Identification of Ehrlichia spp. and Borrelia burgdorferi in Ixodes Ticks in the Baltic Regions of Russia

Abstract

The presence and distribution of Ehrlichia spp. and Borrelia burgdorferi sensu lato was demonstrated among ixodid ticks collected in the Baltic regions of Russia, where Lyme borreliosis is endemic. A total of 3,426 Ixodes ricinus and 1,267 Ixodes persulcatus specimens were collected, and dark-field microscopy showed that 265 (11.5%) I. ricinus and 333 (26.3%) I. persulcatus ticks were positive. From these samples, 472 dark-field-positive and 159 dark-field-negative ticks were subjected to PCR and subsequent reverse line blot hybridization. Fifty-four ticks (8.6%) carried Ehrlichia species, and 4 (0.6%) carried ehrlichiae belonging to the Ehrlichia phagocytophila complex, which includes the human granulocytic ehrlichiosis agent. The E. phagocytophila complex and an Ehrlichia-like species were detected only in I. ricinus whereas Ehrlichia muris was found exclusively in I. persulcatus, indicating a possible vector-specific infection. Borrelia garinii was found predominantly in I. persulcatus, but Borrelia afzelii was evenly distributed among the two tick species. Only two I. ricinus ticks carried B. burgdorferi sensu stricto, while Borrelia valaisiana and a newly identified B. afzelii-like species were found in 1.7 and 2.5% of all ticks, respectively. Of the dark-field-positive ticks, only 64.8% yielded a Borrelia PCR product, indicating that dark-field microscopy may detect organisms other than B. burgdorferi sensu lato. These observations show that the agent of human granulocytic ehrlichiosis may be present in ticks in the Baltic regions of Russia and that clinicians should be aware of this agent as a cause of febrile disease.

Borrelia burgdorferi sensu lato, the causative agent of Lyme disease, is frequently found in a variety of tick species throughout the world. The widespread distribution has made Lyme borreliosis the most prevalent tick-transmitted zoonotic disease in humans (29). In Europe, five different B. burgdorferi sensu lato species are found, of which Borrelia afzelii and Borrelia garinii are present throughout the continent. In countries of central Europe, such as The Netherlands, Germany, Italy, and France, B. burgdorferi sensu stricto and Borrelia valaisiana are found (28). Borrelia lusitaniae is mainly found in Ixodes ricinus ticks in Portugal but has also been detected in ticks from the Czech Republic, Moldavia, Ukraine, and Belorussia (10). Previous surveys have shown 10 to 30% of the Ixodes ticks from the St. Petersburg and Kaliningrad regions of Russia carried B. afzelii and B. garinii, but not B. burgdorferi sensu stricto (1). In a similar study it was shown that the prevalence of B. burgdorferi sensu lato infection of I. ricinus ticks from nearby Helsinki, Finland, varied from 19 to 55% (7). In addition, various other studies have shown that ticks in different regions of Russia are infected with B. burgdorferi sensu lato (8, 11, 26).

Ehrlichioses are known as important tick-borne diseases in animals, particularly in wildlife and domestic ruminants (25). During the last decade, two previously unknown Ehrlichia species have emerged as agents causing considerable public health problems in the United States (2, 5, 32). The Ehrlichia species that was identified in 1986 as causing human monocytic ehrlichiosis is designated Ehrlichia chaffeensis. Initially it was assumed that this species was transmitted by Amblyomma americanum only, but in a recent study researchers have also identified this Ehrlichia species in Ixodes ticks (9). Until now there have been no reports of confirmed cases of human monocytic ehrlichiosis outside the United States. More recently, the agent causing human granulocytic ehrlichiosis (HGE) was identified. This species belongs to the Ehrlichia genogroup 2, which includes Ehrlichia phagocytophila, an organism isolated from ruminants like sheep, cattle, and deer, and Ehrlichia equi, which is found in horses. These organisms, designated the E. phagocytophila complex, are closely related and may even represent the same species. There have been several reports that described the presence of the HGE agent in ticks in Europe (3, 12, 20, 27). However, there seems to be a paucity of reports of confirmed cases of HGE in Europe, and most of these cases are found in Slovenia where Lyme borreliosis is highly prevalent (14, 17, 19, 20, 30).

The aim of this study was to determine the prevalence of Ehrlichia infection in ticks in a region where diseases like tick-borne encephalitis and Lyme borreliosis are highly prevalent. For this reason, I. ricinus and Ixodes persulcatus ticks were collected from the Baltic regions of Russia and screened for the presence of B. burgdorferi sensu lato by dark-field microscopy, followed by PCR and subsequent reverse line blot hybridization to identify the Borrelia and Ehrlichia species present in these ticks.

MATERIALS AND METHODS

Ticks.

From 1997 to 1998, 4,693 ticks were collected by flagging in forested areas near Morskaja and Lisy Nos in the vicinity of St. Petersburg and along the Curonian Spit in the Kaliningrad enclave of Russia. Ticks were stored alive and analyzed by dark-field microscopy within hours after collection.

Dark-field examination of ticks.

For dark-field inspection the contents of the gut from a dissected tick were ejected into a drop of saline. After being immediately covered with a thin cover glass, the slide was inspected under a microscope for the presence of live spirochetes, with 250 fields viewed. The remainder of the tick was stored in 70% ethanol at 4°C for PCR analysis.

Preparation of DNA extracts from ticks.

Ticks were processed as described before (23). Briefly, ticks were taken from the 70% ethanol solution, briefly dried, and boiled for 20 min in 100 μl of 0.7 M ammonium hydroxide to free the DNA. After being cooled, the vial with the lysate was left open and incubated for 20 min at 90°C to evaporate the ammonia. The tick lysate was either used directly for PCR or stored at −20°C until use.

Construction of spike DNA.

To assess the efficiency of the PCRs we constructed plasmids that carried the sequences complementary to the primer sequences and used these as internal spikes in the Ehrlichia and Borrelia PCRs. For construction of the Ehrlichia spike, primers were designed that carried a hybrid of the sequences of a restriction site, the Ehrlichia 16S rRNA priming sites, and primer sequences encompassing a 508-bp region of the tmpB gene of Treponema pallidum. The restriction sites were incorporated into the primers to facilitate cloning and subcloning of the PCR products but were not used in this study. For the construction of a spike control for the Borrelia PCR, a similar approach was used. In the latter case, hybrids between the sequences of a restriction site, the Borrelia primer sequences, and primer sequences encompassing a 589-bp region of the tmpA gene of T. pallidum were used. PCRs were run with T. pallidum DNA, and the PCR products were cloned directly into a TA-TOPO vector (Invitrogen, Groningen, The Netherlands). The plasmids were isolated and purified with the Qiagen Plasmid Minikit (Hilden, Germany) and used as spike controls with the Ehrlichia (tmpB spike) and Borrelia (tmpA spike) PCR. The appropriate spike concentration was determined by titrating the amount of spike in a PCR with serial dilutions of Ehrlichia and Borrelia DNA. The amount of spike DNA used allowed the detection of a single copy of the specific target. The composition of the spike construction primers and the spike probes is presented in Table 1.

TABLE 1.

Oligonucleotide primers and probes used in PCR and hybridization assays

Oligonucleotide primer or probe	Sequence	Target organism	Target gene	Nucleotide position	Reference or source
Borrelia-specific primers
B-5SBor	5′biotin-GAGTTCGCGGGAGAGTAGGTTATT	B. burgdorferi sensu lato	5S rRNA (rrfA)	82–105	This study
23SBor	TCAGGGTACTTAGATGGTTCACTT	B. burgdorferi sensu lato	23S rRNA (rrlB)	208–185	This study
Probes
SL	5′amino-CTTTGACCATATTTTTATCTTCCA	B. burgdorferi sensu lato	23S-5S spacera	182–167	24
SS	5′amino-AACACCAATATTTAAAAAACATAA	B. burgdorferi sensu stricto	23S-5S spacer	59–36	24
GA	5′amino-AACATGAACATCTAAAAACATAAA	B. garinii	23S-5S spacer	58–35	24
AF	5′amino-AACATTTAAAAAATAAATTCAAGG	B. afzelii	23S-5S spacer	44–21	24
VS	5′amino-CATTAAAAAAATATAAAAAATAAATTTAAGG	B. valaisiana	23S-5S spacer	51–21	24
Ruski	5′amino-GAATAAAACATTCAAATAATATAAAC	B. afzelii like	23S-5S spacer	67–42	This study
Ehrlichia-specific primers
16S8FE	GGAATTCAGAGTTGGATCMTGGYTCAG	Eubacteria	16S rRNA gene	8–27	27
B-GA1B	5′biotin-CGGGATCCCGAGTTTGCCGGGACTTCTTCT	Ehrlichia genus	16S rRNA gene	476–456	27
Probes
A-EhrAll	5′amino-TTATCGCTATTAGATGAGCC	Ehrlichia genus	16S rRNA gene	203–222	27
A-HGE	5′amino-GCTATAAAGAATAGTTAGTGG	HGE agent	16S rRNA gene	87–107	27
A-Phago	5′amino-TTGCTATAAAGAATAATTAGTGG	E. phagocytophila	16S rRNA gene	85–107	27
A-D-HGE	5′amino-GCTATGAAGAATAGTTAGTG	HGE agent variant	16S rRNA gene	87–106	27
A-D-Phago	5′amino-TTGCTATGAAGAATAATTAGTG	E. phagocytophila variant	16S rRNA gene	87–106	27
A-ESchot	5′amino-GCTGTAGTTTACTATGGGTA	Ehrlichia like	16S rRNA gene	76–95	27
A-ECan	5′amino-TCTGGCTATAGGAAATTGTTA	Ehrlichia canis	16S rRNA gene	85–105	27
A-EChaf	5′amino-ACCTTTTGGTTATAAATAATTGTTA	E. chaffeensis	16S rRNA gene	83–107	27
A-EmurisC	5′amino-GCTATAGGTTCGCTATTAG	E. muris C variant	16S rRNA gene	85–103	This study
A-EmurisT	5′amino-AGCTATAGGTTTGCTATTAGT	E. muris T variant	16S rRNA gene	86–104	This study
Spike primers
23SBortmpA	CGGGATCCC-TCAGGGTACTTAGATGGTTCACTT-ACCCCGAACTTTTCTCCb	T. pallidum	tmpA	199–215	This study
5SBortmpA	GGAATT-CGAGTTCGCGGGAGAGTAGGTTATT-CTTCCCCAGTTCTGTGTAGA	T. pallidum	tmpA	725–706	This study
EhrFtmpB	GCTCTAGAG-AGAGTTTGATCCTGGCTCAG-CGCCTATCCGACCGAGTAT	T. pallidum	tmpB	308–326	This study
EhrRtmpB	CGGGATCCC-GAGTTTGCCGGGACTTCTTCTA-CCACGTAGTCACCTTGTATT	T. pallidum	tmpB	756–736	This study
Probes
A-TmpABor	5′amino-TACAATCTAAGATCCAGACT	T. pallidum	tmpA	346–365	This study
A-TmpBEhr	5′amino-GAAAACTCGAAGAACAAAGAA	T. pallidum	tmpB	502–522	This study

^aThe SL probe targets the spacer and the first eight bases of the 23S gene sequence (rrlB). 

^bThe dashes in the spike primers indicate the separation between the restriction site sequences, the Borrelia or Ehrlichia sequence, and the tmpA- or tmpB-specific sequence. 

PCR amplifications.

PCR amplifications were performed with an Omnigene thermal cycler (Hybaid, Ltd., Teddington, United Kingdom). DNA amplification was done with 50-μl reaction volumes. For the amplification of Ehrlichia DNA, each reaction mixture contained 10 fg of tmpB spike DNA, 80 pmol of primers 16S8FE and B-GA1B, 1.2 U of SuperTaq DNA polymerase (HT Biotechnology, Ltd., Cambridge, United Kingdom), 0.26 μg of the TaqStart antibody (Clontech Laboratories, Palo Alto, Calif.), 0.1 U of uracil-DNA glycosylase (UDG) (GibcoBRL, Life Technologies B.V., Breda, The Netherlands), deoxynucleoside triphosphates (100 μM dUTP, 100 μM dTTP, 200 μM dATP, 200 μM dGTP, and 200 μM dCTP), and SuperTaq buffer (10 mM Tris-HCl [pH 9.0], 50 mM KCl, 1.5 mM MgCl₂, 0.01% stabilizer, 0.1% Triton X-100). A 25-μl overlay of paraffin oil was added to the tubes, followed by 5 μl of the tick DNA extract. The mixture was incubated for 3 min at 37°C to promote UDG activity, followed by a 10-min incubation at 94°C to inactivate the UDG. To minimize nonspecific amplification a touchdown-up PCR program was used: two cycles of 20 s at 94°C, 30 s at 65°C, and 30 s at 72°C; and then two cycles with conditions identical to the previous cycles, but with an annealing temperature of 63°C. During the subsequent two cycle sets, the annealing temperature was lowered by 2°C until it reached 55°C. Then, an additional 20 cycles, each 20 s at 94°C, 30 s at 55°C, and 30 s at 72°C, and 20 cycles, each 20 s at 94°C, 30 s at 63°C, and 30 s at 72°C, followed the touchdown program. The PCR was ended by an extra incubation for 7 min at 72°C and held at 65°C to keep the UDG inactive. For the amplification of B. burgdorferi sensu lato DNA, similar conditions were used. However, in this PCR, 1 fg of tmpA spike DNA, 20 pmol of the primers 23SBor and B-5SBor, 2.5 U of SuperTaq, and 0.55 μg of TaqStart were used. In addition, the DNA was amplified in a regular touchdown PCR ranging from 60 to 50°C, with 40 cycles at the touchdown temperature.

To monitor for the occurrence of false-positive PCR results, negative controls were included during extraction of the tick samples: one control sample for each five tick samples with a minimum of two controls. In addition, each time that PCR was performed, negative- and positive-control samples were included. To minimize contamination, the reagent setup, the extraction and sample additions, and the PCR and sample analyses were performed in three separate rooms, the first two rooms of which were kept at positive pressure and had airlocks.

Reverse line blot hybridization.

The reverse line blotting technique was performed as described before (23, 24) with some modifications. Briefly, solutions with 5′-amino-linked oligonucleotide probes ranging from 6 to 800 pmol were coupled covalently to an activated Biodyne C membrane in a line pattern with a miniblotter (Immunetics, Cambridge, Mass.). After the oligonucleotide probes were bound, the membrane was taken from the miniblotter, washed in 2× SSPE (360 mM NaCl, 20 mM Na₂HPO₄·H₂O, 2 mM EDTA) with 0.1% sodium dodecyl sulfate (SDS) at 60°C and placed in the miniblotter again with the oligonucleotide lines perpendicular to the slots. Ten microliters of the biotin-labeled PCR product was diluted in 150 μl of 2× SSPE–0.1% SDS, denatured for 10 min at 99°C, and cooled rapidly on ice. The slots of the miniblotter were filled with the denatured PCR product, and hybridization was performed for 1 h at 42°C. The membrane was removed from the miniblotter and washed twice for 10 min in 2× SSPE–0.1% SDS at 51°C. Subsequently, the membrane was incubated for 30 min at 42°C with streptavidin-peroxidase (Boehringer GmbH, Mannheim, Germany), diluted 1:4,000 in 2× SSPE–0.5% SDS, and washed twice for 10 min in 2× SSPE–0.5% SDS. Hybridization was visualized by incubating the membrane with ECL detection liquid (Amersham International plc, Den Bosch, The Netherlands) and exposing an X-ray film (Hyperfilm; Amersham) to the membrane. For species identification the biotinylated Ehrlichia PCR product was hybridized with seven different oligonucleotide probes in the reverse line blot assay. Similarly, the biotinylated spacer fragment of the Borrelia PCR was hybridized with five B. burgdorferi genospecies-specific oligonucleotide probes. All primers and probes are displayed in Table 1.

DNA sequencing and data analysis.

PCR products, used for DNA sequencing, were purified with Qiaquick PCR purification kits (Qiagen). For DNA sequencing reactions, the fluorescence-labeled dideoxynucleotide technology was used. The sequenced fragments were separated, and the data was collected with ABI 377 and ABI 3700 automated DNA sequencers (PE Biosystems, Nieuwerkerk aan de Ijssel, The Netherlands). The collected sequences were assembled, edited, and analyzed with the DNAStar package (Madison, Wis.). The phylogenetic tree and multiple alignments were constructed with the alignment and clustering modules in the Bionumerics package (Applied Maths, Kortrijk, Belgium).

Nucleotide sequence accession numbers.

The partial 16S rRNA gene sequence of Ehrlichia muris and the 5S-23S intergenic spacer region of the B. afzelii-like organism found in this study are available in the GenBank database under accession numbers AF312907 and AF312906, respectively.

RESULTS

PCR detection of Borrelia and Ehrlichia DNAs in ticks.

In this study 2,305 I. ricinus and 1,267 I. persulcatus ticks were collected by flagging, and screening by dark-field microscopy for the presence of spirochetes showed that 265 (11.5%) of the I. ricinus ticks and 333 (26.3%) of the I. persulcatus ticks were dark-field positive. By random choice, 215 dark-field-positive and 80 dark-field-negative I. ricinus ticks and 257 dark-field-positive and 79 dark-field-negative I. persulcatus ticks were selected and analyzed by PCR and subsequent reverse line blot hybridization to detect and identify Borrelia and Ehrlichia species. In 338 (53.6%) of 631 ticks analyzed, Borrelia DNA was detected whereas Ehrlichia DNA was found in 54 (8.6%) ticks (Table 2). In 296 samples (46.9%), only Borrelia DNA was found, and in 12 samples (1.9%) only Ehrlichia DNA was detected. Forty-two (6.7%) extracts of all 631 ticks carried both Borrelia and Ehrlichia DNA.

TABLE 2.

Comparison of dark-field microscopy and reverse line blot detection of Borrelia and Ehrlichia DNA in ticks

Tick species and result by dark-field microscopy	No. of ticks	Reverse line blota
		Borrelia		Ehrlichia
		Positive	Negative	Positive	Negative
I. ricinus
Positive	215	97	118	24	191
Negative	80	15	65	1	79
Total	295	112	183	25	270
I. persulcatus
Positive	257	209	48	29	228
Negative	79	17	62	0	79
Total	336	226	110	29	307
Total	631	338	293	54	577

^aValues are numbers of ticks. 

Comparison of dark-field results and detection of Borrelia and Ehrlichia by PCR.

In 209 of the 257 (81.3%) dark-field-positive I. persulcatus ticks, B. burgdorferi sensu lato DNA was detected (Table 2). In contrast, only 97 of the 215 (45.1%) dark-field-positive I. ricinus ticks carried B. burgdorferi sensu lato DNA. Fifteen of the 80 (18.8%) dark-field-negative I. ricinus and 17 of the 79 (21.5%) dark-field-negative I. persulcatus ticks carried B. burgdorferi sensu lato DNA. Ehrlichia DNA was detected in 29 (11.2%) dark-field-positive I. persulcatus ticks and in 24 (11.1%) dark-field-positive I. ricinus ticks. Only one (1.3%) dark-field-negative I. ricinus tick carried Ehrlichia, and none of the dark-field-negative I. persulcatus ticks carried detectable Ehrlichia DNA.

Distribution of Borrelia and Ehrlichia species in ticks.

The majority of the Borrelia-positive ticks carried B. afzelii and/or B. garinii DNA. There was a remarkable difference in distribution of the various Borrelia species between I. persulcatus and I. ricinus (Table 3). Of the 226 Borrelia-positive I. persulcatus ticks 107 (47.3%) contained B. garinii, 50 (22.1%) carried B. afzelii, and 57 (25.2%) carried both B. afzelii and B. garinii. In contrast, 80 (71.4%) out of 112 Borrelia-positive I. ricinus ticks carried B. afzelii, 13 (11.6%) carried B. garinii DNA, and only 1 tick (0.9%) was infected with both B. afzelii and B. garinii. Only one I. persulcatus tick was triple infected by B. afzelii, B. garinii, and B. valaisiana, and two I. ricinus ticks carried B. burgdorferi sensu stricto DNA.

TABLE 3.

Distribution of Ehrlichia and Borrelia spp. in I. ricinus and I. persulcatus ticks

Reverse line blot result	Tick species		Total
	I. ricinus	I. persulcatus
Borrelia positive	112	226	338
B. afzelii	80	50	130
B. afzelii + B. valaisiana	6	2	8
B. afzelii like	9	7	16
B. garinii	13	107	120
B. garinii + B. afzelii	1	57	58
B. garinii + B. afzelii + B. valaisiana		1	1
B. garinii + B. valaisiana	1	2	3
B. burgdorferi sensu stricto	2		2
Borrelia negative	183	110	293
Ehrlichia positive	25	29	54
E. muris		29	29
E. phagocytophila	1		1
HGE	3		3
Ehrlichia like	21		21
Ehrlichia negative	270	307	577
Total no. of ticks	295	336	631

Sixteen of all Borrelia-positive ticks reacted with the B. burgdorferi sensu lato probe only. Therefore, we sequenced the 5S-23S intergenic spacer of six of these ticks and found that they carried Borrelia species with identical spacer sequences which were similar to but distinct from the spacer of B. afzelii (Fig. 1). We therefore designated this species as B. afzelii like, designed a new probe (Ruski) for use with the reverse line blot, and hybridized the PCR products of all 16 samples with the new probe. As expected, all 16 samples reacted with this probe, and no cross-hybridization with other species-specific probes was seen.

FIG. 1.

Multiple aligment and clustering of the 23S-5S intergenic spacer region of B. burgdorferi sensu lato and reverse line blot result using the amplified spacer sequences. Multiple alignment and clustering were performed with the complete 174- to 182-bp spacer region and are only partially displayed. The coordinates of the B. burgdorferi sensu stricto spacer sequence are indicated above the sequences, and the probe regions are denoted by grey boxes. Reverse line blot probes are indicated as follows: VS, B. valaisiana; RS, B. afzelii like; AF, B. afzelii; GA, B. garinii; SS, B. burgdorferi sensu stricto; SL, B. burgdorferi sensu lato.

Of the 25 Ehrlichia-positive I. ricinus ticks, 3 carried HGE DNA, 1 contained E. phagocytophila DNA, and 21 carried DNA from an organism that was identified before in Dutch ticks and designated as an Ehrlichia-like organism (27). In 29 I. persulcatus ticks, DNA from Ehrlichia species was found that initially reacted with the Ehrlichia genus-specific probe only. From these samples the PCR products were sequenced, and comparison showed that their sequences were identical. Alignment with DNA sequences in the GenBank data bank revealed a close similarity (>99%) with E. muris 16S rRNA gene sequences. We thereafter designed two E. muris-specific probes that differed only in one residue targeting position 95 of the 16S rRNA gene and hybridized all 29 above-mentioned PCR products with these probes. All samples reacted with the E. muris C probe but not with the E. muris T probe.

There was no significant difference between Borrelia or Ehrlichia infection rates in the different sexes or stages of the I. ricinus ticks investigated (Table 4). However, in I. persulcatus ticks approximately 75% of the adults and only 35.5% of the nymphs were infected with Borrelia. Similarly, about 10% of adult I. persulcatus ticks carried Ehrlichia DNA, whereas only 1.6% of nymphs did. None of the 24 larvae carried any detectable Borrelia or Ehrlichia DNA.

TABLE 4.

Distribution of infection type, stratified by tick species, sex, and stage

Infection status	Tick species	Sex		Stage		Total
		Female	Male	Nymph	Larva
Infected + noninfected	I. ricinus	107	89	77	22	295
	I. persulcatus	154	118	62	2	336
Borrelia infected	I. ricinus	47	35	30	0	112
	I. persulcatus	112	92	22	0	226
Ehrlichia infected	I. ricinus	13	4	8	0	25
	I. persulcatus	15	13	1	0	29
Double infected	I. ricinus	8	4	6	0	18
	I. persulcatus	13	11	0	0	24

DISCUSSION

In the study presented here we have shown that Ehrlichia species were found in 8.6% of ixodid ticks collected from vegetation in the Baltic region of Russia. Members of the granulocytic E. phagocytophila group, including some reacting with our HGE agent probe, were found in I. ricinus but not in I. persulcatus. However, the rate of HGE infection in these ticks was low (1%). This prevalence of infection is comparable to that found by several other research groups in Europe (13, 16, 18, 21). However, other European studies have reported infection rates ranging from 3.2% in Slovenia to 28.9% in The Netherlands (6, 20, 27, 31), and some studies from the United States even reported a prevalence of infection in Ixodes scapularis up to 50% (4, 15). The explanation for these large differences in prevalence of infection needs to be established with comparative studies. About 7.1% of the I. ricinus ticks were infected with an Ehrlichia-like organism previously identified in I. ricinus ticks found in The Netherlands and Italy (27). Remarkably, this species was not found in I. persulcatus ticks from the same regions, suggesting a vector specificity of this Ehrlichia species. Although the true nature of this species remains unknown, phylogenetic analysis based on the 16S rRNA gene sequence indicates a close relationship with the monocytic group of ehrlichiae. Nearly 9% of the I. persulcatus ticks were infected with an Ehrlichia species with a 16S rRNA gene sequence nearly identical to one of the published E. muris sequences. This monocytic Ehrlichia species was identified only in I. persulcatus ticks, again suggesting a vector-specific Ehrlichia infection. In a recent study, I. persulcatus ticks from Perm, Russia (region of Ural Mountains), were tested for the presence of Ehrlichia by PCR using Ehrlichia-specific primers (22). None of the ticks carried detectable HGE DNA. However, 5 out of 35 ticks tested yielded a 16S rRNA PCR product of which the DNA sequence was shown to be identical to that of E. muris. This result confirms our finding that E. muris was the only Ehrlichia species found in I. persulcatus ticks.

Analysis of the ticks for the presence of Borrelia species corroborated earlier findings on infection rates of ticks in the Baltic region. Speciation by reverse line blot hybridization showed that many ticks (10.6%) carried more than one Borrelia species. The majority of the ticks were infected with B. afzelii and/or B. garinii; B. burgdorferi sensu stricto was found in only two tick extracts. In 2.5% of the ticks a B. burgdorferi species was found which carried a 5S-23S rRNA spacer sequence similar to, but distinct from, that of B. afzelii. This B. afzelii-like species was found both in I. persulcatus and I. ricinus. We have not investigated whether coinfection of this species with other B. burgdorferi sensu lato species occurs. Furthermore, we have no data that indicates that this species is pathogenic for humans. There was no significant difference between infection rates of I. persulcatus and I. ricinus with B. afzelii. In contrast, of the 182 ticks infected with B. garinii only 15 (8.2%) were I. ricinus ticks. As with the distribution of the Ehrlichia species this suggests a vector-specific infection. It is uncertain whether this represents true vector specificity or a relationship between the vector and its host range. To determine whether an association between pathogen and tick species exists, additional studies, including studies of the vector hosts such as large and small mammals and birds, are required.

In our study, none of the larvae contained any detectable Borrelia or Ehrlichia DNA, but the number of larvae included in this study is too low to draw any conclusions from this result. However, a more significant observation was that I. persulcatus adults are infected with either Borrelia or Ehrlichia species more than twice as often as nymphs. This was true if either the dark-field examination or the PCR result was used as an indicator of Borrelia infection. Such a stage-dependent prevalence of infection was not seen in the I. ricinus ticks. In fact, 60 to 80% of the I. ricinus adults and nymphs included in the study were dark-field positive, and 38 to 43% of the same ticks were positive with the Borrelia PCR.

The latter finding showed that in a large proportion of the dark-field-positive ticks no Borrelia DNA was detected. A possible explanation for this observation is that dark-field microscopy is more sensitive than Borrelia PCR in detecting Borrelia. However, the detection of Borrelia DNA in 18% of the dark-field-negative samples argues against that. Furthermore, we have shown in previous studies that PCR, followed by the reverse line blot method, is both sensitive and specific. False-negative results due to inhibition of the PCR were excluded by the use of internal spike controls. There is a possibility that the DNA of the Borrelia species was degraded during storage, but earlier experiments in our lab have shown that Borrelia DNA in ticks stored in ethanol is extremely stable. In addition, the fact that there is a significant difference in Borrelia DNA detection between I. persulcatus and I. ricinus dark-field-positive ticks also suggests that the microorganisms seen in the dark-field analysis may represent species other than B. burgdorferi sensu lato and that this species is found more frequently in I. ricinus than in I. persulcatus. If this hypothesis is correct, studies that use dark-field microscopy as the sole method of determining the rate of Borrelia infection in ticks may yield an overestimated prevalence of infection. Further PCR studies using primers that will amplify the 16S rRNA gene of a broad range of bacteria and subsequent sequence analyses will have to corroborate this theory and identify the species.

To our knowledge, there have been no reports until now in the scientific literature in English of cases of ehrlichiosis in any part of Russia. However, in the city of St. Petersburg, many cases of Lyme disease are reported (250 cases per year; 5 cases per 100,000 inhabitants), and the majority of these cases are patients who sustained tick bites in the suburban region of the city. It is this region where we collected the Borrelia- and Ehrlichia-infected ticks used in this study. It is therefore conceivable that some people in the St. Petersburg region may have been infected with the HGE agent. Due to the lack of sufficient diagnostic tools and the lack of awareness, these cases of human ehrlichiosis may have gone unnoted. However, the rate of Ehrlichia infection in the ticks is about 10- to 20-fold lower than the rate of Borrelia infection. This would suggest that the chance of infection with Ehrlichia after a tick bite is significantly lower than the chance of being infected with Borrelia but that such exposure may still account for 10 to 25 cases of ehrlichiosis in the St. Petersburg area. Therefore, awareness of possible cases of ehrlichiosis remains important in tick-infested areas like the suburban region of St. Petersburg.