Genetic base and inbreeding of Canadienne, Brown Swiss, Holstein and Jersey cattle in Canada - C. Hansen*, J.N.B. Shrestha**, R.J. Parker^[1], G.H. Crow^[2] & J.N. Derr^[3]

Present address:

* Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Alberta T6G 2P5, Canada
** Dairy and Swine Research and Development Centre, Agriculture and Agri-Food Canada, P. O. Box 90, 2000 Route 108 est, Lennoxville, Québec J1M 1Z3, Canada

Summary

The narrowing of the genetic base is of major concern to many cattle breeders and this is a consequence of a small effective population size and an increase in the inbreeding coefficient in cattle populations. Studies of pedigrees of cattle populations found in Canada showed that the effective population size and rate of inbreeding over a 12-year period (1983-94) were 123 and 0.412 percent for Canadienne, 975 and 0.054 percent for Brown Swiss, and 2 183 and 0.024 percent for Jersey cattle populations. Over the same period, the year to year trends in inbreeding coefficients based on co-ancestry was 0.11 percent for Canadienne and 0.07 percent for Brown Swiss cattle populations, and 0.37 percent for Holstein bulls used by the artificial insemination industry. Inbreeding was not found to be a major problem in any of the populations examined. However, only 10 bulls of their respective breed sired 61 percent of the Canadienne, 21 percent of the Brown Swiss, and 29 percent of the Jersey cattle populations, and 41 percent of the Holstein bulls in the bull studs. More and more breeders are demanding proven sires to increase milk production from fewer cows resulting in the narrowing of the genetic base of the national cattle populations. Newer technology that provides precise genetic modification could further contribute to the narrowing of the genetic base compromising the ability to sustain current production and respond to changing markets in the long-term.

Résumé

Le rapprochement de la base génétique est une des questions qui inquiète le plus à la plupart des éleveurs et est la conséquence du nombre restreint de la population et de l’augmentation du coefficient de consanguinité dans les populations bovines. Les études de pedigree des populations bovines réalisées au Canada ont démontré que la taille effective de la population et le niveau de consanguinité sur une période de 12 ans (1983-1994) était de 123 et 0,412 pour cent pour la race Canadienne, 975 et 0,054 pour cent pour la Brune Suisse, et 2.183 et 0,024 pour cent pour les populations de Jersey. Sur la même période, la tendance annuelle des coefficient de consanguinité sur la base de co-ancêtre était de 0,11 pour cent pour la Canadienne et 0,07 pour cent pour la Brune Suisse, tandis que pour les taureaux Holstein utilisés par l’industrie d’insémination artificielle le pourcentage était de 0,37. On n’a pas considéré le consanguinité un problème important parmi les populations examinées. Cependant, sur 10 taureaux de chacune des races seulement 61 pour cent de la Canadienne, 21 pour cent de la Brune Suisse et 29 pour cent de la Jersey se sont reproduits, et 41 pour cent des tauraux Holstein. Toujours plus les éleveurs demandent des tauraux contrôlés pour augmenter la production de lait de quelques vaches provenants de la base génétique nationale. Les nouvelles technologies qui proportionnent des modifications génétiques précises pourraient contribuer à augmenter le rapprochement de la base génétique compromettant ainsi la possibilité de soutenir la production actuelle et de répondre aux changements des marchés à long terme.

Keywords: Canadian dairy cattle, Genetic base, Rate of inbreeding.

Introduction

The Canadian dairy cattle industry, like those in many industrialised nations, extensively relies on artificial insemination (AI) for the widespread use of genetically superior sires. This technology has enabled breeders to achieve rapid genetic progress for the economically important morphological characteristics and production performance in the short-term. Embryo transfer (increasingly based on embryos produced from in vitro techniques) has become increasingly common over the last 10 years and is beginning to demonstrate an impact on dairy cattle populations. This is especially true for the more popular breeds of cattle, including the Holstein and Jersey.

Currently, only a few proven bulls produce the majority of offspring in the purebred populations (Rutter and Pearson, 1981) and sons of these bulls provide the best opportunity for genetic improvement and eventually replace their sire in the AI units. The relationship among animals in the breed and among bulls in the AI units will increase over time, unless unrelated bulls with exceptional genetic merit can be identified and utilised.

Less populous breeds such as the Canadienne (Fortin, 1940) and Brown Swiss, however, also make significant and ever increasing use of these breeding and reproductive practices. In general, inbreeding has been associated with increased homozygosity, redistribution of genetic variances, a greater probability of the expression of lethal recessive genes and reduction in performance of inbred animals.

As a consequence, there is today, a concern that the use of only a few outstanding bulls will eventually contribute to the narrowing of the genetic base of the cattle populations (Young, 1984; Young et al., 1988; Miglior and Burnside, 1995) and one will observe the detrimental effects of inbreeding on fitness and performance traits (Young et al., 1969; Miglior et al., 1992). The narrowing of the genetic base of a breed or species can have devastating results in the long-term and must be avoided if the current practice of improvement in production is to be sustained and changing demands for future markets are to be addressed. This study was undertaken to investigate the status of the genetic base and inbreeding in the Canadienne, Brown Swiss, Holstein and Jersey cattle populations in Canada.

Materials and Methods

Pedigree records were obtained from the Canadian Livestock Records Corporation in Ottawa for 31 621 Canadienne and Brown Swiss cattle, and 256 945 Jersey cattle. Records on the Canadienne and Brown Swiss breeds were combined into a single file because many of today’s Canadienne cattle have Brown Swiss parentage as a consequence of a decision made in 1969 by the Québec Ministry of Agriculture following increasing concern over inbreeding and a requirement to improve productivity (Hansen et al., 2000).

Due to the large number of Holstein cattle in Canada, only an excerpt of the total pedigreed population was utilised for the current study. A pedigree file consisting of animals tracing back to four generations was obtained from Holstein Canada in Brantford, Ontario for 5 539 Holstein cattle representing the parentage of bulls found in the AI units (Eastern Breeders Inc., United Breeders, Western Ontario Breeders and the Centre d’Insemination Artificielle du Québec).

The tabular method for calculating relationship based on co-ancestry (Cruden, 1949; Emik and Terrill, 1949) was utilised to determine inbreeding coefficients because of the presence of overlapping generations in cattle populations. It was also assumed that animals in the base population (the earliest year in the current pedigree file) were unrelated. The average inbreeding coefficient estimated from the diagonal of the additive relationship matrix was regressed on years to determine the year to year trend in inbreeding for the Canadienne and Brown Swiss cattle and Holstein bulls. Unfortunately, the size of the Jersey file exceeded the computer resources available at the time of the study. Therefore, it proved impossible to use this technique to evaluate the Jersey cattle population as a whole.

The effective population size and rate of inbreeding due to random mating among relatives of finite population size was determined according to procedures proposed by Wright (1931) for natural populations and Gowe et al. (1959) for experimental poultry populations. The actual number of parents (sires and dams) for the Canadienne, Brown Swiss and Jersey cattle populations was utilised. Records with missing parents had to be excluded. This analysis was, therefore, restricted to the years 1983 to 1994 for which the most complete pedigree records were available.

Results

The total number of registrations and the number of male and female parents for Canadienne, Brown Swiss and Jersey cattle are presented in Table 1. The average number of registrations in the Canadienne, Brown Swiss and Jersey cattle was 389±39, 1 756±105 and 7 074±123 animals, respectively. The mean number of male and female parents, respectively, was 24±1 and 184±21 for Canadienne, 193±9 and 1 321±160 for Brown Swiss, and 418±18 and 6 832±98 for Jersey.

The inbreeding coefficients from 1983 to 1994 were based on co-ancestry and ranged from 0 to 1.7 percent for the Canadienne breed, from 0.01 to 0.7 percent for the Brown Swiss breed, and from 0.9 to 4.8 percent for the Holstein bulls (Figure 1). Corresponding year to year trends in inbreeding over the 12-year period were 0.11±0.01, 0.07±0.01 and 0.37±0.02 percent, respectively.

The effective population size and rate of inbreeding estimates (Wright, 1931) ranged from 61 to 110, and 0.46 to 0.82 percent for Canadienne breed, from 368 to 792, and 0.06 to 0.14 percent for Brown Swiss, and from 1 209 to 1 829, and 0.03 to 0.04 percent for Jersey cattle (Table 2). The mean effective population sizes and rates of inbreeding over the 12-year period were 85±4 and 0.606±0.029 percent for Canadienne, 665±40 and 0.079±0.006 percent for Brown Swiss, and 1 573±65 and 0.032±0.001 percent for Jersey cattle. Correspondingly, the effective population size and rate of inbreeding estimates (Gowe et al., 1959) ranged from 97 to 159, and 0.32 to 0.52 percent for Canadienne, from 575 to 1 141, and 0.04 to 0.09 percent for Brown Swiss, and from 1 667 to 2 552, and 0.02 to 0.03 percent for Jersey cattle (Table 2). The mean effective population sizes and rates of inbreeding over the 12-year period were 123±5 and 0.412±0.015 percent for Canadienne, 975±52 and 0.054±0.004 percent for Brown Swiss, and 2 183±92 and 0.024±0.001 percent for Jersey cattle.

Table 1. Total number of registrations and number of male (N_m) and female (N_f) parents, in Canadienne, Brown Swiss and Jersey cattle populations.

Figure 1. Average of inbreeding coefficients derived from the additive relationship matrix for the Canadienne and Brown Swiss cattle and Holstein bulls

Discussion

There appears to be a striking difference among the Canadienne, Brown Swiss and Jersey breeds in the size of the national breeding populations (Table 1). One can clearly see that the size of the Canadienne population is relatively small compared to the Jersey population, which corresponds with the average number of annual registrations. In the last two years, there has been a decline in registrations for the Canadienne and Jersey breeds. In contrast, there has been a marginal increase in the registrations for the Brown Swiss breed.

A closer examination of the number of parents in the Canadienne breed shows that there is, on average, one sire for every eight dams and 16 animals registered. Corresponding estimates for the Brown Swiss breed were one sire for every seven dams and nine animals registered and for the Jersey breed, one sire for every 16 dams and 17 animals registered. As these figures indicate, the ratio of dams to sires in the Canadienne and Brown Swiss breeds is approximately half that of the more populous Jersey breed. This observation has implications for the future of these breeds as it reflects a definite trend towards a narrowing of the current genetic base in these two breeds.

The annual increase in inbreeding coefficients relative to their base population was negligible for the Canadienne and Brown Swiss cattle, while, a more substantial increase was noted in the Holstein bull population (Figure 1). The inbreeding coefficient between 1987 and 1990, which steadily increased at an annual rate of 0.21 percent for all Holstein bulls in Canada (Miglior and Burnside, 1995) was lower than the 0.37 percent in the present study for Holstein bulls in all AI units between 1983 and 1994.

Table 2. The effective population size (N_e and N_e') and rate of inbreeding (DF and DF') in the Canadienne, Brown Swiss and Jersey cattle populations.

(Wright, 1931)

(Gowe et al., 1959)

The rise in inbreeding in the Holstein bull population in recent years, however, is more than three times that observed in the Canadienne population and more than five times that in the Brown Swiss population as a whole. This suggests that inbreeding may be increasing more rapidly in bulls used by the artificial breeding industry compared to all Holstein bulls in general. Although, the inbreeding coefficient in the base population was not considered, it appears that at the present time the overall rise in inbreeding is not of great concern in any of the breeds studied. This conclusion is in agreement with those reported by Young et al. (1988) for the Holsteins in the United States, and Miglior and Burnside (1995) for the Canadian Holstein.

The estimate of effective population size and the rate of inbreeding based on Wright’s approximations were smaller in size and the inbreeding coefficient larger compared to similar estimates based on Gowe and coworkers (Table 2). The latter may be more appropriate in commercial populations where the number of cows are more numerous than bulls, and the number of individuals chosen as parents are relatively similar in ratio of sexes over generations. Thus, reducing the variances among families contributes to a higher estimate of effective population size. The rate of inbreeding in the Canadienne breed when based on Wright’s approximation that ignored over lapping generations was considerably higher (Table 2) compared to corresponding estimates based on co-ancestry (Figure 1). In contrast, the estimates for the Brown Swiss breed were similar.

The increased use of artificial insemination in Japan has been shown to contribute to the decrease in the effective population size in a closed herd of the Japanese Black cattle population in Hyogo prefecture from 262 in 1960 to eight in 1988 (Takayanagi et al., 1996), and in a closed herd of the Japanese Brown cattle population in Hyogo prefecture from 725 in 1960 to 40 in 1997 (Ibi et al., 1996). The current Kerry breed in Ireland, derived from less than 5 percent of the animals registered as foundation stock in 1887, also has a small effective population size fluctuating close to 50 and a high level of cumulative inbreeding of nearly 15 percent (O’-Huigin and Cunninghan, 1990).

An examination of the 10 most frequently sought sires between 1980 and 1995 revealed that the most popular Canadienne bulls sired approximately 61 percent of the Canadienne cattle born during that time period. Corresponding estimates for the Brown Swiss and Jersey bulls were 21 and 29 percent, respectively. In contrast, 41 percent of the Holstein bulls in the four Bull studs were sired by 10 bulls which is slightly lower than the 47 percent of progeny tested US bulls from 1975 to 1979 that were sired by the 10 most popular Holstein bulls (Rutter and Pearson, 1981). There are currently less than 50 Canadienne breeders in the province of Québec and only a small number outside of the province. An examination of the herds of six of the largest breeders revealed that only 39 animals could be found that were not half or full-sibs. The fact that in recent years as many as 61 percent of Canadienne cattle were sired by just 10 bulls and that the effective number of male parents used in the Canadienne breed is extremely small is cause for serious concern over expression of the possible detrimental effects of inbreeding in this breed in the near future. This poses a rather significant problem for the Société des Éleveurs de Bovins Canadiens and for all Canadienne breeders. Recent efforts to utilise unrelated bulls derived from previously stored embryos is an important activity to broaden the existing genetic base (Hansen et al., 2000).

Historically, mild inbreeding and selection directed towards greater use of prominent ancestors have provided the foundation population of current breeds. It is important to note that the breeding strategy that has been employed by the dairy industry in recent years involves mating less closely related animals due to the increased use of Holstein sires from the US than would occur at random. This fact and the increasing number of grade cows entering the herdbook (Miglior and Burnside, 1995) has almost certainly helped keep the overall rate of inbreeding in populations such as the ones studied here lower than would have been the case if such action had not been initiated. Nevertheless, the actual inbreeding coefficient may be higher than the present estimates because inbreeding in the base populations was not considered. The present breeding strategy for the less populous breeds may not be adequate in the future. Their smaller effective population size results in increased variance due to drift contributing to increased homozygosity.

Lush (1948) stated that in most pure breeds of livestock the effective number of males was in the order of 30 to 50 before AI was a factor in dairy cattle improvement. Similarly, Young (1984) argued that popular bulls will have little if any greater influence than did most prominent bulls of past generations with no reason for concern about inbreeding in dairy cattle populations in the United States. However, this may not remain to be true with the advent of emerging technologies that provide opportunities for increasing the precision of estimating transmitting ability, sampling of bulls on a global basis, the possibility of sex selection, multiplication of embryos by nuclear transfer, production of embryonic stem cells, and the isolation and cloning of specific genes for introduction into the bovine genome. Thus, the newer technologies will enable breeders to achieve rapid genetic progress for economically important production traits in the shortterm, contributing to the increased use of a few proven bulls on a global scale, further narrowing the genetic base of the national cattle populations.

In conclusion, it appears that at the present time the rise in inbreeding is not high enough to be of concern, although inbreeding has been an important issue for many dairy cattle breeders for a number of years. Signs of possible future danger do exist, however. The level of inbreeding of Holsteins bulls in the A.I. studs has increased rapidly over the last few years. The effective number of breeding parents in the Canadienne and Brown Swiss breeds and the number of registrations seen in recent years suggest that not only the rare breeds are at risk of becoming endangered, but also the genetic diversity in the more popular breeds is being depleted.

References

Cruden, D. 1949. The computation of inbreeding coefficients in closed populations. J. Hered.40, 248-251.

Emik, L.O. & Terrill, C.E. 1949. Systematic procedures for calculating inbreeding coefficients. J. Hered. 40, 51-55.

Fortin, L. de G. 1940. Histoire de la race bovine Canadienne. La Bonne Terre, Ecole Supérieure d’Agriculture, Ste-Anne de La Pocatière, Québec, Canada, pp. 286.

Gowe, R.S., Robertson, A. & Latter, B. D.H. 1959. Environment and poultry breeding problems. 5. The design of poultry control strains. Poult. Sci. 38, 462-471.

Hansen, C., Bernier, J-.G. & Shrestha, J. N.B. 2000. Broadening of the genetic base in Canadienne cattle. Proc. Fourth Global Conf. Conserv. Domest. Anim. Genet. Resour., Nepal Agricultural Research Council and Rare Breeds International, Kathmandu, Nepal, 152-155.

Ibi, T., Moriya, K., Matsumoto, M., Koba, S. & Sasaki, Y. 1997. Population structure of the Japanese Brown cattle. J. Anim. Breed. Genet. 114, 43-48.

Lush, J.L. 1948. The genetics of populations. Mimeo., Iowa State University, Ames, Iowa.

Miglior, F. & Burnside, E.B. 1995. Inbreeding of Canadian Holstein cattle. J. Dairy Sci. 78, 1163-1167.

Miglior, F., Szkotnicki, B. & Burnside, E.B. 1992. Analysis of levels of inbreeding and inbreeding depression in Jersey cattle. J. Dairy Sci. 75: 1112-1118.

O’-Huigin, C. & Cunningham, E.P. 1990. Analysis of breeding structure of the Kerry breed. J. Anim. Breed. Genet. 107, 452-457.

Rutter, C.R. & Pearson, R.E. 1981. Breeding structure in artificial insemination Holsteins. J. Dairy Sci. 64 (Suppl. 1), 74 (Abstr.).

Takayanagi, S., Moriya, K., Nomura, T., Dohgo, T. & Sasaki, Y. 1996. Population structure of Japanese black cattle in Hyogo prefecture. Anim. Sci. Tech., Japan 67, 286-290.

Wright, S. 1931. Evolution in Mendelian populations. Genetics 16, 97-159.

Young, C.W. 1984. Inbreeding and the gene pool. J. Dairy Sci. 67, 472-477.

Young, C.W., Bonczek, R.R. & Johnson, D.G. 1988. Inbreeding and relationship among registered Holsteins. J. Dairy Sci. 71, 1659-1666.

Young, C.W., Tyler, W.J., Freeman, A. E., Voelker, H.H., McGilliard, L. D. & Ludwick, T.M. 1969. Inbreeding investigations with dairy cattle in the North Central region of the United States. North Central Regional Publ. 191, University of Minnesota, St. Paul, Minnesota.