asite of white-tailed deer (Ba. odocoilei) that is distributed widely throughout the southern United States.5, 6 The closely related black-legged tick (I. scapularis) ...
Am. J. Trop. Med. Hyg., 58(6), 1998, pp. 739–742 Copyright q 1998 by The American Society of Tropical Medicine and Hygiene
DIVERSITY OF BABESIA INFECTING DEER TICKS (IXODES DAMMINI) PHILIP M. ARMSTRONG, PAULA KATAVOLOS, DIANE A. CAPORALE, ROBERT P. SMITH, ANDREW SPIELMAN, SAM R. TELFORD III Department of Tropical Public Health, Harvard School of Public Health, Boston, Massachusetts; Research Department, Maine Medical Center, Portland, Maine
AND
Abstract. To determine whether the presence of nonpathogenic piroplasms may confound field estimates of risk of Babesia microti infection, we identified sporozoites infecting the salivary glands of deer ticks (Ixodes dammini) by parallel microscopy and polymerase chain reaction assays. Piroplasms were evident in 14.4% of adult ticks from sites in the northcentral and northeastern United States. Of these, 83.3% contained DNA characteristic of Ba. odocoilei. This cervid piroplasm was detected in all of the sites examined and generally was more prevalent than was Ba. microti. Because deer ticks transmit both Ba. odocoilei and Ba. microti, estimates of pathogen prevalence based solely on microscopy may overestimate the risk of human babesiosis. The agents of human babesiosis (Babesia microti) and Lyme disease (Borrelia burgdorferi) co-occur in particular sites in the northeastern and northcentral regions of the United States, where they perpetuate in a cycle involving deer tick (Ixodes dammini) vectors and white-footed mouse (Peromyscus leucopus) reservoir hosts.1 Where both kinds of pathogens infect these hosts, human babesiosis is less frequently diagnosed than is Lyme disease. On Nantucket Island, for example, the annual incidence of symptomatic human babesiosis averages about one-tenth that of Lyme disease.2 Moreover, sera of local residents react to antigens of Ba. microti about one-third as often as to those of B. burgdorferi. Although they are acquired similarly, human babesial infection occurs less frequently than does borrelial infection. In spite of this apparent difference in the relative frequency of symptomatic infection, the agent of human babesiosis seems to infect vector ticks in nature about as often as does the agent of Lyme disease.3, 4 These estimates, however, assume that any sporozoites seen microscopically in the salivary glands of deer ticks would be Ba. microti. Yet, Ixodes ticks transmit a diverse array of piroplasms, including a parasite of white-tailed deer (Ba. odocoilei) that is distributed widely throughout the southern United States.5, 6 The closely related black-legged tick (I. scapularis) transmits Ba. odocoilei in Oklahoma, Texas, and Florida.7 We do not know, however, whether diverse Babesia parasites infect deer ticks in the northeastern and upper midwestern United States. It may be that where Ba. microti is zoonotic, other piroplasms may also infect the salivary glands of deer ticks. To evaluate this suggestion, we identified the sporozoites infecting the salivary glands of deer ticks in the northeastern and northcentral United States. In particular, we examined a salivary gland microscopically from each of a series of ticks, screened the other by amplifying its DNA with genus-specific primers and identified the amplification products by restriction enzyme analysis and sequencing.
obtained from hunter-killed deer during the 1995 hunting season. Ticks were dissected and their salivary glands were examined microscopically to determine whether sporozoites were present. Thus, non-engorged female ticks were placed on laboratory rabbits and removed after 4–5 days to stimulate sporogony. Ticks taken from deer were reserved at 48C for study if they appeared to have fed for 4–5 days. Ticks were dissected individually on a new slide using flame-sterilized forceps to avoid cross-contamination. Salivary glands were removed in a drop of 10% fetal bovine serum in phosphate-buffered saline and one of each pair of glands was stained by the Feulgen reaction for examination by brightfield microscopy.3 The other gland was pooled with the salivary glands from four other ticks for polymerase chain reaction (PCR) analysis. Sporozoite-infected salivary gland pools were further analyzed by PCR amplification. The salivary gland pools were homogenized in 100 ml of a lysis buffer (4 M guanidine thiocyanate, 25 mM sodium citrate, 0.5% N-lauroylsarcosine) with a heat-sealed pipet tip and extracted with an equal volume of phenol-chloroform-isoamyl alcohol (24:24:1). The DNA was precipitated by adding 0.15 volumes of 2 M sodium acetate, 2 volumes of 100% ethanol, and 10 mg of glycogen after incubation at -708C for 30 min. The DNA pellets were suspended in 50 ml of water and 5 ml of this preparation were added to a 50-ml of PCR mixture using the Elongase System (Life Technologies, Gaithersburg, MD) at a final Mg11 concentration of 1.5 mM and containing 25 pmols of primers PIRO-A(5’-AATACCCAATCCTGACACAGGG-3’) and PIRO-B(5’-TTAAATACGAATGCCCCCAAC-3’). Primers PIRO-A and PIRO-B were designed to amplify 408- and 437-basepair (bp) fragments from the 18S rRNA gene of Ba. odocoilei and Ba. microti, respectively (Figure 1). Amplification was performed as follows: 1 min at 948C for (one cycle) followed by 40 cycles for 45 sec at 948C, 45 sec at 558C, and 45 sec at 728C. To prevent DNA contamination, we prepared samples in dedicated rooms for setting up PCRs and used only autoclaved barrier tips, tubes, and solutions. We are aware of the possibility that some of the ticks collected off of deer may contain Ba. odocoilei DNA from the current blood meal, which may overestimate the prevalence of Ba. odocoilei. To monitor samples for potential DNA contamination, micros-
METHODS
Deer ticks were sampled from intensely infested sites in the northeastern (Maine and Massachusetts) and the upper midwestern (Wisconsin) United States. Host-seeking ticks were collected by dragging a piece of flannel cloth over the vegetation during 1995 and 1996. In addition, samples were
739
740
ARMSTRONG AND OTHERS
FIGURE 1. DNA alignment of the 18S rRNA gene corresponding to polymerase chain reaction primers PIRO-A and PIRO-B and restriction enzyme sites Bst EII and Hinf I as indicated by the black arrows.
copy-negative salivary glands were included with every batch of samples analyzed by PCR. Amplification products were identified by digestion with selected restriction enzymes in separate reactions. Restriction products were separated by electrophoresis on 3% agarose gels and visualized by staining with ethidium bromide. Representative PCR products derived from infected ticks collected in Ipswich, (MA), Great Island, (MA), Wells, (ME), Monhegan Island, (ME), and Spooner, (WI) were sequenced by the dideoxy-chain termination method using an automated DNA sequencer (Applied Biosystems, Foster City, CA). Laboratory animals used in this study were maintained in compliance with the National Institutes of Health guidelines for humane use of laboratory animals. RESULTS
By means of sequence analysis, we identified restriction sites that might be used to distinguish Ba. odocoilei from
Ba. microti. The enzyme Bst EII appeared likely to digest within the PIRO-A/B amplification products of Ba. odocoilei as well as Ba. divergens, thereby generating 78- and 330-bp fragments (Figure 1). In contrast, Hinf I, would digest such sites within the products of Ba. microti and Ba. rodhaini, and this would generate 81- and 356-bp fragments. Our targeted sequences appear to contain diagnostic restriction sites. In a preliminary experiment, we evaluated the specificity of the proposed PCR–restriction fragment length polymorphism (RFLP) assay with respect to related Babesia and Theileria organisms. Our primers amplified DNA from Ba. microti, Ba. odocoilei, and Theileria cervi, a piroplasm prevalent in Amblyomma americanum (Armstrong PM, unpublished data). The electrophoretic mobility of the amplification products varied slightly among different piroplasms and ranged from 408 to 437 bps in size (Figure 2). Babesia microti was readily distinguished from other piroplasms when digested with Hinf I whereas Bst EII cut at sites solely within Ba. odocoilei amplification products. The PCR amplification coupled with restriction enzyme digestion may readily distinguish Ba. odocoilei from Ba. microti. To determine the prevalence of Babesia infection in I. dammini, one of the salivary gland pairs from each tick was prepared for microscopy (Table 1). Sporozoites were evident in ticks sampled from each of the sites examined in this study. Babesia was most prevalent in ticks from Wells, ME, Monhegan Island, ME, and Great Island, MA (36.7%,
TABLE 1 Prevalence of piroplasms in deer ticks by microscopy Ticks analyzed Ticks sampled in State
FIGURE 2. Restriction digests of Babesia and Theileria PIRO-A/ PIRO-B amplification products. Lane 1, DNA size marker; lane 2, Ba. microti no digest; lane 3, Ba. odocoilei no digest; lane 4, T. cervi no digest; lane 5, water control; lane 6, Ba. microti Hinf I digest; lane 7, Ba. odocoilei Hinf I; lane 8, T. cervi Hinf I; lane 9, Ba. microti Bst EII; lane 10, Ba. odocoilei Bst EII; lane 11, T. cervi Bst EII; lane 12, DNA size marker. Values on the left are in basepairs.
Massachusetts
Maine Wisconsin
Site
Ipswich Martha’s Vineyard Great Island Nantucket Wells Monhegan Island Spooner
Stage
No.
% with sporozoites
Adult Adult Adult Adult Adult Adult Adult
137 62 23 130 30 53 32
6.6 14.5 21.7 7.7 36.7 32.1 18.8
BABESIA INFECTING DEER TICKS
TABLE 2 Polymerase chain reaction–restriction fragment length polymorphism (PCR-RFLP) analysis of sporozoite-infected salivary gland pools No. of pools reacted with Babesia PCR primers*
No. of amplification products cut by
Site
Tested
Amplified
Bst EII
Hinf I
Bst EII and Hinf I
Ipswich Martha’s Vineyard Great Island Nantucket Island Wells Monhegan Island Spooner
8 5 4 8 11 15 5
8 4 4 8 11 15 4
7 1 2 8 9 15 2
1 3 2 0 1 0 1
0 0 0 0 0 0 1
* Some pools contained salivary glands from more than one infected tick.
32.1%, and 21.7%, respectively). Piroplasms frequently infect Ixodes ticks in each designated study site. To further identify the Babesia detected in these ticks, we analyzed by PCR-RFLP salivary gland preparations of ticks in which piroplasms were recognized microscopically. The PCR assay verified infection in all of the salivary gland pools that contained at least one microscopy-positive tick, except for a pool from Spooner, WI and another from Martha’s Vineyard, MA (Table 2). Pools of noninfected salivary glands did not yield visible amplification products. The PCR amplification products were then digested with Bst EII and Hinf I. Babesia odocoilei restriction sites were detected in ticks from all sites examined whereas Ba. microti was detected from all but two sites. In addition, one salivary gland pool from Spooner, WI ticks contained an amplification product that was digested by both restriction enzymes, suggesting coinfection by Ba. odocoilei and Ba. microti. Babesia odocoilei appears to be more widespread and prevalent than is Ba. microti. To confirm the identity of Babesia parasites analyzed by PCR-RFLP analysis, we sequenced the amplification products from 1–2 infected ticks representing each geographic region (Genbank accession numbers AF028342-AF028347). Amplification products containing Hinf I sites of infected Spooner, WI and Great Island, MA ticks were identical to Ba. microti. Amplification products derived from Spooner, WI, Ipswich, MA, Wells, ME, and Monhegan Island, ME ticks that were digested with Bst EII were identical to Ba. odocoilei. Deer tick–derived sporozoites represent at least two distinct piroplasm species. DISCUSSION
We found that Ba. odocoilei is more widely distributed than previously recognized, including foci of transmission in the northeast and upper midwest United States. Moreover, Ba. odocoilei appears to infect deer ticks more frequently than does Ba. microti in sites where these piroplasms cooccur (76.9% of sporozoites observed in adult deer ticks). In previous studies, we assumed that all sporozoites observed in deer ticks were Ba. microti largely because no Babesia had been detected in deer by blood smear or hamster inoculation.8, 9 The deer parasites, however, sparsely infect the peripheral blood and often require several days of
741
culture to become detectable.6 Our failure to routinely infect hamsters with salivary gland suspensions from ticks with Feulgen-detectable sporozoites (unpublished data) suggested the possibility that another species might confound our observations. The prevalence of Ba. odocoilei infection in adult deer ticks is about three-fold greater than Ba. microti infection in sites where these parasites coexist. This discrepancy in infection rates may indicate that subadult deer ticks feed frequently on deer, as previously suggested.9 Alternatively, these disparate rates may reflect differences in vector competence for each kind of Babesia parasite. Indeed, the acquisition and trans-stadial passage of Ba. microti by deer ticks is relatively inefficient. Fewer than 50% of nymphal deer ticks acquire Ba. microti infection after feeding on infectious hosts as larvae, and infection generally is lost after subsequent feeding on an uninfected host and molting to the adult stage.10, 11 Nymphs appear to acquire infection less efficiently than do larvae; only 25% of adults acquiring infection as nymphs become infectious.11 Because the ticks examined in this study were adults, our observations may overestimate the intensity of transmission of Ba. odocoilei relative to Ba. microti. The limited capacity of Ba. microti to disperse and establish new foci of transmission may explain its infrequent detection in sites throughout the northeast. Avian hosts may serve as an effective vehicle for long-range dispersal of infected ticks,12 although birds are not competent hosts for known mammalian Babesia. Babesia microti is not maintained by transovarial transmission (TOT) (Telford III SR, unpublished data); therefore, importation requires that ticks must acquire the infection in enzootic sites as larvae, be transported to new sites by avian hosts as nymphs, and subsequently transmit the infection to a suitable host as adults. This scenario seems unlikely because adult deer ticks almost never feed on rodent hosts.13 In addition, Ba. microti is not maintained within the tick after a second molt; Ba. microti parasites acquired by larval ticks disappears before the adult stage.11 Babesia odocoilei, in contrast, is maintained in deer, the main host of adult deer ticks.6 Transovarial transmission of Ba. odocoilei has not been demonstrated experimentally, although it occurs in the closely related species Ba. divergens.14 Babesia divergens infections that are passed by TOT persist to the adult stage of the next generation. Birds, therefore, might readily introduce Ba. odocoilei into new sites. People may differ in their susceptibility to each of the three babesias that have been implicated in human disease in North America. Babesia microti, which commonly is zoonotic in the northern midwest and northeast, is pathogenic mainly among the elderly and immunocompromised including hundreds of known cases.15 Strain WA1, which is closely related to Ba. gibsoni, produces disease in West Coast residents, but only rarely.16, 17 In Europe, only 21 people are known to have been infected by Ba. divergens. Virtually all were asplenic, and about half died of this fulminating infection. Recently, an autochthonous infection with an agent virtually identical to Ba. divergens (designated M01) was implicated in a fatal episode of a Missouri resident.18 Although Ba. odocoilei has not been implicated in human illness, its presence in Ixodes ticks suggests that people may be exposed. Because this piroplasm is so closely related to Ba.
742
ARMSTRONG AND OTHERS
divergens, we anticipate that Ba. odocoile may similarly cause illness in immunocompromised people. These epidemiologic considerations suggest that sporadic cases of nonBa. microti babesiosis may emerge throughout the extensive eastern United States range of I. dammini and perhaps I. scapularis, but Ba. microti remains the main species of public health concern. Entomologic measures of risk need to consider the prevalence of infection in vector ticks, the frequency of tickhuman contact, and the efficiency of pathogen transmission. Although our study estimated pathogen prevalence in ticks, we do not know whether Ba. odocoilei may be efficiently transmitted by deer ticks. Moreover, the prevalence of Babesia odocoilei infection in nymphal deer ticks requires further analysis. Although adult ticks may occasionally feed long enough to transmit infection to people, most human infections are derived from the bites of infectious nymphs. Deer tick nymphs collected from our site in Maryland were exclusively infected with Ba. odocoilei (Armstrong PM, unpublished data). Therefore, it is likely that Ba. odocoilei will be represented frequently in nymphs from Ba. microti-enzootic sites. Our observation that Ba. odocoilei sporozoites frequently infect deer ticks may help reconcile the paucity of human symptomatic Ba. microti infection relative to its apparent frequency in field-collected ticks. Estimates of risk of human infection based solely on microscopic examination of sampled ticks, therefore, may be misleading. Where these piroplasms coexist, Ba. odocoilei may inflate entomologic inoculation rates derived for Ba. microti. Acknowledgments: We thank Peter W. Rand, Eleanor LaCombe, and Mary Holman for technical assistance. This is a contribution of the University of Massachusetts Nantucket Field Station. Financial support: This work was supported by NIH grants AI39002, AI-37993, and AI- 19693; the Gibson Island Corporation; the Chace Fund; and David Arnold. Authors’ addresses: Philip M. Armstrong, Paula Katavolos, Andrew Spielman, and Sam R. Telford, Department of Tropical Public Health, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115. Diane A. Caporale, Department of Biological Sciences, University of Wisconsin-Whitewater, Whitewater, WI 53190. Robert P. Smith, Research Department, Maine Medical Center, 125 John Roberts Road, Suite 5, South Portland, ME 04106. Reprint requests: Sam R. Telford III, Department of Tropical Public Health, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115. REFERENCES
1. Spielman A, Wilson ML, Levine JF, Piesman J, 1985. Ecology of Ixodes dammini-borne human babesiosis and Lyme disease. Annu Rev Entomol 30: 439–460.
2. Spielman A, 1994. The emergence of Lyme disease and human babesiosis in a changing environment. Ann NY Acad Sci 740: 146–156. 3. Piesman J, Mather TM, Donahue JG, Levine J, Campbell JD, Karakashian SJ, Spielman A, 1986. Comparative prevalence of Babesia microti and Borrelia burgdorferi in four populations of Ixodes dammini in eastern Massachusetts. Acta Trop 43: 263–270. 4. Piesman J, Mather TN, Dammin GJ, Telford III SR, Lastavica CC, Spielman A, 1987. Seasonal variation of transmission risk of Lyme disease and human babesiosis. Am J Epidemiol 126: 1187–1189. 5. Telford Jr SR, Forrester DJ, 1991. Piroplasms of white-tailed deer (Odocoileus virginianus) in Florida. Fla Field Nat 19: 49–51. 6. Waldrup KA, Kocan AA, Qureshi T, Davis DS, Baggett D, Wagner GG, 1989. Serological prevalence and isolation of Babesia odocoilei among white-tailed deer (Odocoileus virginianus) in Texas and Oklahoma. J Wildl Dis 25:164–201. 7. Waldrup KA, Kocan AA, Barker RW, Wagner GG, 1990. Transmission of Babesia odocoilei in white-tailed deer (Odocoileus virginianus) by Ixodes scapularis (Acari:Ixodidae). J Wildl Dis 26: 390–391. 8. Piesman J, Spielman A, Etkind P, Ruebush II TK, Juranek DD, 1979. Role of deer in the epizootiology of Babesia microti in Massachusetts, USA. J Med Entomol 15: 537–540. 9. Telford III SR, Mather TN, Moore SI, Wilson ML, Spielman A, 1988. Incompetence of deer as reservoirs of the Lyme disease spirochete. Am J Trop Med Hyg 39: 105–109. 10. Mather TN, Telford III SR, Moore SI, Spielman A, 1990. Borrelia burgdorferi and Babesia microti: efficiency of transmission from reservoir hosts to vector ticks (Ixodes dammini). Exp Parasitol 70: 55–61. 11. Piesman J, Karakashian SJ, Lewengrub S, Rudzinska MA, Spielman A, 1986. Development of Babesia microti sporozoites in adult Ixodes dammini. Int J Parasitol 16: 381–385. 12. Anderson JF, 1988. Mammalian and avian reservoirs for Borrelia burgdorferi. Ann NY Acad Sci 539: 180–191. 13. Carey AB, Krinsky WL, Maim AJ, 1980. Ixodes dammini and associated Ixodid ticks in southcentral Connecticut, USA. J Med Entomol 17: 89–99. 14. Donnelly J, Pierce MA, 1975. Experiments on the transmission of Babesia divergens to cattle by the tick Ixodes ricinus. Int J Parasitol 5: 363–367. 15. Telford III SR, Spielman A, 1998. Human babesiosis. Collier L, Balows A, Sussman M, Kreier J, eds. Topley and Microbiology and Microbial Infections. Ninth edition. London: Arnold, 349–359. 16. Persing DH, Herwaldt BL, Glaser C, Lane RS, Thomford JW, Mathiesen D, Krause PJ, Phillip DF, Conrad PA, 1995. Infection with a babesia-like organism in northern California. N Engl J Med 332: 298–303. 17. Thomford JW, Conrad PA, Telford III SR, Mathiesen D, Bowman BH, Spielman A, Eberhard ML, Herwaldt BL Quick RE, Persing DH, 1994. Cultivation and characterization of a newly recognized human pathogenic protozoa. J Infect Dis 169: 1050–1056. 18. Herwaldt BL, Persing DH, Precigout EA, Goff WL, Mathiesen DA, Taylor PW, Eberhard ML, Gorenflot AF, 1996. A fatal case of babesiosis in Missouri: identification of another piroplasm that infects humans. Ann Intern Med 124: 643–650.
