Immunodiagnosis of experimental Parelaphostrongylus tenuis infection in elk

Abstract

Elk infected with the meningeal worm, Parelaphostrongylus tenuis (Protostrongylidae), do not consistently excrete larvae in feces, making the current method of diagnosing live animals using the Baermann fecal technique unreliable. Serological diagnosis could prove more useful in diagnosing field-infected animals but depends on the identification and availability of good quality antigen. To mimic field infections, 2 elk were inoculated with 6 infective L3 larvae of P. tenuis, and another 2 with 20 L3 larvae. Fecal samples were examined for nematode larvae using the Baermann technique and serum samples taken were tested for anti-P. tenuis antibody with ELISAs by using the excretory-secretory (ES) products of L3, and sonicated adult worms as antigens. One animal passed first-stage larvae in its feces 202 days postinoculation, but passed none thereafter. The remaining 3 inoculated animals did not pass larvae. In contrast to parasite detection, antibodies against larval ES products were detected in all animals starting from 14 to 28 days postinoculation and persisted until the termination of the experiment on day 243 in 2 animals that harbored adult worms. Antibodies against somatic antigens of the adult worm were not detected until day 56 but also persisted until the end of the experiment in the animals with adult worms. In 2 elk that had no adult worms at necropsy, anti-ES antibodies were detected transiently in both, while anti-adult worm antibodies were present transiently in one. These findings confirm the superiority of P. tenuis larval ES products over somatic adult worm antigens as serodiagnostic antigens, as previously observed in studies of infected white-tailed deer, and extend the application of the newly developed ELISA test in diagnosing and monitoring cervids experimentally infected with P. tenuis.

Introduction

The meningeal worm, Parelaphostrongylus tenuis, is widely distributed in eastern North America and is considered one of the most pathogenic protostrongylid nematodes of cervids. It rarely causes disease in white-tailed deer, regarded as the normal definitive host, but is responsible for neurological signs and death of other infected cervid hosts, including woodland caribou (Rangifer tarandus caribou), moose (Alces alces), elk (Cervus elaphus canadensis), fallow deer (Dama dama), mule deer (Odocoileus hemionus hemionus), and black-tailed deer (O. h. columbianus). Non-cervid domestic livestock, including llamas (Lama glama), goats, sheep, and cattle, are also susceptible (1,2).

Compared with other cervids and domestic livestock, elk appear to be intermediate in their susceptibility to the meningeal worm. Clinical signs range from imperceptible to severe; their severity can be attributed to the infecting dose. Experimental infection of elk by Samuel and colleagues (3) demonstrated that animals given a low dose (15 infective third-stage larvae (L3)), did not show clinical signs, whereas most of those inoculated with a moderate dose (25 to 75 L3) developed patent, clinical infections. Larger doses of 125 to 300 L3 invariably resulted in severe neurological signs and death. Similar signs have been observed in elk naturally infected with P. tenuis on range in Pennsylvania and confirmed by finding adult worms at necropsy (4,5). Elk translocated within the state of Oklahoma from the P. tenuis-free Wichita mountains to an enzootic area later exhibited signs such as “circling” and ataxia, and were found at postmortem to have meningeal worms in the cerebral cortex (6). Neurological signs observed in Michigan's elk for years (7) were eventually attributed to meningeal worm infestation (8) and the incidence of clinical signs correlated with the presence of adult worms and histological lesions in the central nervous system of infected animals (6).

Some elk survive P. tenuis infection long enough for worms to mature and for larvae to be passed in feces. An elk calf inoculated with 2 relatively large doses of P. tenuis (150 L3 on each occasion, 23 d apart) showed only transient neurological signs and passed small numbers of first-stage larvae (L1) (7). Several elk given moderate doses of P. tenuis (25-75 L3) passed larvae, while those given only 15 L3 did not (9). Dorsal-spined larvae, morphologically indistinguishable from P. tenuis L1, were recovered from the feces of wild elk in Minnesota (10) and those found in elk feces in Manitoba were shown by infection experiments to be P. tenuis (11). Although shed only intermittently and in small numbers, larvae passed by elk can develop to the infective stage in terrestrial gastropods and be a source of new infection (3). To prevent the spread of the parasite by this means, the commercial movement of elk from eastern North America to the western part of the continent is presently under restrictions.

The only available antemortem test for diagnosing P. tenuis infection is the Baermann technique, which relies on the detection of L1 in the feces of patent animals by microscopic examination. Conclusive designation of dorsal-spined larvae as P. tenuis can only be achieved through complicated and expensive infection experiments or, more recently, by a polymerase chain reaction (PCR) test performed on recovered larvae (12). Nevertheless, the chance of recovering larvae from an infected elk is poor because of the low numbers of larvae excreted, intermittent larval excretion, and the bulk of feces produced by elk. The problems associated with the use of the Baermann technique to diagnose P. tenuis infection cannot be overcome by repeatedly testing an infected animal (9).

A serological test based on antibody detection could prove more reliable in diagnosing meningeal worm infections in elk. However, attempts to do so have had limited success. Test sensitivity was poor when somatic adult nematode antigens were used as antigen for enzyme-linked immunosorbent assay (ELISA) and was only slightly improved using somatic larval antigen, and only in animals inoculated with large parasite doses (13). Recently, we evaluated different antigen preparations for the serological diagnosis of white-tailed deer experimentally infected with P. tenuis (14,15). In the present report, ES products of L3 and somatic antigens of adult worms were evaluated as ELISA diagnostic antigens in an attempt to serologically diagnose elk experimentally infected with various doses of P. tenuis larvae, judiciously chosen to mimic natural infections.

Materials and methods

Animals and infection

Three elk calves acquired from elk farms located in central Saskatchewan and one from Alberta where P. tenuis does not occur (2) were transported to Thunder Bay, Ontario, held on concrete flooring and infected with P. tenuis L3. Animals were 6 to 12 mo old at the start of the experiment, at which time feces from all animals were examined for nematode eggs using a sugar flotation method and confirmed to be free of dorsal-spined nematode larvae by the modified Baermann-beaker method (16); small numbers of trichostrongylid eggs were found in all animals. For purposes of infection and blood collection, animals were anesthetized with xylazine hydrochloride (Rompun; Miles, Etobicoke, Ontario) administered using a blowpipe and lightweight syringe. Two calves were each orally inoculated with 6 P. tenuis L3 while 2 others received 20 L3. Blood samples were collected from each animal before inoculation, 1 to 2 wk after inoculation, and at regular intervals thereafter until the end of the experiment at day 194 to day 243. One of the calves given 20 L3 passed small numbers of Dictyocaulus-like larvae in the feces during the experiment, and at necropsy 8 adult worms identified as Dictyocaulus sp. were recovered from the lungs. After day 80, about 20 g of feces from each inoculated animal were examined twice weekly for L1. Every 4 to 6 wk, 2 to 3 g of feces from the animals were examined for parasite eggs. At necropsy, the brain and entire spinal cord were removed. The cranial venous sinuses and surface of the brain were examined for adult worms using a compound microscope (4X magnification). The spinal cord was pressed between heavy glass plates and examined for parasites at 16X magnification. Two additional calves born and kept on an elk farm in central Saskatchewan served as uninfected, control animals. Blood and fecal samples were collected from the control animals on days 0, 28, and 194. The feces were examined for P. tenuis-like larvae and parasite eggs. Serum was extracted from clotted blood and kept at −20°C until used. All elk calves were observed daily for clinical signs.

Parasites and antigen preparation

Using the modified Baermann-beaker method, P. tenuis L1 were extracted from the feces of a white-tailed deer killed by vehicle collision near Grand Marais, Minnesota, and found to harbor 1 female and 2 male adult P. tenuis in the cranium. Snails (Triodopsis multilineata) were injected with L1 and kept at room temperature for 8 wk. Infective L3 were obtained from the snails following digestion in 0.7% pepsin solution containing 0.8% HCl. Larvae were used to infect elk as described above or for antigen preparation as follows: 2400 L3 were cultured in RPMI-1640 solution (5 mL) containing fetal bovine serum (2.5%), penicillin (100 U/mL), streptomycin (100 μg/mL) and amphotericin B (1.25 μg/mL) at 37°C for 52 h. Culture supernatant, hereafter referred to as larval excretory-secretory (ES) products, was aspirated and clarified at 12 000 × g for 5 min. Adult worms were obtained from the brain of a necropsied white-tailed deer 147 d after infection with 100 to 150 P. tenuis L3 (15). Four adult worms were rinsed 3 times with phosphate-buffered saline (PBS; pH 7.4), resuspended in 0.5 mL PBS, placed on ice and sonicated (Braunsonic Melsungen; Braun, Allentown, Pennsylvania, USA) at 300 W for 1 min at a time, for a total of 5 min. The sonicated parasite suspension was centrifuged at 12 000 × g for 5 min, the supernatant, hereafter referred to as the somatic adult worm antigen preparation, was harvested and the protein content determined using the bicinchinonic acid (BCA) assay (Pierce, Rockford, Illinois, USA). Both antigen preparations were stored at −20°C until used.

Alkaline phosphatase-labeled rabbit anti-elk IgG

Elk serum (1 mL) was clarified at 16 000 × g for 20 min and mixed with 0.5 mL of Protein G sepharose (Sigma-Aldrich, Oakville, Ontario) that was suspended in 40 mM Tris-HCl (pH 7.8) containing 500 mM NaCl and the mixture was incubated on a rocking shaker for 1 h at room temperature. Sepharose-bound elk IgG was separated from non-bound serum proteins by centrifuging the mixture at 3800 × g for 2 min, followed by 3 washes with 5 mL of 40 mM Tris, pH 7.8, containing 500 mM NaCl. Elk IgG was released from the sepharose by treatment with 0.1 M glycine-HCl (pH 2.7) and immediately pH-neutralized with 75 μL of 1 M Tris-HCl, pH 9.0. The purified elk IgG was dialyzed against PBS and the protein concentration determined by the BCA assay. Elk IgG (60 μg) was resuspended in 500 μL PBS, mixed with an equal volume of Freund's Complete Adjuvant (Sigma Chemical, St. Louis, Missouri, USA) and subcutaneously inoculated into a rabbit. A booster dose of elk IgG (60 μg) in PBS mixed with Freund's Incomplete Adjuvant (Sigma Chemical) was administered subcutaneously to the rabbit 2 wk later. The rabbit was anesthetized with methoxyfluorane (Metofane; Janssen, Toronto, Ontario) 2 wk after the booster immunization and exsanguinated for the purpose of serum collection. Rabbit IgG was purified from the rabbit antiserum by applying the latter to a new protein G sepharose column and the bound rabbit IgG eluted with glycine-HCl as described above. Purified rabbit IgG containing anti-elk IgG activity was conjugated to alkaline phosphatase (AP; Sigma Chemical) with the aid of 1% glutaraldehyde and following standard procedures (17). After treatment with 1 M ethanolamine (pH 7.4), the rabbit anti-elk IgG-AP conjugate was dialyzed against PBS (pH 7.4) at 4°C, mixed with an equal volume of glycerol and stored in aliquots at −20°C until used.

Enzyme-linked immunosorbent assay

Indirect ELISAs using somatic adult antigen preparation (somatic adult-ELISA) or larval ES products (ES-ELISA) were performed according to a previously published procedure (14) with minor modifications. Wells of ELISA plates (Immulon 4; Dynatech Laboratories, Chantilly, Virginia, USA) were coated with somatic adult worm antigen preparation, 0.5 μg/well, or ES products prepared as described above and diluted 1:3 in PBS. Alkaline phosphatase-rabbit anti-elk IgG conjugate was diluted 1:250 in PBS containing Tween 20 (PBST). All reagents were used at 50 μL per well. Color development was stopped with the use 5% ethylene diamine tetraacetic acid, and the optical density (OD) read at 405 nm with a spectrophotometer (Titertek Multiskan; Labsystems, Helsinki, Finland). Results are presented as either antibody titer units or mean OD values and standard deviations (SD). Titer units were calculated from a straight line equation derived by the regression analysis of OD values against reciprocal serum dilutions (18). Pre-infection serum titer was set at 10 units and a 4-fold increase over the negative serum (40 units) was used in scoring a sample as positive, following conventional clinical immunological practice (19,20). Cut-off OD values were set at the mean value of triplicate determinations plus 2 SD of the OD values of serum samples collected from all 6 animals before the inoculation of the experimental animals. Significant differences between mean OD values collected from the same animal at different times were detected by using t-test analysis using statistical software (SigmaStat; SPSS Inc., Chicago, Illinois, USA).

Results

Clinical signs

No clinical signs were observed in any of the inoculated or control elk throughout the study.

Recovery of larvae from the feces and adult worms from the CNS of inoculated elk

Two motile, dorsal-spined larvae were recovered from elk no. 1 (inoculation dose = 6 L3) at day 202, but not at any other time from this animal or from the remaining 3 inoculated and 2 control animals. At necropsy on day 243, 1 female adult P. tenuis was recovered from the cranium of elk no. 1, while an adult female and a male were recovered from elk no. 3 (inoculation dose = 20 L3) (Table I). Adult worms were not recovered from elk nos. 2 or 4 at necropsy on day 194.

Table I.

Detection of serum anti-parasite antibodies in inoculated elk

Antibody titers against the ES products of P. tenuis L3 were first detectable 14 to 15 d postinoculation in elk nos. 2, 3, and 4, and by day 28 in no. 1. Anti-ES antibody titers waned in all 4 animals during the course of infection, but persisted at detectable levels until day 243 in the 2 elk (nos. 1 and 3) from which adult worms were recovered (Table I and Figure 1). In both animals, anti-adult worm antibodies were first detected at day 56 and persisted until the end of experiment at day 243. In the 2 elk from which no adults were recovered, anti-adult worm antibodies were transiently detected only at day 56 in elk no. 2 and not at all in elk no. 4 (Table I and Figure 2). For purposes of comparison, OD values of ELISAs performed on sera from all 4 experimental and both control, uninfected animals on days 0, 28, and 193–194 are presented in Table II. Cut-off values were 0.221 for ES-ELISA and 0.579 for somatic adult-ELISA. Serum samples collected from elk nos. 1 and 3 on days 28 and 193 gave significantly higher ES-ELISA OD readings than their respective day 0 samples, and than the corresponding serum samples collected from control animals (P 0.05) and were scored positive. In contrast, samples obtained from elk no. 1 and 3 on day 193 but not the day 28 samples, were significantly higher than the day 0 and control samples in somatic adult-ELISA (P 0.05). Elk nos. 2 and 4 were positive on day 28 but not on day 194 in ES-ELISA. In the somatic adult-ELISA, elk no. 2 serum samples were similar to one another and significantly lower than the readings of control animals (P 0.05). Both postinoculation serum samples from elk no. 4 gave readings similar to the preinoculation serum sample (P > 0.05), even though they were higher than the cut-off OD value. Background OD readings of serum samples from uninoculated animals (controls and preinoculation) were lower in ES-ELISA than somatic adult-ELISA (P 0.05).

Figure 1. Antibodies against the excretory-secretory products of third-stage larvae (L3) of Parelaphostrongylus tenuis in experimentally infected elk. Animals were inoculated with 6 or 20 L3 and bled before and after inoculation as indicated. Anti-P. tenuis IgG antibodies present in the sera of inoculated animals were tested by ELISA using ES products of P. tenuis L3. Pre-inoculation antibody titer for each animal was set at 10 units and samples with 40 units or more were scored positive.

Figure 2. Antibodies against the somatic antigens of adult Parelaphostrongylus tenuis in experimentally infected elk. Animals were inoculated with 6 or 20 L3 and bled before and after inoculation as indicated. Anti- P. tenuis IgG antibodies present in the sera of inoculated animals were tested by ELISA using somatic antigens from adult P. tenuis. Pre-inoculation antibody titer for each animal was set at 10 units and samples with 40 units or more were scored positive.

Table II.

Discussion

The development of a serological test for P. tenuis was previously attempted in elk inoculated with doses of infective larvae higher than are likely in the field (13). However, a useful test for P. tenuis must be able to detect animals harboring few parasites. Field-infected elk may harbor as few as 1 to 3 adult worms (6), as do naturally infected white-tailed deer (1 to 9 worms) (1,21). This is partly a consequence of the low numbers of infective larvae found in gastropod intermediate hosts (2,22) but also possibly due to the effect of concomitant immunity that prevents the accumulation of worms in deer following initial infection (21). Since most elk receiving high doses will display typical disease signs (3), a useful test for this host must be capable of identifying patent and non-patent infections in animals that harbor few parasites and show no, or only transient, clinical signs.

The present study was designed to assess the possibility of detecting persistent anti-parasite antibodies in elk exposed to numbers of parasites that approximate field exposure, and to compare the results with the current parasitological method of larval detection. The elk that shed larvae in feces had only a single female adult worm in the CNS, indicating that a male adult worm was present at some time to fertilize the female but could have died before the animal was examined postmortem. On microscopic examination of the female worm, eggs in the uterus appeared normal, suggesting that larvae could have developed. A second elk had both a male and female worm but no larvae were recovered from its feces. Larval detection may have been hindered by a combination of factors, including inherent low sensitivity of the Baermann technique, frequency of testing (twice weekly), or because of unidentified host effects on the parasite resulting in the intermittent shedding of few larvae (9). Elk are atypical hosts of P. tenuis when compared with white-tailed deer, in which the co-adaptation of the parasite and host allows both to thrive. White-tailed deer rarely show signs of disease, yet the parasite enjoys high reproductive fitness as measured by the production of larvae. A female worm can produce as many as 24.2 larvae per gram of feces, and only 8% of infected white-tailed deer harboring both female and male worms fail to excrete larvae (21).

In addition to low larval output in elk, delayed patency was also observed in this study, compared with published prepatent periods of days 83 to 165 in animals inoculated with 15 to 75 L3 (3). Patency requires the reproductive maturation and fertilization of a female worm by a male. Low-dose infections may result in delayed copulation and explain extended prepatent periods observed experimentally (3,23,24). Field infections may similarly show extended prepatent periods because of low infecting dose. Therefore, the use of larval recovery methods for detecting and controlling the spread of P. tenuis are often ineffective and impractical. Quarantine procedures have also been used to prevent the spread of nematodes (25), but as a tool against P. tenuis, its effectiveness is diminished by the logistical difficulty of restricting the movement of wild or ranched ungulates in anticipation of patent infections.

In contrast to parasite recovery methods, antibodies against the ES products of the meningeal worm were detected as early as 2 to 4 wk following experimental infection and in animals given as few as 6 L3. Although the anti-ES antibody levels waned in all infected animals, they remained detectable in 2 elk that had adult worms, and in which antibodies against adult worm antigens were also demonstrable. The remaining 2 animals showed a temporary anti-ES seroconversion, starting from day 14, that disappeared after day 147, indicating that infection of both was achieved but was overcome before patency was attained. The detection of antibodies against adult worms on day 56 only in 1 of the 2 animals suggests that worm(s) in this animal died after becoming adult(s). This fits with the known development of P. tenuis in the white-tailed deer, where the adult stage is reached about day 40 (26). Thus, testing of infected elk sera with both ES products and somatic adult antigens of adult worms by ELISA was useful in detailing the progression of antibody levels in animals exposed to low dose infection and in mapping eventual antibody decay attributable to the termination of an infection. The use of somatic adult worm antigens did not permit detection of infected animals before day 56. Previous work done with somatic adult worm antigens showed that animals exposed to 300 L3 could be detected at day 83, but not those exposed to 15 L3, including 2 elk found to harbor adult worms at necropsy (13). In the current study, both animals with adult worms were detectable with the use of somatic adult worm antigens. The use of a higher concentration of the antigen (10 times higher), and of rabbit anti-elk IgG conjugate which helped in eliminating another layer of reagent, may have served to increase the sensitivity of the assay when compared to the previous report. Nevertheless, the usefulness of somatic antigen preparation for routine serological testing is undermined by high background reactivity and by propensity for false positive results as observed in our study. For instance, without the benefit of a preinoculation serum sample, one of the inoculated elk (no. 4) would have been scored positive from the result of the somatic adult-ELISA (Table II). Indeed, when serum samples from 16 animals from P. tenuis-free areas were tested in ELISA using somatic adult worm antigen, 9 animals showed moderate to strong reactions (range, OD = 0.294-1.194), compared to only 4 animals reacting moderately to larval ES products (range, OD = 0.297–0.403; Ogunremi et al, unpublished observations). Thus, somatic adult worm antigen preparations may be less amenable for routine serological diagnosis of P. tenuis-infected elk.

This study shows that an ELISA based on ES products of L3 is useful for diagnosing elk experimentally infected with meningeal worms and for monitoring parasite development in this host. The persistence and detectability of antibodies long after the larvae have transformed into adult worms is attributable, at least in part, to antigens shared between the different parasite stages. Elk living in enzootic areas are likely to become exposed repeatedly over time to meningeal worms and while this may not result in an increased parasite load as has been shown in white-tailed deer (21), it may serve to boost antibody levels and increase the chance of a successful serodiagnosis, as demonstrated in moose (Ogunremi et al, manuscript submitted).

To ensure reliability of the ELISA utilizing ES products in the serodiagnosis of meningeal worms in field-infected elk, some important issues need to be addressed, namely diagnostic test sensitivity and specificity. Sensitivity values will require the testing of many animals known to be infected with the parasite, and to this end serum samples are being assembled from colleagues. Assessment of test specificity, on the other hand, requires testing animals known to be uninfected with the parasite, e.g., elk in areas free of P. tenuis. Following extensive testing, if a significant proportion of sera from elk living in P. tenuis-free areas react with the ES products, i.e., non-specifically, efforts will be made to reduce the non-specific results by procedures that have been used by others to improve test specificity including treatment of antigens or serum samples with various detergents or enzymes (27,28). An alternative approach will be to identify a cloned P. tenuis gene from a cDNA library that was recently created for P. tenuis (Ogunremi et al, manuscript under preparation).

A reliable serological test will be a useful tool for regulators concerned with the spread of P. tenuis through translocation of elk into P. tenuis-free regions of Canada. Such a tool should also be useful for wildlife management and may have application in the monitoring of elk translocated into P. tenuis enzootic areas. Since 1900, most elk translocations into enzootic areas have either failed or met with limited success. This has been attributed to the debilitating neuromotor effects of P. tenuis (6,29) but could also result from subclinical effects of infection that nonetheless affect reproduction. Given this history, the translocation of elk from the Elk Island National Park, Alberta, to several locations in eastern Canada, which started in 1998 (30), might benefit from diligent serological monitoring of animals released in areas with infected white-tailed deer.