Fitter happier: the never-ending quest for a better cow

Authors


A clear challenge is now facing the food animal industry. We must devise methods for producing more food, using fewer inputs and minimizing environmental impact and at the same time ensure the welfare of animals. In this, dairy products provide an efficient and sustainable approach to meet the global food demand because of their efficient production, high nutritional value, diverse manufacturing capabilities and palatability. Achievement of these goals is too often in direct conflict with the short-term needs of farmers. If the contribution of genetic improvement to higher dairy productivity has been essential, with annual increases per milk produced of approximately 200lbs achieved in most of Western Europe and North America [Windig et al. (2005) J. Dairy Sci., 88, 335–347], dairy operations remain a low profit margin industry with the main (and often only) source of income tied to the price of milk. Increasing the net profit of the farmers remains an inescapable step towards the creation of a competitive and sustainable agriculture. Improving productivity has been the major goal of nearly all dairy cattle breeding programmes for a long time. This has neglected fitness and fertility parameters that are related to decreased costs of production and have a large impact on the farms' net profitability. Individual cow diseases particularly are associated with increased culling, loss of production and labour costs [Zwald et al. (2004) J. Dairy Sci., 87, 4287–4294; Hansen (2000) J. Dairy Sci., 83, 1145–1150]. Improving animal health needs to be a top priority in the dairy industry both from an economic as well as an ethical standpoint. In this, genetic selection for improved health will grant a permanent improvement in performance and profitability.

Genetic improvement of traits related to survival has for the large part focused on longevity. Currently, only seven countries incorporate direct health information into their selection programme [Steine et al. (2008) J. Dairy Sci., 91, 418–426]. There has, however, been a growing interest in the health data, and the International Committee for Animal Recording's Functional Traits Working Group released a document detailing recommendations and best practices for the collection of producer-recorded health event data [Cole et al. (2012) J. Dairy Sci., 95 (Suppl. 2), 443]. Several obstacles remain to the widespread routine implementation of selection programmes for health traits.

From a scientific perspective, the broad definition of ‘disease’ or ‘direct health’ traits makes little sense. The heterogeneity and complexity of these traits needs to be dealt with in much greater detail than has been done so far. The dynamics between resistance (the ability of a cow to avoid getting sick altogether) and tolerance (the ability of coping with a disease and maintaining reasonable production) need to be considered [Bishop et al. (2012) Front. Genet., 3, 114], and the interrelation among different diseases and between these and production practices is far from being completely elicited [Appuhamy et al. (2009) J. Dairy Sci., 92, 1785–1795; Parker Gaddis et al. (2012) J. Dairy Sci., 95, 5422–5435]. However, while the effort in understanding and refining statistical methods and biological understanding of health traits is vital, this cannot be separated from the development of meaningful and applicable tools for farmers and organizations. While selecting breeding animals to produce the next generation with known improvements for direct health traits represents the ultimate goal, it is important to understand that given the nature of most of these traits, a large role in the improvement of cow health and the reduction of disease incidence will be played by the managing elements of animal husbandry.

While there is a virtually endless pool of phenotypes that could be potentially considered for selection, there needs to be a concerted effort in establishing a selection programme to identify a few key parameters for which a consistent and demonstrable improvement can be achieved, to avoid the risk of mixed bag of results undermining the perception of the usefulness of selecting for such traits. Moreover, the inherent discrepancy between what is conceivably and reliably measured in the field and the real underlying traits needs to be acknowledged.

Special attention needs to be paid by both the scientific community and the industry in communicating what is realistically attainable through selection. Over-optimistic representation of selection efforts will ultimately undermine the credibility of any programme. Improvement has been recently made for several functional traits like SCS, stillbirth, dystocia and fertility. For some of these, the bottoming of a dangerous decline already represents a significant achievement. Realizing that for health traits, the short-term gain might be small and possibly limited to the curbing of a general decline is essential.

There is an intrinsic heterogeneity of players, and a complex infrastructure in the collection and flow of information connected to health traits. On-farm computer management systems provide an efficient means for collecting health-related data for genetic analysis. These records currently provide one of the few, if not the only, opportunity for direct selection for disease resistance for countries where recording of health disorders is not mandatory [Zwald et al. 2004 J. Dairy Sci., 87, 4295–4302]. Furthermore, even where veterinary data collection is in place the data are seldom utilized as the only source of direct health information [Cole et al. (2012) J. Dairy Sci., 95 (Suppl. 2), 443]. A lively scientific discussion on the helpfulness and quality of the data is undergoing [Koeck et al. (2012) J. Dairy Sci., 95, 4099–4108]. Nevertheless, among the reasons for the slow implementation of health selection programmes, data privacy concerns are at the top of the list. Yet, discussions on this issue are often carried out behind closed doors. A frank and transparent approach is needed in addressing this theme, additionally keeping in mind that the collection, entry and correction of the data has often apparent as well as hidden additional burdens and costs for the farmers that need to be acknowledged.

Genomic selection is largely redefining the notion of what traits we can select for. The success of genomic selection programmes in dairy is often used as a yardstick to gauge the gain achievable with the inclusion of molecular information in a selection programme. Yet, it is often forgotten that this success hinges on highly accurate conventional estimation of breeding values based on hundreds or thousands of progeny for sires that have been genotyped. Genomic selection can be tremendously rewarding for health traits, but its success cannot be granted without building a growing and integrated flow of phenotypic information. It cannot be stressed enough that a few scattered mid-sized resource populations cannot replace a concerted and continuous effort.

Concerns about health and fertility in dairy cattle are widespread, and a general consensus exists on placing more emphasis on selection for health and welfare. In spite of this, the inclusion of factual information on selection programmes remains scarce, particularly for disease resistance traits. The time has come for all the different players, scientists, AI organizations, breed societies and private companies, to seize the opportunity for a broader collaborative effort in making the next step in selection for a profitable healthy cow possible.

Comments and discussions with John B. Cole at USDA-AIPL are gratefully acknowledged.

Ancillary