. Thus, transit in its
arthropod vector exposes Y. pestis to favourable conditions for efficient genetic exchange with microbial
flora of the flea gut. Horizontal gene transfer in the
flea may be the source of antibiotic-resistant Y. pestis
strains recently isolated from plague patients in
Madagascar.
Introduction
Human plague is often rapidly fatal without prompt
Accepted 12 July, 2002. *For correspondence. E-mail jhinnebusch@
niaid.nih.gov; Tel. (+33) 406 363 9260; Fax (+33) 406 363 9394.

2002 US Government

diagnosis and antibiotic intervention, and prophylactic

antibiotic treatment is also critical for persons exposed to
bubonic or pneumonic plague. No vaccine is currently
available. After decades of quiescence, plague reemerged in Madagascar during the 1990s to cause an
estimated 2001600 cases per year (Chanteau et al.,
2000). In 1995, a strain of Y. pestis resistant to multiple
antibiotics, including all the drugs of choice for plague
(streptomycin, gentamycin, tetracycline, chloramphenicol
and sulphonamides) was isolated from a human in Madagascar (Galimand et al., 1997). The Y. pestis isolate contained a foreign conjugative plasmid that encoded all of
the resistance determinants. This was the first report of
multidrug resistance in a vector-borne bacterial pathogen.
The same year, a streptomycin-resistant Y. pestis strain
in which the drug-inactivating aminoglycoside phosphotransferase genes were also present on a conjugative
plasmid was isolated from another plague patient in
Madagascar (Guiyoule et al., 2001). Both resistance (R)
plasmids mediated efficient conjugative self-transfer
between Yersinia spp. and between Y. pestis and E. coli
in vitro, but the two plasmids were otherwise unrelated.
Nucleotide sequence analysis demonstrated that the
150 kb multidrug-resistant plasmid (designated pIP1202)
was related to Inc6-C group plasmids common to the
Enterobacteriaceae, whereas the 40 kb streptomycin resistance plasmid (pIP1203) was composed of a portion
of transposon Tn5393 inserted into the broad-hostrange IncP group plasmid R751 (Galimand et al., 1997;
Guiyoule et al., 2001). In addition, the Y. pestis isolates
harbouring the R plasmids were of different ribotypes.
These molecular epidemiology data indicate that the two
antibiotic-resistant Y. pestis strains arose independently,
presumably by conjugation with unknown Gram-negative
donor bacteria.
Yersinia pestis is an obligate parasite that alternates
between mammal (principally rodent) and flea; its life
cycle does not include a free-living stage. Thus, transfer
of the R plasmids to Y. pestis must have happened in one
of these eukaryotic hosts. Here we demonstrate that Y.
pestis can acquire an R plasmid from Escherichia coli
donors at high frequency during the normal course of
infection in the digestive tract of Xenopsylla cheopis rat
fleas. In addition to implicating conjugation in the flea as
a possible source of recently emerged antibiotic-resistant

350 B. J. Hinnebusch, M.-L. Rosso, T. G. Schwan and E. Carniel

strains, our results may be relevant to past horizontal gene
transfers from insect-associated microorganisms that
were detected in the Y. pestis genome (Parkhill et al.,
2001).
Results
High-frequency conjugative transfer in the flea midgut
To test the potential for conjugative gene transfer in the
flea, we first infected X. cheopis rat fleas (the principal
vector of plague in Madagascar) by feeding them on blood
containing one of two different K-12 E. coli (pIP1203)
donor strains. We previously have found that not only E.
coli K-12 but also O9 and R1 strains are able to establish
comparable low-level chronic infection of the flea digestive
tract (data not shown). Five to seven days after the first
infectious blood meal, the same fleas were secondarily
infected with the streptomycin-sensitive, non-virulent Y.
pestis 6/69c strain. The infection rate and bacterial load
in the fleas was determined 3, 7, 14 and 28 days after the
second infectious blood meal (Fig. 1). Both donor and
recipient were capable of colonizing the flea digestive
tract, even though a certain percentage of fleas rapidly
cleared the infection by excretion. The E. coli donor strains
established a chronic infection in 54% of the fleas. The
average number of E. coli donor bacteria per infected flea

Fig. 2. Incidence of streptomycin-resistant Y. pestis transconjugants

in the digestive tract of X. cheopis fleas after infection with an E. coli
conjugative donor and streptomycin-sensitive Y. pestis.
A. Percentage of fleas containing Y. pestis 6/69c(pIP1203)
transconjugants.
B. The average number of transconjugants per flea, at different times
after the infectious blood meals. The average and range of two
experiments are indicated.

Fig. 1. Percentage of fleas colonized by the E. coli donor (black

bars), the Y. pestis recipient (white bars), by both donor and recipient
(grey bars) and by neither bacteria (hatched bars) at different times
after co-infection. As indicated by the arrows, fleas were infected
serially, first with E. coli and then, 57 days later (designated day 0),
with Y. pestis. The average and range of two independent experiments are shown.

increased from 2.0 103 at the time of the second infectious blood meal containing Y. pestis to 8.4 104 after
28 days. Y. pestis 6/69c colonized the digestive tract of
68% of the fleas, the expected infection rate (Hinnebusch
et al., 1996; Hinnebusch et al., 1998); 40% of these fleas
were also infected with the E. coli donor (Fig. 1). The
average number of Y. pestis per flea was 1.4 103 on day
3 and 5.3 104 on day 28 after infection. After only 3 days
of co-infection, streptomycin-resistant Y. pestis transconjugants that had received pIP1203 were recovered from
the fleas (Figs 2 and 3). The estimated frequency of transfer during this time was 1.0 10-3, based on the average
2002 US Government, Molecular Microbiology, 46, 349354

R-plasmid transfer to Y. pestis in the flea 351

the proventriculus) that is prerequisite for efficient transmission (Pollitzer, 1954). This blockage rate is normal
(Hinnebusch et al., 1996; Hinnebusch et al., 1998),
indicating that co-infection is unlikely to interfere with
transmission. Seven of the 15 blocked fleas examined
contained streptomycin-resistant transconjugants, which
constituted 0.054% of the total Y. pestis recovered from
these fleas.
Mixed aggregates of Y. pestis and co-infecting bacteria
in the flea midgut

Fig. 3. Plasmid DNA content of Y. pestis 6/69c recipient (lane R), E.

coli K802N (pIP1203) donor (lane D), and streptomycin-resistant Y.
pestis 6/69c (pIP1203) transconjugants isolated from fleas at 3, 7, 14
and 28 days after co-infection (lanes 14). Y. pestis 6/69c lacks the
70 kb Yersinia virulence plasmid, but contains the 9.6 kb pPCP and
101 kb pMT plasmids that are unique to Y. pestis (Galimand et al.,
1997). (cc, covalently closed circular isoform; oc, open circular
plasmid isoform).

number of E. coli donor and Y. pestis transconjugant bacteria per flea on day 3. Surprisingly, this value is close to
the conjugation frequency of 6 10-2 between E. coli
and Y. pestis observed in vitro under optimal filter-paper
mating conditions (Guiyoule et al., 2001). The percentage
of fleas that contained streptomycin-resistant Y. pestis
increased with time of co-infection, and at the end of the
4-week observation period 21 of the 22 fleas colonized by
both donor and recipient contained Y. pestis transconjugants (Fig. 2A). The number of resistant Y. pestis per
flea also increased with time (Fig. 2B), averaging 2%
of the total Y. pestis population per flea at day 28.
pIP1203 is stably maintained by both E. coli and Y. pestis
in vitro, even in the absence of selective pressure
(Galimand et al., 1997; Guiyoule et al., 2001). Therefore,
cell division of Y. pestis transconjugants and secondary
conjugation between Y. pestis, as well as continued conjugation between E. coli and Y. pestis, could contribute
to this increase. Plasmid analysis confirmed that the
streptomycin-resistant Y. pestis cultured from fleas had
acquired pIP1203 (Fig. 3).
During the 4 weeks following the infectious blood meal
containing Y. pestis, 22% of fleas developed the characteristic foregut blockage (caused by Y. pestis infection of
2002 US Government, Molecular Microbiology, 46, 349354

Our results show that the flea midgut provides a very

favourable environment for conjugative exchange of an R
plasmid. The manner of Y. pestis growth in the flea may
facilitate conjugation: as they multiply in the midgut, the
bacteria aggregate to form dense, coherent microcolonies
(Bacot and Martin, 1914; Hinnebusch et al., 1998). If coinfecting donor bacteria were incorporated into these
aggregates, prolonged physical contact with Y. pestis
recipients conducive to conjugation would result. To evaluate this, digestive tracts dissected from fleas 1031 days
after co-infection with Y. pestis 6/69c carrying the gene
for green fluorescent protein (GFP) and E. coli K802N
(pIP1203) carrying the gene for red fluorescent protein
(DsRed2) were examined. E. coli was detected within
most Y. pestis aggregates that had developed in the
midgut and proventriculus (Fig. 4). The dense growths of
Y. pestis in the flea digestive tract are surrounded by an
extracellular matrix and resemble a bacterial biofilm
(Hinnebusch et al., 1998; Darby et al., 2002). Highfrequency bacterial conjugation within biofilms has been
reported previously (Hausner and Wuertz, 1999). The
large concentration of Y. pestis that is achieved in the
small volume of the flea gut would in itself increase
chance contacts between co-infecting bacteria that lead
to the formation of stable mating pairs, and high rates of
plasmid exchange have also been reported among gut
bacteria of other insects (Jarrett and Stephenson, 1990;
Hoffmann et al., 1998; Watanabe and Sato, 1998).
Discussion
The acquisition of foreign DNA by horizontal transfer from
unrelated organisms is a major source of variation leading
to new strains of bacterial pathogens (Ochman et al.,
2000). The discovery in 1959 of self-transmissible resistance (R) plasmids, which carry genes for antibiotic resistance and for conjugative transfer to recipient bacteria,
was an early indication of the significance of horizontal
gene transfer to bacterial pathogenesis (Watanabe,
1963). An alarming recent example of this phenomenon
is the emergence of antibiotic-resistant strains of Y. pestis,
an obligate parasite that alternates between insect and

352 B. J. Hinnebusch, M.-L. Rosso, T. G. Schwan and E. Carniel

Fig. 4. Coaggregation of Y. pestis and E. coli in the flea midgut.

A. Phase-contrast microscopy of a typical bacterial aggregate in the midgut of a flea dissected 10 days after infection with Y. pestis 6/69c (pGFP)
and E. coli K802N (pIP1203) (pDsRed2). The indicated portion of the aggregate was examined at higher magnification by fluorescence microscopy
using FITC (B) and TRITC (C) filter sets to visualize Y. pestis and E. coli cells respectively.
D. Superimposed image of (B) and (C) showing incorporation of E. coli within the dense aggregate of Y. pestis. Bar = 20 mm.

mammalian hosts. Conjugation depends on several environmental and bacterial factors that influence the development of stable mating pairs (Curtiss, 1976; Guiney,
1984). Evidence for conjugative transfer between different
bacteria in mammalian intestine, urinary tract, respiratory
tract and wounds has been reported, but estimated transfer rates are very low, even in the presence of antibiotic
selective pressure (Stotzky and Babich, 1986).
The genesis of the two antibiotic-resistant Y. pestis
strains in Madagascar cannot be determined retrospectively, but it has been speculated that the initial conjugative
transfers occurred in a rat or a human (Dennis, 1997;
Chanteau et al., 2000). In the mammal, Y. pestis infects
the dermis, lymphoid tissue and blood, sites that are
normally sterile. Opportunities for conjugative genetic
exchange would be rare, occurring only during simultaneous infections with Y. pestis and another invasive bacterium, and then only if close contact between donor and
recipient were subsequently established. In contrast, Y.
pestis may regularly encounter potential donor bacteria
among the resident flora of the flea digestive tract. The

symbiotic microbial flora of the flea gut has not been

well characterized, but can include Enterobacteriaceae,
pseudomonads and other Gram-negative bacilli (Strand,
1977; Savalev et al., 1978; Beard et al., 1990). For
example, 87% of flea pools collected from individual
rats in Antananarivo, Madagascar, contained a mixture
of Gram-positive and -negative bacteria (M.-L. Rosso and
S. Chanteau, unpublished data). Fleas could take up bacteria as their mouthparts penetrate the skin, or by feeding
on a septicaemic host. Alternatively, donor bacteria could
derive from the gut flora of the vermiform flea larvae,
which feed on detritus and the faeces of adult fleas.
Insects that feed exclusively on blood as adults typically
depend on a permanent digestive tract flora, which can
be acquired during the detritus-feeding and coprophagic
larval stage, to supply required nutrients and vitamins
(Ribeiro, 1996; Durvasula et al., 1997). Eight of 10 X.
cheopis larvae collected from laboratory colonies contained a varied microbial flora: yeast, Gram-positive cocci
and bacilli, and Gram-negative bacilli, including Proteus
mirabilis. P. mirabilis was also a constituent of the micro 2002 US Government, Molecular Microbiology, 46, 349354

R-plasmid transfer to Y. pestis in the flea 353

bial flora of up to 5% of adult fleas, suggesting that some
larval gut symbionts persist through the metamorphosis
from larva to pupa to adult. In considering possible donors
to the Madagascar isolates, it is noteworthy that the
antibiotic resistance locus of pIP1203 derived from a
transposon that to date has been found only in soil- and
plant-associated bacteria (Guiyoule et al., 2001). Y. pestis
is unlikely to encounter soil bacteria in its mammalian
infection sites, but such bacteria acquired from the larval
diet may be present in adult fleas.
The Y. pestis chromosome contains several pathogenicity islands and other loci with heterogeneous nucleotide composition indicative of recent introduction by
horizontal transfer (Parkhill et al., 2001). Among these
are homologues of insecticidal toxin genes of the insectassociated bacteria Serratia entomophila, Xenorhabdus
nematophilus and Photorhabdus luminescens and a
homologue of a baculovirus protease gene required for
insect pathogenesis. Given the known homologies, it
seems likely that these loci were acquired from insect
microbial flora, suggesting that horizontal gene transfer in
an arthropod host, by conjugation or other means, has
been part of the evolutionary history of Y. pestis. The
insecticidal toxin genes are also present in at least some
Y. pseudotuberculosis strains (Parkhill et al., 2001), so
their transfer may also have occurred in the environment
or in the mammalian host before the divergence of Y.
pestis.
We used a defined in vivo laboratory model in this study
to determine whether conjugation could occur in the flea.
That it occurred so readily and frequently under our experimental conditions suggests that it could also occur naturally with smaller numbers of donor bacteria. Figure 4
indicates that if even a single donor cell were incorporated
into a Y. pestis aggregate in the flea gut, the potential for
conjugative transfer would be high. In particular, demonstration of facile transfer of an R plasmid in the flea midgut
suggests that Y. pestis can readily acquire antibiotic resistance genes, or other genes that affect transmission or
pathogenicity, from unrelated bacteria during transit in its
arthropod vector. The similarity between the resistance
loci of soil bacteria and one of the Madagascar Y. pestis
strains (Guiyoule et al., 2001) further suggests that
R-plasmid transfer was not a clinical phenomenon but
occurred within natural wild rodent-flea enzootic foci. Antibiotic treatment of humans with bubonic plague would
select for plasmid-carrying strains, but whether R plasmids would be retained by Y. pestis in rodent reservoirs
is unclear (Lester et al., 1990; Dennis, 1997). However,
the high percentage of R-plasmid-containing bacteria that
occur in humans who are not taking antibiotics indicates
that, once acquired, resistance genes can be stably maintained in the absence of selective pressure (Lester et al.,
1990; Shoemaker et al., 2001). In any case, the separate
2002 US Government, Molecular Microbiology, 46, 349354

appearance of two distinct strains of antibiotic-resistant Y.

pestis during the current outbreak in Madagascar suggests that conjugative transfer to Y. pestis in its natural
environment is not an isolated phenomenon. The potential
for continued emergence and spread to other parts of the
world of Y. pestis strains that have acquired antibiotic
resistance genes or other new virulence factors poses a
global threat to public health.
Experimental procedures
Bacteria and plasmids
The non-virulent Y. pestis 6/69c strain used lacks the Yersinia
virulence plasmid (Galimand et al., 1997), which is not
required for infection of the flea (Hinnebusch et al., 1996).
The E. coli K-12 strains K802N (F l hsdR-hsdM + gal met
supE44 gyrA) and C600R (F l thr leuB6 thi-1 lacY1 supE44
rpoB rif were transformed with the conjugative R plasmid
pIP1203 isolated from a streptomycin-resistant Y. pestis
isolated from a human plague patient in Madagascar
(Guiyoule et al., 2001). The streptomycin phosphotransferase
locus of pIP1203 has been sequenced (Guiyoule et al.,
2001; GenBank accession number AJ249779). Y. pestis
6/69c and E. coli K802N (pIP1203), containing pGFP and
pDsRed2 respectively (both plasmids from Clontech, Palo
Alto, CA, USA), were obtained by electroporation. Plasmid
DNA was isolated using a modified alkaline lysis procedure
(Qiagen, Valencia, CA, USA).

Flea infections and recovery of transconjugants

Xenopsylla cheopis fleas were infected with E. coli
C600R(pIP1203) on day -7 (experiment 1) or E. coli
K802N(pIP1203) on day -5 (experiment 2). On day 0, fleas
were secondarily infected with the non-virulent Y. pestis strain
6/69c (Galimand et al., 1997; Guiyoule et al., 2001). Fleas
were infected by allowing them to feed on heparinized mouse
blood containing approximately 7.5 109 E. coli or 5.0 108
Y. pestis per ml using a membrane feeder apparatus
(Hinnebusch et al., 1996). Fleas that took both infectious
blood meals were kept at 21C, 75% relative humidity, subsequently fed twice weekly on uninfected mice and monitored
for proventricular blockage (Hinnebusch et al., 1996). At 3, 7,
14 and 28 days after infection with Y. pestis, 20 female and
10 male fleas were triturated in 100 ml of brainheart infusion
(BHI) broth and immediately plated onto MacConkey or
BHI agar containing 100 mg ml-1 streptomycin, Yersiniaselective agar (Difco) and Yersinia-selective agar containing
100 mg ml-1 streptomycin to differentially determine colonyforming units (CFU) per flea of the E. coli donor, Y. pestis
recipient and Y. pestis transconjugants respectively. Twenty
random uninfected control fleas were also triturated and
plated on BHIstreptomycin agar; none of these fleas
contained streptomycin-resistant normal flora.
To visualize the association between Y. pestis and coinfecting E. coli in the flea midgut, fleas were fed blood
containing 2 108 Y. pestis 6/69c(pGFP) and E. coli K802N
(pIP1203) (pDsRed2) per ml. On days 10, 13 and 31 after
infection, flea digestive tracts were dissected intact in water,

354 B. J. Hinnebusch, M.-L. Rosso, T. G. Schwan and E. Carniel

coverslipped and examined by phase contrast microscopy
and by fluorescence microscopy using FITC and TRITC filter
sets.

Conjugative transfer rate

Conjugation frequency was estimated by dividing the average
number of Y. pestis transconjugants per flea by the average
number of E. coli donor bacteria per flea. Colony-forming unit
counts from 60 fleas examined 3 days after co-infection in
two separate experiments were used in the calculation. CFU
data from all fleas, including those that contained zero
transconjugants, were used to calculate the average number
of transconjugants per flea, whereas only E. coli-positive
fleas were included in the average donor per flea term.

Acknowledgements
We thank J. M. Musser, M. Achtman and P. Rosa for review
of the manuscript. M.-L. Rosso was supported by the
Dlgation au Rseau International des Instituts Pasteur et
Instituts Associs.

References
Bacot, A.W., and Martin, C.J. (1914) Observations on the
mechanism of the transmission of plague by fleas. J Hyg
Plague Suppl. 3: 423439.
Beard, C.B., Butler, J.F., and Hall, D.W. (1990) Prevalence
and biology of endosymbionts of fleas (Siphonaptera:
Pulicidae) from dogs and cats in Alachua County, Florida.
J Med Entomol 27: 10501061.
Chanteau, S., Ratsitorahina, M., Rahalison, L., et al. (2000)
Current epidemiology of human plague in Madagascar.
Microbes Infect 2: 2531.
Curtiss, R. III (1976) Genetic manipulation of microorganisms: potential benefits and biohazards. Annu Rev
Microbiol 30: 507533.
Darby, C., Hsu, J.W., Ghori, N., and Falkow, S. (2002) Caenorhabditis elegans: plague bacteria biofilm blocks food
intake. Nature 417: 243244.
Dennis, D.T. (1997) Multidrug resistance in plague. N Engl J
Med 337: 702704.
Durvasula, R.V., Gumbs, A., Panackal, A., et al. (1997)
Prevention of insect-borne disease: an approach using
transgenic symbiotic bacteria. Proc Natl Acad Sci USA 94:
32743278.
Galimand, M., Guiyoule, A., Gerbaud, G., Rasoamanana, B.,
Chanteau, S., Carniel, E., and Courvalin, P. (1997)
Multidrug resistance in Yersinia pestis mediated by a
transferable plasmid. N Engl J Med 337: 677680.
Guiney, D.G. Jr (1984) Promiscuous transfer of drug resistance in Gram-negative bacteria. J Infect Dis 149: 320
329.
Guiyoule, A., Gerbaud, G., Buchrieser, C., et al. (2001)
Transferable plasmid-mediated resistance to streptomycin

in a clinical isolate of Yersinia pestis. Emerg Infect Dis 7:

4348.
Hausner, M., and Wuertz, S. (1999) High rates of conjugation
in bacterial biofilms as determined by quantitative in situ
analysis. Appl Environ Microbiol 65: 37103713.
Hinnebusch, B.J., Fischer, E.R., and Schwan, T.G. (1998)
Evaluation of the role of the Yersinia pestis plasminogen
activator and other plasmid-encoded factors in temperature-dependent blockage of the flea. J Infect Dis 178:
14061415.
Hinnebusch, B.J., Perry, R.D., and Schwan, T.G. (1996) Role
of the Yersinia pestis hemin storage (hms) locus in the
transmission of plague by fleas. Science 273: 367370.
Hoffmann, A., Thimm, T., Droge, M., Moore, E.R., Munch,
J.C., and Tebbe, C.C. (1998) Intergeneric transfer of conjugative and mobilizable plasmids harbored by Escherichia
coli in the gut of the soil microarthropod Folsomia candida
(Collembola). Appl Environ Microbiol 64: 26522659.
Jarrett, P., and Stephenson, M. (1990) Plasmid transfer
between strains of Bacillus thuringiensis infecting Galleria
mellonella and Spodoptera littoralis. Appl Environ Microbiol
56: 16081614.
Lester, S.C., del Pilar Pla, M., Wang, F., Perez Schael, I.,
Jiang, H., and OBrien, T.F. (1990) The carriage of Escherichia coli resistant to antimicrobial agents by healthy children in Boston, in Caracas, Venezuela, and in Qin Pu,
China. N Engl J Med 323: 285289.
Ochman, H., Lawrence, J.G., and Groisman, E.A. (2000)
Lateral gene transfer and the nature of bacterial innovation.
Nature 405: 299304.
Parkhill, J., Wren, B.W., Thomson, N.R., et al. (2001)
Genome sequence of Yersinia pestis, the causative agent
of plague. Nature 413: 523527.
Pollitzer, R. (1954) Plague. Geneva: World Health
Organization.
Ribeiro, J.C.M. (1996) Common problems of arthropod vectors of disease. In: The Biology of Disease Vectors. Beaty,
B.J., and Marquardt, W.C. (eds.) Niwot, Colorado:
University Press of Colorado, pp. 2533.
Savalev, V.N., Kozlov, M.P., Nadeina, V.P., and Reitblat,
A.G. (1978) A study of the microflora of fleas of rodents.
Med Parazitol Parazit Bol 47: 7374.
Shoemaker, N.B., Vlamakis, K., Hayes, K., and Salyers, A.A.
(2001) Evidence for extensive resistance gene transfer
among Bacteroides spp. and among Bacteroides and other
genera in the human colon. Appl Environ Microbiol 67:
561568.
Stotzky, G., and Babich, H. (1986) Survival of, and genetic
transfer by, genetically engineered bacteria in natural
environments. Adv Appl Microbiol 31: 93138.
Strand, M.A. (1977) Pathogens of Siphonaptera (fleas). Bull
World Health Org 55 (Suppl. 1): 279287.
Watanabe, T. (1963) Infective heredity of multiple drug resistance in bacteria. Bacteriol Rev 27: 87115.
Watanabe, K., and Sato, M. (1998) Plasmid-mediated gene
transfer between insect-resident bacteria, Enterobacter
cloacae, and plant-epiphytic bacteria, Erwinia herbicola, in
guts of silkworm larvae. Curr Microbiol 37: 352355.

2002 US Government, Molecular Microbiology, 46, 349354