Phylogenetic relationships among the European and American bison and seven cattle breeds reconstructed using the BovineSNP50 Illumina Genotyping BeadChip

Here we present the first attempt to use the BovineSNP50 Illumina Genotyping BeadChip for genome-wide screening of European bison Bison bonasus bonasus (EB), two subspecies of American bison: the plains bison Bison bison bison (PB), the wood bison Bison bison athabascae (WB) and seven cattle Bos taurus breeds. Our aims were to (1) reconstruct their evolutionary relationships, (2) detect any genetic signature of past bottlenecks and to quantify the consequences of bottlenecks on the genetic distances amongst bison subspecies and cattle, and (3) detect loci under positive or stabilizing selection. A Bayesian clustering procedure (STRUCTURE) detected ten genetically distinct clusters, with separation among all seven cattle breeds and European and American bison, but no separation between plain and wood bison. A linkage disequilibrium based program (LDNE) was used to estimate the effective population size (N _e) for the cattle breeds; N _e was generally low, relative to the census size of the breeds (cattle breeds: mean N _e = 299.5, min N _e = 18.1, max N _e = 755.0). BOTTLENECK 1.2 detected signs of population bottlenecks in EB, PB and WB populations (sign test and standardized sign test: p = 0.0001). Evidence for loci under selection was found in cattle but not in bison. All extant wild populations of bison have shown to have survived severe bottlenecks, which has likely had large effects on genetic diversity within and differentiation among groups.     

Omasal ciliated protozoa in cattle, bison, and sheep.

Omasal contents were collected from slaughtered cattle (n = 54), bison (n = 15), and sheep (n = 40) to determine numbers and generic distribution of ciliated protozoa. Total protozoan numbers were significantly lower in omasal contents than in ruminal contents of all three species, but the percent composition of all protozoan genera was similar between omasal and ruminal populations. The highest numbers of omasal protozoa found were 7.61 X 10(5)/g in cattle, 7.01 X 10(5)/g in bison, and 1.29 X 10(6)/g in sheep. Omasal dry matter was significantly higher than ruminal dry matter in all species and ranged up to 51.5% in cattle fed high-concentrate diets. The omasal pH was similar to the ruminal pH in all species. The number of omasal laminae averaged 149, 145, and 74 for cattle, bison, and sheep, respectively. Although protozoan concentrations in omasal contents were approximately 80% lower than those in ruminal contents, the omasum harbored relatively high numbers of ciliated protozoa. The resident omasal protozoa are extremely difficult to remove, particularly in cattle, and apparently are responsible for reinoculating transiently defaunated rumens.     

Omasal ciliated protozoa in cattle, bison, and sheep

Omasal contents were collected from slaughtered cattle (n = 54), bison (n = 15), and sheep (n = 40) to determine numbers and generic distribution of ciliated protozoa. Total protozoan numbers were significantly lower in omasal contents than in ruminal contents of all three species, but the percent composition of all protozoan genera was similar between omasal and ruminal populations. The highest numbers of omasal protozoa found were 7.61 X 10(5)/g in cattle, 7.01 X 10(5)/g in bison, and 1.29 X 10(6)/g in sheep. Omasal dry matter was significantly higher than ruminal dry matter in all species and ranged up to 51.5% in cattle fed high-concentrate diets. The omasal pH was similar to the ruminal pH in all species. The number of omasal laminae averaged 149, 145, and 74 for cattle, bison, and sheep, respectively. Although protozoan concentrations in omasal contents were approximately 80% lower than those in ruminal contents, the omasum harbored relatively high numbers of ciliated protozoa. The resident omasal protozoa are extremely difficult to remove, particularly in cattle, and apparently are responsible for reinoculating transiently defaunated rumens.     

Superovulation and embryo transfer in wood bison (Bison bison athabascae)

Two experiments were done to develop an effective superovulatory treatment protocol in wood bison for the purpose of embryo collection and transfer. In experiment 1, donor bison were assigned randomly to four treatment groups (N = 5 per group) to examine the effects of method of synchronization (follicular ablation vs. estradiol-progesterone treatment) and ovarian follicular superstimulation (single slow-release vs. split dose of FSH). Recipient bison were synchronized with donor bison by either follicular ablation (N = 8) or estradiol-progesterone treatment (N = 9). In experiment 2, bison were assigned randomly to four treatment groups (N = 5 per group) to examine the ovarian response to two versus four doses of FSH, and the effect of progesterone (ovarian superstimulation with or without an intravaginal progesterone-releasing device). Donor bison were inseminated with fresh chilled wood bison semen 12 and 24 hours after treatment with GnRH (experiment 1) or LH (experiment 2). The ovarian response was assessed using ultrasonography. In experiment 1, the number of large follicles (≥7 mm) increased in response to both FSH treatments, but the diameter of the largest follicle detected 4 and 5 days after the start of ovarian superstimulation was greater in bison treated with a single dose of FSH than in those treated with two doses (P < 0.05). A total of 10 ova and/or embryos were collected. One blastocyst was transferred to each of five recipient bison resulting in the birth of two live wood bison calves. In experiment 2, two doses of FSH resulted in a greater number of large follicles (≥9 mm) on Days 4, 5, and 6 (P < 0.05) after beginning of superstimulation (Day 0), and more ovulations than four doses of FSH (11.2 ± 2.4 vs. 6.4 ± 0.8; P < 0.05). Embryo collection was performed on only five donors, and a total of 19 ova and/or embryos were recovered. In summary, fewer FSH treatments were as good or better than multiple treatments, consistent with the notion that minimizing handling stress improves the superovulatory response in bison. Follicular ablation and estradiol plus progesterone treatment were effective for inducing ovarian synchronization in embryo donor and recipient bison, and an intravaginal progesterone-releasing device during superstimulatory treatment did not influence the superovulatory response or embryo collection. Delaying ovulation-inducing treatment (GnRH or LH) to 5 days after superstimulatory treatment resulted in a greater number of ovulations and improved embryo collection efficiency (experiment 2). Embryo collection and transfer resulted in live offspring from wild wood bison.     

Sperm characteristics in plains (Bison bison bison) versus wood (Bison bison athabascae) bison

The objective was to compare sperm characteristics between the two subspecies of North American bison, plains bison (Bison bison bison) and wood bison (Bison bison athabascae). Frozen-thawed ejaculated sperm from age-matched plains (n = 3) and wood (n = 2) bison were evaluated for morphometry, motility, viability, protein profile, and in vitro fertilization characteristics. Sperm morphometry and motility were assessed with computer-based systems, viability was assessed with SYBR-14 and propidium iodide, and fertilizing ability was determined using a heterologous in vitro fertilization system (using bovine oocytes). For plains versus wood bison, there were significant differences for head width (4.76 ± 0.22 vs 4.71 ± 0.19 μm; mean ± SD), head area (35.64 ± 1.91 vs 34.72 ± 2.64 μm²), head perimeter (23.61 ± 0.68 vs 23.31 ± 0.98 μm), midpiece length (14.58 ± 0.4 vs 14.36 ± 0.51 μm), midpiece width (0.81 ± 0.06 vs 0.79 ± 0.07 μm), and tail length (46.61 ± 2.15 vs 45.98 ± 2.08 μm). However, there was no significant difference in head length (overall, 9.04 ± 0.37 μm), progressive motility (41.16 ± 8.39%), or viability (41.58 ± 5.58%). Based on two-dimensional gel electrophoresis, 93 out of 113 protein spots were similar in their expression patterns. Furthermore, we inferred that differences in sperm biometry between these subspecies did not affect in vitro fertilization percentage (overall, 82.62 ± 12.13%). Based on these findings, we concluded that plains bison were an appropriate research model for developing reproductive technologies for wood bison.     

DNA fingerprinting and genetic diversity of the European bison (Bison bonasus), American bison (Bison bison), and gray Ukrainian cattle (Bos taurus)

To describe genetic variability and population diversity in domesticated populations of American bison (Bison bison), aurochs (Bison bonasus), and gray Ukrainian cattle (Bos taurus) different variants of DNA fingerprinting technique (utilizing the M13 phage DNA, (TTAGGG)4 synthetic oligonucleotide, and three arbitrary primers as hybridization probes) were used. Several parameters characterizing polymorphism and genetic diversity levels in each population (species) were evaluated on the basis of the profiles obtained. Dendrograms reflecting similarities between individual animals were constructed. Genetic variability of minisatellite and telomeric markers observed in the gray Ukrainian cattle flock was higher than that in aurochs and bisons. Comparison of the intrapopulation similarity (S) and gene diversity (H) indices along with the analysis of clusters in the dendrograms showed that the relatedness between the aurochs individuals was much higher than between the individual animals in the bison and gray Ukrainian cattle flocks. Furthermore, the gray Ukrainian cattle flock was represented by more distant relatives than the bison flock. It is suggested that reduced genetic variability and the appearance of deviant genotype observed in the two bison lines under selection, resulted from close inbreeding and the founder effect. The diagnostic value and efficacy of utilization of different molecular markers for estimation of genetic diversity and relatedness in domesticated animal populations is discussed.     

A comprehensive evaluation of cattle introgression into US federal bison herds

Genetic introgression, especially from interspecies hybridization, is a significant threat to species conservation worldwide. In this study, 11 US federal bison populations were comprehensively examined for evidence of both mitochondrial and nuclear domestic cattle (Bos taurus) introgression. Mitochondrial introgression was examined using established polymerase chain reaction methods and confirmed through analysis of D-loop sequences. Nuclear introgression was assessed in 14 chromosomal regions through examination of microsatellite electromorph and sequence differences between bison and domestic cattle. Only one population was identified with domestic cattle mitochondrial DNA introgression. In contrast, evidence of nuclear introgression was found in 7 (63.6%) of the examined populations. Historic accounts of bison transfers among populations were corroborated with evidence of introgressed DNA transmission. While neither nuclear nor mitochondrial domestic cattle introgression was detected in bison from Grand Teton National Park, Sully's Hill National Game Preserve, Wind Cave National Park, or Yellowstone National Park, adequate sample sizes were available only from the last 2 populations to allow for statistical confidence (>90%) in nuclear introgression detection limits. The identification of genetically unique and undisturbed populations is critical to species conservation efforts, and this study serves as a model for the genetic evaluation of interspecies introgression.     

Ruminal ciliated protozoa in bison

Ruminal contents from 79 slaughtered bison and 2 ruminally cannulated bison were collected to obtain information on total numbers and species distribution of ciliated protozoa. The bison originated from numerous herds throughout the Great Plains and were grouped into three dietary categories: (i) only forage; (ii) forage with moderate levels of supplementation; and (iii) feedlot concentrate-silage diet. Total ciliate counts were highest in bison receiving grain supplementation (210.1 x 10(4)/g) and lowest in bison consuming only forage (27.1 x 10(4)/g). All protozoan species found in bison have been reported in domestic livestock, although Ophryoscolex sp., a relatively common protozoan in cattle, was detected at low concentrations in only eight bison. The uncommon holotrich Microcetus lappus was present in five bison in concentrations reaching 8.4% of the total ciliate population. Charonina ventriculi, another infrequently observed species, was present in 18 bison, with the highest concentrations in forage-fed animals. Thirty bison possessed a type B protozoan population, characterized by Epidinium sp., Eudiplodinium maggii, and Eudiplodinium bovis. Thirty-eight bison possessed a mixed A-B population, characterized by Polyplastron sp. coexisting with low numbers of Eudiplodinium maggii or Epidinium sp. or both. Thirteen bison possessed populations lacking any remnant type B ciliate species. At least 29 of the bison possessing Polyplastron sp. were known to have been in contact with cattle, whereas all bison isolated from cattle had type B populations. The reduction of type B populations in bison becomes increasingly likely as bison production expands into areas inhabited by domestic livestock.     

Conservation genomics: disequilibrium mapping of domestic cattle chromosomal segments in North American bison populations

Introgressive hybridization is one of the major threats to species conservation, and is often induced by human influence on the natural habitat of wildlife species. The ability to accurately identify introgression is critical to understanding its importance in evolution and effective conservation management of species. Hybridization between North American bison (Bison bison) and domestic cattle (Bos taurus) as a result of human activities has been recorded for over 100 years, and domestic cattle mitochondrial DNA was previously detected in bison populations. In this study, linked microsatellite markers were used to identify domestic cattle chromosomal segments in 14 genomic regions from 14 bison populations. Cattle nuclear introgression was identified in five populations, with an average frequency per population ranging from 0.56% to 1.80%. This study represents the first use of linked molecular markers to examine introgression between mammalian species and the first demonstration of domestic cattle nuclear introgression in bison. To date, six public bison populations have been identified with no evidence of mitochondrial or nuclear domestic cattle introgression, providing information critical to the future management of bison genetic resources. The ability to identify even low levels of introgression resulting from historic hybridization events suggests that the use of linked molecular markers to identify introgression is a significant development in the study of introgressive hybridization across a broad range of taxa.     

Ruminal ciliated protozoa in bison.

Ruminal contents from 79 slaughtered bison and 2 ruminally cannulated bison were collected to obtain information on total numbers and species distribution of ciliated protozoa. The bison originated from numerous herds throughout the Great Plains and were grouped into three dietary categories: (i) only forage; (ii) forage with moderate levels of supplementation; and (iii) feedlot concentrate-silage diet. Total ciliate counts were highest in bison receiving grain supplementation (210.1 x 10(4)/g) and lowest in bison consuming only forage (27.1 x 10(4)/g). All protozoan species found in bison have been reported in domestic livestock, although Ophryoscolex sp., a relatively common protozoan in cattle, was detected at low concentrations in only eight bison. The uncommon holotrich Microcetus lappus was present in five bison in concentrations reaching 8.4% of the total ciliate population. Charonina ventriculi, another infrequently observed species, was present in 18 bison, with the highest concentrations in forage-fed animals. Thirty bison possessed a type B protozoan population, characterized by Epidinium sp., Eudiplodinium maggii, and Eudiplodinium bovis. Thirty-eight bison possessed a mixed A-B population, characterized by Polyplastron sp. coexisting with low numbers of Eudiplodinium maggii or Epidinium sp. or both. Thirteen bison possessed populations lacking any remnant type B ciliate species. At least 29 of the bison possessing Polyplastron sp. were known to have been in contact with cattle, whereas all bison isolated from cattle had type B populations. The reduction of type B populations in bison becomes increasingly likely as bison production expands into areas inhabited by domestic livestock.     

Genetic variation within and relatedness among wood and plains bison populations.

There are two recognized subspecies of bison, wood (Bison bison athabascae) and plains (Bison bison bison) bison. The establishment of most bison populations from a small number of individuals has raised concerns about their genetic variation. To this end, 11 bison populations were surveyed with 11 microsatellite loci in order to calculate genetic variation and genetic distances. Mean number of alleles ranged between 3.18 at Antelope Island State Park (Utah) and 6.55 at Wood Buffalo National Park (Alberta and Northwest Territories). Mean heterozygosity ranged from 0.295 at Antelope Island State Park to 0.669 at Custer State Park (South Dakota). The amount of genetic variability present in the bison populations as measured by mean number of alleles and overall probability of identity was found to correlate with the number of founders for all sampled populations. The G-test for heterogeneity revealed some evidence for the existence of subpopulations at Wood Buffalo National Park, however very small genetic distances between these subpopulations suggest that nuclear material from the plains bison introduced into Wood Buffalo National Park has diffused throughout the park. Genetic distances between the sampled populations were generally larger between than within the two bison subspecies.     

Polymorphism of bovine microsatellite DNA sequences in the lowland European bison

Investigations of genetic polymorphism of microsatellite DNA sequences were conducted in 22 individuals of the European bisonBison bonasus (Linneaus, 1758) from Bia?owie?a Primeval Forest. For this purpose 27 cattle microsatellite primer pairs were used. Among the 27 microsatellite markers examined, an amplification product was obtained for 21 loci. This rendered it possible to identify total of 40 alleles in the bison population tested. In addition, eight loci were proved to be monomorphic. A majority of the 40 alleles identified was identical with the alleles identified at the corresponding loci in cattle. Only two alleles seem to be specific for the European bison. The value of heterozygosity for the examined loci in bison population from Bia?owie?a was low and ranged from 0.13 to 0.53. Hence, the polymorphism information content was low as well. Based on our results the microsatellite DNA markers identified in cattle may be used to analyse the genetic structure of the population of European bison.     

Validation of 15 microsatellites for parentage testing in North American bison, Bison bison and domestic cattle

Fifteen bovine microsatellites were evaluated for use in parentage testing in 725 bison from 14 public populations, 178 bison from two private ranches and 107 domestic cattle from five different breeds. The number of alleles per locus ranged from five to 16 in bison and from five to 13 in cattle. On average, expected heterozygosity, polymorphism information content (PIC) and probability of exclusion values were slightly lower in bison than in cattle. A core set of 12 loci was further refined to produce a set of multiplexed markers suitable for routine parentage testing. Assuming one known parent, the core set of markers provides exclusion probabilities in bison of 0.9955 and in cattle of 0.9995 averaged across all populations or breeds tested. Tests of Hardy-Weinberg and linkage equilibrium showed only minor deviations. This core set of 12 loci represent a powerful and efficient method for determining parentage in North American bison and domestic cattle.     

Serial ovarian ultrasonography in wild-caught wood bison (Bison bison athabascae)

The objectives of this study were to determine the feasibility of daily examination of wild-caught wood bison and to characterize the ovarian function using serial transrectal ultrasonography and blood hormone analysis. Ten 2-year-old wood bison heifers obtained from Elk Island National Park were placed in a corral adjacent to a handling system designed for restraining bison. The handling system was left open to the corral allowing the bison to explore it freely for 2 months. Active acclimation followed for a 2-week period, during which the bison were herded daily through the handling system and rewarded with whole oats. Finally, the bison were restrained in the handling system and rewarded with whole oats upon release. Once conditioned, daily transrectal examination of the ovaries was completed in 100% of attempts for 30 days (January–February) using a B-mode scanner with a 5 to 10-MHz linear array. Follicle size and numbers were recorded, and individual follicles were identified serially. Blood samples were collected daily and the serum was analyzed for FSH concentrations. Nonrandom changes were detected in the number of follicles ≥4 mm in diameter per day (P < 0.05). Each peak in follicle numbers was associated with the development of a single dominant follicle. The interval between the emergence of successive dominant follicles was 6.8 ± 0.6 days (mean ± SEM). The maximum diameter of the dominant follicle was 9.9 ± 0.4 mm. In conclusion, wild-caught wood bison were amenable to daily examination and blood sampling, and ovarian dynamics were characterized by wave-like development of anovulatory antral follicles. The demonstrated success of this approach to the study of ovarian function will be useful for characterizing the annual reproductive pattern in wood bison, which is necessary for the development of bison-specific protocols for controlling ovarian function for species conservation.     

Differential introgression of uniparentally inherited markers in bison populations with hybrid ancestries

Historical hybridization between Bison bison (bison) and Bos taurus (cattle) has been well documented and resulted in cattle mitochondrial DNA (mtDNA) introgression, previously identified in six different bison populations. In order to examine Y chromosome introgression, a microsatellite marker (BYM-1) with non-overlapping allele size distributions in bison and cattle was isolated from a bacterial artificial chromosome (BAC) clone, and was physically assigned to the Y chromosome by fluorescence in situ hybridization. BYM-1 genotypes for a sample of 143 male bison from 10 populations, including all six populations where cattle mtDNA haplotypes were previously identified, indicated that cattle Y chromosome introgression had not occurred in these bison populations. The differential permeability of uniparentally inherited markers to introgression is consistent with observations of sterility among first generation hybrid males and a sexual asymmetry in the direction of hybridization favouring matings between male bison and female cattle.     

Reproductive success of bison bulls (Bison bison bison) in semi-natural conditions

Sexual size dimorphism has been related to a high variance in male reproductive success. Research on American bison (Bison bison bison), a species that shows moderate sexual dimorphism, has put forward the assumption that under wild conditions the largest part of the female herd is bred by a limited number of males. Thus far, this hypothesis has never been supported by genetic data. The aim of this study is therefore to test the hypothesis of extreme male reproductive variance among bison through the determination of the reproductive success of the bulls in private herds. We analysed genetic paternity in four private herds (n=19, 33, 39 and 54) kept under semi-natural conditions. By analysing blood, hair and tissue samples, using 8-10 polymorphic microsatellites, the reproductive success of the bulls was quantified. Additional genetic data were gathered through a survey distributed among bison breeders with private herds over the world. Results show a highly differential reproductive success for bison males in private herds kept under semi-natural conditions. In each study group, a single male sires at least 70% of the calves. In the survey groups, a single male sires 81-100% of the calves. Only one survey group displayed a deviating pattern showing a more evenly distribution of paternity (28.57, 33.33 and 38.10%) among three same aged subadult males.     

Hormonal control of estrous cyclicity and attempted superovulation in wood bison (Bison bison athabascae)

The wood bison (Bison bison athabascae) is a threatened Canadian species that has faced extinction twice in the last 100 yr. Development of assisted reproductive technologies could help ensure the long-term propagation and genetic management of this species. The objectives of this study were to refine estrus synchronization techniques and evaluate superovulatory responses after FSH or eCG administration. In Experiment 1, females were fitted with Syncro-mate B (SMB) implants for 9 d and received an injection of either estradiol valerate (E2V; n = 9) or cloprostenol (PGF; n = 9) at implant insertion (Day-9). In Experiment 2, estrus was synchronized with SMB implants and a PGF injection of Day-9, and superovulation was attempted on Day-2 with either 2500 IU eCG (n = 5) or 400 mg Folltropin-V (n = 5). In each experiment, biosin were examined daily for estrual behavior. Ultrasonography was used during the luteal phase to detect ovulation and assess ovarian status; feces were analyzed by ELISA for immunoreactive progestogens (P) to study ovarian endocrine responses. In Experiment 1, a closer synchrony of estrus was observed between Days 2 to 4 among the PGF-treated (77.8%) than the E2V-treated (66.7%) females. Corpora lutea (CL) were detected in 55% of E2V- and PGF-treated females. In Experiment 2, neither treatment successfully induced superovulation, with only a single female per treatment producing > or = 1 CL. In both experiments, progestogen profiles were similar for each treatment (P < 0.05).     

Ruminal microbial populations and fermentation characteristics in bison and cattle fed high- and low-quality forage

Ruminal microbial populations and fermentation products were compared between two ruminally cannulated bison (375 kg) and two ruminally cannulated Hereford steers (567 kg) on alfalfa or prairie hay diets. Differential media were used to enumerate carbohydrate-specific bacterial subgroups. Voluntary dry matter intake was higher (P=0.006) for cattle than for bison fed alfalfa, but prairie hay intake was not different (P=0.16) between the two species. Volatile fatty acid concentrations, pH, and ruminal ammonia were similar between bison and cattle on both diets. Total anaerobic bacteria and xylanolytic bacterial counts were higher (P<0.02) in bison than in cattle fed alfalfa. However, with the prairie hay diet, no differences in bacterial counts on any medium were observed between ruminant species. Both bison and cattle possessed a mixed A-B protozoan population with nearly identical protozoan numbers and distribution of genera. The similarities between bison and cattle consuming either high-or low-quality forage suggest that any differences in putative forage digestibility between the species are not due to differences in microbial counts.     

Comparisons of ruminal fermentation characteristics and microbial populations in bison and cattle.

Ruminal microbial populations, fermentation characteristics, digestibility, and liquid flow rates in two ruminally cannulated bison and two ruminally cannulated Hereford steers fed a prairie hay diet were compared. No significant differences in anaerobic bacterial counts, volatile fatty acid concentrations, or ruminal pHs were evident between bison and cattle. Also, no significant differences in neutral detergent fiber digestibility, indigestible fiber retention time, or intake were detected between bison and cattle, although cattle had higher levels (P less than 0.08) of ruminal dry matter and indigestible fiber than bison. Bison had a smaller (P = .02) ruminoreticular volume, faster liquid dilution rates, and faster liquid turnover times than cattle. The average ruminal ammonia nitrogen concentration was higher (P = 0.02) in bison (1.17 mg/dl) than in cattle (0.79 mg/dl). Total ciliate protozoal counts and cell volume were greater (P = 0.07) in bison (32.8 x 10(4)/g and 407.1 x 10(-4) ml/g, respectively) than in cattle (15.7 x 10(4)/g and 162.2 x 10(-4) ml/g, respectively). Bison harbored higher (P less than 0.02) numbers of Dasytricha spp., Eudiplodinium maggii, Eudiplodinium bursa, and Epidinium spp. than cattle and possessed a type B protozoan population. The cattle possessed a mixed type A-type B population that was characterized by Ophryoscolex spp. and Polyplastron spp. in association with low concentrations of Epidinium spp. and Eudiplodinium maggii.     

Resistance to malignant catarrhal fever in American bison (Bison bison) is associated with MHC class IIa polymorphisms

The Rhadinovirus ovine herpesvirus-2 (OvHV-2) is the most common causative agent of malignant catarrhal fever (MCF) in clinically susceptible ruminants including cattle and bison. American bison (Bison bison) are highly susceptible to clinical MCF. Nevertheless, approximately 20% of bison on ranches or in feedlots become infected with the virus without developing clinical disease. Defining the genetic basis for differences in susceptibility between bison could facilitate development of improved control strategies for MCF. One genetic region that influences susceptibility to infectious diseases is the major histocompatibility complex (MHC). In this study, a Bison bison (Bibi) DRB3 oligonucleotide microarray was used to type 189 bison from 10 herds where MCF outbreaks had occurred. Binary logistic regression was used to classify DRB3 alleles as resistant (R), susceptible (S) or neutral (N). Animals were reclassified using six DRB3 genotype categories: N/N, N/R, N/S, R/S, R/R and S/S. Analysis of homogeneity across herds showed that there was a herd effect. Consequently, a penalized logistic regression model was run with herd and genotype categories as the explanatory variables. The R/R genotype was associated with resistance to MCF (P = 0.0327), while the S/S genotype was associated with clinical MCF (P = 0.0069). This is the first evidence that MHC class IIa polymorphism is associated with resistance or susceptibility to OvHV-2-induced MCF.