Frequency and Distribution of Rickettsiae, Borreliae, and Ehrlichiae Detected in Human-Parasitizing Ticks, Texas, USA

To describe the presence and distribution of tickborne bacteria and their vectors in Texas, USA, we screened ticks collected from humans during 2008–2014 for Rickettsia, Borrelia, and Ehrlichia spp. Thirteen tick species were identified, and 23% of ticks carried bacterial DNA from at least 1 of the 3 genera tested.

T icks are vectors for a variety of microorganisms, many of which are known agents of zoonotic disease. Although much current research is focused on areas where these diseases are common, it is crucial to collect data from areas with fewer diagnoses of tickborne illness. In Texas, USA, tickborne diseases caused by Rickettsia, Borrelia, and Ehrlichia bacteria are diagnosed less frequently than in some areas of the United States (1); however, those agents have been documented to occur (2), and many medically relevant tick species, capable of carrying and transmitting these pathogens, are established in various geographic areas of Texas (1). Long-term surveillance data encompassing consecutive seasons and a wide geographic range are necessary to ascertain disease transmission risks associated temporally or geographically with established or emerging tickborne pathogens and their vectors. The University of North Texas Health Science Center Tick-Borne Disease Research Laboratory (UNTHSC-TBDL), the primary ticktesting facility for Texas Department of State Health Services Zoonosis Control (TX DSHS), receives ticks continually throughout the year. The data collected from this testing provide an assessment of the prevalence of tick species and associated tickborne bacterial agents collected in Texas.

The Study
From October 1, 2008, through September 30, 2014, ticks removed from humans were sent by TX DSHS to UNTHSC-TBDL, where they were tested by using PCRbased methods, then underwent by DNA sequence analysis to determine the presence of Rickettsia, Borrelia, and Ehrlichia spp. Morphologic identification of tick species was implemented by entomologists at TX DSHS. Ticks that could not be classified morphologically were identified at UNTHSC-TBDL by sequencing mitochondrial 16S rDNA (data not shown).
Each tick was sent to UNTHSC-TBDL in an individual collection tube. Upon arrival, ticks were processed according to the laboratory's standard protocol, as described by Williamson et al. (2). After bead pulverization, we extracted DNA using the E.Z.N.A. Mollusc DNA Isolation Kit (Omega Bio-Tek, Norcross, GA, USA) following the manufacturer's protocol.
DNA from each specimen was screened in duplicate by PCR for Rickettsia, Borrelia, and Ehrlichia spp. as previously described (2) by using primers listed in Table 1. PCR products were evaluated, and presumptive-positive amplicons were purified for sequencing (2). Cycle sequencing reactions were performed in both directions by using Big-Dye Terminator version 3.1 chemistry (Life Technologies, Carlsbad, CA, USA). Dideoxy chain termination products were detected electrophoretically on an ABI 310 or 3130xL Genetic Analyzer (Life Technologies). Sequence analysis was performed by using Sequencher version 4.8/5.0 (Gene-Codes, Ann Arbor, MI, USA). Analyzed sequences were compared with reference data in GenBank (http://blast. ncbi.nlm.nih.gov/). Sequences were submitted to GenBank under accession nos. KP861333-KP861347.
The  (1,10), so this finding could suggest a novel geographic association. The total prevalence of borreliae detected was 1.1%. DNA sequences sharing 100% identity to B. lonestari were found in 8 A. americanum ticks (1.4%). As seen by Stromdahl et al., the B. lonestari isolates matching sequences in this study depended on the insertion or deletion of a nucleotide triplet, AAG (11). Sequences from 7 tick samples matched 100% with B. lonestari flaB isolates containing the additional triplet (accession no. AY850063), and 1 sequence was identical to B. lonestari flaB isolates lacking the triplet (accession no. AY850064). Of the 8 A. americanum    (12). Borrelia 16S rDNA primers produced nonspecific amplification with these 2 samples. Phylogenetic analysis was performed by using MEGA version 5.1 (http://www.megasoftware.net) using GenBank reference sequences to examine relationships between the Borrelia sp. from this study, B. turcica, and both Lyme disease-associated and relapsing fever borreliae (Figure). The results supported findings by Lee et al. that the novel Borrelia sp. flaB sequences were more closely related to the reptilian Borrelia than the other 2 Borrelia groups (12).
Two A. americanum ticks contained DNA sharing 100% identity with Ehrlichia chaffeensis dsb (accession no. CP000236). One of these ticks was co-infected with Candidatus R. amblyommii. Prevalence of E. chaffeensis in the A. americanum specimens tested was 0.34%. In addition, 2 of 42 A. maculatum ticks tested for the emerging pathogen Panola Mountain Ehrlichia sp. (PME) (7) each produced a map1 sequence that was 100% homologous to 2 separate PME reference sequences (accession nos. EU272356, EU272358). These sequences differed from each other by 1 single nucleotide polymorphism. This finding represents a novel association, as A. americanum is the known vector for PME (7). A subset of 141 A. americanum ticks was also tested for PME, with negative results.

Conclusions
Frequency of tickborne zoonoses in Texas remains low compared with some regions of the United States. We report the detection of known pathogens along with bacteria of unknown pathogenicity in human-parasitizing ticks commonly found in Texas. Our findings underscore the importance of better characterization and continued surveillance of the frequency and distribution of tick species and the bacterial agents they carry. Continued monitoring in low-risk areas provides data regarding the presence of potential emerging pathogens and vectors not yet commonly identified, which could pose unidentified threats to public health.

Figure.
Maximum-likelihood tree showing that the novel Borrelia sp. identified in Amblyomma maculatum ticks from Texas in this study (box) and from Mississippi (12) shares a closer phylogenetic relationship to B. turcica than to to other Borreliae groups. Analysis is based on flaB sequences (267 bp). GenBank accession numbers are shown in parentheses. Tree was constructed using the Tamura 3-parameter model with a bootstrap value of 1,000 replicates. Scale bar indicates substitutions per nucleotide position.
Sign up for Twitter and find the latest information about emerging infectious diseases from the EID journal. @CDC_EIDjournal