Complete validation of a unique digestion assay to detect Trichinella larvae in horse meat demonstrates the reliability of this assay for meeting food safety and trade requirements.

A tissue digestion assay using a double separatory funnel procedure for the detection of Trichinella larvae in horse meat was validated for application in food safety programs and trade. The assay consisted of a pepsin-HCl digestion step to release larvae from muscle tissue and two sequential sedimentation steps in separatory funnels to recover and concentrate larvae for detection with a stereomicroscope. With defined critical control points, the assay was conducted within a quality assurance system compliant with International Organization for Standardization-International Electrotechnical Commission (ISO/IEC) 17025 guidelines. Samples used in the validation were obtained from horses experimentally infected with Trichinella spiralis to obtain a range of muscle larvae densities. One-, 5-, and 10-g samples of infected tissue were combined with 99, 95, and 90 g, respectively, of known negative horse tissue to create a 100-g sample for testing. Samples of 5 and 10 g were more likely to be positive than were 1-g samples when larval densities were less than three larvae per gram (lpg). This difference is important because ingested meat with 1 lpg is considered the threshold for clinical disease in humans. Using a 5-g sample size, all samples containing 1.3 to 2 lpg were detected, and 60 to 100% of samples with infected horse meat containing 0.1 to 0.7 lpg were detected. In this study, the double separatory funnel digestion assay was efficient and reliable for its intended use in food safety and trade. This procedure is the only digestion assay for Trichinella in horse meat that has been validated as consistent and effective at critical levels of sensitivity.

Overview of food- and water-borne zoonotic parasites at the farm level.

Zoonotic parasites found in food animals include a wide variety of protozoa, nematodes, trematodes, and cestodes. Many of these parasites are emerging or already occur globally due to changes in farming practices and the increased movement of animals, food, and people. Some of the emerging or ubiquitous parasites, including Toxoplasma, Cryptosporidium, Trichinella, and Taenia, present enormous risks to global food production and consumer health. The parasite life cycle stages, such as eggs, oocysts, and cysts, typically resist adverse temperatures, desiccation, natural irradiation, chemicals, and disinfectants that are commonly used for controlling bacteria and viruses. Other important parasites include trematodes such as Clonorchis and Paragonimus, which are transmitted via fish or crustaceans and cause serious human disease in specific regions of the world. The potential for global occurrence of these parasites is increasing. Control of zoonotic parasites at the producer level requires education and the development and implementation of effective measures to eliminate the contamination of agricultural water and feed with viable stages of parasites. Standardisation, implementation, and documentation of control measures should increase confidence in global food trade.

Experimental Brucella abortus infection in wolves.

Four juvenile male wolves (Canis lupus) each received an oral dose of 1.6-1.7 x 10(12) colony-forming units of Brucella abortus biovar 1 isolated from a bison (Bison bison) in Wood Buffalo National Park (Canada), and two others served as negative controls. Infected wolves did not show clinical signs of disease but did develop high Brucella antibody titers. Small numbers of B. abortus were excreted sporadically in feces until day 50 postinoculation (PI). Very small numbers of the bacterium were isolated from urine of only one wolf late on the same day that it was infected, and very small numbers of colonies of B. abortus were obtained from buccal swabs of three wolves for up to 48 hr PI. Two infected wolves euthanized 6 mo after the start of the experiment had no lesions, and colonies of B. abortus were isolated from thymus and most major lymph nodes. The other two infected wolves euthanized 12 mo after the start of the experiment had no lesions, and smaller numbers of brucellae were recovered from fewer lymph nodes compared with the wolves killed 6 mo earlier. The sporadic excretion of very small numbers of brucellae by the wolves was insignificant when compared with the infective dose for cattle. Brucella abortus, brucellosis, Canis lupus, pathogenesis, serology, wolf.

Experimental Toxoplasma gondii infection in grey seals (Halichoerus grypus).

Laboratory-reared animals were used to assess the susceptibility of seals (Halichoerus grypus) to Toxoplasma gondii infection. Four seals were each orally inoculated with 100 or 10,000 oocysts of T. gondii (VEG strain), and another 4 seals served as negative controls. Occasionally, mild behavioral changes were observed in all inoculated seals but not in control animals. A modified agglutination test revealed the presence of antibodies to T. gondii in sera collected from inoculated seals and mice inoculated as controls. No evidence of the parasite was found on an extensive histological examination of seal tissues, and immunohistochemical staining of tissue sections from inoculated seals revealed a single tissue cyst in only 1 seal. Control mice inoculated with 10 oocysts from the same inoculum given to seals became serologically and histologically positive for T. gondii. Cats that were fed brain or muscle tissue collected from inoculated seals passed T. gondii oocysts in feces. This study demonstrates that T. gondii oocysts can establish viable infection in seals and supports the hypothesis that toxoplasmosis in marine mammals can be acquired from oocysts in surface water runoff and sewer discharge.

The occurrence and ecology of Trichinella in marine mammals.

Trichinella in marine mammals has a circumpolar arctic distribution and a narrow range of host species. It is commonly found in polar bears (Ursus maritimus), and increasingly in walruses (Odobenus rosmarus) where it presents a significant zoonotic hazard. This has resulted in the implementation of food safety programs in some arctic communities to test harvested walrus meat for Trichinella larvae prior to consumption. Trichinella has been reported infrequently in bearded seals (Erignathus barbatus) and ringed seals (Phoca hispida), and was once observed in a Beluga whale (Delphinapterus leucas). Cannibalism is probably the most important factor in maintaining a Trichinella cycle in polar bears. Arctic carnivores such as polar bears, arctic foxes (Alopex lagopus) and domestic dogs (Canis familiaris) have a high prevalence of Trichinella infection and the carcasses of at least some of these animals are deposited in the ocean. Scavenging of these carcasses by walruses probably occurs, but may not account for the high prevalence the parasite seen in this host species. Predation, carrion feeding and cannibalism have been documented for walruses and a sylvatic cycle similar to that of bears may exist in walrus populations. Seals and whales are likely infected through infrequent exposure to infected carcasses, either directly by scavenging or indirectly by consuming amphipods or fish that have fed on infected carcasses. The inefficiency of this mechanism may account for the low prevalence of Trichinella in seals and whales. It is known that isolates from marine mammals are cold tolerant, and infectious for man, and have been identified as Trichinella nativa (T2). Molecular and other phylogenetic studies would be useful to facilitate studies on the inter-relationship of Trichinella cycles involving marine and terrestrial mammals in the arctic and subarctic, and in the investigation of human outbreaks of trichinellosis in these areas.

Brucellosis in ringed seals and harp seals from Canada. lforbes@em.agr.ca.

A novel Brucella sp. was isolated from lymph nodes of four ringed seals (Phoca hispida) collected near Pangnirtung (Baffin Island, Canada) in January and February 1995 and in one harp seal (Phoca groenlandica) collected near the Magdalen Islands (Gulf of St. Lawrence, Canada) in March 1996. Bacteriological characteristics were the same for all five isolates. The colonies were typical of Brucella spp., but took 2 to 5 days longer than the traditional species to appear on primary isolation media. Biotyping results did not match any of the known biovars of Brucella, but were similar to isolates of the genus Brucella previously reported from marine mammals inhabiting other areas of the northern hemisphere. This is the first confirmed report of brucellosis in marine mammals from Canada, and the first report of this organism in ringed and harp seals.

A validated Trichinella digestion assay and an associated sampling and quality assurance system for use in testing pork and horse meat.

A revised digestion method, developed for efficiency and quality assurance, was validated for the detection of Trichinella larvae in pork and horse meat to meet requirements for food safety testing and facilitate access to international markets. The method consisted of a tissue homogenization step and a spin bar digestion procedure conducted at 45 degrees C to free larvae from muscle tissue, followed by two sequential separatory funnel steps to concentrate the larvae for detection using a stereomicroscope. Critical control points were determined for the method and monitored during testing. Under conditions of a defined protocol, test capacity was suitable for industrial applications, since multiples of up to 100 g of tissue could be analyzed at one time. The overall sensitivity of the test system depended on the size and origin of the sample taken from individual infected carcasses. Data from swine indicated that the currently accepted sample size of 1 g from individual carcasses consistently detected larval loads of > or =3 larvae per gram. Larval loads of 1.0 to 1.9 larvae per gram required 3- to 5-g samples of muscle tissue for reliable detection. Five-gram samples were considered optimal, because they consistently detected more tissues than 3-g samples, although the difference was not statistically significant. Tissue localization studies in experimental pigs indicated that the tongue and diaphragm were the tissues of choice for the most sensitive larval recovery. A system of analyst training, laboratory certification based on ISO guide 25, and on-site proficiency panel testing was used to ensure that external laboratories would consistently produce reliable test results. The system developed for pork was successfully modified for the testing of horse meat.

Proficiency samples for quality assurance in Trichinella digestion tests.

A reliable method to produce proficiency samples containing known numbers of Trichinella spiralis cysts for use in quality assurance systems for Trichinella digestion tests was developed and validated. A filtrate containing Trichinella cysts was produced by homogenizing and filtering the muscles of an experimentally infected rat. Using a stereomicroscope and micropipette, intact cysts were removed from the filtrate and were transferred onto an agar substrate to allow accurate counting and subsequent transfer into a sample matrix. The proficiency sample matrix consisted of 20-g balls of lean ground beef and was combined with 80 g of a Trichinella-free muscle tissue to obtain the required 100-g sample weight for the assay. The mean overall larval recovery from 404 proficiency samples was 93.0%. Larval recoveries > or = 95, 85, and 75% occurred in 52.4, 84.4, and 94.3%, respectively, of the 404 samples tested. Results indicated that, after a short training period, technicians with no prior experience in digestion techniques performed as well as experienced technicians. The maximum shelf life of proficiency samples was not determined but was at least 3 weeks. Validation data were used to develop panels composed of proficiency samples prepared as described above and to establish guidelines for the interpretation of proficiency panel results.

Infection of cattle with Brucella abortus biovar 1 isolated from a bison in Wood Buffalo National Park.

An experiment was conducted to determine if cattle could be infected with a strain of Brucella abortus biovar 1 isolated from a bison in Wood Buffalo National Park. Three pregnant cows inoculated conjunctivally with 5.7 x 10(8) cfu of the bacterium, and their subsequent calves, showed seroconversion on standard serological tests for bovine brucellosis, and large numbers of the bacterium were isolated from numerous tissues at necropsy. A 4th cow that was moved into the pen that previously contained the inoculated cows subsequently showed seroconversion, and the same strain of B. abortus biovar 1 was isolated from numerous tissues. Although this strain from bison in Wood Buffalo National Park has existed in isolation from cattle for over 60 years, it remains infectious and contagious for cattle.

Experimental studies on Brucella abortus in moose (Alces alces).

Four moose (Alces alces) were inoculated conjunctivally with B. abortus biovar 1 to determine their susceptibility to brucellosis, and to describe the serology, bacteriology, hematology, clinical chemistry, and pathology associated with infection. All moose became infected. Two moose were killed at day 70 post-exposure, one (83F) died acutely at day 85, and one was killed at day 166. None of the moose had clinical signs, except for 83F immediately before death. Infected moose were readily detected serologically by the buffered antigen plate test, Brewer card test, standard tube agglutination test, and complement fixation test as used for brucellosis in cattle. With the exception of samples from 83F taken 24 hours before death, clinical chemistry, and hematology results were stable for all moose, and similar to normal values seen in cattle. Lesions seen in all moose were indicative of endotoxemia, and moose 83F died of acute endotoxic shock. Brucella abortus biovar 1 was isolated from several tissues in all moose, most notably from lymph nodes where counts often exceeded 4 x 10(4) colony forming units per g of tissue. Thus infection with B. abortus will kill moose, and progression of the disease is likely rapid under field conditions. Moose appear to be a dead-end host for brucellosis.

The brucellosis and tuberculosis status of wood bison in the Mackenzie Bison Sanctuary, Northwest Territories, Canada.

Postmortem examinations were done on 51 wood bison (Bison bison athabascae) killed as part of a multidisciplinary research project in the Mackenzie Bison Sanctuary, Northwest Territories, Canada, between 1986 and 1988. There was no gross, histological or bacteriological evidence of brucellosis or tuberculosis in these bison. Traumatic lesions were seen in one calf that had been attacked by wolves and a second calf that had been gored. Antibody titers to Brucella abortus were not found in sera from these 51 animals or an additional 112 wood bison that were chemically-immobilized or killed in the Sanctuary between 1986 and 1990. The combined prevalence of the diseases in the population could not have exceeded 5.95% for the necropsy survey to have missed finding at least one infected animal, and the prevalence of brucellosis in the population would have had to be less than 1.95% for the broader serological survey to have failed to find at least one reactor animal on the battery of tests. These results, and the cumulative epidemiological information on brucellosis and tuberculosis in bison, indicate that bovine brucellosis and tuberculosis are not enzootic in the wood bison population in and around the Mackenzie Bison Sanctuary, and suggest that the population is free of these diseases. However, this expanding population is at risk of contracting both diseases from the infected bison population in and around nearby Wood Buffalo National Park.

Transmission of brucellosis from reindeer to cattle.

Sixteen reindeer (Rangifer tarandus tarandus) naturally infected with Brucella suis biovar 4 were penned with 6 male and 2 female cattle for 30 days, then removed and euthanatized. During this period, 5 reindeer had fawns, and 2 reindeer aborted. Brucella suis biovar 4 was recovered from all adult reindeer at necropsy. Nine reindeer had B suis biovar 4 in uterus, udder, and/or milk. The cattle were euthanatized 2 months after the reindeer were removed. Clinical or pathologic signs of disease were not observed in the cattle. Brucella suis biovar 4 was isolated from 2 male and from both female cattle at necropsy. The female cattle had positive reactions on the buffered plate antigen test, brucellosis card test, tube agglutination test, complement fixation test, and indirect enzyme immunoassay for most of the experiment, but the males had inconsistent reactions on these tests. The indirect enzyme immunoassay was the only test to detect all cattle from which bacteria were cultured. This study revealed that caution is warranted before moving reindeer or caribou into areas of traditional agriculture.

A survey of brucellosis and tuberculosis in bison in and around Wood Buffalo National Park, Canada.

Examinations of complete or partial remains of 72 bison found dead in and around Wood Buffalo National Park, Canada, revealed evidence of brucellosis in 18 (25%) and tuberculosis in 15 (21%), with a combined prevalence of 42%. Urease-positive and ureasenegative strains of Brucella abortus biovar 1, and strains of biovar 2, were isolated from tissues of bison, including synovium and exudate from severe arthritic lesions. Mycobacterium bovis was isolated from a range of granulomatous lesions that were similar to those reported in tuberculous cattle. Diseased bison had a broad geographical distribution, and were found outside the park on at least three natural corridors. The diseases have a deleterious effect on this population of bison, and pose a health risk to other bison herds, livestock, and native hunters in the region.

Brucella abortus infection in 14 farm dogs.

Fourteen dogs were obtained from 10 farms with Brucella-infected cattle and were studied for periods ranging from 2 to 81 days. At necropsy, Brucella abortus biovar 4 was isolated from all 14 dogs. Mandibular, medial retropharyngeal, tracheobronchial, and mesenteric lymph nodes yielded the highest rate of recovery. Urogenital infection with active shedding was seen in a single aged bitch. Fecal samples (291 from 13 dogs) were B abortus culture negative. Ten dogs monitored serologically over time had standard tube agglutination test titers that fluctuated between 1:50 with incomplete reaction and greater than or equal to 1:200 with positive reaction. At necropsy, the magnitude of seroconversion was not directly related to the number of culture-positive tissues. The maximal duration of infection, based on the time interval between the last possible exposure to infected cattle and recovery of the organism at necropsy, was 464 days. If it were assumed that infection occurred at about the same time as seroconversion in the cattle herd of origin, maximal observed duration of infection was 539 days. The actual maximal duration of infection could not be determined from this study, but would have exceeded the values reported here. The data indicate that dogs have the potential to infect cattle and could pose a threat for longer duration of disease transmission than had previously been assumed. Although the risk of transmission appears small, infection of human beings or cattle associated with Brucella-infected dogs would cause unfavorable political and economic consequences, particularly in low-incidence areas. Removal of contact dogs from infected herds should prevent such transmission.

An outbreak of Brucella abortus biovar 2 in Canadian cattle.

An outbreak of brucellosis caused by Brucella abortus biovar 2 was identified in cattle in Alberta in December 1986. This was the only clinical infection discovered since the national cattle herd was declared brucellosisfree in 1985. It was the first report of B. abortus biovar 2 in Canadian cattle. The outbreak, involving three herds containing purebred Hereford cattle, was spread by the private treaty sale of untested cattle, and was identified following investigation of an abortion. The source of infection for the outbreak was not established, but several possibilities were identified including infected herds present in the area during the mid-1970's, latent infection originating in a Saskatchewan herd during the early 1960's, American cattle imported during the early 1970's, and brucellosis-infected bison in Wood Buffalo National Park. The containment and elimination of this nidus of infection appears to have been successful, and the national cattle herd at the time of writing is free of the disease.

Characterization of an atypical biotype of Brucella abortus.

Brucella abortus strains were isolated from bovine tissue and milk samples from seven Ontario herds. The isolates were characterized by colonial morphology, requirement of CO2 for growth, lysis by Tbilisi phage, biochemical tests and agglutination in monospecific sera. They resembled B. abortus biotype 2 (on the basis of sensitivity to thionin and basic fuchsin) and biotype 4 (on the basis of agglutination with anti-Brucella "M" but not anti-Brucella "A" absorbed sera). Sodium dodecyl sulphate-polyacrylamide gel electrophoresis of these isolates and B. abortus biotypes 1, 2 and 4 showed similar profiles. Immunoblots with anti-A and anti-M absorbed sera showed different antigenic regions reacting with the specific sera and also confirmed that the atypical B. abortus isolates were serologically similar to biotype 4.

Brucella canis Isolates from Canadian Dogs.

Eleven Brucella canis isolates from Canadian dogs were characterized by dye and antibiotic sensitivity, phage susceptibility, urease and H(2)S production, CO(2) requirement, and reaction with monospecific A,M, and R anti-Brucella antiserum. The isolates could be separated into two distinct groups. One group had a sensitivity pattern similar to that seen with the American type strain RM666, while the other group had a pattern identical to that of a Mexican strain, Mex 51. Epidemiological studies supported contraction of infections in the United States and Mexico respectively. The characteristics of all isolates were stable after repeated subculture indicating that strain differences could serve as useful epidemiological markers and supporting division of the species into two biovars.

Atypical isolates of Brucella abortus from Canada and the United States characterized as dye sensitive with M antigen dominant.

A total of 41 Brucella isolates, examined by standard biotyping procedures, were found to be similar to Brucella abortus biovar 2 in dye sensitivity but had a dominant M antigen. Oxidative metabolic tests performed on 39 of the isolates confirmed them as B. abortus. Additional biochemical and bacteriophage susceptibility studies were performed on 35 of the isolates. The isolates had identical reactions in the various tests, except for one isolate which was resistant to lysis by all phage strains used. Two isolates were injected into guinea pigs and shown to be virulent. The isolates described in this study appear similar to atypical Brucella isolates previously reported in the United Kingdom and the United States and may form the basis of a new biovar, B. abortus biovar 10.

A comparison of five serological tests for bovine brucellosis.

Five serological assays: the buffered plate antigen test, the standard tube agglutination test, the complement fixation test, the hemolysis-in-gel test and the indirect enzyme immunoassay were diagnostically evaluated. Test data consisted of results from 1208 cattle in brucellosis-free herds, 1578 cattle in reactor herds of unknown infection status and 174 cattle from which Brucella abortus had been cultured. The complement fixation test had the highest specificity in both nonvaccinated and vaccinated cattle. The indirect enzyme immunoassay, if interpreted at a high threshold, also exhibited a high specificity in both groups of cattle. The hemolysis-in-gel test had a very high specificity when used in nonvaccinated cattle but quite a low specificity among vaccinates. With the exception of the complement fixation test, all tests had high sensitivities if interpreted at the minimum threshold. However, the sensitivities of the standard tube agglutination test and indirect enzyme immunoassay, when interpreted at high thresholds were comparable to that of the complement fixation test. A kappa statistic was used to measure the agreement between the various tests. In general the kappa statistics were quite low, suggesting that the various tests may detect different antibody isotypes. There was however, good agreement between the buffered plate antigen test and standard tube agglutination test (the two agglutination tests evaluated) and between the complement fixation test and the indirect enzyme immunoassay when interpreted at a high threshold. With the exception of the buffered plate antigen test, all tests were evaluated as confirmatory tests by estimating their specificity and sensitivity on screening-test positive samples.(ABSTRACT TRUNCATED AT 250 WORDS)