S1040: Genetic Selection and Crossbreeding to Enhance Reproduction and Survival of Dairy Cattle (S-284)
Statement of Issues and Justification
Over the past century, dairy cows have been largely bred and selected for milk production traits (milk yield, fat and protein content) and conformation. The application of genetic/selection tools to the US dairy population has resulted in tremendous genetic improvement in milk yield, an estimated change in genetic merit for milk production of 7,122 pounds per lactation in Holsteins from birth years 1957 through 2004 (http://aipl.arsusda.gov/eval/summary/trend.cfm). Similar genetic improvement has been documented for Ayrshire (4,372), Brown Swiss (5,546), Guernsey (6,011), Jersey (6,630) and Milking Shorthorn (4,253) breeds. The genetic improvement has been responsible for over 50% of the total improvement observed in milk yield. It is reasonable to assume that advanced management practices (i.e., housing, veterinary care, improved nutrient supply, feed mixing and delivery, etc.) are responsible for the remainder. However, emphases on selection and management for higher yields have negatively affected reproductive and survival traits of different breeds in general and the Holstein breed in particular. Decline in conception rates, fertility and survival and incremental increase in number of services, number of days open and susceptibility to diseases, induce higher maintenance and management costs for the already struggling dairy farmers. Specific evidence shows an undesirable decrease in genetic merit for daughter pregnancy rate of from 2.8 to 7% for these dairy breeds and an increase of from .05 to .2 in somatic cell score. Average inbreeding in 2006 was 5.3% for US Holsteins and 7.1% for US Jerseys and has increased in the last 10 years at an annual rate of .14% and .20%, respectively in these breeds. Even though adjustments for inbreeding are considered in national animal model genetic evaluations (at least partially through previous efforts of this regional project), increases in average inbreeding for these breeds cannot continue without substantial inbreeding depression for cow fertility and health.The current S-1008 project focuses on selection/crossbreeding as genetic avenues to improve dairy cattle performance. The proposed project will continue that focus but with emphasis on field data on reproduction, morbidity, mortality, somatic cell score, and body condition score to develop new selection/management tools and criteria to enhance reproduction and survival in dairy herds.
Dairy producers are turning to crossbreeding as a solution to health and fertility problems with purebreds and in response to increased market emphasis on milk components. The economic justification for crossbreeding has not been well established under modern management systems or with several breeds of cows that have experienced dramatic genetic change since crossbreeding trials of 30-40 years ago. Crossbreeding trials, currently underway in S-1008, are evaluating calving ease, heifer growth, reproduction, immune function, and reproduction, survival, health and lactation yields of lactating cows. These trials will provide the data to test the lifetime economic value of crossbred dairy cows against purebreds, primarily Holstein, under both research herd and commercial conditions and allow determination of the optimum combination of additive genetic merit and non-additive genetic merit in improving lifetime economic performance.
Multi-trait breeding goals and genetic indexes allow producers to combine traits of high economic importance into sound breeding practices. This is an evolving process as critical new traits become available for inclusion in indexes. Prior and current work of the S-1008 project has contributed significantly to the Lifetime Net Merit Dollar (LNM$) index, a measure of lifetime profit, published on US dairy cows and bulls by the Animal Improvement Programs Laboratory (AIPL) of USDA. Most recent enhancements of this index came from collaboration of AIPL and other S-1008 cooperators through revision of economic weights and credits for productive life and inclusion of genetic evaluations for stillbirths. More accurate measures of overall genetic merit will be developed and released to dairy cattle breeders, and incomes and expenses associated with new traits will be estimated. Changes in economic values of existing traits will be monitored to reexamine the hypothesis that current selection policies should continue. Future research will identify new/indicator traits that should be added to national selection indexes. Of particular interest will be performance and economics of crossbred animals in regard to fitness and health traits. Resulting indexes will enable producers and industry personnel to make intelligent use of genetic evaluations on an expanding array of traits and large numbers of bulls and cows available for selection. A major objective of continued regional research collaboration is to give more detailed definition of lifetime economic performance of dairy cattle and ways to use the indexes to bring about change in the traits encompassed by it.
Today's commercial dairy farmers are seeking to improve the health/fertility/survival of their cattle through genetic selection. Systems for genetic evaluation of length of productive life, daughter pregnancy rate, somatic cell score, and direct and maternal components of calving ease and stillbirths have been implemented, and development of systems for genetic evaluation of resistance to specific infectious diseases and metabolic disorders is underway. As such, within-breed selection for health/fertility/survival is now relatively straightforward. However, development of systems for across-breed selection is lagging due to lack of information regarding breed means and heterosis parameters for many key health and fertility traits, particularly for European dairy breeds. The adoption of crossbreeding at an increasing rate apparently seeks to improve these health and fitness traits. At present, selection indices provided to US dairy farms are based on relative net income over opportunity costs (RNIOC), which reflect net lifetime income per animal. Indices based on economic efficiency, which reflects net return per dollar invested, may be more appropriate for genetic improvement programs that involve a combination of purebred selection and crossbreeding. The tools developed within the proposed project will allow objective, profit-based comparisons of alternative purebred selection (with attention to avoidance of inbreeding) and crossbreeding programs or combinations thereof. As a result of the proposed project, farmers, extension agents, and breeding advisors will be able to make informed decisions regarding the expected short and long-term performance, risk and net profit of their genetic programs from the protocols recommended.
A search of the CRIS system found only selection and crossbreeding projects involving dairy cattle that were already a part of S-1008. Two multi-state projects were identified that have a reproduction component: NC-1038: FY 07-12 Methods to Increase Reproductive Efficiency in Cattle; S-1023: FY 05-10 Enhancing Production and Reproduction Performance of Heat-Stressed Cattle. Collaboration with NC-1038 is planned as described in the methods of the proposed project for Objective 1. Genetic and environmental factors influencing death rates and sub-optimal reproduction. The genetic element of a project on mastitis resistance NE-1009 Mastitis Resistance to Enhance Dairy Food Safety focuses on the possible genomic identification of genetic variation with potential genetic associations with mastitis and related traits.
Major Accomplishments of S-1008: The S-1008 research group was extremely productive during the most recent 5-year period, with 102 peer-reviewed publications (an average of 7.3 per station) that were directly related to one or more of the project objectives. Many of these were the result of collaborative projects involving multiple stations or agencies, while the remainders were complementary projects that were carried out at individual stations but supported the common objectives of our regional project. It is important to recognize the close collaborations between university researchers in the S-1008 group and government laboratories, breed associations, milk recording agencies, and breeding companies, as these relationships were critical in ensuring implementation and practical application of the research carried out in the present project. Accomplishments corresponding to each of the three project objectives are described below.
Objective 1: Develop selection tools to enhance reproduction and survival using field data.
Our work in Objective 1 focused on three key topics: development of selection tools for improvement of fertility and calving performance, refinement of selection tools for improvement of dairy cow longevity, and investigation of opportunities for genetic evaluation of clinical mastitis and other early postpartum diseases and disorders. The decline in fertility of lactating dairy cows over the past forty years has been extremely costly to US farmers. Female fertility is a complex trait that is influenced by many management and environmental factors, but despite low heritability there exists significant genetic variation between sire families. VanRaden et al. (2004) described the implementation of a national genetic evaluation system for daughter pregnancy rate, a measure of female fertility. This development allowed US dairy farmers, for the first time, to select AI bulls proven to transmit genes leading to enhanced reproductive performance. Refinement of this process continues at the present time.
Recording, editing, and analysis of fertility data poses many unique challenges, as noted by Weigel (2004), who discussed the opportunities and pitfalls associated with selection for improved male and female fertility. Oseni et al. (2003, 2004a) evaluated seasonal differences in days open in US Holsteins and examined the influence of data editing criteria on genetic parameter estimates for days open and pregnancy rate. Differential utilization of reproductive management tools between farms is a key challenge, and Goodling et al. (2005) assessed the impact of hormonal synchronization programs on genetic parameters for female fertility. Caraviello et al. (2006a, 2006b) characterized the reproductive management practices used on large, modern commercial dairy farms and sought to identify management and environmental factors that were associated with poor reproductive performance using machine learning algorithms. The binary nature of fertility data, as well as the censoring of fertility records for cows that have not achieved pregnancy, create additional challenges in genetic evaluation of reproductive traits. Averill et al. (2004, 2006) developed methodology for genetic evaluation of male and female fertility for traits that are measured as a series of binary responses (e.g., success or failure in each of a series of insemination events). Gonzalez-Recio et al. (2005, 2006) assessed the impact of censoring of records. Chang et al. (2007) developed an ordinal censored threshold model in which pregnancy status was assessed in each of a series of 21-day opportunity periods, commencing at the end of the voluntary waiting period (which was also estimated from the data). Calving difficulties (also known as dystocia events) and stillbirths lead to considerable economic losses on US dairy farms, and these traits are particularly problematic in the Holstein breed. Wiggans et al. (2003) and Van Tassell et al. (2003) described estimation of genetic parameters and implementation of a national genetic evaluation system for calving ease that allowed selection of Holstein sires for both the direct and maternal components of dystocia. Johanson and Berger (2003) assessed the potential of birth weight as a predictor of dystocia and stillbirths. Cole et al. (2007a, 2007b) described estimation of genetic parameters for direct and maternal stillbirth rate and implementation of national genetic evaluation system that allowed US dairy farmers identify sire families that transmit superior calving ability. Genetic evaluations for length of productive life have been available since 1994 but, like female fertility, recording and analysis of productive life data poses many unique challenges. Caraviello et al. (2004a, 2004b) investigated the potential for improving genetic evaluations for dairy cow longevity through the use of Weibull proportional hazards models, which can accommodate time-dependent covariates and censored longevity records of cows that are still alive at the time of evaluation. Weigel et al. (2003) investigated factors associated with culling in expanding dairy herds in Wisconsin, and Hare et al. (2006a, 2006b) examined trends in dairy cow survival on US farms, as well as trends in age at first calving and calving interval. Norman et al. (2007a) studied the influence of milk production and fitness traits on the likelihood of culling of Holstein cows during the first three lactations. Management practices and culling policies can change over time, and Tsuruta et al. (2004) noted that estimated genetic correlations between milk yield, body size, udder traits, and productive life of Holsteins had changed over time. Tsuruta et al. (2005) also noted that changes in the definition of productive life, in particular restrictions on length of the lactation, could lead to differences in estimated correlation parameters. Subsequently, VanRaden et al. (2006) and Dematawewa et al. (2007) described procedures for modeling extended lactation records and incorporating months in milk beyond 305 days postpartum into national genetic evaluations for length of productive life.
Genetic evaluations for lactation average somatic cell score, an indicator of mastitis susceptibility, have also been available since 1994. Additional improvement could come from direct selection for resistance to clinical mastitis, which is typically recorded in a binary manner. Nash et al. (2003) estimated genetic parameters for clinical mastitis and assessed relationships with other traits, including somatic cell score, udder conformation, productive life, and protein yield. Zwald et al. (2006) used a multiple-trait threshold model to estimate genetic correlations between liability to clinical mastitis in different lactations and at differing times within the lactation. Additional opportunities may exist for recording and genetic evaluation of other health, fertility, and fitness traits that contribute to overall farm profitability. Norberg et al. (2004) estimated genetic parameters for electrical conductivity of test-day milk samples and also assessed the usefulness of electrical conductivity as an indicator of mastitis susceptibility (Norberg et al. (2006).
Zwald et al. (2004a, 2004b) documented the possibilities for genetic evaluation of susceptibility to key health disorders, including mastitis, ketosis, lameness, displaced abomasums, and metritis, using farmer-recorded incidence data from on-farm herd management software programs. Differences in fitness may be associated with genetic variation in production-related traits such as lactation persistency (Cole and Van Raden, 2006; Appuhamy et al., 2007), maturity rate (Norman et al., 2005, 2007b), or milking speed (Zwald et al., 2005, Wiggans et al., 2007) and on which genetic evaluations may be developed. In addition, Gonda et al. (2006) reported differences in susceptibility to Mycobacterium avium ssp. paratuberculosis (i.e., Johnes disease) between sire families within the Holstein breed.
Objective 2: Explore the impact of crossbreeding on lifetime performance of cows. Dairy cattle breeding programs have traditionally focused on within-breed selection, which can lead to considerable improvement of fertility, calving ability, length of productive life, and other fitness traits over time. However, interest in dairy crossbreeding, which can provide more rapid improvement in such traits, has increased significantly in recent years. As such, our work in Objective 2 focused on evaluation of the performance of crossbred cattle on commercial farms, as well as the creation of crossbred populations in experimental herds.
Using data from commercial dairy herds in California, Heins et al. (2006a, 2006b, 2006c) evaluated fertility, survival, milk production, calving difficulty, and stillbirth rate in Holstein cows, as compared with Normande x Holstein, Montbeliarde x Holstein, and Scandinavian Red x Holstein cows. More recently, Dechow et al. (2007) assessed the lactation performance, udder health, and fertility of Holsteins, Brown Swiss, and Holstein x Brown Swiss crosses on commercial farms. This work culminated in the implementation of a national, multi-breed genetic evaluation system for genetic evaluation of dairy cattle, as described by VanRaden et al. (2007). This system will allow US dairy farmers to practice both within- and across-breed selection for key production and fitness traits and will allow the creation of optimal dairy crossbreeding programs.
In addition to analyses of field data, several universities initiated the development of crossbred resource populations in which novel phenotypes, such as detailed measures of health, fertility, milk composition, and feed efficiency can be assessed. These projects, at the University of Kentucky, the University of Minnesota, Virginia Tech, North Carolina State University and the University of Wisconsin, are ongoing and will be featured in numerous publications in the coming years. Preliminary results from several studies have been published, including an assessment by Kasimanickam et al. (2007) of factors in dairy sire semen that are associated with differences in fertility among Holstein and Jersey sires in the Virginia Kentucky crossbreeding project, as well as an assessment of differences in conception rate, calving performance, and calf health and survival in Holstein x (Holstein x Jersey) crossbred calves, relative to their pure Holstein contemporaries, in the Wisconsin project by Maltecca et al. (2006).
The Virginia-Kentucky-North Carolina State University project is a major collaborative regional effort which has ties to the Minnesota and Wisconsin projects. Joint analyses and publication of results are planned.
Objective 3: Develop breeding goals and appropriate indexes for optimum improvement of health, survival, reproduction, and production. Implementation of national genetic evaluation systems for fertility, calving ease, stillbirth rate, somatic cell score, and productive life, as well as implementation of a multi-breed genetic evaluation system, are critical steps in creating significant and permanent improvement in the fitness of US dairy cattle. However, such tools are of little practical value unless they can be incorporated into properly conceived economic indices, and selection programs based on these indices must also consider maintenance of genetic diversity. Furthermore, investigation of the manner in which the expression or economic value of such traits differs between herd management systems is necessary to ensure that genetic and management improvements go hand-in-hand. Work on these and related topics comprised our contributions in Objective 3, as noted below. VanRaden (2004) reviewed development of the USDA-ARS Animal Improvement Programs Laboratorys Lifetime Net Merit index, which is the primary total merit index used by US dairy farmers. A key development in the formulation of economic indices in the past 5 years has been revision of our views regarding selection for dairy form. Dechow et al. (2002, 2004a) noted that body condition scores were associated with differences in milk yield and reproductive performance. Dechow et al. (2004b) computed genetic correlations between body condition score, dairy form scores, and health traits using data from two countries, whereas Dechow et al. (2003, 2004b) assessed correlations between body condition score, dairy form, and other conformation traits and estimated genetic parameters for body condition score and dairy form at various ages and stages of the lactation. Heat stress leads to significant economic losses on US dairy farms, particularly in the Southeast. Ravagnolo and Misztal (2002a, 2002b) examined the impact of heat stress on non-return rate in Holstein cattle using temperature-humidity index data from nearby weather stations. Later, Oseni et al. (2004) estimated genetic parameters for days open in models that allowed differential losses in performance between sire families due to heat stress. More recently, Bohmanova et al. (2007) investigated possibilities for using temperature-humidity index data to account for differential losses in milk yield between sire families due to heat stress. Similarly, genotype by environment interactions between confinement and pasture-based systems could lead to selection of certain sire families for performance in specific environments, a possibility that was investigated by Boettcher et al. (2003), Kearney et al. (2004a, 2004b), and Fahey et al. (2007). Lastly, several recent studies addressed inbreeding depression and maintenance of genetic diversity. Cassell et al. (2003a) quantified the importance of complete pedigree information when evaluating the impact of inbreeding on dairy cow performance, whereas Cassell et al. (2003b) measured the impact of maternal and fetal inbreeding depression on female fertility traits. Caraviello et al. (2003) examined the impact of inbreeding on dairy cow longevity in Jersey cattle using a proportional hazards model. Vallejo et al. (2003) assessed genetic diversity and linkage disequlibrium in US Holstein cattle. More recently, Adamec et al. (2006) assessed the impact of inbreeding on dystocia and stillbirth rate, whereas VanRaden and Miller (2006) investigated the impact of dominance, inbreeding, and inherited defects on embryonic loss in dairy cattle. Gulisija et al. (2007) examined the impact of inbreeding on the performance of Jersey cows using nonparametric methods.
Summary: In summary, the last 5 years marked a watershed period with respect to the use of genetic selection and crossbreeding to improve the health, fertility, and survival of US dairy cattle. Work of this committee contributed directly to the development and implementation of a multi-breed genetic evaluation system that can be used by US dairy farmers who wish to improve fitness traits through crossbreeding. Likewise, the work of this committee contributed directly to the development and implementation of national genetic evaluation systems for daughter pregnancy rate, maternal calving ease, and direct and maternal stillbirth rate, as well as a significant improvement in the genetic evaluation system for length of productive life. Furthermore, major modifications to Lifetime Net Merit, the economic index use by most US dairy farmers for routine sire selection decisions, have their roots in the work of this research group. Based on this work, US dairy farmers now have a vast array of genetic tools at their disposal for improving the health, fertility, and survival of their cattle.
Back to Top