| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The agent of human granulocytic ehrlichiosis (HGE), Anaplasma phagocytophilum, appears to be maintained in the same enzootic cycle as that for the agents of Lyme disease and human babesiosis (Babesia microti). Mice are frequently infected, are competent reservoirs, and serve as important hosts for subadult deer ticks.5 However, A. phagocytophilum seems to have as great a vertebrate host range as does B. burgdorferi, infecting mice, deer, sheep, raccoons, skunks, and cattle.6–10 Accordingly, it may also be perpetuated in parallel cycles. Similar to Lyme disease spirochetes, such additional cycles would buffer the agent of HGE against extinction during the years of low mouse density, and may help explain the wide distribution, including Europe and the southern and western United States, of enzootic HGE.11–15 In contrast, B. microti appears to mainly infect small rodents and remains limited as a zoonosis to certain coastal New England and upper Midwestern sites and has not been found in I. scapularis from the south or I. pacificus from California.13,16
It may be that cottontail rabbits help maintain the agent of HGE in a parallel cycle as they do that of Lyme disease. Rabbits are among the most common animals on Nantucket Island, along with mice and deer. Introduced to coastal New England in the early 1900s, the eastern cottontail, Sylvilagus floridanus, has proliferated and displaced the native New England cottontail S. transitionalis.17 Rabbits have great reservoir capacity inasmuch as they are abundant in peridomestic settings, long-lived, and are infested by ticks with great host-specificity. All three stages of I. dentatus, as well as those for Haemaphysalis leporispalustris, feed mainly on rabbits. In addition, infections may be easily bridged to humans or other animals because subadult I. dammini may also feed on rabbits. We described the reservoir capacity of cottontail rabbits for the agent of HGE on Nantucket Island by determining the prevalence of infection in these hosts and the ticks that infest them. In addition, we evaluated the competence of rabbits and rabbit-feeding ticks as hosts for the agent of HGE.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
PCR amplification. DNA was extracted from whole blood and spleens using the Isoquick kit (Orca Research Inc., Bothell, WA) or the DNEasy Tissue kit (Qiagen, Inc., Valencia, CA) following the instructions of the manufacturer. Ticks were crushed individually in 50 µL of phosphate-buffered saline within wells of a 96-well plate, and a 10-µL aliquot of each tick homogenate was pooled with five others. The homogenate pools were extracted using a standard phenol/ chloroform procedure. The PCR was performed using Taq polymerase and buffer reagents from Qiagen as recommended. A nested reaction was used to identify infected rabbits as described previously.18 Ten microliters of product were analyzed by electrophoresis on a 1.5% agarose gel that included a 100-basepair DNA ladder (Gibco-BRL, Gaithersburg, MD). Some of the positive samples were purified using Qiagen spin-columns and sent to the University of Maine (Orono, ME) sequencing facility for sequencing. Because we are always aware of the possibility of PCR contamination, we used appropriate positive and negative template controls and barrier pipette tips, and physically separated each step. Rabbits were considered infected if either blood or spleen yielded amplification products of the expected molecular mass. Confidence intervals (CIs) for prevalence estimates were calculated using the exact binomial method (Stata software package; Stata Corp., College Station, TX).
Serology. Sera were screened by the indirect fluorescent antibody test (IFAT) as described previously.19 Sera were screened at an initial dilution of 1:64. The NCH-1 strain of A. phagocytophilum cultivated in HL60 cells was used as the antigen. Fluorescein-labeled anti-rabbit IgG (gamma chain-specific: Sigma, St. Louis, MO) at a dilution of 1:256 was used as the secondary antibody. Slides were examined under epi-fluorescence at a magnification of x 600. A sample was scored as positive if 3+ fluorescence was observed at a dilution of 1:64 or greater.
Culture. We attempted in vitro cultivation for some rabbit blood samples that were PCR positive. Whole citrate-anticoagulated blood (100 µL) was added to 6-mL polystyrene culture tubes containing 2.5 mL of RPMI 1640 medium (supplemented with 1% L-glutamine and 20% fetal bovine serum) and 106 HL60 cells (CCL240; American Type Culture Collection, Manassas, VA) in growth phase. Cultures were kept loosely capped in a humidified 37°C incubator in an atmosphere of 5% CO2, and fresh medium was added every three days. With each medium change, Cytospin (Thermo-Shandon, Pittsburgh, PA) preparations were stained with Giemsa and examined for typical ehrlichial morulae.
Ticks. Fully engorged ticks removed from rabbits were placed in mesh-capped, three-dram vials containing moist, hardened plaster of paris, kept at a humidity of 95% and at 18°C, and allowed to molt. They were subsequently tested by the PCR as described earlier for evidence of infection. Partially engorged ticks were dissected and their salivary glands were stained by the Feulgen reaction as described;5 remnants of dissected ticks were also tested for infection by the PCR as described earlier. Ticks removed from rabbits whose blood or spleen was positive by the PCR were excluded from the prevalence estimates.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Incidence and seroprevalence in rabbits.
Transmission of the agent of HGE is intense in cottontail rabbits on Nantucket Island. Each year, except for 1998, at least 20% of our rabbits were demonstrated to have active infections as suggested by the presence of amplifiable DNA (Table 1
). In 1998, rabbits were sampled only in May and September, but the wide CI for our estimated prevalence would imply that 1998 did not differ from other years with respect to transmission. The rate of HGE infection, as assessed by PCR, for the one February collection was not significantly different than what we detected the rest of the year (12.5%, 95% CI = 0.3–52). The majority of the rabbits tested each year were seropositive by the IFAT, indicating that they had been exposed to infection. To determine whether the force of HGE transmission differed on a seasonal basis, incidence (as measured by PCR) was stratified by month (Figure 1). Rabbits generally appeared be actively infected regardless of month.
|
|
). The estimated minimum infection rates (MIRs) for all species range from 2% to 6% with overlapping CIs. These estimates are well within the range that we expect to see in host-seeking I. dammini in this field site. Interestingly, the known vector of this agent, I. dammini, was less frequently infected than were the other two ticks.
|
). Although infection rates appear to differ between the two methods, these differences are not statistically significant. We conclude that HGE is capable of infecting the salivary glands of I. dentatus and that our MIRs of ticks tested by the PCR are not grossly inflated due to contaminating blood meal.
|
). In contrast, none of the H. leporispalustris remained infected after the molt. A similar proportion of I. dentatus (5%) become infected regardless of the PCR status of the rabbit that provided the blood meal. However, none of the I. dammini tested from PCR-negative rabbits became infected, suggesting that I. dentatus may be a more competent vector.
|
). Most rabbits that were PCR positive were also seropositive. Furthermore, a robust humoral response, as defined by a high serum antibody titer, did not appear to diminish the probability that infection would be detected by the PCR.
|
). None of the H. leporispalustris was positive. We conclude that both I. dammini and I. dentatus may maintain HGE, whereas H. leporispalustris may not.
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Transmission over the summer months is consistently high (Figure 1), suggesting that either the rabbits maintain infection for a long period of time or they are constantly reinfected throughout the summer months.
Reservoir competence may be affected by host immunity.20 Seroconversion of a host often signals the onset of immunity, clearance of organisms from the blood, and the loss of the ability to infect ticks. White-footed mice have been characterized as poorly competent reservoirs of HGE because they seroconvert and seem to lose their infection within a few weeks.10 We examined this issue with rabbits, specifically asking the question whether rabbits that were seropositive could be actively rickettsemic, i.e., whether rabbits positive by the IFAT were also positive by PCR. Every PCR-positive rabbit (except one) was seropositive, and increasing IFAT titer did not influence the proportion of PCR-positive animals (Figure 2). We conclude that a robust antibody response does not necessarily reduce rabbit reservoir competence, assuming that the presence of amplifiable DNA within peripheral blood represents the availability of viable organisms to feeding ticks.
We identified three different ticks feeding on rabbits: I. dammini, I. dentatus, and H. leporispalustris, as previously reported.21 Ixodes dammini is the main vector of HGE on Nantucket,5 maintaining the cycle among white-footed mice as well as transmitting infection to people. Although vector competence is classically demonstrated under laboratory conditions, the sum of our observations allow us to conclude that I. dentatus is a competent vector for HGE, but that H. leporispalustris is not. We were able to identify HGE in the salivary glands of female I. dentatus, implying the capacity to transmit during feeding. The agent of HGE was efficiently maintained through the molt of larvae to nymphs, demonstrating transstadial transmission. Approximately 5% of I. dentatus and I. dammini nymphs collected from PCR-positive rabbits remained infected after the molt (Table 4), but none of the H. leporispalustris did so. This suggests that although Haemaphysalis were exposed to infectious organisms, they seem to be incompetent vectors for HGE.
Because ehrlichiae do not seem to be maintained by transovarial transmission,22,23 analysis of nymphs derived from larvae engorging on field-collected rabbits provides a xenodiagnosis, which is the definitive assay for reservoir competence.24 A similar proportion of I. dentatus (5%) became infected regardless of the PCR status of the rabbit that provided the blood meal, suggesting that xenodiagnosis is more sensitive than the PCR. Xenodiagnosis, of course, is well documented as a sensitive method for detecting Lyme disease spirochetes.25,26 It may be that ehrlichiae concentrate within neutrophils infiltrating the site of tick feeding, and that peripheral blood neutrophils are less likely to be infected. Interestingly, xenodiagnosis on rabbits demonstrated that I. dammini is less competent than is I. dentatus. Perhaps rabbits serve as relatively poor hosts for deer ticks, thereby providing diminished blood meal quality that may affect competence. Alternatively, rabbit-infecting ehrlichiae may be specifically adapted to rabbits and rabbit ticks. Recently, unique 16S rDNA sequence variants from an agent of HGE that do not appear to be associated with human disease have been described from Rhode Island.27 The limited number of amplification products that we sequenced, however, demonstrates that the rabbit-infecting agent does not differ from that of NCH-1, the index case of HGE in the northeastern United States. Furthermore, blood from field-collected rabbits efficiently infects laboratory mice (a sensitive model for HGE infection5) with the agent of HGE, producing infections that do not appear to differ morphologically or behaviorally from those initiated from white-footed mouse or infected human blood.
Whether this rabbit-maintained enzootic cycle of HGE contributes to human risk is not known. Often, multiple cryptic cycles exist in nature that help buffer the organism from local extinction, but may not directly contribute to zoonotic risk. Subadult I. dammini may serve as the bridge between the rabbit cycle and the mouse cycle. Indeed, more I. dammini are found on rabbits in years in which the local mouse population is at a cyclic nadir (Goethert HK, unpublished data). Accordingly, during such times interventions focused solely on mice would be relatively ineffective.
The rabbit-feeding I. dentatus is not thought to frequently bite humans, although a recent study of Lyme disease risk in a Maryland community found that one in four Ixodes that were submitted for identification by tick-bitten residents were I. dentatus.28 Perhaps I. dentatus attack people more often than is assumed. To quantify the risk associated with I. dentatus, we determined entomologic inoculation rates (EIR) for I. dammini as well as I. dentatus within our study site. Infected I. dammini were five times more prevalent than I. dentatus (Table 5). This was unexpected because our data suggested that I. dentatus seem to become infected more readily than I. dammini (Table 4). Perhaps HGE infection is less stable within I. dentatus than it is in I. dammini. Alternatively, our estimates may be a result of the sampling method. Because I. dentatus are much more patchily distributed than I. dammini, it may be that our low EIR estimate for I. dentatus derives from the collection of patches of negative ticks averaged with patches of positive ticks. Regardless, the presence of I. dentatus may have implications for risk because their hosts tend to accumulate around peoples’ lawns, and the ticks themselves are readily found on lawns, unlike I. dammini.
Finally, the competence of I. dentatus for the agent of HGE has implications for the potential distribution of this pathogen. Immature I. dentatus are known to feed on birds. As with Lyme disease spirochetes,29 the agent of HGE could easily be transported to start new foci of disease. Cottontail rabbits are common throughout much of the eastern and central United States, and therefore the probability that a founder infected tick will find a suitable host seems great. In addition, HGE risk may be apparent in areas where I. dentatus is present but I. dammini is not.
Received November 12, 2002. Accepted for publication March 3, 2003.
Acknowledgment: This is a contribution of the University of Massachusetts Nantucket Field Station.
Financial support: Our work was supported by grants from the National Institutes of Health (AI-39002) and the Centers for Disease Control and Prevention (U50/CCU119560). Heidi K. Goethert was supported, in part, by National Institute of Health NIH training grant AI-07535 "Epidemiology of Infectious Diseases."
Ehrlichia [Cytoecetes] phagocytophilum has recently been reclassified within the genus Anaplasma.4 Although the basis for this decision remains controversial, the Bacteriological Code [unlike the International Code of Zoological Nomenclature] mandates the use of the revised names that have been accepted by the International Society for Systematic Bacteriology.
Authors’ address: Heidi K. Goethert and Sam R. Telford III, Division of Infectious Diseases, Department of Biomedical Sciences, Tufts University School of Veterinary Medicine, 200 Westboro Road, North Grafton, MA 01536.
Reprint requests: Sam R. Telford III, Division of Infectious Diseases, Department of Biomedical Sciences, Tufts University School of Veterinary Medicine, 200 Westboro Road, North Grafton MA 01536, Telephone: 508-887-4236, Fax: 508-839-7911, E-mail: sam.telford{at}tufts.edu.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  U. Blas-Machado, J. de la Fuente, E. F. Blouin, C. Almazan, K. M. Kocan, and J. V. Mysore Experimental Infection of C3H/HeJ Mice with the NY18 Isolate of Anaplasma phagocytophilum Vet. Pathol., January 1, 2007; 44(1): 64 - 73. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.-C. CAO, L. ZHAN, J. HE, J. E. FOLEY, S. J. DE VLAS, X.-M. WU, H. YANG, JAN. H. RICHARDUS, and J. D. F. HABBEMA NATURAL ANAPLASMA PHAGOCYTOPHILUM INFECTION OF TICKS AND RODENTS FROM A FOREST AREA OF JILIN PROVINCE, CHINA. Am J Trop Med Hyg, October 1, 2006; 75(4): 664 - 668. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Tate, D. G. Mead, M. P. Luttrell, E. W. Howerth, V. G. Dugan, U. G. Munderloh, and W. R. Davidson Experimental Infection of White-Tailed Deer with Anaplasma phagocytophilum, Etiologic Agent of Human Granulocytic Anaplasmosis J. Clin. Microbiol., August 1, 2005; 43(8): 3595 - 3601. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. K. GOETHERT and S. R. TELFORD III ENZOOTIC TRANSMISSION OF BABESIA DIVERGENS AMONG COTTONTAIL RABBITS ON NANTUCKET ISLAND, MASSACHUSETTS Am J Trop Med Hyg, November 1, 2003; 69(5): 455 - 460. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
