Acari: Ixodidae - Oxford Journals

MODELING/GIS, RISK ASSESSMENT, ECONOMIC IMPACT

Human Behaviors Elevating Exposure to Ixodes pacificus (Acari: Ixodidae) Nymphs and Their Associated Bacterial Zoonotic Agents in a Hardwood Forest ROBERT S. LANE, DENISE B. STEINLEIN,

AND

JEOMHEE MUN

Division of Insect Biology, Department of Environmental Science, Policy, and Management, University of California, Berkeley, CA 94720

J. Med. Entomol. 41(2): 239Ð248 (2004)

ABSTRACT Epidemiological evidence suggests that the nymph of the western black-legged tick, Ixodes pacificus Cooley and Kohls, is the primary vector of the Lyme disease spirochete, Borrelia burgdorferi Johnson, Schmid, Hyde, Steigerwalt, and Brenner, to humans in northwestern California. In spring 2002, six different human behaviors were evaluated as potential risk factors for acquiring I. pacificus nymphs in a deciduous woodland in Mendocino County, California. Also, the prevalence of B. burgdorferi sensu lato (s.l.) and the causative agents of human granulocytic (Anaplasma phagocytophilum [Foggie] Dumler, Barbet, Bekker, Dasch, Palmer, Ray, Rikihisa, and Rurangirwa) and monocytic ehrlichioses (Ehrlichia chaffeensis Anderson, Dawson, Jones, and Wilson) was determined in nymphs that had been collected from subjects or by dragging leaf litter. Activities involving a considerable degree of contact with wood resulted in greater acquisition of nymphs than those involving exposure solely to leaf litter. Time-adjusted tick-acquisition rates demonstrated that sitting on logs was the riskiest behavior, followed, in descending rank, by gathering wood, sitting against trees, walking, stirring and sitting on leaf litter, and just sitting on leaf litter. The number of ticks acquired appeared to be unrelated to the type of footwear worn (hiking boots, hiking sandals, or running shoes). Overall, 3.4% (n ⫽ 234) of the nymphs were infected with A. phagocytophilum, 3.9% (n ⫽ 181) with B. burgdorferi s.l., and none (n ⫽ 234) with E. chaffeensis. Of 13 nymphs infected with either A. phagocytophilum or B. burgdorferi s.l., 2 (15.4%) were coinfected with both bacteria, as were 1.3% of 158 nymphs obtained from leaf litter, the Þrst report of coinfection in this life stage of I. pacificus. Four unattached, infected nymphs were removed from subjects, including two acquired while sitting on logs that contained A. phagocytophilum, another with the same bacterium obtained while walking, and one acquired while gathering wood that was infected with B. burgdorferi s.l. Despite the use of extreme personal preventive measures by both subjects, two attached, uninfected nymphs were removed from one of them ⱖ1Ð2 d postexposure. The public health implications of these Þndings are discussed. KEY WORDS Ixodes pacificus, risk factors, Anaplasma phagocytophilum, Borrelia burgdorferi, Ehrlichia chaffeensis IN CALIFORNIA, THE WESTERN BLACK-LEGGED tick, Ixodes pacificus Cooley and Kohls, is the primary bridging vector of the Lyme disease spirochete, Borrelia burgdorferi Johnson, Schmid, Hyde, Steigerwalt, and Brenner, to humans. The two life stages of I. pacificus that have been implicated as vectors are the nymphs and the adult females, which readily attach to people and collectively account for ⬇60% of all recognized human attachments by ixodid ticks in the state (Clover and Lane 1995). In contrast, host-seeking larvae and adult males seldom attach to people and therefore are unimportant epidemiologically. The individual risk for contracting Lyme disease depends on the local abundance of spirochete-infected vector ticks and the degree of human-tick contact (CDC 1999). In one area highly endemic for Lyme disease in Mendocino County, California, most incident cases of the disease clustered in spring and sum-

mer coincident with the host-seeking period of I. pacificus nymphs and the spirochetal infection prevalence in nymphs (13.6%) exceeded that in adult ticks (4%) by several fold (Clover and Lane 1995). The mean densities of spirochete-infected nymphs collected subsequently from some leaf-litter areas in the same region approached densities reported for infected Ixodes scapularis Say nymphs from highly endemic areas in the northeastern United States (Ta¨lleklint-Eisen and Lane 1999). In the northeastern United States, most cases of Lyme disease are contracted from June to August (Sonenshine 1993, Falco et al. 1999, CDC 2000) when questing I. scapularis nymphs reach their annual peak densities (e.g., Sonenshine 1993, Stafford and Magnarelli 1993, Falco et al. 1999). Furthermore, the density of B. burgdorferi-infected I. scapularis nymphs has been correlated with the number of human cases in

0022-2585/04/0239Ð0248$04.00/0 䉷 2004 Entomological Society of America

240

JOURNAL OF MEDICAL ENTOMOLOGY

that region (e.g., Mather et al. 1996, Nicholson and Mather 1996, Stafford et al. 1998). In Maryland, activities involving close contact with fallen logs and leaf litter in deciduous woodlands were found to elevate the likelihood of exposure to I. scapularis nymphs (Carroll and Kramer 2001). In California, nothing deÞnitive has been reported about human behaviors that speciÞcally increase the risk of encountering the nymphal stage of I. pacificus in deciduous hardwood forests. However, cutting wood and the use of wide maintained trails for more than 5 h per week were identiÞed as risk factors for contracting Lyme disease (Lane et al. 1992, Ley et al. 1995), and seroreactivity to antigens from one or more of four tick-borne disease agents was signiÞcantly associated with younger age (⬍16 yr), longer residence (11Ð20 yr), and having had a physicianÕs diagnosis of Lyme disease (Fritz et al. 1997). In the latter study, separate analyses performed for children revealed that seroreactive individuals were signiÞcantly older (median age, 11 versus 7 yr) and were more likely to have spent time in wooded areas around the home and on vacation than seronegative individuals. It cannot be overemphasized that without knowing how people are exposed to arthropodal vectors, it is difÞcult to develop and implement effective preventive strategies and educational programs. To Þll in this important gap in our knowledge, the current study was undertaken to determine how people recreating or working in a deciduous hardwood forest are apt to be exposed to I. pacificus nymphs. SpeciÞcally, we evaluated six human behaviors as potential risk factors for acquiring nymphs, and we also determined the prevalence of various bacterial zoonotic agents in nymphs that were collected from humans versus leaf litter. Materials and Methods The Study Area. This investigation was conducted at the University of California Hopland Research and Extension Center (HREC) in southeastern Mendocino County, California. The HREC is a 2,168-ha multipurpose, agricultural sciences research facility located on the western slopes of the Mayacmas Mountains in the Russian-River Valley. Cool, moist winters and hot, dry summers characterize the climate. Rolling hills interspersed with ravines typify the topography, with grassland, woodland-grass, dense woodland, and chaparral composing the principal vegetational types. The behavioral risk-assessment trials were conducted in a small portion of a hardwood grove comprised predominantly of California black oak, Quercus kelloggii Newberry, with scattered elements of PaciÞc madrone, Arbutus menziesii Pursh, at an elevation of ⬇665 m. The entire grove encompasses an area of ⬇6.3 ha and is surrounded by patches of chaparral to the east, south, and west, and by grassland to the north. The forest ßoor is carpeted densely with leaf litter and contains scattered branches, limbs, logs, and low-lying herbaceous vegetation. The sampling area (Fig. 1) was centered about a 27 ⫻ 50-m dragging/walking grid (1,350 m2) estab-

Vol. 41, no. 2

lished amid, and heavily shaded by, black oaks, except for three madrones situated at its northern edge. The grid consisted of 10 parallel 50 ⫻ 1-m linear transects spaced 2 m apart and oriented northwest to southeast. Each 50-m transect was divided in half, which yielded a total of 20 linear transects 25 m in length. Dragging and walking samples were taken within the conÞnes of the grid, whereas the other Þve behavioral samples were performed in leaf-litter areas around its immediate periphery and up to a distance of ⬇30 Ð35 m away on slight to moderate slopes. To determine the depth of leaf litter within the grid, Þve measurements were made at 5-m intervals within all 20 transects. Vertebrates or their sign observed in the area included Columbian black-tailed deer (Odocoileus hemionus columbianus [Richardson]), western gray squirrel (Sciurus griseus Ord), dusky-footed wood rat (Neotoma fuscipes Baird), mole (Scapanus sp.), acorn woodpecker (Melanerpes formicivorus [Swainson]), BewickÕs wren (Thryomanes bewickii [Audubon]), plain titmouse (Parus inornatus Gambel), spotted towhee (Pipilo maculatus Swainson), warbling vireo (Vireo gilvus [Vieillot]), and western fence lizard (Sceloporus occidentalis Baird and Girard). Micrometeorologic Factors. On all dates, temperature and relative humidity were recorded with a thermo-hygrometer (model HI 8564; Hanna Instruments, Woonsocket, RI) at the beginning and end of each individual sampling period ⬇2.5 cm above ground in the nymphal biotope. Experimental Design. The same two subjects performed the Þeld trials on back-to-back dates each week between 28 May and 13 June 2002 for a total of six sampling occasions. These dates were selected because nymphal host-seeking activities in leaf-litter areas peak around mid- to late spring in the general study area (Eisen et al. 2002). Subjects wore a pair of white socks and paintersÕ gloves, a long-sleeved white sweatshirt, and a pair of white paintersÕ overalls made of 100% cotton (Fig. 1). The bottoms of the pant legs were tucked beneath the socks, and the tops of the socks and the bottom of each sleeve were sealed with duct tape. Three types of footwear were worn. One subject always wore a pair of high-top hiking boots (Clarion Impact, Vasque, Red Wing, MN), whereas the other subject alternated between two types, a low-cut running shoe (Hurricane Ridge GTX, Montrail, Seattle, WA) and a sport sandal (Teva Sport Sandals, Flagstaff, AZ). To estimate the relative density of nymphs, drag sampling was conducted concomitantly with walker sampling during the Þrst day of the 2-d weekly sampling occasions. On those dates, sampling was initiated between 1107 and 1137 h. One subject performed drag sampling during wk 1 and 3, the other during wk 2. The subject who was responsible for dragging doubled as a second walker because the tick drag was pulled behind the subject. This permitted a comparison of the efÞcacy of different footwear for acquiring ticks while subjects traversed leaf litter during both dragging and walking operations. The drag was attached to a wooden handle and consisted of a 1.0 ⫻ 1.25-m white

March 2004

LANE ET AL.: EXPOSURE TO Ixodes pacificus NYMPHS

241

Fig. 1. The black-oak grove in which six human behaviors were evaluated as potential risk factors for acquiring I. pacificus nymphs. Subject wearing standardized garb is pointing to the drag-sampling grid.

ßannel cloth with a chain hemmed into its trailing edge. The dragger and the designated walker slowly traversed each of the 25-m transect lines while spaced

1.5 m apart. Upon completion of each replicate, which took ⬇25 s, both subjects inspected their shoes and socks for presence of ticks. To minimize overlooking

242


the minute nymphs, an adhesive roller (Evercare; Helmac Products, Flint, MI) also was used to sample the footwear. Next, the dragger inspected the tick drag and recorded the number of ticks discovered by life stage. On the Þrst sampling occasion only (28 May), all ticks collected by dragging were retained for speciÞc identiÞcation and testing. During the second and third drag-sampling occasions, all ticks detected were Þeld identiÞed to life stage and promptly released within the same transect from which they had been collected to avoid depleting the population further. We consider that it is highly unlikely that nymphs released on the second and third sampling occasions belonged to a species other than I. pacificus for two reasons: all 161 nymphs collected by dragging on the Þrst occasion were determined to be I. pacificus, as were all 86 nymphs removed from subjects throughout the study. Paired walking samples (n ⫽ 20 per subject) alone were taken on each of the three dates when dragging was not carried out, i.e., during the second day of the weekly back-to-back sampling occasions. On those occasions, the diurnal sequence of behaviors was reversed from that performed the previous day so that walker samples were taken last beginning between 1454 and 1525 h. Thus, subjects traversed a combined distance of 6 km in leaf litter while conducting drag/ walker sampling on all dates. Four of the Þve remaining behaviors, the exception being gathering wood (see below), were replicated 10 times per day and timed for 5 min. Sitting versus stirring and sitting on leaf litter, and sitting on logs versus sitting against trees, were performed concurrently with both subjects in close proximity. Subjects carried out these behaviors alternately on different dates, but within each sampling occasion only one subject served as the sitter, the other as the sitter/ leaf-litter stirrer, and so forth. Upon completion of each replicate, subjects stood up and carefully inspected their own clothing Þrst and then that of the other person. The bodily location of each tick found was recorded, and all ticks were saved in 70% ethanol for later identiÞcation and testing. The subject who sat in leaf litter did so in one spot for the entire 5-min period, whereas the leaf-litter stirrer/sitter spent 2.5 min stirring ⬇2 m2 of litter with gloved hands and then 2.5 min sitting. The black-oak trunks selected for sitting against were surrounded by leaf litter, and their bark was colonized basally by moss. Similarly, the logs chosen for sitting on represented black oaks, were surrounded by leaf litter, and were colonized by variable amounts of moss. Subjects always had some contact with moss while sitting against trees, whereas subjects sitting on logs selectively chose either bare or moss-covered areas as seats. Aspect was considered when choosing trees to sit against so that all four cardinal points would be represented. Either one, or in some cases two aspects (e.g., north versus south) of the same tree were evaluated for presence of host-seeking ticks. The portion of subjects that was in direct contact with tree trunks, i.e., the lower back to the top of the head,

Vol. 41, no. 2

reached a height of ⬇0.9 m above ground. The diameters and circumferences of such trees (n ⫽ 47 measurements, which include repeated measures on different dates) averaged 0.70 ⫾ 0.15 m (range, 0.42Ð1.00 m) and 2.25 ⫾ 0.54 m (range, 1.55Ð3.40 m), respectively. To expand the number of potentially risky behaviors evaluated, gathering wood was added to the experimental design during the third sampling occasion, and therefore it was performed on just four dates. Nevertheless, it was replicated 10 times per subject on those dates for a total of 80 replicates. Subjects searched around the perimeter of the dragging/walking grid for Þve fallen branches or small tree limbs that they then cradled in their arms while walking a 50-m transect line. An individual branch/limb was used only once and then discarded. The branches/limbs (n ⫽ 20) averaged 68.5 ⫾ 18.6 cm (range, 44 Ð100 cm) in length and 5.4 ⫾ 2.5 cm (range, 1.5Ð11.0 cm) in width. It took ⬇50 s to walk the transect, but the time spent Þnding suitable branches or limbs and carrying them to the start of the transect line took several minutes. We estimate that the entire process averaged roughly 5 min. DNA Extraction and Pathogen Detection in Ticks. Nearly all (98%) of the 161 I. pacificus nymphs collected from leaf litter during the Þrst dragging episode and 88% of the 86 nymphs obtained from the clothing or skin of subjects engaged in the various behaviors were tested for presence of DNA representative of three bacterial zoonotic agents with the polymerase chain reaction (PCR). These included the etiologic agents of human granulocytic ehrlichiosis (Anaplasma phagocytophilum [Foggie] Dumler, Barbet, Bekker, Dasch, Palmer, Ray, Rikihisa, and Rurangirwa), human monocytic ehrlichiosis (Ehrlichia chaffeensis Anderson, Dawson, Jones, and Wilson), as well as the Lyme disease spirochete (B. burgdorferi). Unfortunately, 53 of the 76 nymphal-tick samples from subjects became contaminated with B. burgdorferi extract, and therefore were excluded from this analysis. DNA was extracted from ticks using the Qia DNeasy extraction kit (Qiagen, Valencia, CA) with the following modiÞcations. Ticks were soaked in ddH2O for 10 min before triturating them in 180 ␮l of buffer ATL. The DNA was eluted in a Þnal volume of 100 ␮l of AE buffer. To verify that DNA had been extracted successfully, a PCR was run using the tick-speciÞc primers 16S ⫹ 2 and 16S ⫺ 1 (Black and Piesman 1994) and 1 ␮l of the Þnal elutions of the tick-DNA preparations. The primary PCR assays used 3 ␮l of DNA extracts as the template in a total reaction volume of 25 ␮l. All PCR mixtures contained 2.5 ␮l of 10⫻ PCR buffer (PE Applied Biosystems, Foster City, CA), 2.5 ␮l of 8 mM dNTP, 1.5 ␮l of 25 mM MgCl2, 1 ␮l of 10 ␮M primers, and 0.2 ␮l of 5 U/␮l Taq polymerase (PE Applied Biosystems). The cycling conditions involved an initial 4-min denaturation at 94⬚C, followed by 40 ampliÞcation cycles, each consisting of a 1-min denaturation at 94⬚C, a 1-min annealing at 55⬚C, and a 1-min extension at 72⬚C. These cycles were followed by a 10-min extension at 72⬚C. Annealing temperatures and

March 2004


times were varied for each primer set to determine optimal PCR conditions. To monitor for the occurrence of false-positive PCR results, negative controls, consisting of the same amount of PCR water instead of template DNA, were included for each run. Positive controls consisting of pathogen DNA were used to assess amplicon size and to demonstrate the relative efÞciency of amplicon production. B. burgdorferi sensu lato (s.l.) infection in ticks was assessed with two sets of primers in a nested PCR format that speciÞcally targets the 5S-23S rRNA spacer region of all borreliae belonging to this group. The outer primers were rrf-rrl 1 and rrf-rrl 2 (Postic et al. 1994), which produce a product of ⬇250 bp. The internal primers were 5⬘-GAGTAGGTTATTGCCAGGGTTTTATT-3⬘ and 5⬘-TATTTTTATCTTCCATCTCTATTTTGCC-3⬘, forward and reverse, respectively. The design of these internal primers, which produce 200-bp products, was based on the sequences obtained using the outer primers. To detect the presence of A. phagocytophilum, the 16S rRNA gene of the genogroup was targeted with two sets of primers in a nested PCR format (Massung et al. 1998). The outer and internal primers produce products having ⬇900 and 500 bp, respectively. Finally, to assess ticks for infection with E. chaffeensis, a primer set that targets the 16S rRNA gene was used (Anderson et al. 1992). For sequencing, all of the positive PCR products were puriÞed with a QIAquick PCR PuriÞcation Kit (Qiagen). Each 10 ␮l cycle-sequencing reaction contained 3.5 ␮l of PCR-grade water, 2 ␮l of Big Dye Terminator Ready Reaction Mix (PE Applied Biosystems), 0.5 ␮l of 3.2 ␮M primer, 2 ␮l of sequencing dilution buffer, and 2 ␮l of puriÞed PCR product. Amplicons were sequenced using their PCR primers. All cycle-sequencing products were puriÞed with Sephadex Centri-Sep columns (Princeton Separations, Adelphia, NJ) and run on an ABI 377 sequencer (PE Applied Biosystems). The sequences were aligned using the software Sequencher 3.1 (Gene Code, Ann Arbor, MI). Statistical Analyses. ␹2 analyses with Yates correction for continuity (␹2adj) were used to test all prevalence data, and the 5% level of probability was set for rejection of the null hypothesis. The prevalence rates for nymphs infected with A. phagocytophilum versus B. burgdorferi s.l. from either subjects or leaf litter were not compared statistically by McNemarÕs test because of the similar and very low rates of infection with each organism. To determine whether the spatial distribution of ticks within the dragging/walking grid was aggregated, the variance to mean ratio was computed for data gathered during the Þrst drag-sampling occasion. Results Environmental Measures. I. pacificus nymphs were collected from mid-morning to late afternoon (i.e., ⬇1000 Ð1900 h) over a wide range of temperatures (20 Ð32⬚C) and relative humidities (38 Ð71%). The mean depth (n ⫽ 5) of the leaf litter within each of the

243

20 transect lines was fairly uniform, as it ranged between 4 and 6 cm in 80% of the cases. Thus, no attempt was made to correlate the depth of leaf litter with the abundance of nymphs obtained by drag sampling. Drag Sampling. In total, 275 nymphs and 2 adults (1 male, 1 female) of I. pacificus plus 15 Ixodes sp. larvae and 3 adult males of Dermacentor occidentalis Marx were collected by dragging. Averages of 8.1 ⫾ 8.5 (range, 1Ð29), 4.0 ⫾ 2.1 (range, 1Ð8), and 1.8 ⫾ 1.3 (range, 0Ð4) I. pacificus nymphs were enumerated per 25-m transect on 28 May, 3 June, and 12 June, respectively, which correspond to densities of ⬇32, 16, and 7 ticks per 100 m2. On the Þrst sampling occasion, the spatial distribution of ticks within the grid was highly aggregated because the variance to mean ratio was 8.92. Acquisition of Nymphs by Behavior. Sitting against trees, gathering wood, and particularly sitting on logs were the riskiest behaviors, with 17Ð30% of the encounters yielding one or more ticks on either clothing or skin (Table 1). None of the three pairwise comparisons for the wood-associated behaviors differed signiÞcantly with respect to the proportion of ticks acquired by subjects (␹2adj ⫽ 0.66 Ð2.28, P ⫽ 0.13Ð 0.42). However, activities involving contact with wood collectively resulted in signiÞcantly greater acquisition of nymphal ticks than those entailing exposure solely to leaf litter (Table 1). SpeciÞcally, the mean (23%) for the combined prevalence data for the woodassociated activities was signiÞcantly higher than that for the combined prevalence data (mean, 10.8%) for the two activities involving exposure to leaf litter alone (␹2adj ⫽ 6.60, P ⫽ 0.0102). Data for the walking samples, which also involved contact solely with leaf litter, were excluded from Table 1 because the exposure time during each replicate was one-twelfth of what it was for the other Þve behaviors, i.e., 25 s versus 5 min. Nevertheless, 240 walking replicates produced 4 nymphs over a distance of 6 km for an overall mean of 1 tick per 1.5 km. Three nymphs were collected from running shoes (n ⫽ 60 replicates) for a prevalence of 5% (mean ⫾ SD ⫽ 0.05 ⫾ 0.22, range ⫽ 0Ð1), one from hiking boots (n ⫽ 120 replicates) for a prevalence of 0.8% (mean ⫾ SD ⫽ 0.008 ⫾ 0.091, range ⫽ 0Ð1), and none from hiking sandals (n ⫽ 60 replicates). We conclude that the numbers of ticks acquired while walking were similar for all three types of footwear (␹2adj ⫽ 1.57, P ⫽ 0.211). As anticipated, there was an obvious association between speciÞc behaviors and the bodily regions infested (Table 2). Thus, 70% of ticks acquired while sitting on logs were found on the lower body, especially the buttocks, whereas 95% of ticks acquired while gathering and carrying wood were detected on the upper body. While sitting against trees, subjects acquired about 42Ð50% of the nymphs that infested them from the trunks and the remainder from leaf litter surrounding the bases of such trees (Table 2). Moreover, 38.5% of the nymphs were acquired from the northern aspect, 7.7% from the southern aspect, 15.4% from the eastern aspect, 23.1% from the western aspect, and 15.4% from intermediate aspects (e.g., northeastern).

244


Vol. 41, no. 2

Table 1. Prevalence, mean number, and mean intensity of I. pacificus nymphs acquired by subjects as they engaged in various behaviors in a black-oak woodland in spring 2002 Behavior (n)a

Total no. nymphs

Prevalence (%)b

Mean no. ⫾ SD (range)

Mean intensityc ⫾ SD

Sitting on log (n ⫽ 60) Gathering wood (n ⫽ 80) Sitting against tree (n ⫽ 60) Stirring and sitting on litter (n ⫽ 60) Sitting on litter (n ⫽ 60)

36 19 13 8 6

30.0 22.5 16.7 13.3 8.3

0.60 ⫾ 1.20 (0Ð6) 0.24 ⫾ 0.46 (0Ð2) 0.22 ⫾ 0.56 (0Ð3) 0.13 ⫾ 0.34 (0Ð1) 0.10 ⫾ 0.35 (0Ð2)

2.00 ⫾ 1.41 1.06 ⫾ 0.24 1.30 ⫾ 0.67 1.00 ⫾ 0.10 1.20 ⫾ 0.45

a

n, Number of replicates. Number of replicates yielding at least one tick divided by the total number of replicates. Each replicate was exactly 5 min long, except for gathering wood, in which 5 min is an approximation of the time involved per replicate. c Mean number of nymphs for all samples producing at least one tick. b

Time-Adjusted Acquisition Rates. When the number of hours spent engaged in each behavior is summed across all replicates, the number of nymphs collected per hour by activity was as follows: sitting on logs, 7.2 (n ⫽ 5 h); gathering wood, 2.8 (n ⫽ 6.7 h); sitting against trees, 2.6 (n ⫽ 5 h); walking, 2.4 (n ⫽ 1.7 h); stirring and sitting on litter, 1.6 (n ⫽ 5 h); and sitting on litter, 1.2 (n ⫽ 5 h). Sitting on logs therefore resulted in acquisition of 2.6 Ð 6 times as many ticks as the other behaviors, and although walkers acquired few nymphs, the rate of tick acquisition while walking was the highest of the three behaviors involving exclusive contact with leaf litter. Prevalence of Bacterial Zoonotic Agents in Ticks. In total, 86 nymphs were collected from subjects as they partook in all six behaviors; of these, 76 (88%) were tested for A. phagocytophilum and E. chaffeensis versus 27% for B. burgdorferi s.l. because the latter DNA samples became contaminated (Table 3). Approximately 3Ð 4% of the nymphs assayed by PCR tested positive for A. phagocytophilum or for B. burgdorferi s.l. Of 13 nymphs infected with either A. phagocytophilum or B. burgdorferi s.l., 2 (15.4%) were coinfected with both bacteria, as were 1.3% of 158 nymphs obtained from leaf litter. Statistically comparable percentages of nymphs were infected with A. phagocytophilum (␹2adj ⫽ 0.00, P ⫽ 1.00) or B. burgdorferi s.l. (␹2adj ⫽ 0.00, P ⫽ 1.00) despite the collection method, i.e., dragging as opposed to removal from subjects. None of the nymphs was infected with E. chaffeensis, however. With regard to the 4 infected ticks removed from subjects, 2 (3.3%) of 60 replicates representative of sitting on logs yielded individual nymphs containing A. phagocytophilum DNA, as did 1 (0.4%) of 240 repliTable 2.

cates spent walking. One (1.25%) of 80 replicates involved in gathering wood produced a nymph infected with a spirochete related to, but distinct from B. burgdorferi sensu stricto (s.s.). The prevalence data for spirochete-infected nymphs with respect to activity doubtless contain one or more underestimates because only 27% of the 86 nymphs removed from subjects engaged in all six behaviors were tested for presence of B. burgdorferi s.l. Six of the nine DNA extracts from nymphs that tested positive by PCR for A. phagocytophilum were reconÞrmed by sequencing analyses. Four sequences showed 100% homology with A. phagocytophilum, and two had but a single nucleotide substitution. To determine which genospecies of B. burgdorferi s.l. was present in each of the seven infected nymphs, a 201-bp region of the 5S-23S rRNA intergenic spacer was ampliÞed using species-speciÞc primers. Species were characterized by comparing their DNA sequences with those of strains available in GenBank. Two different Borrelia genotypes were identiÞed. Six of the seven sequences were 98.5% similar to B. burgdorferi s.s. isolate SI-1 (AF221680), which was isolated from a cotton mouse (Peromyscus gossypinus [Le Conte]) from Sapelo Island, Georgia (Lin et al. 2001). All six of these specimens exhibited a 2-bp deletion relative to B. burgdorferi s.s. SI-1 isolate. Three of the six were identical, two differed by a single nucleotide substitution, and one differed by a single nucleotide substitution and a single nucleotide deletion. The other sequence was identical with B. burgdorferi s.l. isolate CA31 (AJ006372), which is distinct from the B31 type strain of B. burgdorferi s.s. (Postic et al. 1999). The latter isolate originated from a California kanga-

Number of I. pacificus nymphs collected from two subjects by behavior and bodily region Percentage of ticks collected by bodily region (% of total)

Behavior (n)a Sitting on log (36) Gathering wood (19) Sitting against treeb (12) Stirring and sitting on litter (8) Sitting on litter (6) Walking (4) a b

Foot 2.8 8.3 33.3 100

Lower leg

Knee

2.8 25.0

8.3 37.5

n, Number of ticks collected. Datum for one additional nymph not recorded.

Upper leg 22.2 5.3 12.5 33.3

Crotch

Buttocks

Back

2.8

38.9

11.1

8.3

25.0

16.7

16.7

Abdomen

Chest

Shoulder

Arm

Hand

5.3

26.3

5.6 5.3

2.8 31.6 16.7 12.5

11.1 26.3 33.3 12.5

March 2004


245

Table 3. Prevalence (%) of A. phagocytophilum, B. burgdorferi s.l., and E. chaffeensis infection in I. pacificus nymphs collected either from subjects or leaf litter in a black-oak woodland in spring 2002 No. nymphs positive/no. tested (%) by agent

Source of nymphs

A. phagocytophilum

B. burgdorferi s.l.

E. chaffeensis

Subjectsa Dragging litterc Totals

3/76 (3.9) 5/158 (3.2) 8/234 (3.4)

1/23 (4.3)b 6/158 (3.8) 7/181 (3.9)

0/76 0/158 0/234

a

Subjects acquired nymphal ticks on their clothing or skin while engaged in various behaviors on six dates. In total, 53 of 76 samples inadvertently were contaminated with B. burgdorferi extract, and therefore were excluded from this analysis. All ticks tested were collected by dragging during the Þrst sampling occasion only (28 May). Two nymphs were coinfected with A. phagocytophilum and B. burgdorferi s.l. b c

roo rat (Dipodomys californicus Merriam) captured at the HREC. Discussion Acquisition of Nymphs by Behavior. In a predominantly black-oak woodland in northern California, three behaviors involving direct contact with wood for only a few minutes were found to elevate the risk of exposure to I. pacificus nymphs and their associated bacterial zoonotic agents. In particular, sitting on logs entails considerable risk and is likely to lead to multiple tick exposures if performed repeatedly during the nymphal activity period. However, two of three behaviors involving exclusive contact with leaf litter posed considerably less risk (Table 1). The third such behavior, walking, resulted in acquisition of a mean of only one tick per 1.5 km, but the time-adjusted acquisition rate (2.4 nymphs per hour) for walker samples was higher than the rates for either sitting on litter (1.2 nymphs per hour) or stirring and sitting on litter (1.6 nymphs per hour). The speciÞc behaviors evaluated were chosen because they represent a wide range of activities that people might engage in while recreating or working in deciduous hardwood forests. Thus, sitting against trees was selected because it imitates the behavior of turkey hunters who frequently assume that position while calling toms during the spring hunting season; amassing tree branches and limbs, and then carrying them was chosen to mimic gathering Þrewood by campers; and sitting in and stirring leaf litter with oneÕs hands might reßect gardening activities by individuals residing in semirural woodland settings. The density of I. pacificus nymphs in leaf litter adjacent to logs (n ⫽ 32) at the HREC previously was reported to be about 3 times higher than it was just 3 m distant, which presumably was related to the utilization of logs by western fence lizards and rodents (Ta¨lleklint-Eisen and Lane 2000a). After we discovered that logs were being used by nymphs as host-seeking sites, a log that infested one of us with six nymphs during experimentation was examined carefully for presence of additional nymphs. Host-seeking nymphs were readily observed on the upper surface of this log, which was 20 cm in diameter, devoid of moss, and surrounded by leaf litter. In an area encompassing 0.6 m2 on the same log, sampling with a lint roller produced 13 nymphs. Similarly, large fallen logs in de-

ciduous woods were found to be important questing sites for I. scapularis nymphs in Maryland (Carroll and Kramer 2001). In that study, nymphs were detected on 87% of the logs examined and in 36% of the samples taken by appressing a 0.25-m2 (0.5 ⫻ 0.5-m2) ßannel cloth to their upper surfaces for ⬇5 s. Thus, the two primary tick vectors of B. burgdorferi in North America both use logs in deciduous hardwoods as questing sites, and the public should be advised to either avoid or at least minimize sitting on or resting against them when nymphal ticks are active. Furthermore, cutting wood and gathering fallen branches or limbs impose elevated risk for acquiring Lyme disease and I. pacificus nymphs, respectively (Lane et al. 1992; current study). In California and Maryland (Carroll and Kramer 2001), walking through leaf litter appears to pose modest risk for acquiring Ixodes spp. nymphs, even at times when nymphal populations are quite high. During our Þrst drag-sampling occasion (28 May) when the relative abundance of I. pacificus nymphs was determined to be 32 ticks per 100 m2, only 3 nymphs were detected on the footwear of both subjects as they traversed a combined 1 km of leaf litter. We consider the mean density of nymphs present then to be extraordinarily high, at least for the HREC, because in a previous investigation there (Ta¨lleklint-Eisen and Lane 1999), the area-wide mean density of nymphs collected by dragging leaf litter at 12 sites was 6.1 ticks per 100 m2 (range, 2.4 Ð9.4). To put the 2002 data in even better perspective, the efÞciency of a single drag-sampling occasion for estimating the population size of I. pacificus nymphs in leaf litter is only ⬇5.9% (Ta¨lleklint-Eisen and Lane 2000b). Thus, the actual number of nymphs present in the leaf litter sampled on 28 May in the current study may have approximated 540 nymphs per 100 m2. Of the three types of footwear compared in the current study, a prevalence of acquisition exceeding 1% per replicate (⬇25-s walk) was recorded only for the running shoes. In Maryland, however, nymphs of I. scapularis were acquired and retained on 22 or 24% of 30-s walks when boots or sneakers were worn (Carroll and Kramer 2001). The relative density of nymphs collected by dragging in that study was about 3 times higher than that recorded from leaf litter by us, and was equivalent to roughly 96 ticks per 100 m2. That is, 239 ticks were collected from Þve sites in which they had dragged ⬇100 m per site with a ßannel cloth

246


having a diameter of 0.5 m. The greater abundance of ticks recorded in the Maryland sites alone may account for the higher prevalence of nymphs on the footwear evaluated by Carroll and Kramer (2001), aside from any differences related to either the particular types of footwear worn or the distinct biologies of the two ticks. Nymphs of I. pacificus in California and I. scapularis in Maryland have been recorded from tree trunks before by appressing their bark with ßannel cloth (Slowik and Lane 2001, Carroll 2002). In a woodlandgrass habitat at the HREC, medium- and large-sized trees representing several species, and having basal leaf-litter and moss-covered trunks, were signiÞcantly associated with presence of nymphs (Slowik and Lane 2001). Apart from the lack of tree diversity, these are the same attributes possessed by the black oaks in the dense woodland that we assessed, so it was anticipated that sitting against trees would result in acquisition of some ticks. Slowik and Lane (2001) reported that 30% of 600 tree trunks yielded an overall mean of 0.79 I. pacificus nymphs per tree. Unsurprisingly, we found that sitting against tree trunks rather than appressing them with ßannel produced both a lower prevalence (17%) and mean number (0.22) of nymphal ticks. In Maryland, the percentage of trees (6%) producing 1 or more I. scapularis nymphs was even lower, as only 6 ticks were collected in toto from 5 of 83 trees (Carroll 2002). The discovery that nearly 40% of the ticks infesting subjects did so while they sat against the northern aspect of tree trunks may have biological signiÞcance; it may be an artifact of our experimental design; or it may reßect the few ticks enumerated (Table 1). Our design did not allow us to deÞnitively address this question or to analyze the data statistically because we sampled only one or two aspects of each tree. In hindsight, a more informative design would have been to consistently sample all four aspects of every tree on each sampling occasion. In Maryland, over one-half of the I. scapularis nymphs collected from leaf litter adjacent to the bases of living trees were from their northern aspects, a Þnding that suggests that the nymphs preferentially avoid even limited direct insolation and its covariants present in the other aspects (Carroll 2002). Prevalence of Bacterial Zoonotic Agents and Associated Acarologic Risk. Because 3.2 and 3.8% of the I. pacificus nymphs tested positive for A. phagocytophilum or B. burgdorferi s.l. during the Þrst drag-sampling occasion when the density of nymphs was 32.4 per 100 m2, the mean density of A. phagocytophilum-infected nymphs would have been ⬇1.04 per 100 m2, and that of B. burgdorferi s.l.-infected nymphs ⬇1.23 per 100 m2. The mean number of infected ticks per unit area is an indirect measure of the potential risk of exposure to vector ticks. With such seemingly low densities of infected ticks in the black-oak grove, the likelihood of anyone being exposed to the attachment of a nymph infected with either A. phagocytophilum or B. burgdorferi would have been low in 2002, unless an individual had been bitten repeatedly. Our Þndings with

Vol. 41, no. 2

respect to ticks infesting subjects also support this conclusion inasmuch as only four infected nymphs were detected among those removed from both subjects during the performance of all six behaviors, and the one nymph infected with B. burgdorferi s.l.-infected spirochetes contained DNA representing a genospecies not known to be pathogenic for humans. We realize, however, that had all 86 ticks discovered on subjects been tested for presence of these bacterial agents, the prevalence of infected ticks, particularly spirochete-infected nymphs, most likely would have been higher for one or more of the behaviors. In that regard, all but one of the B. burgdorferi s.l.-infected nymphs were determined to harbor the Lyme disease spirochete, B. burgdorferi s.s. Similarly, 10 of 10 borrelial isolates obtained from I. pacificus nymphs that were collected from leaf litter in a mixed-hardwood forest in nearby Sonoma County, California, were identiÞed as B. burgdorferi s.s. by restriction fragment length polymorphism and analysis of rrf (5S)-rrl (23S) intergenic spacer amplicons (Lane et al. 2001). The Þndings that 1.3% of the nymphs (n ⫽ 158) collected from leaf litter by dragging were coinfected with A. phagocytophilum and B. burgdorferi, and that 15.4% of those (n ⫽ 13) infected with either agent, irrespective of the method of collection, were coinfected, represent the Þrst such reported coinfection in this life stage and the second for any life stage of I. pacificus (Lane et al. 2001). Although I. pacificus nymphs commonly are infected with B. burgdorferi s.l. and infection prevalences in different localities can reach 12Ð 41% in this stage (Clover and Lane 1995, Ta¨lleklint-Eisen and Lane 1999, Lane et al. 2001), only one nymph had been found to be infected before with A. phagocytophilum (Barlough et al. 1997). Interestingly, 95 I. pacificus adults from Sonoma County, California, that were tested individually for presence of A. phagocytophilum by PCR and for B. burgdorferi by culture yielded an overall coinfection prevalence of 4.2%. Among infected ticks (n ⫽ 14), the coinfection prevalence was 28.6% (Lane et al. 2001), which is nearly double that observed for nymphs in the current study. To date, we have assayed 646 nymphs from two counties (Mendocino, Sonoma) for presence of E. chaffeensis by PCR without detecting it (Lane et al. 2001; current study). Therefore, if this rickettsial agent indeed does occur in populations of I. pacificus nymphs in northwestern California, its spatial distribution must be highly focal. Of this total, 412 of the nymphs, as well as 202 adult ticks that also tested negative, were obtained from a semirural community in Sonoma County, where 4.6% of the 219 residents surveyed had seroreacted to E. chaffeensis antigen (Fritz et al. 1997). However, E. chaffeensis was detected in I. pacificus adults and Dermacentor variabilis (Say) ticks from central coastal California by nested PCR (Kramer et al. 1999), and 24 serum samples from Californians reacted to E. chaffeensis or to the closely related E. canis (Donatien and Lestoquard) Moshkovski antigen between 1986 and 1997 (McQuiston et al. 1999). During a more recent survey of three species

March 2004


of adult human-biting ixodid ticks in recreational areas in central coastal California, E. chaffeensis was detected by nested PCR in 0.64% of 776 I. pacificus, 6.9% of 58 D. variabilis, and 0.28% of 353 D. occidentalis (Holden et al. 2003). Furthermore, one (0.12%) of the I. pacificus was found to be coinfected with B. burgdorferi and E. chaffeensis, and eight (1.03%) other I. pacificus contained both A. phagocytophilum and B. burgdorferi (Holden et al. 2003). Obviously, much remains to be learned about the basic transmission cycle of E. chaffeensis in the far-western United States, including what species of ticks and vertebrates maintain and distribute this human pathogen in different habitats. Public Health Implications. In certain areas of Mendocino County, individuals who spend considerable time in deciduous hardwood forests may be at elevated risk for exposure to tick-borne bacterial zoonotic agents. In a small rural community at high risk for Lyme disease in the Ukiah area, for instance, the mean density of B. burgdorferi s.l.-infected nymphs was 1.8 per 100-m2 area-wide; it varied a remarkable 76-fold between 12 separate properties and, on one of those properties, it was 22.2 per 100 m2 (Ta¨lleklint-Eisen and Lane 1999). The cumulative frequency of seroreactivity to B. burgdorferi in that community was ⱖ24%, and 31 of 83 residents examined physically were diagnosed as having either deÞnite (13) or probable (18) Lyme disease (Lane et al. 1992). In 2001, an intensive survey was conducted at 12 sites scattered throughout inland areas of the county to determine the seasonality and density of B. burgdorferi s.l.-infected I. pacificus nymphs in three primary habitat types (oak, oak/Douglas Þr, and redwood/tanoak woodlands), in which humans are at risk of exposure to this vector tick (Eisen et al. 2003). The amount of variation detected at this scale was extraordinary too: peak nymphal densities varied 35-fold from 0.8 to 30.1 per 100 m2 (median, 2.95); infection prevalences ranged 40-fold from 0.4 to 16.4% (median, 7.2%); and the peak density of infected nymphs varied 490-fold from 0.01 to 4.9 infected nymphs per 100 m2 (median, 0.26). There is a deÞnite need for similar kinds of risk-assessment investigations to be carried out on the nymphal stage of I. pacificus in other regions of California, where its associated bacterial zoonotic agents are known to be a threat to human health. Also, behaviors that potentially may elevate the risk of human exposure to I. pacificus nymphs, such as those evaluated during the current study, should be assessed in other deciduous hardwood forests (e.g., oak/Douglas Þr woodlands) in endemic regions of the state to determine whether contact with wood generally poses higher risk than contact with leaf litter. In northwestern California, most incident cases of Lyme disease are contracted in spring and summer when populations of I. pacificus nymphs are reaching their annual peak densities (Ley et al. 1994, Clover and Lane 1995). The difÞculty inherent in detecting attached Ixodes spp. nymphs is exempliÞed by data from one series of patients in New York who presented with erythema migrans, the pathognomonic skin lesion of

247

early-stage Lyme disease (Berger 1989). Of 237 patients with this condition, 85% were diagnosed in spring and summer (May through September) when I. scapularis nymphs are active, but only 14% recalled a tick bite. By the time that some attached I. scapularis nymphs happen to be discovered, they usually will have been feeding on humans for a longer period than the more conspicuous female ticks (Yeh et al. 1995, Falco et al. 1996). Although comparable epidemiological data have not been recorded systematically for I. pacificus, it is known that the nymphs do not begin to efÞciently transmit B. burgdorferi to naive mice until they have been attached for 3Ð 4 days (Peavey and Lane 1995). Despite the extreme personal preventive measures we took in terms of both the attire worn (Fig. 1) and the thoroughness of the inspections after each exposure period, which far exceeds whatever precautions the average person might take, one subject was bitten twice over the 6-d sampling period and did not discover the partially replete ticks until they had been attached for at least 1Ð2 d. Our Þndings underscore the advisability of the oft-cited recommendation that individuals recreating or working in areas endemic for tick-borne diseases should inspect themselves frequently during seasons when vector ticks are active. They also argue convincingly that, after exposure to nymphal-tick biotopes such as deciduous hardwood forests, individuals should continue to examine themselves for another 2Ð3 d because partially replete nymphs are much more easily detected than are recently attached, unfed ticks. Acknowledgments We gratefully acknowledge Lars and Rebecca Eisen for their helpful comments on the manuscript, Colin Brooks for habitat-related information, Durland Fish for providing a pertinent reference, Robert J. Keiffer for identifying birds within the study area, Andrew Muss for preliminary testing of some nymphal ticks, and Hans Peeters for identiÞcation of wood-rat droppings. This research was funded in part by Cooperative Agreement U50/CCU906594 from the Centers for Disease Control and Prevention and by Grant AI-22501 from the National Institutes of Health to R.S.L.

References Cited Anderson, B. E., J. W. Sumner, J. E. Dawson, T. Tzianabos, C. R. Greene, J. G. Olson, D. B. Fishbein, M. OlsenRasmussen, B. P. Holloway, E. H. George, and A. F. Azad. 1992. Detection of the etiologic agent of human ehrlichiosis by polymerase chain reaction. J. Clin. Microbiol. 30: 775Ð780. Barlough, J. E., J. E. Madigan, V. L. Kramer, J. R. Clover. L. T. Hui, J. P. Webb, and L. K. Vredevoe. 1997. Ehrlichia phagocytophila genogroup rickettsiae in ixodid ticks from California collected in 1995 and 1996. J. Clin. Microbiol. 35: 2018 Ð2021. Berger, B. W. 1989. Dermatologic manifestations of Lyme disease. Rev. Infect. Dis. 11 (Suppl. 6): S1475Ð1481. Black, W. C., IV, and J. Piesman. 1994. Phylogeny of hardand soft-tick taxa (Acari: Ixodida) based on mitochon-

248


drial 16S rDNA sequences. Proc. Natl. Acad. Sci. USA 91: 10034 Ð10038. Carroll, J. F. 2002. A survey for nymphs of host-seeking Ixodes scapularis and Amblyomma americanum (Acari: Ixodidae) on tree trunks in deciduous forests. J. Med. Entomol. 39: 237Ð240. Carroll, J. F., and M. Kramer. 2001. Different activities and footwear inßuence exposure to host-seeking nymphs of Ixodes scapularis and Amblyomma americanum (Acari: Ixodidae). J. Med. Entomol. 38: 596 Ð 600. [CDC] Centers for Disease Control and Prevention. 1999. Recommendations for the use of Lyme disease vaccine: recommendations of the Advisory Committee on Immunization Practices (ACIP). Morbid. Mortal. Wkly. Rep. 48 (RR-7): 1Ð25. [CDC] Centers for Disease Control and Prevention. 2000. Surveillance for Lyme disease: United States, 1992Ð1998. Morbid. Mortal. Wkly. Rep. 49 (SS-3): 1Ð11. Clover, J. R., and R. S. Lane. 1995. Evidence implicating nymphal Ixodes pacificus (Acari: Ixodidae) in the epidemiology of Lyme disease in California. Am. J. Trop. Med. Hyg. 53: 237Ð240. Eisen, L., R. J. Eisen, and R. S. Lane. 2002. Seasonal activity patterns of Ixodes pacificus nymphs in relation to climatic conditions. Med. Vet. Entomol. 16: 235Ð244. Eisen, R. J., L. Eisen, M. B. Castro, and R. S. Lane. 2003. Environmentally related variability in risk of exposure to Lyme disease spirochetes in northern California: effect of climatic conditions and habitat type. Environ. Entomol. 32: 1010 Ð1018. Falco, R. C., D. Fish, and J. Piesman. 1996. Duration of tick bites in a Lyme disease-endemic area. Am. J. Epidemiol. 143: 187Ð192. Falco, R. C., D. F. McKenna, T. J. Daniels, R. B. Nadelman, J. Nowakowski, D. Fish, and G. P. Wormser. 1999. Temporal relation between Ixodes scapularis abundance and risk for Lyme disease associated with erythema migrans. Am. J. Epidemiol. 149: 771Ð776. Fritz, C. L., A. M. Kjemtrup, P. A. Conrad, G. R. Flores, G. L. Campbell, M. E. Schriefer, D. Gallo, and D. J. Vugia. 1997. Seroepidemiology of emerging tickborne infectious diseases in a northern California community. J. Infect. Dis. 175: 1432Ð1439. Holden, K., J. T. Boothby, S. Anand, and R. F. Massung. 2003. Detection of Borrelia burgdorferi, Ehrlichia chaffeensis, and Anaplasma phagocytophilum in ticks (Acari: Ixodidae) from a coastal region of California. J. Med. Entomol. 40: 534 Ð539. Kramer, V. L., M. P. Randolph, L. T. Hui, W. E. Irwin, A. G. Gutierrez, and D. J. Vugia. 1999. Detection of the agents of human ehrlichioses in ixodid ticks from California. Am. J. Trop. Med. Hyg. 60: 62Ð 65. Lane, R. S., S. A. Manweiler, H. A. Stubbs, E. T. Lennette, J. E. Madigan, and P. E. Lavoie. 1992. Risk factors for Lyme disease in a small rural community in northern California. Am. J. Epidemiol. 136: 1358 Ð1368. Lane, R. S., J. E. Foley, L. Eisen, E. T. Lennette, and M. A. Peot. 2001. Acarologic risk of exposure to emerging tick-borne bacterial pathogens in a semirural community in northern California. Vector Borne Zoonotic Dis. 1: 197Ð210. Ley, C., I. H. Davila, N. M. Mayer, R. A. Murray, G. W. Rutherford, and A. L. Reingold. 1994. Lyme disease in northwestern coastal California. West. J. Med. 160: 534 Ð 539. Ley, C., E. M. Olshen, and A. L. Reingold. 1995. Case-control study of risk factors for incident Lyme disease in California. Am. J. Epidemiol. 142: S39 ÐS47.

Vol. 41, no. 2

Lin, T., J. H. Oliver, Jr., L. Gao, T. M. Kollars, Jr., and K. L. Clark. 2001. Genetic heterogeneity of Borrelia burgdorferi sensu lato in the southern United States based on restriction fragment length polymorphism and sequence analysis. J. Clin. Microbiol. 39: 2500 Ð2507. Massung, R. F., K. Slater, J. H. Owens, W. L. Nicholson, T. N. Mather, V. B. Solberg, and J. G. Olson. 1998. Nested PCR assay for detection of granulocytic ehrlichiae. J. Clin. Microbiol. 36: 1090 Ð1095. Mather, T. N., M. C. Nicholson, E. F. Donnelly, and B. T. Matyas. 1996. Entomologic index for human risk of Lyme disease. Am. J. Epidemiol. 144: 1066 Ð1069. McQuiston, J. H., C. D. Paddock, R. C. Holman, and J. E. Childs. 1999. The human ehrlichioses in the United States. Emerg. Infect. Dis. 5: 635Ð 642. Nicholson, M. C., and T. N. Mather. 1996. Methods for evaluating Lyme disease risks using geographic information systems and geospatial analysis. J. Med. Entomol. 33: 711Ð720. Peavey, C. A., and R. S. Lane. 1995. Transmission of Borrelia burgdorferi by Ixodes pacificus nymphs and reservoir competence of deer mice (Peromyscus maniculatus) infected by tick-bite. J. Parasitol. 81: 175Ð178. Postic, D., M. V. Assous, P.A.D. Grimont, and G. Baranton. 1994. Diversity of Borrelia burgdorferi sensu lato evidenced by restriction fragment length polymorphism of rrf (5S)-rrl (23S) intergenic spacer amplicons. Int. J. Syst. Bacteriol. 44: 743Ð752. Postic, D., N. Marti Ras, R. S. Lane, P.-F. Humair, M. M. Wittenbrink, and G. Baranton. 1999. Common ancestry of Borrelia burgdorferi sensu lato strains from North America and Europe. J. Clin. Microbiol. 37: 3010 Ð3012. Slowik, T. J., and R. S. Lane. 2001. Nymphs of the western black-legged tick (Ixodes pacificus) collected from tree trunks in woodland-grass habitat. J. Vector Ecol. 26: 165Ð 171. Sonenshine, D. E. 1993. Biology of ticks, vol. 2. Oxford University Press, New York Stafford, K. C., III, M. L. Cartter, L. A. Magnarelli, S.-H. Ertel, and P. A. Mshar. 1998. Temporal correlations between tick abundance and prevalence of ticks infected with Borrelia burgdorferi and increasing incidence of Lyme disease. J. Clin. Microbiol. 36: 1240 Ð1244. Stafford, K. C., III, and L. A. Magnarelli. 1993. Spatial and temporal patterns of Ixodes scapularis (Acari: Ixodidae) in southeastern Connecticut. J. Med. Entomol. 30: 762Ð 771. Ta¨ lleklint-Eisen, L., and R. S. Lane. 1999. Variation in the density of questing Ixodes pacificus (Acari: Ixodidae) nymphs infected with Borrelia burgdorferi at different spatial scales in California. J. Parasitol. 85: 824 Ð 831. Ta¨ lleklint-Eisen, L., and R. S. Lane. 2000a. Spatial and temporal variation in the density of Ixodes pacificus (Acari: Ixodidae) nymphs. Environ. Entomol. 29: 272Ð280. Ta¨ lleklint-Eisen, L., and R. S. Lane. 2000b. EfÞciency of drag sampling for estimating population sizes of Ixodes pacificus (Acari: Ixodidae) nymphs in leaf litter. J. Med. Entomol. 37: 484 Ð 487. Yeh, M.-T., J. M. Bak, R. Hu, M. C. Nicholson, C. Kelly, and T. N. Mather. 1995. Determining the duration of Ixodes scapularis (Acari: Ixodidae) attachment to tick-bite victims. J. Med. Entomol. 32: 853Ð 858.

Received for publication 22 May 2003; accepted 3 October 2003.