Tick-borne Agents in Rodents, China, 2004–2006

Many rodent species may be involved in the enzootic maintenance of these agents.

in Xinjiang Autonomous Region was a forest with a rural resident population.Rodents were trapped by using peanuts as bait.After captured rodent species were identifi ed, spleen specimens were collected and stored at -20°C until DNA was extracted.

Extraction of DNA
Total DNA was extracted from spleen samples by using Trizol agent (Invitrogen, Carlsbad, CA, USA) following the instructions of the manufacturer.Briefl y, ≈300 mg of spleen tissue from each rodent was crushed in Trizol reagent, and DNA was separated from RNA by centrifugation.DNA was precipitated after washing twice in a solution containing 0.1 M sodium citrate in 10% ethanol.The DNA pellet was then washed in 75% ethanol and kept at room temperature for 10-20 min.After centrifugation at 2,000 × g at 2-8°C for 5 min, DNA was dissolved in 8 mmol/L NaOH and centrifuged to remove insoluble material.The supernatant containing DNA was removed and adjusted with HEPES buffer to a pH of 7-8.

PCR
Nested PCR was conducted with primers designed to amplify part of the 16S rRNA gene of A. phagocytophilum, as described (8).For amplifi cation of B. burgdorferi DNA, a nested PCR was performed with primers derived from B. burgdorferi 5S-23S rRNA intergenic spacer (9).PCR was performed by using primers Rr 190.70p and Rr 190-701n to amplify a fragment of the gene encoding a 190-kDa outer membrane protein A (ompA) gene specifi c for SFG rickettsiae (10).Samples were tested for F. tularensis by a nested PCR specifi c for the outer membrane protein (fopA) gene, as described (11).All PCRs were performed by using a model 2700 thermal cycler (Perkin-Elmer, Waltham, MA, USA).PCR products were separated by agarose gel electrophoresis, stained with ethidium bromide, and examined under UV light.To avoid contamination, we performed DNA extraction, reagent setup, amplifi cation, and agarose gel electrophoresis in separate rooms and included negative controls (distilled water) were in each amplifi cation.

DNA Sequencing and Analysis
PCR products of positive samples were sequenced directly by using a dideoxynucleotide cycle sequencing method with an automated DNA sequencer (ABI PRISM 377; Perkin-Elmer).To limit errors in sequencing, we performed 2 sequencing reactions of each PCR product.When different sequences were obtained, additional sequencing reactions were conducted to generate a consensus sequence.Sequences obtained in the present study were compared with the corresponding sequences deposited in GenBank by using the BLAST program of the National Center for Biotechnology Information (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Statistical Analysis
Chi-square or Fisher exact tests were used to compare proportions.p values <0.05 were considered statistically signifi cant.

Results
A total of 705 rodents were captured.The number of rodents tested and infectivity rates at different survey sites are shown in the In Heilongjiang Province, all 4 agents were detected in rodents at similar frequencies (χ 2 2.80, df 3, p = 0.424).No SFG rickettsiae were detected in rodents from Jilin Province.The infectivity rates for the 3 agents in Jilin Province did not signifi cantly differ (χ 2 2.23, df 2, p = 0.328).Infectivity rates for SFG rickettsiae were signifi cantly higher than those for B. burgdorferi and F. tularensis in rodents from Inner Mongolia Autonomous Region (χ 2 39.76, df 2, p<0.001).Infectivity rates for the 3 agents in Xinjiang Autonomous Region did not differ signifi cantly (χ 2 5.01, df 2, p = 0.082).Except for F. tularensis, the other 3 agents showed similar infectivity rates for Zhejiang Province (χ 2 1.30, df 2, p = 0.523).Only B. burgdorferi and SFG rickettsiae were found in Guizhou Province, and the difference A total of 18 (2.6%,95% confi dence interval 1.4%-3.8%)rodents from all survey sites except Xinjiang Autonomous Region were positive for 2 or 3 agents, among which 15 were positive for 2 agents.To confi rm the presence and determine genotypes of detected organisms, PCR products were sequenced and analyzed.A 919-bp partial 16S rRNA gene fragment for A. phagocytophilum was obtained from each positive specimen (8) For identifi cation of SFG rickettsiae, partial nucleotide sequences of the ompA gene were obtained from positive specimens in Heilongjiang Province and Inner Mongolia Autonomous Region.All sequences were identical to those of the R. sibirica genotype (GenBank accession no.U43807).Nucleotide sequences of 35 specimens positive for F. tularensis were identical to each other and to published sequences for the F. tularensis subsp.holarctica strain (GenBank accession no.AF247642.2).

Discussion
We detected A. phagocytophilum, B. burgdorferi, SFG rickettsiae, and F. tularensis in diverse species of rodents from different areas of China.Our fi ndings and previous evidence (6,9,(12)(13)(14)(15) suggest that several tick-borne agents cocirculate in mainland China, and a variety of rodent species may be involved in enzootic maintenance of these agents.
This study was not intended to be a comprehensive survey on active infections with A. phagocytophilum, B. burgdorferi, SFG rickettsiae, and F. tularensis.Rather, it was designed to investigate the presence and extent of these   1), we could not determine the geographic diversity of these agents in rodents.The number of rodents examined was limited; therefore, infectivity rates in the current study could be biased.In addition, because intensity of circulation of any vectorborne agent fl uctuates dramatically throughout the year and from year to year, even at the same location (20,21), we could not justify comparing infectivity rates between different sites on the basis of unsynchronized single collections over a 3-year period.A randomized sampling scheme and further collection of rodents are required to clarify this issue.Unfortunately, we did not collect the ticks from captured rodents for additional testing of the tick-transmitted agents.This limitation prevented us from understanding vector potential.
In this study, A. phagocytophilum was detected only in eastern China (Table, Figure), where it coexists with the other 3 agents (online Appendix Table 1).A. phagocytophilum detected in Heilongjiang, Jilin, and Zhejiang provinces were closely related to each other by 16S rRNA gene sequence analysis, but less related to other known strains in other countries.B. burgdorferi was detected in rodents from all 6 survey sites.As observed in a previous study (9), B. garinii was the dominant genospecies in mainland China, and the B. valaisiana-related group was present in southern China.
SFG rickettsiae, including ≈20 species of rickettsiae, can be transmitted to animals and humans not only by ticks but also by other arthropods such as infected lice, fl eas, and mites (10).In this study, we amplifi ed the ompA gene, which is present in most SFG rickettsiae (10,22).The overall infectivity rate for SFG rickettsiae was highest (9.1%) among the 4 agents tested (online Appendix Table 1).Sequence analysis identifi ed the Rickettsia sp.detected in Heilongjiang Province and Inner Mongolia Autonomous Region as a genotype of R. sibirica, which is known to cause Siberian tick typhus (18).However, we did not sequence PCR products amplifi ed from rodents at other study sites because of a limited amount of samples.Although sequence analysis of the ompA gene fragment is not suffi cient to identify the agent (22), it is commonly used to recognize tick-borne Rickettsia spp. in fi eld surveys (23).
F. tularensis was found only in northern China, which verifi es our belief that F. tularensis is present only north of 30°N latitude.In many disease-endemic areas, ticks are known to play a role in transmitting F. tularensis from animal hosts to humans, although other arthropods such as deer fl ies, fl eas, mites, and mosquitoes are known to carry the bacterium.Sequence analysis showed that all F. tularensis detected in this study belong to the subspecies holarctica.
Interference of infections among A. phagocytophilum, B. burgdorferi, SFG rickettsiae, and F. tularensis in rodent hosts is not clear.Our fi ndings indicate that infection with A. phagocytophilum does not intensify risk for transmission of the other 3 agents and vice versa.B. burgdorferi in rodents appears to increase risk for infection with F. tularensis but does not increase the possibility of infection with SFG rickettsiae or A. phagocytophilum.Further investigations are needed to demonstrate positive or negative interactions of the pathogens and to establish whether this interference is associated with the animal species.
Of 705 rodents tested in this study, 15 were infected with 2 agents and 3 were infected with 3 agents.These fi ndings indicate that mixed natural foci of tick-borne agents are present at the study sites.Because A. phagocytophilum, B. burgdorferi, SFG rickettsiae, and F. tularensis were found in ticks collected in the study areas (6)(7)(8)(9)(12)(13)(14), it is not surprising that multiple agents were detected in rodents.Coexistence of multiple agents might be caused by a single bite of a tick infected with several agents or multiple bites of ticks infected with at least 1 agent.The presence of 4 pathogens in the study areas demonstrates the risk for multiple infections in humans, which may lead to variations and exacerbation of clinical signs (1).Therefore, differential diagnoses should be made for febrile patients with a history of tick bites in these areas, particularly when clinical signs are atypical for 1 disease or a related disease.
Among 23 rodent species trapped in this study, 21 were infected with >1 agent (online Appendix Table 2).Only 2 species (Cricetulus migratourius and N. fulvescens) were negative for all 4 agents.Which species is the main host of each agent remains unknown, because none of the agents are predominantly associated with 1 or a few related rodent species, regardless of their geographic origin.However, A. phagocytophilum, B. burgdorferi, SFG rickettsiae, and F. tularensis in various rodent species illustrate the potential roles of various rodents in maintaining these tickborne agents.Systematic epidemiologic studies that investigate characteristics of natural foci and the role of small wild animals in transmission of these agents to humans are needed.

RESEARCH
This study was supported by the National Science Fund for Distinguished Young Scholars (no.30725032) and the National Natural Science Foundation of China (no.30600506).Dr Zhan is an epidemiologist at the Beijing Institute of Microbiology and Epidemiology.Her primary research interests are epidemiology and control of emerging and reemerging infectious diseases.

Figure .
Figure.Study sites (triangles) in the People's Republic of China where rodents were collected, 2004-2006.Numbers in parentheses are co-infection rates of rodents with 2 or 3 tick-borne agents
. A. phagocytophilum sequences detected in rodents from Heilongjiang and Jilin provinces (GenBank accession no.DQ342324) were identical and differed from those from Zhejiang Province (GenBank accession no.DQ458808) by 2 bp, from those from ticks in United Kingdom and Sweden (GenBank accession nos.AY082656 and AJ242784.1,respectively) by 2 bp, and from other known A. phagocytophilum sequences by >3 bp.Sequence analysis of the B. burgdorferi 5S-23S rRNA intergenic spacer showed that agents isolated from rodents in Heilongjiang Province, Inner Mongolia Autonomous Region, Jilin Province, and Xinjiang Autonomous Region belonged to the B. garinii genospecies, similar to agents detected in ticks (GenBank accession no.DQ150540) in northern China.Of 16 B. burgdorferi detected in Zhejiang Province, 12 belonged to the B. garinii genospecies and the other 4 belonged to the B. valaisiana-related group (GenBank accession nos.EU160458 and EU160459).The 2 strains found in Guizhou Province also belonged to the B. valaisiana-related group (GenBank accession no.EU247840).

Table .
(16)(17)(18)(19)for 4 tick-borne agents in rodents, People's Republic of China, 2004-2006* No. (%) rodents positive in China.If one considers that human infections with A. phagocytophilum, B. burgdorferi, SFG rickettsiae, and F. tularensis have been reported in various regions of China(16)(17)(18)(19), the presence of these agents in rodents in the study areas suggests a potential threat to humans, and the public health role of these fi ndings should be further investigated.Although infectivity rates varied at different survey sites (Table, online Appendix Table agents