You have requested a machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Neither BioOne nor the owners and publishers of the content make, and they explicitly disclaim, any express or implied representations or warranties of any kind, including, without limitation, representations and warranties as to the functionality of the translation feature or the accuracy or completeness of the translations.
Translations are not retained in our system. Your use of this feature and the translations is subject to all use restrictions contained in the Terms and Conditions of Use of the BioOne website.
1 December 2013
Ixodid Ticks Associated with Feral Swine in Texas
David M. Sanders, 1 Anthony L. Schuster, 2 P. Wesley McCardle, 2 Otto F. Strey, 3 Terry L. Blankenship, 4 Pete D. Teel 3
1 U.S. Air Force Research Laboratory/USAFSAM, Wright-Patterson AFB, OH 45433-7408, U.S.A. 2 U.S. Army Medical Component, Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand 3 Department of Entomology, Texas A&M AgriLife Research, Texas A&M University, College Station, TX 77843-2475, U.S.A., [email protected] 4 Welder Wildlife Foundation, P.O. Box 1400, Sinton, TX 78387-1400, U.S.A.
SAVE TO MY LIBRARY
PURCHASE THIS CONTENT
PURCHASE SINGLE ARTICLE
Includes PDF & HTML, when available
Ixodid ticks were collected from feral swine in eight Texas ecoregions from 2008–2011. Sixty-two percent of 806 feral swine were infested with one or more of the following species: Amblyomma americanum, A. cajennense, A. maculatum, Dermacentor albipictus, D. halli, D. variabilis, and Ixodes scapularis. Juvenile and adult feral swine of both sexes were found to serve as host to ixodid ticks. Longitudinal surveys of feral swine at four geographic locations show persistent year-round tick infestations of all gender-age classes for tick species common to their respective geographic locations and ecoregions. Amblyomma americanum, A. cajennense, A. maculatum and D. variabilis were collected from 66% of feral swine harvested through an abatement program in seven ecoregions from March to October in 2009. These results indicate westward geographic expansion of D. variabilis. Summary results show feral swine are competent hosts for ixodid species responsible for the transmission of pathogens and diminished well-being in livestock, wildlife, and humans.
Received: 14 February 2013; Accepted: 1 September 2013; Published: 1 December 2013
PURCHASE THIS CONTENT
PURCHASE SINGLE ARTICLE
This article is only available to subscribers. It is not available for individual sale.
Experimental Infection of Ponies with Borrelia burgdorferi by Exposure to Ixodid Ticks
Dr. Y.-F. Chang, Department of Population Medicine and Diagnostic Science, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853 (USA). E-mail: [email protected] .
Seven specific-pathogen-free (SPF) ponies, 1–5 years old, were exposed to Borrelia burgdorferi–infected adult ticks while being treated with dexamethasone over 5 consecutive days. One SPF pony (pony No. 178) was first exposed to laboratory-reared nymphs without B. burgdorferi infection and 3 weeks later was exposed to B. burgdorferi–infected adult ticks with concurrent dexamethasone treatment for 5 consecutive days. Four uninfected ponies treated with dexamethasone, exposed to laboratory-reared ticks without B. burgdorferi infection served as uninfected controls. Clinical signs, bacteriologic culture, polymerase chain reaction (PCR) for bacterial DNA, immunologic responses, and gross lesions and histopathologic changes were investigated during the experiment or at necropsy 9 months after tick exposure. In all of the seven challenged ponies, infection with B. burgdorferi was detected from monthly skin biopsies and various tissues at postmortem examination by culture and by PCR. However, pony No. 178 exposed to laboratory-reared nymphs (without B. burgdorferi infection) and challenged with B. burgdorferi–infected adult ticks 2 months later did not develop a B. burgdorferi infection. All of the infected ponies seroconverted. Control ponies and pony No. 178 were negative by culture, PCR, and serology. Except for skin lesions, we failed to induce any significant histopathologic changes in this study. This is the first report of successful tick-induced experimental infection in ponies by exposure to B. burgdorferi–infected ticks. This Lyme disease model will be very useful to evaluate efficacy of vaccines against the Lyme agent and the effect of antibiotic therapy on horses infected with B. burgdorferi.
Lyme disease (LD) is the most important arthropod-borne bacterial infection in the United States. Affecting humans, dogs, horses, cattle, and cats, LD is caused by the spirochete Borrelia burgdorferi transmitted primarily by ticks of the genus Ixodes.11,16,22,24–26,30 The incidence of equine Borrelia infection seems to be increasing in the northeastern United States, the Midwest, Texas, and California.30
A broad spectrum of clinical manifestations has been attributed to equine LD including chronic weight loss, sporadic lameness, hepatitis, laminitis, low-grade fever, swollen joints, muscle tenderness, and anterior uveitis.6–8,10–12,23 Neurologic signs, including depression, behavioral changes, dysphagia, head tilt, facial paralysis, and encephalitis, can also occur.6–8,10–12,23 Infection of pregnant mares has been suggested to cause fetal resorption, abortion, or the birth of weak foals that die within the first few days of life. Foals that survive may develop neurologic signs when they get older. In contrast, clinical signs in seropositive mares are limited to stiffness and lameness within 3 weeks of foaling.9 Serologic surveys show that 12–75% of clinically normal horses in New England are positive for B. burgdorferi by indirect immunofluorescence antibody (IFA) testing.23,27 The rising incidence and geographic spread of equine LD have raised concerns among horse owners and equine practitioners. Because of the variable clinical signs and the lack of experimental data, the significance of serologic results in horses is difficult to assess. The purpose of the present study was to develop a protocol for reliably inducing B. burgdorferi infection in horses in order to evaluate responses to chemotherapeutic agents, protection by vaccination, and to aid in interpretation of serologic results obtained by enzyme-linked immunosorbent assay (ELISA) and western blotting.
Materials and Methods
Twelve specific-pathogen-free (SPF) ponies, 1–5 years old (Table 1), from Cornell University, College of Veterinary Medicine, were kept in P2 isolation units, fed a commercial ration, and provided water ad libitum. The protocol of this study was approved by the Institutional Animal Care and Use Committee at Cornell University to comply with federal law (PL99-198). All work was conducted in compliance with regulations, policies, and principles of the Animal Welfare Act, the Public Health Service for Policy on Humane Care and Use of Laboratory Animals used in Testing, Research, and Training, the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and the New York State Department of Public Health regulations. All ponies were observed for clinical signs and their body temperatures were recorded daily. Body weights were measured weekly.
Table 1 Experimental design of equine Lyme infection model.
Table 1 Experimental design of equine Lyme infection model.
Adult ticks (Ixodes scapularis) infected with B. burgdorferi were collected by flagging in a forested area of Westchester County, New York. Ticks were maintained at the Cornell Entomology Laboratory at 94% relative humidity and 10 C for 2 months. To determine the percentage of ticks infected with B. burgdorferi, 20 male or female ticks were ground and cultured individually in BSK-2 medium with 8 mg/ml kanamycin and 50 mg/ml rifampicin as previously described.4,13,15,21,35 The cultures, examined weekly over a 6-week period for B. burgdorferi by dark-field microscopy and IFA testing showed a 55–60% infection rate. Twenty laboratory-reared adult ticks and nymphs, evaluated in the same manner, were negative for B. burgdorferi.
Exposure of ponies to ticks
Ponies were exposed to 20 female and 10 male field-collected adult ticks (I. scapularis) by placing the ticks onto the clipped side of each pony as previously reported.13,15 The four control ponies were exposed to laboratory-reared uninfected ticks in a similar manner. Pony No. 178 was first exposed to uninfected nymphs, and was reexposed to infected adult ticks 2 months later. Dexamethasone (0.2 mg/lb/day [0.44 mg/kg], Schering-Plough Animal Health, Kenilworth, NJ) was given intramuscularly for 5 consecutive days starting on the first day of adult tick exposure and again for 5 days starting at 5 months post–tick exposure. Ticks were allowed to feed and engorge for 7 days, when at least 50% of the female ticks were fully engorged; at this time they were removed from the ponies.
Serum and tissue samples
A serum sample was obtained from each pony at the time of tick exposure and then at 2-week intervals for 9 months. Sera were tested by western blotting and kinetic ELISA (KELA). After tick exposure, skin biopsies were taken from the site of tick exposure at monthly intervals for isolation of spirochetes. Nine months after challenge, all ponies were euthanized and tissues were removed for culture and polymerase chain reaction (PCR) analysis for B. burgdorferi, and for histopathology.
Isolation of B. burgdorferi
To test for infection, attempts were made to isolate B. burgdorferi from skin biopsies (monthly) at the site of tick bite and from various tissues at postmortem examination. Samples from skin punch biopsies (4 mm) collected at monthly intervals after tick exposure and pieces of tissue (approximately 0.2–1 g) obtained aseptically at necropsy (Table 1) were homogenized in 5 ml of BSKII medium in a tissue homogenizer (stomacher, Tekmar, Cincinnati, OH) and then transferred to 25 ml of prewarmed BSKII medium. The cultures held at 34 C were checked weekly for up to 6 weeks for the presence of B. burgdorferi by dark-field examination and IFA testing.
Serology: KELA and immunoblots
KELA for measuring the relative quantity of serum antibody to B. burgdorferi was performed as described previously.4,13,15,21 Briefly, diluted serum was added to duplicate wells in microtiter plates containing antigens of French-pressed B. burgdorferi lysate. Bound antibody was detected by using horseradish peroxidase (HRP) conjugated to goat anti-horse immunoglobulin G (IgG) (Cappel Research Products, Durham, NC). Color development using the chromogen tetramethylbenzidine with H2O2 as a substrate was measured kinetically and expressed as the slope of the reaction rate between enzyme and substrate solution. Each unit of slope was designated as a KELA unit.4,13,15,21
Western blot analysis was performed as previously described.13,15,21 Briefly, French-pressed B. burgdorferi lysate was used as an antigen and subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Western blot analysis was performed in a miniblotter. Test sera from experimental animals were used as the primary antibody, followed by goat anti-horse IgG conjugated to HRP as a second antibody. Bands were developed by using the substrate solution (4-chloro-1-naphthol, 24 μg in 8 ml of methyl alcohol, 40 ml Tris-buffer solution [pH 7.5], and 24 μl of 30% H2O2).
DNA from biopsy samples (skin) or from postmortem tissues (Table 1) was extracted by standard procedures.4,15 Also, 25 tissues including synovial membranes, lymph nodes, muscles, peritoneum, pericardium, and skin collected from uninfected ponies were used as negative controls. The DNA from B. burgdorferi was isolated and PCR was performed as described4,15 using the SL primer set for the ospA gene (sense, SL1: 5′-AATAGGTCTAATAATAGCCTTAATAGC-3′; antisense, SL2: 5′-CTAGTGTTTTGCCATCTTCTTTGAAAA-3′; probe, SL3: 5′-GGCAAGTACGATCTAATTGCAACAGT-3′),17 and the primers from 23S rRNA gene (sense, JS1: 5′-AGAAGTGCTGGAGTCGA-3′; antisense, JS2: 5′-TAGTGCTCTACCTCTATTAA-3′; probe, FS1: 5′-AGTCTGTTTAAAAAGGCA-3′).34 The primers were synthesized using an Applied Biosystems 380A DNA Synthesizer (Foster City, CA) at the Analytical and Synthetic Facility, Cornell University.
To prevent contamination, the preparation of reaction mixtures, DNA extraction, amplification, and detection of PCR products were all performed in different rooms. Also, aerosol-resistant filter pipette tips were used throughout the experiment. Amplification of B. burgdorferiospA and 23S rRNA-specific target sequences was carried out in a 50-μl reaction mixture of 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl2, 0.5% NP4O, 0.5% Tween 20, 200 mM each of deoxynucleoside triphosphates, 2 mM of primer sets (SL, or 23 S rRNA) and 2 U of the thermostable Taq DNA polymerase (Perkin-Elmer Cetus, Foster City, CA) containing 100 ng of DNA from the specimens listed in Table 1. B. burgdorferi genomic DNA (1 ng) was used as a positive control and distilled water was used as a negative control, pipetted and handled identically to the samples. The reaction mixture was subjected to 40 cycles of amplification by using an automated DNA thermal cycler (Perkin-Elmer Cetus 9600). Each cycle involved heating to 94 C for 1 minute (DNA denaturation), cooling to 39 C for 1 minute (primer annealing), and again heating to 72 C for 2 minutes (primer extension). Negative controls, which consisted of distilled water substituted for the DNA template in the reaction mixture, were included in each PCR run.
Visualization of the PCR amplification products was performed by gel electrophoresis on a 1.5% agarose gel with pBH2O-HinfI cut as a size marker. For Southern blot analysis, the PCR amplification product was run on a 1.5% agarose gel, stained by ethidium bromide, denatured (1.5 M NaCl, 0.5 M NaOH) for 1 hour, neutralized (1 M Tris-HCl pH 8.0, 1.5 M NaCl), and transferred to a nitrocellulose membrane as previously described.15 The oligonucleotide probes (SL3 and FS1, respectively) were 3′-labeled with a nonradioactive labeling kit (ECL 3′-oligolabelling system, Amersham, Little Chalfont, Buckinghamshire, UK) as previously described.15 Southern blot hybridization and detection were performed as described by the manufacturer (ECL 3′-oligolabelling system, Amersham).15
All ponies were euthanized approximately 9 months after tick exposure and necropsied (Table 1). The following tissues were fixed in 10% neutral buffered formalin: joint capsules (right and left elbow, shoulder, stifle, carpus, tarsus, fetlock), cerebellum, cerebrum, meninges, spinal cord, myocardium, urinary bladder, thyroid, liver, spleen, kidney, lung, stomach, intestine, skeletal muscles, aorta, eyes, nerves (left and right brachial plexus, trigeminal ganglion, cervical and thoracic nerve root, median, ulnar, radial, tibial, fibular, sciatic, and facial), and lymph nodes (axillary, prescapular, and popliteal). Tissues were embedded in paraffin wax, sectioned, and stained with hematoxylin and eosin by conventional methods for histopathologic evaluation.
No significant or obvious clinical signs were detected in either the control ponies or ponies exposed to B. burgdorferi–infected ticks.
Isolation of B. burgdorferi from skin biopsy
Two or more of the monthly skin biopsies from the site of tick exposure were positive in culture for B. burgdorferi for every pony, except for pony No. 178 and those in the control group (Table 1).
Isolation of B. burgdorferi from postmortem tissues
Various tissues collected from seven ponies at the time of postmortem examination were positive in culture. B. burgdorferi was most frequently isolated from skin, fascia, and muscle (Table 1) and sporadically isolated from the joint capsules, skin, and lymph nodes (Tables 1, 2).
Table 2 Culture and polymerase chain reaction (PCR) of B. burgdorferi from tissues of 12 ponies 9 months after tick exposure.∗
Table 2 Culture and polymerase chain reaction (PCR) of B. burgdorferi from tissues of 12 ponies 9 months after tick exposure.∗
Tissues taken from ponies exposed to infected ticks and those exposed to uninfected ticks were subjected to PCR. All tissues from the control ponies and pony No. 178 were negative, whereas all other challenged ponies were PCR positive in many tissues. For these seven ponies, the total number of culture-positive ponies and those positive by PCR-1 (SL primers) and PCR-2 (23S rRNA primers), were 45, 156, and 132, respectively (Table 2, Fig. 1 ). Thus, PCR was more sensitive for detecting B. burgdorferi DNA than the culture technique for viable organisms.
Fig. 1. An agarose gel showing representative polymerase chain reaction (PCR) results using tissues from an infected pony (pony No. 172). A1 (using SL primers) and B1 (using 23S rRNA primers) and B2 (hybridization). A2 and B2, Southern blot hybridization of representative PCR products from A1 and B2, respectively. Template DNA (100 ng) was extracted from muscles (lanes 2, 3), lymph node (lane 4), skin (lane 5), kidney capsule (lane 6), urinary bladder (lane 7), meninges (lane 8), joint capsules (lanes 9–15), fascias (lanes 16, 17), myocardium (lane 18), and pericardium (lane 19). Positive control (lane 20), 1 ng of Borrelia burgdorferi DNA was used as a template. Negative control (lane 1); instead of a DNA template, distilled water was used.
KELA and western blotting
All ponies exposed to B. burgdorferi–infected ticks developed detectable antibodies 5–6 weeks after exposure. The antibody titers steadily increased for 3–4 months to 200–300 KELA units and remained at this level until necropsy ( Fig. 2 ). The four control ponies and pony No. 178 remained KELA negative throughout the experiment.
Fig. 2. Antibody levels of ponies exposed at day 0 to Borrelia burgdorferi–infected adult ticks (Ixodes scapularis) as determined by kinetic enzyme-linked immunosorbent assay (KELA). The line at 100 KELA units represents the cutoff between positive and negative sera. Open symbols indicate that ponies were not exposed to B. burgdorferi–infected ticks (control group). Pony No. 142 (∗——∗) had an intermediate KELA titer before exposure to B. burgdorferi–infected ticks. Western blot analysis indicated that this pony had a high level of nonspecific flagellar antibodies (41 kd, not shown). Pony No. 178 (•——•) was exposed to 25 laboratory-reared nymphs without B. burgdorferi infection at day 0, and then exposed to B. burgdorferi–infected ticks at day 49.
Western blot analysis of sera from ponies exposed to B. burgdorferi–infected ticks all showed bands in molecular weight regions of p83, p65, p60, p41, and p39 ( Fig. 3 ). These bands are diagnostic for B. burgdorferi in horses (Chang, unpublished data). Many other bands were visible but were not always present nor were they always specific. These sera showed no antibody to OspA ( Fig. 3 ). Western blots on the four control animals and pony No. 178 were negative.
Fig. 3. Western blot of sera from a tick-infected pony (pony No. 179). Lanes 1–13, immune response against Borrelia burgdorferi at the time of challenge (lane 1) and 15, 22, 29, 35, 43, 57, 84, 99, 128, 156, 193, and 225 days after tick exposure (lanes 2–13). Specific bands for OspA and OspB are missing from the 32- and 34-kd regions, although monoclonal antibodies to OspA and OspB detect the antigens (not shown). The dominant bands were 83, 65, 60, 41, and 39 kd. The biotinylated sodium dodecyl sulfate–polyacrylamide gel electrophoresis standard broad-range molecular weight markers (Bio-Rad Laboratories, Richmond, CA) were used. The numbers at the right indicate molecular weights.
Gross pathology and histopathology
Lesions were restricted to the skin and peripheral lymph nodes of ponies exposed to B. burgdorferi–infected ticks. Changes consisted of lymphohistiocytic nodules up to 2 mm in diameter scattered about the middle and deep dermis ( Figs. 4 , 5 ). Occasionally, a chain of large lymphoid nodules with prominent germinal centers and thin mantle zones (pseudolymphoatous reaction) was present in the deep dermis. In one infected pony (pony No. 180) a single large nodule in the middle dermis had several central multinucleated giant cells and moderate numbers of peripheral eosinophils mixed with rare neutrophils. No tick mouth parts (chelicerae or hypostomes) were noted in any skin sections examined.
Fig. 4. Skin; pony No. 142. Perivascular and perineural lymphohistiocytic aggregates in the superficial and deep dermis near the attachment sites of Borrelia burgdorferi–infected ticks. Note the row of secondary follicles in the deep dermis. HE. Bar = 800 μm.
Fig. 5. Skin, pony No. 142. Secondary follicles in the deep dermis. HE. Bar = 160 μm.
Changes in peripheral lymph nodes were confined primarily to the prescapular lymph nodes. The prescapular lymph nodes of control ponies had follicles that were evenly spaced along the outer rim of the cortex and averaged approximately 200 μm in diameter. About 25% of the follicles had round germinal centers and moderately wide mantle zones of uniform width. The follicles were separated by paracortex composed of loosely packed small lymphocytes and a smattering of eosinophils. Medullary cords were about five to eight cells wide and populated by a mixture of plasma cells, lymphocytes, and histocytes. The subcapsular and medullary sinuses had small numbers of mature lymphocytes and occasional histocytes. In contrast, prescapular lymph nodes from B. burgdorferi–infected ponies had marked lymphoid hyperplasia with left and right sides of the horses affected equally. The paracortical zones were encroached upon by numerous large follicles (750–1,000 μm in diameter) with large, often misshapened germinal centers surrounded by thin mantle zones. The germinal centers were often polarized with a basally positioned dark zone and an apical light zone.
The primary aim of this study was to establish clinical and pathologic LD in the horse to facilitate evaluation of vaccines and antimicrobial therapy. We successfully induced infection in these ponies by exposure to B. burgdorferi–infected ticks with concurrent dexamethasone treatment. Our equine LD model can be used to evaluate immunologic responses and Lyme vaccines (such as bacterin, OspA-enriched, recombinant subunit, or DNA vaccine) for protection against infection and disease or for therapeutic activity. The model can also be used to investigate the interaction of B. burgdorferi with other tick-borne agents, such as human granulocytic ehrlichiosis, Babesia microti, and/or other agents.
This is the first evidence of induced infection in ponies by exposure to adult ticks infected with B. burgdorferi. Although these ponies were concurrently given dexamethasone at the time of infection and again 5 months later, no significant clinical signs were observed. This differs markedly from the adult dog model of LD where dexamethasone given concurrently with infection routinely elicits clinical signs.14 Dexamethasone is primarily used as an anti-inflammatory agent in various human and animal diseases.2,3 The precise mechanisms of glucocorticoid action are not completely understood. Monocytes and macrophages possess glucocorticoid receptors41 and their function can be altered in response of glucocorticoid treatment.33,37 Dexamethasone may enhance mRNA and protein expression of the calcium- and phospholipid-binding proteins (lipocortins), which partially mediate an antiproliferative effect on lymphocytes,2,3 inhibit the release of arachidonic acid (precursor of leukotriene B4) and prostaglandins,19 inhibit cytokine production via blockade of cytokine gene transcription,2,3 and/or decrease the stability of cytokine mRNA.18 We injected dexamethasome intramuscularly for 5 consecutive days starting at day 1 of tick exposure. Based on previous studies14,40 in an adult canine LD model, clinical lameness can be induced in B. burgdorferi–infected dogs that had not previously shown clinical signs by injecting dexamethasome intramuscularly 150 days after tick exposure (0.4 mg/lb [0.88 mg/kg]).14 Skin biopsy of these dogs confirmed they were positive for B. burgdorferi before dexamethasome treatment. Whether the second dose of dexamethasome (0.2 mg/lb [0.44 mg/kg]) given to these ponies had the effect of preventing lesions and clinical signs is unknown. Dexamethasone does not seem to exacerbate infections in mice as it does in dogs. The horse model may more closely resemble the mouse rather than the dog model of LD.20 In our previous studies, we found that asymptomatic adult dogs with persistent Lyme infections developed clinical lameness after dexamethasone treatment (Y.-F. Chang, V. Novosel, B. Summers, unpublished data). Dogs with persistent Lyme infection that were treated with prednisone were recently reported to develop clinical lameness.38 The role of immunosuppression in the development of clinical disease warrants further study.
The skin lesions found in the infected ponies are similar to erythema chronicum migrans, the characteristic skin rash associated with human LD.29 Controversy remains over how B. burgdorferi induces clinicopathologic changes, but tissue damage is likely secondary to inflammation directed at persistent borrelial antigen.36 We have not yet addressed the issue of whether B. burgdorferi organisms are present within the foci of inflammation in infected ponies. Although some researchers report that silver stains are adequate to identify borrelial spirochetes in tissue sections,5 silver stains are nonspecific and difficult to interpret. B. burgdorferi may assume a variety of morphologic forms in vivo and the plane of section rarely coincides with the long axis of the organism.1 Thus, determination of whether a small irregular stained structure truly represents a B. burgdorferi organism becomes virtually impossible. Studies utilizing in situ hybridization to study the distribution of B. burgdorferi organisms are planned.
The sensitivity of PCR is greater than that of cultures. For example, although a total of 41 samples from seven ponies were positive by culture, 134 samples were positive by PCR using primers derived from the ospA gene, and 120 samples were positive by PCR using primers derived from 23S rRNA (Table 2). Culture can only detect viable spirochetes, whereas PCR can detect both live and dead organisms, and both intact and fragmented spirochete DNA. PCR testing could improve the sensitivity of LD diagnoses in equine patients. A difference was encountered by using primers specific for ospA or for 23S rRNA. This might be due to the overrepresentation of this B. burgdorferi plasmid DNA sequence in the clinical samples, which has been referred to as target imbalance.31 In a previous study, we used primers derived from the flagellin gene and found that the results of culture and PCR are comparable.15 However, this study and others indicate that a higher sensitivity can be obtained by using primers derived from ospA gene or 23S rRNA.17,34 This suggests that it may be necessary to use at least two different primer sets to increase the PCR sensitivity in detecting B. burgdorferi DNA in tissues of infected animals.
Positive antibody responses to B. burgdorferi by western blot analysis and KELA were found in all seven ponies after tick exposure ( Figs. 2 , 3 ). Antibodies reached highest levels by approximately 3 months after tick exposure and remained high at the time of necropsy. Western blot analysis indicated that the significant bands are 83, 65, 60, 41, and 39 kd with a few faint bands (28, 30/31, 35, and 45 kd). By comparison, the Centers for Disease Control criterion for diagnosis of human LD is that at least five of the following bands should be present: 18, 21, 28, 30, 41, 45, 58, 66, and 83.39 Although these infected ponies had high antibody titers to B. burgdorferi, the spirochetes were still isolated from various postmortem tissues, especially fascia and muscle. This indicated that once the ponies were infected by a tick bite, they remained persistently infected. Lyme infection is now established to cause persistent infection in humans,32 dogs,4 and horses (this study). Whether antibiotic treatment can eliminate the persistent infection of equine LD is unknown. This horse model should be very useful in answering this question.
Pony No. 178 was first exposed to laboratory-raised nymphs not harboring B. burgdorferi and then 2 months later was challenged with B. burgdorferi–infected adult ticks. Fifteen of the adult ticks were fully engorged 8 days after exposure and 10 of them still harbored B. burgdorferi after engorgement as determined by the IFA test. However, this pony was not infected by B. burgdorferi, as demonstrated by skin biopsies, postmortem culture of tissues, PCR, and serology (KELA and western blot analysis). When mice are exposed to SPF I. scapularis nymphs four times and then reexposed to B. burgdorferi–infected nymphs, only 16.7% of the mice become infected compared to 100% of control mice that were exposed to B. burgdorferi–infected nymphs only.42 Recently, Nazario et al.28 reported that 1 out of 18 tick-immune guinea pigs challenged with B. burgdorferi–infected nymphs was infected compared to 10 out of 18 from the non–tick-immune controls. Several explanations have been suggested. Changes in the cutaneous environment at the tick-bite site may interrupt feeding or be deleterious to incoming spirochetes. Active immunosuppressive components of the tick saliva possibly are neutralized because of host immunity to saliva induced by the first tick exposure. The exact mechanism of this vector-impairing transmission and/or establishment of B. burgdorferi within the host is still unknown. These intriguing results suggest that specific antigen(s) from ticks can induce nonspecific host defense against tick-borne pathogens. Whether these antigens can be used to develop an antitick vaccine will require further studies.42
In conclusion, we have successfully developed an equine LD model that could be used to evaluate the efficacy of Lyme vaccines or chemotherapy in horses. The increasing importance of equine LD in endemic areas argues for a safe and efficacious vaccine against B. burgdorferi infection. At the present time, two types of Lyme vaccines (a bacterin and a recombinant OspA subunit vaccine) are commercially available for dogs. We are in the process of evaluating a recombinant OspA subunit vaccine in ponies by using the equine LD model we developed. Further studies on the pathogenesis of Lyme arthritis in horses by using younger ponies may be necessary because we were unable to induce clinical lameness in these adult ponies.
We are grateful to Allyn Vonderchek, John Daley, David Dierrerich, and Dale Strickland for animal care and Patti Easton for KELA and western blot analyses. This work was supported by grants from the Zweig Fund and the Cornell Biotechnology program, Center of Advanced Technology.