Growth Dynamics and Antibiotic Elimination of Symbiotic Rickettsia buchneri in the Tick Ixodes scapularis (Acari: Ixodidae)

This paper describes the growth of symbiotic Rickettsia buchneri within Ixodes scapularis through the life cycle of the tick and provides methods to eliminate R. buchneri from I. scapularis ticks.

ABSTRACT

Rickettsia buchneri is the principal symbiotic bacterium of the medically significant tick Ixodes scapularis. This species has been detected primarily in the ovaries of adult female ticks and is vertically transmitted, but its tissue tropism in other life stages and function with regard to tick physiology is unknown. In order to determine the function of R. buchneri, it may be necessary to produce ticks free from this symbiont. We quantified the growth dynamics of R. buchneri naturally occurring in I. scapularis ticks throughout their life cycle and compared it with bacterial growth in ticks in which symbiont numbers were experimentally reduced or eliminated. To eliminate the bacteria, we exposed ticks to antibiotics through injection and artificial membrane feeding. Both injection and membrane feeding of the antibiotic ciprofloxacin were effective at eliminating R. buchneri from most offspring of exposed females. Because of its effectiveness and ease of use, we have determined that injection of ciprofloxacin into engorged female ticks is an efficient means of clearing R. buchneri from the majority of progeny.

IMPORTANCE This paper describes the growth of symbiotic Rickettsia buchneri within Ixodes scapularis through the life cycle of the tick and provides methods to eliminate R. buchneri from I. scapularis ticks.

INTRODUCTION

Ixodes scapularis, the blacklegged tick of North America, is an important vector of the agents of Lyme disease, human granulocytic anaplasmosis, babesiosis, ehrlichiosis associated with Ehrlichia muris subsp. eauclairensis, and Borrelia miyamotoi disease, as well as Powassan virus (1). Although I. scapularis is not known to transmit pathogens in the genus Rickettsia, spotted-fever-group rickettsiae are commonly detected in these ticks by PCR. This is due to the presence of a rickettsial endosymbiont, recently described as Rickettsia buchneri (2). This bacterium appears to be very widely distributed in I. scapularis, with confirmed examples found in New England (3–5), the Upper Midwest (2, 6), Tennessee (7), Texas (8), Pennsylvania, Indiana, and Maine (9). There is growing interest in the microbiome associated with tick vectors of pathogens, as there is evidence that tick microbial communities may affect their vectorial capacities (10, 11). However, a large part of the microbiota associated with ticks is related to their environment (12, 13) and is variable, and its effects are not well defined.

In the present study, we considered that endosymbiont microbes obligately associated with ticks have formed these stable partnerships because of a mutualistic interdependence. For example, Rickettsia peacockii, an endosymbiont of the Rocky Mountain wood tick, Dermacentor andersoni, shares an ancestor with human-infectious, tick-borne Rickettsia rickettsii, but it lost its ability to infect vertebrates following extensive genome rearrangements due to acquisition of a mobile element (14). However, it has retained its ability to infect ticks and has been linked to disruption of transovarial transmission of R. rickettsii. It is thought to contribute to the survival of its tick host’s offspring by interfering with infection of the developing oocytes by R. rickettsii, which is destructive to ticks (15, 16) by reducing fertility and reproductive fitness (17, 18). Similarly, Rickettsia buchneri (previously known as the rickettsial endosymbiont of I. scapularis [REIS]) is associated primarily with the ovarian tissue of I. scapularis. Although it is regularly detected by PCR in males and immature ticks, tissue tropism in these life stages is unknown. Because of the conserved nature of this symbiont and its widespread persistence in geographically divergent I. scapularis populations, its presence most likely provides a selective advantage to the tick host (19, 20). One possibility is that R. buchneri provides nutritional supplementation to its host, an arrangement common in other, exclusively blood-feeding arthropods (21). Genetic analysis has demonstrated a functionally complete 6-gene biotin synthesis operon on the pREIS2 plasmid of R. buchneri, supporting the hypothesis that the symbionts contribute essential nutrients to their tick hosts that feed on nutritionally deficient vertebrate blood (22). A further potential benefit provided by R. buchneri could result from rickettsiae that have gained access to the hemocoel as extracellular bacteria, where they could be ingested by hemocytes and stimulate the innate immune responses of the vector. Presumably, this would render the tick less susceptible to parasitism or infection by pathogens, a situation referred to as immune priming, which has been documented in a variety of symbiont-insect relationships (23–25). In tsetse flies (Glossina spp.), symbiont immune priming is necessary for the normal development of the immune system (26–28).

Limited experimental research has been performed to determine the contributions of tick-borne symbionts to their vectors. A few studies have described attempts to manipulate symbiotic bacteria within ticks by exposure to antibiotics. Injection of engorged female ticks with rifampin, kanamycin, or tetracycline reduced the quantity of symbiotic Rickettsia or Coxiella spp. present, and a concomitant decline in egg laying and hatching was notable (29, 30). On the other hand, injection of engorged ticks with ciprofloxacin minimally affected egg production and hatchability but significantly reduced or eliminated the ovarian Rickettsia or Coxiella endosymbionts of engorged female ticks. This suggested that tetracycline and kanamycin might be toxic to developing oocytes, which would obscure any effect of removing symbiotic bacteria. In contrast, ciprofloxacin drastically reduced symbiont numbers without affecting egg viability, indicating that the presence of the bacteria was not necessary for the development and survival of the offspring (29, 30). Elimination of symbiotic bacteria is an important step in determining their function in the context of the symbiotic relationship.

The purpose of this study was to determine the growth dynamics of the symbiotic R. buchneri bacterium within its host as that host proceeds through its own life cycle. To assist future studies into the actual function of R. buchneri in I. scapularis, we present the results of our attempts to eliminate the bacteria from their host. In particular, the most successful method for eliminating R. buchneri, injection of mother ticks with ciprofloxacin, is described in detail.

RESULTS

Lifetime quantification of R. buchneri citrate synthase gene (gltA) copies.

The quantity of R. buchneri gltA copies varied substantially within each life stage through the life cycle of captive-reared and wild-collected I. scapularis (Fig. 1 and 2). Copy number generally increased throughout the tick life cycle, although the number of copies present between individual ticks was quite variable. Nearly all captive-reared ticks (cohort 1) (Fig. 1) had gltA quantities above the quantitative PCR (qPCR) assay threshold of detection. Numbers of cohort 1 ticks that had copy numbers of gltA above the threshold of detection were 23/23 (100%) for flat larvae, 93/96 (96.9%) for engorged larvae, 124/129 (96.1%) for flat nymphs, 37/43 (86.1%) for engorged nymphs, and 24/24 (100%) for flat adults. Among the F₁ offspring of this cohort of ticks, 47/53 (88.7%) eggs and 64/64 (100%) larvae had gltA copy numbers above the threshold of detection.

FIG 1.

Number of R. buchneri gltA copies per whole tick through the life cycle of captive-reared cohort 1 I. scapularis. Data are means and errors from R. buchneri-positive ticks only. The timing of important life events, such as feeding and molting, is indicated by a vertical box showing the duration of the event. The number (“n tested”) of individual ticks evaluated by qPCR at each time point is given beneath each time point, as is the number of ticks with gltA copies above the threshold of detection (“n positive”). The tick life stages corresponding to each section of data points are indicated at the top of the chart, with unfed larvae unlabeled at the far left (week 0). The “L,N,F” line shows copy numbers for larvae, nymphs, and female ticks. Copy numbers from males are shown by the diverging “M” line. “Eggs” indicates eggs of the subsequent, F₁ generation of ticks, and “Larvae” refers to the larvae that eclosed from those eggs. Error bars show standard errors. For F₁, standard error is not mappable because it descends below log 10¹.

FIG 2.

Number of R. buchneri gltA copies per whole tick through the life cycle of the wild-collected tick cohort. Data are means and errors from R. buchneri-positive ticks only. The line labeled “Rb+/total” indicates numbers of R. buchneri gltA-positive ticks and total numbers of ticks tested. “Days since detachment or last eclosion” shows the time since feeding detachment or last life cycle change, i.e., 54 days since hatching, 24 days since feeding detachment, 12 days since molting, etc. Absolute age is the age of each group of ticks since hatching. Dots indicate individual ticks, wide horizontal lines indicate means, and error bars show standard errors. Nymphs were fed in two groups, accounting for the difference between time since detachment and absolute age (days since emergence).

The offspring of wild-collected I. scapularis followed a pattern of infection similar to that in the captive-reared ticks, with a high proportion of R. buchneri gltA detection across all life stages (Fig. 2). Of unfed larvae, 20/20 (100%) had copy numbers of gltA above the threshold of detection. This proportion was 6/6 (100%) in engorged larvae, 6/6 (100%) in younger flat nymphs, 9/10 (90%) in older flat nymphs, 8/10 (80%) in the youngest engorged nymphs,10/10 (100%) in medium-aged engorged nymphs, 9/10 (90%) in the oldest engorged nymphs, 6/10 (60%) in males, and 10/10 (100%) in unfed females. Males and females were housed together, so adult ticks were likely to have mated.

Sequence identification of R. buchneri.

Two cohort 1 ticks, both cohort 2 ticks, and all wild-type progeny ticks produced gene sequences with a 100% match to R. buchneri strain ISO7 contig 169 (whole-genome shotgun sequence; NCBI accession number JFKF01000169.1). One tick from cohort 1 had a sequence that differed by 2 bp and had 100% sequence identity with I. scapularis endosymbiont isolates MOBP07 and TX136 (NCBI accession numbers EU544296.1 and EF689737.1, respectively).

Quantification of R. buchneri gltA following ciprofloxacin exposure.

For each of the four treatment groups fed using an artificial membrane, 2 or 3 females successfully engorged and detached within the allotted feeding interval. All of these successfully laid fertile egg masses. This proportion of successful engorgement and reproduction is in line with that observed for adult I. scapularis fed on rabbits under laboratory conditions (11/36; 30.6%) (31). The mean number of R. buchneri gltA copies per egg, larva, or nymph is shown in Fig. 3. Rickettsia buchneri was detected in the no-treatment control samples and in 100% of the eggs of one of the 10-μg/ml-ciprofloxacin-fed females. Eggs from the other 10-μg/ml-ciprofloxacin-fed female were all negative for R. buchneri gltA. Larvae which developed from the previously R. buchneri-positive eggs of the 10-μg/ml-fed female were all negative for R. buchneri gltA. Of the progeny of mothers treated with ciprofloxacin (40 μg/ml fed and 30 μg injected), all other F₁ life stages were entirely negative for R. buchneri gltA, regardless of dosages and delivery methods.

FIG 3.

Number of R. buchneri gltA copies per whole tick following exposure of the mother tick to various doses and delivery methods of ciprofloxacin. All but one of the 10-μg/ml-ciprofloxacin-fed F₀ mother ticks yielded progeny with gltA copy numbers below the threshold of detection. No larvae or nymphs contained R. buchneri gltA.

We further examined larvae (F₁ progeny) derived from mother ticks injected with 10 mg/ml ciprofloxacin and the nymphs that developed after feeding on mice or hamsters. A set of 20 unfed larvae from this cohort were tested by conventional PCR for the presence of R. buchneri. Following engorgement, the resulting nymphs were tested for the presence of R. buchneri 16S rRNA and for tick mitochondrial 16S rRNA as a positive control (32) by conventional PCR. All larvae (20/20) were positive for tick mitochondrial 16S rRNA, and 3/20 (15%) were positive for R. buchneri 16S rRNA. All nymphs (78/78) were positive for tick mitochondrial 16S rRNA, and 16/78 (20.5%) were positive for R. buchneri 16S rRNA.

Comparison of R. buchneri gltA copies between sexes and cohorts.

The distribution of R. buchneri between sexes varied between cohorts 1 and 2 and the lab-reared progeny of wild-collected ticks (Fig. 4). Mean quantities of gltA and standard errors were calculated based on adult ticks assayed individually in cohort 1. gltA copies of cohort 2 were calculated as weighted averages from groups of pooled ticks. In cohort 1, the mean number of R. buchneri gltA copies was somewhat lower in adult males (mean, 551,615 copies; n = 12; weeks 19 to 23) than in adult females (mean, 780,028 copies; n = 12; weeks 19 to 23), though this difference was not statistically significant (t-test 2-tailed P value, >0.4; df = 22). Cohort 2 adults demonstrated a marked difference in the quantity of gltA copies between females (weighted average, 428,560 copies) and males (weighted average, 12,790 copies) with more than an order of magnitude more bacteria present in females than males. Statistical analysis was not performed on the cohort 2 samples, because the pools varied in the number of ticks included. Lab-reared wild tick adults also showed a large difference in both the proportion of R. buchneri gltA-positive ticks and the quantity of bacteria present, with females (mean, 10,048,600 copies) carrying more than an order of magnitude more bacteria than males (mean for positive ticks, 412,881 copies), a significant difference (P < 0.00001).

FIG 4.

Rickettsia buchneri gltA copies varied between sexes in cohort 2 and wild-collected Camp Ripley F₁ ticks but not in cohort 1. “n” indicates the total number of individual or pooled ticks tested. The cohort 1 mean was based on gltA quantification of numerous individual ticks, but the cohort 2 mean is a weighted mean based on pooled adult ticks. Thus, no error bars are included for cohort 2. The asterisk indicates a significant difference (P < 0.00001).

DISCUSSION

In adult female I. scapularis, R. buchneri is found in the ovarian tissue, primarily in oocytes. Ovaries of female ticks are systemically infected with the symbionts, raising the possibility that, in immature life stages that will develop into females, the bacteria may reside primarily in the ovarian primordia (33, 34). If this is the case, the dearth of host tissue available in the immature ticks may explain the order-of-magnitude jump in copy number that occurs between larval and nymphal stages and between nymphal and adult stages (Fig. 1).

In captive-reared cohort 1, the number of R. buchneri gltA copies increased substantially (Fig. 1) following engorgement, as immature ticks developed into the next life stage, although this occurred in a 2-step process. Initially, the number of gltA copies in newly emerged nymphs declined immediately after molting (weeks 5 to 7) but rapidly increased soon after (weeks 7 to 8). Following nymphal feeding (weeks 12 to 13), there was a small decline in the number of gltA copies but, as the molting process progressed in the engorged nymphs (weeks 13 to 17), the number of copies increased by an order of magnitude. Once eclosion of female ticks was complete (weeks 17 to 19), copy numbers continued to increase until they plateaued at a mean of about 7.5 × 10⁵ during weeks 21 to 23.

The distribution of R. buchneri among female versus male ticks differed between the two captive-reared cohorts. In cohort 1, all ticks harbored R. buchneri and there was no significant difference in the quantity present. In cohort 2, the pattern differed substantially with all female ticks testing positive for R. buchneri gltA while many males tested negative, and overall quantities in the males were substantially lower than in females. The literature indicates geographic variation between wild populations of I. scapularis in regard to their rickettsial endosymbiont loads (5, 35). In populations in which R. buchneri is overall abundant, a declining proportion of ticks harbor the symbionts as the life cycle progresses, but adult females retain high levels of infection (36). The offspring of ticks collected at Camp Ripley, MN, demonstrated a substantial difference between males and females in both the quantity of bacteria and the proportion of infected ticks (Fig. 4). In our study, males of both captive-reared cohorts, albeit from different laboratory generations, were examined at the same age, so the observed variation cannot be explained by tick age.

Ciprofloxacin was included among the antibiotics to be tested due to its demonstrated effectiveness in eliminating the rickettsial symbiont of Ixodes pacificus (REIP) from I. pacificus (37). Ciprofloxacin exposure of I. scapularis mothers, either through feeding or by direct injection, successfully eliminated R. buchneri from most offspring. Practically, injecting the antibiotic into engorged females proved to be an effective and simpler method than feeding antibiotics to the ticks in an artificial feeder. Female fecundity did not appear to be severely affected by ciprofloxacin injection. Eggs from low-dose-ciprofloxacin-fed mothers maintained a mean number of gltA copies closer to that of the control group, but R. buchneri was absent by the time these eggs had hatched into larvae. It may be that bacteria exposed to lower dosages of the antibiotic were killed more slowly and that residual DNA from killed bacteria, which later degraded, was responsible for those detections. Altogether, these findings are consistent with those of Kurlovs et al. (37), who showed that eggs from I. pacificus females cured of REIP hatched normally.

Eliminating native R. buchneri symbionts from I. scapularis will enable future research to determine the contribution of these bacteria to the physiology and immunology of this important pathogen vector. Future research with aposymbiotic black-legged ticks may elucidate the relationship of R. buchneri with other constituents of the I. scapularis microbiome, including other symbionts or tick-borne pathogens. The I. scapularis microbiome is complex and highly variable between populations and life stages (38). Most larvae and nymphs harbor R. buchneri, and it is found in nearly all females; when present, it is highly abundant and represents the majority of bacteria present in the tick (36). Other endosymbionts have not been described, though some taxa are found at high prevalence in some populations of I. scapularis (Enterobacteriaceae and Pseudomonas in particular [4, 10, 39]). Transformation and reintroduction of R. buchneri into the ovaries of females may provide a mechanism for introducing genes of interest into ticks to facilitate innovative strategies for tick population control. Gaining the ability to manipulate the relationship between I. scapularis and its most consistently occurring symbiont will potentially aid in the development of new mechanisms or technologies for controlling tick-borne pathogens.

MATERIALS AND METHODS

Tick maintenance.

Captive-reared Ixodes scapularis ticks were obtained from the Oklahoma State University Tick Rearing Facility. Wild-collected I. scapularis ticks were the offspring of engorged females collected in October of 2018 from hunter-killed white-tailed deer at Camp Ripley, MN. Ticks were maintained in sealed humidors containing a saturated solution of K₂SO₄ to keep humidity levels at >95%, in a room kept at 22°C with a 16-h:8-h light-dark cycle. Ticks from two captive-reared cohorts were used: the first to quantify R. buchneri gltA in untreated ticks and the second to quantify it in ciprofloxacin-fed or injected ticks. For experiments in which ticks were fed or injected with ciprofloxacin, ticks were maintained on naive Syrian hamsters or C57BL/6 mice. All rodents were maintained and utilized in accordance with the regulations of the University of Minnesota and its institutional animal care and use committee.

Sampling for lifetime quantification of R. buchneri gltA.

Two cohorts of captive-reared ticks were fed upon hamsters and evaluated regularly by qPCR for R. buchneri gltA copies to determine the typical growth dynamics of R. buchneri throughout the life cycle of its host tick. Ticks were sampled weekly until molting to adults, after which adult females were sampled every 2 weeks. Rickettsia buchneri gltA copy numbers in engorged females were not quantified. qPCR testing was performed upon individual ticks. F₁ eggs, larvae, and engorged larvae were obtained by feeding adult ticks using the artificial membrane system without antibiotics to avoid potentially impacting R. buchneri numbers. F₁ ticks were all qPCR tested individually.

Rickettsia buchneri gltA copies were quantified in eggs and in the resulting larvae from the next (F₁) generation of ticks. Eggs were collected 7 days after laying began and were taken from different locations in the egg mass to sample eggs at different stages of oogenesis. Larvae were collected 1 week after the initiation of eclosion. Five individual eggs were tested from each of 10 F₀ mother ticks, and 3 eggs were tested from one additional F₀ tick. Five F₁ larvae from each of 12 F₀ ticks and 4 larvae from one additional F₀ tick were tested.

Rickettsia buchneri gltA copies in wild-collected I. scapularis were quantified at several time points in the life cycle for comparison against captive-reared populations.

Ciprofloxacin treatment.

Female ticks were treated with ciprofloxacin by injection or by artificial membrane feeding on cow blood spiked with the antibiotic. Membrane feeding was performed as previously described (40). Briefly, female ticks were fed to engorgement on blood through a 100-μm-thick silicone-and-rayon membrane. Defibrinated bovine blood (4.5 ml) supplemented with 2% glucose, ciprofloxacin, and 0.1% 3 mM ATP was replaced every 12 h. Four groups of adults were used: a group that was membrane fed low-dose ciprofloxacin (45 μg ciprofloxacin in 4.5 ml blood; 10 μg/ml), a group that was membrane fed high-dose ciprofloxacin (180 μg ciprofloxacin in 4.5 ml blood; 40 μg/ml), a no-ciprofloxacin negative control group, and a group that was not fed ciprofloxacin but was injected with 30 μg following engorgement and detachment. Each group comprised 9 female I. scapularis ticks, all from a single cohort. Eight males were also included in each feeding chamber to ensure mating and complete engorgement.

Previous studies have shown that 10 μg/ml ciprofloxacin successfully eliminates Mycoplasma spp. from vertebrate cell culture (41), justifying the low-dose feeding treatment. Because we were unsure how much spiked blood would be ingested, the high-dose treatment of 40 μg/ml was also used. Ticks were allowed 4 days to attach to the membrane prior to ciprofloxacin being added to the feeding reservoirs. Ciprofloxacin was then added to the reservoirs for five consecutive days. Following ciprofloxacin exposure, the ticks were allowed another week to engorge. Any ticks that did not fully engorge and detach by this time were discarded. Upon engorgement, ticks were removed from the feeding chamber, washed, and stored in humidors for egg laying.

Ticks intended for ciprofloxacin injection were membrane-fed on ciprofloxacin-free blood. Following engorgement, these ticks were intrahemocoelically injected using pulled-glass capillary needles glued over 28-gauge needles. All injections comprised 3 μl of 10 mg/ml ciprofloxacin. Doses were measured by micropipette and loaded individually into the tip of the capillary needle. The needle was then placed on an insulin syringe mounted on a Narishige MM-3 manipulator/micromanipulator (Narishige Scientific Instrument Lab, Tokyo, Japan). Injections were made shallowly on the ventral side of the tick in the less-sclerotized integument between the genital aperture and coxae 2 and 3. Double-sided tape or a backstop was used to keep ticks positioned properly. After injection, neutral pressure was maintained on the injection dosage to avoid backfilling of the capillary needle and the needle was kept in place for 20 s to ensure diffusion into the hemolymph prior to withdrawal.

F₁ eggs, larvae, and nymphs were tested by qPCR for R. buchneri. Five eggs were randomly selected from different parts of each egg mass for individual extraction and testing. After the eggs hatched, 5 larvae were removed from each mass for testing. Larvae from 2 egg masses from each treatment group were separately fed on naive hamsters to obtain nymphs. After the engorged larvae molted to nymphs, five nymphs were randomly selected for testing from each group (10 nymphs tested per treatment). To verify the successful extraction of DNA in samples negative for R. buchneri citrate synthase, a portion of tick mitochondrial 16S rRNA was amplified using the 16S+1 and 16S−1 primers (32). Three larval samples from each treatment group were tested, all of which were positive for 16S rRNA.

In a second set of experiments, 6 membrane-engorged female ticks were injected with 3 μl of 10 mg/ml ciprofloxacin. The F₁ larvae were fed on hamsters or mice and allowed to molt into nymphs. Twenty larvae and 78 nymphs were then individually tested for the presence of R. buchneri 16S rRNA genes using conventional PCR.

Quantification of R. buchneri gltA copies by qPCR.

DNA was extracted from each life stage of I. scapularis, using the DNeasy blood and tissue kit (Qiagen, Valencia, CA) as previously described (42), with the exception that the cuticle material was not retained. The DNA for each life stage was eluted into the following volumes of DNeasy AE buffer (10 mM Tris-Cl, 0.5 mM EDTA [pH 9.0]): eggs, 30 μl; larvae, 45 μl; engorged larvae, 50 μl; nymphs and engorged nymphs, 100 μl; adult females and males, 200 μl. gltA copy numbers were extrapolated to determine the total number of genomic copies of R. buchneri per tick as previously described (43), except that the annealing temperature was set at 58°C, and 5 μl of undiluted tick DNA was used for each reaction. The specific primers CS-F and CS-R (44) were used to amplify the single-copy rickettsial gltA gene (44, 45). Data retrieval and analysis were done with MxPro software v. 4 (Agilent, Santa Clara, CA). The relative number of copies of R. buchneri per tick was determined by referencing tick DNA gltA cycle threshold (C_T) values to the standard curve as previously described (43). The threshold of detection was equal to the number of plasmid copies present in the least concentrated dilution of the standard curve (2.07 copies).

PCR detection of R. buchneri 16S rRNA genes in immature I. scapularis.

Rickettsia buchneri presence in ciprofloxacin-exposed I. scapularis larvae and nymphs was determined by PCR. Unfed ticks were immersed in 70% ethanol for 12 h, rinsed in 1× Tris-acetate-EDTA buffer, cut longitudinally, and incubated overnight at 56°C in 180 μl of tissue lysis buffer containing 0.4 mg of proteinase K (Qiagen). DNA was extracted using the DNeasy blood and tissue kit according to the manufacturer’s protocol. Infection with R. buchneri was determined by PCR amplification with the primers REIS rt F2 (5′-GTCTAGGTTCACGCTTTCGC-3′) and REIS rt R2 (5′-TTGCAGTTCACAAAAGCAGG-3′), targeting the 16S rRNA gene of R. buchneri. Each 50-μl reaction was performed with Go-Taq DNA polymerase (Promega, Madison, WI) as per the manufacturer’s recommendations using 100 ng of template and cycling parameters as follows: 1 cycle of 95°C for 1 min; 35 cycles of 95°C for 30 s, 54°C for 30 s, and 72°C for 45 s; and a final extension of 1 cycle of 72°C for 5 min. PCR products were electrophoresed on 1% agarose gels and stained with GelGreen (Biotium, Fremont, WI).

Identification of Rickettsia species by ompA sequencing.

To confirm that ticks contained R. buchneri rather than another species of Rickettsia, DNA from a subset of unfed females from each group of ticks was amplified using primers for the ompA gene (46) and sequenced at the University of Minnesota Genomics Center (2 from cohort 1, 3 from cohort 2, and 10 wild-type progeny of Camp Ripley-collected ticks). A microbial nucleotide NCBI BLAST search was performed on the gene sequences.

Regulatory compliance.

Tick feeding upon animal hosts was performed in accordance with the U.S. National Institutes of Health guidelines (47) and the University of Minnesota Institutional Animal Care and Use Committee guidelines.

Data availability.

Sequence data for ompA were uploaded to NCBI GenBank. All but one tick had identical nucleic acid sequences (NCBI accession number MW221785). One tick varied by 2 bp from the other sequences (NCBI accession number MW232419).