Use of Quantitative PCR To Measure Density of Borrelia burgdorferi in the Midgut and Salivary Glands of Feeding Tick Vectors

Abstract

Quantitative real-time PCR was used to assay spirochetes in feeding ticks. Spirochetes in tick midguts increased sixfold, from 998 per tick before attachment to 5,884 at 48 h of attachment. Spirochetes in tick salivary glands increased >17-fold, from 1.2 per salivary gland pair before feeding to 20.8 at 72 h postattachment. The period of the most rapid increase in the number of spirochetes in the salivary glands occurred from 48 to 60 h postattachment; this time period coincides with the maximal increase in transmission risk during nymphal tick feeding.

Nymphal Ixodes scapularis ticks are the principal vectors of Lyme disease spirochetes (Borrelia burgdorferi sensu stricto) in North America. These spirochetes appear to be well adapted to their hosts (1) and tick vectors (12). Questing nymphal I. scapularis ticks contain spirochetes restricted mainly to the tick midgut. When an infected tick successfully finds a host and starts to take a blood meal, dramatic changes occur in the spirochete population: spirochetes multiply rapidly in the midgut of infected ticks, thus increasing the overall spirochete density (3). In parallel with increased density, a shift occurs in the dominant outer surface protein expression of the spirochetes, from OspA to OspC (4, 17). Other proteins, such as the vlsE protein, become more heterogeneous upon tick attachment (12). Spirochetes then migrate through the hemolymph of the feeding nymphal tick to the salivary glands, where they are subsequently transmitted to the skin of the vertebrate host (2, 12, 21).

A basic understanding of spirochete dynamics within feeding ticks is central to an appreciation of why nymphal I. scapularis ticks infected with B. burgdorferi sensu stricto removed during the first 2 days of attachment do not transmit infection to tick bite victims whereas those feeding for longer periods efficiently transmit infectious spirochetes (6, 10, 12). In addition, the efficacy of the newly licensed recombinant OspA human vaccine in use in the northeastern United States is apparently based on the ability of anti-OspA antibodies to enter the tick midgut and kill spirochetes before they can migrate to the salivary glands to be transmitted to the victims of tick bite (4, 5). This vaccine does not appear to act against the spirochetes within the vertebrate host (4, 20). Thus, studies on the dynamics of spirochete populations in feeding ticks will improve our understanding of how ticks transmit spirochetes and how to prevent such transmission. A better appreciation of the interactions between ticks, hosts, and spirochetes may also lead to a better understanding of the epidemiology and ecology of these important human pathogens (11).

Spirochete populations in ticks have been quantified principally through the use of microscopic tools. Rough estimates of the number of spirochetes viewed in tick smears through the use of standard epifluorescent, electron, or confocal microscopy have been reported (2, 12, 21). Recently, extremely sensitive and accurate quantitative fluorogenic-detection PCR assays for measuring B. burgdorferi have become available (9, 13, 18, 19). In the present study, we used these new assays to quantify spirochete populations in feeding ticks as these bacteria move from the midgut to the salivary glands of their vectors.

I. scapularis ticks were infected with the B-31 strain of B. burgdorferi sensu stricto as previously described (15). Briefly, larval ticks from a Borrelia-free colony originating from Bridgeport, Conn., were allowed to feed to repletion on mice previously infected via tick bite with B-31. Replete ticks were held at 21°C in saturated humidity and allowed to molt into nymphs. This method routinely produces batches of ticks with a >80% infection rate. A total of six flat nymphs were tested from each batch used in these experiments by homogenization and culture in Barbour-Stoenner-Kelly medium. Cultures were examined for spirochetes under dark-field microscopy. Only batches where six of six nymphs produced positive cultures (100% infected) were used in these experiments.

Flat nymphal ticks were allowed to feed ad libitum on outbred 4-week-old female mice from the Centers for Disease Control and Prevention pathogen-free mouse colony. At selected intervals, nymphs were removed from mice and dissected at 40× magnification under a Zeiss dissecting microscope. Midguts and salivary glands were removed and placed in a drop of phosphate-buffered saline on a microscope slide, washed in a second drop of phosphate-buffered saline and transferred to a tube containing 150 μl of commercial lysis buffer (ATL buffer; Qiagen, Inc., Valencia, Calif.) for DNA isolation. Midguts and salivary gland pairs from 15 nymphs at each time point were placed in the buffer solution. Samples from groups of ticks were obtained before attachment and at 24, 48, 60, and 72 h after attachment. Ticks completed feeding on the host after 72 h, and groups of replete ticks were examined at 4 and 8 days after attachment.

DNA from tick midguts and salivary glands was extracted using a commercial DNA isolation kit (Qiagen, Inc.) with minor modifications. After the dissected midguts and salivary glands were placed in 150 μl of commercial lysis buffer, 400 μg of protease K (Gibco Life Technologies, Gaithersburg, Md.) was added and the mixture was allowed to incubate at 56°C for 18 h. The DNA isolation procedure and recovery from spin columns then proceeded according to the manufacturer's instructions. DNA from each individual experiment (repeated three times) was extracted at the same time. The quantitative PCR (q-PCR) assay to quantify DNA from tick tissues was then run as previously described (19), using forward and reverse primers to amplify conserved sequences within the flagellin gene. Amplifications were performed using the model 7700 sequence detector system in optical tubes (Perkin-Elmer, Foster City, Calif.). The standard curve, generated each time the assay was performed, was comprised of data for DNA extracted from dilutions of B. burgdorferi strain B31 ranging from 10⁶ to 10⁰ spirochetes as previously described (19). To be certain tick tissues did not interfere with the generation of standard curves, cultures of B. burgdorferi were placed directly in buffer or in buffer containing the salivary glands or midguts from 15 uninfected ticks. The curves generated from data for cultured spirochetes with or without tick tissues present were virtually (≥99.8%) identical (data not shown). The detection limit for B. burgdorferi DNA was between 10 and 1 spirochetes, as previously reported (19). Determinations for all samples were done in triplicate wells, and all data were analyzed using the model 7700 sequence detection system software, version 1.63 (Perkin-Elmer). The numbers of spirochetes are reported as means plus standard deviations. The mean value of the results for three wells was considered the mean for each trial; the mean plus standard deviation was calculated from a total of three trials for each time point and each organ.

Spirochete populations in tick midguts increased rapidly (sixfold) from a total of 998 per tick before attachment to 5,884 at 48 h of attachment (Fig. 1). Spirochete numbers in the midgut leveled off at 48 to 72 h, increasing by less than 1,000 to a peak of 6,876 during this third day of feeding. Spirochete numbers in the tick midgut dropped precipitously upon repletion, to 2,076 at 4 days postattachment. By 8 days postattachment, spirochete numbers in the midgut had rebounded to 12,961.

FIG. 1.

Density of spirochetes in tick midguts determined by q-PCR. Spirochete numbers were determined for pools containing 15 midguts dissected from 15 individual ticks. The total in a pool was divided by 15. Three trials were conducted for each time point; bars represent the averages for the three trials, and lines represent the standard deviations; an asterisk indicates a P of <0.05 as determined by Student's t test in comparison with the values observed at 72 h.

Spirochete loads in tick salivary glands followed a slightly different pattern. Spirochete numbers increased >17-fold, from 1.2 per salivary gland pair before feeding commenced to 20.8 at 72 h postattachment (Fig. 2). The period of the most rapid spirochete increase in the salivary glands was from 48 to 60 h postattachment. After repletion, the number of spirochetes began to decrease steadily, falling to 18.3 at 4 days and 6.9 at 8 days.

FIG. 2.

Density of spirochetes in tick salivary glands determined by q-PCR. Spirochete numbers were determined for pools containing 15 salivary gland pairs dissected from 15 individual ticks. The total in a pool was divided by 15. Three trials were conducted for each time point; bars represent the averages for the three trials, and lines represent the standard deviations; an asterisk indicates a P of <0.05 as determined by Student's t test in comparison with the values observed at 72 h.

Spirochete populations increase rapidly once an infected nymph starts to feed. During the first 2 days of tick feeding, spirochetes (B. burgdorferi sensu stricto) in the tick midgut shift their outer surface proteins, changing from a population that uniformly expresses OspA to include populations that express OspA and OspC, OspC alone, or neither outer surface protein (4, 12, 17). Curiously, OspA binding activity has been observed in the midgut tissue of I. scapularis ticks (14). Pal et al. (14) suggested the hypothesis that repression of OspA during tick feeding may facilitate detachment from the tick midgut and migration to the salivary glands. The pattern of spirochetal abundance in the midgut observed during the present study is consistent with this hypothesis, but formal proof of the role of midgut receptors is beyond the scope of this study and awaits future research. In addition, European genospecies of B. burgdorferi present a more complex pattern of Osp expression in their tick vectors (7, 8).

Although small numbers of spirochetes are found in tick salivary glands during the first 2 days of tick feeding, a rapid increase in spirochete numbers in the salivary glands occurs from 48 to 60 h. Our observations using q-PCR mirrored a similar increase observed in a previous study which used fluorescent microscopy (12). This rapid increase in spirochetes in the salivary glands during the third day of tick feeding occurs at the same time (60 h) that homogenates of salivary glands become infectious to mice (16) and coincides with a dramatic increase in the risk of transmission of infectious spirochetes to hosts (6, 10, 12). After tick repletion, the salivary glands begin to deteriorate and spirochete numbers in this organ decrease steadily.

A frequently asked question is how many spirochetes an infected nymphal tick inoculates when feeding on tick bite victims. Our results do not allow a precise calculation of the total amount of spirochetes inoculated by a feeding nymph since we have only ex vivo estimates of the number of spirochetes present in the salivary glands at discrete points in time. To make an accurate calculation, one would have to know the turnover of spirochetes in the glands and the proportion actually inoculated.

The q-PCR procedure outlined here offers some real advantages relative to conventional PCR methodology. This real-time procedure monitors DNA amplification using a gene-specific probe, which allows for consistent, rapid, and reliable quantification of a specific gene product without postamplification handling of the DNA. Because the assay includes binding of a B. burgdorferi-specific, FAM (6-carboxyfluorescein)-labeled probe before amplification and subsequent fluorescence can be measured, there is minimal concern for the generation of false-positive results. This assay was sensitive (detecting between 1 and 10 spirochetes), and our data compare favorably with those resulting from more conventional assays, such as culture and fluorescence confocal microscopy (12, 16). Future studies will include the application of this assay to monitor and quantify a number of physiologically relevant B. burgdorferi-specific gene targets during the feeding of I. scapularis on mammalian hosts.