Detection and differentiation of Borrelia burgdorferi sensu lato in ticks collected from sheep and cattle in China.

BACKGROUND: Lyme disease caused by Borrelia burgdorferi sensu lato complex is an important endemic zoonosis whose distribution is closely related to the main ixodid tick vectors. In China, isolated cases of Lyme disease infection of humans have been reported in 29 provinces. Ticks, especially ixodid ticks are abundant and a wide arrange of Borrelia natural reservoirs are present. In this study, we developed a reverse line blot (RLB) to identify Borrelia spp. in ticks collected from sheep and cattle in 7 Provinces covering the main extensive livestock regions in China.
RESULTS: Four species-specific RLB oligonucleotide probes were deduced from the spacer region between the 5S-23S rRNA gene, along with an oligonucleotide probe which was common to all. The species specific probes were shown to discriminate between four genomic groups of B. burgdorferi sensu lato i.e. B. burgdorferi sensu stricto, B. garinii, B. afzelii, and B. valaisiana, and to bind only to their respective target sequences, with no cross reaction to non target DNA. Furthermore, the RLB could detect between 0.1 pg and 1 pg of Borrelia DNA.A total of 723 tick samples (Haemaphysalis, Boophilus, Rhipicephalus and Dermacentor) from sheep and cattle were examined with RLB, and a subset of 667 corresponding samples were examined with PCR as a comparison. The overall infection rate detected with RLB was higher than that of the PCR test.The infection rate of B. burgdoreri sensu stricto was 40% in south areas; while the B. garinii infection rate was 40% in north areas. The highest detection rates of B. afzelii and B. valaisiana were 28% and 22%, respectively. Mixed infections were also found in 7% of the ticks analyzed, mainly in the North. The proportion of B. garinii genotype in ticks was overall highest at 34% in the whole investigation area.
CONCLUSION: In this study, the RLB assay was used to detect B. burgdorferi sensu lato in ticks collected from sheep and cattle in China. The results showed that B. burdorferi senso stricto and B. afzelii were mainly distributed in the South; while B. garinii and B. valaisiana were dominant in the North. Borrelia spirochaetes were detected in Rhipicephalus spp for the first time. It is suggested that the Rhipicephalus spps might play a role in transmitting Borrelia spirochaetes.

Background

Lyme borreliosis has been recognized as the one of most common vector-borne diseases in the world. The disease was first reported in the USA by Steere in 1975. Subsequently its clinical manifestations were described by a scholar in Yale University in 1977, and named Lyme disease in 1980 [1]. Burgdorfer and his colleagues isolated a spirochete and confirmed it as the causative agent of Lyme disease [2]. The spirochaete was formally named Borrelia burgdorferi in 1984 [3].

Lyme disease is distributed in over 30 countries and regions of Asia, Europe, America, Africa and Oceania, with more than 0.3 million clinical cases per year [4]. In the USA, it is particularly severe with approximately 20-100 cases per 100,000 [5]. The number of cases is increasing, and the disease was listed as a key target for prevention and control by WHO in 1992 [6].

In China, the disease was first identified in the forest regions of Hailin county in Heilongjiang province in 1986 [7]. Liang and Zhang latter confirmed that three B. burgdorferi isolates from the region caused Lyme disease [8]. To date, serological investigations have confirmed that Lyme disease is present in more than 29 provinces and autonomous regions, and more than 130 isolates of B. burgdorferi have been recovered from patients, ticks or animals in 19 provinces and autonomous regions [9,10], especially in the northeast forest areas. The prevalence of Lyme disease correlates with the geographic distribution and activity of vector ticks [11]. Of the 109 species of ticks identified in China, Lyme disease pathogens have been isolated from 9 Ixodid ticks, including Ixodes persulcatus, I. granulatus, I. acutitarsus, H. longicornis, H. bispinosa, H. concinna, H. formosensis, Boophilus microplus and Dermacentor silvarum[12].

The B. burgdorferi sensu lato complex comprises at least 13 species, B. burgdorferi sensu stricto, B. garinii, B. afzelii, B. lusitaniae, B. valaisiana, B. bissettii, B. andersonii, B. japonica, B. tanukii, B. turdi, B. sinica, B. spielmanii and B. californiensis. The following five species of B. burgdorferi have been isolated from ticks: B. burgdorferi sensu stricto (I. scapularis, I. dammini, I. ricinus and I. pacificus), B. garinii (I. ricinus and I. persulcatus), B. afzelii (I. ricinus and I. persulcatus), B. lusitaniae (I. ricinus) and B. valaisiana (I. ricinus and I. columnae) [13-17].

In this study, we developed a reverse line blot (RLB) to detect and differentiate 4 B. burgdorferi sensu lato species on the basis of the variable spacer region regions between 5S and 23S rRNA genes sequences. We then used the RLB to investigate the distribution and prevalence of B. burgdorferi sensu lato in China.

Results

Specificity of the RLB

Amplification of the 5S-23S rRNA gene internal transcribed spacer was performed by nested PCR on four isolates (Figure 1.). The primers specifically amplified the spacer region between 5S-23SrRNA genes sequence from Borrelia and no cross-reaction occurred with DNAs from M. pneumoniae, C. psittasi, Anaplasma. marginale and A. ovis. All probes bound to their respective target sequence only; no cross reaction was observed at dilutions used, resulting in the clear recognition of individual strains, species, or groups. Each of the 4 B. burgdorferi species was identified by one of the oligonucleotide probes (Table 1). Isolate BO23 was identified by two oligonucleotide probes: Sl (453-430), and a specific probe recognizing Af (305-278), isolate B31 was identified by two oligonucleotides probes: the Sl (453-430), and a specific probe recognizing Ss (322-299). Isolates SZ, T25, PBr, 20047, IP90 and TN were recognized by two oligonucleotides as well: Sl (453-430), and a specific probe recognizing Ga (322-298). Isolate VS116 was identified by two oligonucleotide probes: Sl (453-430), and a specific probe recognizing Vs (303-278), No signal was detected from genomic DNA of M. pneumoniae, C. psittasi, A. marginale and A. ovis, or water (Figure 2).

Figure 1

The second PCR amplification of the 5S-23S intergenic spacer region. Lanes 1-14:BO23: Borrelia. afzelii; B31: Borrelia.burgdorferi sensu stricto; SZ, T25, PBr, 20047, IP90, TN: Borrelia. garinii;VS116: Borrelia. valaisiana; Mycoplasma ovipneumoniane,, Chlamydia psittacii, Anaplasma marginale, Anaplasma ovis and pathogen-free tick DNA as negative controls.

Table 1

Species and their recognition pattern by oligonuclotide probes

Isolate designation	Species	Oligonucleotide*
		Species-specific	Group-specific
BO23	B. afzelii	Af
B31	B. burgdorferi s.s.	Ss
T25	B. garinii
PBr	B. garinii
SZ	B. garinii	Ga	Sl
20047	B. garinii
IP90	B. garinii
TN	B. garinii
VS116	B. valaisiana	Vs

*See Table 3 for oligonucleotide sequence data.

Figure 2

Reverse line blot (RLB) assay specificity test for . Sl: Probe of Borrelia burgdorferi sensu lato; Ga: Probe of Borrelia.garinii; Ss: Probe of Borrelia.burgdorferi sensu stricto; Af: Probe of Borrelia. afzelii;Vs: Probe of Borrelia. valaisiana; row 1-14: BO23: Borrelia. afzelii; B31: Borrelia.burgdorferi sensu stricto; SZ, T25, PBr, 20047, IP90, TN: Borrelia. garinii;VS116: Borrelia. valaisiana; Mycoplasma ovipneumoniane, Chlamydia psittacii infections, Anaplasma marginale, Anaplasma ovis and pathogen-free tick DNA as negative control.

Comparison of sensitivity of nested PCR and RLB

Sensitivity of the RLB was assessed using 10 fold diluted genomic DNA (10-1~10-12) of B. burgdorferi sensu stricto, B. garinii, B. afzelii, and B. valaisiana. The RLB system was capable of detecting 1pg of B. burgdorferi sensu stricto, B. afzelii, DNA and 0.1pg of B. garinii, B. valaisiana DNA. The detection level of the nested PCR carried out in parallel was also restricted to about 1pg~0.1pg (Figure 3). Unlike the nested PCR, the RLB was able to identify which species B. burgdorferi sensu stricto, B. garinii or B. afzelii, the tick was infected with.

Figure 3

The sensitivity of the RLB assay for B. burgdorferi sensu lato. Plate 1 shows the sensitivity of the nested-PCR using Ba and Bb probes; Plate 2 shows the sensitivity of RLB using Ba and Bb probes; Plate 3 shows the sensitivity of the nest-PCR using Bg and Bv probes; Plate 4 shows the sensitivity of the RLB of using Bg and Bv probes. Lane1-12: 100-11 10 × fold dilution DNA of Genome for B.burgdorferi sensu lato Row 1-6: oligonucleotide probe of B. burgdorferi sensu lato (Sl); oligonucleotide probe of B. burgdorferi sensustricto; (Ss); oligonucleotide probe of B. garinii( Ga); oligonucleotide probe of B. afzelii (Af); oligonucleotide probe of B. valaisiana (Vs); oligonucleotide probe of B. burgdorferi sensu lato (Sl).

Detection of B. burgdorferi sensu lato in ticks collected from sheep and cattle in China

The prevalence of each B. burgdorferi species was identified by RLB and PCR (Table 2). The number of samples in which B. burgdorferi sensu stricto, B. garinii, B. afzelii, or B. valaisiana alone were detected were 201 (28%), 245 (34%), 160 (22%), 85 (12%) respectively (Table 2), and 7% (50/723) were mixed infections and mainly in northern China. For the positive samples from southern China, it was found that the majority (40%) of them were infected with B. burgdorfei sensu stricto, while B. garinii was dominant (40%) in the positive samples from northern China. The results have revealed that most of the positive samples were infected by B. burgdorfei sensu stricto, B. garinii and/or B. afzelii. Sixty-nine samples (Shangzhi and Huichun), 15 samples (Lintan) and 1 sample (Huaihua) were detected from Dermacentor, Haemaphysalis and Boophilus, which belonged to B. valaisiana.

Table 2

Comparison of the results of examination of field tick samples by RLB and PCR

Origin	Genra of tick	No. of tick infected/examined
		RLB					PCR
		BB	BG	BA	BV	Mixa
Huaihua	Boophilus	62/146(42)	54/146(37)	33/146(23)	1/146(0.7)		44/146(30)
Nanping	Haemaphysali	3/39(7.6)	16/39(41)	1/39(2.6)			1/39(2.6)
Huizhou	Rhipicephalus	4/48(8.3)	7/48(14.6)	3/48(6.3)			14/45(31)
Laibin	Boophilus	41/42(98)	3/42(7)	41/42(98)			15/39(38)
Total in south		110/275(40)	80/275(29)	78/275(28)	1/275(0.4)		74/269(27)
Shangzhi	Dermacentor	44/154(29)	91/154(59)	44/154(29)	54/154(35)	33/154(21)	7/134(5)
Huichun	Dermacentor	24/160(15)	34/160(21)	16/160(10)	15/160(9.4)	9/160(5.6)	3/130(2.3)
Total in Northeast		68/314(22)	125/314(40)	60/314(19)	69/314(22)	42/314(13)	10/264(4)
Lintan	Haemaphysalis	23/134(17)	43/134(32)	22/134(16)	15/134(11)	8/134(6)	10/134(7)
Total		201/723(28)	248/723(34)	160/723(22)	85/723(12)	50/723(7)	94/667(14)

BB: B. burgdorferi sensu stricto, BG: B. garinii, BA: B. afzelii, BV: B. valaisiana

Mix: a tick is infected by more than one species of B. burgdorferi.

Discussion

Lyme disease is a tick-borne disease caused by Borrelia. Animals such as cattle, sheep, horses, dogs and rats can be infected by the pathogen, and in most cases, play a role as reservoirs with varied clinical symptoms [18-20]. Lyme disease spirochetes have been isolated and detected from the following tick vectors: Ixodes persulcatus, I. crenulatus, Haemaphysalis longicornis, H. japonica, Boophilus microplus, Dermacentor silvarum and D. nuttalli [21]. The main mode of transmission is via the salivary gland when infected ticks feed on mammalian hosts. The infection rate of I. persulcatus, I. granulates and H. bispinosa were 40%-50%, 16%-40% and 24% respectively in north of China, suggesting that I. persulcatus is the principal vector of Lyme disease spirochete in northern China, while I. granulates and H. bispinosa are the main vectors in the south of China [22].

Although Lyme disease was found more than 20 years ago in China, knowledge of the epidemiology of the causative organism in the tick vectors is limited. We therefore considered that it would be very useful to develop a validated method to detect all four Lyme disease pathogens associated with ticks to better understand the epidemiology of the disease in different regions of China.

The RLB assay was here developed into a useful diagnostic tool to simultaneously detect and differentiate B. burgdorferi subspecies in ticks. Each species can be identified by a species-specific oligonucleotide probe using a line-blotter apparatus which is quickly becoming a standard molecular tool for diagnostic and epidemiological studies in an increasing number of laboratories all over the world. In this study, a RLB assay was established based on the spacer region between 5S and 23S rRNA genes of B. burgdorferi senso lato. B. burgdorferi sensu stricto, B garinii, B. afzelii, and B. valaisiana as decribed Barandika et al [23]. All four Borrelia species could be differentiated with 4 species-specific oligonucleotides while the oligonucleotide probe Sl designed specifically for the genera of B. burgdorferi hybridized with all 4 species. Any new species or genotype that may present would therefore be detected by the SI probe but not by the species-specific probes. Meanwhile, in order to confirm whether there was false positives, the sequence analysis of 120 samples was conducted for amplicons (PCR product) from these samples by random sampling. In all cases, it was found that the fragment was about 400 bp and 250bp, and the homology was 99.5% and 99.2% with the 5S-23S rRNA intergenic spacer region gene of B. burgdorferi by DNAstar analysis, suggesting that the ticks were infected by B. burgdorferi sensu lato. The RLB was a reliable diagnostic tool and when PCR amplification steps were included, the specificity was further improved. Different strains can be detected concurrently, and coinfection of different strains of B. burgdorferi sensu lato could be distinguished.

We have shown here that the infection rate of Borrelia spirochaetes in ticks varies considerably between different geographical regions. B. burgdorferi sensu stricto is found more frequently in the South (Laibin and Huaihua) and B. garinii more frequently in the Northeast (Shangzhi and Huichun) and Northwest (Lintan), this is not accordance with the recent reports [24]. Our results do not agree with previous findings which suggest that B. afzelii is the dominant species in northeast China. However, in the present study, the B. afzelii infection rate of samples from the north was lower than those samples infected with B. garinii and B. burgdorferi sensu stricto. Although the infection rate of ticks with B. afzelii was 19%, which was much higher than that previously reported [25], the low infection rate of B. afzelii might be associated with the limited number of ticks collected from field. Another explanation could be that the tick infection rate of each of the four Borrelia varies among samples from different regions, and it is possible that samples in this study were collected from areas where the tick infection rate of B. afzelii is lower.

To the best of our knowledge, this is the first study in which Lyme disease species of B. burgdorferi sensu lato were detected by RLB in field ticks in China. The prevalence and distribution of Borrelia spirochaetes in ticks are key factors for risk assessment of Lyme borreliosis. From our results, it could be seen that B. garinii is the dominant species and widely distributed in China, which agrees with the previous data [26-28]. B. afzelii and B. burgdorferi sensu stricto have been found only in a limited number of ticks, however, the tick infection rate with B. burgdorferi sensu stricto was higher than that of B. afzelii, which is different from previously reported.

B. valaisiana has only been isolated from rats from Guizhou Province in China and it has been assumed that distribution of B. valaisiana is confined to the southwest of China [29]. In this study B. valaisiana was detected in relatively high levels in the samples from northeast China (Table 2). Masuzawa et al [30] have confirmed that B. valaisiana is present in Korea which is geographically very close to northeast China. The prevalence of B. valaisiana varied among our study sites. It could be inferred that the distribution of B. valaisiana is much wider than originally thought. Previous reports suggested that B. valaisiana was more difficult to isolate from clinically diagnosed cases of Lyme Borreliosis than genomospecies B. burgdorferi sensu stricto, B. afzelii and B. garinii [14,31]. As most of the current prevalence data on Lyme disease are from human medical institutions and identification of Borrelia is based on the isolation, it is reasonable to believe that in northeast China, B. valaisiana is more prevalent but may be misdiagnosed due to the difficulty of isolation of the pathogen.

We collected samples of four genera of ticks from sheep and cattle, which were Haemaphysalis (n = 173), Boophilus (n = 188), Rhipicephalus (n = 48) and Dermacentor (n = 314) from seven regions, where the primary forests are preserved. The infection rates of the four genera of ticks with the four species of Borrelia burgdorferi l were 15%, 34%, 13% and 0.8% (Haemaphysalis), 54%, 28%, 39% and 0.05% (Boophilus), 0.8%, 1.4%, 0.6% and 0% (Rhipicephalus), and 21%, 39%, 19% and 22% (Dermacentor); while the infection rates detected by PCR were 6% (Haemaphysalis), 32% (Boophilus), 31% (Rhipicephalus) and 4% (Dermacentor) [31]. Recently, Richter and Matuschka (2010) have found that although about a quarter of I. ricinus ticks questing on the pasture were infected by spirochetes, the positive rate of the ticks collected from goats and cattle was much lower, and no molted ticks that had previously engorged or repletion on ruminant harbored Lyme disease spirochetes [32]. The authors also concluded that the spirochetes are cleared from I. ricinus ticks during feeding on ruminants. To confirm this hypothesis, more engorged ticks of these four species should be collected to test the status of spirochete infection of their offspring. In the present study, the results indicated that the positive detection rates of the RLB technique were higher than that by PCR. We further confirmed the sensitivity of the RLB, and the detection results of the infection state of three genera of ticks are also consistent with two methods on the whole, except for the results with Rhipicephalus. It has been still unknown yet why the positive rate by PCR was higher than by RLB with the Rhipicephalus samples. Our data suggests that the infection rate of different genera of ticks is also variable. However, the differences in infection rates among tick species were not statistically significant, and these differences could be due to geographical and seasonal variations of tick and host populations rather than tick vector competence. Because the ticks were collected from animals which could have been already infected with Borrelia burgdorferi, we can therefore only say that these ticks are potential vectors. The data showed the infection rate of Boophilus with Borrelia was relatively high and this species might play a role in transmission of Lyme diseases [21,33]. As Boophilus does not readily attach to humans, it could transmit the pathogen between cattle and perhaps sheep, and these infected animals would become reservoirs for other tick species which will attach to humans. These results are consistent with previous studies and confirm that Boophilus plays an important role in Lyme disease transmission amongst livestock. It has been reported that Haemaphysalis and Dermacentor are the vectors of Lyme disease [27,34] yet Rhipicephalus spp. have not previously been confirmed as vectors. In this study, we detected Borrelia burgdorferi in Rhipicephalus spp suggesting that Rhipicephalus spp. might transmit Borrelia burgdorferi. In the present study, it was not possible to confirm that these four genera of ticks were capable of transmitting Borrelia sporochaetes, and the competence of the ticks as vector will need to be determined by further experimental study.

Conclusion

This study reports development of a RLB assay that is able to detect 4 species of B. burgdorferi sensu lato in field ticks simultaneously. The genomic DNA of B. burdorferi senso stricto and B. afzelii, B. garinii and B. valaisiana have been detected in samples prepared from 4 genera of ticks collected from seven different areas in China. These findings extend the knowledge of the epidemiology of Borrelia burgdorferi and its possible transmission by adult ticks in China. RLB is a powerful tool for such epidemiological studies and for further investigation of the association between tick vectors B. burgdorferi sensu lato species and clinical manifestations of Lyme disease. It can also be used for screening ticks and can easily be expanded to include additional Borrelia species. We also demonstrated that Rhipicephalus spps are the potential vectors of Borrelia burgdorferi. Several tick species may be involved in the transmission of Borrelia in China and the high infection rate suggests that there is a high risk of human infection. Little is known of the epidemiology, diagnosis and control of Lyme borreliosis in livestock, and more research is needed.

Methods

Borrelia strains and culture

Three B. burgdorferi sensu lato strains, B. burgdorferi sensu stricto (B31), B. afzelii (BO23) were purchased from American Type Culture Collection. B. garinii (SZ) was isolated from Dermacentor ticks as described by Niu et al. [35]. The spirochaetes were cultured in BSKII medium at 33°C as described previously [18]. DNA of standard Borrelia genotypes used in the study was provided by Dr. Fingerle (Nationales Referenzzentrum für Borrelien Max von Pettenkofer-Institut, LMU München) in Germany. The designations and origins of the strains are given in Table 1[34,13,36].

Control samples

DNA is from four species of bacteria, Mycoplasma ovipneumoniane, Chlamydia psittacii infections, Anaplasma marginale and Anaplasma ovis, were provided by colleagues in Lanzhou Veterinary Research Institute, CAAS and used as negative controls.

DNA from adult Haemaphysalis qinghaiensis was also used a control. The ticks were originally collected from sheep, goats and cattle in Lintan county, Gansu province. Prior to DNA extraction, the final generation of ticks was fed on splenectomised sheep to confirm absence of any tick borne pathogens. The examination of the sheep by microscopy and PCR proved that the sheep has not been infected by Borrelia. Furthermore, the same DNA was tested by both PCR and RLB, and the results were negative. Therefore, the remaining unfed adult ticks of H. qinghaiensis were designated a Borrelia-free strain of H. qinghaiensis.

Field samples

From March to August in 2009 (Figure 4), ticks (n = 723) were collected on cattle and sheep from Nanping in Fujian province, Huizhou in Guangdong province, Laibin in Guangxi province, Huaihua in Hunan province, Shangzhi in Heilongjiang province, Huichun in Jilin province and Lintan in Gansu province. Ticks were identified to genus using keys previously described by Deng and Jiang 1991 [37].

Figure 4

Map of China showing the number of ticks examined in each of the seven regions in this study.

Tick DNA extraction

Each tick was soaked in 70% ethanol in 15 min, dried, and ground in a separate 1.5 ml Eppendorf tube to avoid cross contamination. The sample was incubated with proteinase K for 2h at 56°C, and then boiled at 100°C for 10 min to inactivate proteinase K. After centrifugation, the supernatant was transferred to a fresh sterile microtube, and DNA was extracted using a Genomic DNA Purifcation Kit (Gentra, USA) according to the manufacturer's instructions.

Primers and probes

Two pairs of primers were designed from 5S-23S rRNA intergenic spacer region [38,39]. They are: 23SN1 (5'-ACCATAGACTCTTATTACTTTGAC-3' 469-446), 23SC (5'-TAAGCTGACTAATACTAATTACCC-3' 92-115) and 5SCB (5'-biotin-GAGAGTAGGTTATTGCCAGGG-3' 243-263), 23SN2 (5'-ACCATAGACTCTTATTACTTTGACCA-3' 469-444).

Species-specific RLB oligonucleotide probes were deduced from the hypervariable 5S-23S rRNA gene intergenic spacer region (Table 3). All the specific oligonucleotide probes contained a N-(trifluoroacetamidohexyl-cyanoethyl, N, N-diisopropyl phosphoramidite [TFA])-C6 amino linker and were diluted to 10-1,200 pmol/150 μl in 500 mM NaHCO3 (pH 8.4) to provide optimal sensitivity and specificity (summarized in Table 4).

Table 3

Sequence, concentration and position of specific oligonucleotide probes for B. burgdorferi sensu lato

Type of primers and isolate	Oligonucleotide Sequence(5'-3')	Concentration used(pmol)	Position on 5S-23Sintergenic spacer region
23SN1	ACCATAGACTCTTATTACTTTGAC	50	469-446
23SC	TAAGCTGACTAATACTAATTACCC	50	92-115
5SCB	biotin-GAGAGTAGGTTATTGCCAGGG	50	243-263
23SN2	ACCATAGACTCTTATTACTTTGACCA	50	469-444
B.burgdorfei sensu lato (S1)	a-CTTTGACCATATTTTTATCTTCCA	800	453-430
B. burgdorferi sensu stricto (Ss)	a-AACACCAATATTTAAAAAACATAA	20	322-299
B. garinii (Ga)	a-AACATGAACATCTAAAAACATAAA	10	322-298
B. afzelii (Af)	a-AACATTTAAAAAATAAATTCAAGG	200	305-278
B. valaisiana (Vs)	a-CATTAAAAAAATATAAAAAATAAATTTAAGG	10	303-278

Table 4

Strain designation, species, and geographic origin of isolates used in this study

Strain	Species	Origin	Geographic origin	Reference
BO23	B. afzelii	Skin	Germany	13
B31	B. burgdorferi s.s.	Ixodes dammini	United States	13
SZ	B. garinii	Dermacentor	China	35
T25	B. garinii	Ixodes ricinus	Germany	13
PBr	B. garinii	Human (CSF)	Germany	13
20047	B. garinii	I. ricinus	France	36
IP90	B. garinii	I. persulcatus	Russia	36
TN	B. garinii	Ixodes ricinus	Germany	13
VS116	B. valaisiana	Ixodes ricinus	Switzerland	17

Nested PCR amplification

All DNA samples of ticks and B. burgdorferi sensu lato isolates were amplified in duplicate by nested PCR. Genomic DNA of M. pneumoniae, Chlamydia psittacii, A. marginale and A. ovis, and water were used as controls. Primers, Taq DNA polymerase and buffers were obtained from TakaRa (Dalian, China). The first PCR was performed in a reaction volume of 50 μl. For each sample, the PCR mixture was prepared as follows: H2O 38.7 μl, 10 × reaction buffer (200 mM Tris-HCl (pH 8.55), 160 mM (NH4)2SO4 and 20 mM MgCl2) 5 μl, 10 mM dNTP 4 μl, 50 pmol each of primers 23 SN1 and 23 SC, Taq polymerase 1.8 U and 1 μl of DNA sample or water. PCR amplification was performed in an automatic DNA thermocycler (Eppendorf). The program for the first PCR consisted of an initial denaturation at 94°C for 3 min followed by the thermal cycle reaction program of 1 min at 94°C, 90 s at 50°C, and 90 s at 72°C for 40 cycles with a final extension step at 72°C for 5 min. Samples were held at 4°C until analysis. 1 μl of the product from the first PCRs was added to the second PCR tubes containing reaction mixture which were then were briefly vortexed, centrifuged and then transferred to a thermal cycler. After a denaturation step, (1 min at 94°C), 40 rounds of temperature cycling (94°C for 30 s, 55°C for 30 s, and 72°C for 1 min) were performed.

Analysis of amplified products and sequence analysis

All PCR products were separated electrophoretically in 1.5% agarose and visualized under UV light after ethidium bromide staining.

Nine independent positive clones for each B. burgdorferi sensu lato isolates were sequenced using the BigDye Terminator Mix (TaKaRa Company, China). The sequences of the first PCR was size of approximately 362-392 bp, and the second PCR product was 218-235 bp, respectively. The homology was 96.8-98.7% amongst the other isolates in GenBank by DNAstar analysis, confirming that they were from genomic DNA of Borrelia species.

Reverse line blot hybridization

Protocols for preparation of RLB membrane and hybridization were carried out as previously described [40] with the following modifications: Biodyne C membrane was cut into a 14.5 cm square and activated by 10 min incubation in 10 ml freshly prepared 16% (w/v) 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDAC) in demineralized water, in a rolling bottle at room temperature. The membrane was then washed with agitation in demineralized water for 2 min and placed on a supporting cushion in a clean miniblotter system. Each probe was diluted in with water to give a range of concentrations (800 pmol, 400 pmol, 200 pmol, 100 pmol, 50 pmol, 20 pmol and 10 pmol), aliquoted into the miniblotter slot and incubated for 1 min. The solutions were aspirated and the membrane incubated in 100 mM NaOH for 10 min, and then washed at 42°C for 5 min in 2 × SSPE, 0.1% (SDS). Subsequently, the membrane was placed into the miniblotter, perpendicular to its previous orientation. The denatured PCR samples were aliquoted into the slots of the miniblotter for 10 min at 42°C, then aspirated and the membrane washed at 42°C for 10 min in 2 × SSPE, 0.1% SDS. Subsequently, the membrane was treated at 42°C for 45~60 min with peroxidase-labeled streptavidin diluted 1:4,000 in 2 × SSPE/0.1% SDS, washed twice at 42°C for 10 min, and washing twice at room temperature for 5 min in 2 × SSPE, 0.1% SDS. Finally detection of binding was by chemiluminescence performed according to manufacturer's recommendations (Santa Cruz Biotechology ).

Specificity and sensitivity of the RLB

The RLB was performed by 20 μl of 100ng amplified product of B. burgdorferi sensu stricto, B. garinii, B. afzelii, and B. valaisiana extracted. Meanwhile, Mycoplasma ovipneumoniane, Chlamydia psittacii, Anaplasma marginale, Anaplasma ovis and pathogen-free tick DNA as negative control were used as negative controls. The sensitivity of the RLB was tested by hybridizing 20 μl aliquots of 10-fold serially diluted PCR product. The genomic DNA of the four pathogens was serially diluted from 100ng/μl to 0.000001pg/μl by 10-fold dilution as templates for the PCR, and then hybridized by RLB.

Epidemiological study

A total of 723 ticks were examined by the established RLB assay and the positive and negative samples were recorded to assess the Borrelia spirochaetes infection rate in the four genera of ticks. Furthermore, 667 of these same samples were tested with the PCR.

Authors' contributions

QLN carried out the samples detection, RLB analysis and drafted the manuscript. HY, JXL participated in the design of the study. GQG, JFY, YGF, ZKX, YQL, MLM, JLL, AHL and QYR participated in sampling. WJ reviewed and commented on the manuscript. All authors read and approved the final manuscript.