Assessment of Polymicrobial Infections in Ticks in New York State

Introduction

P olymicrobial infections of ticks may result in transmission of several pathogens following a single tick bite. In the United States, the tick responsible for the majority of reported tick-borne infections is Ixodes scapularis. This tick is a vector of several human pathogens, most notably the bacteria Borrelia burgdorferi and Anaplasma phagocytophilum, as well as the protozoan Babesia microti (Spielman et al. 1979, Burgdorfer et al. 1982, Pancholi et al. 1995). B. burgdorferi is the etiological agent of Lyme disease, the most common vector-borne disease in the United States with approximately 20,000 cases reported to the Centers for Disease Control and Prevention each year (Bacon et al. 2008). Infections with A. phagocytophilum result in approximately 500 reported annual cases of human granulocytic anaplasmosis (HGA) (CDC 2008). B. microti causes babesiosis, an erythrocytic infection of unknown prevalence due to it not being a reportable disease. In addition, several other important microbes can be transmitted by I. scapularis. Among these is Powassan virus, a flavivirus in the tick-borne encephalitis complex (Telford et al. 1997). Two distinct genotypes of Powassan virus exist: genotype II found in I. scapularis and genotype I found only in other Ixodid ticks (Ebel et al. 2001, Kuno et al. 2001). Although rare, Powassan encephalitis is nonetheless clinically important; only 40 cases have been diagnosed since its initial description in 1958; however, Powassan encephalitis has a case fatality rate of 10–15%. Further, the incidence may be increasing; seven cases were reported in 2007 (CDC 2001, Hinten et al. 2008). Along with B. burgdorferi, another Borrelia species, Borrelia miyamotoi, has been detected in I. scapularis (Fukunaga et al. 1995, Scoles et al. 2001). Genetically, B. miyamotoi is very similar to “Borrelia lonestari,” a spirochete detected in Amblyomma americanum ticks (Fukunaga et al. 1996, Burkot et al. 2001). Phylogenetic analysis placed both B. miyamotoi and B. lonestari with Borrelia species associated with relapsing fever, although the pathogenicity of both species is unknown.

We recently described a multiplex method for rapid and economic screening of tick-borne pathogens (Tokarz et al. 2009). This method, MassTag polymerase chain reaction (PCR), enables efficient, sensitive screening of individual ticks for polymicrobial infection. Here we report its application for tick surveillance in Lyme disease endemic areas in New York State.

Materials and Methods

All PCRs were performed using cDNA generated from individual adult ticks. Live ticks were collected from vegetation in five separate locations in the two counties (Westchester and Suffolk) of New York State in the fall of 2008 (Fig. 1). Locations 1 and 2 are in Westchester County, whereas locations 3, 4, and 5 are in Suffolk County. Prior to nucleic acid extraction, ticks were immersed in 75% ethanol for 15 min, washed twice with phosphate-buffered saline, and then homogenized in Tri-reagent LS (Molecular Research Center, Cincinnati, OH). Total RNA was resuspended in 20 μL of H₂O. cDNA was generated in a 20 μL reaction with Superscript II Reverse Transcriptase (Invitrogen, Carlsbad, CA) using 10 μL of total RNA. MassTag PCR was performed using the tick panel (Tokarz et al. 2009), with additional primers for the detection of Powassan virus and relapsing fever Borrelia species (Table 1). All MassTag PCR assays utilized 3 μL of cDNA. Reaction conditions were 95°C for 5 min, one cycle at 95°C for 20 s, 65°C for 20 s (annealing), 72°C for 30 s, followed by 11 cycles with annealing temperature decreased by −1°C at each cycle. The final PCR was run for 37 cycles at the annealing temperature of 54°C. For sequencing, PCR products were size-fractionated on ethidium bromide–stained 1% agarose gel, observed under ultraviolet illumination, excised, and gel purified for dideoxy sequencing.

OPEN IN VIEWER

FIG. 1.

Map indicating tick collection sites within New York State.

Results

MassTag PCR was used to screen adult I. scapularis ticks collected from five locations in two different counties surrounding New York City (Fig. 1) for polymicrobial infection. The number of ticks screened from each location ranged from 24 to 98, with a total of 286 ticks analyzed. We detected at least one organism in 204 ticks (71%), of which 85 (30% of total) had a polymicrobial infection (Table 2). About 82 ticks (29%) did not have an infection with any of the microbes surveyed.

B. burgdorferi coinfection with B. miyamotoi and Powassan virus

B. miyamotoi is a relapsing fever-like Borrelia species. This organism was much less frequently detected; only seven ticks (2%) were positive for this organism but all were also coinfected with B. burgdorferi. Infection with Powassan virus was also infrequent. We detected the virus in seven ticks (2% of total). Five of these were collected at location 2 and a single Powassan virus–positive tick each was collected from locations 1 and 3. The majority of these did not exhibit a coinfection; only two of the Powassan virus–positive ticks were coinfected with B. burgdorferi, and another was coinfected with A. phagocytophilum, whereas the remaining four did not harbor any of the other infecting agents we screened for. For confirmation of Powassan virus–positive samples, we amplified by PCR and sequenced a 395-base pair fragment of Powassan virus (Fig. 2A) from each positive tick. All sequences represented genotype II of Powassan virus.

OPEN IN VIEWER

FIG. 2.

Confirmation of MassTag polymerase chain reaction (PCR) results. (A) Powassan virus PCR using E gene primers on seven tick samples positive for Powassan virus by MassTag PCR (NS 5 gene; see Table 1). Lanes 1–7 represent tick samples, NTC, no template control; L, 100 base pair DNA ladder. (B) Confirmatory PCR of a polymicrobial infection in a single tick. Lane 1, Borrelia burgdorferi ospA; lane 3, Babesia microti 18S rRNA; lane 5, Anaplasma phagocytophilum gltA; lane 7, Borrelia miyamotoi flaB. Lanes 2, 4, 6, and 8 represent ospA, 18S rRNA, gltA, and flaB no template controls, respectively.

Discussion

We screened adult I. scapularis ticks collected from five different areas for pathogen infections. We found that 71% of the ticks were infected with at least one organism. In addition, we also detected a large number of coinfections; the number of ticks coinfected with at least two microbes was higher than the number of uninfected ticks. We also detected that 5% of the ticks were infected with triple infections and a single tick with four pathogens. Our results indicate that in the areas surveyed, coinfection of adult I. scapularis may be just as common as the lack of infection.

A. phagocytophilum and B. microti were each detected in 20% of the ticks. Surveillance studies indicated that the prevalence of these organisms in ticks varies greatly, though infection rates of >20% have been shown (Piesman et al. 1986, Schwartz et al. 1997, Schauber et al. 1998). High coinfection rates can increase the likelihood of multiple infections in humans following a single tick bite, a phenomenon already documented (Benach et al. 1985, Magnarelli et al. 1998, Belongia et al. 1999, De Martino et al. 2001, Krause et al. 2002). Lyme disease and HGA are the two most common diseases reported resulting from a bite of I. scapularis, although reported Lyme disease cases outnumber HGA by >30 to 1 (CDC 2008). In our analysis, tick infection by these pathogens was not dramatically different, with a ratio of approximately 3:1. Nearly a quarter of ticks infected with B. burgdorferi were also coinfected with A. phagocytophilum, implying that in the areas sampled, B. burgdorferi transmission may be accompanied by A. phagocytophilum transmission in up to 25% of the cases. Similar rates of coinfection with B. burgdorferi and B. microti also indicate high probability of cotransmission of these two organisms as well. The large discrepancy in HGA and Lyme disease cases may reflect failure in diagnosis, underreporting of HGA cases, or lower rates of A. phagocytophilum prevalence in some geographic areas. HGA may also be mild or subclinical. Further, little is known about the transmission dynamics of coinfecting microbes. In ticks hosting multiple pathogens, the microbial interaction may result in variable microbial loads, and such differences could lead to alteration in the transmission efficiency for a given microbe.

In addition, we screened for Powassan virus and B. miyamotoi, two microbes of much lower prevalence and not detected at all surveyed sites. Powassan virus has been reported to infect 1–5% of the ticks in the North Central United States, but prevalence studies of other areas are lacking (Ebel et al. 1999, Brackney et al. 2008). We detected Powassan virus and B. miyamotoi in 2% of the ticks we screened. The two microbes had different frequency of coinfection with other pathogens. All seven I. scapularis infected with B. miyamotoi were also coinfected with B. burgdorferi. This is in contrast to Powassan virus–positive ticks, where in four out of the seven infected ticks we did not detect a coinfection. Whether this was a chance result or whether those ticks picked up the virus by feeding on an animal refractory to B. burgdorferi infection is unclear.

Polymicrobial infection following tick bites is an occurrence of which clinically much is still unknown. Our study indicates high coinfection prevalence in ticks within the areas surveyed. How these organisms interact during transmission and disease remains to be determined, but may have important impact on diagnosis and treatment of tick-borne diseases.