ATTENTION on NOVEL TRAITS: Needed Or Novelty

ATTENTION ON NOVEL TRAITS: Needed or Novelty

46th Beef Improvement Federation Annual Meeting and Research Symposium

Co-hosted by the University of Nebraska–Lincoln Department of Animal Science, the US Meat Animal Research Center (USMARC), and the Nebraska Cattlemen.

The University of Nebraska–Lincoln is an equal opportunity educator and employer. TABLE OF CONTENTS

Schedule of Events...... 4-6

General Sessions Speaker Biographies ...... 8-11

Clay Mathis Lorna Marshall Dan Shike Donagh Berry Rick Funston Holly Neibergs Kim Vonnahme Raluca Mateescu Tom Field Susan Duckett Donnell Brown Galen Erickson J.D. Radakovich Harvey Freetly

2014 BIF Commercial Producer of the Year Award Nominees...... 12-13

BIF Commercial Producer of the Year Past Award Recipients...... 14

2014 BIF Seedstock Producer of the Year Award Nominees...... 15-17

BIF Seedstock Producer of the Year Award Past Recipients...... 18

Pioneer Award Past Recipients...... 19-21

Continuing Service Award Past Recipients...... 22-23

Ambassador Award Past Recipients...... 24

BIF Travel Felloweship Recipients...... 25

Frank Baker Memorial Scholarship...... 26-27

Frank Baker Memorial Scholarship Award Recipient Heather Bradford...... 28-35

Frank Baker Memorial Scholarship Award Recipient Xi Zeng...... 36-43

Frank Baker Memorial Scholarship Award Past Recipients...... 44

Roy A. Wallace Memorial Fund...... 45

Roy A. Wallace Memorial Fund Past Recipients...... 46

Tours ...... 47

2 Proceedings...... 48-159

General Session I Papers...... 48-97

Economic Considerations For The Cow Herd, Clay Mathis...... 48

Heifer Intake and Efficiency as Indicators of Cow Intake and Efficiency, Dan Shike...... 50

Beef Heifer Development and Lifetime Productivity, Rick Funston...... 56

The Long-Lasting Impact of Nutrition: Developmental Programming, Kim Vonnahme...... 62

General Session II Papers...... 68-115

Selection for Novel Traits: An International Genomics Perspective, Donagh Berry...... 68

Economic Benefits of Using Genetic Selection to Reduce the Prevalence of Bovine Respiratory Disease Complex in Beef Feedlot Cattle, Holly Neibergs...... 82

It Is Possible to Genetically Change the Nutrient Profile of Beef, Raluca Mateescu...... 87

Changes in Dietary Regime Impact Fatty Acid Profile of Beef, Susan Duckett...... 93

Improving Feed Efficiency in the Feedlot: Opportunities and Challenges, Galen Erickson...... 101

Relationship Between Selection for Feed Efficiency and Methane Production, Harvy Freetly...... 112

Technical Committes...... 116-159

Healthfulness of Beef: A Genome-Wide Association Study Using Crossbred Cattle, Cashley Ahlberg and Lauren Schiermiester...... 116

Breeding for Reduced Environmental Footprint in Beef Cattle, Donagh Berry...... 126

Across-Breed EPD Tables for the Year 2014 Adjusted to Breed Differences for Birth Year of 2012, Larry Kuehn...... 134

Mean EPDs Reported by Different Breeds, Larry Kuehn...... 155

Sponsors...... 160-165

3 SCHEDULE ATTENTION ON NOVEL TRAITS: OF EVENTS Needed or Novelty

11:00 The Long Lasting Impact of Nutrition: Developmental Programming June 18 Kim Vonnahme, North Dakota State University 12:00-7:00 Registration 11:30 Merging Genetics and Management for Improved Profitability— 1:00-4:00 Board Meeting Tom Field, Moderator 5:00-6:00 Welcome Reception Panelists: Donnell Brown, 6:00-9:00 USMARC Symposium: 50 Years of JD Radakovich, Lorna Marshall Service to the Beef Industry 12:30-2:30 Awards Luncheon History of the Germplasm Evaluation Commercial Producer of the Year Award Project: Pioneer Award Past: Larry Cundiff Continuing Service Award Present: Mark Thallman The Genomics Era: Steve Kappes 2:30-5:30 Technical Breakout Sessions The USMARC / UNL partnership: Advancements in Cowherd Efficiency Ronnie Green and Adaptability Panel Discussion 2:30-3:15 Mike MacNeil, Delta G, — The Challenge: Seedstock: Bill Rishel Evaluating Cow Lifetime Productivity and Genomics: Dave Nichols Efficiency Commercial: Burke Teichert 3:15-4:00 Mike Davis, The Ohio State University — Feedlot: Chuck Folken Measuring Cow Efficiency and Productiv- ity: What Do We Know from the Research? 4:00-4:45 Larry Kuehn, USDA MARC — Cow Efficiency/Productivity: June 19 The MARC Perspective 4:45-5:30 Scott Speidel, CSU — National Cattle Evaluation: Approaches to Cow 6:00 a.m. - Registration Productivity and Efficiency 6:00 p.m. Advancements in Emerging Technology 8:00-12:30 General Session I: Focus on the Cowherd 2:30-2:45 Ronnie Green, University of Nebraska– Lincoln — Following the Yellow Brick 8:00 Welcome -- Lt. Gov. Heidemann, Road of Beef Cattle Genomics – 25 Years Ronnie Green, Steve Whitmire of Perspective 8:30 Economic Considerations for Profitable 2:45-3:15 Daniel Pomp — The Evolution of Cowherds Commercial DNA Analysis in the Cattle Clay Mathis, King Ranch Institute Industry 9:15 Heifer Intake and Feed Efficiency as 3:15-3:45 Michael Bishop, Illumina — Accelerating Indicators of Cow Intake and Efficiency Agrigenomics: The Business of Cattle Dan Shike, University of Illinois Genetics 10:00 Break 3:45-4:15 Matt Spangler, University of Nebraka– 10:30 Decreasing Costs through Improved Lincoln — Using Genomics to Heifer Development Strategies Pick High-hanging Fruit: Rick Funston, University of Nebraska– Integrated Projects Update Lincoln

4:15-4:45 Scott Fahrenkrug, Recombinetics — 9:00 It is Possible to Genetically Change the Molecular Breeding to Accelerate Livestock Nutrient Profile of Beef Improvement Raluca Mateescu, University of Florida 4:45-5:15 Harvey Blackburn, National Center for 9:30 Changes in Dietary Regime Impact Fatty Genetic Resources Preservation —Design Acid Profile of Beef and Function of a Genomics Database for Susan Duckett, Clemson University the Cattle Industry 10:00 Break Advancements in Selection Decisions* 10:30 Improving Feed Efficiency in the Feed- Advancements in Producer Applications* lot: Opportunities and Challenges Brief statement on BIF’s roll in standardization and what we Galen Erickson, University of hope to accomplish in session Nebraska–Lincoln 2:45-3:30 Bruce Golden, California Polytechnic State 11:00 Relationship Between Selection for Feed University —ERTs for the ‘New Beef Efficiency and Methane Production Industry’ Harvey Freetly, US Meat Animal Research Center 3:30-4:15 Dan Moser, Kansas State University — Don't Blame the Bull: Rethinking 11:30 Wrap up Contemporary Groups Starting at or Before 11:45 Annual meeting, regional caucuses, Conception election of directors 4:15-5:00 Producer Panel: EPD Wish List and 12:30-2:30 Awards Luncheon Farm-level Data Collection Challenges Seedstock Producer of the Year Award 5:00-5:30 Breed Association Panel: Performance Data Frank Baker Award Wish List Now and in the Future, Response Roy Wallace Scholarship Award Ambassador Award to Producer Panel Presidents Address * Committee sessions will be joined Elections 6:00-10:00 p.m. Dinner at Lincoln Station 2:30-5:30 Technical Breakout Sessions Advancements in Live Animal, Carcass and End Product 2:30-3:00 Ken Odde, Kansas State University — Diminishing Beef Cow Research Herds … June 20 Their Role in the Past and Their Role in the Future 6:00 a.m.- Registration 3:00-3:30 Tommy Wheeler, Meat Animal Research 6:00 p.m. Center, Clay Center, NE — Meat Quality 8:00-12:30 General Session II: Focus on the Feedlot Research at MARC 8:00 Selection for Novel Traits: An International 3:30-4:15 John Gonzalez, Kansas State Universi- Genomics Perspective ty — Measuring and Quantifying the Donagh Berry, Animal & Grassland Role of Collagen Crosslinks in Research and Innovation Centre, Beef Tenderness Teagasc, Moorepark, Ireland 4:15-4:45 Lauren Schiermiester & Cashley 8:30 Economic Benefits of using Genetic Ahlberg, University of Nebraska– Selection to Reduce the Prevalence Lincoln — Healthfulness of Beef: of Bovine Respiratory Disease Complex in A Genome Wide Association Study Beef Feedlot Cattle Using Crossbred Cattle. Holly Neibergs, Washington State University 5 SCHEDULE OF EVENTS

4:45-5:30 Donagh Berry, Animal & Grassland Research and Innovation Centre, Teagasc, Moorepark, Ireland — Breeding for Reduced Environmental Footprint Advancements in Genetic Prediction 2:30-3:00 Dorian Garrick, Iowa State University — Opportunities and Challenges for a New Approach to Genomic Prediction 3:00-3:45 Breed Association Representatives — Updates on Implementation of Genom- ic-enhanced National Cattle Evaluation 3:45-4:00 Bruce Golden, California Polytechnic State University — Eliminating the approximation bias in NCE accuracy computations with high-performance Gibbs Sampling 4:00-4:30 Mark Thallman, US Meat Animal Research Center — Things that Annoy Me About National Cattle Evaluation

5:30-8:00 Board Meeting

June 21

Post Conference Tour 6:45 a.m. Depart hotel Circle Five Feedlot US Meat Animal Research Center GeneSeek 5:00 p.m. Arrive back to hotel

6 7 GENERAL SESSION SPEAKERS

CLAY MATHIS was named of Illinois. Shike is an Assistant Professor in Animal Scienc- Director and Endowed Chair of the es at the University of Illinois and is responsible for teaching King Ranch® Institute for Ranch and research in beef cattle nutrition and management in Management in July, 2010. As Di- addition to serving as the coordinator for the Livestock and rector, Mathis leads faculty and staff Meats judging teams. appointed to the King Ranch® Institute for Ranch Management and oversees Shike’s research is focused on identifying nutritional teaching and outreach efforts of the strategies and management practices that improve efficien- Institute. He maintains and develops cy, reproduction and profitability in beef cow/calf produc- curriculum for the M.S. in Ranch Management degree tion. Specifically, his work focuses on understanding the program, which includes more than 42 hours of business relationship of intake, measures of efficiency, and methane and animal production coursework and intensive project production in the developing heifer and measures of intake work tackling issues on large partnering ranches across and efficiency in the mature cow. He also evaluates how the United States. Mathis works closely with the KRIRM nutrition and management of the cow during gestation and Management Council to identify topics and speakers for the lactation not only impacts the reproductive performance of entire suite of KRIRM lectureships and the annual Holt Cat® the cow but also what impacts this may have on the devel- Symposium on Excellence in Ranch Management. oping fetus and early postnatal life.

Mathis is a native of New Braunfels, TX. He received a RICK FUNSTON is a professor B.S. in Animal Science and M.S. in the Physiology of Repro- and Reproductive Physiologist at the duction from Texas A&M University. In 1998, he earned a University of Nebraska–Lincoln. He Ph.D. from Kansas State University in Ruminant Nutrition received his B.S. from North Dakota where his research focused on supplementing grazing cattle. State University, M.S. from Montana From 1998 to 2010, Dr. Mathis worked as a Professor and State University his Ph.D. from the Extension Livestock Specialist at New Mexico State Univer- University of Wyoming, and complet- sity. ed a Post Doc at Colorado State Uni- versity in Reproduction/Biotechnolo- Outside of his professional activities, he and his family gy. He divides his time between extension and research. His have fed cattle in growing, finishing, and cull-cow feeding research on lighter heifer development is receiving national enterprises. Mathis and his wife, Rhonda, have three chil- attention/adoption; research on fetal programming effects dren: Morgan, Miles, and Amy Kaye. on postnatal calf performance including carcass characteristics and reproduction has received national and interna- DAN SHIKE lives in Sadorus, tional recognition; and he is a team member of nationally Illinois with his wife, Jennifer and 3 recognized beef systems research. In the extension capacity, children; Olivia, Hunter, and Harper. he provides leadership and subject matter expertise for Shike grew up on a diversified grain educational programs in cow-calf production management and livestock operation in western for the West Central District and statewide expertise in beef Illinois. Shike’s family owns and reproductive management programs operates Shike Cattle Company and he is actively involved with his father and brother in the management and marketing of the cattle. Shike received his A.S. degree from Black Hawk College – East Campus, his B.S. from Kansas State University, and his M.S. and Ph.D. from the University

8 KIMBERLY VONNAHME DONNELL BROWN and his received her B.S. degree in Animal wife Kelli are the fifth generation to Science from Iowa State University, own and manage the R.A. Brown M.S. at Oklahoma State University, Ranch in Throckmorton, TX, a family and Ph.D. from the University of Wy- business since 1895. They raise reg- oming. Kim moved to North Dakota istered Angus, Red Angus and Si- State University for a post-doctorate mAngus cattle and sell 600 bulls each program in 2003 and accepted an October. assistant professor position in the Department of Animal Sciences in 2004. She was promoted The ranch’s #1 goal is to improve the profitability and to Associate Professor in 2010. Kim served as the Co-Di- sustainability of their commercial customers. The R.A. rector of the Center for Nutrition and Pregnancy from 2009 Brown Ranch has been honored with the NCBA Cattle until April 2012. Vonnahme’s research programs focuses Business of the Century Award as well as being named BIF on the impacts of maternal nutrition on fetal and placental Seedstock Producer of the Year. development in sheep and cattle. More specifically, Kim is interested in how the maternal nutrition impacts uteropla- The strength of their program is shown by the high cental blood flow, development of the placenta, and nutrient percentage of repeat commercial bull customers, as well as transfer. She has generated over $1.8 million in research by having more than 25 bulls in major AI studs. Their trade- grants, has 100 peer review publications, > 210 abstracts, 3 mark is in their unique bull development system that tests book chapters, and 1 patent. their bulls for performance, efficiency and carcass superior- ity while developing them for longevity with the use of high Kim is married to Michael Kangas and they have 2 forage diets on the rocky hills of west Texas. children, Katie and Joe. Donnell is a graduate of Texas Tech University. Prior to TOM FIELD is a passionate that, he served as President of the Texas FFA & as the Na- advocate for education, agriculture, tional FFA President. He has served in a Strategic Planning free enterprise, engaged citizenship, capacity for four different breed associations as well as the and the potential of young people. He National Cattlemen’s Beef Association. Donnell’s wife Kelli serves the people of Nebraska as the served as the National FFA President in 1988 and President Director of the Engler Agribusiness of the Red Angus Association of America in 2009 & 2010. Entrepreneurship Program and holder of the Engler Chair in Entrepreneur- J.D. RADAKOVICH was ship at the University of Nebraska– raised on a purebred and composite Lincoln. He is also a noted agricultural author with works seedstock cattle operation in Iowa. including his column “Out of the Box” and featured com- Radakovich earned a Bachelor’s de- mentator of “The Entrepreneurial Minute” on the Angus gree in Animal Science from Colora- Report on RFD-TV. He is the author of two agricultural text do State University and then worked books which have been adopted in both domestic and inter- for nine months on ranch stations national markets. A frequent speaker at agricultural events throughout eastern Australia. Follow- in the U.S. and abroad, he has consulted with a number of ing that, Radakovich spent ten years agricultural enterprises and organizations, and has served in northern Nevada working on ranches before completing on numerous boards related to education, agriculture, and the King Ranch Institute for Ranch Management Masters athletics. He is the co-owner of Field Land and Cattle Com- Degree program. Currently, J.D. is the Manager of the pany, LLC in Colorado. He and his wife Laura watch over Hoodoo Ranch in Cody, Wyoming. a brood that includes one year old twins, a set of twins in college, and one starting his career in Teach for America. 9 GENERAL SESSION SPEAKERS HOLLY NEIBERGS is an As- sociate Professor at Washington State University in the Department of LORNA MARSHALL grew up Animal Sciences where she conducts on a small registered Simmental and research investigating the genetic alfalfa hay operation near Wichita, basis of complex traits in cattle such Kansas. Following graduation from as disease (bovine paratuberculosis, Kansas State University and Colorado bovine viral diarrhea-persistent infec- State University, Marshall served as tion, and bovine respiratory disease), Director of Performance Programs feed efficiency, and reproduction. She received a B.S. and and Youth Activities for the American M.S. (reproduction) in Animal Science at Washington State Gelbvieh Association (1993-1995), University and a Ph.D. at Texas A&M University in genet- Manager of Beef Sire Acquisition for ABS Global, DeForest, ics. She completed a post-doctoral fellowship at the Nation- WI (1995-2011), and is currently the U.S. Beef Marketing al Animal Disease Center in Ames, Iowa prior to her tenure Manager for Genex Cooperative, Shawano, WI where she at the University of Louisville College of Medicine where works with Genex’s Large Herd Beef Initiative. she studied the genetic basis of hereditary cancers and directed the Norton Hereditary Cancer Institute which offered Lorna and her husband, Troy, are first generation genetic testing and counseling to families with histories of ranchers with a 300 head Angus and SimAngus seedstock cancer. Neibergs returned to livestock genetics in 2007. operation that holds an annual bull sale each March. Both Troy and Lorna have been active in various beef industry RALUCA MATEESCU recently organizations and associations. The Marshall’s ranch near joined the faculty in the Department of Burlington, CO along with their children (and work force), Animal Science at University of Flori- Wyatt, Justis, and Wynn. da, after serving on the Animal Science faculty at Oklahoma State University DONAGH BERRY is a part- for 7 years. As an Associate Professor time beef and sheep farmer but also a of Genetics at Oklahoma State Univer- principal investigator in quantitative sity, she developed a research program genetics at the semi-state research focused on applying the most recent center, Teagasc, Moorepark in Ireland. genomic technologies to improving animal production He is responsible for the research on efficiency and enhancing animal products for improved genetics in dairy and beef cattle and human health. She has also dedicated much of her time to has recently become involved in sheep incorporating the latest genomic discoveries in teaching, at and plant genetics. His main interests both undergraduate and graduate level, to ensure that the are in the derivation of breeding goals, genetic and genomic student population is well prepared to become participants evaluations, statistics, decision support tools and breeding in the genetic revolution and informed users or consumers programs. of biotechnology. She received a B.S. degree in Molecular Biology and Genetics from Bucharest University, Romania and received her M.S and Ph.D. in Animal Breeding and Genetics from Cornell University.

10 SUSAN DUCKETT is currently a HARVEY FREETLY is the Re- Professor in the Department of Animal search Leader for the Nutrition and and Veterinary Sciences at Clemson Environmental Management Re- University where she holds The Ernest search Unit at the USDA, Agricultural L. Corley, Jr. Trustees Endowed Chair. Research Service, U.S. Meat Animal She received her B.S. degree in Animal Research Center in Clay Center, Ne- Science from Iowa State University, braska. In 1990, he joined USDA after and M.S. and Ph.D. degrees in Animal receiving his Ph.D. in Nutrition from Science from Oklahoma State Univer- the University of California – Davis. sity. She held faculty positions at the University of Idaho His research has focused on nutrient management during and University of Georgia prior to her appointment at heifer development, pregnancy, and lactation. He is cur- Clemson University. Duckett’s research integrates ruminant rently conducting research to determine the role of nutri- nutrition and meat science to alter lipid metabolism, fatty tion on developmental programming of heifers, and the acid composition and palatability of animal products. microbiome of the gastro-intestinal tract of cattle that differ in feed efficiency. GALEN ERICKSON is the Nebraska Cattle Industry Professor of Animal Science in the Department of Animal Science at the Universi- ty of Nebraska–Lincoln, as well as Beef Feedlot Extension Specialist. He received his Ph.D. in 2001 from the University of Nebraska. Research and extension activities focus on utilization of byproducts for growing and finishing beef cattle, utilization of alternatives to grain for finishing cattle, the interaction between nutrition, management, and environmental issues including air quality and nutrient management, and growth promoters that include use of implants, feed additives, and beta-agonists. Along with graduate students, we have published approximately 275 Nebraska Beef Report articles and over 90 scientific journal articles over the past 13 years.

11 AWARD COMMERCIAL PRODUCER NOMINEES

CB Farms Family Partnership Owners and Managers: Berry, Carla and Brandon Bortz Preston, Kansas

A week after Berry and Carla Bortz graduated from Kansas State University in 1982, they got married and began farming in eastern Pratt County near Preston, Kansas. They started with two irrigated circles, two dryland quarters and about 300 acres of grass, which they used to background calves. Their family began in 1986 with the birth of their son Brandon, followed by their daughter, Amber, in 1987 and son, Darnell, in 1991. As the kids grew, so did the farm. In 2001, they started their cowherd and reduced the number of cattle purchased per year. Brandon and his wife, Cari, returned to the farm in 2012. Today, they farm 19 irrigated circles, 2,500 acres of dryland and 2,000 acres of native grass. They have 550 spring-calving cows, of which 150 are registered Black Angus. They also operate a 1,500 head feed yard. They grow corn, wheat, soybeans, milo, sunflowers, cotton, alfalfa, bermuda and feed. They finish all their calves at home along with some calves they purchase from their bull customers. The calves are marketed through U.S. Premium Beef (USPB). They believe that if the beef industry is going to survive and be something besides a niche in the protein market, operating costs must be reduced in all sectors of the industry. Their mission is to deliver a desirable product to the consumer from which they can derive an acceptable standard of living. To do this, they are going to control or participate in as many practices as they can from solar interception to product delivery on the plate. The Kansas Livestock Association is proud to nominate CB Farms Family Partnership. Hansen Family Ranches Owners: Circle Ranches – Ed and Marilyn Hansen Lone Pine Livestock – Carl and Debbie Hansen Quarter Circle Lazy H Ranch – Chris and Janeth Hansen D Dart Ranch – Cheri and Scott Dent Managers: Hansen Family Livermore, Colorado

The Hansen Family, from Livermore Colorado, is a fifth-generation family ranch originally purchased in 1940 by Sam and Castor Hansen; father and grandfather of Ed Hansen and grandfather and great-grandfather of Carl, Cheri, and Chris. The ranch headquarters have been located in Livermore, CO since 1940. In 1994, the Hansen family acquired a summer ranch in Grover, CO, replacing summer forest permits. The Grover Ranch consists of approximately 15,000 acres and provides reliable pasture and more flexible management. Currently, the herd consists of 250 head of Angus cattle. Three years ago, due to extreme drought, the family was forced to reduce one third of the herd. They raise grass hay, between 700-800 tons, to use for winter feed and sell for supplemental income. They also raise replacement heifers. The calving season runs between January 15th and March 15th. What makes the Hansen’s unique is that they implement a 100% AI breeding program. They raise their own clean-up bulls and purchase high quality bulls which they collect semen from and then use for inseminating the herd. This allows them to spread bull costs over more cows, and consequently, pay a higher dollar for quality sires. The cow herd spends from June 1st through October 15th at the Grover ranch where calves are sold and shipped. The cows spend the balance of the year at the home ranch in Livermore for calving and breeding. There are three generations involved in showing livestock at various expositions throughout the state. Hansen Family Ranches is proudly nominated by the Colorado Cattlemen’s Association and Colorado State University.

12 James Kean Owner/Manager: James Kean Louisa, Virginia

James Kean runs a 300 head fall-calving cow/calf beef operation in Louisa, Virginia. James has a reputation for having high quality beef cattle with his calves always bringing a premium in the local state graded sales or Central Virginia Cat- tlemen Association (CVCA) special tele-auction sales. James has always paid special attention to maintaining high quality genetics in his herd through the use of AI with his heifers, as well as using top quality bulls many times from the Beef Cattle Improvement Association (BCIA) sales on the cows. The cow herd consists primarily of Angus and black baldy cows. James also raises small grain and corn row crops on his operations, which is used primarily for feed resources in his cattle operation. James is a leader for the local farming community serving on several boards of directors. James has been an active member of the Louisa Farm Bureau Board of Directors for over 25 years serving as President for several years. He has been a director on the Thomas Jefferson Soil and Water Conservation District for the past eleven years. James served on the -Or ange Madison Coop Board of Directors for 20 years. He has been a member of the CVCA since it started in the late 1980’s and served on the board of directors since January of 2011. James is also a 4-H volunteer and has helped with the Louisa Agriculture Fair for the past 15 years. James is married to Dr. Kate Hussman who recently retired from being a large animal veterinarian in Louisa County. James has two sons Brian and John. The Virginia Beef Cattle Improvement Association is proud to nominate James Kean. XA Cattle Company Owners: Bill, Marie and Levi Farr Manager: Bill Farr Moorefield, Nebraska

Bill, Marie and Levi Farr have lived at their current location in Lincoln County Nebraska for the past 7 years. Before that, they lived between Stockville, and Farnam, but, ran their cows there in the summer time. Currently, their operation consists of around 700 Balancer® cows and 150 registered Hereford cows; they also farm 2,000 acres of a combination of dry and irrigated corn, soybeans, and wheat. Their heifers start calving in February and the cows the last week of March. In November they gather their cattle out of the hills and begin weaning the calves. The calves are backgrounded at their headquarters in their feedlot and then sold the end of February or the first week of March, depending on weather and markets. The cows run on winter pasture until they are worked. They are then moved to stocks until calving time. The replacement heifers are grown and developed there at the ranch and developed to gain 1.5 pounds per day. XA Cattle Company is proudly nominated by the American Gelbvieh Association.

13 PAST AWARD COMMERCIAL PRODUCER OF THE YEAR RECIPIENTS

Name State Year Name State Year

Darnall Ranch, Inc. Nebraska 2013 Bob & Sharon Beck Oregon 1984 Maddux Cattle Company Nebraska 2012 Al Smith Virginia 1983 Quinn Cow Company Nebraska 2011 Sam Hands Kansas 1982 Downey Ranch Kansas 2010 Henry Gardiner Kansas 1981 JHL Ranch Nebraska 2009 Jess Kilgore Montana 1980 Kniebel Farms and Kansas 2008 Bert Hawkins Oregon 1979 Cattle Company Mose Tucker Alabama 1978 Broseco Ranch Colorado 2007 Mary & Stephen Garst Iowa 1977 Pitchfork Ranch Illinois 2006 Ron Baker Oregon 1976 Prather Ranch California 2005 Gene Gates Kansas 1975 Olsen Ranches, Inc. Nebraska 2004 Lloyd Nygard North Dakota 1974 Tailgate Ranch Kansas 2003 Pat Wilson Florida 1973 Griffith Seedstock Kansas 2002 Chan Cooper Montana 1972 Maxey Farms Virginia 2001 Bill & Claudia Tucker Virginia 2000 Mossy Creek Farm Virginia 1999 Giles Family Kansas 1999 Mike & Priscilla Kasten Missouri 1998 Randy & Judy Mills Kansas 1998 Merlin & Bonnie Kansas 1997 Anderson Virgil & Mary Jo Kansas 1996 Huseman Joe & Susan Thielen Kansas 1995 Fran & Beth Dobitz South Dakota 1994 Jon Ferguson Kansas 1993 Kopp Family Oregon 1992 Dave & Sandy Oregon 1991 Umbarger Mike & Diana Hopper Oregon 1990 Jerry Adamson Nebraska 1989 Gary Johnson Kansas 1988 Rodney G. Oliphant Kansas 1987 Charles Fariss Virginia 1986 Glenn Harvey Oregon 1985

14 AWARD SEEDSTOCK PRODUCER NOMINEES

Marshall Cattle Company Owners/Managers: Troy and Lorna Marshall Burlington, Colorado

Marshall Cattle Company is a first generation Angus and SimAngus seedstock operation that includes 270 registered females located in Burlington, Colorado. Their mission statement is “to provide revolutionary genetic solutions that provide value and maximize profits for our customers. We are dedicated to ensuring a thriving beef industry for the next generation while conducting ourselves in a manner that reflects our faith in God.” They purchased their ranch 13 years ago and recently hosted their 10th annual production sale in eastern Colorado. They have both a spring and fall breeding program to spread out labor and the use of high quality clean-up herd sires. Their program relies heavily on AI and ET to multiply the impact of their best genetics and high accuracy, balanced trait sires that excel for the economically-relevant traits of beef production. Marshall Cattle Company is the home of “SuperMamas”; to them, that means females that are designed to thrive in the harsh eastern Colorado environment under commercial conditions. Their females must be moderate framed, easy fleshing, structurally sound, good dispositioned, nice uddered, and reproductively efficient with appropriate levels of milk. Recent- ly, they have collected feed efficiency and PAP data in addition to the more traditional traits. The labor force on the ranch consists of Troy and Lorna Marshall and their three children, along with a spring intern. Because they are a first generation operation, they know they would not be where they are today without the help and support of many neighbors and friends. Each year at the beginning of their bull sale, they present the “Above and Beyond” award to someone who has truly gone above and beyond the call of duty to make their success possible – friends, neighbors, extension agents, veterinarians, customers, cooperators, and this year their auctioneer! They know their future success depends on customer satisfaction, honesty and integrity. Repeat customers, long-standing relationships and genetic value are what they strive to produce. Marshall Cattle Company is proudly nominated by the Colorado Cattlemen’s Association and Colorado State Universi- ty. Schuler Red Angus Owner: The Darrell Schuler Family Manager: Butch Schuler Bridgeport, Nebraska

Schuler Red Angus is located in the panhandle of western Nebraska on Lawrence Fork Creek. Darrell and Mary Lou began raising commercial Herefords there in 1959. By the 1970’s, they were using Red Angus bulls and discovered that the crossbred calves were superior to the ranch’s traditional straight-bred cattle. A registered Red Angus herd was started in 1976 to develop genetics for the ranch’s commercial herd and provide seedstock for neighboring ranches. The seedstock herd expanded in the 1980s and continued to improve through the use of artificial insemination, performance-testing and a data-based breeding program utilizing EPDs. Customer input and feedback from meat-packers regarding the ranch’s finished commercial cattle encouraged the Schuler’s to begin collecting carcass data in 1991. They later developed structured carcass tests utilizing their own and customer cattle. Today, over 25% of Red Angus’ high-accuracy carcass trait sires have been proven by Schuler Red Angus. A composite seedstock herd called “Schuler Reds” was started in 1992 which utilizes Red Angus, Simmental and Gelbvieh genetics giving Schuler Red Angus customers the opportunity to add heterosis and breed complementarity via a simple crossbreeding system. The current ranching operation encompasses 17,000 acres including 2,000 acres of private pasture leases and 1,250 acres of irrigated farm ground. Butch and Susan Schuler with their children Stephanie and David manage the operation of approximately 1,000 head of spring calving females. The Schuler’s hosted their 32nd production sale this spring selling 150 registered Red Angus and Schuler Red composite bulls. The Nebraska Cattlemen and the Red Angus Association of America are proud to nominate SchulerRed Angus.

15 AWARD SEEDSTOCK PRODUCER NOMINEES

Shelton Angus Farm Owners/Managers: W.H. “Buddy” Shelton Gretna, Virginia

Shelton Angus Farm is a family operated registered Angus seedstock operation in Gretna, Virginia. The farms are located in Pittsylvania County which is historically one of the largest tobacco producing regions in the southeast. The registered herd was established in 1963 by Walter H. and Ruby Shelton. Management of the cattle became the responsibility of W.H. “Buddy” Shelton Jr. in 1988 after he returned to home post-graduation from Virginia Tech. One hundred twenty brood cows are maintained on an all fescue grazing system. The herd is exclusively fall calving, which is typical in south-central Virginia. The herd has been on a 100% A.I. breeding program since 1988. Bulls are developed collaborative with other seedstock breeders in the region, and historically marketed through the Virginia BCIA program, until five years ago when Shelton Angus initiated their own annual fall bull sale. Genetic improvement in the herd centers on functionality and adaptability to the region’s fescue environment, along with economically relevant traits to their feeder cattle-producing customers. Embryo transfer and genomics are key technologies which have assisted in the advancement of the herd. Additionally, Shelton Angus focuses on customer service by providing group backgrounding and marketing opportunities to their clients, as well as facilitating retained ownership and collection of carcass data which benefits both their customers’ and their own breeding programs. Buddy Shelton has been an active leader in agriculture, including serving as President of Virginia Angus, President of the Pittsylvania County Cattleman’s Association, along with being engaged in Farm Bureau, 4-H and youth groups, and his local church. Shelton Angus Farm is proudly nominated by the Virginia Beef Cattle Improvement Association. Wedel Red Angus Owners/Managers: Frank and Susan Wedel Leoti, Kansas

Wedel Red Angus is located in the short-grass country known as the High Plains of western Kansas. Frank and Susan Wedel own the seedstock operation, which is located 15 miles northwest of Leoti, KS. Their introduction to Red Angus began in 1989 when they purchased their first Red Angus bulls to help solve calving-ease problems in their commercial cowherd. They purchased their first Red Angus heifers in 1990 and began selling bulls in 1993. The first few years they sold 30 to 40 bulls per year private treaty. In 2001, they held their first production sale. At that time, they determined that to be a viable seedstock supplier they needed to expand their business. They sold their commercial cows and focused on the seedstock business. This year they will sell more than 140 bulls and make 500 matings. Their cowherd includes Red Angus as well as Sim/Red Angus and Char/Red Angus hybrids. The mature cows graze year-round northeast of Wallace, KS, using an intensive rotational grazing program. The young heifers spend their first two years at Leoti until their first calf is weaned, then join the mature cows. Each year, they purchase steer and heifer calves from their bull customers. The steers are finished at commercial feed yards. The heifers are developed for replacements. Those that don’t make the cut are finished and carcass and performance data are collected and shared with producers. They sell 150 to 160 heifers in their production sale each year and about that many are bred and sold in the fall. Their motto is “Knowing our customers is our highest priority because their success becomes our success.” Wedel Ranch is proudly nominated by the Kansas Livestock Association.

16 Wells Farm Owner/Manager: Mike Wells Selma, Alabama

Wells Farm, owned by Dr. and Mrs. Mike Wells, has been producing registered cattle in Dallas County, Alabama for 23 years on farmland purchased by Mike’s grandparents in 1942. The original commercial cattle were mostly Angus based bred to Hereford bulls. The value of the Simmental breed was quickly seen when introduced in the late 1970’s. Today, the Wells Farm cattle herd consists of approximately 90 breeding females, with 60% purebred Simmental and 40% SimAngus. Calves are born from late August until October to produce superior yearling bulls for commercial cattlemen. EPD’s, carcass ultrasound and performance data are all heavily relied on to select herd sires, whether AI or clean up. All Purpose and Terminal Indices and carcass EPD’s are of particular importance in the selection process. Relationships with contemporary Simmental breeders have been built over the years, whose opinions are also valued in sire selection. As a full time veterinarian and the sole employee of Wells Farm, easy keeping cattle with good dispositions are a priority. Wells Farm has participated in seven different Alabama BCIA bull evaluations and marketing opportunities, but currently market almost exclusively by private treaty, with a select few going to the Wiregrass BCIA Forage Based Bull Evalua- tion each year. A favorite part of being in the Simmental business is being able to meet commercial cattlemen from all over the southeast and let them select their own bull in a relaxed environment. Many of their customers prefer to purchase bulls this way, and Wells Farm has been fortunate to sell a high percentage of bulls to repeat customers each year. Future goals of Wells Farm are to continue to replace older females with top replacement heifers to accelerate genetic improvement and to continue to strive to produce the best Simmental and Sim-Angus bulls. The Alabama Beef Cattle Improvement Association is proud to nominate Wells Farm.

17 PAST AWARD SEEDSTOCK PRODUCER OF THE YEAR RECIPIENTS

Name State Year Name State Year

Bradley 3 Ranch Texas 2013 J. David Nichols Iowa 1993 V8 Ranch Texas 2012 Leonard Wulf & Sons Minnesota 1992 Mushrush Red Angus Kansas 2011 Summitcrest Farms Ohio 1991 Sandhill Farms Kansas 2010 Douglas & Molly Hoff South Dakota 1990 Harrell Hereford Ranch Oregon 2009 Glynn Debter Alabama 1989 Champion Hill Ohio 2009 W.T. “Bill” Bennett Washington 1988 TC Ranch Nebraska 2008 Henry Gardiner Kansas 1987 Pelton Simmental Red Kansas 2007 Leonard Lodoen North Dakota 1986 Angus Ric Hoyt Oregon 1985 Sauk Valley Angus Illinois 2006 Lee Nichols Iowa 1984 Rishel Angus Nebraska 2005 Bill Borror California 1983 Camp Cooley Ranch Texas 2004 A.F. “Frankie” Flint New Mexico 1982 Moser Ranch Kansas 2003 Bob Dickinson Kansas 1981 Circle A Ranch Missouri 2002 Bill Wolfe Oregon 1980 Sydenstricker Angus Missouri 2001 Jim Wolf Nebraska 1979 Farms James D. Bennett Virginia 1978 Fink Beef Genetics Kansas 2000 Glenn Burrows New Mexico 1977 Morven Farms Virginia 1999 Jorgenson Brothers South Dakota 1976 Knoll Crest Farms Virginia 1998 Leslie J. Holden Montana 1975 Flying H Genetics Nebraska 1998 Jack Cooper Montana 1975 Wehrmann Angus Ranch Virginia 1997 Carlton Corbin Oklahoma 1974 Bob & Gloria Thomas Oregon 1997 Mrs. R. W. Jones, Jr. Georgia 1973 Frank Felton Missouri 1996 John Crowe California 1972 Tom & Carolyn Perrier Kansas 1995 Richard Janssen Kansas 1994 R.A. “Rob” Brown Texas 1993

18 PAST AWARD PIONEER AWARD RECIPIENTS

Name State Year Name State Year

Keith Bertrand Georgia 2013 Frank Felton Missouri 2004 Ignacy Misztal Georgia 2013 Tom Jenkins Nebraska 2004 Glenn Selk Oklahoma 2013 Joe Minyard South Dakota 2004

Sally Buxkemper Texas 2012 George Chiga Oklahoma 2003 Donald Franke Louisiana 2012 Burke Healey Oklahoma 2003 Leo McDonnell Montana 2012 Keith Zoellner Kansas 2003

Mike Tess Montana 2011 H.H. “Hop” Dickenson Kansas 2002 Mike MacNeil Montana 2011 Martin & Mary Jorgensen South Dakota 2002 Jerry Lipsey Montana 2011 L. Dale Van Vleck Nebraska 2002

Richard McClung Virginia 2010 Larry Benyshek Georgia 2001 John and Bettie Rotert Missouri 2010 Minnie Lou Bradley Texas 2001 Daryl Strohbehn Iowa 2010 Tom Cartwright Texas 2001 Glen Klippenstein Missouri 2010 J. David Nichols Iowa 2000 Bruce Golden California 2009 Harlan Ritchie Michigan 2000 Bruce Orvis California 2009 Robert R. Schalles Kansas 2000 Roy McPhee (posthu- California 2009 mously) Joseph Graham Virginia 1999 John Pollak New York 1999 Donald Vaniman Montana 2008 Richard Quaas New York 1999 Louis Latimer Canada 2008 Harry Haney Canada 2008 John Crouch Missouri 1998 Bob Church Canada 2008 Bob Dickinson Kansas 1998 Douglas MacKenzie Canada 1998 Rob Brown Texas 2007 Fraser David and Emma Danciger Colorado 2007 Jim Gosey Nebraska 2007 Larry V. Cundiff Nebraska 1997 Henry Gardiner Kansas 1997 John Brethour Kansas 2006 Jim Leachman Montana 1997 Harlan & Dorotheann Mississippi 2006 Rogers A.L. “Ike” Eller Virginia 1996 Dave Pingrey Mississippi 2006 Glynn Debter Alabama 1996

Jack and Gini Chase Wyoming 2005 James S. Brinks Colorado 1995 Jack Cooper Montana 2005 Robert E. Taylor Colorado 1995 Dale Davis Montana 2005 Les Holden Montana 2005 Don Kress Montana 2005 19 PAST AWARD PIONEER AWARD RECIPIENTS

Name State Year Name State Year

Tom Chrystal Iowa 1994 Mick Crandell South Dakota 1985 Robert C. DeBaca Iowa 1994 Mel Kirkiede North Dakota 1985 Roy A. Wallace Ohio 1994 Bill Graham Georgia 1984 James D. Bennett Virginia 1993 Max Hammond Florida 1984 M.K. “Curly” Cook Georgia 1993 Thomas J. Marlowe Virginia 1984 O’Dell G. Daniel Georgia 1993 Hayes Gregory North Carolina 1993 Jim Elings California 1983 Dixon Hubbard Virginia 1993 W. Dean Frischknecht Oregon 1983 James W. “Pete” Patterson North Dakota 1993 Ben Kettle Colorado 1983 Richard Willham Iowa 1993 Jim Sanders Nevada 1983 Carroll O. Schoonover Wyoming 1983 Frank Baker Arkansas 1992 Ron Baker Oregon 1992 Gordon Dickerson Nebraska 1982 Bill Borror California 1992 Mr. & Mrs. Percy Powers Texas 1982 Walter Rowden Arkansas 1992 F.R. “Ferry” Carpenter Colorado 1981 Bill Long Texas 1991 Otha Grimes Oklahoma 1981 Bill Turner Texas 1991 Milton England Texas 1981 L.A. Maddox, Jr. Texas 1981 Donn & Sylvia Mitchell Canada 1990 Charles Pratt Oklahoma 1981 Hoon Song Canada 1990 Clyde Reed Oklahoma 1981 Jim Wilton Canada 1990 Richard T. “Scotty” Clark Colorado 1980 Roy Beeby Oklahoma 1989 Bryon L. Southwell Georgia 1980 Will Butts Tennessee 1989 John W. Massey Missouri 1989 Robert Koch Nebraska 1979 Mr. & Mrs. Carl Roubicek Arizona 1979 Christian A. Dinkle South Dakota 1988 Joseph J. Urick Montana 1979 George F. & Mattie Ellis New Mexico 1988 A.F. “Frankie” Flint New Mexico 1988 James B. Lingle Maryland 1978 R. Henry Mathiessen Virginia 1978 Glenn Burrows New Mexico 1987 Bob Priode Virginia 1978 Carlton Corbin Oklahoma 1987 Murray Corbin Oklahoma 1987 Max Deets Kansas 1987

Charles R. Henderson New York 1986 Everett J. Warwick Maryland 1986

20 Name State Year

Mick Crandell South Dakota 1985 Ralph Bogart Oregon 1977 Mel Kirkiede North Dakota 1985 Henry Holsman South Dakota 1977 Marvin Koger Florida 1977 Bill Graham Georgia 1984 John Lasley Missouri 1977 Max Hammond Florida 1984 W. L. McCormick Georgia 1977 Thomas J. Marlowe Virginia 1984 Paul Orcutt Montana 1977 J.P. Smith Missouri 1977 Jim Elings California 1983 H.H. Stonaker Colorado 1977 W. Dean Frischknecht Oregon 1983 Ben Kettle Colorado 1983 Forrest Bassford Colorado 1976 Jim Sanders Nevada 1983 Doyle Chambers Louisiana 1976 Carroll O. Schoonover Wyoming 1983 Mrs. Waldo Emerson Wyoming 1976 Forbes Gordon Dickerson Nebraska 1982 C. Curtis Mast Virginia 1976 Mr. & Mrs. Percy Powers Texas 1982 Glenn Butts Missouri 1975 F.R. “Ferry” Carpenter Colorado 1981 Keith Gregory Nebraska 1975 Otha Grimes Oklahoma 1981 Braford Knapp, Jr. Montana 1975 Milton England Texas 1981 L.A. Maddox, Jr. Texas 1981 Reuben Albaugh California 1974 Charles Pratt Oklahoma 1981 Charles E. Bell, Jr. Virginia 1974 Clyde Reed Oklahoma 1981 John H. Knox New Mexico 1974 Paul Pattengale Colorado 1974 Richard T. “Scotty” Clark Colorado 1980 Fred Wilson Montana 1974 Bryon L. Southwell Georgia 1980 Ray Woodward Montana 1974

Robert Koch Nebraska 1979 Jay L. Lush Iowa 1973 Mr. & Mrs. Carl Roubicek Arizona 1979 Joseph J. Urick Montana 1979

James B. Lingle Maryland 1978 R. Henry Mathiessen Virginia 1978 Bob Priode Virginia 1978

21 PAST AWARD CONTINUING SERVICE RECIPIENTS

Name Year Name State Year Lisa Kriese-Anderson Alabama 2006 Ben Eggers Sydenstricker Ge- 2013 Dave Notter Virginia 2006 netics Brian House Select Sires 2013 Jerry Lipsey Montana 2005 Lauren Hyde American Simmental 2013 Michael MacNeil Montana 2005 Association Terry O’Neill Montana 2005 Jerry Taylor University of Mis- 2013 Robert Williams Missouri 2005 souri Jack Ward American Hereford 2013 Chris Christensen South Dakota 2004 Association Robert “Bob” Hough Texas 2004 Steven M. Kappes Nebraska 2004 Tom Field Nebraska 2012 Richard McClung Virginia 2004 Stephen Hammack Texas 2012 Brian McCulloh Wisconsin 2012 Sherry Doubet Colorado 2003 Larry Olson South Carolina 2012 Ronnie Green Colorado 2003 Connee Quinn Nebraska 2003 Tommy Brown Alabama 2011 Ronnie Silcox Georgia 2003 Mark Enns Colorado 2011 Joe Paschal Texas 2011 S.R. Evans Mississippi 2002 Marty Ropp Montana 2011 Galen Fink Kansas 2002 Bob Weaber Missouri 2011 Bill Hohenboken Virginia 2002 Bill Bowman Missouri 2010 William Altenburg Colorado 2001 Twig Marston Nebraska 2010 Kent Andersen Colorado 2001 David Patterson Missouri 2010 Don Boggs South Dakota 2001 Mike Tess Montana 2010 Ron Bolze Kansas 2000 Darrh Bullock Kentucky 2009 Jed Dillard Florida 2000 Dave Daley California 2009 Renee Lloyd Iowa 2009 Bruce Golden Colorado 1999 Mark Thallman Nebraska 2009 John Hough Georgia 1999 Gary Johnson Kansas 1999 Doug Fee Canada 2008 Norman Vincil Virginia 1999 Dale Kelly Canada 2008 Duncan Porteous Canada 2008 Keith Bertrand Georgia 1998 Richard Gilbert Texas 1998 Craig Huffhines Missouri 2007 Burke Healey Oklahoma 1998 Sally Northcutt Missouri 2007 Glenn Brinkman Texas 1997 Jimmy Holliman Alabama 2006 Russell Danielson North Dakota 1997

22 Name State Year Name State Year Gene Rouse Iowa 1997 James Bennett Virginia 1984 M.K. Cook Georgia 1984 Doug L. Hixon Wyoming 1996 Craig Ludwig Missouri 1984 Harlan D. Ritchie Michigan 1996 Art Linton Montana 1983 Paul Bennett Virginia 1995 Pat Goggins Montana 1995 J.D. Mankin Idaho 1982 Brian Pogue Canada 1995 Mark Keffeler South Dakota 1981 Bruce E. Cunningham Montana 1994 Loren Jackson Texas 1994 Glenn Butts Missouri 1980 Marvin D. Nichols Iowa 1994 Jim Gosey Nebraska 1980 Steve Radakovich Iowa 1994 C.K. Allen Missouri 1979 Doyle Wilson Iowa 1994 William Durfey Missouri 1979 Robert McGuire Alabama 1993 James S. Brinks Colorado 1978 Charles McPeake Georgia 1993 Martin Jorgensen South Dakota 1978 Henry W. Webster South Carolina 1993 Paul D. Miller Wisconsin 1978 Jack Chase Wyoming 1992 Lloyd Schmitt Montana 1977 Leonard Wulf Minnesota 1992 Don Vaniman Montana 1977 John Crouch Missouri 1991 A.L. Eller, Jr. Virginia 1976 Robert Dickinson Kansas 1990 Ray Meyer South Dakota 1976

Roger McCraw North Carolina 1989 Larry V. Cundiff Nebraska 1975 Dixon D. Hubbard Virginia 1975 Bruce Howard Canada 1988 J. David Nichols Iowa 1975

Bill Borror California 1987 Frank H. Baker Oklahoma 1974 Jim Gibb Missouri 1987 D.D. Bennett Oregon 1974 Daryl Strohbehn Iowa 1987 Richard Willham Iowa 1974

Larry Benyshek Georgia 1986 F. R. Carpenter Colorado 1973 Ken W. Ellis California 1986 Robert DeBaca Iowa 1973 Earl Peterson Montana 1986 E.J. Warwick Maryland 1973

Jim Glenn Iowa 1985 Clarence Burch Oklahoma 1972 Dick Spader Missouri 1985 Roy Wallace Ohio 1985

23 PAST AWARD AMBASSADOR RECIPIENTS

Name Publication State Year

A.J. Smith Oklahoma Cowman Magazine Oklahoma 2013 Burt Rutherford BEEF Magazine Texas 2012 Jay Carlson BEEF Magazine Kansas 2011 Larry Atzenweiler and Missouri Beef Cattlemen Missouri 2010 Andy Atzenweiler Kelli Toldeo Cornerpost Publications California 2009 Gren Winslow and Larry Canadian Cattleman Magazine Canada 2008 Thomas Angie Denton Hereford World Missouri 2007 Belinda Ary Cattle Today Alabama 2006 Steve Suther Certified Angus Beef LLC Kansas 2005 Kindra Gordon Freelance Writer South Dakota 2004 Troy Marshall Seedstock Digest Missouri 2003 Joe Roybal BEEF Magazine Minnesota 2002 Greg Hendersen Drovers Kansas 2001 Wes Ishmael Clear Point Communications Texas 2000 Shauna Rose Hermel Angus Journal & BEEF Magazine Missouri 1999 Keith Evans American Angus Association Missouri 1998 Bill Miller Beef Today Kansas 1997 Ed Bible Hereford World Missouri 1996 Nita Effertz Beef Today Idaho 1995 Hayes Walker III America’s Beef Cattleman Kansas 1994 J.T. “Johnny” Jenkins Livestock Breeder Journal Georgia 1993 Dick Crow Western Livestock Journal Colorado 1991 Robert C. DeBaca The Ideal Beef Memo Iowa 1990 Forrest Bassford Western Livestock Journal Colorado 1989 Fred Knop Drovers Journal Kansas 1988 Chester Peterson Simmental Shield Kansas 1987 Warren Kester BEEF Magazine Minnesota 1986

24 RECIPIENTS TRAVEL SCHOLARSHIP

Name University Kristi Allwardt Oklahoma State University Ryan Boldt Colorado State University Heather Bradford Kansas State University Miranda Culbertson Colorado State University Erika Downey Texas A & M University Kara Marley South Dakota State University Jamie Parham South Dakota State University Kelli Retallick Kansas State University Jason Warner University of Nebraska–Lincoln Xi Zeng Colorado State University

25 MEMORIAL FRANK H. BAKER SCHOLARSHIP

Dr. Frank Baker is widely recognized as the “Founding Father” of the Beef Improvement Federation (BIF). Frank played a key leadership role in helping establish BIF in 1968, while he was Animal Science Department Chairman at the University of Nebraska, Lincoln, 1966-74. The Frank Baker Memorial Scholarship Award Essay competition for graduate students provides an opportunity to recognize outstanding student research and competitive writing in honor of Dr. Baker. Frank H. Baker was born May 2, 1923, at Stroud, Oklahoma, and was reared on a farm in northeastern Oklahoma. He received his B.S. degree, with distinction, in Animal Husbandry from Oklahoma State University (OSU) in 1947, after 2ó years of military service with the US Army as a paratrooper in Eu- rope, for which he was awarded the Purple Heart. After serving three years as county extension agent and veterans agriculture instructor in Oklahoma, Frank returned to OSU to complete his M.S. and Ph.D. degrees in Animal Nutrition. Frank’s professional positions included teaching and research positions at Kansas State University, 1953-55; the University of Kentucky, 1955-58; Extension Livestock Specialist at OSU, 1958-62; and Extension Animal Science Programs Coordinator, USDA, Washington, D.C., 1962-66. Frank left Nebraska in 1974 to become Dean of Agriculture at Oklahoma State University, a position he held until 1979, when he began service as International Agricultural Programs Officer and Professor of Animal Science at OSU. Frank joined Winrock International, Morrilton, Arkansas, in 1981, as Senior Program Officer and Director of the International Stockmen’s School, where he remained until his retirement. Frank served on advisory committees for Angus, Hereford, and Polled Hereford beef breed associations, the National Cattlemen’s Association, Performance Registry International, and the Livestock Conservation, Inc. His service and leadership to the American Society of Animal Science (ASAS) included many committees, election as vice-president and as president, 1973-74. Frank was elected an ASAS Honorary Fellow in 1977, he was a Fellow of the American Association for the Ad- vancement of Science, and served the Council for Agricultural Science and Technology (CAST) as president in 1979. Frank Baker received many awards in his career, crowned by having his portrait hung in the Saddle and Sirloin Club Gallery at the International Livestock Exposition, Louisville, Ken- tucky, on November 16, 1986. His ability as a statesman and diplomat for the livestock industry was to use his vision to call forth the collective best from all those around him. Frank was a “mover and shaker” who was skillful in turning “Ideas into Action” in the beef cattle performance movement. His unique leadership abilities earned him great respect among breeders and scientists alike. Frank died February 15, 1993, in Little Rock, Arkansas.

FRANK H. BAKER

Born: May 2, 1923, Stroud, Oklahoma Died: February 15, 1993, Little Rock, Arkansas

Frank H. Baker

Photograph of portrait in Saddle and Sirloin Club

Gallery – EvereƩ Raymond Kinstler, ArƟst

167 27 AWARD FRANK BAKER MEMORIAL SCHOLARSHIP RECPIENT

THE EFFECT OF QUANTITY AND BREED COMPOSITION OF GENOTYPES FOR GENOMIC PREDICTION IN PUREBRED OR CROSSBRED CATTLE Heather Bradford1 1Kansas State University, Manhattan

Introduction the intent that all QTL affecting a trait are in LD with at least 1 marker (Hayes and Goddard, 2010). The implementation of genomics enabled producers to more accurately select young animals for Because of the LD between SNP and QTL, the breeding resulting in a decrease in generation interval. association between SNP and QTL affecting a trait Beef cattle typically have long generation intervals of interest can be used to create genomic predictions. compared with species like poultry and swine, and These genomic predictions are the sum of the effect genomic selection can increase response to selec- of each SNP on the trait of interest. Across-breed LD tion. Genomic selection should have the most benefit is much more restricted than within-breed LD due to for traits that are hard to measure, measured late in differential selection since the divergence of individu- life, sex-limited, and measured after harvest (Hayes al breeds (Hayes and Goddard, 2010). Because of the and Goddard, 2010). Traits like female fertility are difference in LD across breeds, genomic predictions sex-limited and difficult to measure while being eco- historically needed to be breed-specific. The accuracy nomically relevant to producers. Selection for female of genomic predictions was largely the result of LD, fertility would benefit greatly from the inclusion of and the loss of LD resulted in less accuracy in sub- genomic data to increase accuracy. There are many sequent generations (Habier et al., 2007). Thus, SNP economically relevant traits that beef producers could effects have to be periodically re-estimated because of better select for by using genomics. the erosion of LD. A population of genotyped animals with phe- Review of Literature notypes or very accurate breeding values is typically used for estimating SNP effects. This group is referred to as a training or reference population while a sepa- Linkage rate group of animals with genotypes and phenotypes A quantitative trait loci (QTL) is a gene that or breeding values is the validation population. The affects a quantitative trait. Phenotype results from the SNP effects estimated from the training population are total of the effects of all QTL including dominance used to predict genetic merit in the validation popula- and any gene interactions, environment, and the inter- tion. The accuracy of the genomic prediction, referred action of genetics and environment. A single nucleo- to as a genomic breeding value (GBV), direct genomic tide polymorphism (SNP) is a single base difference value (DGV), or molecular breeding value (MBV), in a DNA sequence that may or may not be located can then be assessed in the validation population, within a gene. Linkage disequilibrium (LD) results because the validation population was independent of when a SNP allele and QTL allele are linked and in- the training population that was used to develop the herited together more often than expected (Hayes and predictions. Genetic correlations between the genomic Goddard, 2010). Meusissen et al. (2001) first proposed prediction and phenotypic trait data can be estimated genomic selection using all SNP markers simultane- with a two trait animal model using REML (Kachman, ously. This method relies on dense SNP panels with 2008). The square of this genetic correlation is the

28 FRANK BAKER MEMORIAL SCHOLARSHIP

percent of the additive genetic variance that was Accuracies continue to improve as the refer- explained by the genomic test (Thallman et al., 2009). ence population grows from 1,000 to 1,000’s of indi- Genomic results are then incorporated into traditional viduals. When combining 3 Nordic Red populations genetic evaluations resulting in genomic-enhanced with individual reference populations of 1,562 animals expected progeny differences (GE-EPD). or fewer, reliabilities increased by a magnitude of 3 to 8% on average with a total reference population of 3,735 animals (Brøndum et al., 2011). Increasing the Number of Genotyped Animals training population from 1,300 to 5,250 animals while As more animals are genotyped, researchers using the same methodology resulted in predictions can better estimate SNP effects resulting in more ac- that on average explained 18% more genetic variation curate genomic predictions. Simulations with various and increased accuracy by 0.40 (Boddhireddy et al., training population sizes and relationships to the vali- 2014). Reliabilities were 5 to 32% greater when using dation population showed accuracy increases when the a combined Chinese and Nordic Holstein population size of the training population increases, even if those of 7,387 instead of only 2,171 Chinese Holsteins animals are many generations removed from the val- (Zhou et al., 2013). Predictions were more accurate idation population (Saatchi et al., 2010). In a study of even when combining populations of the same breed 9 breeds for feed efficiency and carcass traits, breeds from different countries. There has been a consistent with larger training populations had greater accuracies increase in accuracy as more animals were added to than average (Bolormaa et al., 2013). The number of small to moderate sized reference populations. animals in training had a greater impact on lowly her- Research in dairy cattle has evaluated the itable traits, and the relationship to the training popu- impact of larger training populations when there are lation became less important for those traits (Saatchi et many genotyped animals. Reliabilities increased al., 2010). As beef cattle training populations increase, 10% on average when combining European Holstein the greatest impact on accuracy should be for lowly populations to create reference populations with more heritable traits. than 9,000 bulls compared with individual country As the adoption of genomic technology has reference populations with 3,000 to 4,000 bulls (Lund increased, there has been the opportunity to evaluate et al., 2011). Improvements can still be made when the realized increase in accuracy resulting from an reference populations contain several thousand head. increase in the size of the reference population. Di- Adding 3,593 foreign bulls to the U.S. Holstein evalu- rect genomic values for Uruguayan Herefords were ation with over 10,000 genotypes increased reliability more accurate when using predictions for American by 2 to 3% (VanRaden et al., 2012). As the number (n = 1,081) instead of Uruguayan Herefords (n = 395; of genotypes increases, the improvement in accuracy Saatchi et al, 2013). The difference in accuracy likely from a larger reference population isn’t as substantial resulted from the larger training population for the and would be expected to continue to decline. American Hereford prediction and not the relationship Because of the accuracy increase that results to the population used in training. There was a linear from larger training populations, combining geno- increase in accuracy exceeding 0.10 as the size of the typed populations to develop genomic predictions is reference population increased from 500 to 2,500 head of interest. The main focus in the beef industry has of crossbred sheep (Daetwyler et al., 2012b). Howev- been genotyping purebred populations to develop er, the increase in accuracy would not be expected to predictions that are then used within that breed. Due to continue to increase linearly as the size of the refer- the cost of genomic testing and the number of proven ence population continues to increase. Predictions animals needed for training, combining these popu- from small reference populations with fewer than lations could improve prediction accuracy. However, 1,000 individuals become considerably more accurate a very diverse training population could result in less as more animals are included in the reference popula- accurate predictions because the training population is tion. now less related to the individual breeds that are being

29 FRANK BAKER MEMORIAL SCHOLARSHIP predicted. Many simulations have been performed in genetic correlations between GBV and the respective addition to research in beef cattle and other species traits were 0.14 to 0.81 in Angus (Saatchi et al, 2011; to evaluate the impact of the relationship between the Northcutt, 2013; Boddhireddy et al., 2014), 0.18 to training and validation populations. 0.52 in Hereford (Saatchi et al., 2013), 0.39 to 0.76 in Limousin (Saatchi et al., 2012), and 0.29 to 0.65 in Simmental (Saatchi et al., 2012). Generally, there was Relationship between Training and Validation sufficient LD between SNP and genes for traits in- The relationship between the reference animals cluded in national cattle evaluations to achieve strong and the populations that the genomic predictions will genetic correlations. Because of these results, several be used in affects accuracy. Daetwyler et al. (2012a) beef breed associations currently publish GE-EPD. demonstrated that a large proportion of the accuracy Connectedness within-breed can also affect the of predictions results from the strong relationship be- accuracy of predictions for animals that are distantly tween the reference and validation populations. When related to the training population. Genomic predictions the training population consisted of generations that developed for American Herefords were less accurate were more similar to the validation population, predic- when used in Argentinian, Canadian, or Uruguayan tion accuracy was greater than if distant generations Herefords, possibly resulting from lesser relationships were used (Saatchi et al., 2010; Pszczola et al 2012). to the training population or genetic by environment The animals in the more recent generations tend to interactions (Saatchi et al., 2013). When comparing be more related to the young animals in which the reliabilities for Red dairy cattle from 3 European genomic tests are being used. The importance of the countries, the within country predictions were always relationship of the training and validation populations more reliable than if predictions were developed in 1 likely resulted from recombination that took place country and used in the others (Brøndum et al., 2011). between generations and reduced the LD between the Because the reference populations were of similar markers and QTL (Saatchi et al., 2010). As LD erodes, size, the loss in accuracy again resulted from the lack accuracy decreases when a different SNP is associated of connectedness between countries. Further analysis with the QTL than in the reference population. When revealed that the correlation of LD phase between animals had a greater average squared relationship to countries ranged from 0.46 to 0.86 (Brøndum et al., the reference population, those animals had greater 2011). This correlation suggests there has been some reliabilities (Pszczola et al., 2012). If an animal has divergence in the Red breed in these countries which relatives in the reference population, more confidence impacts the ability to develop across-country genom- can be placed on the resulting genomic predictions ic predictions. When using the American Hereford because relatives with similar LD between SNP and prediction in Argentinian Herefords, animals with QTL were used to estimate SNP effects. Because of American Herefords in their pedigree had on average the importance of the relationship between the refer- greater correlations between DEBV and GBV than ence and validation populations on accuracy, there has those without American genetics (Saatchi et al., 2013). been much research on the breed specificity of genom- Similarly, DGV accuracies were less when using An- ic predictions. gus predictions in an Angus herd that was closed for many generations and was less related to the training Within-breed Prediction population (Saatchi et al., 2011). These studies demonstrated the importance of the training population being In the beef industry, much emphasis has been a representative sample of a breed to obtain accurate placed on developing predictions for use within-breed estimates of genomic merit across the population. with the results incorporated into national cattle evaluations. A simulation study by Kizilkaya et al. (2010) found slightly greater correlations between Across-breed Prediction true and estimated breeding values when training and It would be convenient if predictions could be validating in purebreds compared with training and developed in 1 breed and used for other breeds, but validating in a multibreed population. Estimates for 30 this approach has produced very poor accuracies. Sim- purebred Simmental. Only calving ease maternal and ulations trained in one breed and predicted in another weaning weight maternal were not more accurate with resulted in significantly less accuracy than training in the multi-breed training population (Saatchi and Gar- the breed of interest (Toosi et al., 2010). When Angus rick, 2013). Stayability was unchanged because there trained predictions were used in other breeds, simula- was no information for this trait in the other breeds. tions accounted for less than one-third of the genetic With the additional breeds, the size of the reference variation that was achieved in Angus (Kizilkaya et al., population more than doubled compared with only 2010). These simulations did not demonstrate favor- Simmental animals (Saatchi and Garrick, 2013). These able results for using breed-specific predictions across studies suggest a benefit from combining single breed breeds. When breed-specific predictions for Angus, reference populations. Not only is the size of the ref- Hereford, or Limousin were used across breeds, in erence population greater, but the reference population most cases the genetic correlation was not significant can capture more of the variation within the breed of and in a few instances was slightly negative despite a interest. moderate, positive genetic correlation when validating within-breed (Kachman et al., 2013). On average, the Predictions that were developed without in- genetic correlation from using predictions developed cluding the breed of interest were less accurate than if in Herefords on Angus sires was not different from that breed had been included in training. The accura- 0 while predictions developed specifically for Angus cy of multi-breed prediction in Australian sheep was had the greatest accuracies (0.24 to 0.61; Weber et al., always less if the breed for prediction was excluded 2012b). Similar results were observed in Holstein and from the training population (Daetwyler et al., 2012a). Jersey dairy cattle (Hayes et al., 2009). Using 50,000 However, if the breed to be predicted was included in SNP has not been sufficient density to use predictions the reference population, multi-breed predictions were across breeds. Breeds have different LD which erodes no more accurate than single-breed predictions (Pryce the accuracy of predictions developed in one breed et al., 2011). If the within-breed predictions are based when the prediction was used in another breed. The on a large enough reference population, the potential use of genomic predictions across breeds has not been benefit from multi-breed predictions might be very feasible due to the limited prediction accuracy. minimal. If multi-breed predictions someday achieved equivalent or greater accuracies than within-breed predictions, all breeds of interest would need to be Multi-breed for Purebred Prediction included in the training population. Another approach would be to combine data There was very little difference in the accuracy for many purebreds to develop predictions that were of GBV when using a Holstein or Holstein and Jersey then used for individual breeds. Combining popu- reference population to validate in Holsteins, but there lations resulted in greater accuracy, especially for was an increase in accuracy when using the combined lowly heritable traits, than training on each population instead of the Jersey reference population to validate individually (de Roos et al., 2009). Predictions were in Jerseys (Hayes et al., 2009). This could result from more accurate when trained in a multi-breed popula- the very small number of Jersey bulls with genotypes. tion instead of a purebred population and validated The addition of more genotypes, despite the breed, in the same purebred (Bolormaa et al., 2013). This allowed for more accurate genomic predictions. Using possibly results from capturing more of the variants a Holstein, Jersey, and Brown Swiss reference pop- within the breed of interest. Genetic correlations ulation, resulted in an increase in accuracy for some averaged 0.47 (0.10 to 0.73) when trained on only traits in Jersey and Brown Swiss above that of the sin- Simmentals and 0.55 (0.18 to 0.91) when trained on gle-breed prediction (Olson et al., 2012). Again, there Simmental, Angus, Red Angus, Gelbvieh, Brangus, was no benefit for Holsteins to use multi-breed predic- Hereford, and Charolais (Saatchi and Garrick, 2013). tions but some benefit for smaller breeds with fewer This is of interest because animals registered with the genotypes. Incorporating 2 breeds in the reference American Simmental Association do not have to be population to predict a third breed increased prediction

31 FRANK BAKER MEMORIAL SCHOLARSHIP accuracy compared with using 1 of the breeds to pre- with using crossbred genotypes in the beef industry, dict a different breed (Pryce et al., 2011). Again, the mainly the lack of complete pedigree and performance improvement in accuracy could result from the larger recording outside of research herds. The use of crossreference population used to predict marker effects bred predictions for many breeds appears less feasible. in a different breed. Given a reference population of sufficient size, there has been no consistent benefit to Another approach to using crossbred genotypes using a multi-breed population to develop predictions has been to model breed-specific SNP effects. Model- for use in purebreds. Yet, while training populations of ling with breed-specific compared with across-breed sufficient size are being collected, multi-breed predic- SNP effects resulted in similar prediction accuracies tions could help improve accuracy until enough ani- for a variety of simulation scenarios (Ibán z-Escriche mals were genotyped to produce reliable predictions. et al., 2009). As marker density increased up to 2,000 markers on one chromosome, there was less value in using breed-specific SNP effects (Ibán z-Escriche et Crossbred for Purebred Prediction al., 2009). The use of breed-specific SNP effects required large breed differences to justify the additional Another scenario is collecting data on cross- effects in the model, and this model had an advantage bred animals for use in purebreds although this sit- when large training populations were used (Ibán z-Es- uation is unlikely in the beef industry with a lack of criche et al., 2009). Developing reference populations pedigree and performance recording in crossbred cat- of sufficient size to justify the use of breed-specific tle. Simulations demonstrated, as the number of breeds SNP effects will be challenging in the beef industry. represented in the crossbred population increased, the Very few of the 2,500 SNP with the largest effect accuracy of predicting one of the purebreds decreased were common to the GPE and 2,000 Bulls populations (Toosi et al., 2010). This decrease in accuracy could (Weber et al., 2012a). These results suggest a potential result from a decrease in the prevalence of haplotypes need for breed-specific effects to better account for from the breed of interest in the training population as both differences in LD and the magnitude of the SNP population size was held constant. That breed would effect across breeds. then have a lesser contribution to the estimation of marker effects. Using the crossbred U.S. Meat An- imal Research Center Germplasm Evaluation Pro- Crossbred Prediction gram (GPE) population for training and validating in a purebred population resulted in MBV accuracies Although genetic evaluation of crossbred beef generally ranging from 0.20 to 0.40 with less accurate cattle is not common, a cheap genomic test for eco- predictions in Charolais for most traits, likely a result nomically relevant traits would be a valuable genetic of limited Charolais influence in the training popula- selection tool for commercial producers. In addition, tion (Weber et al., 2012a). Validation in the 2,000 Bull many beef breed associations include hybrid animals Project animals, consisting of influential bulls repre- in their genetic evaluations, and breed-specific pre- senting 16 beef breeds, resulted in genetic correlations dictions might not be as accurate in those composite ranging from 0.19 to 0.37, which were similar to animals. Genomic predictions based on 3,000 SNP validation in purebreds (Weber et al., 2012a). Greater for feed efficiency in Angus-Brahman crosses had accuracies are being achieved in the beef industry by accuracies ranging from -0.13 to 0.36 (Elzo et al., using within-breed predictions (Saatchi et al., 2011; 2012). The small number of SNP could have contrib- Saatchi et al., 2012; Northcutt, 2013; Saatchi et al., uted to the limited accuracy that was achieved from 2013; Boddhireddy et al., 2014). An analysis of cross- those genomic predictions. The accuracy of crossbred bred sheep of primarily Merino decent resulted in predictions was numerically less in most cases than greater accuracies for Merinos than for terminal breeds within-breed predictions; however, those estimates (Daetwyler et al., 2010). Thus, the breed makeup for had large standard errors (Mujibi et al., 2011). Larger the crossbred population was important, and the breed reference populations incorporating a broader sample of interest needed to be well represented in the cross- of the possible breed crosses might improve accuracy bred genetics. There are many challenges associated as more of the population of interest would be used to 32 develop predictions. more accurate as a result of the increase in genomic testing. Research on using genomic predictions developed in 1 breed for use in another has not been favor- Purebred or Multi-breed for Crossbred Prediction able. Yet, pooling genotypes from multiple breeds to develop predictions for a purebred was promising for Because most phenotypes in the beef industry increasing accuracy past that achieved with only the are collected on purebreds, creating predictions based purebred genotypes. The use of multi-breed predic- on the purebred data for use in selecting crossbreds tions could be of interest to smaller breeds with fewer could be beneficial. Training on Angus, Angus and genotyped animals and breeds that register percentage Red Angus, or Hereford resulted in weak MBV accu- animals. Smaller breeds could benefit from the larger racy (0.01 to 0.43) for growth and carcass traits when reference population that could be assembled from validating in the crossbred GPE population (Weber combining genotypes. Breeds with hybrid animals et al., 2012a). MBV accuracy tended to be less than could benefit from the inclusion of the LD from other that achieved with a multi-breed training population breeds to develop more accurate predictions. consisting of sires in the 2,000 Bull Project (Weber et al., 2012a). Using a multi-breed instead of purebred There is potential benefit from genetic eval- training population should produce better predictions uation at the commercial level. Because there is no for crossbred animals because breeds differ in the LD infrastructure for performance recording in commer- between SNP and QTL. Training on 2,000 Bull Proj- cial cattle, genomic testing is a more feasible option to ect and validating on GPE yielded moderate genetic identify the genetically superior crossbred cattle. Pre- correlations (0.13 to 0.42) with little or no improve- liminary research has demonstrated the feasibility of ment from including breed effects in the DEBV for the developing accurate genomic predictions from pure- 2,000 Bull Project (Weber et al., 2012a). Training on bred or multi-breed populations to use in crossbred a multi-breed population instead of a purebred popu- individuals. As genomic predictions become more lation increased accuracy more for composite breeds refined in the seedstock industry, there is the potential than purebreds (Bolormaa et al., 2013). Genotypes to develop cheaper genomic tests for economically and phenotypes on purebreds can be useful to develop relevant traits in crossbred cattle. predictions for crossbreds. The American Angus Asso- ciation and Zoetis currently market a genomic test for Literature Cited commercial Angus-influence cattle. This test provides predictions for a couple economically relevant Boddhireddy, P., M. J. Kelly, S. Northcutt, K. C. traits, and commercial producers are using this test to Prayaga, J. Rumph, and S. DeNise. 2014. Genomic add value to feeder cattle and to select replacement predictions in Angus cattle: Comparisons of sam- females. If the beef industry were to move toward ple size, response variables, and clustering meth- larger scale crossbred genetic evaluation, establishing ods for cross-validation. J. Anim. Sci. 92:485-497. genomic predictions from existing purebred databas- Bolormaa, S., J. E. Pryce, K. Kemper, K. Savin, B. J. es appears to be the most feasible method. Including Hayes, W. Barendse, Y. Zhang, C. M. Reich, B. many breeds in the reference population would help A. Mason, R. J. Bunch, B. E. Harrison, A. Revert- make these predictions more relevant for a wider array er, R. M. Herd, B. Tier, H.-U. Graser, and M. E. of commercial producers. Goddard. 2013. Accuracy of prediction of genomic breeding values for residual feed intake and Conclusions and Implications to Genetic Improve- carcass and meat quality traits in Bos taurus, Bos ment of Beef Cattle indicus, and composite beef cattle. J. Anim. Sci. 91:3088-3104. As the use of genomic testing in the beef industry grows, reference populations of greater size are being established by individual breed associations. Breed-specific genomic predictions are becoming

33 FRANK BAKER MEMORIAL SCHOLARSHIP

Brøndum, R. F., E. Rius-Vilarrasa, I. Strandén, G. Su, Ibán z-Escriche, N., R. L. Fernando, A. Toosi, and J. B. Guldbrandtsen, W. F. Fikse, and M. S. Lund. C. M. Dekkers. 2009. Genomic selection of pure- 2011. Reliabilities of gneomic predictions using breds for crossbred performance. Genet. Sel. Evol. combined reference data of the Nordic Red dairy 41:12. cattle populations. J. Dairy Sci. 94:4700-4707. Kachman, S. D. 2008. Incorporation of marker scores Daetwyler, H. D., J. M. Hickey, J. M. Henshall, S. into national genetic evaluations. In: Proc. Beef Dominik, B. Gredler, J. H. J. van der Werf, and Improv. Fed. 9th Genet. Pred. Workshop., Kansas B. J. Hayes. 2010. Accuracy of estimated genom- City, MO. p. 92–98. ic breeding values for wool and meat traits in a Kachman, S. D., M. L. Spangler, G. L. Bennett, K. multi-breed sheep population. Anim. Prod. Sci. J. Hanford, L. A. Kuehn, W. M. Snelling, R. M. 50:1004-1010. Thallman, M. Saatchi, D. J. Garrick, R. D. Schna- Daetwyler, H. D., K. E. Kemper, J. H. J. van der Werf, bel, J. F. Taylor, and E. J. Pollak. 2013. Compari- and B. J. Hayes. 2012. Components of the accu- son of molecular breeding values based on within- racy of genomic prediction in a multi-breed sheep and across-breed training in beef cattle. Genet. Sel. population. J. Anim. Sci. 90:3375-3384. Evol. 45:30. Daetwyler, H. D., A. A. Swan, J. H. J. van der Werf, Kilzikaya, K., R. L. Fernando, and D. J. Garrick. 2010. and B. J. Hayes. 2012b. Accuracy of pedigree and Genomic prediction of simulated multi-breed and genomic predictions of carcass and novel meat purebred performance using observed fifty thou- quality traits in multi-breed sheep data assessed by sand single nucleotide polymorphism genotypes. J. cross-validation. Genet. Sel. Evol. 44:33. Anim. Sci. 88:544-551. de Roos, A. P. W., B. J. Hayes, and M. E. Goddard. Lund, M. S., A. P. W. de Roos, A. G. de Vries, T. 2009. Reliability of genomic predictions across Ducrocq, S. Fritz, F. Guillaume, B. Guldbrandtsen, multiple populations. Genetics. 183:1545-1553. Z. Liu, R. Reents, C. Schrooten, F. Seefried, and Elzo, M. A., G. C. Lamb, D. D. Johnson, M. G. Thom- G. Su. 2011. A common reference population from as, I. Misztal, D. O. Rae, C. A. Martinez, J. G. four European Holstein populations increases re- Wasdin, and J. D. Driver. 2012. Genomic-polygen- liability of genomic predictions. Genet. Sel. Evol. ic evaluation of Angus-Brahman multi-breed cattle 43:43. for feed efficiency and postweaning growth using Meuwissen, T. H. E., B. J. Hayes, and M. E. God- the Illumina 3K chip. 2012. J. Anim. Sci. 90:2488- dard. 2001. Prediction of total genetic value using 2497. genome-wide dense marker maps. Genetics. Habier, D., R. L. Fernando, and J. C. M. Dekkers. 157:1819-1829. 2007. The impact of genetic relationship informa- Mujibi, F. D. N., J. D. Nkrumah, O. N. Durunna, P. tion on genome-assisted breeding values. Genetics. Stothard, J. Mah, Z. Wang, J. Basarab, G. Pla- 177:2389-2397. stow, D. H. Crews, Jr., and S. S. Moore. 2011. Hayes, B. and M. Goddard. 2010. Genome-wide asso- Accuracy of genomic breeding values for residual ciation and genomic selection in animal breeding. feed intake in crossbred beef cattle. J. Anim. Sci. Genome. 53:876-883. 89:3353-3361. Hayes, B. J., P. J. Bowman, A. C. Chamberlain, K. Northcutt, S. L. 2013. AAA experience with incorpo- Verbyla, and M. E. Goddard. 2009. Accuracy of rating genomics into genetic evaluation. In: Proc. th genomic breeding values in multi-breed dairy cat- Beef Improv. Fed. 10 Genet. Pred. Workshop., tle populations. Genet. Sel. Evol. 41:51. Kansas City, MO. p. 5-7.

34 Olson, K. M., P. M. VanRaden, and M. E. Tooker. Saatchi, M., J. Ward, and D. J. Garrick. 2013. Accura- 2012. Multi-breed genomic evaluations using cies of direct genomic breeding values in Hereford purebred Holsteins, Jerseys, and Brown Swiss. J. beef cattle using national or international training Dairy Sci. 95:5378-5383. populations. J. Anim. Sci. 91:1538-1551. Pryce, J. E., B. Gredler, S. Bolormaa, P. J. Bowman, Thallman, R. M., K. J. Hanford, R. L. Quaas, S. D. C. Egger-Danner, C. Fuerst, R. Emmerling, J. Kachman, R. J. Templeman, R. L. Fernando, L. A. Sӧlkner, M. E. Goddard, and B. J. Hayes. 2011. Kuehn, and E. J. Pollak. 2009. Estimation of the Short communication: Genomic selection using a proportion of genetic variation accounted for by multi-breed, across-country reference population. DNA tests. In: Proc. Beef Improv. Fed., Sacramen- J. Dairy Sci. 94:2625-2630. to, CA. p. 184-209. Pszczola, M., T. Strabel, H. A. Mulder, and M. P. L. Toosi, A., R. L. Fernando, and J. C. M. Dekkers. 2010. Calus. 2012. Reliability of direct genomic values Genomic selection in admixed and crossbred pop- for animals with different relationships within and ulations. J. Anim. Sci. 88:32-46. to the reference population. J. Dairy Sci. 95:389- VanRaden, P. M., K. M. Olson, D. J. Null, M. Sargol- 400. zaei, M. Winters, and J. B. C. H. M. van Kaam. Saatchi, M., S. R. Miraei-Ashtiani, A. Nejati-Java- 2012. Reliability increases from combining remi, M. Moradi-Shahrebabak, and H. Mehreba- 50,000- and 777,000- marker genotypes from four ni-Yeghaneh. 2010. The impact of information countries. Interbull Bull. 46:75-79. quantity and strength of relationship between train- Weber, K. L., R. M. Thallman, J. W. Keele, W. M. ing set and validation set on accuracy of genomic Snelling, G. L. Bennett, T. P. L. Smith, T. G. Mc- estimated breeding values. Afr. J. Biotechnol. Daneld, M. F. Allan, A. L. Van Eenennaam, and L. 9:438-442. A. Kuehn. 2012a. Accuracy of genomic breeding Saatchi, M., M. C. McClure, S. D. McKay, M. M. values in multi-breed beef cattle populations de- Rolf, J. W. Kim, J. E. Decker, T. M. Taxis, R. H. rived from deregressed breeding values and pheno- Chapple, H. R. Ramey, S. L. Northcutt, S. Bauck, types. J. Anim. Sci. 90:4177-4190. B. Woodward, J. C. M. Dekkers, R. L. Fernando, Weber, K. L., D. J. Drake, J. F. Taylor, D. J. Garrick, R. D. Schnabel, D. J. Garrick, and J. F. Taylor. L. A. Kuehn, R. M. Thallman, R. D. Schnabel, W. 2011. Accuracies of genomic breeding values in M. Snelling, E. J. Pollak, and A. L. Van Eenen- American Angus beef cattle using K-means clus- naam. 2012b. The accuracies of DNA-based tering for cross-validation. Genet. Sel. Evol. 43:40. estimates of genetic merit derived from Angus Saatchi, M., R. D. Schnabel, M. M. Rolf, J. F. Tay- or multi-breed beef cattle training populations. J. lor, and D. J. Garrick. 2012. Accuracy of direct Anim. Sci. 90:4191-4202. genomic breeding values for nationally evaluated Zhou, L., X. Ding, Q. Zhang, Y. Wang, M. S. Lund, traits in US Limousin and Simmental beef cattle. and G. Su. 2013. Consistency of linkage disequi- Genet. Sel. Evol. 44:38. librium between Chinese and Nordic Holsteins and Saatchi, M., and D. J. Garrick. 2013. Improving ge- genomic prediction for Chinese Holsteins using a nomic prediction in Simmental beef cattle using a joint reference population. Genet. Sel. Evol. 45:7. multi-breed reference population. In: Proc. West. Sec. Am. Soc. Anim. Sci., Bozeman, MT. p. 174- 177.

35 AWARD FRANK BAKER MEMORIAL SCHOLARSHIP RECPIENT

HIGH ALTITUDE DISEASE AND GENETICS OF BEEF CATTLE AT HIGH ELEVATION REGIONS Xi Zeng1 1Colorado State University

1 Introduction and the herd average weaning weight in 2009 (529.8± 72.4lbs; Neary, 2013). Native cattle at high altitude may be more resistant to HAD than low attitude cattle In high altitude states such as Colorado, Wyoming, due to artificial selection (Will et al., 1975). About 10% New Mexico, and Utah, bovine pulmonary hyperten- to 40% of cattle develop HAD when they were moved sion (BPH) is observed and commonly referred to as from low altitude to high altitude (Grover et al., 1963, “brisket disease” or “high altitude disease (HAD)” Will et al., 1970). (Holt and Callen, 2007). The disease was first studied 2.2 Physiology of HAD by Glover and Newsome (1915) in cattle for the sole purpose of advising Colorado and New Mexico stock- Based on clinical and physiologic principles, three man to protect their herds. The cardinal sign of HAD is major high-altitude diseases were identified (West, swelling of the brisket due to fluid accumulation in the 2004): 1. Acute mountain sickness. The mechanisms thoracic cavity. It is believed that in response to alveo- are not fully understood, but brain swelling may be lar hypoxia, the pulmonary artery constricts resulting in a phenotype. 2. High-altitude pulmonary edema. The hypertension, right heart ventricular hypertrophy, vas- mechanism is probably uneven hypoxic pulmonary cular remodeling, and death from congestive heart fail- vasoconstriction that exposes some capillaries to a ure (Holt and Callan, 2007). Pulmonary arterial pres- high pressure, damaging their walls and leading to a sure (PAP) is a measure indicative of hypertension and high-permeability form of edema. 3. High-altitude has been reported to be moderately heritable in cattle cerebral edema. It closely related to acute mountain (0.34 to 0.46; Enns et al., 1992; Shirley et al., 2007). sickness and that it is the extreme end of the spectrum. Therefore, PAP has been widely used as an indicator These are specific description of different types of High trait of BPH/HAD in recent studies. Therefore, various Altitude Disease. What can be used as a general ref- studies involving PAP have been applied to describe erence to identify HAD? The hypoxia from the high reasons for HAD, including genetics, since 1914. elevation regions is the major cause of HAD/BHP. Al- exander and Jenson (1959, 1963) found that, hypoxia at 2 Literature review high elevation causes pulmonary vasoconstriction, in- 2.1 Economics creased pulmonary arterial pressure (PAP), right ventricle stress, congestive right heart failure, and hydrotho- Why HAD is important and worth to be studied in rax in the chest cavity and brisket. Additionally, Holt cattle at high elevation? There is a high economic rele- and Callen (2007) indicated that HAD is characterized vance to HAD, with an incidence of 3% to 5% typically by the presence of ventral edema in the brisket region in native cattle (Holt and Callen, 2007). However, it is secondary to increased vascular hydrostatic pressure a major cause of calf morbidity for beef cattle ranch- (intravascular hypertension) and the loss of fluid into es and feedyards above 1500 m (Hecht et al., 1962; the extra vascular space. Jensen et al., 1976). A producer losing 20% of his 600 calves equates to $78,864 of lost potential income be- 2.3 Relationship between PAP and HAD tween summer turnout and weaning based on the mar- As an indicator of HAD, PAP scores were used to ket price ($1.24/lb. live weight, November 7th 2011) 36 FRANK BAKER MEMORIAL SCHOLARSHIP

assist selection of cattle to reduce HAD in recent de- been shown to be moderately to highly heritable and cades in high altitude regions. Holt and Callen (2007) repeatable in cattle (Schimmel, 1981; Enns et al., 1992; indicate that: 1. The measured PAP of less than 41 Shirley et al., 2007). The heritability and repeatability mmHg at an elevation greater than 1500 m (5000 ft) of PAP were first estimated in a dissertation work of are likely to maintain an acceptable PAP at high altitude Schimmel (1981). The PAP values in this study were and serve as good breeding stock; 2. Animals with mea- collected from weaning calves and mature cow raised sured PAP larger than 41 mmHg and less than 49mmHg at the San Juan Basin Research Center, Hesperus, Col- at high altitude should be used with caution at high ele- orado (elevation at 2,316m). He reported heritabilities vations; 3. Cattle with PAP larger than 49mmHg at any of PAP as 0.77 ± 0.21, 0.60 ± 0.24, 0.40 ± 0.13 and 0.13 altitude are at risk for developing HAD and should not to 0.23 for bull, heifer, calves and cows. Enns (1992) be maintained or used in breeding programs at high al- reported a heritability estimate as 0.46 ± 0.16 of wean- titude. Therefore, these recommendations serve as phe- ing measured PAP, which were from Angus cattle from notypic selection tools. Also, the information indicates western Colorado. The most recent published heritabil- that higher PAP measures imply higher risk for HAD. ity of PAP was 0.34 ± 0.05 reported by Shirley (2007). In addition, an ongoing study estimated heritability for 2.4 Measurement of PAP PAP measured in yearling Angus cattle was 0.21±0.04, 0.37±0.08, 0.19±0.14 and 0.23±0.03 for bulls, heifers, The procedure used to measure PAP has been used steers and compiled data (Cockrum, unpublished data). for more than 30 years. However, PAP measures can be Similar to many other traits, the estimated heritabili- influenced by any unprofessional action in the process. ty was varied among studies, which may account for Therefore, the PAP score can only be taken by one li- the genetic by environmental effect of age of PAP and censed veterinarian in one herd in order for selection to sex management. However, all of the studies showed a be more effective. With the right equipment and facili- moderate to high heritability for PAP measure (0.23 to ties, a professional veterinarian can take PAP score for 0.77). Furthermore, the study from Cockrum, which will a large number of animals daily, which makes PAP a be published in August 2014 at 10th World Congress on measurable and affordable trait for selection. The PAP Genetics Applied to Livestock production (WCGALP), test is a right heart catheterization procedure, which was similar to results from Schimmel (1981), in that requires jugular venipuncture, catheter insertion and the heritability of PAP of bulls was higher than that of passing flexible catheter tubing through a large bore heifers. This fact may result from the high intensity ar- needle inserted into the jugular vein. The catheter is tificial selection of bulls. The repeatability reports were passed down the jugular vein, through the right atrium, limited in previous studies, the reason for which may be into the right ventricle, and then into the pulmonary ar- that the PAP score is usually measured once (i.e. year- tery. Once the catheter is inside the pulmonary artery, ling). However we can expect a moderate repeatability an average blood pressure (average of systolic and dia- of PAP, based on the repeatability as 0.25 to 0.16 on stolic values) is recorded from the heart monitor, which cows reported by Schimmel (1981). is attached to the catheter via a transducer (Ahola et al., 2007). 2.5.2 Genetic Correlation 2.5 Genetic Parameters for PAP Veit and Farrell (1978) suggested that larger body size and metabolic demands would place stress on the 2.5.1 Heritability and Repeatability bovine pulmonary system; thus pre-disposing cattle to In order to reveal the genetics influences within respiratory disease and pulmonary hypertension. This HAD, heritability, repeatability and genetic correlation viewpoint was supported by the estimated genetic correlated to PAP have been estimated in many studies. relation from Shirley et al. (2007), who reported the ge- Heritability is the proportion of phenotypic variation netic correlation between PAP and birth weight (BW) that is explained by additive genetic variation. Table 1 or weaning weight (WW) to be moderate (i.e. 0.49 to summarized the heritability of PAP reported in previ- 0.51). However, the genetic correlation between PAP ous literature. Pulmonary arterial pressure (PAP) has and post weaning growth traits of yearling weigh (YW) and post weaning gain (PWG) were reported to be 0.22

37 FRANK BAKER MEMORIAL SCHOLARSHIP

± 0.04 and 0.04 ± 0.12, respectively, which is interpret- most interested in the extreme value of PAP. ed as weak, yet positive genetic correlation (Zeng et al., unpublished data). However, Schimmel (1981) report- 2.5.4 Expected Progeny Differences (EPD) for PAP ed a genetic correlation of between bull PAP and YW. The expected progeny differences (EPD) for PAP Based on these results, the genetic parameters of PAP were first estimated with data from the Tybar Ranch, appeared to be varying among studies. Carbondale, CO. Since the first use of a PAP EPD for The varied estimates of genetic correlations may be selection of resistance to HAD at the Tybar Ranch in explained by genetic difference observed among PAP 1992, the EPD for PAP was continuously used in cat- collected at different age (weaning versus yearling) or tle breeding in Colorado (Enns, 2011). Also, the PAP different sex (bulls versus heifers; Cockrum et al., un- EPD has been used in the selection program in John published data). The genetic correlation between PAP E. Rouse Ranch of Colorado State University Beef of weaning and yearling (0.56 ± 0.24), or between year- Improvement Center (CSU-BIC) since 2006. Figure 1 ling PAP of bulls and heifers (0.67 ± 0.15) was not high, presents the genetic trend of PAP EPD from both Tybar which suggested that PAP measurements at weaning and Ranch and CSU-BIC. The genetic trend in PAP score yearling, or in heifers and bulls were potentially differ- has been consistently downward (favorable) since the ent traits (Cockrum et al., unpublished data). These re- use of a PAP EPD in Tybar Ranch beginning in 1992. sults implied a genetic difference between BPH/HAD of The downward (favorable) genetic trend has also been bulls and heifers. However, we must consider these re- seen at CSU-BIC since the use of a PAP EPD in 2006. sults as potentially confound with growth management. Producer reports collected in veterinary health studies Both of these trait measures (i.e. bulls and heifers) were suggest that, in some cases, low PAP cows should sig- based on the data from John E. Rouse Ranch of Colo- nificantly reduce the incidence of HAD within their calf rado State University Beef Improvement Center (CSU- crop. However, report from other producers indicated BIC). In the production system, bulls were developed that the selection on low PAP has no influence on re- within a grain-supplemented performance test, whereas ducing the mortality of pre-weaned beef calves (Neary, heifers and steers were grazed. Therefore, these may be 2013). Therefore, more studies should be executed to a genetic by environmental interaction via two source ensure that genetic selection on low PAP would reduce of information: 1) sex; 2) diet environment. Similarly, the chance of cattle to HAD/BPH. the environmental effect on phenotype of PAP had been Although, PAP has been widely recognized as an reported in earlier literatures, which suggested that age, indicator to study HAD/BPH, there are limitations in gender, temperature and diet influenced PAP phenotype this trait and its interpretation. First, cattle need adap- (Rhodes, 2005; Holt and Callen., 2007). tion period for at least 30 days before PAP scored mea- 2.5.3 Model Used for Genetic Evaluation surement when they move from low altitude to high altitude area (Holt and Callen, 2007). Second, the mea- Both univariate and multivariate animal models sure of PAP needs to be completed by a skilled veteri- were used in previous PAP studies. The fixed effects narian. included in the models included PAP date, sex, age of dam, management contemporary group, and age of PAP 2.6 Genomic wide Association Study (GWAS) as covariate (Shirley et al., 2007). The random effects With the advance of molecular genetics techniques, in these models were animals. In previous studies, the high-density marker maps and tools are available and major software used to execute these mixed animal large number of animals can be genotyped with a rea- models with continuous response variable was ASReml sonable investment. This fact allows genome wide as- (Gilmour, 2009). However, PAP scores are not normal- sociation study (GWAS), which utilizes high-density ly distributed which violate our assumption in evalu- single-nucleotide polymorphisms (SNP). The GWAS is ation of these models. The problem may be solved by an approach to revel common genetic variants in dif- transforming the PAP data to categorical data, and then ferent individuals to assess if any variant is associated execute a threshold animal model for genetic evalua- with a trait. In the beef industry, GWAS can be used tion. We hypothesize the later is reasonable as we are in genomic selection using estimate genomic estimate 38 breeding value (GEBV), whose accuracy is much high- accuracy GEBV. er than traditional EBV. Also, GWAS has been widely used in identifying significant SNP, biological pathways The genome based BLUP, BayesA and BayesB and networks underlying complex traits. Therefore it is were first introduced, compared and discussed in the beneficial to conduct GWAS on PAP and use GEBV or paper of Meuwissen et al. (2001). The BLUP method marker assisted selection (MAS) to conduct selection assumed a normal distribution of SNP effects, which of cattle at both low and high altitude for resistant to suggested a very large number of QTL with small ef- HAD. However, there are few published GWAS studies fects. The BayesA assumed at distribution of SNP ef- on PAP or HAD on cattle, except for the work from fects, which is based on a large number of QTL with Newman et al. (2011) and unpublished work from Col- small effects and a small proportion with moderate to orado State University to be presented in August 2014 large effects. In BayesA, the variance of each SNP ef- at 10th World Congress on Genetics Applied to Live- fect was assumed unequal and under an inverted chi- stock production (WCGALP), Vancouver, BC, Canada square distribution with scale parameter S and v degree of freedom, whereas it is assumed that the error vari- 2.6.1 Response Variable in GWAS ance was under a inverted chi-square with scale parameters 2. BayesB is a method assuming mixture distribu- Information resources used in GWAS can be al- tion of zero effects and t distribution of effects for SNP, ternative sources of information including single or re- which suggest a large number of genome regions with peated measures of individual phenotypic performance, zero effect and a small proportion of QTL with mod- information on progeny, estimated breeding value erate effects. The variance distribution assumption for (EBV) from genetic evaluations, or a pooled mixture QTL loci and error term are the same as BayesA. of more than one of these information sources (Gar- rick et al., 2009). The SNP/marker effects were come Habier et al. (2011) developed BayesC and from the training data and would be used to fit the test BayesCπ. Both assumed that there is πproportion of data to estimate the GEBV. To guarantee the accuracy loci have 0 effect and (1-π) proportion of loci have of GEBV prediction, the ideal data for training would moderate to large effect with common variance across be true genetic merit data observed on unrelated ani- these loci. The π is a fixed value in BayesC while in mals in the absence of selection (Garrick et al., 2009). BayesCπ, π is sampled from a beta distribution based Also, as indicated previously, the PAP scores are not on data. The error variance is assumed under an in- normally distributed, which violate the assumption of verted chi-square distribution with scale parameter 2 statistical methods used in GWAS. Therefore, a dere- as other Bayes methods. These Bayesian methods can gressed estimated breeding value (DEBV) may be the be executed using the GenSel software (Fernando and best response variable used in future GWAS on PAP. Garrick, 2008). 2.6.2 Method Used in GWAS Another method is Bayesian Lasso introduced by Yi and Xu (2008), which also assumed a very large pro- Even though, published GWAS of PAP or HAD on portion of SNP effect close to zero and small proportion cattle are forthcoming, the statistical methods used in with a moderate to large effect. In this method, the SNP GWAS for different traits are generally the same. In or- effect is under a normal distribution and the variance of der to improve the accuracy of GWAS, many statistical QTL is under an exponential distribution. methods have been applied during the past 20 years. Actually, these methods are different kinds of mod- The GBLUP is based on the restricted maximum el selection methods. The most widely used methods likelihood (REML) concept. The SNP effects and vari- include BLUP, BayesA, BayesB, BayeCπ, Bayesian ance can be estimated from mixed model developed by LASSO, GBLUP, machine learning etc. Hayes and Henderson (1976) based on REML with treating the Goddard (2010) concluded that the highest accuracies SNP as random effect and including a genomic rela- of GWAS were achieved when the prior distribution of tionship matrix. Using this methods, fixed effects can SNP effects matches the true distribution. The method be estimated too. This GWAS method can be accom- assuming many SNP effects of zero and a small propor- plished using many software packages including SVS tion of SNPs with moderate to large effects yield higher (Golden Helix, Inc., Bozeman, MT), R, SAS (SAS In- 39 FRANK BAKER MEMORIAL SCHOLARSHIP stitute, Cary NC), ASReml (Gilmour et al., 2009), etc. and the myocardial signaling protein (FKBP1A). Be- In R, some GWAS packages written by other research- sides identification of significant genes, BPH related ers can be used directly. pathway results and gene networks were also explained in the study to help the understanding of the biological In addition to the previous methods, a machine signature of BPH. Qiu et al. (2012) found that gene learning method was developed by Long et al. (2007). families, which were related to sensory perception and This method can be used to classify suspect and healthy energy metabolism, as well as an enrichment of protein animals with high accuracy and identify disease relat- domains involved in sensing the extracellular environ- ed SNPs. Specifically, a case-control experiment is ment and hypoxic, were express different between yak designed, then machine learning method was used to and cattle. This fact can be used to study the adaption select SNPs. Besides the naïve Bayes used in Long’s to high altitude in other animal species and humans. study, the machine learning method has many algo- In addition, a study (Wang et al., 2012) identified a rithms including support vector machine, decision tree, Hypoxia-inducible factor-2alpha (HIF-2α) encoding artificial neural machine, etc. gene, Endothelial PAS domain-containing protein 1 (EPAS-1), which is a key gene mutated in the Tibetan In recent years, a method named as multiple locus population adapted to living at high altitude. mixed model (MLMM) were used in GWAS studies. It is a method using a simple stepwise mixed-model re- In future, since cattle are considered a natural an- gression with forward inclusion and backward elimi- imal model to study HAD and higher density chip are nation of genotypic markers as fixed effect covariates available for genotyping, GWAS should be done on cat- with a genomic relationship matrix (Segura, 2012). The tle. Since the heritable PAP score was widely treated as variance components are re-estimated between each an indicator of HAD. The GWAS study based on PAP forward and backward step. Currently, the MLMM is score can be used to identify the most significant SNP available in the SVS (Golden Helix, Inc., Bozeman, related to HAD, and then related genes can be studied. MT). The method discussed by Fortes et al. (2011) can be used to develop a gene network on PAP with the detect- 2.6.3 Results of GWAS ed SNPs. Furthermore, these genes can be used to con- Few GWAS studies have been conducted for HAD duct a pathway study, which can help reveal the whole or PAP on cattle, but there are some genomic related picture of HAD and provide efficient treatment plan. studies on HAD for yaks and humans. Reviews of these 3 Conclusion and Implications to Genetic Improve- effects give us some genomic information on HAD ment of Beef Cattle across species. Also, these results can be compared to our future findings to help us explain our results. The Selection for resistance to HAD/BPH is important following is a review of potential candidate genes. for beef cattle, because HAD influences calf mortality at high altitudes (above 1500m). Pulmonary arterial In the study of Simonson et al. (2010), they report- pressure can be treated as indicator trait for selection of ed that gene Egl nine homolog 1 (EGLN1) and Per- tolerance to high altitude, especially since it is physio- oxisome proliferator-activated receptor alpha (PPARA) logically related to HAD/BPH and moderately herita- were associated with hypoxia response factor (HIF) ble. Genetic selection for low PAP by beef producers and expressed in high altitude adapted individuals, at high altitudes could potentially improve profitability which can be used to study the high altitude adaption by reducing the mortality rate. However, more genet- pathway in humans. Newman et al. (2011) provided the ic evidence is needed to ensure that selection for low first molecular interrogation on BPH based on a case PAP could reduce the incidence of HAD. The GWAS of control GWAS and gene expression study. The study PAP score can be used to identify the most significant revealed six or more significant genes, among which SNPs or genes potentially related to HAD, and estimate three genes were candidates possibly involved in BPH GEBV to serve as a selection tool. Thus, genomic infor- including NADH dehydrogenase (ubiquinone) flavo- mation can help the selection of cattle for resistance to protein 2 (NDUFV), myosin heavy chain 15 (MYH15) HAD at earlier ages. Besides the benefit of traditional

40 Table 1. Estimated heritability and repeatability for pulmonary arterial pressure (PAP) in previous literature

Author Heritability Repeatability Age of cattle Schimmel (1981) 0.13~0.23 0.25~0.26 Mature Cow Schimmel (1983) 0.20~0.77 - Weaning Enns (1992) 0.46(0.16) - 166d-662d Shirley (2007) 0.34(0.05) - Weaning

Figure 1. Genetic trend pulmonary artery pressure at the Tybar Ranch (Tybar) and the CSU John E. Rouse Beef Improvement Center (BIC) since selection with EPD began in 1992 in Tybar) and 2002 in BIC (Enns et al., 2011). genetic selection on PAP, GWAS of PAP will also help Alexander, A. F., & Jensen, R. 1959. Gross cardiac reveal the genomic architecture of HAD by studying changes in cattle with high mountain (brisket) genes, and increase the selection efficiency for resis- disease and in experimental cattle maintained tance to HAD. However, case/control data of HAD are at high altitudes. Amer. J. Vet. Res. 20:680-689. needed to help expose information of the complex high altitude disease. Thus, it is important and beneficial to Alexander, A. F., & Jensen, R. 1963. Pulmonary collaborate with breeders across mountains regions of vascular pathology of high altitude-induced the country to collect the HAD case in the future. pulmonary hypertension in cattle. Amer. J. Vet. Res. 24:1112-1122. 4 Literature Cited Enns, R. M., Brinks, J. S., Bourdon, R. M., & Field, T. G. 1992. Heritability of pulmonary arterial Ahola, J. K., Enns, R. M., & Holt, T. 2006. Examina- pressure in Angus cattle. Proc. West. Sect. Am. tion of potential methods to predict pulmonary Soc. Anim. Sci. (Vol. 43, pp. 111-112). arterial pressure score in yearling beef cattle. J. Anim. Sci. 84:1259-1264.

41 FRANK BAKER MEMORIAL SCHOLARSHIP

Enns, R. M., Brigham, B. W., McAllister, C. M., Hecht, H. H., Kuida, H., Lange, R. L., Thorne, J. L., & & Speidel, S. E. 2011. Evidence of genetic Brown, A. M. (1962). Brisket disease: II. Clin- variability in cattle health traits: Opportunities ical features and hemodynamic observations in for improvement. Proc. Beef Improvement altitude-dependent right heart failure of cattle. Federation http://www.beefimprovement.org/ Amer. J. Med. 32:171-183. proceedings.html. Holt, T. N. and Callan, R. J. 2007. Pulmonary arterial Fernando, R. L., & Garrick, D. J. 2008. GenSel-User pressure testing for high mountain disease in manual for a portfolio of genomic selection re- cattle. Vet. Clinics of N. Amer.: Food Anim. lated analyses. Animal Breeding and Genetics, Practice. 23:575-596. Iowa State University, Ames. Jensen, R., Pierson, R. E., Braddy, P. M., Saari, D. Fortes, M. R., Reverter, A., Nagaraj, S. H., Zhang, Y., A., Benitez, A., Horton, D. P., ... & Will, D. Jonsson, N. N., Barris, W., & Hawken, R. J. H. (1976). Brisket disease in yearling feedlot 2011. A single nucleotide polymorphism-de- cattle. J. Amer. Vet. Med. Assoc. 169:515-517. rived regulatory gene network underlying pu- Long, N., D. Gianola, G. Rosa, K. Weigel and S. Av- berty in 2 tropical breeds of beef cattle. . Anim. endano. 2007. Machine learning classification Sci. 89:1669-1683. procedure for selecting SNPs in genomic selec- Garrick, D. J., Taylor, J. F., & Fernando, R. L. 2009. tion: application to early mortality in broilers. Deregressing estimated breeding values and J. Anim. Breed. Genet. 124:377-389. weighting information for genomic regression Meuwissen, T. H. E., B. Hayes and M. Goddard. analyses. Genet. Sel. Evol. 41: 44. 2001. Prediction of total genetic value using Gilmour, A. R., Gogel, B. J., Cullis, B. R., & Thomp- genome-wide dense marker maps. Genetics son, R. 2009. ASReml user guide release 3.0. 157:1819-1829. VSN International Ltd, Hemel Hempstead, Neary, J. M. 2013. Pre-weaned beef calf mortality on UK. high altitude ranches in Colorado (Doctoral Glover, G. H., and Newman, I. E. 1915. Brisket dissertation, Colorado State University). Disease (Dropsy of high Altitude). Colorado Newman, J. H., T. N. Holt, L. K. Hedges, B. Womack, Agriculture Experiment Station. 204 Prelimi- S. S. Memon, E. D. Willers, L. Wheeler, J. nary Report, 3:24. A. Phillips III and R. Hamid. 2011. High-alti- Grover, R. F., Reeves, J. T., Will, D. H., & Blount, S. tude pulmonary hypertension in cattle (brisket G. 1963. Pulmonary vasoconstriction in steers disease): Candidate genes and gene expression at high altitude. J. Appl. Physiol. 18:567-574. profiling of peripheral blood mononuclear Habier, D., R. L. Fernando, K. Kizilkaya and D. J. cells. Pulmonary Circulation 1: 462. Garrick. 2011. Extension of the Bayesian Qiu, Q., G. Zhang, T. Ma, W. Qian, J. Wang, Z. Ye, C. alphabet for genomic selection. BMC Bioinfor- Cao, Q. Hu, J. Kim and D. M. Larkin. 2012. matics 12:186. The yak genome and adaptation to life at high Hayes, B. and M. Goddard 2010. Genome-wide asso- altitude. Nature Genetics. ciation and genomic selection in animal breed- Rhodes, J. 2005. Comparative physiology of hypoxic ing. Genome 53:876-883. pulmonary hypertension: historical clues from Henderson, C. 1976. A simple method for computing brisket disease. J. Appl. Phys. 98:1092-1100. the inverse of a numerator relationship matrix used in prediction of breeding values. Biomet- rics : 69-83.

42 Segura, V., Vilhjálmsson, B. J., Platt, A., Korte, A., Seren, Ü., Long, Q., & Nordborg, M. 2012. An efficient multi-locus mixed-model approach for genome-wide association studies in structured populations. Nature Genetics. 44:825-830. Schimmel, J. G. 1981. Genetic aspects of high mountain disease in beef cattle. PhD Diss. Colorado State Univ., Fort Collins Shirley, K. L., Beckman, D. W., & Garrick, D. J. 2008. Inheritance of pulmonary arterial pressure in Angus cattle and its correlation with growth. J. Anim. Sci. 86:815-819. Simonson, T. S., Y. Yang, C. D. Huff, H. Yun, G. Qin, D. J. Witherspoon, Z. Bai, F. R. Lorenzo, J. Xing and L. B. Jorde. 2010. Genetic evidence for high-altitude adaptation in Tibet. Science 329: 72-75. Veit, H. P., & Farrell, R. L. 1978. The anatomy and physiology of the bovine respiratory system relating to pulmonary disease. The Cornell Veterinarian. 68:555-581. Wang, J., Y. Zhang, C. Marian and H. W. Ressom. 2012. Identification of aberrant pathways and network activities from high-throughput data. Briefings in Bioinformatics 13:406-419. West, J. B. 2004. The physiologic basis of high-altitude diseases. Ann. Internal Med. 141:789-800. Will, D. H., & Alexander, A. F. 1970. High mountain (brisket) disease. Bovine Medicine and Sur- gery. WJ Gibbons, EJ Catcott, and JF Smith- cors, ed. Am. Vet. Publ., Wheaton, IL, 412- 430. Will, D. H., Hicks, J. L., Card, C. S., & Alexander, A. F. 1975. Inherited susceptibility of cattle to high-altitude pulmonary hypertension. J. Appl. Phys. 38:491-494. Yi, N. and S. Xu. 2008. Bayesian LASSO for quantitative trait loci mapping. Genetics 179:1045- 1055.

43 PAST AWARD FRANK BAKER MEMORIAL SCHOLARSHIP RECIPIENTS

Name University Year Kelly W. Bruns Michigan State University 1994 William Herring University of Georgia 1994 D. H. “Denny” Crews, Jr. Louisiana State University 1995 Dan Moser University of Georgia 1995 D. H. “Denny” Crews, Jr. Louisiana State University 1996 Lowell S. Gould University of Nebraska–Lincoln 1996 Rebecca K. Splan University of Nebraska–Lincoln 1997 Patrick Doyle Colorado State University 1998 Shannon M. Schafer Cornell University 1998 Janice M. Rumph University of Nebraska–Lincoln 1999 Bruce C. Shanks Montana State University 1999 Paul L. Charteris Colorado State University 2000 Katherine A. Donoghue University of Georgia 2000 Khathutshelo A. Nephawe University of Nebraska–Lincoln 2001 Janice M. Rumph University of Nebraska–Lincoln 2001 Katherine A. Donoghue University of Georgia 2002 Khathutshelo A. Nephawe University of Nebraska–Lincoln 2002 Fernando F. Cardoso Michigan State University 2003 Charles Andrew McPeake Michigan State University 2003 Reynold Bergen University of Guelph 2004 Angel Rios-Utrera University of Nebraska–Lincoln 2004 Matthew A. Cleveland Colorado State University 2005 David P. Kirschten Cornell University 2005 Amy Kelley Montana State University 2006 Jamie L. Williams Colorado State University 2006 Gabriela C. Márquez Betz Colorado State University 2007 Yuri Regis Montanholi University of Guelph 2007 Devori W. Beckman Iowa State University 2008 Kasey L. DeAtley New Mexico State University 2008 Scott Speidel Colorado State University 2009 Lance Leachman Virginia Polytechnic Institute 2009 Kent A. Gray North Carolina State University 2010 Megain Rolf University of Missouri 2011 Brian Brigham Colorado State University 2011 Kristina Weber University of California-Davis 2012 Jeremy Howard University of Nebraska–Lincoln 2012 Heather Bradford Kansas State University 2013 Erika Downey Texas A&M 2013 44 FRANK BAKER MEMORIAL SCHOLARSHIP MEMORIAL ROY A. WALLACE FUND

The Roy A. Wallace BIF Memorial Fund was established to honor the life and career of Roy A. Wallace. Mr. Wallace worked for Select Sires for 40 years, serving as vice-president of beef programs and devoted his life to beef-cattle improvement. He became involved with BIF in its infancy and was the only person to attend each of the first 40 BIF conventions. He loved what BIF stood for - an organization that brings together purebred and commercial cattle breeders, academia, and breed associations, all committed to improving beef cattle. Wallace was honored with both the BIF Pioneer Award and the BIF Continuing Service Award and co-authored the BIF 25-year history, Ideas into Action.

This scholarship was established to encourage young men and women interested in beef cattle improvement to pursue those interests as Mr. Wallace did, with dedication and passion.

Proceeds from the Roy A. Wallace Beef Improvement Federation Memo- rial Fund will be used to award scholarships to graduate and undergraduate students currently enrolled as full-time students in pursuit of a degree related to the beef cattle industry. Criteria for selection will include demonstrated commitment and service to the beef cattle industry. Preference will be given to students who have demonstrated a passion for the areas of beef breeding, genetics, and reproduction. Additional consideration will include academic performance, personal character, and service to the beef cattle industry. Two scholarships will be offered in the amount of $1250 each. One will be awarded to a student currently enrolled as an undergraduate and one will be awarded to a student currently enrolled in a Master of Science or Doctoral program. (From BIF Website/www.beefimprovement.org).

45 PAST AWARD ROY A. WALLACE MEMORIAL FUND RECIPIENTS

UNDERGRADUATES Name University Year Sally Ruth Yon South Carolina 2010 Cassandra Kniebel Kansas State University 2011 Natalie Laubner Kansas State University 2012 Tyler Schultz Kansas State University 2013

GRADUATES Name University Year Paige Johnson Texas Tech University 2010 Jessica Bussard University of Kentucky 2011 Ky Pohler University of Missouri 2012 Loni Woolley Texas Tech University 2013

46 TOURS

ROMAN L. HRUSKA U.S. MEAT ANIMAL RESEARCH CENTER Clay Center Nebraska

Roman L. Hruska U.S. Meat Animal Research Center (MARC) was penned into existence when Congress officially transferred a significant portion of the Naval Ammunition Depot property over to USDA on June 16, 1964. USMARC has had a storied history of research in support of the cattle, swine and sheep industries and today the Center is one of the largest research locations in USDA’s Agricultural Research Service (ARS).

The USMARC is celebrating its 50th anniversary in 2014. USMARC scientists have contributed to the annual BIF meeting since its beginning.

GeneSeek Lincoln Nebraska

GeneSeek, coupled with its Igenity bioinformatics program, provides producers with the information they need to make the best breeding and management decisions early on, saving time and money. With genetic information, producers can more accurately predict key traits, such as marbling and ribeye area.

CIRCLE FIVE BEEF, INC. Henderson Nebraska Alan Janzen, owner

This feeding company started in 1972 with Alan, his father, uncle and three other area producers. The cen- tralized location at Circle 5 Beef makes it easy to access both feed and packers in central Nebraska.

47 PROCEEDINGS GENERAL SESSION I: FOCUS ON THE COWHERD

ECONOMIC CONSIDERATIONS FOR values, and these high prices are not reflected in the THE COW HERD dataset. In fact, the average weaned calf price at 507 C.P. Mathis1, C.T. Braden1, R.D. Rhoades1, and K.C. pounds was only $119/cwt; and is much lower than McCuistion1 current prices. This does not discount the information 1King Ranch® Institute for Ranch Management for those interested in maximizing profit because drivers of profit remain the same regardless of the actual Texas A&M University-Kingsville price of calves. Average net income during this period Introduction was below breakeven (-$65/cow exposed). Cow-calf producers are continually challenged It is discouraging that operations in the bench- to maintain the profitability of their operations despite mark dataset were not profitable on average, but upon the dynamic nature of weather patterns, cattle mar- closer evaluation there are still a portion of the op- kets, and the cost of input commodities and services. erations that were profitable. In fact, some cow-calf Good managers make a multitude of small decisions enterprises were highly profitable (figure 1). Produc- to collectively keep costs low relative to the value of tion systems can vary greatly; however, those herds the weaned calves they produce. However, the real in the top net income quartile (average profit = $159/ separation between “good” and “excellent” manage- cow exposed) generated not only greater gross income ment is that the very best managers also understand from calf sales relative to the other three-fourths of the and find leverage in the production system that have 44 herds (figure 2), but also had the lowest produc- long-standing systematic benefit to the operation. tion costs. The bottom line is that highly profitable Those producers with a clear view of the financial herds typically return more income and have lower position of the ranch and the drivers of net income costs. Producers interested in being among the top net and return on assets will be best prepared to make the income quartile are encouraged to continuously ask high leverage decisions with long-term benefit to the themselves: operation. 1) What are the most profitable herds doing This paper discusses the impact of key cow that makes them different? herd performance criteria on the net income of cow- 2) How can I improve profit the most in my calf enterprises, and is intended to help managers operation? prioritize the areas in their unique operation that will likely yield the largest improvement in profitability A Closer Look at Revenue if altered. Standardized Performance Analysis (SPA) The two sources of revenue for cow-calf oper- benchmark information is used as a basis to estimate ations are calf sales and cull cow and bull sales, with the impact of some management decisions on key cow calf sales being the most important. Calf income is a herd performance criteria and net income. function of quantity (number sold), quality (genetics and condition), and marketing. Table 2 shows calf What is Driving Net Income? weaning measures and revenues by net income quar- Benchmark data from the SPA database offers tile to provide insight into some of the differences that some historical insight into the key performance and exist among profitable and unprofitable operations. financial measures affecting profit of cow-calf enter- Weaning percentage does not show an upward lin- prises. It is also noteworthy that current SPA bench- ear trend parallel to rising net income. This does not mark information only offers regional information mean that weaning percentage is unimportant, but em- from the southwest (TX, OK, and NM; Stan Bevers, phasizes that top net income quartile operations have a personal communication). Table 1 is the Southwest balance between cost and performance that maximize SPA Key measures summary for 44 herds from 2008 net income. The top quartile does not have the highest to 2013. These herds ranged in size from 44 to 2,963 weaning percentage, but these operations have weaned head and represent 17,196 cow years. Calf prices the largest calves by 58 pounds over the second high- in 2013 and 2014 have reached exceptionally high est profit quartile. The advantage in weaning weight 48 PROCEEDINGS GENERAL SESSION I: FOCUS ON THE COWHERD

primarily results from calves in the top quartile being reducing purchase price of breeding stock, increasing approximately 20 days older at weaning (data not salvage values, or increasing longevity of cows and shown). Although not quantifiable from SPA data, it is bulls. Reducing equipment depreciation may be ac- likely that calves from the top quartile operations also complished by sharing, renting, leasing, or contracting have an additional advantage in genetics for growth equipment. However, each of these options has some and/or end product value. tradeoffs in convenience and control. Unlike livestock The overall average weaning rate and weight depreciation, which is a direct cost, the expense of were 83.8 percent and 507 pounds, respectively (table equipment, buildings, and fences depreciation is an 2). Using these values as a foundation, and assuming indirect or overhead cost. While capital purchases and that 507-pound calves are worth $119/cwt (average improvements may have the potential to improve effi- SPA price from 2008-2012), the value of a single ciency and production, the increase of the associated percentage unit change in weaning rate is about $6/ depreciation expense may offset the gain in efficiency cow exposed (calculation: 507 lbs * 1% * $119/cwt from the improvement. The most profitable operations = $6.03). If a more current 507-pound calf price of generally find ways to reduce this depreciation burden $200/cwt is assumed, a single unit increase in weaning as much as possible. percentage raises profit by more than $10/cow exposed. Therefore, any management change that cost Putting the Performance and Financial Pieces To- less than $10/cow exposed to implement and increas- gether es weaning rate by one percentage unit or more will A cow calf enterprise is a complex biological increase net income. system where inputs and outputs are interconnected. Managers interested in maximizing profit are encour- A Closer Look at Expenses aged to focus on optimizing weaning rate and weaning Total cost before non-calf revenue adjust- weight, as well as feed, labor, and depreciation ex- ment averaged $608/cow exposed (table 1), but when penses. However, there is no silver bullet or prescrip- evaluated by net income quartile, the quartile aver- tion that is most effective at accomplishing the perfect age ranged $451 to $700. The top quartile producers balance because of the vast differences in resources simply wean and market more pounds of calf/cow and goals from one ranching operation to the next. exposed at a much lower cost than the less profitable The key is to evaluate potential changes based on unit operations. Figure 3 shows that over half of the ex- cost of production. This measure will merge inputs penses to a cow-calf enterprise can be categorized as and outputs into a single value. In reality, only a small depreciation, labor, or feed. In most cow-calf enter- portion of cow-calf enterprises have an accounting and prises these three expense categories offer opportunity performance measurement system in place to accurate- for high leverage change to the production system that ly calculate unit cost of production. Implementation can yield significant financial improvement. Other of a managerial accounting system should be the ini- expenses like repairs and maintenance, fertilizer, fuel, tial step to improving profit because a clear picture of leases, and veterinary services are important when tak- the current financial status of the operation is needed en together, but independently are generally not high to make the best business decisions for the future. leverage expenses. It will take many small decisions across all Feed and labor expenses are typically well facets of the business to keep cost low, yet still achieve understood, but depreciation is an expense often more performance goals. However, in most systems there difficult to grasp. The result is a considerable amount are a few high-leverage interventions that can make a of unaccounted expense in livestock, equipment, and big impact. These changes will not be the same on all infrastructure depreciation. Managers should be aware operations, but all managers should seek to find these of the effect depreciation of livestock, equipment, and areas in the operation that if changed could yield dra- infrastructure has on the long term equity of an opera- matic improvement. Table 3 lists examples of chang- tion. The ways to decrease livestock depreciation are: es that may have a significant long-standing benefit

49 to an operation. These interventions are included as HEIFER INTAKE AND EFFICIENCY AS examples only, and are not intended to be generalized INDICATORS OF COW INTAKE AND recommendations for all operations. Notice that labor, EFFICIENCY depreciation, and pounds weaned are all affected in Daniel W. Shike1, Chris J. Cassady1, J. W. Adcock1, almost every intervention. A number of other exam- and Keela M. Retallick2 ples could also be included, especially those that affect 1 genetic makeup of the cowherd, which is always a Univeristy of Illinois at Urbana-Champaign 2 long-standing change. California Polytechnic State University, San Luis Cash Flow. Without minimizing the impor- Obispo tance of previously discussed financial principles, operating capital is essential. A yearly financial plan Introduction with projected monthly cash flows adjusted accord- Feed costs account for over 60% of the total ing to operational plans is invaluable in preventing costs associated with maintaining a beef cow and are un-expected asset liquidation out of necessity. Not the largest detriment to profitability for beef producers being able to service short-term liabilities can lead to (Miller et al., 2001). Approximately 60 – 70% of en- the liquidation of revenue producing assets, resulting ergy for beef production is required by the cow herd. in long-term reduced profit potential. While the value Of the energy needed for the cow herd, approximately of cows liquidated today is capitalized on, the value 70% goes to maintenance energy (Ferrell and Jenkins, of future production is lost. The importance of not 1982). Thus, nearly 50% of all energy required by the liquidating assets in order to operate cannot be over beef industry is used simply to maintain the cow herd. emphasized. Although variation in maintenance energy appears to exist, maintenance requirements of cattle have shown Conclusions little no change over the past 100 years (Johnson et The most profitable cow-calf operations are al., 2003). Limited work has been done evaluating the efficient, generally weaning the most pounds of calf relationship between heifer intake and performance per cow exposed with the lowest breakeven. Most im- during the postweaning growing period and cow per- portantly, these operations yield the greatest return on formance and reproduction traits. assets. Success in the cattle industry does not happen The objective of this study was to determine on accident. Decision makers at the most profitable the relationship between residual feed intake (RFI), operations have built production and marketing sys- residual body weight gain (RG), and intake in heifers tems that, most importantly minimize labor, feed, and during the postweaning period and subsequent cow depreciation expenses relative to weaned calf value. performance and reproduction as 2-year-old lactating Producers interested in improving the profitability of and dry cows. their cow-calf operation are encouraged to utilize a managerial accounting system that maintains a clear Materials and Methods picture of the operation financials and allows measure- Postweaning heifer evaluation ment of unit cost of production. Furthermore, manag- A postweaning intake and performance eval- ers should seek practical, high leverage alterations to uation was conducted on Angus and Simmental x the production system with a keen focus on optimizing Angus heifers (n=511) over a 5-yr period at the Beef weaning rate and weaning weight, as well as feed, Field Research Laboratory in Urbana, IL. Heifers were labor, and depreciation expenses. developed on a diet consisting of approximately 70% corn silage, 25% corn co-products, and 5% supplement each year. Heifer intake and performance were monitored for a minimum of 70 d each year; according to BIF standards. Individual intakes were recorded using the GrowSafe® automated feeding system. For years 1, 2, and 3, cattle were weighed on 2 consecutive days at the beginning and end of the test period, and ADG was calculated by dividing total BW gain by the number of days on test. Individual animal mid- 50 test metabolic weight (MWW) was determined by the individual cow relative to several production traits. At average of the beginning and end weights of the test 60 d postpartum, twenty-four hour milk production period. For years 4 and 5, cattle were weighed on 2 estimates were determined using a 12-hr weigh-suck- consecutive days at the beginning and end of the test le-weigh technique (Beal et al., 1990). Individual and biweekly throughout. Heifer ADG was calculat- intake was measured during each evaluation period by ed by regressing each individual weight over all time using the GrowSafe® automated feeding system. At points of the test. Individual MWW was determined the conclusion of each evaluation period, weights were by taking the mid-date test weight via the regression taken on two consecutive days, hip height recorded, equation. Individual animal 12th rib fat thickness (BF) BCS scored (1-9 scale) by a trained technician, and was recorded via ultrasound on years 4 and 5. cows were ultrasound for BF. Heifer RFI and RG were determined for each Calves were weaned at approximately 6 mo of individual animal. For all years, animals were sepa- age. Weaning weights were recorded and submitted to rated into contemporary groups, based on breed type the American Angus Association and American Sim- and source of origin. For years 1, 2, and 3, RFI was mental Association. An adjusted weaning weight was assumed to represent the residuals from a multiple then calculated by the associations. As a measurement regression model regressing DMI on ADG and MWW, of cow efficiency during the lactating period, a cow using pen as a random effect, and RG was assumed RFI value was calculated for each cow. Cow RFI was to represent the residuals from a multiple regression assumed to represent the residuals from a multiple model regressing ADG on DMI and MWW, using regression model regressing DMI on metabolic weight pen as a random effect. For years 4 and 5, RFI was (MW), BF, and 24-hour milk production. assumed to represent the residuals from a multiple re- Statistical Analysis gression model regressing DMI on ADG, MWW, and The MIXED procedure of SAS was used to BF using pen as a random effect, and RG was assumed test the effect of heifer intake and efficiency classifica- to represent the residuals from a multiple regression tion on cow production traits. The model used includ- model regressing ADG on DMI, MWW, and BF using ed the fixed effect of RFI, RG, or intake classification pen as a random effect. group (high, medium, and low.) The GLIMMIX proce- Heifers were classified as low, medium, or dure of SAS was used to test the effect of heifer intake high RFI, RG, or intake. Classification groups were and efficiency classification on reproductive traits established by calculation of the mean and SD of the (binomial data). The model used included the fixed heifers for RFI, RG, and intake. Heifers that were less effect of RFI, RG, or intake classification group (high, than 0.5 SD below the mean were classified as “Low,” medium, and low). Mean values were considered to be heifers that were ± 0.5 SD of the mean were classified significantly different when P < 0.05 and considered a as “Med,” and heifers that were more than 0.5 SD tendency when P > 0.05 and < 0.10. above the mean were classified as “High.” Heifers with structural soundness problems Results and Discussion or very poor performance were culled annually prior Heifers were classified into Low, Med, or to the breeding season. Heifers (n=366) kept as re- High RFI groups, and the effects of the RFI classifi- placements were synchronized and AI. Heifers were cation on female reproductive and performance traits exposed to clean-up bulls for 60 d following AI. are presented in table 1. There were no differences in Reproductive data were collected for first service AI percentage of females kept as replacements, first AI conception and overall pregnancy rates. Calving data conception rate, overall pregnancy rate, or age at calv- was recorded to determine age of cow (days) at first ing between the RFI classifications. The heifer RFI calving and calf birth weight. classification did not affect calf birth weight or wean- 2-year-old cow evaluation ing weight. Heifer RFI classification did not affect Each year, cows were placed in the barns at cow BW, hip height, BF or milk production at 60 d the Beef Field Research Laboratory in Urbana, IL for postpartum, but there was a trend (P = 0.08) for cows two 14 d evaluation phases (60 d (lactating) and 240 d from the Med RFI group to have decreased BCS at 60 (dry) postpartum) where they were fed a common for- d postpartum compared to cows from the High RFI age based diet (~60% TDN). During these evaluation group. Cows classified as Med or High RFI had great- periods, measurements were taken to characterize each er (P < 0.01) DMI than cows in the Low RFI group 51 Table 1. Effects of RFI classification on female reproductive and performance traits Heifer RFI Category Item Low Med High SEM P-value Reproductive traits Retained as replacement, % 69 76 71 - 0.36 First AI conception rate, % 45 50 42 - 0.50 Overall pregnancy rate, % 86 83 85 - 0.80 Cow age at first calf, d 736 734 741 3 0.16 Calf performance1 Calf birth weight, lb 73 73 75 1 0.51 Calf weaning weight, lb 598 586 618 12 0.12 2-year-old cows (lactating)2 Cow BW, lb 1270 1257 1272 14 0.68 Cow hip height, in 52.6 52.8 52.9 0.2 0.54 Cow BCS 5.7xy 5.6x 5.7y 0.1 0.08 Cow BF, in 0.25 0.24 0.25 0.01 0.91 24 h milk production, lb 18 17 18 1 0.70 Cow DMI, lb 32.4a 35.9b 36.9b 1.1 <0.01 Cow RFI, lb -1.67a 0.56b 1.09b 0.65 <0.01 2-year-old cows (dry)3 Cow BW, lb 1378 1368 1384 14 0.67 Cow hip Height, in 53.5 53.5 53.5 0.2 0.99 Cow BCS 5.8 5.8 5.9 0.1 0.81 Cow BF, in 0.27 0.27 0.28 0.01 0.55 Cow DMI, lb 29.0x 30.9xy 33.4y 1.3 0.06 a,b Row means that do not have a common superscript differ, P < 0.05 x,y Row means that do not have a common superscript tend to differ, P > 0.05 and < 0.10 1 Progeny of 2-year-old cows 2 2-year-old cow traits measured at 60 d postpartum 3 2-year -old cow traits measured at 240 d postpartum

at 60 d postpartum. Cows classified as Med and High = 0.06) to reach puberty at a younger age than Med RFI heifers had greater Cow RFI than cows that were or Low RFI but this did not result in any differenc- classified as Low RFI heifers; heifers that ate less than es among RFI classifications for conception rate or predicted during the postweaning evaluation also ate pregnancy. Crowley et al. (2011) reported a negative less than predicted as 2-year-old lactating cows. There genetic correlation between RFI in growing males and were no differences in cow BW, hip height, BCS, or cow age at first calving but did not find any correla- BF at 240 d postpartum among heifer RFI classifications with fertility or calving difficulty. Crowley et tion groups; however, there was a trend (P = 0.06) for al. (2011) also found a negative genetic correlation cows from the High RFI group to have increased DMI between growing male RFI and cow BW but reported compared to cows from the Low RFI group. no correlation between RFI and maternal weaning There has been limited work done evaluat- weight. Black et al. (2013) found that heifers classified ing the effects of efficiency during the postweaning as Med or High RFI had greater DMI as cows than period on cow performance and reproduction. Shaffer heifers classified as Low RFI. et al. (2011) reported that High RFI heifers tended (P 52 Heifers were also classified into Low, Med, or High heifer group were intermediate. Heifer intake classi- RG groups, and the effects of the RG classification on fication did not affect milk production, BCS, or BF at female reproductive and performance traits are shown 60 d postpartum. Cows from the High Intake group in table 2. There were no differences in percentage of had increased (P < 0.01) DMI compared to cows from females kept as replacements, first AI conception rate, the Low Intake group, and cows from the Med Intake overall pregnancy rate, or age at calving between the group were intermediate. Cows from the High Intake RG classifications. The RG classification also did not group also had greater (P = 0.04) cow RFI than the affect calf birth weight or weaning weight. Heifer RG cows from the Low Intake group. Results at 240 d classification did not affect cow BW, BCS, BF, milk postpartum were very similar to results at 60 d post- production, DMI, or cow RFI at 60 d postpartum, but partum. Cows from the Med and High Intake groups there was a trend (P = 0.06) for cows from the High again had greater (P < 0.01) BW at 60 d postpartum RG group to have increased hip heights compared than cows from the Low Intake group. Cows from the to the cows from the Low RG group. There were no High Intake group also again had increased (P < 0.01) differences in cow BW, hip height, BCS, BF, or DMI hip height than cows from the Low Intake group, and at 240 d postpartum among heifer RG classification cows from the Med Intake group were intermediate. groups. Heifer intake classification did not affect BCS or BF Crowley et al. (2011) found that RG in grow- at 240 d postpartum either. Similar to 60 d postpartum, ing males was genetically correlated to age at first cows from the High Intake heifer group had increased calving. Crowley et al. (2011) also reported that grow- (P = 0.02) DMI compared to cows from the Low In- ing male RG was genetically correlated to cow BW take group, and cows from the Med Intake group were and maternal weaning weight (0.67 and 0.57, respec- intermediate. tively). Crowley et al. (2011) reported a negative Heifers were also classified into Low, Med, correlation between growing male concentrate intake or High intake groups, and the effects of the intake and cow age at first calving. Crowley et al. (2011) also classification on female reproductive and performance found a positive correlation between growing male traits are shown in table 3. There were a greater (P < concentrate intake and calving difficulty, cow BW, and 0.01) percentage of heifers retained as replacements maternal weaning weight. from the groups classified as Med or High Intake heifers compared to the heifers classified as Low Conclusions Intake. Heifers were culled prior to breeding for either Results from this study suggest that heifers that structural soundness problems or very poor perfor- are more efficient based off of RFI will consume less mance. We speculate that that the difference in per- DMI as cows with no differences in cow or calf percentage of heifers retained as replacements is likely a formance or reproduction. There were no differences reflection of some of the low intake heifers being the detected between RG and cow performance or repro- smaller, poorer gaining heifers. There were no differ- ductive traits. Heifers that have greater DMI calve at ences in first AI conception rate or overall pregnancy an older age, have larger BW and greater hip height rate; however, heifers classified as Low Intake were as 2-year-old cows, and have increased DMI as cows. younger (P = 0.04) at calving then the heifers classi- Further evaluation of the relationship of heifer intake fied as High Intake. Cows that were classified as High and efficiency measures on cow production traits after Intake heifers had calves with greater (P < 0.01) birth 2 years of age is needed. weights than cows that were classified as Low or Med Intake heifers. However, there were no differences Literature Cited in calf weaning weights among cows from different Beal, W. E., D. R. Notter, and R. M. Akers. 1990. heifer intake classification groups. Cows from the Med Techniques for estimation of milk yield in and High Intake groups had greater (P = 0.02) BW beef cows and relationships of milk yield to calf at 60 d postpartum than cows from the Low Intake weight gain and postpartum reproduction. J. Anim. group. Cows from the High Intake group had in- Sci. 68:937-943. creased (P < 0.01) hip height than cows from the Low Intake heifer group, and cows from the Med Intake

53 Table 2. Effects of RG classification on female reproductive and performance traits Heifer RG Category Item Low Med High SEM P-value Reproductive traits Retained as 71 71 74 - 0.80 replacement, % First AI con- 46 44 47 - 0.91 ception rate, % Overall preg- 86 83 85 - 0.72 nancy rate, % Cow age at 71 71 74 - 0.80 first calf, d Calf performance1 Calf birth 73 74 74 1 0.84 weight, lb Calf weaning 596 600 601 12 0.94 weight, lb 2-year-old cows (lactating)2 Cow BW, lb 1261 1256 1280 14 0.42 Cow hip 52.5x 52.7xy 53.0y 0.2 0.06 height, in Cow BCS 5.7 5.6 5.6 0.1 0.91 Cow BF, in 0.26 0.24 0.24 0.01 0.45 24 h milk 18 17 18 1 0.47 production, lb Cow DMI, lb 35.1 35.0 35.2 1.1 0.99 Cow RFI, lb 0.92 -0.30 -0.54 0.63 0.21 2-year-old cows (dry)3 Cow BW, lb 1372 1359 1398 14 0.12 Cow hip 53.3 53.4 53.7 0.2 0.14 height, in Cow BCS 5.9 5.8 5.8 0.1 0.83 Cow BF, in 0.28 0.27 0.27 0.01 0.67 Cow DMI, lb 29.6 31.7 31.9 1.3 0.36 x,y Row means that do not have a common superscript tend to differ, P > 0.05 and < 0.10 1 Progeny of 2-year-old cows 2 2-year-old cow traits measured at 60 d postpartum 3 2-year -old cow traits measured at 240 d postpartum

54 Table 3. Effects of intake classification on female reproductive and performance traits Heifer Intake Category Item Low Med High SEM P-value Reproductive traits Retained as replacement, % 57a 80b 76b - <0.01 First AI conception rate, % 51 44 45 - 0.62 Overall pregnancy Rate, % 84 84 86 - 0.87 Cow age at first calf, d 731a 738ab 741b 3.1 0.04 Calf performance1 Calf birth weight, lb 71a 73a 77b 1.4 <0.01 Calf weaning weight, lb 605 590 607 13.8 0.47 2-year-old cows (lactating)2 Cow BW, lb 1225a 1273b 1285b 16.2 0.02 Cow hip height, in 52.1a 52.8b 53.2c 0.2 <0.01 Cow BCS 5.6 5.7 5.7 0.1 0.75 Cow BF, in 0.24 0.26 0.24 0.01 0.25 24 h milk production, lb 18 18 17 0.8 0.73 Cow DMI, lb 30.2a 35.4b 38.4c 1.2 <0.01 Cow RFI, lb -1.24a -0.20ab 1.19b 0.74 0.04 2-year-old cows (dry)3 Cow BW, lb 1305a 1377b 1409b 18.6 <0.01 Cow hip height, in 52.9a 53.5b 53.9c 0.2 <0.01 Cow BCS 5.7 5.8 5.9 0.1 0.24 Cow BF, in 0.27 0.27 0.29 0.01 0.44 Cow DMI, lb 27.3a 30.7ab 33.1b 1.5 0.02 a,b,c Row means that do not have a common superscript differ, P < 0.05 1 Progeny of 2-year-old cows 2 2-year-old cow traits measured at 60 d postpartum 3 2-year -old cow traits measured at 240 d postpartum

Black, T. E., K. M. Bischoff, V. R. G. Mercadante, Ferrell, C. L., and T. G. Jenkins. 1982. Efficiency G. H. L. Marquezini, N. DiLorenzo, C. C. of cows of different size and milk production Chase, Jr., S. W. Coleman, T. D. Maddock and G. potential. Pages 12–24 in USDA, ARS, Germ- C. Lamb. 2013. Relationships among performance, plasm Evaluation Program Progress Report No. residual feed intake, and temperament assessed in 10.MARC, Clay Center, NE. growing beef heifers and subsequently as 3-year- Johnson, D. E., C. L. Ferrell, and T. G. Jenkins. 2003. old, lactating beef cows. J. Anim. Sci. 91:2254- The history of energetic efficiency research: Where 2263. have we been and where are we going? J. Anim. Crowley, J. J., R. D. Evans, N. McHugh, D. A. Kenny, Sci. 81:E27–E38 M. McGee, D. H. Crews, Jr., and D. P. Berry. Miller, A. J., D. B. Faulkner, R. K. Knipe, D. R. Stroh- 2011. Genetic Relationships between feed efficien- behn, D. F. Parrett, and L. L. Berger. 2001. cy in growing males and beef cow performance. J. Critical control points for profitability in the cow- Anim. Sci. 89:3372-3381. calf enterprise. Prof. Anim. Sci. 17:295-302. Shaffer, K. S., P. Turk, W. R. Wagner, and E. E. D. Felton. 2011. Residual feed intake, body composition, and fertility in yearling beef heifers. J. Anim. Sci. 2011:1028-1034. 55 BEEF HEIFER DEVELOPMENT AND Heifer Development System and Pubertal Status LIFETIME PRODUCTIVITY1 Association among BW, puberty, and heifer R. L. Endecott1, R. N. Funston2, J. T. Mulliniks3, and A. pregnancy rate appears to have changed over time J. Roberts4 (Funston et al., 2012). Earlier research demonstrat- 1Department of Animal and Range Sciences, Montana ed limiting post-weaning growth negatively affected State University, Bozeman 59717 2University of Ne- age of puberty and pregnancy rates, whereas more braska West Central Research and Extension Center, recent studies demonstrate less of a negative impact North Platte 69101 of delayed puberty on pregnancy rate. Funston et al. 3Department of Animal and Range Sciences, New (2012) hypothesized that changes over time may have Mexico State University, Las Cruces 88003 4US- resulted from: the shift from calving heifers at 3 yr of age DA-ARS, Fort Keogh Livestock and Range Research 1) to calving at 2 yr of age and subsequent Laboratory, Miles City, MT 59301 selection pressure for decreased age at puberty; Introduction The heifer development paradigm is adapting 2) genetic changes in age of puberty to less traditionally inexpensive feed available and resulting from selection for bull scrotal changes in cattle genetics over the last 40 years, mak- circumference; and ing it critical to understand how management practices 3) perhaps a change in fertility of pubertal affect lifetime production efficiency. Increased feed estrus compared with subsequent estrous costs have negatively impacted heifer development cycles. protocols that rely heavily on harvested feeds. Much of the research leading to the paradigm of developing Other factors may also contribute to the change heifers to a target BW of 60 to 65% mature BW at over time. Establishment and use of EPDs in select- breeding was conducted during the late 1960s through ing for growth, milk, and carcass characteristics have the 1980s. However, trait selection based on EPDs contributed to changes in reproductive performance has created substantial genetic change in the last 40 due to genetic associations with these and other traits years. This impact of genetic change on heifer devel- (American Angus Association, 2012; American Here- opment has not been widely considered. Research in ford Association, 2014; American International Cha- the last decade has compared traditional, more inten- rolais Association, 2014). For example, genetic trend sive systems with systems using less feed and relying for increased mature weight would be expected to on compensatory gain. These studies provide evi- correspond with an increase in BW at puberty. Results dence that developing heifers to a lighter target BW at summarized in Figure 1 illustrate that BW at time of breeding, that is, 50 to 57% of mature BW compared puberty has increased over time. Although information with 60 to 65% BW, reduced development costs while concerning mature size is not provided in most studies not impairing reproductive performance (Funston and represented in Figure 1, the progression from a mature Deutscher, 2004; Roberts et al., 2009; Funston and size of 1,100 lb in the initial studies to 1,300 lb in the Larson, 2011; Mulliniks et al., 2012). However, much most recent studies may be reasonable. Even though heifer development research is limited in its consider- different management and feeding practices were ation of long-term applications. Longevity has a rel- implemented within and among studies summarized in atively low heritability; thus, heifer development and Figure 1, the data indicate a majority of heifers would other management strategies have a greater potential achieve puberty at or below 60% mature BW, assum- to impact cow retention in the breeding herd. While ing mature BW of 1,100; 1,200; or 1,300 lb for heif- limited information exists about the impacts of heifer ers used in the 3 time periods. Data in Figure 1 also development strategies on cow longevity, data from indicate average age of puberty was prior to 430 d of non-ruminant and non-livestock species implies that age, which would correspond to the start of breeding limiting caloric intake during juvenile development in order to begin calving at 2 yr of age. Furthermore, it can increase lifespan (Speakman and Hambly, 2007). is expected that selection and management processes implemented over time have contributed to a greater proportion of heifers achieving puberty at lower target 1 Adapted from Endecott et al. (2013). BW. 56 Figure 1. Body weight (BW) and age of heifers at puberty in studies over the last 5 decades where heifers were developed on 2 or more levels of growth during the post weaning period. Data from 1960 to 1971 are depicted with black diamonds (Wiltbank et al, 1966 and 1969; Short & Bellows 1971). Data from 1972 to 1987 are noted with black squares (Ferrell, 1982; Greer et al., 1983; Byerly et al., 1987). Data from 1990-2009 are shown as black triangles (Hall et al., 1995; Lynch et al., 1997; Freetly et al., 1997; Ciccioli et al., 2005; Roberts et al., 2009). The data indicate that BW at puberty has increased over the time periods that different studies were conducted. Horizontal lines represent BW representing 60, 55, and 51% of 1,300 lb mature BW; 60 and 55% of 1,200 lb BW; and 65 and 60% of 1,100 lb BW. The black vertical line at 430 d of age represents the age to start breeding in order to calve at 2 yr of age. Not all heifers achieved puberty in the time frame encompassed by some of the studies depicted. However, the data indicate genetic potential of heifers under different management strategies to achieve puberty at or below 60% of a mature BW predicted to be representative of cows for each time period.

Fertility of the pubertal estrus is another heifers assigned to be bred on the third estrus. Thus, component of the heifer development paradigm that the implications of data from the first estrus group needs to be reevaluated. Industry recommendation that bred at an average age of less than 11 months for the heifers be developed so they experience puberty prior industry where the majority of heifers would tradition- to start of breeding is derived from results of Byerly et ally be bred 13 to 15 mo of age is questionable. Re- al. (1987) who observed 21% lower pregnancy rate in cently, research reported 6% lower pregnancy rates in heifers bred on their first estrus compared with heifers heifers that were not pubertal at the start of the breed- bred on their third estrus. However, mean age and BW ing season compared with heifers that were pubertal of heifers at the time of breeding were confounded by (Roberts et al., 2013; Vraspir et al., 2013). Although estrus status classification. Mean age at breeding for these results are not a direct assessment of first estrus heifers bred at first estrus was 322 d, whereas heifers fertility, the results indicate the magnitude of infertility bred on third estrus averaged 375 d old. Furthermore, is not near the extent indicated in the original study by age of breeding accounted for increased pregnancy in Byerly et al. (1987). heifers classified to be bred at first estrus, but not in 57 Nutrition Following the Start of Breeding and calving season and calf birth date. Decreased winter Through Subsequent Calvings gain in the low input development systems resulted in Establishing impact of heifer development greater gain during the breeding season, which may protocols on longevity is complex, requiring consid- explain similar overall pregnancy rates. eration for nutritional factors following the start of If nutrition following start of breeding is inad- breeding through subsequent calvings. Resulting main- equate, poor reproductive performance may result. tenance requirements and behavior traits associated White et al. (2001) found restricting nutrients to 40% with development protocols must be considered. Most of maintenance prevented ovulation in 70% of heifers longer-term heifer development studies manage re- with no change in BCS. Perry et al. (2009) reported placement heifers as a group on breeding pastures after decreased pregnancy success for heifers moved from development. Heifers developed under conditions of feedlot to summer grazing post-AI. Post-insemination dormant or scarce forage, low precipitation, undulating nutrition may affect embryonic survival through a terrain, and large pastures, or those that are restricted variety of mechanisms. Nutritionally-mediated chang- gain, pen-developed often exhibit compensatory gain es to the uterine environment can occur by changing during summer grazing (Olson et al., 1992; Roberts et components of uterine secretions or by influencing the al., 2009; Funston and Larson, 2011; Mulliniks et al., circulating concentrations of progesterone that regu- 2012). Examples of this include comparisons of heif- late the uterine environment (Foxcroft, 1997). Arias et ers developed in a drylot at 1.52 lb/d ADG from initia- al. (2012) determined yearling heifers that gained BW tion of the study to breeding with heifers developed at had greater AI pregnancy rate (77%) than heifers that 0.57 lb/d on a low-quality pasture with protein supple- maintained (56%) or lost (61%) BW during the first mentation (Mulliniks et al., 2012). Development treat- 21-d period post-AI. Therefore, nutritional plane post- ments resulted in 77-lb difference in weight at start AI may be as or more important than pre-breeding of breeding. However, the pasture-developed heifers nutritional plan in yearling heifers. Collectively, the had greater gain (1.83 lb/d) from start of breeding to studies discussed above provide evidence that devel- pregnancy diagnosis than drylot heifers (1.34 lb/d). oping heifers to lighter weights at start of breeding re- Range-developed heifers compensated for their mini- duces maintenance requirements providing them with mal pre-breeding ADG and gained more weight during greater opportunity to be in positive nutrient balance the breeding season than feedlot-developed heifers, in conditions when forage quality is marginally suffi- due to lower maintenance requirements and the ability cient around the time of breeding. to respond to a seasonal improvement in forage quality Differences in size and corresponding main- (Marston et al., 1995; Ciccioli et al., 2005). Pasture de- tenance requirements may persist over time to result veloped heifers tended to have greater pregnancy rates in greater retention in subsequent years. Pregnancy than heifers developed in a drylot (91 vs. 84%). rates through the 4th calf remained similar between Other research (Funston and Larson, 2011; high- and low-gain heifers developed in Nebraska, Larson et al., 2011) compared heifers grazing on corn where nutrition following the development period was residue or winter range as an alternative to drylot feed- considered adequate (Funston and Deutscher, 2004). ing. Heifers grazing corn residue gained 0.5 lb/d more In contrast results from New Mexico, where nutrition than heifers developed on winter grass or a drylot. may have been limiting. Mulliniks et al. (2012) re- Heifers grazing winter grass or corn residue were supported 68% retention in the breeding herd through 5 plemented with the equivalent of 0.31 lb/d of protein yr of age for range-developed heifers fed a high-RUP and gained between 0.42 and 0.93 lb/d during winter (rumen undegradable protein) supplement compared grazing. Once placed on higher quality spring pasture, with 41% retention for range-raised counterparts fed the heifers gained 1.19 to 1.61 lb/d during the breed- a lower-RUP cottonseed meal-based supplement, and ing season. Heifers grazing corn residue weighed less 42% retention for heifers developed in a feedlot. This prior to breeding than heifers developed in the drylot, relationship tended to be significant as early as 2 and had achieved 56% of their mature BW, had similar 3 yr of age, respectively. These data indicate not only pregnancy rates at the end of the breeding season, and where a heifer is developed (i.e., low-input vs. feed- achieved similar BW prior to calving with a similar lot), but also what she is fed when developed (i.e., percentage (> 60%) calving in the first 21 d of the high-RUP vs. lower-RUP supplement) may influence

58 her longevity in the cow herd. study were lighter prior to calving (871 vs 888 lb) and Nutrition through subsequent calvings may prior to start of breeding (818 vs 842 lb) as 2-yr-olds interact with heifer development protocol to influence compared with pregnant heifers from both devel- cow longevity. In the Nebraska and New Mexico stud- opment groups and non-pregnant heifers developed ies discussed above, heifers were managed in common on ad libitum feed. This primary difference between after the respective heifer development treatments. In lower-input heifer development programs emphasizes contrast to the Nebraska and New Mexico data sets, a the importance of managing extensively developed study in Montana evaluated cows provided different heifers for continued growth after lower inputs during levels of feed inputs during post-weaning development post-weaning development. The data also indicate that and subsequent winter supplementation over a 10-year the way the dams are fed may program the heifer fe- period. Each year following weaning, heifers were de- tuses to respond differently to low input development veloped in dry lots on a corn silage-based diet. Heifers later in life. were fed to appetite (control) or restricted to fed 20% Heifer development protocols may influence re- less than controls at similar weight. In subsequent sulting behavior traits associated with the environment winters, control females were provided supplemental in which the heifer was developed. Range-developed feed expected to be adequate for production on win- heifers may retain better grazing skills and be more ter range, whereas restricted heifers were fed level of productive during the subsequent summer (Olson et supplemental feed expected to be marginal for range al., 1992; Perry et al., 2009). In a recent study at 2 conditions. Heifers used in this study were produced locations in Nebraska (Summers et al., 2013), heifers by dams that had received either marginal or adequate were either developed on winter range vs. corn resi- levels of winter supplemental feed, thereby creating 4 due or drylot vs. corn residue. Pregnancy rate based classifications: restricted heifers from dams provided on heifer development system was similar; however, marginal levels of winter supplemental feed; restricted heifers developed on corn residue exhibited greater heifers from dams provided adequate levels of winter ADG when placed on corn residue as a pregnant heifer supplemental feed; control heifers from dams provided compared with either winter range or drylot developed marginal levels of winter supplemental feed; control heifers (Summers et al., 2013), supporting the hypoth- heifers from dams provided adequate levels of winter esis of a learned behavior for grazing corn residue. supplemental feed. All females were required to wean However, drylot-developed heifers that graze dormant a calf each year of production to remain in the herd. forage during the winter prior to development in a pen Retention at year 1 (heifer pregnancy) and at start of may not exhibit a change in grazing skills upon return- the 2nd breeding season were influenced by the inter- ing to a grazing environment. Mulliniks et al. (2012) action of heifer and dam nutritional treatments; being reported similar ADG in drylot-developed heifers greater for restricted heifers from dams on marginal between the drylot phase (1.52 lb/d) and grazing phase level of supplement than restricted heifers from ad- (1.34 lb/d). Data from other species indicates the en- equately supplemented dams. Retention from 2 to 3 vironment experienced during development can have years of age was less for restricted animals than con- lifetime impacts. trols. No differences in loss were observed between Adequate heifer growth and development to 3 and 4 years of age, but control animals incurred ensure minimal calving difficulty can be important greater loss between year 4 and 5 resulting in similar for longevity (Rogers et al., 2004) however, providing percent retention among the different classification additional supplemental feed during post-weaning groups at 5 years of age. Collectively, rebreeding re- development to accomplish this may be less efficient sults from New Mexico and Nebraska would indicate than later in development. Similar calving difficulty that lower-input heifer development where all heifers has been observed between low- and high-gain heifers are managed together after the post-weaning period developed in confinement (Funston and Deutscher, did not impair rebreeding, but continued subsequent 2004), between heifers developed with low-inputs on restriction in the form of marginal winter supplemen- corn residue and winter range and feedlot-developed tation, as experienced by the Montana heifers, result- heifers (Funston and Larson, 2011), and between ed in lower retention rates in 2 to 3-year-old cows. low-input developed heifers grazing either winter Restricted heifers that failed to rebreed in the Montana range or corn residue (Larson et al., 2011). Within

59 study, all heifers were exposed to a low-birthweight ed in increased economic advantages compared with EPD bull battery in the same breeding pastures. developing heifers at greater rates of ADG to achieve Calving date for first calf heifers may impact a greater target BW. cow longevity and productivity. Calving late in yr 1 increases the proportion of cows that either calve Summary and Conclusions later next year or do not conceive (Burris and Priode, Developing heifers to lighter target BW may 1958). Research has indicated heifers having their be advantageous in maintaining positive energy bal- first calf earlier in the calving season remained in the ance or adapting to negative energy balance through herd longer compared with heifers that calved later the breeding season in many range settings. Likewise, in the calving season (Rogers et al., 2004; Cushman heifers developed under a range setting may be better et al., 2013). Therefore, heifers calving earlier in the adapted to maintain desired metabolic status during calving season have greater potential for longevity and breeding than heifers reared in a pen or developed at a lifetime productivity. However, the above-mentioned high rate of gain. Implications of heifer development studies do not demonstrate that heifer development system on cow longevity must be considered when affected date of calving or longevity. evaluating economics of a heifer enterprise; however, studies evaluating the effects of heifer development Economic Analysis of Heifer Development Systems systems on cow longevity are extremely limited. Mulliniks et al. (2012) evaluated enterprise budgets for the 3 New Mexico heifer development Literature Cited treatments. Assumptions included comparing 100 American Angus Association. 2012. Angus genetic heifers in each treatment, and all heifers would be trend by birth year. Available at :. (Accessed April sold in the fall of their yearling year, regardless of 25, 2014). pregnancy status. Gross returns were greatest for the American Hereford Association. 2014. Hereford RUP-supplemented range heifers and least for heifers genetic trend by birth year. Available at: http://her- developed in the feedlot; feed costs were greatest for eford.org/userfiles/F12_Trend.pdf. (Accessed April feedlot-developed heifers. Compared with feedlot-de- 25, 2014). veloped heifers, net returns were $99.71 and $87.18 greater per heifer developed for the high-RUP and cot- American International Charolais Association. 2014. tonseed meal-supplemented heifers grazing dormant Charolais genetic trend by birth year. Available native range, respectively. The increase in net returns at: http://www.charolaisusa.com/pdf/2012/09.07/ for range-raised heifers was due to greater pregnancy GeneticTrendGraphically.pdf. (Accessed April 25, rates and decreased development costs. 2014). A similar approach was used to evaluate the Arias, R. P., P. J. Gunn, R. P. Lemanager, and S. L. heifer development protocols in the Montana data Lake. 2012. Effects of post-AI nutrition on growth set. Gross returns were greater for control heifers, but performance and fertility of yearling beef heifers. restricted heifers had lower feed costs. This resulted Proc. West. Sec. Amer. Soc. Anim. Sci. 63:117-121. in an increase of $37.24 in net returns per developed Burris, M. J., and B. M. Priode. 1958. Effect of calv- heifer for the restricted group. ing date on subsequent calving performance. J. Research from the University of Nebraska Anim. Sci. 17:527-533. reports similar savings in development costs, where developing heifers on dormant winter forage resulted Byerly, D. J., R. B. Staigmiller, J. G. Berardinelli, and in a $45 savings per pregnant heifer compared with R. E. Short. 1987. Pregnancy rates of beef heifers drylot development (Funston and Larson, 2011), and bred either on pubertal or third estrus. J. Anim. Sci. a similar development cost comparing 2 extensive 65:645-650. development systems, winter range vs. corn residue Christie, M. R., M. L. Marine, R. A. French, and M. S. (Larson et al., 2011). Studies from New Mexico, Blouin. 2012. Genetic adaptation to captivity can Montana, and Nebraska illustrate that restricting gain occur in a single generation. Proc. Natl. Acad. Sci. during post-weaning development by limiting DMI or 109:238-242. developing heifers on dormant winter forage result-

60 Ciccioli, N. H., S. L. Charles-Edwards, C. Floyd, R. P. Larson, D. M., A. S. Cupp, and R. N. Funston. 2011. Wettemann, H. T. Purvis, K. S. Lusby, G. W. Horn, Heifer development systems: A comparison of and D. L. Lalman. 2005. Incidence of puberty in grazing winter range or corn residue. J. Anim. Sci. beef heifers fed high- or low-starch diets for differ- 89:2365:2372. ent periods before breeding. J. Anim. Sci. 83:2653- Lynch, J. M., G. C. Lamb, B. L. Miller, R. T. Brandt, 2662. Jr., R. C. Cochran, and J. E. Minton. 1997. Influ- Cushman, R. A., L. K. Kill, R. N. Funston, E. M. ence of timing of gain on growth and reproductive Mousel, and G.A. Perry. 2013. Heifer calving date performance of beef replacement heifers. J. Anim. positively influences calf weaning weights through Sci. 87:3043-3052. six parturitions. J. Anim. Sci. 91:4486-4491. Marston, T. T., K. S. Lusby, and R. P. Wettemann. Endecott, R. L., R. N. Funston, J. T. Mulliniks and A. 1995. Effects of postweaning diet on age and J. Roberts. 2013. Implications of beef heifer devel- weight at puberty and milk production of heifers. opment systems and lifetime productivity. J. Anim. J. Anim. Sci. 73:63-68. Sci. 91:1329-1335. Mulliniks, J. T., D. E. Hawkins, K. K. Kane, S. H. Ferrell, C. L. 1982. Effects of postweaning rate of gain Cox, L. A. Torell, E. J. Scholljegerdes, and M. on onset of puberty and productive performance of K. Petersen. 2012. Metabolizable protein supply heifers of different breeds. J. Anim. Sci. 55:1272- while grazing dormant winter forage during heifer 1283. development alters pregnancy and subsequent in- Foxcroft, G. R. 1997. Mechanisms mediating nutri- herd retention rate. J. Anim. Sci. 91:1409-1416. tional effects on embryonic survival in pigs. J. Olson, K. C., J. R. Jaeger, and J. R. Brethour. 1992. Reprod. Fert. Suppl. 52:47-61. Growth and reproductive performance of heifers Freetly, H. C. and L. V. Cundiff. 1997. Postweaning overwintered in range or drylot environments. J. growth and reproduction characteristics of heifers Prod. Agri. 5:72-76. sired by bulls of seven breeds and raised on differ- Perry, G., J. Walker, C. Wright, and K. Olson. 2009. ent levels of nutrition. J. Anim. Sci. 75:2841-2851. Impact of method of heifer development and post- Funston, R. N. and G. H. Deutscher. 2004. Compari- AI management on reproductive efficiency. Proc. son of target breeding weight and breeding date for Range Beef Cow Symp. XXI, pp 35-42. replacement beef heifers and effects on subsequent Roberts, A. J., T. W. Geary, E. E. Grings, R. C. Water- reproduction and calf performance. J. Anim. Sci. man, and M. D. MacNeil. 2009. Reproductive per- 82:3094-3099. formance of heifers offered ad libitum or restricted Funston, R. N. and D. M. Larson. 2011. Heifer de- access to feed for a one hundred forty-day period velopment systems: Dry-lot feeding compared after weaning. J. Anim. Sci. 87:3043-3052. with grazing dormant winter forage. J. Anim. Sci. Roberts, A. J., J. Ketchum, R. N. Funston, and T. W. 89:1595-1602. Geary. 2013. Impact of number of estrous cycles Funston, R. N., J. L. Martin, D. M. Larson, and A. J. exhibited prior to start of breeding on reproductive Roberts. 2012. Nutritional aspects of developing performance in beef heifers. Proc. West. Sec. Amer. replacement heifers. J. Anim. Sci. 90:1166-1171. Soc. Anim. Sci. 64:254-257. Greer, R. C., R. W. Whitman, R. B. Staigmiller, and D. Rogers, P. L., C. T. Gaskins, K. A. Johnson, and M. D. C. Anderson. 1983. Estimating the impact of man- MacNeil. 2004. Evaluating longevity of composite agement decisions on the occurrence of puberty in beef females using survival analysis techniques. J. beef heifers. J. Anim. Sci. 56:30-39. Anim. Sci. 82:860-866. Hall, J. B., R. B. Staigmiller, R. B. Bellows, R. E. Short, R. E. and R. A. Bellows. 1971. Relationships Short, W. M. Moseley, and S. E. Bellows. 1995. among weight gains, age at puberty and reproduc- Body composition and metabolic profiles asso- tive performance in heifers. J. Anim. Sci. 32:127- ciated with puberty in beef heifers. J. Anim. Sci. 131. 73:3409-3420. 61 Speakman, J. R. and C. Hambly. 2007. Starving for THE LONG-LASTING IMPACT OF life: What animal studies can and cannot tell us NUTRITION: about the use of caloric restriction to prolong hu- DEVELOPMENTAL PROGRAMMING man lifespan. J. Nutr. 137:1078-1086. Kimberly A. Vonnahme1 Summers, A.F., S. P. Weber, H. A. Lardner, and R. N. 1Department of Animal Sciences, North Dakota State Funston. 2014. Effect of beef heifer development University system on average daily gain, reproduction, and Introduction adaptation to corn residue during first pregnancy. J. Anim. Sci. jas.2013-7225; published ahead of Livestock producers are interested in utilizing print March 25, 2014, doi:10.2527/jas.2013-7225. nutrients in the most efficient way to optimize growth. Often, one tends to focus on the growth that an animal R. A. Vraspir, A. F. Summers, A. J. Roberts, and R. N. achieves after birth, however, the majority of mamma- Funston. 2013. Effect of pubertal status and num- lian livestock (i.e. swine, sheep, and cattle) spend 35- ber of estrous cycles prior to the breeding season 40% of their life (i.e. from conception to consumption) on pregnancy rate in beef heifers. Proc. West. Sec. within the uterus, being nourished solely by the pla- Amer. Soc. Anim. Sci. 64:116-120. centa. The maternal system can be influenced by many White, F. J., L. N. Floyd, C. A. Lents, N. H. Ciccioli, different extrinsic factors, including nutritional status, L. J. Spicer, and R. P. Wettemann. 2001. Acutely which ultimately can program nutrient partitioning and ultimately growth, development and function of the restricting nutrition causes anovulation and alters major fetal organ systems (Wallace, 1948; Wallace et endocrine function in beef heifers. Oklahoma State al., 1999; Godfrey and Barker, 2000; Wu et al., 2006). University Anim. Sci. Res. Report. Oklahoma Ag. The trajectory of prenatal growth is sensitive to direct Expt. Sta. Pub. P986. and indirect effects of maternal environment, partic- Wiltbank, J. N., K. E. Gregory, L. A. Swiger, J. E. ularly during early stages of embryonic life (Robin- Ingalls, J. A. Rothlisberger, and R. M. Koch. 1966. son et al., 1995), the time when placental growth is Effects of heterosis on age and weight at puberty exponential. Moreover, pre-term delivery and fetal in beef heifers. J. Anim. Sci. 25:744-751. growth restriction are associated with greater risk of neonatal mortality and morbidity in livestock and Wiltbank, J. N., C. W. Kasson, and J. E. Ingalls. 1969. humans. Offspring born at an above average weight Puberty in crossbred and straightbred beef heifers. have an increased chance of survival compared with J. Anim. Sci. 29:602-605. those born at a below average weight in all domestic livestock species, including the cow, ewe, and sow. Just as growth-restricted human infants are at risk of immediate postnatal complications and diseases later in life (Godfrey and Barker, 2000), there is increasing evidence that production characteristics in our domestic livestock may also be impacted by maternal diet (Wu et al., 2006). Some of the complications reported in livestock include increased neonatal morbidities and mortalities (Hammer et al., 2011), intestinal and respiratory dysfunctions, slow postnatal growth, increased fat deposition, differing muscle fiber diameters and reduced meat quality (reviewed in Wu et al., 2006). The objective of this proceedings paper is to highlight some of our laboratory’s investigations on how maternal environment can impact fetal and placental development, impacts on uterine and/or umbilical blood flow in cattle and sheep, and potential timing of intervention, or potential therapeutics, which may increase uteroplacental blood flow.

62 Placental Development and Uteroplacental Blood would decrease the incidence of morbidity and mor- Flow tality as well as suboptimal offspring growth perfor- The placenta plays a major role in the regula- mance in livestock species. tion of fetal growth. In ruminants, the fetal placenta Therapeutic agents targeting placental blood attaches to discrete sites on the uterine wall called flow increased fetal growth in compromised pregnan- caruncles. These caruncles are aglandular sites which cies (Reynolds et al., 2006). There is an ever-increas- appear as knobs along the uterine luminal surface of ing wealth of data that are demonstrating how reali- non-pregnant animals, and are arranged in two dor- mentation, or other therapeutic agents, may be used to sal and two ventral rows throughout the length of the rescue at-risk pregnancies. In our laboratory, we have uterine horns (Ford, 1999). The placental membranes investigated the role that realimentation, protein sup- attach at these sites via chorionic villi in areas termed plementation, and melatonin supplementation has on cotyledons. The caruncular-cotyledonary unit is called uteroplacental blood flow and/or vascular reactivity of a placentome and is the primary functional area of the placental arteries. In order to perform the former, physiological exchanges between mother and fetus. we have employed the use of Doppler ultrasonogra- Placental nutrient transport efficiency is direct- phy. Other methods of determining blood flow are ly related to uteroplacental blood flow (Reynolds and effective, but require surgery and increased numbers Redmer, 1995). All of the respiratory gases, nutrients, of animals to determine blood flow at different time and wastes that are exchanged between the maternal points during pregnancy because of the growth of the and fetal systems are transported via the uterus-pla- uterine vasculature as gestation advances. Uterine and centa (Reynolds and Redmer, 1995, 2001). Thus, it is umbilical artery cardiac cycle waveforms were plot- not surprising that fetal growth restriction in a num- ted in Doppler mode by velocity (cm/s; y-axis) and ber of experimental paradigms is highly correlated time (s; x-axis). Fetal or maternal heart rate (beats/ with reduced uteroplacental growth and development min), pulsatility index (PI), resistance index (RI), and (Reynolds and Redmer, 1995, 2001). Establishment blood flow (BF) were calculated using preset functions of functional fetal and uteroplacental circulations is on the ultrasound instrument. Abbreviations for the one of the earliest events during embryonic/placental various instrument-generated functions are as follows: development (Patten, 1964; Ramsey, 1982). It has peak systolic velocity (PSV), end diastolic velocity been shown that the large increase in transplacental (EDV), mean velocity (MnV), and cross sectional area exchange, which supports the exponential increase in of the vessel (CSA). Equations are as followed: PI = fetal growth during the last half of gestation, depends [PSV (cm/s) – EDV (cm/s)] / MnV (cm/s); RI = [PSV primarily on the dramatic growth of the uteroplacental (cm/s) – EDV (cm/s)] / PSV (cm/s); blood flow (BF, vascular beds during the first half of pregnancy (Mes- mL/min) = MnV (cm/s) × cross sectional area of the chia, 1983; Reynolds and Redmer, 1995). Therefore, vessel (cm2) × 60 s. By continuously monitoring the an understanding of factors that impact uteroplacental same animal, which has not undergone surgical ma- blood flow will directly impact placental function and nipulation, we feel that we can effectively determine thus fetal growth. how different interventions may regulate uteroplacental blood flow. Our current animal models are outlined Adequate uteroplacental blood flow is critical below. for normal fetal growth, and therefore, not surprising- ly, experimental conditions designed to investigate Nutrient Restriction fetal growth retardation and placental insufficiency, In normal pregnancies, resistance of the utero- be it over-nutrition, nutrient restriction, hyperthermia, placental arteries have been documented to decrease or high altitude, commonly share reduced uterine and as gestation advances. Our laboratory has reported that umbilical blood flows (for review see Reynolds et when pregnant ewe lambs are nutrient-restricted, lamb al., 2006). Therefore, modifying uterine blood flow birth weight is reduced compared to control fed ewes and nutrient transfer capacity in the placenta allows (Swanson et al., 2008; Meyer et al., 2010). While pla- for increased delivery of oxygen and nutrients to the cental weights are not different, we have demonstrated exponentially growing fetus. Fowden et al. (2006) that when ewes are restricted, there is ~33% decrease reviewed key factors affecting placental nutrient in endothelial nitric oxide synthase mRNA expression transfer capacity, which were size, nutrient transporter on d 130 of gestation in the maternal portion of the abundance, nutrient synthesis and metabolism, and placenta compared to control-fed animals (Lekatz et hormone synthesis and metabolism. Discovery of nov- al., 2010a). We hypothesized that this reduction in el therapeutic agents that improve placental function birth weight was due to a greater placental vascular 63 resistance, and decreased uteroplacental blood flow flow, which was decreased in restricted ewes from day in restricted ewes compared to control ewes. In order 80 through 110 of gestation compared to adequately to evaluate the effects of maternal nutrient restriction fed ewes. Moreover, at day 110 of gestation, restricted on the umbilical hemodynamics, we have a model of ewes had a 23% decrease in umbilical artery blood global restriction that begins on day 50 of gestation flow compared to adequately fed ewes (Lemley et al., until term (~145 days). Restricted ewes had increased 2012). While we are continuing our investigations (P = 0.01) PI and RI compared to control ewes (Lekatz into the impacts of melatonin supplementation in at- et al., 2010a; Lemley et al., 2012). Moreover, we have risk pregnancies, we feel that melatonin treatment may demonstrated that umbilical blood flow is reduced be useful in negating the consequences of intrauterine when a nutrient restriction is applied (Lemley et al., growth restriction that occur due to specific abnormal- 2012). ities in umbilical blood flow. Therapeutic supplements thought to target In cattle, we have recently demonstrated that placental blood flow and nutrient delivery to the fetus nutrient restriction from early to mid-pregnancy (i.e. have been shown to increase fetal growth in animal day 30-140) does not alter uterine blood flow (Cama- models of intrauterine growth restriction (Vosatka et cho et al., 2014). However, upon realimentation, the al., 1998; Richter et al., 2009; Satterfield et al., 2010); uterine artery blood flow increases in those cows however, few studies have addressed uteroplacental that were previously restricted, but only to the horn hemodynamics in models of improved fetal growth. in which the calf is housed (Camacho et al., 2014). For instance, melatonin supplementation was shown to Interestingly, it appears that realimentation alters the negate the decreased birth weight in nutrient- restrict- growth trajectory of the bovine placenta (Vonnahme ed rats (Richter et al., 2009), which was attributed to et la., 2007), something that has not been investigated increased placental antioxidant enzyme expression in in the ewe. Recent data in our laboratory demonstrates nutrient-restricted rats supplemented with melatonin. that the placental artery reactivity to vasoactive agents Our hypothesis was that dietary melatonin treatment in vitro are more responsive to vasodilators (Reyaz during a compromised pregnancy would improve and Vonnahme, unpublished data), and there is an fetal growth and placental nutrient transfer capacity increase in capillary numbers (Mordhorst and Von- by increasing uterine and umbilical blood flow. The nahme, unpublished data), perhaps to allow for more uteroplacental hemodynamics and fetal growth were nutrient uptake. The ability of the uterus-placenta to determined in ewes that received a dietary supplemen- compensate upon realimentation is quite intriguing tation with or without melatonin (5 mg) in adequately and we are continuing our studies to determine which fed (100% of NRC recommendations) or nutrient-re- portions of the placenta (i.e. maternal or fetal) may stricted (60% of control) ewes. Dietary treatments contribute to compensatory prenatal growth of the were initiated on d 50 of gestation and umbilical blood fetus. flow, as well as fetal growth (measured by abdominal Protein Supplementation and biparietal distances) were determined every 10 d from d 50 to d 110 of gestation. By d 110 of gestation, While the literature is now booming with fetuses from restricted ewes had a 9% reduction (P increasing evidence of how nutrient restriction impairs = 0.01) in abdominal diameter compared to fetuses several physiological parameters, few concentrate on from adequately nourished ewes, whereas fetuses from enhancing postnatal growth in livestock species. In melatonin supplemented ewes tended to have (P = a recent series of papers in cattle, cows gestated on 0.08) a 9% increase in biparietal diameter (Lemley et range (where crude protein of forage is < 6%) that al., 2012). were protein supplemented during late gestation had calves similar in birth weight, but had calves with We did observe a significant melatonin treat- increased weaning weight compared to protein un- ment by day interaction (P < 0.001) for umbilical supplemented cows (Stalker et al., 2006, Martin et al., artery blood flow which was increased in melatonin 2007; Larson et al., 2009). It is valuable to note that supplemented ewes from day 60 through 110 of ges- the protein supplementation enhanced growth after tation compared to control (no melatonin supplemen- birth. Furthermore, the pregnancy rates in heifer calves tation). Moreover, at day 110 of gestation melatonin born from protein supplemented cows were enhanced supplemented ewes had a 20% increase in umbilical compared to control cows (93 vs 80%; Martin et al., artery blood flow compared to control ewes. In addi- 2007). It was our hypothesis that the increased fertility tion, a significant nutritional plane by day interaction and growth rate of the calves from supplemented dams (P < 0.0001) was observed for umbilical artery blood may be due to enhanced uterine blood flow and/or 64 placental nutrient transfer. Ongoing studies in our lab- ultimately assist researchers in understanding how the oratory are investigating how protein supplementation maternal environmental impacts placental, and thus during late gestation can impact uterine blood flow. fetal, development. For the past 2 years we have investigated how protein Literature Cited supplementation (in the form of DDGS) can impact uterine blood flow. When we use DDGS with a low quality forage source, uterine blood flow is reduced Camacho, L. E., Lekatz, L. A., VanEnom, M. L., compared to control cows (Mordhorst and Vonnahme, Schauer, C. S., Maddock Carlin, K. R., and Von- unpublished observations). Just recently, we demon- nahme, K. A. 2010. Effects of maternal metabo- strated that when DDGS is given with a corn-stalk lizable protein supplementation in late gestation forage base, we increase uterine blood flow (Kennedy on uterine and umbilical blood flows in sheep.J. and Vonnahme, unpublished observations). We are Anim. Sci. 88: E-Suppl. 2: 106. investigating how specific nutrients differed between these 2 studies in order to tease apart the mechanism Camacho, L.E., Lemley, C.O., Prezotto, L.D., Bauer, that may be impacting how protein influences uterine M. L., Freetly, H. C., Swanson, K.C. and Von- blood flow in the beef cow. nahme, K.A. 2014. Effects of maternal nutrient In order to more fully understand the impacts restriction followed by realimentation during of maternal protein on uteroplacental blood flow and midgestation on uterine blood flow in beef cows. placental vascular development, we also have used Theriogenology. 81:1248-1256. an ovine model where the diets are isocaloric, with differing levels of protein in the diet. Singleton fetuses Ford, S.P. 1999. Cotyledonary placenta. Encyclopedia from ewes consuming the high protein diet are heavier of Reproduction. 1:730-738. on d 130 of gestation compared to fetuses from ewes consuming the low protein diet, with no differences Fowden, A. L., Giussani D. A., and Forhead, A. J. in placental weight apparent (Camacho et al., 2010). When uterine blood flow was obtained from a single 2006. Intrauterine programming of physiology time point (d 130 of gestation), ewes consuming the systems: causes and consequences. Physiology high protein diet had a decrease in uterine blood flow 21:29-37. compared to the low group, with the control being intermediate (Camacho et al., 2010). This is similar to Godfrey, K.M. and Barker, D. J. 2000. Fetal nutrition our first year protein supplementation work with beef and adult disease. Am. J. Clin. Nutr. cattle. Moreover, when investigating the ability of the 71:1344S-1352S. fetal placental arteries to vasodilate to increasing concentrations of bradykinin, placental arteries from high Hammer, C. J., Thorson, J. F., Meyer, A. M., Redmer, protein ewes had a decreased responsiveness com- D. A., Luther, J. S., Neville, T. L., Reed, pared to control and low protein ewes (Lekatz et al., J. J., Reynolds, L. P., Caton, J. S., and Vonnahme, 2010b). Understanding if additional calories (i.e. cow K. A. 2011. Effects of maternal selenium supply study), or a greater proportion of total calories coming and plane of nutrition during gestation on passive from protein (i.e. sheep study), needs to be elucidated, transfer of immunity and health in neonatal lambs. and further work is underway in our laboratory. J. Anim. Sci. 89:3690-3698. Summary and Conclusions We hope to improve approaches to manage- Larson, D.M. Martin, J.L., Adams, D.C., and Funston, ment of livestock during pregnancy which may im- R.N. 2009. Winter grazing system and pact not only that dam’s reproductive success, but her supplementation during late gestation influence offspring’s growth potential and performance later in performance of beef cows and steer progeny. J. life. Future applications of this research may be used Anim. Sci. 87:1147-1155. to develop therapeutics for at-risk pregnancies in our domestic livestock. If these therapeutics can be used on-farm, producers would have the ability to increase animal health while also reducing costs of animal production. While each species is unique in its placental development and vascularity, comparative studies may 65 Lekatz, L.A., Caton, J.S., Taylor, J.B, Reynolds, L.P., Reynolds, L.P., and Redmer, D.A. 1995. Utero-placen- Redmer, D.A., and Vonnahme, K.A. 2010a. Ma- tal vascular development and placental ternal selenium supplementation and timing of function. J. Anim. Sci. 73:1839-1851. nutrient restriction in pregnant sheep: Impacts on maternal endocrine status and placental character- Reynolds, L.P., and Redmer, D.A. 2001. Angiogenesis istics. J. Anim. Sci. 88:955-971. in the placenta. Biol. Reprod. 64:1033- 1040. Lekatz, L. A., Van Emon, M. L., Shukla, P. K., O’Ro- urke, S. T. , Schauer, C. S., Carlin K. M., Reynolds, L.P., Caton, J.S., Redmer, D.A., Grazul-Bil- and Vonnahme, K. A. 2010b. Influence of me- ska, A.T., Vonnahme, K.A., Borowicz, P.P., Luther, tabolizable protein supplementation during late J.S., Wallace, J.M., Wu, G., and Spencer, T.E. gestation on vasoreactivity of maternal and fetal 2006. Evidence for altered placental blood flow placental arteries in sheep. J. Anim. Sci. 88:E-Sup- and vascularity in compromised pregnancies. J. pl. 2:869-870. Physiol. 572:51-58.

Lemley, C.O., Meyer, A. M., Camacho, L. E. , Nev- Richter H.G., Hansell J.A., Raut S., and Giussani, ille, T. L., Newman, D. J., Caton, J. S. ,and Von- D.A. 2009. Melatonin improves placental nahme, K. A. 2012. Melatonin supplementation efficiency and birth weight and increases the alters uteroplacental hemodynamics and fetal placental expression of antioxidant enzymes in development in an ovine model of intrauterine undernourished pregnancy. J. Pineal Res. 46:357- growth restriction (IUGR). Amer. J. Physiol. 364. 302:R454-467. Robinson, J., Chidzanja, S., Kind, K., Lok, F., Owens, Martin, J.L., Vonnahme, K.A. , Adams, D.C. , Lardy, P., and Owen J. 1995. Placental control of G.P. , and Funston, R.N. 2007. Effects of fetal growth. Reprod. Fertil. Dev. 7:333–344. dam nutrition on growth and reproductive performance of heifer calves. J. Anim. Sci. 85:841-847. Satterfield ,M.C., Bazer, F.W., Spencer, T.E., and Wu, G. 2010. Sildenafil citrate treatment Meschia, G. 1983. Circulation to female reproductive enhances amino acid availability in the conceptus organs. In: Handbook of Physiology and fetal growth in an ovine model of intrauterine 3:241-267. growth restriction. J. Nutr. 140,251-258.

Meyer, A.M., Reed, J.J., Neville, T.L. , Taylor, J.B., Stalker, L.A., Adams, D.C. , Klopfenstein, T.J., Feuz, Hammer, C.J., Reynolds, L.P., Redmer, D.M., and Funston, R.N. 2006. Effects of pre- and postpartum nutrition on reproduction in D.A., Vonnahme, K.A., and Caton, J.S. 2010. Effects spring calving cows and calf feedlot performance. of nutritional plane and selenium supply during J. Anim. Sci. 84:2582-2589. gestation on ewe and neonatal offspring performance, body composition, and serum selenium. J. Swanson, T.J., Hammer, C.J., Luther, J.S., Carlson, Anim. Sci. 88:1786-1800. D.B., Taylor, J.B., Redmer, D.A., Neville, T.L., Reed, J.J., Reynolds, L.P., Caton, J.S., and Von- Patten, B.M. 1964. Foundations of Embryology (2nd nahme, K.A. 2008. Effects of plane of nutrition Ed) McGraw-Hill, New York. and selenium supplementation on colostrum quality and mammary development in pregnant ewe Ramsey, E.M. 1982. The Placenta, Human and Ani- lambs. J. Anim. Sci. 86:2415-2423. mal. Praeger, New York.

66 Vonnahme, K.A., Zhu, M. J., Borowicz, P. P., Geary, T.W., Hess, B. W., Reynolds, L. P., J. S. Caton, Means, W. J. and Ford, S. P. 2007. Effect of early gestational undernutrition on angiogenic factor expression and vascularity in the bovine placentome. J. Anim. Sci. 85:2464-2472.

Vosatka R.J., Hassoun, P.M., and Harvey-Wilkes, K.B. 1998. Dietary L-arginine prevents fetal growth restriction in rats. Am. J. Obstet. Gynecol. 178:242-246.

Wallace, L.R. 1948. The growth of lambs before and after birth in relation to the level of nutrition. J. Agric. Sci.,Cambridge. 38:243-300 and 38:367-398.

Wallace, J.M, Bourke, D.A., and Aitken, R.P. 1999. Nutrition and fetal growth: paradoxical effects in the over-nourished adolescent sheep. J. Reprod. Fertil. Suppl. 54:385-399.

Wu, G., Bazer, F.W., Wallace, J.M., and Spencer, T.E. 2006. Board invited review. Intrauterine growth retardation: implications for the animal sciences. J. Anim. Sci. 84:2316-2337.

67 PROCEEDINGS GENERAL SESSION 2: FOCUS ON THE FEEDLOT

Selection for Novel Traits: Genomic selection (GS) has been successfully imple- An international GENOMICS mented into national dairy cattle genetic evaluations in perspective many countries since 2009 (Spelman et al., 2013). Ret- Donagh P. Berry1 rospective analysis (McParland et al., 2014) signifies 1Animal & Grassland Research and Innovation Centre that GS is up to 29% more accurate at predicting an Teagasc, Moorepark, Ireland animal’s true genetic merit (based on progeny performance) compared to just parental average. However, the breeding structures of dairy and beef are quite dif- Introduction Genomic selection is being heralded as the ferent and this has implications for the successful im- “..most promising application of molecular genetics in plementation of genomic selection in beef but also the livestock production since work began almost 20 years justification for international cooperation, especially ago” (Sellner et al., 2007). The objective of genomic for novel traits. selection is to increase the accuracy of identifying ge- The objective of this article is to discuss the netically elite (and inferior) animals at a younger age potential for international collaboration in genomics in but also at a lower cost per animal. Genetic gain may beef cattle; although examples will be given for novel be defined as (Rendel and Robertson, 1950): traits the relevance of the discussion is applicable to all traits although the marginal benefit is greatest for novel ∆G = i ∙ r ∙ σ traits where the population of phenotyped and geno- L where ∆G is annual genetic gain; i is the intensity of typed animals may be smaller (discussed later). selection; r is the accuracy with which you know the Differences between Dairy and Beef Breeding Struc- genetic merit of each animal, σ is the genetic standard deviation (i.e., the square root of the genetic variance or tures and Implications for Genomic Selection simply just a measure of the genetic differences among Many differences exist between dairy and beef breed- animals), and L is the generation interval. Genomic se- ing structures so therefore the approaches applied to- lection attempts to alter i, r and L. It may also influ- date in dairy cattle may not be directly applied in beef, ence the detected genetic variation. Genomic selection, although there are obvious similarities. however, does not necessarily improve all three com- Breed. One breed (i.e., Holstein-Friesian) predom- ponents simultaneously as it may reduce the accuracy inates the dairy cattle populations in most developed of selection (i.e., r) compared to traditional methods countries making it relatively easy and inexpensive to but reduce the generation interval (i.e., L) proportion- develop large informative reference populations for the ally more thereby increasing annual genetic gain. Be- generation of accurate genomic predictions. It is now cause the cost of “testing” a young bull with genomic well known that the stronger the genomic relationship selection is approximately 0.3% (i.e., 0.003) the cost of between the reference population of genotyped and phe- progeny testing the selection intensity can be increased notyped animals with the candidate animals, the greater considerably thus advancing genetic gain. will be, on average, the accuracy of genomic predic- Genomic selection (and genomics in general) is partic- tions (Habier et al., 2007; Pszczola et al., 2012). Ac- ularly advantageous for traits that are: curate across-breed genomic predictions have to-date been elusive (Karoui et al, 2012; Berry 2012) in cattle. • Sex linked (e.g. milk yield and female fertility) Figure 1 shows a genome wide association study for • Take a long time to measure (e.g., cow longevity) direct calving difficulty in Irish Holstein and Charolais animals. The scoring system for calving difficulty is the • Exhibit low heritability (e.g., female fertility) same across both breeds and the genetic evaluations are across breeds. A genomic region with a large association • Difficult and/or expensive to measure (i.e., (2.49% of genetic variation) with calving difficulty was novel traits like feed intake complex, meat detected on chromosome 18 in the Holstein-Friesian quality) population and, although these SNPs were also segre-

68 PROCEEDINGS GENERAL SESSION 2: FOCUS ON THE FEEDLOT

gating in the Charolais population, no association was Effective Population Size. The effective population detected in this region of the genome. Similarly a ge- size globally of Holstein-Friesians is likely to be some- nomic region associated with calving difficulty in Cha- where between 40 and 100 (McParland et al., 2007; rolais (3.13% of the genetic variation) was detected on Saatchi et al., 2011). The global effective population Chromsome 2 but not in Holstein-Friesians despite the size of beef breeds is likely to be larger (McParland SNPs segregating in both populations. Moreover, the et al., 2007; Saatchi et al., 2011) given the vast differ- sign of the allelic effects for 50% of SNP differed when ences in breeding policies implemented in the different estimated in either the Holstein-Friesian population or populations. The accuracy of genomic predictions is a the Charolais population. This is likely due to differ- function of the size of the reference population, the her- ences in linkage phase between breeds and background itability of the trait under investigation, and the effec- polygenic effects and is undoubtedly a contributor to tive population size of the population (Daetwyler et al., the sometimes observed negative correlations between 2008). Larger effective population sizes require larger genomically predicted EPDs and progeny-based EPDs reference populations to achieve the equivalent accu- when the population being tested is not adequately rep- racy of genomic predictions compared to populations resented in the genomic reference population. This dif- with smaller effective population sizes. ference between dairy and beef and the current inability for genomic algorithms and genomic information to be The number of independent genomic segments is likely useful for acrossbreed genomic evaluation implies that to vary with effective population size. The number of

each breed has to generate (and therefore incur the cost) independent loci (Me) in a 30 Morgan genome can be of generating its own reference population. The same derived deterministically for a range of different effec- is true for novel traits implying a large cost for each tive population sizes as (Goddard, 2009): country to implement and genomic selection program. 2Ne L Me= Log10 (4Ne L)

where Ne is the effective population size and L is the length of the genome in Morgans. The number of animals (N) required to achieve a given accuracy (i.e., square root of the reliability) can then be derived as (Calus et al., 2012): r2 M N = e h2 (q2 - r2)

Where q2 is the proportion of genetic variance captured by the SNPs (here assumed to be 0.8) and h2 is the heritability of the traits (here assumed to be 0.20). Figure 2 illustrates the number of animals that need to be phenotyped and genotyped to achieve a given accuracy for different effective population sizes. The larger the effective population size the larger the dataset of phenotyped and genotyped animals that is required to s achieve an equivalent accuracy of genomic predictions Figure 1. Manhattan plots of the single nucleotide compared with populations with smaller effective pop- polymorphisms associated with direct calving diffi- ulation sizes. culty in 770 Holstein-Friesian (Top figure) and 927 Charolais (bottom figure) (Purfield et al., 2014).

69 Therefore, a large population of phenotyped and genotyped animals will be required to achieve an acceptable accuracy of genomic predictions.

Less Phenotypes and Parentage Recording in Beef. Accurate recording of detailed phenotypes on large populations of commercial animals is generally the norm in most dairy cow populations. Furthermore, parentage of most dairy females is known facilitating accurate EPDs of their pedigree. Although phenotypic recording exists in many beef populations it is, how- Figure 2. Number of animals that need to be both ever, lacking (for some traits at least) in some popula- genotyped and phenotyped to achieve different lev- tions. As alluded to previously, genotypes from animals els of accuracy (i.e., square root of the reliability) with high accuracy EPDs can be more informative than when the effective population size is 50 (solid line), genotypes of animals with lower accuracy EPDs. One- 100 (long-dashed line), 200 (shorter dashed line) step genomic procedures will not alleviate this issue and 300 (smallest dashed line) as animals with non-recorded pedigree will still have to be genotyped to allocate the animal to its pedigree. Greater Usage of AI in Dairy. In general, there is a Ultra-low cost genomic tools for parentage assignment greater usage of AI in dairy cattle than in beef. The may aid in allocating animals to parents and thus in- accuracy of an animal’s EPD from traditional genetic crease the accuracy of traditional genetic evaluations evaluations increases with increasing quantity of prog- for some animals. Lack of pedigree information and eny records; therefore the accuracy of progeny tested phenotypes is generally not of concern for animals with bull can be very high. Because the heritability statistic novel phenotypes since if the resources are being ex- measures the strength of the resemblance between the pended in generating the phenotypes then the pedigree phenotypic value of an animal and its true genetic mer- is usually also recorded. it, the effective heritability of high accuracy EPDs is close to unity. Lack of Participation in International Genetic Eval- Figure 3 illustrates the number of genotyped and phe- uations. Many genomic evaluations in dairy cattle, notyped animals required to achieve different accuracy including Ireland, operate a two-step procedure where levels of genomic predictions. Clearly to achieve the same accuracy of genomic predictions, less genotyped and phenotyped animals are required for higher heritability traits (or animals with higher accuracy EPDs). Dairy cattle genomic breeding programs firstly focused on the genotyping of thousands of AI progeny tested bulls because of their greater accuracy and thus greater effective heritability. Using this approach for a trait with a heritability of 0.20, 7903 genotyped animals with own performance records would be equivalent to 1756 bulls (i.e., less than one quarter) with an EPD accuracy (i.e., square root of reliability) of 0.95. Hence, all else being equal, the implementation of genomic se- Figure 3. Number of phenotyped and genotyped lection in beef where less high reliability sires exist will animals that are required to achieve an accuracy be considerably more expensive than in dairy. Collabo- of genomic prediction of 0.4 to 0.8 (length of dashes ration can help reduce this cost. Because novel traits do decrease and the accuracy increases); calculations not generally tend to be measured on large populations are based on the assumption of 1000 independent of animals, the generation of high accuracy EPDs for genomic regions and the genomic markers explain- a large population of sires is generally not achievable. ing 80% of the total genetic variance. 70 derivatives of EPDs are used as input phenotypes for ration in the selection of novel traits is the sharing of the development of the genomic prediction equations. information on what animals have been genotyped and Many countries, especially those with a small breed- on what genotyping panel. An example of the current ing program, exploit MACE evaluations generated by international list compiled (currently only participated INTERBULL. Therefore, phenotypic information is in by Ireland, the UK, France and Australia) is in Table available on bulls even if they do not have any daughter 1; Australian information was deleted to fit in the page performance records in that country. The level of par- as were several other columns with bull aliases. More- ticipation of beef breeds in international genetic evalu- over, whether DNA is available or if the bull is of par- ations is less although initiatives such as BreedPlan and ticular importance in a country for genotyping can be INTERBEEF as well as pan-American are underway. noted. The complete list can be obtained from the au- If participating in international genetic evaluations the thor ([email protected]). Furthermore, requests extent of genotype-by-environment interactions should to join the list can be directed to the author. What is be quantified and the appropriate approach taken there- immediately obvious from Table 1 is that already some after in the genetic evaluation. To estimate precise ge- bulls have been genotyped more than once representing netic correlation between populations, good genetic a squandering of funds. Figure 4 outlines the number connectedness is needed (Berry et al., 2014b). This is of dairy bulls that were genotyped more than once up particularly true for novel traits which tend to be muea- to the year 2010 across 10 different countries. Almost sured in research herds. 700 genotypes were genotyped more than once; each genotype at the time cost approximately €160 implying Type of International Genomic Cooperation Initia- a squandering of over €110,000. tives in Genomics Advantages: Ability to identify animals that have al- Several alternative strategies of international cooper- ready been genotyped and thus engage with sharing of ation in genomics exist and some of these are briefly genotypes to avoid duplication of genotyping; no com- discussed. It should be noted that there is widespread petitive advantage is gained by genotyping the same international collaboration among dairy populations animal twice both in the sharing of genotypes and phenotypes. This Disadvantages: I cannot think of any disadvantage oth- is despite the points previously raised to that genomic er than for some unknown reason not wanting others selection in dairy is arguably considerably easier (i.e., to know what animals have been genotyped in a given less expensive) than in beef. population. Information on what dairy animals are genotyped is generally freely available. Which dairy ani- Sharing of Information on What Animals Have Been mals are genotyped and on what genotype platform in Genotyped. One of the easiest and least controversial the US is freely available at https://www.cdcb.us/eval. approaches to achieving useful international collabo- htm

Table 1. A small section of the international list of beef bulls genotyped ANIMAL_NAME ID DOB Brd IRL UK FRA BLUEBELL AIGLON CHLFRAM007185101623 24/01/1985 CH Want ABOUKIR CHLFRAM007185119662 07/01/1985 CH Illum_HD COMMANDEUR CHLFRAM007187126401 01/01/1987 CH Have DNA BANDIT LMSFRAM008786003322 14/02/1986 LM Want Illum_54K IMPERIAL LMSFRAM008793000421 10/01/1993 LM Want Illum_54K TANHILL RUMPUS LMSIRLM000000FBR092 24/04/1980 LM Illum_HD ESPOIR LMSFRAM008789003720 02/03/1989 LM HIGHLANDER LMSFRAM001692111209 01/01/1992 LM Illum_HD Illum_54K OMAR LMSFRAM001930098242 24/12/1998 LM Illum_HD Illum_54K KILKELLY DUKE AANIRLM272061330257 01/03/2007 AA Illum_HD DELFUR T-BONE SIMGBRM523461601799 09/04/2006 SI Illum_HD

71 types benefit the exporting country more. The greater the genomic relationships between the reference population and the candidate population the greater, on average, will be the accuracy of the genomic predictions (Habier et al., 2007; Pszczola et al., 2012). Therefore, if the back-pedigree of animals being exported into a country exists within that country’s reference population (assuming they also have phenotypic measures such as EPDs from pedigree or other descendants) then the accuracy of genomic predictions for those candidate animals will be greater. Hence the sharing of genotypes benefits both countries; the exporting country receives more accurate genomic proofs on Figure 6. Mean animal allele concordance rate for their candidate bulls while the importing country gains Illumina Bovine50 Beadchip genotypes across dif- access to potentially genetically elite individuals. ferent paternal half-sib progeny group sizes. Er- rors bars represent the, within animal, lowest and greatest mean concordance rate. Also represented (diamonds) is the mean animal allele concordance rate for a subset of the data across different parental half-sib progeny groups sizes when the paternal grand sire’s genotype is also available (Berry et al., 2014a).

Sharing of genotypes is also advantageous even if the animal has no phenotype in the country. This is because the genotype of a non-genotyped animal (but potentially with a phenotype) can be imputed from its progeny genotypes. Figure 6 shows that parental al- Figure 4. Presentation by Donagh Berry at the IN- leles can be imputed with, on average, ≥96% accuracy TERBULL business meeting in 2010 on the number if genotypes on ≥5 progeny are available (Berry et al., of animals that had been genotyped more than once 2014a). Sharing of different genotype platforms (i.e., across ten participating countries. The cost per different densities or different commercial providers) animal genotype at that time was €160 can also relatively easily be facilitated through imputation across genotyping panels (Druet et al., 2010; Berry et al., 2014c). Furthermore, access to genotypes of animals even with no phenotypes can be used to improve the accuracy of imputation of their descendants or pedigree genotyped on lower density panels.

The precedence already exists in the sharing of genomic information. Sharing of genotypes in dairy is occurring among many populations. The 1000 bulls’ genome project has collated to-date sequence data from over 1000 dairy and beef bulls. Furthermore, several Figure 5. Number of bull genotypes included in the thousand SNP and microsatellite genotypes were col- Irish genomic selection reference population for lated for the development of algorithms to convert SNP each year which were genotyped by Irish funding data to microsatellite data for parentage testing (Mc- (black bars) or obtained through bilateral sharing Clure et al., 2013). This approach has benefited the en- (diagonal bars) 72 tire global cattle industry by eliminating the necessity must therefore be (equal) investment. Genotypes can to SNP genotype animals already genotyped back-ped- also be used to achieve more accurate imputation from igree for microsatellite. The introduction of this tool in lower density panels. Ireland in March 2013 has already saved the beef indus- Disadvantages: Populations with a very large reference try €200,000 by not having to re-genotype back-pedi- population may have little to gain from sharing of gen- gree. Ireland is moving to parentage testing using just otypes if they already have most of the other available SNP data in both dairy and beef cattle. The advantage genotypes in their reference population, Furthermore, of this approach is that the SNPs, if undertaken as part the marginal benefit of additional genotypes in a ref- of a larger panel, can also be used for genomic selec- erence population diminishes as the size of the actual tion. Also, in Ireland the custom genotyping panel in- reference population increases (Figure 3). There is still cludes almost 2,000 research SNPs which can be used however a marginal benefit of additional genotypes on to validate in an independent commercial population. phenotyped animals even with many thousands of ani- This can be particularly useful to elucidate, at no cost, mals in the reference population. There is sometimes a if any SNPs strongly associated with novel traits are perception that genotype sharing should not be under- also antagonistically correlated with other performance taken because it was expensive to generate the popu- traits in commercial populations. lation; however it is usually exactly the same expense Sharing of genotypes of young bulls can be of for the other country assuming equal numbers of gen- particular benefit if the genotypes are run through each otypes are shared. Sharing is a less expensive way to country’s genomic prediction equations and the geneti- achieving a large reference population. cally elite animals identified and subsequently import- ed. Such an approach benefits the exporter (sells the Sharing of Phenotypes. Many dairy cow population germplasm) and importer (access to genetically elite share phenotypic information through the international germplasm). However as previously alluded to, this genetic evaluation at INTERBULL. Some beef pop- approach is best achieved if the genotypes and pheno- ulations also share phenotypic data via INTERBEEF, types of the back-pedigree of these candidate animals Breedplan and Pan-American initiatives. are already in the importing country’s genomic refer- Advantages: access to “phenotypes” on a large popula- ence population. tion of animals which increases the accuracy of genom- At the very least the genotypes of each animal ic prediction with the marginal benefit being greatest for the SNP parentage panel should be available with- when the reference population is relatively small as is out restriction. This panel cannot be used in genomic usually the case for novel traits. The sharing of pheno- selection but is extremely useful in parentage testing. types and genotypes can also be used as an independent An example of a document that could be used in population to evaluate the precision and robustness of the bilateral sharing of genotypes is given in Appendix developed genomic selection algorithms. This is partic- I. ularly important for novel traits where the population Advantages: The reference population size can be in- size is small. creased dramatically; in the case of the Irish dairy Disadvantage There is background intellectual property genomic selection breeding program, the size of the associated with the generation of phenotypes, especially reference population was increased 300% (Figure 5). for novel traits and acquiring such phenotypes are usu- This will increase the accuracy of genomic evaluations ally costly. There is therefore reluctance among some (Figure 3). The marginal benefit of additional geno- to provide these data free of charge to others. To over- types is greater when the reference population is small- come this however the approach described previously er (Figure 3) as is usually the case for novel traits. For on equal exchange of genotypes could be imposed for the (larger) exporting country the accuracy of genomic phenotypes. This however is only sensible if undertak- predictions on their candidate animals in the importing ing univariate (within country) genetic/genomic evalu- country is, on average, expected to be greater. The ap- ations and excluding phenotypes from a multi-popula- proach of sharing of genotypes should not be construed tion evaluation would not be recommended. Again to as an approach to facilitate the generation of genomic overcome this, a price per unit phenotype could be gen- breeding values for bodies that decided not to invest in erated; this could be relatively easily achieved using se- genomic selection; it involves bilateral sharing so there lection index theory. Then a value on each population’s

73 correlations with other populations could be generated. year with some crossreference bulls across years) that The consortium may purchase these phenotypes or may should be used in different populations to improve con- pay an annual licensing fee to have access to these data nectedness. Connectedness algorithms could be used for use in the multi-population evaluation. The price to identify populations that could benefit most from paid per population will differ based on the information such an initiative; such algorithms are commonly used content of the phenotypes (i.e., coheritability between to improve connectedness between flocks in sheep populations) but also on how much that population is (Fouilloux et al., 2008). Such an initiative is particular- also contributing to the database of phenotypes. ly important for novel traits which tend to be recorded mainly in research herds; thus a pan-global list of bulls Sharing of Genomic Keys. Collection of novel traits is for recommended use in research herds could be gener- generally a costly exercise and therefore the number of ated. Only a few progeny need to be generated thereby traits collected is usually limited. For example a pop- having a likely minimal impact on the objectives of the ulation may deeply phenotype for one health trait but research projects. not for others. Sharing of genomic keys among popula- Advantages: More precise estimates of genetic cor- tions that have phenotyped for a different suite of novel relations among populations necessary for inclusion in traits could provide potentially useful information on multi-trait genomic evaluations across populations to the likely correlated responses in other (not measured) increase the accuracy of genomic predictions novel traits. The validity of genomic keys from other Disadvantages: Could be difficult to reach a consensus populations could be relatively easily tested by pheno- on such a list of bulls given the likely different breeding typing a smaller number of animals and relating their objectives in different populations and may reduce the phenotypic values to those predicted from the shared statistical power of the experiment. genomic keys. Furthermore, visibility on the genomic keys from other population could help inform genomic Validation or Fine-mapping of Putative Causal Vari- prediction algorithms in that population for the same ants. Genomic selection requires the continual regen- phenotype and therefore place greater emphasis on ge- eration of genomic predictions including more recent nomic regions detected as significant from more than generations of phenotyped animals in the prediction one populations. Combined genomic keys can also process. Moreover, for novel traits generally measured provide a greater in depth knowledge of the underlying in research herds, the genomic relationships between biological pathways governing differences in perfor- these animals and the candidate population may be mance facilitating more powerful biological pathway low; the weaker the genomic relationship between the analysis. animals in the reference population and the candidate Advantages: Could remove the necessity to phenotype population the lower will be the accuracy of genomic for all novel traits possible but could also be useful in predictions (Habier et al., 2007; Pszczola et al., 2012). elucidating the genetic merit of a population for exam- The reason the accuracy of genomic predictions is ex- ple for diseases that currently do not exist in that pop- pected to decline over generations is due to recombi- ulation. nation during meiosis because the SNPs exploited in Disadvantages: There may be discontent in the sharing genomic predictions are very unlikely to be the causal of genomic keys that required considerable resources mutation and thus the linkage disequilibrium between to generate. Agreements can be put in place a priori the genotyped SNP and the true causal mutation can on either selling the genomic keys or direct sharing break down during meiosis. Hence, many research (with some financial remuneration if differences exist projects are engaged with attempted to locate the actual between populations in the size of the reference popu- causal mutation thereby avoiding the necessity to con- lation of the cost of generating the phenotypes). tinually re-estimate the genomic prediction equations. To facilitate the discovery of causal mutations, a very Pan International Bull List to Increase Connected- large population of animals is required to ensure ade- ness. Genetic connectedness is fundamental for the quate statistical power. Therefore, few, if any, animals estimation of precise genetic correlations among pop- exist to validate the discoveries or fine-map the genom- ulations (Berry et al., 2014b). There may be an advan- ic regions further. Different populations tend to have tage of generating a small list of bulls (varying every different linkage phases so therefore using alternative

74 populations could be extremely beneficial in fine-map- predication accuracy will be less subjected to erosion ping further and eventually identifying the causal muta- over generations. tion if segregating in the validation population. Disadvantage: Many think they will make millions out Advantage: Detection of the causal mutation or mu- of patenting of causal mutations. A good example of tations in very strong linkage disequilibrium with the how this does not always materalise is the K232A poly- causal mutation should increase the accuracy of ge- morphism in DGAT1 which has a very large effect nomic predictions (especially across breeds) and the

Table 2. Number of lactations (N) as well as the mean, genetic standard deviation, heritability (h2) and repeatability (t) of dry matter intake in all countries (i.e., all countries) or each individual country.

Mean σg Country N (kg DM/day) (kg DM/day) h2 t Cows All 10,641 19.7 1.13 0.27 (0.02) 0.66 (0.01) Canada 411 22.2 1.01 0.11 (0.11) 0.46 (0.06) Denmark 668 22.1 1.48 0.46 (0.12) 0.62 (0.04) Germany 1141 20.2 0.64 0.16 (0.06) 0.84 (0.05) Iowa 398 23.5 1.48 0.58 (0.12) Ireland 1677 16.7 0.88 0.29 (0.07) 0.64 (0.02) Netherlands 2956 21.4 1.15 0.38 (0.04) 0.54 (0.03) UK 2840 17.4 1.07 0.30 (0.06) 0.72 (0.02) Wisconsin 447 24.9 0.90 0.19 (0.13) Australia 103 15.6

Heifers Australia 8.3 0.77 0.39 (0.08 New Zealand 7.6 0.66 0.25 (0.07)

Table 3. Genetic correlations (below diagonal; standard errors in parenthesis) between dry matter intake measured in groups of countries1 as well as the number of sires common (above diagonal; sires plus maternal grandsires in common in parenthesis) between the groups of countries.

Region North-America EU high-input EU low-input Grazing North-America 39 (72) 4 (10) 6 (8) EU high-input 0.76 (0.21) 125 (144) 23 (28) EU low-input 0.79 (0.38) 0.84 (0.14) 4 (4) Grazing 0.14 (0.43) 0.33 (0.20) 0.57 (0.43) 1 North-America = Iowa + Wisconsin + Canada; EU-high input = Netherlands+Ger- many+Denmark+high input feeding treatment in the UK; EU-low input= low input feeding treatment in the UK; Grazing = Ireland + Australia;

75 on milk production traits in dairy cattle (Berry et al., al., 2011); could a similar approach be adopted in the 2010). Royalties must be paid if this extremely large prediction of meat quality using rapid measures? The genomic effect is to be used in a breeding program but number of countries participating in prediction of milk to my knowledge few, if any bodies actually exploit this quality initiative has since more than doubled. This is mutation in their breeding program. There is a growing because the benefit of collaborating far outweighs the consensus that discovered causal mutations should be benefits of not. published in the scientific literature. An alternative it to retain the mutation as a trade secret within the company Four Simple Steps Required for International Col- that made the discovery. The downside of this approach laboration in Genotype Sharing to Happen! is that others may detect the mutation in the near future 1. Decide whether or not you want accurate and publish it. genomic predictions for your breed or population – if so then international collaboration is the best approach to achieve this Case Study of International Genetic and Genomic Evaluations in Dairy Cows for a Novel Trait– Dry 2. Email [email protected] if you are Matter Intake willing to participate in a publically available Despite the large contribution (~60%) of feed excel spreadsheet on what animals are to the variable costs of production in dairy cattle sys- genotyped in your population and on what tems (Ho et al., 2005; Shalloo et al., 2004), feed in- genotyping platform take is currently not explicitly included in the breeding 3. You can also let it be known whether or not you goal of any dairy cattle population. This omission is are willing to share genotypes, either bilaterally principally due to an absence of sufficient feed intake or multilaterally. information to estimate breeding values of individual animals. Collation of international data on feed intake 4. Exchange genotypes and associated information from research herds and nu- cleus breeding herds is one approach to increase the Literature Cited quantity of feed intake data available for estimation of Berry, DP. 2012. Across-breed genomic evaluations in breeding values. This was the motivation of the glob- cattle. European Association of Animal Production. al Dry Matter Intake initiative (gDMI) participated in Bratislava, Slovakia, 27-31 August 2012. P 127. by 9 countries. A total of 224,174 feed intake test-day Berry, D.P., McParland, S., Kearney, J.F., Sargolzaei, records from 10,068 parity one to five records from M., and Mullen, M.P. 2014a. Imputation of un-gen- 6,957 Holstein-Friesian cows, as well as records from otyped parental genotypes in dairy and beef cattle 1,784 growing Holstein-Friesian heifers were collated from progeny genotypes. Animal (in press). from 9 countries in the US, Europe and Austral-Asia. Berry, D.P., Coffey, M.P., Pryce, J.E., de Haas, Y., Animal and back-pedigree genotypes were also pooled Løvendahl, P., Krattenmacher, N., Crowley, J.J., (Pryce et al., 2014) with the aim of undertaking an in- Wang, Z., Spurlock, D., Weigel, K., Macdonald, K, ternational genomic evaluation for feed intake which and Veerkamp, R.F. 2014b. International genetic is still being researched (de Haas et al., 2014). Genetic evaluations for feed intake in dairy cattle through parameter estimates for the different populations are in the collation of data from multiple sources. J. Dairy Tables 3 and 4, respectively (Berry et al., 2014b). Of Sci. (in press). less specific interest here are the actual results, but the point being made is that nine countries understood that Berry, D.P., Howard, D., O’Boyle, P., Waters, S., Ke- the only way to achieve accurate genomic predictions arney, J.F. and McCabe, M. 2010. Associations for this novel trait was to pool their respective data- between the K232A polymorphism in the diacyl- sets. The same approach can be easily applied to other glycerol-O-transferace 1 (DGAT1) gene and perfor- novel phenotypes or to different breeds. For example mance in Irish Holstein-Friesian dairy cattle. Irish three countries pooled information on milk quality to J. of Agric. & Food Res. 49:1-9. derive more accurate rapid predictors of these novel traits from infrared spectroscopy in milk (Soyeurt et

76 Berry, D.P., McClure M.C., and Mullen, M.P. 2014c. Saatchi M., McClure M.C., McKay S.D., Rolf M.M., Within and across-breed imputation of high densi- Kim J., Decker J.E., Taxis T.S., Chapple R.H., Ra- ty genotypes in dairy and beef cattle from medium mey H.R., Northcutt S.L., Bauck S., Woodward B., and low density genotypes. J. Anim. Breed. Gen. Dekkers J.C.M., Fernando R.L., Schnabel R.D., (in press). Garrick D.J. and Taylor J.F. 2011. Accuracies of ge- Calus, M.P.L., Berry, D.P., Banos, G., de Haas, Y. and nomic breeding values in American Angus beef cat- Veerkamp, R.F. 2013. Genomic selection: the op- tle using K-means clustering for cross-validation. tion for new robustness traits?. In: Advances in Genet. Sel. Evol. 43:40. Animal Biosciences, Cambridge University, 4 : McClure, M.C., Sonstegard, T.S., Wiggans, G.R., Van 618-625. Eenennaam, A.L., Weber, K.L., Penedo, C.,T., Ber- Daetwyler H.D., Villanueva B. and Woolliams J.A. ry, D.P., Flynn, J., Garcia, J., Carmo, A.S., Regi- 2008. Accuracy of predicting the genetic risk of tano, L.C.A., Albuquerque, M., Silva, M.V.G.B., disease using a genome-wide approach. PLoS One Machado, M.A., Coffey, M., Moore, K., Boscher, 3, e3395. M.Y., Gene 2013. Imputation of microsatellite alleles from dense SNP genotypes for parentage veri- de Haas Y., Pryce, J.E., Berry, D.P., and Veerkamp, R.F. fication across multiple Bos taurus and Bos indicus 2014. Genetic and genomic solutions to improve breeds. Frontiers in Genet. 4 : 1-11. feed efficiency and reduce environmental impact of dairy cattle. Proc. World Cong. on Gen. Appl. to McParland, S., Berry, D.P. and Kearney, J.F. 2014. Ret- Livest. Prod. Vancouver. August 2014. rospective analysis of the accuracy of genomic selection in Irish dairy cattle. Proc. Irish Agric. Res. Druet, T., Schrooten, C., and de Roos A.P. 2010. Im- Forum. March 10-11, 2014. Tullamore, Ireland. putation of genotypes from different single nucleotide polymorphism panels in dairy cattle. J. Dairy McParland, S., Kearney, J.F., Rath, M. and Berry, D.P. Sci. 93:5443-5454. 2007. Inbreeding trends and pedigree analysis of Irish dairy and beef cattle populations. J. Anim. Sci. Fouilloux, M.N., Clément, V., and Laloë, D., 2008. 85:322-331. Measuring connectedness among herds in mixed linear models: From theory to practice in large- Prendiville R and McHugh, N. 2014. Comparative live sized genetic evaluations. Genet. Sel. Evol. 40:145- weight, body condition score at breeds, onset of 159. puberty and age at first calving for heifers of high and low maternal Index. Proceedings of the Irish Goddard, M 2009. Genomic selection: prediction of Agricultural Research Forum. March 10-11, 2014. accuracy and maximization of long term response. Tullamore, Ireland. Pp117. Genetica 136: 245–257. Pryce, J.E., Johnston J, Hayes B.J., Sahana G, Weigel Habier D., Fernando R. and Dekkers J. 2007. The K.A., McParland S., Spurlock D., Krattenmach- impact of genetic relationship information on ge- er N., Spelman R.J., Wall E., Calus M.P.L. 2014. nome-assisted breeding values. Genetics 177: Imputation of genotypes from low density (50,000 2389–2397. markers) to high density (700,000 markers) of cows Ho, C., Nesseler R, Doyle P, Malcolm B. 2005. Future from research herds in Europe, North America, and dairy farming systems in irrigation regions. Aus- Australasia using 2 reference populations. J. Dairy tral. Farm Bus. and Manage. J. 2: 59–68. Sci. 97: 1799-1811 Karoui S., Carabaño M.J., Díaz C. and Legarra A. Pszczola M., Strabel T., Mulder H.A. and Calus M.P.L. 2012. Joint genomic evaluation of French dairy 2012. Reliability of direct genomic values for ani- cattle breeds using multiple-trait models. Genet. mals with different relationships within and to the Sel. Evol. 44:39. reference population. J. Dairy Sci. 95: 389–400.

77 Purfield D.C., Bradley, D.G, Evans R.D., Kearney, J.F. and Berry, DP. 2014. Genome-wide association study for calving performance using high density genotypes in dairy and beef cattle. BMC genomics. (submitted). Rendel J., and Robertson A. 1950. Estimation of genetic gain in milk yield by selection in a closed herd of dairy cattle. J. Genetics. 50:1-8. Sellner, E. M., J.W. Kim, M.C. McClure, K.H. Taylor, R.D. Schnabel and J.F. Taylor. 2007. Applications of genomic information in livestock. J. Anim. Sci. 85 : 3148-3158. Shalloo L., Dillon P., Rath M. and Wallace, M. 2004. Description and validation of the Moorepark Dairy Systems Model (MDSM). J. Dairy Sci. 87:1945- 1959. Soyeurt, H., Dehareng, F., Gengler, N., McParland, S., Wall, E., Berry, D.P. and Coffey, M. 2011. Mid-infrared prediction of bovine milk fatty acids across multiple breeds, production systems, and countries. J. Dairy Sci. 94: 1657-1667. Spelman, R., Hayes, B.J. and Berry, D.P. 2013. Use of molecular technologies for the advancement of animal breeding: Genomic selection in dairy cattle populations in Australia, Ireland and New Zealand. Anim. Prod. Sci. 53: 869-875.

78 APPENDIX I. Agreement between

BODY1 NAME BODY1 ADDRESS BODY1 ADDRESS BODY1 ADDRESS hereinafter referred to as “XXXXXX” and

BODY2 NAME BODY2 ADDRESS BODY2 ADDRESS BODY2 ADDRESS

hereinafter referred to as “YYYYYY”

Agreement dated this [insert date] and continuing until this [insert date + three years] or until terminated under guidelines in Article 5.

1 Purpose of the agreement

XXXXXX and YYYYYY agree to collaborate in the area of genomic evaluation and selection namely through:

• exchanging information about methods used for genomic evaluation and selection in cattle, • exchange of animals genotypes to avoid multiple genotyping of the same animals, and • exchange of genotypes of animals in general.

2 Exchange of information about methods used for genomic evaluation and selection

(a) XXXXXX and YYYYYY agree to exchange the following information solely for the purposes referred to herein (the Purpose):

i. their respective methods used for genomic evaluation and selection in cattle; ii. animal genotypes and the identity of animals proposed to be genotyped; iii. the estimation of effects at the single nucleotide polymorphisms (SNP) and the conclusions derived for the corresponding breeding program; and iv. such other information as is referred to herein (the Confidential Information)

(b) The Confidential Information exchange will take place on a regular basis at the conferences and other forums as agreed between the parties.

(c) For the avoidance of doubt any exchange of Confidential Information specifically excludes a license to a software package that one party may use to implement genomic evaluation or selection.

79 (d) Nothing in this agreement shall be construed as assigning or otherwise transferring any proprietary rights including Intellectual property rights in a party’s Confidential Information to the other party.

3 Confidentiality

(a) All Confidential Information given by a party to the other party under the terms of this agreement is valuable information of the disclosing party and the receiving party undertakes to keep the Confidential Information secret and to protect and preserve the confidential nature and secrecy of the Confidential Information.

(b) A receiving party: i. must not disclose Confidential Information of the disclosing parties to any person except as permitted under this Agreement; ii. must not permit unauthorised persons to have access to the disclosing party’s Confidential Information; iii. must not make or assist or permit any person including its officers and employees, agents or advisors (Represen- tatives) to make any unauthorised use, disclosure or reproduction of the disclosing Party’s Confidential Informa- tion iv. must take reasonable steps to enforce the confidentiality obligations imposed or required to be imposed by this Agreement and must co-operate with the disclosing party in any action that it may take to protect the confidentiality of the Confidential Information disclosed under this Agreement.

(c) A receiving party must only use the disclosing party’s Confidential Information for the Purpose and must only disclose Confidential Information to its Representatives for the conduct of the Purpose and then only on a need to know basis.

(d) Each party must ensure that its Representatives do not do or omit to do anything which if done or omitted to be done by the receiving party would be a breach of the receiving party’s obligations under this agreement.

4 Avoid multiple genotyping of the same animals

XXXXXX and YYYYYY will use the same genotyping technology and platform, namely the same SNP-chip. Currently, the Illumina Bovine SNP 50™ BeadChip and the Illumina Bovine SNP HD™ BeadChip will be used. XXXXXX and YYYYYY will exchange the identity of the animals they have or plan to genotype.

5 Exchange of animals genotypes

(a) XXXXXX is granted the right to obtain genotypic information of genotyped sires from YYYYYY and YYYYYY are granted the right to obtain genotypic information of genotyped sires from XXXXXX. Each party shall exchange approximately equal numbers of genotyped sires to the other. Each party, XXXXXX and YYYYYY, will retain ownership of the genotyping information they provided. (b) The genotyping information XXXXXX obtains from YYYYYY may be used by XXXXXX for genetic evaluation in the COUNTRY base and scale and selection purposes only. All results and products originating from genotyping information obtained from YYYYYY belong to XXXXXX. (c) The genotyping information YYYYYY obtains from XXXXXX may be used by YYYYYY for genetic evaluation in the COUNTRY base and scale and selection purposes only. All results and products originating from genotyping information obtained from XXXXXX belong to YYYYYY. (d) XXXXXX and YYYYYY may extract the genotype for parentage testing SNPs and provide these to third parties for the purpose of validating parentage of animals in their respective cattle populations.

80 6. Termination

(a) XXXXXX and YYYYYY have the right to terminate this agreement by giving 3 months written notice to the other party.

(b) Termination of this agreement will be without prejudice to any other rights and remedies of the parties arising out of any default which occurs before the termination and will be without prejudice to any claim for money payable at the time of termination in respect of work done, genotyping information exchanges or liabilities incurred before the termination.

(c) Upon termination or expiration of this agreement, or the request of the disclosing party, the receiving party will deliver to the disclosing party (or with the disclosing party’s prior consent, destroy or erase);

i. all material forms of the other party’s Confidential Information (including biological or other samples) in its possession or the possession of any of its Representatives; and ii. a statutory declaration made by an authorised officer of the party declaring that the provisions of this article have been complied with.

(d) Return of material forms of Confidential Information does not release a party or its Representatives from the confidentiality obligations set out in this agreement

(e) The obligations of confidentiality contained herein survive termination or expiration this agreement.

7 Dispute Resolution

In the event of any dispute between the parties in relation to the terms and conditions of this agreement, the parties will first seek to resolve such dispute by promptly giving notice of the dispute to the other party and in good faith endeavour to resolve such dispute. If the dispute remains unresolved for 20 days, the parties will first seek a resolution through the use of mediation. If the dispute still remains unresolved, a resolution through the use of arbitration shall be tried and only as a last resort, resolution is pursued through courts. Nothing in this Agreement will be interpreted as preventing a party from seeking urgent interlocutory relief through the courts to protect its interest in the Confidential Information disclosed to the other party.

8 Governing Law

This Agreement shall be governed by the laws of COUNTRY.

Signed for XXXXXXX: Signed for YYYYYYYY:

Signature Date Signature Date

81 ECONOMIC BENEFITS OF USING GENETIC year through the implementation of selection for SLEECTION TO REDUCE THE PREVALENCE OF cattle that are less susceptible to BRDC. BOVINE RESPIRATORY DISEASE COMPLEX IN BEEF FEEDLOT CATTLE Introduction Bovine respiratory disease complex is the result of of H.L. Neibergs1, J.S. Neibergs1, AJ Wojtowicz1, viral and bacterial pathogens and is the leading cause of illness and death in feedlot cattle (USDA 2001, Ga- 2 3 J.F. Taylor , C.M. Seabury , gea et al. 2006, Snowder et al 2006). The prevalence of J. E. Womack3 bovine respiratory disease complex (BRDC) detected in feedlot cattle varies by year with a 15 year range

1 of from 5% to 44%, and also by season, with higher Washington State University, Pullman prevalence rates in the fall and winter (Snowder et 2University of Missouri, Columbia al 2005, Miles 2009). The average prevalence rate of 3Texas A&M University, College Station BRDC was 16.2%, with virtually all feedlots (96.9%) reporting one or more cases between July 1, 2010 and June 30, 2011 (USDA 2011). Recent reports have Abstract indicated that greater than 60% of all cattle in the feedlot have lung lesions resulting from BRDC and The prevalence of bovine respiratory disease complex many of these animals were undetected as suffering (BRDC) has remained unchanged for decades despite from an illness (Schneider et al. 2009). Of animals efforts to suppress the disease through prevention that showed noticeable signs of illness, BRDC was the programs aimed at vaccination and metaphylaxis. An most common cause (67% to 82%) of illness detect- additional approach that focuses on host response ed in feedlot cattle (Edwards 1996, USDA 2011). An to infection by the pathogens responsible for BRDC, estimated 1.4% of all feedlot cattle die from BRDC through the selection of animals that are less suscep- prior to harvest. tible to the disease, has been undertaken as part of the USDA funded “Integrated Program for Reduc- The high prevalence of BRDC in feedlot cattle has ing BRDC in Beef and Dairy Cattle” with Dr. James not fallen in spite of best management practices Womack at Texas A&M University as the Project and vaccination programs (Gagea et al. 2006, Miles Director. This study, now beginning its fourth year, 2009). According to the USDA (2011), the majority of has found that estimates of heritability for suscepti- feedlots over 1,000 head used one respiratory vaccine bility to BRDC were greater than 17% for a binary to combat disease. Unfortunately, only about 25% of case-control definition of BRDC and greater than cattle were vaccinated for both the viral and bacterial 29% for a semi-quantitative (clinical score) defini- pathogens associated with BRDC. Specifically, 96.6% tion of BRDC in a commercial feedlot in Colorado. of feedlots vaccinated for bovine viral diarrhea virus The higher heritability estimate for the more precise (BVDV), 93.7% vaccinated for infectious bovine definition of BRDC was anticipated as heritability im- rhinotracheitis virus (IBR), 85.1% vaccinated for proves as accuracy of a measured phenotype (in this parainfluenza 3 virus (PI3) and 89.5% of feedlots vac- case the BRDC phenotype) improves. The estimated cinated for bovine respiratory syncytial virus (BRSV) annual rate of genetic gain due to selection on these (USDA 2011). Approximately 66% of feedlots used phenotypes was estimated at 1.2% (case-control) and vaccines that incorporated BRDC bacterial pathogens 2.1% (clinical scores). The economic cost of $204.10 Hemophilus somnus and Mannheimia haemolyti- per BRDC feedlot steer was determined through the ca (previously named Pasteurella haemolytica) and loss of carcass quality, death and treatment costs. 21.8% vaccinated against Mycoplasma bovis. When this value was combined with the 16.2% national prevalence of BRDC in the feedlot and the es- One strategy used to prevent or minimize an out- timated reduced prevalence of BRDC (rate of genetic break of BRDC is to treat cattle with an injectable gain made by selecting to increase the proportion of antibiotic (metaphylaxis) for BRDC pathogens. Fac- cattle that are more resistant to disease), the feedlot tors that heightened concerns of BRDC and caused industry could gain between $8 and $16 million per feedlots to consider metaphylaxis included: cattle 82 with a poor appearance on arrival (88.4% of feed- To adopt selection as a means of reducing BRDC in lots would consider metaphylaxis), the presence of feedlot cattle, it must be feasible and profitable for one or more animals from the same source affected the feedlot industry. The average cost per treatment with BRDC (83.8%), the presence of BRDC affected for cattle with BRDC in feedlots over 1,000 head cattle in the same pen (70.5%), or if cattle came from was reported as $23.60 by the USDA (2011) but the a sale barn (88.3%) (USDA 2011). In all, 59.3% of average number of treatments given per affected feedlots used metaphylaxis treatment for some cattle animal was not provided. However, for U.S. cattle in the feedlot. Unfortunately this strategy, as well as weighing less than 700 pounds, over 18% did not vaccination, have collectively failed to reduce the respond to their first treatment, 4% died, 14.9% were prevalence of BRDC and further suggests that other retreated and 2.3% were considered chronic and were approaches, such as focusing on the host response subsequently shipped to slaughter prior to reaching to pathogen challenges, may be helpful in reducing a normal slaughter weight. For cattle weighing over the prevalence. The use of new approaches, such as 700 pounds, just over 13% did not respond to their genomic selection, is supported by studies providing first treatment, 3.6% died, 12.4% were retreated and evidence that genetic factors are important in BRDC 1.9% were considered chronic (USDA 2011). For light prevalence rates. cattle weighing less than 700 pounds treated a second time, 63.1% of cattle responded to treatment, 13.3% Although the environment and stress play a major died and 12% were treated a third time. For cattle role in BRDC infection rates, there is increasing evi- heavier than 700 pounds that were retreated, 69.5% dence that susceptibility to BRDC is at least partially responded, 13.2% died and 17.1% were treated a third under direct genetic control. Differences in BRDC time (USDA 2011). Although the exact cost of BRDC susceptibility have been found between cattle breeds to the beef industry is unknown, it has been estimat- and sire lines, and heritability estimates in the low to ed to be responsible for losses of over $800 million moderate range (0.04 to 0.21) have been reported for annually and represents the single most economically BRDC susceptibility in beef and dairy cattle (Lyons important disease of cattle (Chirase and Green 2001, et al 1991, Muggli-Cockett et al. 1992, Snowder et al Snowder et al. 2006a, Gagea et al 2006). The aims of 2005, Heringstad et al. 2008, Schneider et al 2009, this study were to estimate the heritability of BRDC Neibergs et al 2013, Seabury et al 2014). This suggests susceptibility in Bos taurus feedlot cattle at a com- that selecting for BRDC resistant cattle could have a mercial facility that did not treat cattle with meatphy- substantial impact on BRDC prevalence (Snowder laxis, estimate the rate of genetic change that would et al 2009). A limited number of quantitative trait result from selection for cattle that were less suscepti- loci related to bovine health, including resistance to ble to BRDC and determine the economic gain of se- BRDC, have been reported (Casas and Snowder 2008, lecting cattle for reduced BRDC susceptibility in the Settles et al 2009, Zanella et al 2011). New tools are feedlot based on the estimated rate of genetic change. now available to investigate the role of genetics in diseases such as BRDC that were not available just a few Materials and Methods years ago. These resources have now been harnessed to identify the bovine genomic regions associated Nine hundred ninety-five Bos taurus beef cattle were with BRDC susceptibility so that breeding less sus- evaluated using the BRDC diagnostic criteria of ceptible breeding stock can be identified and utilized McGuirk (2008) and determined to be either affect- (Neibergs et al 2013, Seabury et al 2014). The identi- ed with BRDC (n=497) or to be unaffected (n=498). fication of individual genetic differences in cattle that Animals’ health statuses were defined by clinical predispose them to enhanced susceptibility to BRDC signs of fever, cough, nasal discharge, and either serves as the basis for selecting cattle that are less ocular discharge or the ear position or head tilt scores likely to become ill as breeding stock. The develop- (McGuirk 2008). For each clinical sign, a numerical ment of genomic breeding values for sires that are less value of 0 to 3 was assigned based on the severity of susceptible to BRDC is underway as part the ongoing the clinical signs. Values for ocular discharge and ear USDA-funded multi-institutional research project position/head tilt were compared and the largest of “Integrated Program for Reducing BRDC in Beef and these values was summed with all of the other clini- Dairy Cattle” (www.brdcomplex.org). cal score values to reach a cumulative score. Animals with summed cumulative scores ≥5 were deemed 83 BRDC affected and animals with summed cumulative To estimate the rate of genetic change, the equation scores <5 were deemed unaffected. The mean clinical described by Falconer (1989) was used: score for cases was 8.04 ± 1.23 and the mean score for controls was 2.06 ± 0.037. All cases and controls were housed together in the same pens until harvest. Weights of animals at diagnosis, finished weights, days until harvest, treatment costs and estimated feed costs were provided for study animals. Treatment costs were based on a one-time injectable antibiotic treatment for BRDC as cattle were not retreated per where ΔBV/t is the rate of genetic change per year, the policy of the feedlot facility. The steers were mar- which represents the reduction in BRDC prevalence keted as a pen when they reached a finished weight. in feedlot steers, i is the standardized selection inten- Six lots were shipped throughout the study period sity, r is the accuracy of selection, σa is the additive and were followed to processing where the carcasses genetic standard deviation of the trait of interest, and were evaluated for yield and quality grade. Hot car- L is the generation interval in years. cass weight was provided for all study animals. The following parameters were assumed to estimate Heritability estimates for BRDC susceptibility were the model. In a typical beef cow-calf operation, the obtained by GenABEL/GRAMMAR (GenABEL.org) annual cow culling rate is between 13% and 20%, from relationship matrixes obtained from genotypes so for this example, we used a cow culling average of each animal derived from the Illumina BovineHD of 15%. If sexed-semen was not used and half of the assay that contains 778,000 single nucleotide poly- calves were heifers, then 30% of the heifers would morphisms (SNPs). All cattle were steers and consist- need to be retained to maintain a constant herd size. ed of 908 Angus, 18 Charolais, 25 Hereford, and 44 This corresponds to a standardized selection intensi- Red Angus. To account for potential breed differences ty coefficient of 1.16. The accuracy of selection was in susceptibility to BRDC, animal breed was fit as a assumed to equal the square root of the heritability fixed effect in the model used to obtain heritability for BRDC (42% for case-control and 54% for clini- estimates. Data were filtered for quality at both the cal scores) that would be realized from phenotypic animal and SNP level, such that animals with a geno- selection. This would form a conservative estimate typing success rate of less than 90% (n=63 animals), for the accuracy of prediction of molecular breeding or SNPs that failed to genotype greater than 95% of values for susceptibility to BRDC. The genetic stan- the time or that had minor allele frequencies less than dard deviation for BRDC prevalence was based on 1%, were removed. In addition, animals with ambigu- the following assumptions and calculations: BRDC ous genetic gender identification (n=3) were removed prevalence in beef steers will vary between opera- leaving a total of 932 males and 678,895 SNPs for the tions, seasonally and annually. The USDA (2011) av- analyses. Two different phenotypes for BRDC were erage prevalence of BRDC for feedlot cattle of 16.2% used to estimate heritabilities. The first phenotype (with a standard error of 1.4) was used as the BRDC was a binary case-control phenotype where cases had prevalence rate. The binomial phenotypic variance of McGuirk health scores ≥5 and controls had scores BRDC susceptibility with a prevalence rate of (p) can <5 and will be referred to as the ‘case-control’ pheno- be calculated as p(1-p) and, assuming that heritability type. The second BRDC phenotype used numerical is constant (independent of p), the additive genetic values of the McGuirk system (that ranged from 0 variance for a prevalence p is VA = h2p(1-p). Thus, to 12) as a semi-quantitative phenotype and will be for a heritability of 17.7% (case-control) or 29.2% referred to as the ‘clinical score’ phenotype. The her- (clinical score), the additive standard deviations for itability estimate for the case-control phenotype was the prevalence rate of 16.2% are σa = 0.1563 and 17.7% and was 29.2% for the clinical score pheno- 0.1984, respectively. The generation interval (L) for types. These estimates were similar to those estimated beef cows was estimated at 6 years. Biologically, the in previous studies and as estimated by investigators shortest possible generation interval is the sum of age of the BRDC-CAP for dairy calves (Lyons et al. 1991, at sexual maturity and gestation length, or approxi- Neibergs et al 2013, Seabury et al 2014). mately 2 years of age. 84 Results and Discussion The rates of genetic change for the case-control phenotype was 1.28% with a BRDC prevalence rate of 16.2% The rates of genetic change for the clinical score phenotype was even higher at 2.07% for BRDC as defined by clinical scores (Table 1). Direct costs attributable to BRDC include declines in carcass quality, death losses, treatment and labor Table 1. Factors Affecting the Rate of Genetic Change costs, and prevention costs. In this study, losses due in Reducing BRDC Susceptibility to carcass quality and death, and costs for treatment cattle were estimated to be affected with BRDC in were used to estimate direct costs. Prevention costs 2013. A conservative estimate of the cost of BRDC (vaccination and best management practices imple- to the feedlots (based on a single treatment cost, and mented at the feedlot) were identical between BRDC loss of carcass value) was determined to be $204.10 cases and controls and so were not estimated for this per animal or $830,658,210 in total losses to the study. Labor costs for treating BRDC cases were not feedlot industry. With the current estimates of the provided by the feedlot and so were not included in rate of genetic gain (1-2%, see Table 1) that could be the direct costs. Table 2 presents the quality grades achieved through selection for cattle that were less and death loss data for the cattle affected and un- susceptible to BRDC, the feedlot industry could real- affected with BRDC. The BRDC cases had a lower ize gains between $8,306,582 to $16,613,164 per year number of choice animals compared with healthy based on 2013 costs and market prices, by selecting animals (P=0.005), but a similar number of select car- for cattle that are less susceptible to BRDC. casses between cases and controls (P>0.05). The drop in carcass value shown in Table 2 reflects that the loss in quality grade of BRDC affected animals was not due to a simple slip of quality grade from choice to select, but a more extreme loss of carcass value to that of condemned ($0 value), railers (carcasses with quality issues that result in a standard value) or animals that died prior to harvest ($0 value). The average loss in value of BRDC cases compared to controls was Table 2.Carcass Quality of Bovine Respiratory Dis- $162.78 per head in 2013. ease Complex Cases and Controls The average treatment cost of the single BRDC treat- Conclusions ment of an injectable antibiotic was $41.32 per head. Because cases and controls were co-mingled, fed and Genomic selection for health traits, such as BRDC, harvested together, there was no difference (P>0.05) offers new approaches to reduce the prevalence of between cases and controls on rate of gain, hot economically important diseases. New technologies carcass weight or yield grade. When treatment costs allow the identification of cattle that are less sus- were combined with the losses due to carcass quality, ceptible to BRDC and the opportunity to select less the estimated total direct cost of each BRDC case in susceptible breeding stock so that the next generation the feedlot was $204.10. of feedlot cattle will be less likely to be affected with BRDC. The use of molecular breeding values in sires In 2013, 9,131,500 heifers and 16,003,400 steers and elite dams has become common for cattle geno- were harvested from U.S. feedlots that contained typed through commercial companies and/or breed 1,000 or more head (http://quickstats.nass.usda. associations. As part of the aims for the ongoing gov/results/135554B0-FDB3-34F2-A5F9-8ADBFE- “Integrated Program for Reducing BRDC in Beef and BAC18D) for a total of 25,134,900 animals. With the Dairy Cattle” the genomic regions that are predictive most current national estimate of BRDC prevalence of cattle that are less resistant to BRDC will become in feedlots of 16.2% (USDA 2011), 4,071,854 feedlot publicly available. These SNPs will then be 85 freely available to be placed on commercial genotyp- Miles, D.G. 2009. Overview of the North American ing platforms to benefit the beef and dairy industries. beef cattle industry and the prevalence of bovine Molecular or genomic breeding values for susceptibil- respiratory disease (BRD). Anim. Health Res. Rev. ity to BRDC can be computed for genotyped cattle so 10:101-103. that selection decisions based on BRDC susceptibility Muggli-Cockett N.E., L.V. Cundiff, K.E. Gregory. may be made across the industry. The use of genomic 1992. Genetic analysis of bovine respiratory disease selection offers significant opportunities to reduce in beef calves during the first year of life. J. Anim. BRDC prevalence and gain increased profitability in Sci. 70:2013-2019. the beef feedlot industry. Neibergs, H.L., C.M. Seabury, J.F. Taylor, Z. Wang, Acknowledgement E. Scraggs, R.D. Schnabel, J. Decker, A. Wojtowicz, J.H. Davis, T.W. Lehenbauer, A.L. Van Eenennaam, This work was supported by the USDA-NIFA grant S.S. Aly, P.C. Blanchard, B.M. Crossley. 2013. Iden- no. 2011-68004-30367. tification of loci associated with Bovine Respiratory Literature Cited Disease in Holstein calves. 2013. Plant & Animal Genome XXI, San Diego, California. Casas E., G.D. Snowder. 2008. A putative quantita- Schneider, M.J., R.G. Tait, W.D. Busby, J.M. Reecy. tive trait locus on chromosome 20 associated with 2009. An evaluation of bovine respiratory disease bovine pathogenic disease prevalence. J Anim. Sci. complex in feedlot cattle: Impact on performance 86:2455-2460. and carcass traits using treatment records and lung Chirase, N.K., L.W. Grene. 2001. Dietary zinc and lesion scores. J. Anim Sci. 87: 1831-1827. manganese sources administered from the fetal Seabury, C.M., J.F. Taylor, H.L. Neibergs, BRD Con- stage onwards affect immune response of transit sortium. 2014. GWAS for differential manifestation stressed and virus infected offspring steer calves. of clinical signs and symptoms related to bovine Anim. Feed Sci. Tech. 93:217-228. respiratory disease complex in Holstein calves. Edwards, A.J. 1996. Respiratory diseases of feedlot 2014. Plant & Animal Genome XXII, San Diego, cattle in the central USA. Bovine Pract. 30:5-7. California. Falconer, D. S. (1989). Introduction to Quantitative Settles, M., R. Zanella, S.D. McKay, R.D. Schnabel, Genetics. 3rd ed. Longman Scientific and Technical, J.F. Taylor, T. Fyock, R.H. Whitlock, Y Schukken, New York, NY. JS Van Kessel, J Karns, E Hovingh, JM Smith, HL Gagea, M.I., K.G. Bateman, T. van Dreumel, B.J. Neibergs. 2009. A whole genome association anal- McEwen, S. Carman, M. Archambault, R.A. Sha- ysis identifies loci associated with Mycobacterium nahan, J.L. Caswell. 2006. Diseases and pathogens avium subsp. paratuberculosis infection status in US associated with mortality in Ontario beef feedlots. J. Holstein cattle. Anim. Genet .40:655-662. Vet. Diagn. Invest. 18:18-28. Snowder G.D., L.D. Van Vleck, L.V. Cundiff, G.L. Heringstad, B., Y.M. Chang, D. Gianola, O. Steras. Bennett. 2005. Influence of breed, heterozygosity, 2008. Short communication: Genetic analysis of re- and disease prevalence on estimates of variance spiratory disease in Norwegian Red calves. J. Dairy components of respiratory disease in preweaned Sci. 91:367-370. beef calves. J. Anim. Sci. 83:1247. Lyons, D. T., A.E. Freeman, A.L. Kuck. 1991. Genetics Snowder G.D., L.D. Van Vleck, L.V. Cundiff, G.L. of health traits in Holstein cattle. J. Dairy Sci. 74(3): Bennett. 2006a. Bovine respiratory disease in feed- 1092-1100. lot cattle: Environmental, genetic and economic factors. J. Anim. Sci. 84:1999-2008. Lyons, D.T., A.E. Freeman, A.L. Kuck. 1991. Genetics of health traits in Holstein cattle. J. Dairy Sci. 74(3): 1092 1100. McGuirk, S.M. 2008. Disease management of dairy calves and heifers. Vet. Clin. NA: Food Anim. Pract. 24:139-153. 86 Snowder, G.D., L.D. Van Vleck, L.V. Cundiff, G.L. IT IS POSSIBLE TO GENETICALLY CHANGE Bennett, M. Koohmaraie, M.E. Dikeman. 2006b. THE NUTRIENT PROFILE OF BEEF Bovine respiratory disease in feedlot cattle: Pheno- Raluca Mateescu1 typic, environmental, and genetic correlations with 1 growth, carcass, and longissimus muscle palatability University of Florida traits. J. Anim. Sci. 85:1886-1892. USDA. 2001. Treatment of respiratory disease in U.S. Introduction Feedlots. USDA-APHIS-VS, CEAH. Fort Collins, For the last 25 years health professionals have en- CO #N347-1001. couraged people to reduce their intake of red meat USDA, 2011. Feedlot 2011 Part IV: Health and health as a means of reducing saturated fat intake with the management on U.S. feedlots with a capacity of goal of decreasing serum cholesterol level and, hence, 1,000 or more head. USDA-APHIS-VS-CEH- the risk of atherosclerosis and cardiovascular disease NAHMS. Fort Collins, CO #638.0913 (CVD) (Mensink, 2011). This recommendation is Zanella, R., E.G. Casas, G.D. Snowder, H.L. Neibergs. based on the perception that red meat is the major 2011. Fine mapping of loci on BTA2 and BTA26 contributor to both total fat and saturated fat in the associated with bovine viral diarrhea persistent Western diet and that animal fat is a high risk factor infection and linked with bovine respiratory disease for these diseases. Although this perception was sel- in cattle. Front. Livest.Genom. 2:82. dom questioned, it is recently coming under increasing scrutiny and recent studies show that reducing intake of meat may not reduce the risk of CVD (Mc- Neill and Van Elswyk, 2012). In this context, reducing the intake of red meat would only result in reducing the intake of a food with the highest nutritional value per unit of energy (nutritional density) as well as many bioactive components with important health promoting properties. Modern consumers are increasingly aware of the relationship between diet and health, and this awareness is responsible for the trend toward consumption of food perceived to be safe, nutritious and promoting good health and wellbeing. Meat provides valuable amounts of high quality protein containing several essential amino acids, fatty acids, vitamins (E and B complex, being major sources of B12) and minerals (USDA/HHS Dietary guidelines Americans, 2010). Equally important, meat is also a source of many bioactive components with health promoting properties such as conjugated linoleic acid, minerals of high bioavailability such as iron, zinc and selenium, peptides (carnitine, creatine, creatinine, carnosine and anserine), choline, etc. Therefore, the beef industry is in a good position to respond to the demands of health-conscious consumers. To capitalize on this trend, the industry needs to focus its research and promotion efforts toward nutritional and health benefits of meat consumption. A strategy designed to ensure that meat plays the role it deserves as a major component of a healthy diet 87 should include research designed to document the at -20°C. Steaks were cooked and subjected to WBSF relationship between meat consumption and specific and sensory analysis at Oklahoma State University health benefits, to develop the genetic or manage- Food and Agricultural Products Center. Nutrient ment tools needed to increase the components with and bioactive compounds composition analysis was positive and reduce those with negative health conse- conducted at Iowa State University. quences, and to develop consumer education programs to promote nutritional and health benefits of Mineral and Peptide Content of Angus Beef meat consumption. The industry should also emulate Dietary minerals are essential components of hu- the fruit and vegetable blueprint in pursuing scientific man diets, and most dietitians recommend that evidence on positive aspects of meat consumption on these minerals be supplied from foods in which they human health. occur naturally. Meat provides valuable amounts of Why is the Nutrient Profile of Beef Important? important minerals but limited information is available regarding their content and natural variation in While the prevalence of obesity is rapidly increasing beef, the extent to which that variation is the result (Flegal et al., 2012) and has reached a 33.8% high of genetic differences or if it is associated with meat among US adults (Shields et al., 2011), many Ameri- palatability traits. The objectives of our study were to cans are not meeting the recommended daily intake quantify the genetic and environmental components for many nutrients (USDA-ARS, 2011), i.e., they of observed variation in the concentrations of several are “overfed and undernourished”. Among all diet minerals and peptides in LM of Angus beef cattle, to components, meat has the unique status of provid- estimate genetic parameters and their associations ing per unit of energy high amount of high quality with a wide portfolio of beef palatability traits. The protein along with many nutritive factors and other concentrations for several minerals and peptides are components important for human health. Given its shown in Table 1. high nutrient density, red meat can, and should, play a critical role in meeting the nutritional needs of the Iron and Zinc consumers. Beef is already an important food group in the diet of many consumers and improvement of Iron deficiency is one of the most common and its healthfulness will be an efficient way to provide widespread nutritional disorder in the world affecting health benefits to a large proportion of the popula- both developing and industrialized nations (WHO, tion, without dramatically changing dietary habits or 2006). In the U.S. and Europe the iron deficiency is affecting food quality, convenience and costs. greater particularly in pregnant women and infants living in lower socioeconomic groups (Agostoni et al., Animals and Sample Collection. 2008). A recent study from Australia (Samman, 2007) A total of 2,285 Angus sired bulls (n = 540), steers indicates that ~30% of young women had mean daily (n = 1,311), and heifers (n = 434) were used in this iron intakes of less than 70% of the recommended study. All cattle were finished on concentrate diets in daily intake and among young female athletes was Iowa (n = 1,085), California (n = 360), Colorado (n even higher at 51%. The intake of iron was inversely = 388), or Texas (n = 452). Animals were harvested correlated with the amount of red meat consumed on at commercial facilities when they reached typical the day of the survey (Baghurst, 1999). The picture is US market endpoints with an average age of 457 ± 46 similar in the US with iron deficiency anemia being days. Production characteristics including detailed identified by the Centers for Disease Control & Pre- sample collection and preparation of these cattle were vention as the most common nutritional deficiency. reported previously (Garmyn et al., 2011). Briefly, The iron concentration in our data set was 14.44 µg/g external fat and connective tissue were removed muscle, representing on average 1.44 mg iron per 3.5 from 1.27-cm steaks for nutrient and other bioactive oz serving of beef. The current recommended daily compounds composition and 2.54-cm steaks were allowance varies depending on gender and age from removed for Warner-Bratzler shear force (WBSF) and 8 to 18 mg per day. In this context, a 3.5 oz serving of sensory analysis. All steaks were vacuum packaged, beef would provide between 8 and 18% of the recom- aged for 14 d from the harvest date at 2°C and frozen 88 Table 1. Simple statistics for calcium, copper, iron, magnesium, manganese, phosphorus, potassium, sodium and zinc concentrations (µg/g muscle), and carnitine, creatine, creatinine, carnosine and anserine concentrations (mg/g muscle) of steaks from Angus cattle. Trait N1 Mean SD2 CV3 Calcium 2,260 38.71 19.79 .51 Copper 1,980 0.78 0.85 1.09 Iron 2,259 14.44 3.03 0.21 Magnesium 2,274 254.54 43.06 0.17 Manganese 2,000 0.07 0.04 0.57 Phosphorus 2,271 1,968.02 278.36 0.14 Potassium 2,225 3,433.54 494.27 0.14 Sodium 2,273 489.44 92.92 0.19 Zinc 2,261 38.96 7.90 0.20 Carnitine 2,248 3.16 0.94 29.75 Creatine 1,835 5.26 0.53 10.08 Creatinine 2,161 0.21 0.11 52.38 Carnosine 2,140 3.72 0.46 12.37 Anserine 2,139 0.67 0.13 19.40

1Number of cattle 2Standard deviation 3Coefficient of variation mended daily allowance. The amount of iron absorbed to increase iron content. compared with the amount ingested is typically low, and the source of iron is an important factor deter- Zinc is an essential nutrient involved in a number of mining the efficiency of absorption (Kapsokefalou and metabolic processes, including protein and nucleic acid Miller, 1993; Andrews, 2005; West and Oates, 2008; synthesis, insulin and other enzymes, growth and im- Han, 2011). Of particular importance are the results munity, therefore, critically important for good health. reported by Etcheverry and co-workers (2006) which The World Health Organization (WHO) considers indicate that, in adolescents, non-heme iron and zinc zinc deficiency to be a major contributor to the burden absorption from a beef meal is significantly greater of disease in developing countries, especially in those than that from a meal providing soy protein. with a high mortality rate. Based on WHO estimates, it appears that 25% of the populations of South and The obvious and probably most effective dietary South-East Asia and Latin America are at risk of inad- strategy to improve iron status in population groups equate zinc intake, compared with 10% of the popula- exhibiting iron deficiency (especially infants, growing tion of Western Europe and North America. Similar to children and young women) is to increase intake of iron, zinc in animal products is more readily absorbed absorbable iron by increasing consumption of meat than from plant foods. Beef is the major source of zinc and the concentration of iron in meat. Both strategies in the diet. represent opportunities for the beef industry by developing programs focusing on the benefits of meat con- Relatively high heritability for iron (Table 2) and sumption targeting segment of the population at risk moderate heritability for zinc along with their positive of iron deficiency and implementing genetic programs genetic correlation (0.49) indicate that a selection pro

89 Table 2. Genetic (σ2a) and residual (σ2e) variance and heritability (h2) estimates with SE for calcium, copper, iron, magnesium, manganese, phosphorus, potassium, sodium and zinc concentrations (µg/g muscle) and carnitine, creatine, creatinine, carnosine and anserine concentrations in LM from Angus cattle obtained by single-trait REML analysis. Trait 1 σ2a σ2e h2 ± SE Calcium 0.00003 277.74 0.000 ± 0.03 Copper 0.00025 0.49 0.000 ± 0.04 Iron 3.69 3.09 0.544 ± 0.09 Magnesium 36.78 530.83 0.065 ± 0.04 Manganese 0.00006 0.007 0.009 ± 0.03 Phosphorus 1105.10 29630.5 0.036 ± 0.03 Potassium 3989.63 104989.0 0.037 ± 0.03 Sodium 591.32 2574.71 0.187 ± 0.06 Zinc 4.73 47.10 0.091 ± 0.04 Carnitine 0.0055 0.350 0.015 ± 0.03

gram with emphasis on increasing the beef content Seven regions on six chromosomes (1, 2, 7, 10, 15 for these two minerals is feasible and permanent and and 28) were identified to have major effect on iron cumulative genetic improvement should be success- content of LM in Angus cattle. Many of these chro- ful. The associations of iron and zinc concentrations mosomal regions contain, or are in close proximity with several palatability traits (tenderness, juiciness to, genes associated with iron homeostasis or iron and flavor) were all low indicating that increasing metabolism, providing strong candidate genes for the iron and zinc content, no negative consequences further investigation as well as confirming the va- on palatability traits are expected (Mateescu et al., lidity of the genome-wide association results. The 2013a). proportion of phenotypic variance of iron concentration in muscle explained by SNP genotypes (genomic Genome-wide Association Study for Iron Concen- heritability) was 0.37 and the accuracy of genomic tration breeding value (GEBV) was 0.59. This level of accu- Given the difficulty of collecting records for these racy indicates that selection based on genomic merit traits in selection candidates, implementation of a for iron concentration would be as efficient as selec- selection program would require identification of tion based on individual phenotype for a trait with genetic markers associated with iron and zinc content heritability of 0.35. We estimated that in a selection to be used in marker-assisted selection programs. program to improve iron concentration based on Toward this end, a genome-wide association study GEBV, and for each unit (µg/g of meat) improvement using the Bovine SNP50 Infinium II BeadChip was in iron GEBV, 0.73 units (µg/g of meat) improvement conducted to identify chromosomal regions asso- in the actual iron concentration is expected. ciated with concentrations of iron in LM of Angus To assure long-term sustainability of the industry, a beef cattle, to estimate genomic breeding values for beef cattle improvement program should consider iron concentration and assess their accuracy, and to traits that influence production efficiency, traits that determine how other economically important traits influence quality of the eating experience, traits that might be affected by genomic selection to improve influence animal health and well-being, and traits iron concentration (Mateescu et al., 2013b). that would provide health benefits to humans con-

90 suming the product. Increasing the concentration of synthetic response occurring at rest and following iron and zinc in beef muscle through selection should resistance exercise in middle-aged men following the benefit the beef cattle industry as well as consumers ingestion of 4 oz (113 g) of beef protein or an equiva- by producing meat that is healthier for humans to eat lent amount of soy-based protein marketed and sold and, therefore, encouraging consumption. In addi- as a bona-fide replacement for beef (Phillips, 2012). tion, increasing iron concentration in muscle would The results show that meat, with its quality protein contribute to improved functionality of beef (defined and bioactive compounds, is better than plant protein as retention of red color at days 3-4 of retail display) at promotion of myofibrillar (the contractile protein and improved beef flavor. Vitamin E and iron content of skeletal muscle) protein synthesis at rest and also in muscle are the most important factors determin- following resistance exercise. Based on these results ing the functionality of meat, with redness being men over 50 should include lean beef in their diet to positively related to both vitamin E and heme iron prevent or delay the onset of muscle loss. The inclu- content in lamb meat (Ponnampalam et al., 2012). sion of a serving of beef would also provide substan- Increasing iron content in muscle is expected to also tial amounts of iron, zinc, vitamin B-12 as well as improve color stability (shelf life) of beef at retail dis- carnatine, creatine, creatinine, carnosine, anserine play. A significant genetic and phenotypic correlation and other nutrients and bioactive compounds that was reported recently (Garmyn et al., 2011; Mateescu are missing or present in small amounts and with low et al., 2013a) between beef flavor and iron concen- bioavailability in plant-based proteins. tration, indicating an increase in iron concentration would contribute toward an improved beef flavor. Our study found creatine, carnosine and anserine to be moderately heritable (Table 2) whereas almost Other Nutrients no genetic variation was observed in carnitine and creatinine. The additive genetic variation for some of Meat also contains many other compounds with these traits is large enough to be exploited in selec- human health importance. Among these compo- tion or management if changing the concentration nents, some of which are not generally recognized as of these compounds is contemplated but at this point nutrients, we evaluated carnitine, creatine, creatinine, in time such a program may not be necessary. The carnosine and anserine. There is growing evidence re- natural concentrations of these components in beef garding the positive effect these meat bioactive com- seem adequate to enhance the protein quality and pounds play in human health and wellbeing. They prevent the onset of sarcopenia in the segment of the are powerful antioxidants and play important roles in population at risk. Few associations between these muscle metabolism and other metabolic functions. compounds and WBSF or meat quality assessed by The focus in this paper will be on the role meat, with sensory panels were detected, and these associations its quality protein and bioactive compounds, can play were favorable, suggesting that palatability would in preventing and reversing muscle wasting diseases not be compromised if the nutritional profile of beef such as sarcopenia. would be improved by altering the concentration of Sarcopenia refers to the gradual loss of muscle mass these compounds. and strength at a rate of about 1% per year that Conclusions accompany the aging process. Sarcopenia leads to reduced mobility and weakness, increased risk of Following recent trends, consumers are likely to diabetes and weight gain, poor quality of life and continue to pay increased attention to the effect of morbidity. The underlying mechanism is unknown diet on health. Red meat is a very nutritious food and but age-associated changes in diet and exercises are contains numerous compounds with positive health primary suspects. The ‘National Health & Nutrition effects but, unfortunately, the average consumer is not Examination Survey’ estimated that about 25% of familiar with these benefits. An online poll conducted adults over age 50 have low levels of B12 vitamin, for American Meat Institute by Harris Poll revealed strongly suggesting inadequate amount of animal that most consumers do not fully recognize the products in their diet. What is known is that physi- unique nutritional benefits that beef has to offer. For cal exercise, particularly weightlifting, and adequate example, only 12% of consumers correctly identi-

91 fied animal products like beef and poultry as the only Garmyn, A. J. et al. 2011. Estimation of relationships natural source of vitamin B12. In the same poll, 20% between mineral concentration and fatty acid com- of the consumers said cruciferous vegetables such as position of longissimus muscle and beef palatability broccoli and cauliflower while 13% said citrus fruit traits. J. Anim. Sci. 89:2849-2858. were the natural source of vitamin B12, when in fact Han, O. 2011. Molecular mechanism of intestinal neither of these types of foods contains vitamin B12. iron absorption. Metallomics: Integrated Biometal The beef industry needs to increase its efforts to Sci. 3:103-109. document and promote the nutritional and health Kapsokefalou, M., and D. D. Miller. 1993. Lean beef benefits of beef in order to capitalize on the consumer and beef fat interact to enhance nonheme iron ab- trend. The most convincing way to demonstrate the sorption in rats. J. Nutr. 123:1429-1434. positive effects of meat consumption on health is via Mateescu, R. G. et al. 2013a. Genetic parameters for well designed human intervention studies and the concentrations of minerals in longissimus muscle beef industry should take a proactive role and in- and their associations with palatability traits in crease its efforts to promote/support studies address- Angus cattle. J. Anim. Sci. 91:1067-1075. ing the most prevalent chronic diseases, in which dietary intervention using red meat would reduce Mateescu, R. G. et al. 2013b. Genome-wide asso- risks or improve quality of life. ciation study of concentrations of iron and other minerals in longissimus muscle of Angus cattle. J. In this paper two important human health issues, Anim. Sci. 91:3593-3600. iron deficiency and sarcopenia, were discussed. In McNeill, S., and M. E. Van Elswyk. 2012. Red meat in both cases, increasing consumption of red meat to global nutrition. Meat Sci. 92:166-173. meet the recommended daily intake would mitigate the health problem given that the segments of the Mensink, R. P. 2011. Dietary Fatty acids and car- population affected (young women and elderly peo- diovascular health - an ongoing controversy. Ann. ple), have a relatively low red meat consumption. This Nutri. Metabol. 58:66-67. represents a golden opportunity to improve human Phillips, S. M. 2012. Nutrient-rich meat proteins in health and increase red meat consumption. offsetting age-related muscle loss. Meat Sci.92:174- 178. Literature Cited Ponnampalam, E. N., K. L. Butler, M. B. McDonagh, Andrews, N. C. 2005. Understanding heme transport. J. L. Jacobs, and D. L. Hopkins. 2012. Relationship N. England J. Med. 353:2508-2509. between muscle antioxidant status, forms of iron, Baghurst, K. 1999. Red meat consumption in Aus- polyunsaturated fatty acids and functionality (retail tralia: intakes, contributions to nutrient intake and colour) of meat in lambs. Meat Sci. 90:297-303. associated dietary patterns. Euro. J. Cancer Preven- Samman, S. 2007. Red Meat and Iron. Nutri. Dietetics tion 8:185-191. 64:126. Etcheverry, P., K. M. Hawthorne, L. K. Liang, S. A. West, A. R., and P. S. Oates. 2008. Mechanisms of Abrams, and I. J. Griffin. 2006. Effect of beef and heme iron absorption: current questions and con- soy proteins on the absorption of non-heme iron troversies. World J. .Gastroenterol. 14:4101-4110. and inorganic zinc in children. J. Amer. Coll. Nutr. 25:34-40. Flegal, K. M., M. D. Carroll, B. K. Kit, and C. L. Ogden. 2012. Prevalence of Obesity and Trends in the Distribution of Body Mass Index Among US Adults, 1999-2010. J. Amer. Med. Assoc. 307:491- 497.

92 CHANGES IN DIETARY REGIME IMPACT longissimus muscle (LM) are: oleic (C18:1; 40%) acid, FATTY ACID PROFILE OF BEEF palmitic (C16:0; 27%) acid, and stearic (C18:0; 15%) Susan K. Duckett1 acid (Fig. 1, Duckett et al., 1993). Consumption of beef does provide hypercholesterolemic fatty acids in 1 Clemson University the diet, namely palmitic acid, and therefore efforts to reduce its amount would be perceived as beneficial Introduction for human health. Heart disease remains the leading cause of death in Trans fatty acids are receiving attention lately and are the US (CDC, 2011). In addition, over one-third of even being banned from the menu in some U.S. cities. the US population is considered obese (CDC, 2012). Trans fatty acids are produced during the hydrogenation of unsaturated vegetable oils (40-60% of total

Fig. 1 Fatty acid composition of beef. fatty acids as trans) and are found in margarines or 45 processed products that list partially hydrogenated 40 vegetable oil in the ingredient list. This process of 35 30 hydrogenation increases shelf life of the oil by reduc- 25 ing polyunsaturated fatty acid levels. In this process, 20 many short chain trans fatty acids are produced 15 10 (trans bonds in 6-16 position) and consumption of % of total fatty acids 5 these artificial trans fatty acids increases bad (LDL) 0 C16:0 C18:0 C18:1 cholesterol and decreases good (HDL) cholesterol. Fatty acid (no. of carbons:no of double bonds) Results from the Nurses’ Health Study found that women who consumed 4 teaspoons of margarine containing artificial trans fat had a 50% greater risk Obesity is a worldwide epidemic that increases the of heart disease than women who ate margarine risk for developing insulin resistance and sever- only rarely (Willet et al., 1993). Mensink and Katan al chronic diseases such as diabetes, heart disease, (1990) compared the effects of a trans or saturated stroke and non-alcoholic fatty liver disease. A dietary fatty acid rich diet in humans and demonstrated that factor that contributes both to heart disease and obe- trans fats have a more negative effect on serum cho- sity is dietary fat consumption. Of particular interest lesterol levels than saturated fats. Clifton et al. (2004) are the intake of saturated fatty acids (SFA), trans fat- reported high correlations (r = 0.66) between dietary ty acids (TFA), and total fat in the human diet. Con- trans fat intake from margarine and level of trans cerns about dietary saturated fat content are related to fat in adipose tissue, and that the level of trans fat in consumption of diets high in specific SFA raise serum adipose tissue was associated with increased risk of low-density lipoprotein (LDL) or bad cholesterol con- coronary artery disease. centrations. The hypercholesterolemic or cholesterol-elevating SFA are: palmitic (C16:0) acid, myrisitic One strategy to limit SFA intake is to replace these (C14:0) acid, and lauric (C12:0) acid (Mattson and fatty acids with dietary unsaturated fatty acids on an Grundy, 1985; Grundy, 1986; Bonanome and Grundy, isocaloric basis. Certain unsaturated fatty acids are 1988; Mensink and Katan, 1989 & 1990; Denke and considered to be hypocholesterolemic or LDL-choles- Grundy, 1992; Zock et al., 1994). In contrast, stearic terol lowering. These fatty acids include: monounsat- (C18:0) acid, another SFA, does not raise serum cho- urated fatty acids (MUFA) and polyunsaturated fatty lesterol and is considered to be neutral (Bonanome acids (PUFA). Mattson and Grundy (1985) showed and Grundy, 1988; Keys et al., 1965; Hegsted et al., that MUFA were as effective as PUFA in lowering 1965; Grande et al., 1970; Kris-Etherton et al., 1993). LDL-cholesterol. Mensink and Katan (1987) com- Estimates are that these three hypercholesterolemic pared high fat diets containing MUFA versus low fatty acids make up about two-thirds of the saturated fat diets. They found that high fat diets containing fatty acids in the American diet and that dietary in- MUFA were as effective as low-fat diets in lowering take should be reduced to less than 7% of total energy LDL-cholesterol. Consumption of diets rich in mono- (Grundy, 1997). The predominant fatty acids in beef unsaturated fatty acids increases good, high-density lipoproteins (HDL) and lowers bad, LDL-cholester 93 ol levels (Mensink and Katan, 1989; Wardlaw and per day (Ip et al., 1994). Snook, 1990). Canola and olive oils contain predominately MUFA at levels of 58% and 72% of total Dietary intake of specific fatty acids is important to fatty acids, respectively. Grundy (1997) recommend human health. Health professionals recommend con- intakes of oleic acid, the predominant MUFA, at 16% suming a diet low in saturated and trans fatty acids. of total energy. Limiting intake of animal fats is also typically recommended as a way to reduce total fat and saturated fat Polyunsaturated fatty acids (PUFA) are subdivided intake (AHA, 2014). Diet composition provided to into two categories, omega-6 and omega-3, based on the finishing animal can alter fatty acid composition location of the double bonds in the fatty acid chain. of the LM to enhance healthfulness of beef products. Omega-6 fatty acids are common in grains and This paper will review current research examining vegetable oils. Omega-3 fatty acids are common in how finishing system alters fatty acid composition. plant lipids and fish oils. Diets containing omega-6 or omega-3 fatty acids lower blood total and LDL-cho- Grain vs. Grass Finishing Systems lesterol; however, omega-6 PUFA also tend to lower Fatty acid composition as a percentage of total fatty HDL-cholesterol (Mensink and Katan, 1989). Con- acids of beef muscle from concentrate-finished versus sumption of diets high in omega-3 fatty acids is asso- pasture-finished beef is shown in Table 1. The results ciated with reduced risk of heart disease, stroke and are from two experiments that evaluated finishing cancer (Kris-Etherton et al., 2002). Currently, Amer- (final 150 d prior to slaughter) of Angus-cross steers icans consume greater amounts of omega-6 PUFA (n = 326) on a high concentrate diet (82% concen- than omega-3 PUFA, which has dramatically altered trate:18% corn silage) versus pasture (mixed pastures the omega-6 to omega-3 ratio in the human diet. consisting of bluegrass, orchardgrass, endophyte-free Health professionals recommend that we consume tall fescue and white clover; Duckett et al., 2009 & a diet with a more balanced ratio (< 4:1) of omega-6 2013). Steers were fed to an equal animal age end- to omega-3 PUFA. The World Health Organization point in order to minimize confounding of treat- recommends a daily intake of 1.1 g/d of omega-3 fatty ments by animal age or environmental effects. The acids with approximately 0.8 g/d of linolenic acid and percentage of omega-6 PUFA did not differ between 0.3 g/d of EPA and DHA. McAfee et al. (2011) report- concentrate- and pasture-finished beef. Monounsat- ed that consumption of grass-fed red meat products urated fatty acid percentage was greater for concen- increases plasma and platelet n-3 PUFA status, which trate than grass-finished. Omega-3 PUFA percentage indicates that lower n-6:n-3 ratios typically observed was greater for grass- than concentrate-finished. This in forage-finished beef can potentially impact human resulted in a lower, more desirable for human health, health. ratio of omega-6 to omega-3 fatty acids in grass-fin- The predominant fatty acids (70% or greater) in for- ished beef (1.54) compared to concentrate-finished ages are PUFA; however, the predominant fatty acids beef (5.01). The percentage of CLA, cis-9 trans-11 in beef LM are MUFA and SFA due to the extensive isomer, was greater for grass- than concentrate-fin- biohydrogenation of PUFA to SFA by ruminal mi- ished. Trans-11 vaccenic acid percentage was also crobes (Duckett et al., 2002; Sackmann et al., 2003), greater for grass- than concentrate-finished beef. and conversion of SFA to MUFA via adipose tissue Fatty acid composition can be presented in two desaturases (Duckett et al., 2009). Intermediates of ways, 1) as a percentage of total fatty acids or 2) as ruminal biohydrogentation, trans-11 vaccenic acid the total amount per specific steak weight, estimat- (TVA) and conjugated linoleic acid (CLA), can be ed on a cooked basis (assume 25% cooking shrink). found in beef LM. However, the majority of CLA, cis- Table 1 showed the percentage of each fatty acid as 9 trans-11 isomer, in beef comes from desaturation part of the total fatty acid amounts. Table 2 shows of TVA to CLA in adipose tissues (Pavan and Duck- the gravimetric content of total fatty acid types as ett, 2007). Conjugated linoleic acid, cis-9 trans-11 amount per 18.7 oz. cooked serving. The American isomer, has been shown to possess anticarcinogenic Heart Association recommends a 3-oz serving size properties (Ha et al., 1987) that could be beneficial to of beef; however, most retail beef products available human health. Recommendations are that we should exceed this amount. Therefore, the estimates for fatty consume about 300 mg of CLA, cis-9 trans-11 isomer, 94 Table 1. Fatty acid percentage of the longissimus muscle from steers finished on a high-concentrate diet or mixed pasture. CONCEN- PASTURE TRATE n 135 191 Total lipid content (TL), % 5.39a 2.48b Total fatty acid content (TFA), % 4.56a 2.25b Saturated fatty acids (SFA), % 43.23b 44.48a C14:0, Myristic acid, % 2.76a 2.46b C16:0, Palmitoleic acid, % 26.62a 24.92b C18:0, Stearic acid, % 13.83b 17.09a Monounsaturated fatty acids (MUFA), % 42.78a 35.13b C14:1, Myristoleic acid, % 0.61a 0.41b C16:1, Palmitoleic acid, % 3.49a 2.71b C18:1, Oleic acid, % 39.68a 32.02b Polyunsaturated fatty acids (PUFA), n-6, % 3.59 3.61 C18:2, Linoleic acid, % 2.89a 2.67b C20:4, Arachidonic acid, % 0.70b 0.94a Polyunsaturated fatty acids (PUFA), n-3, % 0.78b 2.47a C18:3, Linolenic acid, % 0.36b 1.13a C20:5, EPA, % 0.12b 0.50a C22:5, DPA, % 0.26b 0.76a C22:6, DHA, % 0.04b 0.08a Ratio of n-6 to n-3 fatty acids 5.01a 1.54b Trans-11 vaccenic acid (TVA), % 0.53b 3.37a Conjugated linoleic acid (CLA), cis-9 trans-11, % 0.35b 0.71a abMeans in the same row with uncommon superscripts differ (P < 0.05). acid content per ‘real-world’ serving is for 18.7-oz Forage Species for Finishing assuming that a 6.7-oz cooked hamburger (McDon- Angus-cross steers (n = 60) from the Clemson alds Big Mac) and 12-oz ribeye steak (Longhorn’s University beef herd were used in this 2-yr grazing Outlaw ribeye steak, 18 oz bone-in) were consumed study (Schmidt et al., 2013). Each winter, 30 steers per day. Based on these assumptions, beef from grazed cereal rye/ryegrass and tall fescue pastures steers fed high- concentrate diet would provide prior to being blocked by BW and assigned randomly about 47% more total fat and saturated fat content to 1 of 5 forage-finishing treatments of alfalfa Med( - than beef from steers finished on pasture. Intake of icago sativa L.), bermudagrass (Cynodon dactylon), MUFA would be 58% higher for beef from steers chicory (Cichorium intybus L.), cowpea (Vigna un- fed a high-concentrate diet than pasture. Intake of guiculata L.), and pearl millet (Pennisetum glaucum n-6 PUFA would be 48% greater and intake of n-3 (L. R Br.). Finishing forage treatments started when PUFA would be 64% lower for beef from steers fed forage growth for each individual forage species was high-concentrates versus pasture. The actual amount adequate for grazing. Steers were slaughtered when of CLA provided from both beef sources would be there was either insufficient forage mass for continued similar and meet the recommended daily consump- steer gain or when steer live weight exceeded 568 kg. tion levels. However, CLA can also be produced in The steak from the 12th rib was trimmed of all exter

95 Table 2. Fatty acid amount per ‘real-world’ serving (18.7 oz; lunch = 6.7-oz hamburger cooked; dinner = 12-oz ribeye steak broiled) for beef finished on high-concentrate diet or mixed pasture. CONCEN- PASTURE TRATE Total fat content, g 99 51 Saturated fatty acids, g 42.8 22.7 Monounsaturated fatty acids, g 42.4 17.9 Polyunsaturated fatty acids, omega-6, g 3.55 1.84 Polyunsaturated fatty acids, omega-3, g 0.77 1.26 Trans-11 vaccenic acid, g 0.52 1.72 Conjugated linoleic acid, cis-9 trans-11, g 0.35 0.36

nal fat and epimysial connective tissue for PM) compared to other forage species. The highest subsequent fatty acid analysis. ratios of n-6 to n-3 fatty acids were produced in LM when steers grazed CH and PM. Forage species utilized for finishing did not alter total lipid, fatty acid, saturated, monounsaturat- Corn Grain Supplementation on Grass or Legume ed or polyunsaturated fatty acid content of the LM. Finishing However, individual concentrations of certain fatty acids were altered. Most notably, trans-11 vaccenic Thirty-two Angus x Hereford steers were used (C18:1 trans-11; TVA) acid concentration in the LM (BW = 461 ± 17.4 kg) to evaluate the effects of forage was greater for BG than CH, CO and AL. Conjugated type (legumes [LG, alfalfa and soybeans] vs. grasses linoleic acid (CLA), cis-9 trans-11 isomer, concentra- [GR, non-toxic tall fescue and sudangrass]) with or tion was greatest (P < 0.05) for BG and PM than AL, without daily corn supplementation (none [NONE] CH, and CO. Since the grasses (BG and PM) in this vs. 0.75 % BW/d of corn grain [CORN]) on animal study had higher NDF content than did the legumes performance and beef quality in a 2-yr study (Wright (AL, CO) or forbs (CH), this likely resulted in higher et al., 2014). The finishing period was 98 d in yr 1 and outflow of TVA at the duodenum, which corresponded 105 d in yr 2. Fatty acid composition as a percent of to greater tissue deposition of TVA and CLA in these the total fatty acids in the LM is presented in Table forage treatments. 4. All interactions between forage species and corn supplementations were non-significant P( > 0.05). Steers grazing CH, the forage species with a Finishing on grasses increased stearic acid (C18:0) greater linolenic acid percentage, produced LM with and trans-11 vaccenic acid concentrations compared the greater linolenic acid concentrations compared to legumes. Grazing legumes increased linolenic acid to AL, BG, and PM. In contrast, linolenic acid levels and total n-3 fatty acid concentrations in the LM com- in PM forage were similar to CH but linolenic acid pared to grazing grasses. The concentration of other levels in the LM of PM steers were less than CH and fatty acids in the LM was not altered by forage type. CO. For CO, forage linolenic levels were less than CH and PM but LM linolenic acid concentrations Corn supplementation increased myristic were greater for CO than AL, BG, or PM. Due to the (C14:0) and palmitic (C16:0) acid concentrations process of biohydrogentation in the rumen and desat- but did not alter total saturated fatty acid percentage. uration in the adipose tissues, differences in forage Oleic (C18:1 cis-9) and palmitoleic (C16:1 cis-9) acid fatty acid levels are not directly translated to similar concentrations tended to be increased with corn grain changes in LM fatty acid composition. The n-6 to supplementation. As a result, the total monounsatu- n-3 ratio was greater for CH and PM than AL, BG rated fatty acid (MUFA) percentage in the LM was and CO. For AL, BG, and CO. Anticarcinogenic fatty greater with corn supplementation. Linolenic (C18:3) acids, TVA and CLA cis-9 trans-11 isomer, concentra- acid concentration was reduced with corn grain sup- tions were greater in beef finished on grasses (BG and plementation; however, other individual and

96 Table 3. Fatty acid percentage of the longissimus muscle from steers finished on a five different forage species. Forage Speciesa AL BG CH CO PM n TFA, % 2.35 2.82 2.18 2.38 2.16 SFA, % 43.59 43.12 43.42 44.46 41.54 C14:0, % 2.77 2.39 2.65 2.42 2.32 C16:0, % 26.63 25.42 25.84 26.19 24.54 C18:0, % 14.16d 15.31bc 14.92bcd 15.54b 14.68cd MUFA, % 39.67 39.20 37.46 38.08 39.50 C14:1, % 0.65 0.51 0.58 0.46 0.54 C16:1, % 3.28 3.11 3.07 3.10 3.36 C18:1, % 35.74 35.58 33.81 34.53 35.60 PUFA, n-6, % 4.14 3.60 5.37 4.22 4.41 C18:2, % 2.93c 2.60c 4.12b 3.13c 3.09c C20:4, % 1.22 1.00 1.24 1.09 1.33 PUFA, n-3, % 2.19 1.91 2.52 2.38 1.96 C18:3, % 1.03c 0.90c 1.46b 1.32b 0.86c C20:5, % 0.73 0.63 0.66 0.65 0.68 C22:5, DPA, % 0.73 0.63 0.66 0.65 0.68 C22:6, DHA, % 0.06 0.06 0.05 0.05 0.06 Ratio of n-6:n-3 1.88c 1.90c 2.11b 1.80c 2.26b TVA, % 2.01d 3.03b 2.35cd 2.40cd 2.84bc CLA, cis-9 trans-11, % 0.38c 0.52b 0.40c 0.40c 0.55b aForage species: AL = alfalfa, BG = bermudagrass, CH = chicory, CO = cowpea, and PM = pearl millet. bcdMeans in the same row with uncommon superscripts differ (P < 0.05).

Table 4. Fatty acid percentage of the longissimus muscle from steers finished legume or grass pastures with or without corn grain supplementation (0.75% BW/d). Forage Corn Gain Typea Supplementationa GR LG 0 0.75% n 16 16 16 16 TFA, % 3.42c 5.02b 4.04c 4.40b SFA, % 44.89 44.18 44.16 44.91 MUFA, % 39.69 40.53 39.34c 40.89b PUFA, n-6, % 3.58 3.87 3.88 3.57 PUFA, n-3, % 0.99c 1.25b 1.20 1.04 Ratio of n-6:n-3 3.68 3.29 3.28 3.69 TVA, % 2.62b 2.07c 2.61 2.08 CLA, cis-9 trans-11, % 0.54 0.50 0.56d 0.49e aForage type = GR: grass (tall fescue + sorghum-sudan); LG: legumes (alfalfa + soybean) bcMeans in the same row with uncommon superscripts differ (P < 0.05). deMeans in the same row with uncommon superscripts differ (P < 0.10). 97 Table 5. Fatty acid percentage of the longissimus muscle from steers fed high concentrate diets or grazed pastures during phase 1 (30-d post weaning for 11 d) or phase 3 (final d to slaughter). Phase 1a HC HC PA PA Phase 3a HC PA HC PA n 10 10 9 10 TFA, % 3.63 3.60 3.41 3.32 SFA, % 44.47 44.73 46.26 45.12 MUFA, % + 44.64 42.59 43.62 42.72 PUFA, n-6, % * 3.20 3.28 2.84 2.48 PUFA, n-3, % *+ 0.98 1.57 1.57 1.82 Ratio of n-6:n-3 # 3.28b 2.18c 1.83d 1.36e TVA, % # 0.84c 1.49b 1.29b 1.43b CLA, cis-9 trans-11, % + 0.35 0.47 0.39 0.46 aPhase 1 (30-d postweaning for 111 d) or Phase 3 (final ~100 d before slaughter): HC = high concentrate diet; PA = pasture. *Denotes Phase 1 effect (P < 0.05). +Denotes Phase 3 effect (P < 0.05). #Denotes interaction between Phase 1 and Phase 3 effect (P < 0.05). bcdeMeans with uncommon superscripts differ (P < 0.05).

total n-3 fatty acid concentration were not altered creased the concentration of n-6 PUFA and decreased by supplementation. Conjugated linoleic acid, cis-9 the concentrations of n-3 PUFA. Exposure to HC in trans-11 isomer, concentration tended to be lower for Phase 3 increased MUFA, and decreased PUFA n-3 corn supplemented than non-supplemented. The ratio and CLA cis-9 trans-11 isomer concentrations. Inter- of n-6 to n-3 fatty acids did not differ between CS and actions between Phase 1 and Phase 3 feeding treat- NS. ments were significant for ration n-6 to n-3 fatty acids and trans-11 vaccenic acid. The ratio of n-6 to n-3 Timing of High Concentrate and Forage Finishing was higher in longissmus muscle of steers that spent more time on a high concentrate diet (HC-PA-HC > Research was conducted to determine the PA-PA-PA). In addition, late exposure to HC resulted timing of exposure to a high concentrate diet on sub- in lower ratios than early exposure to HC (HC-PA-PA sequent beef quality and composition (Volpi Lagreca > PA-PA-HC). Timing of exposure to HC or PA diets et al., 2014). Steers (n = 40) were backgrounded for can alter fatty acid composition of the longissimus 30-d post weaning and then randomly assigned to muscle. However, all ratios of n-6 to n-3 fatty acids, high concentrate diet (HC) or pasture (PA) in Phase regardless of the length of time exposed to HC, were 1 (111 d). After the completion of Phase 1, all steers below the recommended 4:1 level for human health. grazed high-quality pastures for 98 d (Phase 2). At the end of Phase 2, steers were randomly divided based Summary on Phase 1 treatments into two treatments of HC or PA for Phase 3. Phase 3 started when steers were Animal nutrition can alter LM fatty acid com- about 454 kg BW and finished when steers reached position. Finishing on high concentrate diets increases 568 kg BW (live weight endpoint). total fatty acid content and MUFA concentrations. The enzyme, stearoyl-CoA desaturase (SCD-1), is Total fatty acid content did not differ due to responsible for the conversion of SFA to MUFA, Phase 1 or Phase 3 treatments even though marbling and is very responsive to energy content of the diet. scores did differ with Phase 1 and Phase 3 treatments. Research comparing gene expression in subcutane- Total SFA percentage was not altered by Phase 1 or ous fat from high-concentrate finished versus pasture Phase 3 treatments. Exposure to HC in Phase 1 in- finished cattle found that SCD-1 was up-regulated by 98 46-fold and MUFA increased concentration by 68% in Duckett, S.K., J. G. Andrae, and F. N. Owens. 2002. high concentrate finished (Duckett et al., 2009). Even Effect of high oil corn or added corn oil or supplementation of corn grain, at a level of 0.75% of added corn oil on ruminal biohydrogenation body weight, can also increase MUFA percentages and conjugated linoleic acid formation in beef in the LM. Finishing on forages typically increases steers fed finishing diets.J. Anim. Sci. 80:3353- n-3 PUFA and lowers total and saturated fat content. 3360. Finishing on different forage species results in minor Duckett, S. K., E. Pavan, and S. L. Pratt. 2009. Corn changes in fatty acid composition of LM. Howev- oil or corn grain supplement to steers grazing er, finishing on grasses will increase TVA and CLA endophyte-free tall fescue. II. Effects on sub- concentrations; whereas, finishing on legumes will cutaneous fatty acid content and lipogenic gene increase n-3 PUFA and total lipid content. Concentra- expression. J. Anim. Sci. 87:1120-1128. tion of n-3 PUFA decreases with the number of days cattle are fed a high-concentrate diet. Forages contain Duckett, S. K., D. G. Wagner, L. D. Yates, H. G. predominately n-3 PUFA (58% C18:3); in contrast to Dolezal, and D. R. Gill. 1993. Effect of time- corn grain which contains predominately n-6 PUFA on-feed on beef nutrient composition. J. Anim. (58% C18:2). Therefore, finishing programs that uti- Sci. 71:2079-2088. lize high concentrate diets for longer time periods will Duckett, S. K., J. P. S. Neel, J. P. Fontenot, and W. have lower n-3 fatty acid concentration. Finishing M. Clapham. 2009. Effects if winter stocker programs that combine periods of pasture and high growth rate and finishing system on: III. Tissue concentrate finishing will provide higher levels of proximate, fatty acid, vitamin, and cholesterol n-3 fatty acids than just high concentrate alone. With content. J. Anim. Sci. 87:2961-2970. certain finishing strategies, fatty acid composition of beef can be altered to enhance specific fatty acids that Duckett, S. K., J. P. S. Neel, R. M. Lewis, J. P. Fonte- consumers find desirable. not, and W. M. Clapham. 2013. Effects of forage species or concentrate finishing on animal Literature Cited performance, carcass and meat quality. J. Anim. Sci. 91:1454-1467. AHA. 2014. The American Heart Association’s Diet and Lifestyle Recommendations. Avail- Grundy, S. M. 1986. Comparison of monounsaturated able at: ttp://www.heart.org/HEARTORG/ fatty acids and carbohydrates for lowering plas- GettingHealthy/NutritionCenter/Health- ma cholesterol. N. Engl. J. Med. 314:745-748. yEating/The-American-Heart-Associa- Grundy, S. M. 1997. What is the desirable ratio of tions-Diet-and-Lifestyle-Recommendations_ saturated, polyunsaturated and monounsatu- UCM_305855_Article.jsp rated fatty acids in the diet? Am. J. Clin. Nutr. Bonanome, A. and S. M. Grundy. 1988. Effect of 66:988S-990S. dietary stearic acid on plasma cholesterol and Ha, Y. L., N. K. Grimm and M. W. Pariza. 1987. An- lipoprotein levels. N. Engl. J. Med. 318:1244-8. ticarcinogens from fried ground beef: heat-al- CDC. 2011. Leading Cause of Death, available at: tered derivatives of linoleic acid. Carcinogene- http://www.cdc.gov/nchs/fastats/lcod.htm. sis 8(12):1881-1887. CDC. 2012. Adult obesity facts, available at: http:// Hegsted, D. M., R. B. McGandy, M. L. Myers, and F. www.cdc.gov/obesity/data/adult.html. J. Stare. 1965. Quantitative effects of dietary Clifton, P. M., J.B. Keogh, and M. Noakes. 2004. fat on serum cholesterol in man. Am. J. Clin. Trans fatty acids in adipose tissue and the food Nutr. 17:281-95. supply are associated with myocardial infarc- Ip, C., M. Singh, H. J. Thompson, and J. A. Scime- tion. J. Nutr. 134:874-879. ca. 1994. Conjugated linoleic acid suppresses Denke, M. A. and S. M. Grundy. 1992. Comparison mammary carcinogenesis and proliferative ac- of effects of lauric acid and palmitic acid on tivity of the mammary gland in the rat. Cancer Res. 54:1212-1215. plasma lipids and lipoproteins. Am. J. Clin. Nutr. 56:895-8.

99 Keys, A., J. T. Andreson, and F. Grande. 1965. Serum Volpi Lagreca, G., J. G. Andrae, and S. K. Duckett. cholesterol response to changes in the diet. 2014. Timing of exposure to high concentrate IV. Particular saturated fatty acids in the diet. diets versus pasture on beef quality and compo- Metabolism 14:776-87. sition. ICoMST Proc. (Submitted). Kris-Etherton, P. M., J. Derr, and D. C. Mitchell. Wardlaw, G. M. and J. T. Snook. 1990. Effect of diets 1993. The role of fatty acid saturation on high in butter, corn oil or high-oleic sunflow- plasma lipids, lioproteins, and apolipoproteins. er oil on serum lipids and apolipoproteins in I. Effects of whole food diets high in cocoa mean. Am. J. Clin. Nutr. 50:1382-1388. butter, olive oil, soybean oil, dairy butter, and Willet, W. C., M. J. Stampfer, J. E. Manson, G. A. milk chocolate on the plasma lipids of young Colditz, F. E. Speizer, B. A. Rosner, and C. H. men. Metabolism 42:121-9. Hennekens. 1993. Intake of trans fatty acid and Mattson, F. H., and S. M. Grundy. 1985. Comparison risk of coronary heart disease among women. of effects of dietary saturated, monounsaturat- Lancet 341:581-585. ed, and polyunsaturated fatty acids on plasma Wright, A. M., C. Fernandez Rosso, M. C. Miller, E. lipids and lipoproteins in man. J. Lipid Res. Pavan, J. G. Andrae, and S. K. Duckett. 2014. 26:194-202. Effect of forage type with or without corn Mensink, R. P. and M. B. Katan. 1989. Effect of a diet supplementation on beef fatty acid composition enriched with monounsaturated or polyunsat- and palatability. Meat Sci. (Submitted). urated fatty acids on levels of low-density and Zock, P. L., J. H. M. de Vries, and M. B. Katan. 1994. high-density lipoprotein cholesterol in healthy Impact of myristic acid versus palmitic acid on women and men. N. Engl. J. Med. 321:436-41. serum lipid and lipoprotein levels in healthy Mensink, R. P. and M. B. Katan. 1990. Effect of women and men. Arterioscler. Thromb. 14:567- dietary trans fatty acids on high-density and 75. low-density lipoprotein cholesterol levels in healthy subjects. N. Engl. J. Med. 323:439-45. Pavan, E. and S. K. Duckett. 2013. Fatty acid composition and interrelationships among eight retail cuts of grass-fed beef. Meat Sci. 93:371-377. Pavan, E. and S. K. Duckett. 2007. Corn oil supplementation to steers grazing endophyte-free tall fescue. II. Effects on longissmus muscle and subcutaneous adipose fatty acid composition and stearoyl-CoA desaturase activity and expression. J. Anim. Sci. 85:1731-1740. Sackmann, J. R., S. K. Duckett, M. H. Gillis, C. E. Realini, A. H. Parks, and R. B. Eggelston. 2003. Effects of forage and sunflower oil levels on ruminal biohydrogenation of fatty acids and conjugated linoleic acid formation in beef steers fed finishing diets.J. Anim. Sci. 81:3174- 3181. Turpeinen, A. M., M. Mutanen, I. Salimen, S. Basu, D. L. Palmquist, and J. M. Griinari. 2002. Bioconversion of vaccenic acid to conjugated linoleic acid in humans. Am. J. Clin. Nutr. 76:504-10.

100 IMPROVING FEED EFFICIENCY IN rumen and intestinal tract of cattle unless broken due THE FEEDLOT: OPPORTUNITIES AND to mastication. To avoid passage of whole kernels and CHALLENGES thus aid in starch digestion, corn grain is commonly Galen E. Erickson1 processed. The three most common corn processing 1University of Nebraska-Lincoln methods are dry-rolling, ensiled high-moisture, or steam-flaking. How corn is processed (and which grain source is fed) can have dramatic impacts on Introduction Feedlots focus heavily on feed efficiency or feed efficiency. Based on individual studies and feed conversion and evaluate pens of cattle as a tool reviews, in diets with 80 to 85% corn grain inclusion, of how well management, nutrition, weather, and feeding HMC is only 1 to 2% better than DRC. How- cattle purchasing decisions are performing. Feed ever, feeding SFC improves feed efficiency by 12 to efficiency would generally be gain divided by intakes 15% (Cooper et al, 2002; Owens et al., 1997). (G:F), whereas conversions generally refer to intakes Byproducts divided by gains. Conversion would be more typical for discussions with producers. Regardless of which Numerous summaries are available on the is used, intakes should always be on a DM basis. An- impact of feeding distillers grains and corn gluten other concern is that feed efficiency is based on gain feed to beef cattle. We have a summary available which requires measuring initial and final weights. at our http://beef.unl.edu website. Historically, pro- Body weights are important currency to use when ducers have been able to purchase distillers grains at measuring efficiency; however, these weights can 70 to 80% of corn price (DM basis). This price was have errors that impact accuracy. Live cattle weights considerably greater in 2013 and 2014 at 100 to 130% are dramatically impacted by gastrointestinal fill. of corn price. The relatively strong prices on distillers Most yards will use receiving weight or pay weight grains are likely a reflection of strong demand for dry for initial body weight and fill is likely less than once distillers (DDGS) for export and for use in non-rumi- cattle arrive and consume hay and water. Lastly, use nants. Our data suggest that wet distillers grains plus of live final body weights are less meaningful than solubles (WDGS) has 143% the value of corn to the carcass weights as final prices are based on hot car- feedlot producer at 20% inclusion, and approximately cass weights, even when selling live because packers 130% at 40% inclusion (Bremer et al., 2011; Table are evaluating red meat yield and dressing percent 1). When distillers grains are dried partially to make when negotiating price. Therefore, when evaluating modified distillers grains (MDGS), the feeding value management or nutrition in feedlots, the impact on decreases to 117 to 124% of corn (at 20 to 40% inclu- carcass weight is the ultimate outcome. We believe sions). When distillers are completely dried to make targeting gain and efficiency on a carcass basis is the DDGS, the value to the feeder is 112% of corn. The direction for the beef industry. Unfortunately, carcass concept that WDGS results in better feed efficiency weights are still converted back to live weights to than MDGS which is better than DDGS has also been calculate gains and efficiency. documented in individual studies (Nuttelman et al., 2011, 2013) OPPORTUNITIES Nutritional Methods Other byproducts such as wet corn gluten Corn processing feed, distillers solubles or syrup, and Sweet Bran Corn grain has been a staple in feedlot diets all have different impacts on feed efficiency of the due to abundance, low prices, and serving as the cattle. Predicting impact of these byproducts should cheapest source of energy. Corn is the most common be based on performance data as most experiments grain fed in the U.S.; however, other grains can be uti- compare the value to corn it is replacing. lized in a similar manner such as grain sorghum, bar- ley, or wheat. Corn contains about 2/3 or 70% starch which is readily digested once the kernel is broken. Whole kernels are quite resistant to digestion in the

101 Table 1. Meta-analysis of finishing steer performance when fed different dietary inclusions of corn wet distillers grains plus solubles (WDGS), modified distillers grains plus soluble (MDGS) or dried distillers grains plus soluble (DDGS) replacing dry rolled and high moisture corn. (Bremer et al., 2011)

DGS Inclusion a: 0DGS 10DGS 20DGS 30DGS 40DGS Linb Quadb

WDGS DMI, lb/day 23.0 23.3 23.3 23.0 22.4 0.01 < 0.01 ADG, lb 3.53 3.77 3.90 3.93 3.87 < 0.01 < 0.01 F:G 6.47 6.16 5.96 5.83 5.78 < 0.01 < 0.01 Feeding value, % d 150 143 136 130 MDGS DMI, lb/day 23.0 23.8 24.1 24.0 23.4 0.95 < 0.01 ADG, lb 3.53 3.77 3.90 3.92 3.83 < 0.01 < 0.01 F:G 6.47 6.29 6.17 6.10 6.07 < 0.01 0.05 Feeding value, % d 128 124 120 117 DDGS DMI, lb/day 23.0 24.0 24.6 24.9 24.9 < 0.01 0.03 ADG, lb 3.53 3.66 3.78 3.91 4.03 < 0.01 0.50 F:G 6.47 6.39 6.32 6.25 6.18 < 0.01 0.45 Feeding value, % d 112 112 112 112

a Dietary treatment levels (DM basis) of distillers grains plus solubles (DGS), 0DGS = 0% DGS, 10DGS = 10% DGS, 20DGS = 20% DGS, 30DGS = 30% DGS, 40DGS = 40% DGS. b Estimation equation linear and quadratic term t-statistic for variable of interest response to DGS level. d Percent of corn feeding value, calculated from predicted F:G relative to 0WDGS F:G, divided by DGS inclusion.

Distillers grains plus solubles are the most ing does not improve efficiency as much when diets common byproduct used today and some discussion contain distillers grains plus solubles. It is unclear is warranted. Besides whether you are using dry or why this occurs, but is quite repeatable. wet distillers, another major factor affecting how well distillers grains work for finishing cattle is related to Forage Concentration how corn is processed. Unlike historical corn-based Forages fed in feedlot diets are often referred diets with 80 to 85% grain where feeding SFC works to as roughages. Forages are also routinely used for best, diets that contain distillers grains do not respond grain adaptation or the gradual (18 to 28 days) switch similarly (Table 2). Numerous studies have illustrated of cattle diets from a primarily forage-based diet that feeding distillers grains appears to fit better with to primarily a concentrate-based diet. While grain diets that contain DRC or HMC, and not as well with adaptation is very important, especially for teaching SFC (Corrigan et al, 2009; Vander Pol et al., 2008; cattle to eat differently, the focus of this section is on Buttrey et al., 2012). Feeding SFC is better than DRC the amount of forage in the final, high-concentrate with diets containing distillers solubles (Titlow et al., finishing diet. Roughages are bulky ingredients with 2013; Harris et al., 2014) and with Sweet Bran (Scott large shrink losses that feedlots would prefer to avoid. et al., 2003; Macken et al., 2006). The conclusion is In general, as forage concentration is decreased in that steam-flaking corn will normally improve feed feedlot diets, feed efficiency improves. Cattle can be efficiency (grain based diets, diets with distillers solu- fed no roughage in feedlot diets. However, risk of bles, Sweet Bran, or corn gluten feed) but steam-flak- ruminal acidosis increases and results in lower DMI,

102 Table 2. Effect of corn processing in diets containing increasing amounts of wet distillers grains plus solubles (Corrigan et al., 2009)1.

0.0 15.0 27.5 40.0

Dry-rolled corn DMI, lb/d 3 22.3 22.2 21.4 21.3 ADG, lb 2 3.64 3.77 3.87 3.92 G:F 2 0.163 0.170 0.181 0.185 High-moisture corn DMI, lb/d 3 20.1 21.0 20.2 20.0 ADG, lb 3 3.68 3.96 3.97 3.86 G:F 2 0.183 0.189 0.197 0.194 Steam-flaked corn DMI, lb/d 3 20.2 20.2 19.8 18.8 ADG, lb 3 3.67 3.74 3.60 3.44 G:F 0.182 0.186 0.182 0.183

1 For ADG: Effect of corn processing method, P < 0.01; effect of WDGS level, P = 0.01, and effect of corn processing method × WDGS level, P < 0.01. For G:F: Effect of corn processing method, P < 0.01; effect of WDGS level, P < 0.01, and effect of corn processing method × WDGS level, P < 0.01. 2 Linear effect of WDGS level within corn processing method (P < 0.05). 3 Quadratic effect of WDGS level within corn processing method (P < 0.05)

lower ADG, and equal or often times improved effi- studies (Table 3), feeding 20% alkaline treated corn- ciency. Conventional inclusions of roughage would be stalks (treated with 5% calcium oxide) made the cattle approximately 4% neutral detergent fiber (NDF) from 2.3% less efficient (greater F:G) yet is often profitable the roughage source. This equates to about 7 or 8% depending on corn and cornstalks prices. We have alfalfa hay, 5% crop residues like straw or stalks, and also evaluated feeding 15, 30, or 45% corn silage in 10 to 12% corn silage. Exchanging these roughages diets with distillers grains (Table 4; Table 5). Feeding on an equal NDF basis is the logical approach (Gal- 45% silage instead of 15% decreased feed efficiency yean and Defoor, 2003; Benton et al., 2007) to main- by 5 to 5.5% (Burken et al., 2013a; Burken et al., tain DMI and ADG. unpublished) yet increased profits (Burken et al., 2013b). Most times, increased feed efficiency means There are a few examples where increasing increased profits, but not always. forage will negatively impact feed efficiency, yet improve profitability. Two examples are with alkaline treated crop residues and feeding elevated dietary inclusions of corn silage. Across a series of six feedlot

103 Table 3. Summary of F:G across experiments with 20% treated stalks (TRT) compared to a 5% stalks control (CON) or not treating (NONTRT). See literature cited for trial references.

Treatments CON vs TRT CON vs NONTRT CON TRT NONTRT DIFF % DIFF % DIFF Johnson calf 6.36a 6.22a 7.05b -0.14 -2.2% 10.8% Johnson yrlgs 6.42a 6.85b 7.65c 0.43 6.7% 19.2% Shreck 3” 6.54 6.55 7.72 0.01 0.2% 18.0% Peterson 40% 5.79 5.88 - 0.09 1.6% - Cooper 5.53 5.83 - 0.30 5.4% - Average 2.34%

Table 4. Impact of feeding 15, 30, 45, or 55% dietary corn silage in diets with 40% distillers grains on feedlot performance (Burken et al., 2013a) Treatment P-value 15 30 45 55 Lin. Quad. DMI, lb/day 23.15 22.77 22.70 21.92 0.01 0.45 ADG, lb3 4.04 3.92 3.76 3.53 <0.01 0.19 Feed:Gain 0.175 0.172 0.166 0.161 <0.01 0.33 115:40= 15% Corn Silage, 40% MDGS; 30:40= 30% Corn Silage, 40% MDGS; 45:40= 45% Corn Silage, 40% MDGS; 55:40= 55% Corn Silage, 40% MDGS; 30:65= 30% Corn Silage, 65% MDGS; 45:0= 45% Corn Silage, 0% MDGS.

2Lin. = P-value for the linear response to corn silage inclusion, Quad.= P-value for the quadratic response to corn silage inclusion, 30 = t-test comparison of treatments 30:40 and 30:65, 45 = t-test comparison of treatments 45:40 and 45:0.

3Calculated from hot carcass weight, adjusted to a common 63% dressing percentage.

104 Table 5. Effect of feeding 15 or 45% corn silage with 20 or 40% MDGS inclusion on cattle performance and carcass characteristics (Burken et al., unpublished). Treatment1 P-value2 Control 15:20 15:40 45:20 45:40 F-test Int. Silage MDGS DMI, lb/day 27.2 26.1 26.4 26.9 26.7 0.13 0.41 0.07 0.86 ADG, lb3 4.32 4.26 4.42 4.19 4.22 0.11 0.18 0.01 0.06 Feed:Gain3 0.159bc 0.163ab 0.167a 0.156c 0.158c <0.01 0.61 <0.01 0.07 HCW, lb 879 874 885 866 869 0.18 0.41 0.01 0.12 1Control = 5% cornstalks, 40% MDGS; 15:20 = 15% Corn Silage, 20% MDGS; 15:40 = 15% Corn Silage, 40% MDGS; 45:20 = 45% Corn Silage, 20% MDGS; 45:40 = 45% Corn Silage, 40% MDGS

2F-test= P-value for the overall F-test of all diets. Int. = P-value for the interaction of corn silage X MDGS. Silage = P-value for the main effect of corn silage inclusion. MDGS = P-value for the main effect of MDGS inclusion.

3Calculated from hot carcass weight, adjusted to a common 63% dressing percentage.

4Marbling Score: 400 = Small00, 500 = Modest00.

abcWithin a row, values lacking common superscripts differ (P < 0.10). provide 60to 90mg/animaldailyfor thelast20to be fedataconcentration of7.56g/tondietDMto and target 28days.Zilmaxwasapprovedin2006to data, mostfeedlotswillfeed 200to300mg/animal drawal time(FDA,NADA 141-221,2003).Basedon last 28to42dofthefeedingperiodwithno DM andbetween70to430mg/animaldailyfor the in 2003tobefedatarateof8.2to24.6g/tondiet from Merck Animal Health).Optaflexx wasapproved Animal Health)andzilpaterol(tradenameZilmax U.S.: ractopamine(tradenameOptaflexxfromElanco ciency. Two betaagonistsareapproved foruseinthe crease carcassweights,gain,andimprovefeedeffi fed tocattleattheendoffeedingperiodin press estruswhichimprovesgainandfeedefficiency. ing MGA (melengesterolacetate)iscommonto sup clearly gainisdecreased.Forfinishingheifers,feed is unknownontheseindividualcattlewithabscesses, er trimatthepackingplantpresumably. While intake carcass weightwithadheredabscessesisduetogreat and Lawrence,2010). The muchgreaterdecrease in if adheredtothecarcass(Davisetal.,2007;Brown lb decreaseswhennotadhered,26to30 databases, cattlewith A+ liverabscesseshad7to10 impact onperformance.Inafewlarge summariesof most severeabscesscategory(A+)causesthebiggest weight andthusgainlikelyduetotrimlosses. The high-grain diets.Feedingtylosinincreasescarcass to decreaseliverabscessesthatresultfromfeeding tylosin (Tylan, Elanco Animal Health). Tylosin isfed al., 2010). The secondmostcommonfeedadditiveis efficiency by2.5to3.5%inrecentstudies(Duffieldet a recentreview, feedingmonensinimprovedfeed is monensin(Rumensin,Elanco Animal Health).In The mostcommonionophorefedtofinishingcattle important additivesusedinhumanmedicine. moved inthefutureifthereiscrossovertomedically proved feedefficiency, thoselabelclaimswillbere many areapprovedforgrowthpromotionandim approved, meaningnooff-label useisallowed. While follow legalguidelinesestablishedwhentheywere options. FeedadditivesareFDA approvedandmust implants. Within bothcategories,therearemany tor. The twomaincategoriesarefeedadditivesand technologies arecommonlyusedinthefeedlotsec Use oftechnology(implantsandbetaagonists) Beta agonistsaretheothermajorfeedadditive The firstcommonadditiveisionophores. For conventionalbeefproduction,numerous - - - - - 105 - - - - 40 d before harvest, with a three day with- not depress quality grades of cattle if compared at drawal time (FDA NADA 141-258, 2006). Zilmax is equal fatness, but does with equal days fed in the not commercially available today. When it was fed, feedlot. No other technology used today in feedlot 20 days were targeted followed by a 3 day withdraw- cattle has as great of a return as use of implants. al. Because these products have dramatic increases in carcass weight and weight gain and are fed at the CHALLENGES end of the feeding period when cattle normally have Measuring feed efficiency in pen settings poorer feed efficiency, they dramatically improve the While we think about feed efficiency of efficiency of the beef industry and also profitability. individual cattle, we don’t measure individual feed Feeding Optaflexx increases carcass weights efficiency in feedlots. Cattle are fed in pens. While by 13.4 to 20.3 lb depending when fed at 200 to 300 gains are estimated (note estimated due to weighing mg daily to steers (Pyatt et al., 2013) with a 10 to conditions) for individuals, there is no sound, sci- 15% improvement in feed efficiency during the final entific method for accurately predicting individual 28 days. Feeding Zilmax for the last 20 days increases feed intakes. Obtaining individual dry matter intake live weights by 19 lb, but increases carcass weights is critical but is generally limited to research set- by 33 lb primarily by shifting bodyweight from less tings or small-scale evaluations using Calan gates, internal fat to greater muscle mass (Elam et al., 2009). GrowSafe, or other systems. Based on these data, we On a live basis, there is not a dramatic improvement know dry matter intakes vary by 20% or more within in feed efficiency. However, if adjusted for carcass groups of “like” cattle, gains vary by more than 30%, weight gain and increased yield of red meat, feeding which leads to tremendous variation in feed efficien- Zilmax dramatically improves efficiency of the beef cy (+/- 20%). If feed efficiency varies by 20% from industry as well. the mean, then you cannot calculate intake based on The last major technology used by feedlots to a gain measurement of individuals very effective- improve feed efficiency is the use of implants. Ste- ly. While variation is good for selection purposes, roid implants are approved to be placed in the middle variation makes comparing individuals within pens third of the ear, just below the skin and slowly release extremely difficult to predict, even with sophisticated hormone over a set period of days (usually 90 to 120 calculations. days but some last more 200 days). Implants can be classified into two major categories, estrogenic or Age combination implants and can further be classified Cattle age when entering the feedlot has based on strength or overall amount of steroid hor- dramatic impacts on performance while in the feedlot mone. Combination implants provide both estrogen phase. Numerous comparisons between feeding year- and trenbolone acetate (TBA) which is an analog of lings versus calf-feds are available. One important testosterone. There is no withdrawal on implants as component of these comparisons is whether the cattle the location used in the animal is discarded at slaugh- are genetically similar or not. Normal procedures ter although it is economically wise to use the last im- in commercial production would be large framed, plant 90 to 120 days prior to slaughter to fully capture heavier weaned calves would be targeted to be fed the value. Guiroy et al. (2002) summarized the impact as calf-feds whereas smaller framed, lighter weaned of different implant strengths and concluded that final calves are traditionally “grown” into yearlings by live body weight is increased by 40 to 100 lb depend- backgrounding and/or grazing prior to entering the ing on strength. More recently, stronger combinations finishing phase. Griffin et al. (2007) compared per- and longer payout periods have likely lead to even formance and economics of feeding calf-feds or greater increases in weights within approximately the yearlings that were not similar in genetics at weaning. same number of days. In general, implanting increas- Calf-feds had fall receiving weights of 642 lb in mid es ADG for the entire feeding period by 10 to 15% November whereas the group “grown” into yearlings and improves feed efficiency by 8 to 12%. Preston et weighed 526 lb at that time. After backgrounding al. (1990) concluded that implanted cattle require a through the winter by grazing cornstalks and some few more days (7-10 depending on gender) to reach drylotting, grazing pasture in the summer, the year- similar body composition or fatness. Implanting does lings were 957 at feedlot entry the following fall. This

106 Table 6. Cattle background impact on feedlot performance for calf-feds, summer yearlings, and fall yearlings originating from the same pool of cattle as weaned calves (Adams et al., 2010)1.

Calf-fed Summer Yrlg Fall Yrlg

Initial BW 576 789 928 DMI, lb/d 20.1a 25.1b 29.0c ADG, lb 3.59a 4.10b 4.28b G:F 0.179a 0.164b 0.147c Hot carcass weight, lb 774 856 919

was a 7 year comparison. Yearlings ate more feed per nomic impact. Numerous studies illustrate that BRD day, had greater daily gains, but were less efficient negatively impacts gains and these are based on how (F:G = 6.76) than calf-feds (F:G=5.63). Yearlings individuals within pens that were diagnosed with finished heavier with 50 heavier carcasses at about the BRD (and presumably have BRD) gained compared same fat thickness. Using similar cattle (i.e., starting to those not treated. Gardner et al. (1999) observed a with the same “pool” of cattle each fall, Adams et 12% decrease in gain and 44 lb lighter carcasses for al. (2010) fed those cattle as either calf-fed, summer cattle treated more than once compared to not treat- (short) yearlings, or fall (i.e., long) yearlings and ed at all for BRD. Treating once didn’t have much compared performance (Table 6). In their study, they impact. Ranch-to-rail data from New Mexico found a imposed two treatments that included either sorting 14% decrease in daily gain and 15 lb lighter carcasses or not which had little impact on performance during for cattle treated more than once (Waggoner et al., finishing. Evaluating just finishing performance, 2007). The study with the greatest number of cattle yearlings eat more per day, gain more per day, but are (about 21,000 head total) was by Reinhardt et al. less efficient than calf-feds (Table 4). Summer fed (2009) using cattle in Iowa feedlots. They observed yearlings are intermediate. Meaning reasons exist for a 20% decrease in daily gain for steers and a 27% either feeding cattle as calf-feds or growing them into decrease for heifers treated more than once versus not yearlings including forage resources available, opti- at all. Treating once was intermediate in their study. mizing finish weight (i.e., carcass weight), and eco- Cattle treated more than once were also 15 lb lighter nomics. Even though yearlings are less efficient while at slaughter. All of these data are on pen-fed cattle in the feedlot, grazing or utilizing forage is unique to where intakes, and thus feed efficiency, are unknown. ruminant production is makes these systems very eco- When economics are applied, most have assumed nomical despite poorer feed efficiency while in just average pen intakes which means a 20% decrease in the feedlot phase. Feeding yearlings also increases gain translates to cattle being 20% less efficient. saleable weight per weaned calf as they “grow” frame Some data are available on the impact on during the backgrounding phase. feed efficiency with intakes measured. An excellent Table 6. Cattle background impact on feedlot per- study was done at Oklahoma State where cattle were formance for calf-feds, summer yearlings, and fall received and then after receiving, cattle were penned yearlings originating from the same pool of cattle as (grouped) based on whether they got treated 0, 1, 2, 3, weaned calves (Adams et al., 2010)1. or 3+ times during the first 60 days. If they were sick during receiving, gain decreased dramatically during Bovine Respiratory Disease the first 60 days. Interestingly, gains were very similar Bovine respiratory disease (BRD) is detri- from day 60 to finish after being penned or grouped mental to the cattle industry and is perceived to have based on how many times they had gotten sick. Cattle large economic impacts from treatment costs and lost consumed less feed after the receiving period if they performance. However, many studies that evaluate had gotten sick so cattle were actually more efficient the impact of BRD on feedlot cattle performance are during finishing if they had gotten sick, and improved incorrect and lead to erroneous estimates of eco- linearly as number of times treated increased. Clearly,

107 the majority of BRD occurs within a few weeks of weights on cattle once fill is replenished at the yards receiving (Babcock et al., 2009). will deflate gains some and cattle will appear less We evaluated data from our individual feeding efficient. True biological efficiency of cattle is not -im facility at UNL (Calan gates), as well as individual pacted by shrink as very little carcass weight is ever feeding data from the University of Illinois using lost in normal situations of shrink due to transport and GrowSafe. In those two datasets, if cattle contract- handling cattle at marketing. ed BRD within the first 30 days (out of 120 or more total days), there was no impact on intakes, gains, or Carcass weight gain for efficiency efficiency. If treated after the first 30 days on feed for As discussed earlier, feeding Zilmax dra- BRD, cattle tended to eat less, gain less, but efficien- matically increases carcass weight (33 lb) yet only cy was the same as healthy cohorts. This suggests to increases live weight by 19 lb compared to controls us that cattle gain less and eat less when they are sick. within studies. As a result, dressing percentage (car- After a receiving period (especially if only treated cass weight divided by live weight) is dramatically once or twice), gains come back some, but cattle still increased (usually by about 1.5 percentage units). eat less which is interesting. Getting sick early in the Genetics can dramatically influence the relative feeding period or at receiving probably has little im- amount of carcass gain compared to live weight gain. pact overall if treated and they recover (treated only One example of this are some recent data collected at once). These changes in intake and efficiency (or lack UNL using Piedmontese and active, inactive, or het- thereof) should be taken into account when applying erogenous myostatin allele cattle. Table 7 shows two economics to cattle that are affected by BRD. In all years of data feeding calf-fed steer calves with these these studies, it is important to point out that data are three genotype variations and the impact on finishing based on visual observation, body temperature, and performance (Moore et al., 2013). Live gains were a diagnosis of BRD which may not always be 100% decreased with the inactive myostatin genetic back- accurate. ground but when carcass-adjusted, gains were not decreases and cattle were dramatically more efficient (in Weighing conditions both scenarios). Cattle were much leaner but dressing Most people in the beef industry take for percentage increased 4.25 percentage units. Similar granted that when weights are collected, cattle weigh results were observed with yearling heifers finished. whatever the scale reads. While that is true, this The beef industry should begin evaluating weight may not be repeatable. The main factor affect- efficiency on a carcass weight basis, including calcu- ing weights and particularly variation in weights is lation of carcass gain (new measure of average daily gut fill. How this impacts feed efficiency in feedlots is gain) and feed efficiency from that gain calculation. probably less of a concern, but can be a real concern More evidence for this is recent work illustrating when establishing a weight and price for sale of cattle the economic benefits of feeding cattle longer and (entering or leaving a feedlot) and also when calcu- larger when marketing on a carcass weight basis or lating gains from initial and final weights. Length of grid basis because gain of carcass does not decrease time when gains are measured improves these esti- at the end of the feeding period like live weight gain mates of gain. Watson et al. (2013) summarized the does (MacDonald et al., 2014). The biggest challenge impact of different weighing conditions on gain esti- is lack of accurate carcass weights at the beginning mates for growing cattle. Equalizing gut fill by limit of the feeding period to use in calculating carcass feeding and multiple day weights improved accuracy gain. However, for feedlots selling cattle on a carcass in gain estimates for growing cattle especially when weight basis, collection of live weights at the end of measured over short durations. The reason for being the feeding period when loaded for transport to the aware is that cattle that are severely shrunk when slaughter plant are meaningless as well. If not collect- arriving at the feedlot due to transport distance and reed, an estimate has to be made for final live weight moval of feed and water for long periods of time will from carcass weight anyway to calculate closeouts. certainly “refill” when given access to feed and water. Using severely shrunk weights will inflate gains and make cattle appear more efficient. Likewise, using

108 Table 7. Live and carcass-adjusted BW performance, and carcass traits of calf-fed steers varying in allele copies of myostatin using Piedmontese. Myostatin1 P – Value2 Performance traits ACTIVE HET INACTIVE Lin. Quad. DMI, lb/d 18.9 17.1 15.0 < 0.01 0.69 Final BW, lb3 1132 1099 1015 < 0.01 0.27 ADG, lb/d 2.56 2.35 2.26 < 0.01 0.43 F:G 7.30 7.25 6.67 < 0.01 0.07 Carcass-adjusted BW4 ADG, lb/d 2.53 2.39 2.58 0.72 0.05 F:G 7.41 7.09 5.88 < 0.01 < 0.01 Carcass traits HCW, lb 712 699 684 0.18 0.93 Dress, % 63.0 63.7 67.3 < 0.01 < 0.01 Marbling5 597 453 283 < 0.01 0.57 LM area, in2 12.4 14.6 15.5 < 0.01 0.05 12th rib Fat, in 0.51 0.28 0.13 < 0.01 0.26 1Myostatin: homozygous active (ACTIVE), heterozygous (HET), and homozygous inactive (INACTIVE) 2P-value: Lin. = linear response to inactive myostatin and Quad. = quadratic response to inactive myostatin 3Live BW collected on 2 consecutive d prior to shipment, shrunk 4 % 4Carcass-adjusted BW calculated at 63 % dressing 5Marbling score: 500 = SM, 400 = SL, 300 = TR, 200 = PD

Comparisons at equal body composition LITERATURE CITED Body composition influences overall feed efficiency due to the energetics of depositing fat or Adams, D. R., T. J. Klopfenstein, G. E. Erickson, W. muscle. Some of the impact of age (calf-feds versus A. Griffin, M. K. Luebbe, M. A. Greenquist, yearlings) is due to composition of growth. However, and J. R. Benton. 2010. Effects of sorting when cattle are not finished to the same endpoint in steers by body weight into calf-fed, summer terms of carcass fatness, leaner animals are more ef- yearling, and fall yearling feeding systems. ficient. Likewise, as cattle grow during finishing and Prof. Anim. Sci. 26:587-594. deposit more fat, their efficiency (of live weight gain) decreases as well. The reason cattle efficiency is im- Babcock, A. H., B. J. White, S. S. Dritz, D. U. Thom- pacted by composition of gain is because it requires son, and D. G. Renter. 2009. Feedlot health about the same amount of calories (energy) to deposit and performance effects associated with the protein and fat in the carcass. However, muscle is timing of respiratory disease treatment. J. about 75% water and 25% protein. As a result, mus- Anim. Sci. 87:314-327. cle growth (not protein) is about 3 times greater in efficiency of growth which is logical so cattle deposit Benton, J. R., G. E. Erickson, T. J. Klopfenstein, K. muscle first and retain energy as fat only when ad- J. Vander Pol, and M. A. Greenquist. 2007. ditional energy is consumed above that required to Effects of roughage source and level with the grow muscle. The point is that cattle sold “early” that inclusion of wet distillers grains on finishing are leaner may be more efficient than cattle fed later cattle performance and economics. Neb. Beef and at least some of the efficiency difference is due to Cattle Rep. MP 90:29-32. composition of gain. 109 Bremer, V. R., A. K. Watson, A. J. Liska, G. E. Erick- Davis, M. S., W. C. Koers, K. J. Vander Pol, and O. A. son, K. G. Cassman, K. J. Hanford, and T. J. Turgeon. 2007. Liver abscess score and car- Klopfenstein. 2011. Effect of distillers’ grains cass characteristics of feedlot cattle. J. Anim. moisture and inclusion level in livestock diets Sci. 85(E Suppl):84 abstr. on greenhouse gas emissions in the corn-eth- anol-livestock life cycle. Prof. Anim. Sci. Duffield, T. F., J. K. Merriill, and R. N. Bagg. 2012. 27:449-455. Meta-analysis of the effects of monensin in beef cattle on feed efficiency, body weight gain, and Brown, T. R., and T. E. Lawrence. 2010. Association dry matter intake. J. Anim. Sci. 90:4583-4592. of liver abnormalities with carcass grading performance and value. J. Anim. Sci. 88:4037- Elam, N. A., J. T. Vasconcelos, G. Hilton, D. L. 4043. VanOverbeke, T. E. Lawrence, T. H. Mont- gomery, N. T. Nichols, M. N. Streeter, J. P. Burken, D. B., B. L. Nuttelman, T. J. Klopfenstein, Hutcheson, D. A. Yates, and M. L. Galyean. and G. E. Erickson. 2013a. Feeding elevated 2009. Effect of zilpaterol hydrochloride du- levels of corn silage in finishing diets containing ration of feeding on performance and carcass MDGS. Neb. Beef Cattle Rep. MP98:74-75. characteristics of feedlot cattle. J. Anim. Sci. 87:2133-2141. Burken, D. B., T. J. Klopfenstein, and G. E. Erickson. 2013b. Economics of feeding elevated levels of FDA. 2003. Freedom of information summary. corn silage in finishing diets containing MDGS. Original new drug application NADA 141-221. Neb. Beef Cattle Rep. MP98:76-77. Ractopamine hydrochloride (OPTAFLEXX 45). Type A medicated article for cattle fed in con- Buttrey, E. K., N. A. Cole, K. H. Jenkins, B. E. finement for slaughter. Found at: http://www. Meyer, F. T. McCollum, S. L. M. Preece, B. W. fda.gov/downloads/AnimalVeterinary/ Products/ Auvermann, K. R. Heflin, and J. C. MacDon- ApprovedAnimalDrugProducts/FOIADrugSum- ald. 2012, Effects of twenty percent corn wet maries/UCM236239.pdf. Accessed 11 February distillers grains plus solubles in steam-flaked or 2014. dry-rolled corn-based finishing diets on heifer performance, carcass characteristics, and ma- FDA. 2006. Freedom of information summary. nure characteristics. J. Anim. Sci. 90:5086-5098. Original new drug application NADA 141-258. Zilpaterol hydrochloride (ZILMAX). Type A Cooper, R., B. Dicke, D. J. Jordon, T. Scott, C. Mack- medicated article for cattle fed in confinement en, and G. E. Erickson. 2014. Impact of feeding for slaughter. Found at: http://www.fda.gov/ alkaline-treated corn stover at elevated amounts downloads/AnimalVeterinary/Products/ Approve- in commercial feedlot cattle. Neb. Beef Cattle dAnimalDrugProducts/FOIADrugSummaries/ Rep. MP99:69-71. ucm051412.pdf. Accessed 11 February 2014

Cooper, R. J., C. T. Milton, T. J. Klopfenstein, and D. Galyean, M. L., and P. J. Defoor. 2003. Effects of J. Jordon. 2002a. Effect of corn processing on roughage source and level on intake by feedlot degradable intake protein requirement of finish- cattle. J. Anim. Sci. 81 (Suppl 2):E8-E16. ing cattle. J. Anim. Sci. 80:242–247. Gardner, B. A., H. G. Dolezal, L. K. Bryant, F. N. Corrigan, M. E., G. E. Erickson, T. J. Klopfenstein, Owens, and R. A. Smith. 1999. Health of M. K. Luebbe, K. J. Vander Pol, N. F. Meyer, finishing steers: Effects on performance, car- C. D. Buckner, S. J. Vanness, and K. J. Han- cass traits, and meat tendernsess. J. Anim. Sci. ford. 2009. Effect of corn processing method 77:3168-3175. and corn wet distillers grains plus solubles inclusion level in finishing steers.J. Anim. Sci. 87: 3351-3362. 110 Griffin, W. A., T. J. Klopfenstein, G. E. Erickson. D. Nuttelman, B. L., D. B. Burken, C. J. Schneider, G. M. Feuz, J. C. MacDonald and D. J. Jordon. E. Erickson, and T. J. Klopfenstein. 2013. 2007. Comparison of performance and eco- Comparing wet and dry distillers grains plus nomics of a long-yearling and calf-fed system. solubles for yearling finishing cattle.Neb. Prof. Anim. Sci. 23:490-499. Beef Cattle Rep. MP98:62-63.

Guiroy, P. J., L. O. Tedeschi, D. G. Gox, and J. P. Nuttelman, B. L., W. A. Griffin, J. R. Benton, G. E. Hutcheson. 2002. The effects of implant strat- Erickson, and T. J. Klopfenstein. 2011. Compar- egy on finished body weight of beef cattle.J. ing dry, wet, or modified distillers grains plus Anim. Sci. 80:1791-1800. soluble on feedlot cattle performance. Neb. Beef Cattle Rep. MP94:50-52. Harris, M. E., G. E. Erickson, K. H. Jenkins, and M. K. Luebbe. 2014. Evaluating corn condensed Owens, F. N., D. S. Secrist, W. J. Hill, and D. R. distillers solubles concentration in steam-flaked Gill. 1997. The effect of grain source and grain corn finishing diets on cattle performance and processing on performance of feedlot cattle: A carcass characteristics. Neb. Beef Cattle Rep. review. J. Anim. Sci. 75:868-879. MP99:86-87 Peterson, S. J., B. L. Nuttelman, C. J. Schneider, D. B. Johnson, J. M., D. B. Burken, W. A. Griffin, B. L. Burken, J. C. MacDonald, and G. E. Erickson. Nuttelman, G. E. Erickson, T. J. Klopfenstein, 2014a. Optimum inclusion of alkaline-treated M. J. Cecava, and M. J. Rincker. 2013. Effect of cornstalks and distillers grains fed to calf-fed feeding greater amounts of calcium oxide treat- steers. Neb. Beef Cattle Rep. MP99:72-74. ed corn stover and Micro-Aid on performance and nutrient mass balance. Neb. Beef Cattle Preston, R. L., S. J. Bartle, A. C. Brake, and R. E. Rep. MP98:70-73. Castlebury. 1990. Effect of anabolic growth implants on the time required to reach quality MacDonald, J. C., C. J. Schneider, K. M. Rolfe, S. D. grade equivalent to nonimplanted cattle. Texas Kachman, T. J. Klopfenstein, and G. E. Erick- Tech Univ. Agric. Sci. Tech. Rep. No T-5-283. son. 2014. Optimal marketing date of steers depends on marketing strategy. Neb. Beef Cattle Pyatt, N. A., G. J. Vogel, J. W. Homm, R. L. Botts, Rep. MP99:92-96 and C. D. Bokenkroger. 2013. Effects of ractopamine hydrochloride on performance and car- Macken, C. N., G. E. Erickson, T. J. Klopfenstein and cass characteristics in finishing steers: 32-trial R. A. Stock. 2006. Effects of corn processing summary. J. Anim. Sci. 91(Suppl. 2):79 abstr. method and protein concentration in finishing diets containing wet corn gluten feed on cattle Reinhardt, C. D., W. D. Busby, and L. R. Corah. performance. Prof. Anim. Sci. 22:14-22. 2009. Relationship of various incoming cattle traits with feedlot performance and carcass Moore, S. K., C. J. Schneider, K. M. Rolfe, B. L. traits. J. Anim. Sci. 87:3030-3042. Nuttelman, D. B. Burken, W. A. Griffin, J. R. Benton, G. E. Erickson, and M. L. Spangler. Shreck, A. L., B. L. Nuttelman, W. A. Griffin, G. E. 2013. Association of inactive myostatin in Erickson, T. J. Klopfenstein, and M. J. Cecava. Piedmontese-influenced steers and heifers 2012b. Reducing particle size enhances chemi- on performance and carcass traits at different cal treatment in finishing diets.Neb. Beef Cattle endpoints. Neb. Beef Cattle Rep. MP98:53-55. Rep. MP95:108-109.

111 Scott, T. L., C. T. Milton, G.E. Erickson, T. J. Klop- RELATIONSHIP BETWEEN fenstein, and R. A. Stock. 2003. Corn process- SELECTION FOR FEED EFFICIENCY ing method in finishing diets containing wet AND METHANE PRODUCTION corn gluten feed. J. Anim. Sci. 81:3182-3190. Harvey Freetly1 1USDA, ARS, U. S. Meat Animal Research Center, Titlow, A. H., A. L. Shreck, S. A. Furman, K. H. Jen- Clay Center, NE 68933* kins, M. K. Luebbe, and G. E. Erickson. 2013. *USDA is an equal opportunity provider and employ- Replacing steam-flaked corn and dry-rolled corn er. with condensed distillers solubles in finishing diets. Neb. Beef Cattle Rep. MP98:51-52. Where Does Methane Come From? Enteric methane is a product of fermentation Vander Pol, K. J., M. A. Greenquist, G. E. Erickson, in the gastro-intestinal tract of ruminants. A group of T. J. Klopfenstein, and T. Robb. 2008. Effect archaea bacteria collectively called “methanogens” of corn processing in finishing diets containing are responsible for the synthesis of methane. Metha- wet distillers grains on feedlot performance nogens live in environments that are void of oxygen and carcass characteristics of finishing steers. and are frequently involved with the fermentation of Prof. Anim. Sci. 24:439-444. organic material. In addition to being found in the gastro-intestinal tract of animals, they are found in Waggoner, J. W., C. P. Mathis, C. A. Loest, J. E. other sites where fermentation occurs such as bogs, Sawyer, F. T. McCollum and J. P. Banta. 2007. marshes (marsh gas), landfills, waste water contain- Case Study: Impact of morbidity in finishing ment ponds, and feedlot surfaces. Methanogens beef steers on feedlot average daily gain, car- typically use acetate or carbon dioxide and hydrogen cass characteristics, and carcass value. Prof. as substrate to grow and produce methane as a by- Anim. Sci. 23:174-178. product. In ruminants, the majority of the methanogen species use carbon dioxide and hydrogen. In Watson, A. K., B. L. Nuttelman, T. J. Klopfenstein, L. ruminants, the methanogens grow in the reticulum-ru- W. Lomas, and G. E. Erickson. 2013. Impacts men complex and in the cecum. Most of the methane of a limit feeding procedure on variation that a ruminant produces is in the reticulum-rumen and accuracy of cattle weights. J. Anim. Sci. (87%), and is released into the environment through 91:5507-5517. the mouth (Murray et al., 1976). Most of the methane produced in the cecum (89%) is absorbed in the blood and travels to the lungs where it is exhaled during respiration (Murray et al., 1976). About 3% of the methane is released from the rectum (Murray et al., 1976; Muñoz et al., 2012). Methanogens live in a symbiotic relationship with the other bacteria in the rumen; however, they make up a relative small proportion of the total rumen microbes (Krause and Russell, 1996; Mosoni et al., 2011). Methanogens help maintain a zero net hydrogen balance in the rumen by releasing hydrogen in the form of methane rather than other microbes producing longer chained volatile fatty acids such as propionate.

The Problem Methane is a greenhouse gas. Depending on the size and level of feed intake, cattle will produce 10 to 16 kg of methane per year (Hristov et al., 2013).

112 Methane represents a lost opportunity to capture feed Residual feed intake (RFI) is the difference energy. If captured, this lost energy could potential- in amount of feed consumed by an animal from that ly be used for maintenance, growth, and lactation. predicted for its rate of body weight gain and size. There is a lot of variation in the fraction of intake Negative RFI are more efficient since they ate less energy released as methane (Johnson and Johnson, feed than is predicted to be needed for a given rate 1995). This variation can partially be explained by of production. Residual feed intake has been used the composition of the diet. About 3% of intake ener- as a measure of feed efficiency and has been used in gy consumed by steers fed a high-corn diet is lost as selection programs to improve feed efficiency. Selec- methane energy (Archibeque et al., 2007). The per- tion on RFI decreases feed intake (Herd et al., 2002). centage increases when cattle are eating a high-forage Methane production increases with increased feed in- diet. Increasing the forage:concentrate ratio increased take; however, the methane per unit of feed decreases methane production (Reynolds et al., 1991; Sauvant (Blaxter and Clapperton, 1965). Hegarty et al. (2007) and Giger-Reverdin, 2007). Methanogens are sensi- reported that cattle selected for low RFI have a re- tive to low rumen pH and their prevalence decreases duced daily methane production, and Nkrumah et al. (Van Kessel and Russell, 1996). Pregnant beef cows (2006) found that steers that ranked low for RFI had eating a corn silage based diet will release 5 to 7% a reduced methane production. Zhou et al. (2009) de- of their gross energy intake as methane (Freetly et termined the relative proportion of different species of al., 2008). A number of strategies have been used methanogens differ between cattle classified as having to reduce methane production including chemical less or greater RFI which may influence the potential inhibitor, ionophores, and manipulation of the rumen to produce methane. The studies of Nkrumah et al. ecology. A potential approach for reducing methane (2006) and Hegarty et al. (2007) differ when methane production is to select for increased feed efficiency. productions are expressed per unit of feed fed. He- garty et al. (2007) found that cattle selected for a low Methane and Feed Efficiency RFI also had a reduced total feed intake, but they did The relationship between methane production not differ in the amount of methane produced per unit and feed efficiency is dependent on how feed effi- of feed. In our studies, we found RFI did not ac- ciency is defined. Selecting cattle for greater resid- count for differences in methane production when we ual gain or greater gain:feed ratios may result in an adjusted for feed intake (Freetly and Brown-Brandl, increase in methane production. Residual gain is the 2013). Nkrumah et al. (2006) found that steers with difference in amount of body weight gain an animal a low RFI produced less methane per unit fed than achieves compared to what it is predicted to gain for a other steers. Collectively, these studies suggest that given feed intake. Cattle that more completely digest selection for low RFI does not inherently mean that their feed will get more nutrients per unit of feed methane production per unit of feed is decreased, and produce more methane. In our studies in cattle but methane production is reduced by decreasing the not selected for feed efficiency, methane production amount of feed consumed. increased with increased gain:feed ratios when they were fed a roughage diet, but there were no differenc- Other Factors to Consider es when they were fed a concentrate diet (Freetly and Factors other than feed efficiency contribute to Brown-Brandl, 2013). The different response in the the methane footprint of cattle. Hristov et al. (2013) two experiments may have been due to the relative has reviewed several management strategies used to digestibility of the two diets. The concentrate diet reduce methane production. These include the feed- was highly digestible and the variance in the rate of ing of inhibitors, electron receptors, ionophores, plant digestibility may have been lower than compared to bioactive compounds, enzymes, yeast products, and cattle consuming the less digestible roughage diet. oils. Other approaches have included decreasing the Goopy et al. (2014) found that methane production rumen protozoa and manipulating the rumen archaea increased with increased rumen retention times. They and bacteria ecology. One of the biggest factors that also determined that sheep that produced more meth- determine the lifetime methane production of calves ane had greater rumen volume. is the number of days from birth to harvest. If we

113 assume a 160-day finish period and cattle consume 35 Herd, R. M., P. F. Arthur, and R. S. Hegarty. 2002. Mcal/day and 3% of the consumed energy is released Potential to reduce greenhouse gas emissions as methane, then total methane release is 168 Mcal. from beef production by selection to reduce Using the same assumptions on a 150-day finishing residual feed intake. Communication 10–22 period, methane production is 158 Mcal. The 10- in Proc. 7th World Congr. Genet. Anim. Prod., day decrease on feed results in a 6% drop in methane Montpellier, France. production. Similarly, backgrounding programs that Hristov, A. N., J. Oh, J. L. Firkins, J. Dijkstra, E. prolong the age at harvest will increase lifetime meth- Kebreab, G. Waghorn, H. P. S. Makkar, A. T. ane production. Management and selection programs Adesogan, W. Yang, C. Lee, P. J. Gerber, B. that decrease the age at harvest will reduce lifetime Henderson, and J. M. Tricarico. 2013. Spe- methane production. cial topics--Mitigation of methane and nitrous The bulk of the annual methane production oxide emissions from animal operations: I. A from cattle can be attributed to the cow herd. If we review of enteric methane mitigation options. consider the measure of methane efficiency to be the J. Anim. Sci. 91:5045-5069. amount of calf marketed per unit of methane produced in a cow’s lifetime, then factors that make a cow Johnson, K. A., and D. E. Johnson. 1995. Methane economically efficient are the same that makes her emissions from cattle. J. Anim. Sci. 73:2483- efficient with regard to methane production. Selecting 2492. and managing cattle for prolonged lifetime produc- Krause, D. O. and J. B. Russell. 1996. Symposium: tivity, and pounds of calf marketed per unit of feed Ruminal microbiology. How many ruminal consumed will improve methane efficiency. bacteria are there? J. Dairy Sci. 79:1467- 1475. Literature Cited Mosoni, P., C. Martin, E. Forano, D. P. Morgavi. 2011. Long-term defaunation increases the Archibeque, S. L., H. C. Freetly, N. A. Cole, and C. abundance of cellulolytic ruminococci and L. Ferrell. 2007. The influence of oscillating methanogens but does not affect the bacteri- dietary protein concentrations on finishing cattle. al and methanogen diversity in the rumen of II. Nutrient retention and ammonia emissions. J. sheep. J. Anim. Sci. 89:783-791. Anim. Sci. 85:1496-1503. Muñoz, C., T. Yan, D. A. Wills, S. Murray, and A.W. Blaxter, K. L. and J. L. Clapperton. 1965. Prediction Gordon. 2012. Comparison of sulfur hexa- of the amount of methane produced by rumi- fluoride tracer and respiration chamber tech- nants. Br. J. Nutr. 19:511-522. niques for estimating methane emissions and Freetly, H. C., J. A. Nienaber, and T. Brown-Brandl. correction for rectum methane output from 2008. Partitioning of energy in pregnant beef dairy cows. J. Dairy Sci. 95:3139-3148. cows during nutritionally induced body weight Murray, R. A. A. M. Bryant , and R. A. Leng. 1976. fluctuation. J. Anim. Sci. 86:370-377. Rates of production of methane in the rumen Freetly, H. C., and T. Brown-Brandl. 2013. Enteric and large intestine of sheep. Br. J. Nutr. 36:1- methane production from beef cattle that vary 14. in feed efficiency. J. Anim. Sci. 91:4826-4831. Nkrumah, J. D., E. K. Okine, G. W. Mathison, K. Goopy, J. P., A. Donaldson, R. Hegarty, P. E. Vercoe, Schmid, C. Li, J. A. Basarab, M. A. Price, Z. F. Haynes, M. Barnett, and V. Hutton Oddy. Wang, and S. S. Moore. 2006. Relationships 2014. Low-methane yield sheep have smaller of feedlot feed efficiency, performance, and rumens and shorter rumen retention time. Br. feeding behavior with metabolic rate, methane J. Nutr. 111:578-585. production, and energy partitioning in beef Hegarty, R. S., J. P. Goopy, R. M. Herd, and B. Mc- cattle. J. Anim. Sci. 84:145-153. Corkell. 2007. Cattle selected for lower residual feed intake have reduced daily methane production. J. Anim. Sci. 85:1479-1486. 114 Reynolds, C. K., H. F. Tyrrell, and P. J. Reynolds. 1991. Effects of dietary forage-to-concentrate ratio and intake on energy metabolism in growing beef heifers: Whole body energy and nitrogen balance and visceral heat production. J. Nutr. 121: 994-1003. Sauvant, D., and S. Giger-Reverdin. 2007. Empirical modeling meta-analysis of digestive interactions and CH4 production in ruminants, pp. 561-563 in Energy and Protein Metabolism and Nutrition. I. Ortigues-Marty, N. Miraux, and W. Brand-Williams, ed., Wageningen, The Netherlands. Van Kessel, J. A., and J. B. Russell. 1996. The effect of pH on ruminal methanogenesis. FEMS Microbiol. Ecol. 20:205-210. Zhou, M., E. Hernandez-Sanabria, and L. L. Guan. 2009. Characterization of variation in rumen methanogenic communities under different dietary and host feed efficiency conditions, as determined by PCR-denaturing gradient gel electrophoresis analysis. Appl. Environ. Micro. 75:6524-6533.

115 TECHNICAL COMMITTEES

HEALTHFULNESS OF BEEF: A Materials and Methods GENOME-WIDE ASSOCIATION STUDY Experimental Design USING CROSSBRED CATTLE Crossbred steers and heifers of unknown pedigree and breed fractions (n= 236) with varying per- C.M. Ahlberg1, L.N. Schiermiester1, M. L. Spangler1 centages of Angus, Simmental and Piedmontese were 1Animal Science Department, University of Nebraska– placed in a Calan gate facility at the Agricultural Re- Lincoln search and Development Center (ARDC) feedlot facil- Introduction ity near Mead, NE. Prior to arrival, animals were geno- Consumers are becoming increasingly typed for the Piedmontese-derived myostatin mutation health-conscious and demand healthy and palatable (C313Y) to determine their myostatin genotype (MG) meat, both of which are affected by lipid composition as either homozygous normal (313C/313C, 0 copy, (Dunner et al., 2013). Red meat has relatively high n=83), heterozygous (313C/313Y, 1-copy, n=96), levels of saturated fatty acids and beneficial oleic or homozygous for inactive myostatin (313Y/313Y, acid, and low concentrations of beneficial polyunsatu- 2-copy, n=57). Cattle were fed in four groups over rated fatty acids (Dunner et al., 2013). However, fats a 2-yr period. Groups 1 and 3 consisted of calf-fed are not the only nutrients that affect the nutritional steers and groups 2 and 4 consisted of yearling heifers value of beef. Beef is an excellent source of iron as described by Howard et al., (2013). required in the human diet, yet the consistency of iron content in beef products is highly variable (Duan et Animals had ad libitum access to water and al., 2009). Considerable attention has been placed were fed a diet that met or exceeded National Re- on improving the nutritional value of beef and the search Council (NRC) (1996) requirements. The fin- development of products that are beneficial to human ishing ration for steers and heifers in year 1 included health and disease prevention (Scollan et al., 2006). wet distillers grain with solubles, a 1:1 blend of high moisture and dry rolled corn, grass hay and supple- It has been illustrated that animal nutritional ment at 35, 52, 8, and 5 % of the diet on a dry matter regime differences can alter the nutrient profile of basis. The finishing ration for steers and heifers in beef (Realini et al., 2004) and that genetic factors can year 2 included modified distillers grain with solu- also play a role (De Smet et al., 2004; Mateescu et al., bles, sweet bran, a 1:1 blend of high moisture and dry 2013a,b). Identification of genetic variants that would rolled corn, grass hay and supplement at 20, 20, 48, allow producers to select for optimum nutritional val- 8, and 4 % of the diet on a dry matter basis. Animals ues with respect to fatty acids, minerals, and vitamins, were on an all-natural program and were not implant- without sacrificing performance or product quality, ed or fed growth-promoting additives. Cattle were could ultimately increase value and consumer satis- harvested as a group based on average body weight faction of beef. Genetic selection aided by genomic and external fat. Steaks were sampled from the M. predictors may serve as an important and highly Longissimus thoracis et lumborum (LTL) and the M. applicable tool in improving the nutritional value of Semitendinosus (ST) three days post mortem. Steaks beef given the expensive and difficult nature of phe- were cut to ½ inch thick and trimmed to 1/8 inch of notypic data collection. The objectives of the current subcutaneous fat. Steaks were shipped to Midwest study were to determine the proportion of phenotypic Laboratories, Inc. (Omaha, NE) for further analysis. variation explained by the Ilumina BovineSNP50K- Lipid, and mineral analysis results were reported for a Bead-Chip for cholesterol (CH), polyunsaturated fatty 113.40 gram serving size. acids (PUFA), monounsaturated fatty acids (MUFA), protein, potassium, iron and sodium, to identify chro- Statistics for carcass traits are summarized in mosomal regions that harbor major genetic variants Table 1. Fatty acids (MUFA and PUFA) and CH were underlying the variation of these traits. analyzed as both a percentage of total lipid content

116 TECHNICAL COMMITTEES

and mg/100g of whole (wet) tissue. Omega 3, 6 and tional candidate gene approach was conducted using 9 fatty acids were reported as MUFA or PUFA. The Bos taurus build UMD_3.1 assembly (Zimin et al., interpretation of these two measurement scales is 2009). Due to the limited functional annotation of the dramatically different, as a sample with relatively low Bos taurus genome, human orthologs of beef cattle PUFA content as measured in mg/100g of whole (wet) positional candidate genes were obtained and used for tissue would likely have low total lipid content and functional characterization by using Ensembl Genes as a consequence would have relatively high PUFA 69 database and the BioMart data mining tool (http:// content when measured as a percentage of total lipids. www.ensembl.org/biomart/martview/dd0c118c99ed- Potassium, iron and sodium were analyzed as ppm of 15210cc6e97131d873fb). Overrepresented gene whole tissue. ontology terms, and pathway analysis were identified using DAVID (http://david.abcc.ncifcrf.gov). Statistical Analysis Myostatin genotype has been shown to have Results an effect on fatty acid composition. Consequently, Genomic Heritabilities outliers, adjusted for group and MG, classified as The posterior mean (standard deviation; SD) being > 3 SD from the mean of the residual variance genomic heritability estimates (proportion of pheno- (zero), were removed from the analysis. Summary typic variation explained by the markers) are present- statistics for fatty acid and mineral traits after editing ed in Table 3. For both cuts, heritability estimates for are detailed in Table 2. A genome wide association protein and mineral traits ranged from 0.05 to 0.75. study (GWAS) using the BovineSNP50K Bead-Chip The posterior mean (SD) genomic heritability esti- was conducted via the GenSel platform (Version mates for CH, PUFA and MUFA as a percentage of 0.9.2.045; Fernando and Garrick, 2011). A Bayes C total lipid content for both cuts ranged from 0.40 to model was employed (Habier et al., 2011) with group 0.70. When analyzed as mg/100g of total wet tissue, (concatenation of year (i.e. feeding regime) and sex; the posterior mean (SD) genomic heritability esti- 4 classes) fitted as a fixed effect. The proportion of mates for CH, PUFA and MUFA for both cuts ranged markers having a null effect was set to 0.95. A chain from 0.45 to 0.85. length of 150,000 iterations was run with the first 50,000 discarded as burn-in. The genomic estimated Mateescu et al. (2013a) estimated the herita- breeding value (GEBV) was estimated by summing bility based on pedigree information and phenotypic posterior mean marker effects by marker genotype data to be 0.48, 0.00, and 0.15 for LTL iron, potas- across all SNP. Phenotypic correlations were estimat- sium, and sodium, respectively. The proportions of ed using multivariate analysis of variance (MANO- phenotypic variation explained by the BovineSNP50 VA) procedures with group fitted as a fixed effect. assay were 0.37, 0.03, and 0.09 for iron, potassium To estimate potential GEBV re-ranking, correlations and sodium, respectively (Mateescu et al., 2013b). between GEBV were estimated across traits within These results are in general agreement with the find- a cut (i.e. ST or LTL) and between cuts within each ings of the current study for the traits of iron and sodi- trait. Additionally, the cattle genome was separated um. The vastly different estimates for potassium may into 1 Megabase (Mb) windows and SNP variance be attributed to the admixed population or the small within a window was summed to give an estimate of sample size, and the fact that this population was the total SNP variance for each window (n=2,677). segregating the C313Y mutation. One SNP within one The percentage of top 5% (n=134) windows in com- of the top 1Mb windows for potassium was in perfect mon across traits and cuts were then compared with LD with the myostatin mutation. Lower posterior GEBV correlations among traits and between cuts. mean estimates of genomic heritability for ST sodium The top 0.5% 1-Mb windows (n=13) for each trait is likely a function of the lower phenotypic variation were extended by 1-Mb in both directions and a posi- of sodium content, which can be explained biological-

117 ly by the body highly regulating sodium levels (Hol- a percentage of fat and as mg/100g of wet tissue. lenberg, 1980). The MUFA was negatively correlated with PUFA and CH within and across cuts when measured on a For LTL and ST CH, LTL PUFA and ST percentage of total fat. However, when measured as MUFA posterior mean estimates of genomic herita- mg/100g of wet tissue, MUFA and PUFA were strong bility remained the same regardless of the scale of positively correlated and CH was moderate negatively measurement (percentage of total lipids or mg/100g correlated with MUFA and PUFA between and across of whole (wet) tissue). The genomic heritability esti- cuts. Consequently, from a selection perspective, the mate for LTL MUFA was higher when measured on phenotype used (percentage or mg/100g) would lead mg/100g of whole (wet) tissue than on a percentage to the selection of different animals. This is primar- of total lipids. ST PUFA genomic heritability was ily because increases in fat content dilute fatty acids lower when measured on mg/100g whole (wet) tissue found in membranes, notably CH and PUFA. Expres- basis. The coefficients of variation for ST PUFA were sion of results as mg/100g of wet tissue thus reflects 0.61 and 0.34 when measured as a percentage of overall increases in fat content. total lips and mg/100g, respectively. This increase in variation could partially explain the increase in the The interpretation of results relative to fat- proportion of variation explained by the markers. Al- ty acids is conditional on understanding the scale though the ST had lower concentrations of PUFA as of the phenotypes (percentage of total fatty acids measured in mg/100g of wet tissue, it also had lower or mg/100g of wet tissue). When the gravimetric values for total lipids. Consequently when PUFA amount of PUFA, for instance, is low the amount of was adjusted for total lipid content, the mean PUFA PUFA relative to total fatty acids (percentage of total as a percentage of total lipid content was actually fatty acids) can be high simply because the amount higher than the LTL. The same general trend of the of total fatty acids was also very low. Similarly, when ST containing a higher proportion PUFA and MUFA PUFA content is relatively high as a percentage of as a percentage of total fatty acids was also reported total fatty acids (i.e. when the amount of total fatty by Sexton et al. (2012). Estimates of heritability for acids is also low) CH would also be expected to be fatty acids are sparse in the literature. Pitchford et al. relatively high when measured as a percentage of total (2002) reported low to moderate estimates of heri- fatty acids. The expectation that with the increase in tability for fatty acid traits in beef cattle. However, adipose tissue that CH increases, PUFA decreases and Cameron (1990) reported high (0.53-0.71) heritability MUFA increases on a percent fat basis is challenged estimates for palmitic, stearic, oleic, and linoleic fatty in the case of cattle with the double muscling geno- acids. This is consistent with the estimate of 0.75 for type. Raes et al. (2001) have shown that the double the heritability of C18:1 in a population of Japanese muscling genotype within the Belgian Blue breed has black cattle (Uemoto et al., 2010), and supports a low proportions of MUFA and high proportions of moderate to high level of genetic control of fatty acids PUFA in muscle lipid compared with normal geno- within meat. type animals. This is due to the low concentration of total lipid in the muscle and a high ratio of phospho- lipid and total lipid. Phospholipids are high in PUFA Genomic Estimated Breeding Value and Phenotyp- content in order to perform the function as a con- ic Correlations stituent of cellular membranes (Wood et al., 2008). Correlations between GEBV follow the However, when PUFA content is high in mg/100g phenotypic correlation trends as reported by Ahlberg of whole (wet) tissue, total fatty acid content is also et al., (2014). Phenotypic correlations are presented likely high leading to a reduction in the proportionate in Tables 4 and 5. Among the protein and mineral, amount of CH. as well as the mineral and protein with lipids, correlations were low to moderate between and within Significant correlations between GEBV sug- the two cuts and were varied in the direction of the gest that selection for increased iron concentration correlation when measured as a percentage of fat in the ST would lead to increased levels of MUFA and as mg/100g of wet tissue. Phenotypic correla- and decreased levels of both CH and PUFA as a tions among lipid traits were moderate to strong as percentage of total lipids. In both cuts, selection for

118 increased levels of potassium would have the oppo- myostatin animals had lower proportions of MUFA site effects leading to increased PUFA and CH and and higher proportions of PUFA illustrating that this decreased MUFA as a percentage of total lipids. On mutation has a measureable impact on these traits. a total tissue basis, selection for increased potassium This is supported by Wiener et al. (2009) who showed in both cuts would lead towards a correlated decrease a significant effect of the myostatin mutation in South in PUFA and MUFA and increase in CH. Selection for Devon cattle for both PUFA and MUFA concentra- increased iron would lead to a correlated decrease in tions. Outside of the myostatin mutation, Mateescu CH and an increase in MUFA and PUFA in the ST on et al. (2013c) reported 16 SNP in a single Mb region a total wet tissue basis. (103-104 Mb) on BTA2 to explain 1.33% of the phenotypic variation of iron content, although the region reported by Mateescu et al. (2013) does not overlap Sodium was lowly to moderately correlated with the regions reported in the current study. with all traits measured, in agreement with Matees- cu et al. (2013) who also reported low to moderate Conclusions correlations between sodium and other mineral traits. In general, the mean estimates of the posteri- However, correlations between GEBV between the or heritability were moderate to high for fatty acids, different cuts for sodium was high despite the low suggesting that significant progress could be made proportion of variation explained by the markers. This through selection with the aid of genomics. The strong GEBV correlation may be due to markers pick- proportion of variation for mineral traits was more ing up breed/family relationships, which would give variable, although a moderate proportion of variation rise to a larger positive GEBV correlation. was explained by the markers for iron and potassium Candidate Gene Annotation content. Differences did exist for fat traits depending Functional annotation analysis resulted in a on the scale of measurement (mg/100g or percentage common gene found among lipid traits was GULP1 of total lipid content), in terms of relationships be- (Engulfment adaptor PTB domain containing 1). tween traits, chromosomal regions underlying genetic GULP1 is an adaptor protein that binds and directs variation, and in some cases the proportion of varia- the trafficking of LRP1 Low( density lipoprotein tion explained by the markers. The choice between receptor-related protein 1), which is involved in lipid these two scales would impact the ranking of animals. homeostasis (He and Lin, 2010). ITGAV is associated Further investigation of fatty acid and mineral con- with metabolic processes and negative regulation of centrations need to be conducted in other popula- lipid transport and storage (Kim et al., 2013). tions to fully understand the proportion of variation explained by markers and better predict candidate Some significant SNP from the top 0.5% 1-Mb genes. Potential candidate genes, GULP1 and ITGAV windows that were on BTA2 for each trait were in located on BTA2 in close proximity to C313Y, were high LD with the myostatin C313Y alleles. Conse- identified and involve regulation of lipids. Further quently, these SNP may simply be an artifact of the analysis of expression of these genes will allow for importance of the myostatin mutation for some for better understanding of lipid transport and regulation the traits analyzed. Between all traits and cuts there in muscle and their subsequent role in determining was a wide range in the number of 1-Mb windows meat quality of livestock. that were on BTA2, ranging from 1 to 9 windows. Traits with few top windows on BTA2 are likely not impacted as much by C313Y. Previous work by Aldai et al. (2005) showed significant differences between animals of the Asturiana de los Valles breed of cattle that were homozygous for the myostatin deletion and those that were homozygous normal for protein percentage. The authors also showed that homozygous

119 Table 1. Summary statistics for carcass traits. 0 1 2 Standard De- Trait n Minimum Maximum Mean copya copya copya viation HCW, kg. Group 1c 59 19 28 12 253.55 372.85 305.88 25.42 Group 2c 60 25 26 9 265.80 385.55 319.85 24.96 Group 3c 58 20 22 16 268.52 400.98 332.19 26.84 Group 4c 59 19 20 20 271.25 434.00 346.24 34.19 Back Fat, cm. Group 1 59 19 28 12 0.10 1.40 0.73 0.37 Group 2 60 25 26 9 0.10 2.03 0.84 0.41 Group 3 58 20 22 16 0.25 2.29 0.86 0.55 Group 4 59 19 20 20 0.25 3.05 1.02 0.68 Marbling Scoreb Group 1 59 19 28 12 100 470 294.92 100.75 Group 2 60 25 26 9 100 860 373.00 118.40 Group 3 58 20 22 16 250 880 533.79 166.97 Group 4 59 19 20 20 270 730 426.78 114.75 a Refers to the number of copies of the inactive Myo- statin allele. b Marbling score units: 400 = Sm00, 500 = Modest00 c Group 1 refers to year 1 steers, group 2 refers to year 1 heifers, group 3 refers to year 2 steers and Group 4 refers to year 2 heifers.

120 Table 2. Summary statistics for nutrient traits. Standard Trait Units n Mean Minimum Maximum Deviation LDa MUFA (% of fat) 224 46.25 33.2 55.00 4.31 LTL MUFA (mg/100g) 227 6087.70 270.97 13849.38 3233.42 STb MUFA (% of fat) 223 45.11 26.6 56.70 5.55 ST MUFA (mg/100g) 227 2461.14 37.24 10308.06 1977.84 LTL PUFA (% of fat) 223 5.27 2.66 15.30 2.21 LTL PUFA (mg/100g) 224 572.60 149.86 1197.99 180.07 ST PUFA (% of fat) 222 8.50 1.14 25.60 5.20 ST PUFA (mg/100g) 227 378.87 36.24 735.02 132.31 LTL Cholesterol (% of fat) 222 0.50 0.14 2.84 0.45 LTL Cholesterol (mg/100g) 225 45.76 33.00 59.00 4.48 ST Cholesterol (% of fat) 223 1.94 0.22 17.10 2.48 ST Cholesterol (mg/100g) 225 46.26 32.00 58.00 4.73 LTL Sodium (ppm) 226 418.69 336.50 491.20 32.14 ST Sodium (ppm) 227 393.92 317.40 478.60 29.02 LTL Potassium (ppm) 227 3015.18 2283.00 3614.00 268.71 ST Potassium (ppm) 226 3484.30 2867.00 4087.00 227.99 LTL Iron (ppm) 224 13.65 8.99 19.56 2.06 ST Iron (ppm) 226 13.92 7.50 25.50 2.62 LTL Protein (%) 225 21.69 17.34 27.44 1.86 ST Protein (%) 227 22.91 18.58 26.17 1.33 a M. Longissimus dorsi (LTL) b M. Semitendinosus (ST)

121 Table 3. Genomic heritabilities Trait Units Heritability (SE) LTL a MUFA (% of fat) 0.40 (0.10) LTL MUFA (mg/100g) 0.85 (0.04) STb MUFA (% of fat) 0.60 (0.07) ST MUFA (mg/100g) 0.60 (0.10) LTL PUFA (% of fat) 0.70 (0.06) LTL PUFA (mg/100g) 0.70 (0.08) ST PUFA (% of fat) 0.65 (0.06) ST PUFA (mg/100g) 0.45 (0.04) LTL Cholesterol (% of fat) 0.50 (0.09) LTL Cholesterol (mg/100g) 0.50 (0.06) ST Cholesterol (% of fat) 0.45 (0.10) ST Cholesterol (mg/100g) 0.45 (0.11) LTL Sodium (ppm) 0.15 (0.08) ST Sodium (ppm) 0.05(0.05) LTL Potassium (ppm) 0.75 (0.08) ST Potassium (ppm) 0.65 (0.09) LTL Iron (ppm) 0.35 (0.13) ST Iron (ppm) 0.35 (0.09) LTL Protein (%) 0.70 (0.08) ST Protein (%) 0.75 (0.06) a M. Longissimus dorsi (LTL) b M. Semitendinosus (ST)

122 - 0.48 0.28 0.56 0.71 0.70 0.20 0.51 0.82 -0.36 -0.65 -0.71 -0.20 (0.01) (0.01) (0.91) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) -0.008 LTLPUFA - - 0.74 0.15 -0.44 -0.32 -0.07 -0.37 -0.50 -0.58 -0.60 -0.24 -0.55 -0.63 (0.01) (0.01) (0.27) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.02) LTLMUFA - - - 0.45 0.03 0.28 0.75 0.74 0.71 0.23 0.49 -0.11 -0.27 -0.66 (0.11) (0.01) (0.01) (0.83) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) LTLCH - - - - 0.49 0.02 0.46 0.29 0.47 0.64 0.18 0.50 -0.30 -0.52 (0.01) (0.01) (0.82) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) LTLPO - - - - - 0.15 0.25 0.12 0.17 0.19 0.25 0.20 -0.06 -0.21 LTLS (0.03) (0.36) (0.01) (0.07) (0.01) (0.01) (0.01) (0.01) (0.01) ------0.35 0.02 0.20 LTLI -0.05 -0.06 -0.12 -0.21 -0.08 (0.46) (0.01) (0.73) (0.37) (0.07) (0.01) (0.01) (0.22) . abcdef ------0.59 0.39 0.50 0.68 -0.34 -0.05 -0.61 (0.01) (0.01) (0.01) (0.01) (0.45) (0.01) (0.01) LTLPR

------0.59 0.44 0.68 -0.39 -0.03 -0.85 (0.01) (0.01) (0.70) (0.01) (0.01) (0.01) STPUFA ------0.32 -0.53 -0.09 -0.48 -0.59 (0.01) (0.01) (0.16) (0.01) (0.01) STMUFA ------0.35 0.02 0.24 -0.31 (0.06) (0.01) (0.78) (0.01) STCH ------0.64 0.24 M. Semitendinosus (ST) -0.16 b (0.01) (0.02) (0.01) STPO ------STS 0.30 -0.09 (0.01) (0.19) Standard errors for correlations were 0.067. f ------STI -0.32 (0.01) ------STPR Phenotypic correlations with lipid traits measured as a percent of total fat

abc STI STS LTLI LTLS STPR STPO STCH Trait LTLPR LTLPO LTLCH STPUFA STMUFA LTLPUFA LTLMUFA STCH, STMUFA, STPUFA, LTLCH, LTLMUFA and LTLPUFA units as percent of total fat and LTLPUFA LTLMUFA LTLCH, STPUFA, STCH, STMUFA, et lumborum (LTL) M. Longissimus thoracis ST protein (STPR), ST iron (STI), ST sodium (STS), ST potassium (STPO), ST cholesterol (STCH), ST monounsaturated cholesterol (STCH), ST potassium (STPO), ST sodium (STS), ST iron (STI), ST protein (STPR), ST ST value). Phenotypic correlations (P Table 4. Table a c d e fatty acids (STMUFA), ST polyunsaturated fatty acids (STPUFA), LTL protein (LTLPR), LTL iron (LTLI), LTL sodium LTL iron (LTLI), LTL protein (LTLPR), LTL polyunsaturated fatty acids (STPUFA), ST fatty acids (STMUFA), poly - and LTL monounsaturated fatty acids (LTLMUFA), LTL cholesterol (LTLCH), LTL potassium (LTLPO), LTL (LTLS), unsaturated fatty acids (LTLPUFA)

123 124 Table 5. Phenotypic correlations with lipid traits measured as mg/100g of total (wet) tissueabcdef. Traitabc STPR STI STS STPO STCH STMUFA STPUFA LTLPR LTLI LTLS LTLPO LTLCH LTLMUFA LTLPUFA

-0.32 -0.09 0.64 0.40 -0.48 -0.43 0.59 -0.05 0.14 0.49 0.31 -0.65 -0.47 STPR - (0.01) (0.19) (0.01) (0.01) (0.01) (0.01) (0.01) (0.46) (0.03) (0.01) (0.01) (0.01) (0.01) STI - - 0.20 -0.16 -0.33 0.22 0.19 -0.34 0.35 -0.06 -0.30 -0.29 0.40 0.31 (0.01) (0.02) (0.01) (0.01) (0.01) (0.01) (0.01) (0.36) (0.01) (0.01) (0.01) (0.01) STS - - - 0.24 0.02 0.05 0.02 -0.05 0.02 0.25 0.02 0.01 0.04 0.03 (0.01) (0.78) (0.47) (0.81) (0.45) (0.73) (0.01) (0.82) (0.84) (0.56) (0.63) STPO - - - - 0.29 -0.36 -0.29 0.39 -0.06 0.12 0.46 0.28 -0.46 -0.27 (0.01) (0.01) (0.01) (0.01) (0.37) (0.07) (0.01) (0.01) (0.01) (0.01) STCH ------0.27 -0.21 0.41 -0.07 0.15 0.33 0.34 -0.45 -0.27 (0.01) (0.01) (0.01) (0.29) (0.02) (0.01) (0.01) (0.01) (0.01) STMUFA ------0.82 -0.47 0.06 -0.15 -0.36 -0.24 0.53 0.46 (0.01) (0.01) (0.35) (0.02) (0.01) (0.01) (0.01) (0.01) STPUFA ------0.39 0.16 -0.11 -0.24 -0.19 0.42 0.45 (0.01) (0.02) (0.12) (0.01) (0.01) (0.01) (0.01) LTLPR ------0.08 0.25 0.64 0.46 -0.82 -0.76 (0.22) (0.01) (0.01) (0.01) (0.01) (0.01) LTLI ------0.20 0.18 -0.03 0.09 0.08 (0.01) (0.01) (0.67) (0.20) (0.22) LTLS ------0.50 0.23 -0.24 -0.28 (0.01) (0.01) (0.01) (0.01) LTLPO ------0.39 -0.70 -0.62 (0.01) (0.01) (0.01) LTLCH ------0.45 -0.35 (0.01) (0.01) LTLMUFA ------0.83 (0.01) LTLPUFA ------

a M. Longissimus thoracis et lumborum (LTL) b M. Semitendinosus (ST) cST protein (STPR), ST iron (STI), ST sodium (STS), ST potassium (STPO), ST cholesterol (STCH), ST monounsaturated fatty acids (STMUFA), ST polyunsaturated fatty acids (STPUFA), LTL protein (LTLPR), LTL iron (LTLI), LTL sodium (LTLS), LTL potassium (LTLPO), LTL cholesterol (LTLCH), LTL monounsaturated fatty acids (LTLMUFA), and LTL polyunsaturated fatty acids (LTLPUFA) dSTCH, STMUFA, STPUFA, LTLCH, LTLMUFA, LTLPUFA units as mg/100g of total wet tissue. ePhenotypic correlations (P value). fStandard errors for correlations were 0.067. Literature Cited Hollenberg, N. K. 1980. Set point for sodium homeostasis: Surfeit, deficit, and other implications. Ahlberg, C. M., L. N. Schiermiester, J. T. Howard, C. Kidney Int. 17: 423-429. Calkins, M. L. Spangler. 2014. Genome wide Howard, J. T., S. D. Kachman, M. K. Nielsen, T. L. association study of cholesterol and poly- and Mader, & M. L. Spangler. 2013. The effect monounsaturated fatty acids, protein, and of myostatin genotype on body temperature mineral content of beef from crossbred cattle. during extreme temperature events. J. Anim. Meat Sci., (under review). Sci. 91: 3051-3058. Aldai, N., B. E Murray, M. Olivan, A. Martinez, D.J. Illumina, Inc. 2010. Technical Note: DNA Analysis, Troy, K. Osoro, & A. I. Najera. 2005. The in- Infinium® Genotyping Data Analysis. fluence of breed andmh -genotype on carcass San Diego (CA). Illumina, Inc., 9 p. Pub. No. conformation, meat physic-chemical charac- 970-2007-005. teristics, and the fatty acid profile of muscle from yearling bulls. Meat Sci. 72: 486-495. Kim, T., S. Shin, J. Park, & C. Park. 2013. Genome wide identification and expression profile in Cameron N.D. 1999. Genetic and phenotypic parame- epithelial cells exposed to TiO particles. J. ters for fatty acid composition of subcutaneous 2 Environ.Toxicol. 29: 4. fat, meat quality, carcass composition and eating quality traits in pigs. Proc. 4th WCGALP, Mateescu, R. G., A. J. Garmyn, R. G. Tait, Jr., Q. Edinburgh, UK, XV: 549–552. Duan, Q. Liu, M. S. Mayes, A. L. Van Eenen- naam, D. L. VanOverbeke, G. G. Hilton, De Smet, S., K. Raes, & D. Demeyer. 2004. Meat D. C. Beitz, & J. M. Reecy. 2013a. Genetic fatty acid composition as affected by fatness parameters for concentrations of minerals in and genetic factors: a review. Anim. Res. 53: longissimus muscle and their associations with 81-98. palatability traits in Angus cattle. J. Anim. Sci. Duan, Q., J. M. Reecy, R. Tait, J. P. Schoonmaker, D. 91: 1067-1075. C. Beitz, A. L. Van Eenennam, R. Mateescu, Mateescu, R. G., A. J. Garmyn, R. G. Tait, Jr., Q. & A. Garmyn. 2009. Phenotypic variation of Duan, Q. Liu, M. S. Mayes, A. L. Van Eenen- mineral contents in beef. Fed. Am. Soc. Exp. naam, D. L. VanOverbeke, G. G. Hilton, Biol. J. 23: LB412. D. C. Beitz, & J. M. Reecy. 2013b. Genetic Dunner, S., N. Sevane, D. Garcia, H. Leveziel, J. L. parameters for concentrations of minerals in Williams, B. Mangin, A. Valentini, & GeM- longissimus muscle and their associations with Qual Consortium. 2013. Genes involved in palatability traits in Angus cattle. J. Anim. Sci. muscle lipid composition in 15 European Bos 91: 1067-1075. taurus breeds. Animal Genetics 44: 493-501. Mateescu, R. G., A. J. Garmyn, R. G. Tait, Jr., Q. Fernando, R. L., & D. J. Garrick. 2009. GenSel User Duan, Q. Liu, M. S. Mayes, A. L. Van Eenen- manual for a portfolio of genomic selection naam, D. L. VanOverbeke, G. G. Hilton, D. C. related analysis. Animal Breeding and Genet- Beitz, & J. M. Reecy. 2013c. Genome-wide ics, Iowa State University, Ames. association study of concentrations of iron Habier, D., R. L. Fernando, K. Kizilkaya, & D. J. and other minerals in longissimus muscle of Garrick. 2011. Extension of the Bayesian Angus cattle. J. Anim. Sci. 91: 3593-3600. alphabet for genomic selection. BMC Bioin- NRC. 1996. Nutrient Requirements of Beef Cattle. formatics 12:186. 7th Revised Edition, 1996: Update 1996. Na- He, Q., & D. Lin. 2010. A variable selection method tional Academy Press, Washington, D. C. for genome-wide association studies. J. Bioin- Pitchford, W. S., M. P. B. Deland, B. D. Siebert, A. E. formatics 27: 1-8. O. Malau-Aduli, & C. D. K. Bottema, 2002. Genetic variation in fatness and fatty acid composition of crossbred cattle. J. Anim. Sci. 80: 2825–2832. 125 Raes, K., S. De Smet, & D. Demeyer, 2001. Effect of BREEDING FOR REDUCED ENVIRONMENTAL double-muscling in Belgian Blue young bulls FOOTPRINT IN BEEF CATTLE on the intramuscular composition with empha- Donagh P. Berry1 sis on conjugated linoleic acid and polyunsatu- 11 rated fatty acids. Animal Science 73: 253-260. Animal & Grassland Research and Innovation Cen- tre Realini, C. E., S. K. Duckett, G. W. Brito, M. D. Rizza, D. De Mattos, 2004. Effect of pasture Teagasc, Moorepark, Ireland vs. concentrate feeding with or without anti- Introduction oxidants on carcass characteristics, fatty acid composition, and quality of Uruguayan beef. There is considerable commentary in recent years on Meat Sci. 74: 567-577. “sustainable intensification”. The world human population is expanding and becoming more affluent and thus Scollan, N. D., J. Hocquette, K. Nuernberg, D. Dan- the demand for animal-derived protein (i.e., milk and nenberger, I. Richardson, & A. Moloney. meat) is increasing. Global production of meat from 2006. Innovations in beef production systems livestock is expected to double over the fifty years from that enhance the nutritional and health value 1999/2001 (229 million tonnes) to 2050 (465 million of beef lipids and their relationship with meat tonnes; Steinfeld et al., 2006). This therefore implies quality. Meat Sci. 74: 17-33. that, all else being equal, the environmental footprint Sexten, A. K., C. R. Krehbiel, J. W. Dillwith, R. D. per unit meat produced will need to be halved if the Madden, C. P. McMurphy, D. L. Lalman,R. total environmental footprint of livestock production G. Mateescu. 2012. Effect of muscle type, is not to increase. Most commentary on livestock sire breed, and time of weaning on fatty acid environmental footprint, however, tends to focus on composition of finishing steers.J. Anim. Sci. greenhouse gas emissions. O’Mara (2011) stated that 90: 616-625. animal agriculture is responsible for 8.0 to 10.8% of Uemoto, Y., T. Abe, N. Tameoka, H. Hasebe, K. global greenhouse gas emissions. If, however, complete Inoue, H. Nakajima, N. Shoji, M. Kobayashi, lifecycle analysis (i.e., accounting for the production of and E. Kobayashi. 2011. Whole-genome inputs to animal agriculture as well as change in land association study for fatty acid composition use such as deforestation) is undertaken this figure of oleic acid in Japanese Black cattle. Animal can be up to 18%. Cattle are the largest contributors to Genetics, 42: 141–148. global greenhouse gas emissions (O’Mara, 2011). Wiener, P., J. A. Wooliams, A. Frank-Lawale, M. Livestock production systems are, nonetheless, also Ryan, R. I. Richardson, G. R. Nute, & J. D. implicated for pollution of freshwater supplies (e.g., ni- Wood. 2009. The effects of a mutation in the trogen and phosphorus) as well as depleting water re- myostatin gene on meat and carcass quality. serves. Livestock is implicated for 32-33% of Nitrogen Meat Sci. 83: 127-134. and Phosphorus contamination of freshwater supplies (Steinfeld et al., 2006). Moreover, 64% of the world’s Wood, J.D., M. Enser, A.V. Fisher, G. R. Nute, P. R. population is expected to reside within water-stressed Sheard, R. I. Richardson, S. I. Hughes, F. M. areas by the year 2025 (Steinfeld et al., 2006). Livestock Whittington. 2008. Fat deposition, fatty acid production accounts for 8% of the water used by the composition and meat quality: A review. Meat human population (Steinfeld et al., 2006). Thus water Sci. 78: 343-358. use efficiency in animal production, as well as pollutant Zimin, A. V., A. L. Delcher, L. Florea, D. R. Kelley, potential of water supplies, is also a crucial characteris- M. C. Schatz, D. Puiu, F. Hanrahan, G. Pertea, tic of animals for environmental footprint. C. P. van Tassell, T. S. Sonstegard, G. Marcais, M. Roberts, P. Subramanian, J. A. Yorke, & S. Much research is focusing particularly on differences L. Salzberg. 2009. A whole-genome assembly of the domestic cow, Bos taurus. Genome Biol. 10(4): R42.

126 among individuals in greenhouse gas emissions. Here I EBI was criticized at the time for not accurately reflect- discuss other, often easier and more holistic approach- ing the prevailing market signals. In 2006/2007 how- es to potentially reduce the environmental footprint of ever, the milk payment system in Ireland changed to modern day cattle production systems while simulta- be strongly reflective of the relative economic weights neously improving profitability. This article should be in the EBI; thus the EBI had been identifying the most viewed more for provoking discussion than a definitive suitable germplasm for this payment system for the solution to how best to reduce the environmental foot- previous 5-6 years. Irish beef breeding programs will print of modern-day production systems. soon include EPDs for carcass cuts with the anticipa- Animal breeding programs may be summarized graph- tion that the carcass payment system will change in the ically as in Figure 1. near future to better reflect carcass quality. Moreover, research is also underway on the inclusion of meat quality traits in the Irish national beef index, again in anticipation of financial incentives for superior meat quality in the future; the precedence already exists through incentives for meat from Angus cattle. Therefore goals of breeding programs should be expanded to not only include profit but to do so in an “environmentally and socially responsible and sustainable manner”. Although difficult, cognizance must be given to the likely policy enforced in several years. Breeding objective The breeding objective lists traits and their respective Figure 1. Schematic of an animal breeding program. relative weightings to best describe the goal of the breeding program. Such traits should include revenue Goal generating traits (e.g., carcass yield and value) as well The goal of most cattle production systems in the as cost of production traits (e.g., feed intake, health developed world is profit. Profit is dictated by revenue and survival and in the case of maternal breeding and cost of production. Some traits however current- objectives reproduction and longevity). Ideally such ly have no monetary value in most countries but are objectives should also include direct environmental deemed to have “public good” attributes. Moreover, characteristics like daily (or lifetime) methane emis- animal breeding strives to identify and select germ- sions, nitrogen (and other minerals) excretion, as well plasm that will be most profitable in several years. as water intake. It is vitally important at this stage that Thus, although some attributes may have little (e.g., cognizance is not given to whether or not these traits water) or no (e.g., greenhouse gases) current monetary can be (easily) measured. Such details will be resolved value in most countries, the same may not be true in in the later steps of the breeding program. If the ge- the future when the (grand-)progeny of the animals se- netic variation present in a trait cannot be adequately lected today will be producing. A good example is the captured then, if not deemed sufficiently important, it evolution of the milk payment system in Ireland. The can be discarded in the iterative process of the breed- Irish national dairy cow breeding objective, the EBI, ing program (Figure 1). launched in 2001 penalized higher producing animals The relative weighting on each trait in the breeding with lower milk composition. This was during a time goal can be derived using several approaches including when Irish producers were paid on a differential milk economic values (i.e., from bioeconomic models or pricing system with no penalty for milk volume. The

127 profit functions), choice experiments or willingness to is being achieved in holistic breeding objectives; this pay experiments (e.g., 1,000 minds) or desired gains therefore is expected to reduce the environmental approaches. The contribution however of increased an- footprint of the growing cattle sector but cognizance imal performance to reduced environmental footprint must of course be taken of the cattle production sys- of the entire production system must be recognized. tem in its entirety (i.e., cow-calf production system). Goddard et al. (2011) defined herd feed conversion Table 1 describes the national breeding objectives efficiency (FCE) for a beef herd as: for beef cattle in Ireland; the genetic evaluations are undertaken across breed and there is a single national breeding objective which operates across all breeds. W (wean − loss) Herd FCE = Off The breeding objectives have a positive weight on car- DMI + wean ⋅ DMI cass weight (i.e., reducing age at slaughter for a fixed Cow Off carcass weight) and a negative economic weight on feed intake. The expected responses to selection based where Woff is the slaughter weight of the offspring, on the terminal index are in Figure 1. Gains in carcass wean is the weaning rate, loss is the cow loss rate, weight (i.e. earlier age at slaughter for the same carcass DMIcow is the total feed intake of the cow, and DMI- weight) are expected despite an expected reduction in off is the total feed intake of the offspring. This clearly daily feed intake. This therefore is a double-whammy shows that factors other than feed intake or direct of reduced feed intake per day and reduced number of environmental measures such as fertility (i.e., weaning days of feeding. This is clearly exemplified by the mean rate) and cow loss rate can also affect herd efficiency performance of slaughtered animal divergent for the and thus environmental footprint. Of key importance Irish terminal index (Table 2; Connelly et al., 2014); here is that DMI reflects total DMI and not daily DMI. genetic merit for each animal was based on a genet- Average daily DMI is almost always used in the defi- ic evaluation that did not include the animal’s own nition of feed efficiency traits like residual feed intake performance record. The genetically elite animals were (RFI; Berry and Crowley, 2013). Berry and Crowley slaughtered 54 days younger despite their carcasses (2012) however clearly demonstrated that animals weighing 17% more than the lowest genetic merit superior for RFI, although eating less per day, may group. Moreover, the EBVs for daily feed intake of the require a longer period of time to reach a target weight highest and lowest genetic merit group were -0.08 kg/ and thus eat more during this finishing period com- day and 0.48 kg/day, respectively. These characteristics pared to animals ranked on their proposed index trait combined suggest that not only do the genetically elite which included growth rate. A similar conclusion was animals eat food (and therefore require less associated reached in poultry (Willems et al., 2013). Although the labour and capital costs) for 54 days less but they are direct translation to reduced environmental footprint also eating potentially more than half a kg less per day is not clear, it is logical to assume that animals with less than their genetically inferior counterparts. These lower total feed intake are also likely to have a reduced characteristics are likely to result in a lower environ- environmental footprint. This is because feed intake mental footprint of these genetically elite animals; it and daily methane emissions are positively correlated is important to remember that this is being achieved (Fitzsimons et al., 2013) and there is an expectation without any direct inclusion of an environmental therefore also that lower feed intake (achieved through trait in the breeding goal. Further reductions in en- genetic gain without compromising performance), on vironmental footprint are no doubt possible with the average, results in less water intake, as well as less feces direct inclusion of environmental trait in the breeding and uterine produced.. objective but such inclusions will likely come at a cost and it is currently not clear what marginal gains could The dual objective of reducing feed intake per-day and actually be achieved by such endeavors. number of days on feed (i.e., growth rate) is what Improving cow fertility and longevity can reduce the environmental load of the entire beef production system as described in the equation above. Garnsworthy (2004) documented, using modelling, that if dairy cow

128 fertility in the UK national herd could be restored to 1995 level from 2003 levels then herd methane emissions could be reduced by 10 to 11% while ammonia emissions could be reduced by 9% under a milk quota environment; the respective reductions were 21 to 24% and 17% if ideal fertility levels were achieved. A reduction of 4 to 5% in herd methane emissions was expected in the UK if fertility levels were restored to 1995 levels from 2003 levels where no milk quota ex- isted (Garnsworthy, 2004). These improvements were due primarily to a reduced number of non-producing Figure 1. Expected annual gain in genetic standard replacement animals and to a lesser extent greater milk deviation units (assuming an annualized genetic gain yield (i.e., in beef would result in greater calf growth of 0.15 standard deviation units) for direct calving rate) when fertility was improved. No cognizance was difficulty (DCD), gestation length (GEST), perinatal taken here of the impact of replacement rate on genetic mortality (MORT), docility, dry matter intake (DMI) gain. carcass weight (Ccwt), carcass conformation (Cconf) and carcass fat (Cfat), assuming 100 progeny records for the calving traits and docility, 6 progeny records for feed intake and 85 progeny records for the three carcass traits.

Daily methane emissions per animal were sampled from a normal probability distribution with a mean of 300 g/day and a standard deviation of 40 g/day. Meth- ane intensity was defined as daily methane emissions divided by actual recorded daily feed intake available on those animals. As expected the heritability of the simulated daily methane emissions was zero; the heritability of feed intake was 0.49 (Crowley et al., 2010). The heritability of methane intensity was 0.19 (0.05). Berry (2012) proposed that to measure the potential of genetic selection to alter methane emissions without compromising performance, a statistical approach analogous to that used to define residual feed intake Table 1. Relative emphasis on traits in the Irish nation- (Koch et al., 1963), should be used. This trait may be al beef maternal and terminal breeding objectives termed residual methane production (RMP) and could be defined as the residuals from a model regressing Many studies focus on methane intensity as a breed- individual animal daily methane emission on energy ing goal trait. Methane intensity may be described as sinks like growth rate and metabolic live-weight. Feed the total (daily) methane output per unit feed intake. intake may also be included as covariate in the statis- Heritable genetic variation in methane intensity has tical model. If this approach was used in the example been reported (Donoghue et al., 2013) but Berry above, then the heritability of the residual methane (2012) cautioned strongly on the interpretation of such trait was, as expected, zero. It is the genetic variation heritability estimates as it is unclear what proportion in this RMP trait that is of crucial importance as this of the heritability originates from the numerator or depicts the scope for genetic improvement while still denominator of the intensity equation. Berry (2012) continuing to produce meat for the growing human used a dataset of 2,605 growing beef bulls, described in demand. The heritability of this statistic merely de- detail by Crowley et al. (2010), to justify his concerns.

129 scribes how much one would have to invest to generate could explain 72% of the genetic variation in daily feed accurate genetic evaluations for this trait for individual intake in growing cattle from live-weight, growth rate animals. A similar approach should be undertaken for and ultrasound fat measures all of which are relatively other environment traits like water use efficiency and easily measurable. The proportion of genetic variation nitrogen use efficiency. increased to 90% when a subjective measure of muscularity was also included in the selection index; this is Selection criterion because RFI and muscularity are genetically correlated Traits and their respective weights in the selection cri- (Berry and Crowley, 2013). There is no denying that terion are chosen to maximize the correlation between variation in RFI exists, my question is how much is the overall criterion and the overall breeding objec- there that cannot be captured through other means. It tive. Many are investing in high-tech facilities for the is very likely that a larger proportion of the variation accurate measurement of feed intake (and efficiency) total animal intake can be captured with easy to record as well and environmental traits (and other non-envi- traits since total feed intake will also be determined by ronmental traits). To my knowledge days on feed. Moreover, animals not at the feed bunk are simply not eating. RFID is now routinely used in a detailed peer-reviewed cost-benefit of such endeav- many feedlots, and if not is relatively inexpensive. Al- ors, taking cognizance of selection index theory, has though differences in bite rate and bite size among ani- not be undertaken. One must remember that current mals exists (Chen et al., 2014), total time spent feeding carcass trait genetic evaluations (if the country actually (from using sensors at the feed face and transponders has one!) are not perfect. Most countries use imprecise on animals) must explain additional genetic variation approaches to predict actual carcass value through in RFI. (Robinson and Oddy, 2004; Chen et al., 2014). themeasurement of carcass conformation which does The difference between metabolizable energy intake not directly take cognizance of individual meat cut and net energy intake is heat increment. Promising yields, let alone meat quality. Therefore, why such an research from Guelph (Montanholi et al., 2009) sug- emphasis on attempting to generate extremely accurate gests that measurement of animal heat produced from measures for other traits? I am not saying it is incor- simple infra-red cameras can be used to predict RFI. rect, but at least the true cost-benefit should be eluci- Some will say that 70% prediction accuracy is too low; dated and a discussion should be had. Such an exercise however the genetic correlation between carcass con- must take cognizance of the ability to predict some of formation and total meat yield (adjusted to a common these traits, with reasonable accuracy, using selection carcass weight) is 0.55 (Pabiou et al., 2009) implying index theory. Therefore, of real importance is what “re- that the current carcass grading system in the EU ex- sidual” variation in the trait of interest remains that is plains only 30% of the genetic variation in carcass meat not already captured by other easy to record traits. Ber- proportion! Would resources not be better spent on

Index Age (days) Carcass Conformation Fat Price per kg Value weight (kg) (scale 1 – 15) (scale 1-15) (€/kg) (€) Very High 726 371 8.68 (R+) 6.33 (3=) 3.85 1412 High 775 327 5.04 (O=) 6.40 (3=) 3.60 1174 Low 779 321 4.97 (O=) 6.44 (3=) 3.60 1153 Very Low 780 316 4.88 (O=) 6.33 (3=) 3.57 1123

SE 0.89 0.31 0.01 0.97 0.24 1.65

Table 2. Association between terminal EBV (Index) and age at slaughter, carcass weight, conformation and fat score as well as price per kg and overall animal value; pooled standard error (SE) also included. 130 improving this and selecting animals with more meat production, total water intake) in the breeding objec- yield to feed the growing human demand? tive but the adjusted trait (either as EPDs or categorized as high, average, or low depending on the accura- Breeding Scheme Design cy of the EPDs) as a stand-alone trait. By categorizing The breeding scheme design incorporates the genet- traits (or the stand alone trait as a monetary value like ic and genomic evaluations as well as the breeding feed cost saved) issues with which sign is desirable scheme used to ensure long-term and sustainable (i.e., apparently negative RFI) is removed. One could genetic gain. There is an expectation among some that simply change the sign but this will cause confusion if genomic selection will solve all issues in beef cattle (international) scientists are discussing with producers breeding. On the contrary, genomic selection can actu- since they will subconsciously say that genitive RFI is ally exacerbate any issues that exist in a breeding pro- better. By categorizing, issue with fluctuation EPDs gram. For example, genomic selection is expected to because of low reliability will be minimized. A similar increase genetic gain approximately 50% implying that categorization of traits is undertaken in Ireland for the rate of genetic deterioration in a given (non-mon- beef cattle where animals are grouped into 5 categories itored) trait will also likely increase by 50%. Moreover, (termed stars in Ireland) where the top category (i.e., although genomics can be used to reduce the accu- 5-star) are animals in the top 20% for genetic merit for mulation of inbreeding, in most instances in dairying, that trait. Although knowledge if the animal resides in inbreeding is increasing as breeding companies battle 1% percentile or the 19% is useful, getting producers to to increase the rate of short term genetic gain. The use and engage with animal breeding may actually be greater use of young bulls may minimize the ability, more beneficial. or increase the difficulty, to purge out unfavourable Dissemination characteristics. Furthermore, inaccurate, imprecise or non-pertinent genetic evaluations will not be solved Arguably the link in a successful animal breeding pro- with genomic selection. The input variables for genom- gram that is most often ignored is dissemination. There ic selection are either (a derivative) of EPDs from the is not much point having the best genetic evaluation genetic evaluation systems (i.e., two step) or the direct system and breeding program in the world if nobody phenotypes themselves with the genetic relevant evalu- understands it or is willing to use the elite germplasm. ation model (i.e., one-step). Therefore, implementation Animal breeders find it difficult to understand why of genomic selection will be most optimal once the the best germplasm is not used; even if individual bull fundamentals of a successful animal breeding program reliability is low, on average, if producers use the elite are in place. bulls the entire population will make gains. However, individual producers are more concerned with theper- There have been long discussions on how best to in- formance of their own animals and herd rather than corporate feed efficiency in a breeding program (Berry the national population. It is still remarkable how and Crowley, 2013) and to-date no consensus exists. many producers globally do not believe genetic evalua- Table 3 (Berry and Pryce, 2014) outlines the advantag- tions. One has to question the investment in genomics es of disadvantages of including a residual feed intake to produce more accurate EPDs when the EPDs are or dry matter intake itself in the breeding objective. sometimes not even used in the first place. Of course The same discussions are likely to prevail for environ- genomics will increase the accuracy of these genetic mental footprint traits especially if residual-based traits evaluations but resources must be put into explaining are derived. In other words should residual methane and demonstrating the impact of genetic differences production, total daily methane production, or meth- on phenotypic performance. Dairy cattle breeders did ane intensity be included in the breeding objective or an excellent job in convincing (mostly) non-geneti- as a stand-alone trait. The disadvantages of selection cists that breeding can actually improve reproductive on ratio traits (i.e., methane intensity) like feed conver- performance. This was achieved (eventually!) through sion efficiency has been discussed at length (Berry and demonstration, not structured demonstration, but be- Crowley, 2013) suggesting that methane intensity (or cause of widespread use of elite genetics in dairying the any other environment trait like water intake per unit results across so many herds were impossible to ignore. average daily gain or per unit feed intake) may be not Nonetheless, controlled experiments, although costly,

131 Table 3. Reasons in favor and against including DMI or RFI in a breeding goal

(Buckley et al., 2014) and beef (Prendiville et al., 2014). tailed environmental footprint of animals selected to Controlled experiments on feed efficiency and the be genetically divergent for a given selection strategy mean methane emissions per stratum also exist (Nkru- can be very useful in elucidating the impact of cur- mah et al., 2006). rent breeding strategies on expected genetic trends in environmental footprint. Moreover the ideal reference Structured demonstration herds or research herds population for accurate genomic evaluations should can also be informative for breeding to reduce envi- be genomically and phenotypically diverse (as well as ronmental footprint. Because the routine capture of related to the candidate population of animals). Ani- data on most direct environmental traits for genetic mals divergent for the breeding strategy employed can evaluations can be expensive, it may not make eco- therefore be very useful for the development of genomic sense to collect such information. Estimating nomic predictions. This is especially true for difficult the impact of current breeding strategies on genetic to measure traits such as direct environmental traits. change in environmental traits can be achieved using selection index theory. However, procuring sufficient Economic analysis data to estimate precise genetic parameters can also be Animal breeders (either academic or seedstock pro- costly. Evaluating in a controlled environment, the de- 132 ducers) must not be afraid to discontinue certain Campion, B., Keane M.G., Kenny D.A., and Berry D.P. paths if it is not economically advantageous or if more 2009. Evaluation of estimated genetic merit for car- economic gain can be realized with a different strat- cass weight in beef cattle: Live weights, feed intake, egy. Such economic analyses however must include body measurements, skeletal and muscular scores, long-term impact, discounted to current day equiva- and carcass characteristics. Livest. Sci. 126: 87-99 lents. Economic analyses of breeding programs can be Chen L, Mao F., Crews D.H., Jr., Vinsky M., and Li, C. undertaken at the producer level, the breeding com- 2014. Phenotypic and genetic relationships of feed- pany level, or at the national/global level. Moreover, ing behavior with feed intake, growth performance, as previously alluded to, the economic cost of most feed efficiency, and carcass merit traits in Angus and environmental traits can be difficult to quantify unless Charolais steers. J. Anim. Sci. 92:974-983; there is some financial incentive (e.g., carbon trading) Clarke, A.M., Drennan M.J., McGee M., Kenny D.A., or penalty (e.g., nitrates directive) for same. Many of Evans R.D., and Berry D.P.. 2009. Intake, growth and the benefits of reduced environmental footprint of cat- carcass traits in male progeny of sires differing in tle production systems will be realized at the national genetic merit for beef production. Animal 3:791-801 or even global level. Research in this area is on-going (Wall et al., 2010) Coleman, J., Berry D.P., Pierce K.M., Brennan A., and Horan, B. 2010. Dry matter intake and feed effi- Conclusions ciency profiles of 3 genotypes of Holstein-Friesian within pasture-based systems of milk production. J. Environmental footprint of modern-day production Dairy Sci. 93:4318-4331 systems will undoubtedly become more important in the near future as global food production increase Connolly, S.M., Cromie A.R., and Berry D.P. 2014. and the ramifications of such are contemplated. Many Genetic differences in beef terminal traits and Index approaches exist to possibly reduce the environmental is reflected in phenotypic performance difference footprint of animal production systems. Animal breed- in commercial beef herds. Proc. World Cong. Gen. ing has the advantage of being cumulative and per- Appl. Lives. Prod.. Vancouver. manent; the main disadvantage of the long generation Crowley, J.J., McGee M., Kenny D.A., Crews D.H. Jr, interval in breeding is being ameliorated with the ad- Evans R.D., and Berry D.P. 2010. Phenotypic and vent of genomic selection. Nonetheless, the alternative genetic parameters for different measures of feed strategies in the animal breeder’s toolbox to achieve the efficiency in different breeds of Irish performance objective of reduced environmental footprint of animal tested beef bulls. J. Anim. Sci. 88:885-894 production without compromising animal perfor- Donoghue K.A., Herd R.M., Bird S.H., Arthur P.F. and mance must be thoroughly investigated taking cogni- Hegarty R.G. 2013. Preliminary genetic parameters zance of the cost of each strategy. for methane production in Australian beef cattle. Literature Cited Proc. Assoc. Advancement Anim. Breed. Gen. 20:290- 293 Berry, D.P., and Crowley, J.J. 2012. Residual intake and Goddard, M.E., Bolormaa S., and Savin K. 2011. Selec- gain; a new measure of efficiency in growing cattle. J. tion for feed conversion efficiency in beef cattle. In; Anim. Sci. 90:109-115 recent advances in Animal Nutrition 18, Australia. Berry, D.P. and Crowley, J.J. 2013 Genetics of feed effi- Ed. P. Cronje. University of New England. Australia ciency in dairy and beef cattle. J. Anim. Sci. 91:1594- Fitzsimons, C., Kenny D.A., Deighton M.H., Fahey 1613 A.G. and McGee M. 2013. Methane emissions, body Berry, DP and Pryce J.E. 2014. Feed Efficiency in composition, and rumen fermentation traits of beef Growing and Mature Animals. Proc. World Cong. heifers differing in residual feed intake. Journal of Gen. Appl. Livest. Prod. Animal Science. 91: 5789-5800 Buckley, F., McParland S. and Brennan A. 2014. The Koch, R.M., Swiger, L.A., Chambers, D. and Grego- Next Generation Herd –Year 1 results. Moorepark ry, K.E. 1963. Efficiency of feed use in beef cattle. J. Open Day Booklet. 9th April 2014. Moorepark, Ire- Anim. Sci. 22: 486-494 land. 133 Montanholi Y.R., Swanson K.C., Schenkel F.S., McBride ACROSS-BREED EPD TABLES FOR B.W., Caldwell T.R., Miller S.P.. 2009. On the deter- THE YEAR 2014 ADJUSTED TO mination of residual feed intake and associations BREED DIFFERENCES FOR BIRTH YEAR of infrared thermography with efficiency and ultra- OF 2012 sound traits in beef bulls. Livest. Sci. 125: 22–30 L. A. Kuehn1 and R. M. Thallman1 Nkrumah, J.D., Okine E.K., Mathison G.W., Schmid K., 1Roman L. Hruska U.S. Meat Animal Research Center, Li C., Basarab J.A., Price M.A., Wang Z. and Moore USDA-ARS, Clay Center, NE 68933 S.S.. 2006. Relationships of feedlot feed efficiency, performance, and feeding behavior with metabolic Summary rate, methane production, and energy partitioning in Factors to adjust the expected progeny differ- beef cattle. J. Anim. Sci. 84: 145-153 ences (EPD) of each of 18 breeds to the base of Angus EPD are reported in the column labeled 6 of Tables O’Mara, F.P. 2011 The significance of livestock as a 1-7 for birth weight, weaning weight, yearling weight, contributor to global greenhouse gas emissions today maternal milk, marbling score, ribeye area, and fat and the near future. Anim. Feed Sci. Tech. 166-167: thickness, respectively. An EPD is adjusted to the 7-15. Angus base by adding the corresponding across-breed Robinson, D.L. and Oddy V.H. 2004. Genetic param- adjustment factor in column 6 to the EPD. It is critical that this adjustment be applied only to Spring 2014 eters for feed efficiency, fatness, muscle area and feed- EPD. Older or newer EPD may be computed on dif- ing behaviour of feedlot finished beef cattle. Livest. ferent bases and, therefore, could produce misleading Prod. Sci. 90: 255–270. results. When the base of a breed changes from year to Pabiou, T., Fikse, W.F., Nasholm, A., Cromie, A.R., year, its adjustment factor (Column 6) changes in the Drennan, M.J., Keane, M.G. and Berry, D.P. 2009. Ge- opposite direction and by about the same amount. netic parameters for carcass cut weight in Irish beef Breed differences are changing over time as cattle. J. Anim. Sci. 87: 3865-3876 breeds put emphasis on different traits and their genetic trends differ accordingly. Therefore, it is necessary Steinfeld, H, Gerber P, Wassenaar T., Castel V., Rosales to qualify the point in time at which breed differenc- M., and de Haan C.. 2006.Livestock’s Long Shad- es are represented. Column 5 of Tables 1-7 contains ow. The Livestock, Environment and Development estimates of the differences between the averages of Initiative (LEAD). Rome: FAO. http://www.fao.org/ calves of each breed born in year 2012. Any differ- docrep/010/a0701e/a0701e00.HTM ences (relative to their breed means) in the samples of Wall E., Simm G. and Moran D. 2010. Developing sires representing those breeds at the U.S. Meat Ani- mal Research Center (USMARC) are adjusted out of breeding schemes to assist mitigation of greenhouse these breed difference estimates and the across-breed gas emissions. Animal 4: 366-376. adjustment factors. The breed difference estimates are Willems O.W., Miller S.P. and B.J. Wood. 2013. Assess- reported as progeny differences, e.g., they represent ment of residual body weight gain and residual intake the expected difference in progeny performance of and body weight gain as feed efficiency traits in the calves sired by average bulls (born in 2012) of two turkey (Meleagris gallopavo). Gen. Sel. Evol. 2013: 45: different breeds and out of dams of a third, unrelated 26. breed. In other words, they represent half the differences that would be expected between purebreds of the two breeds. Introduction This report is the year 2014 update of estimates of sire breed means from data of the Germplasm Evaluation (GPE) project at USMARC adjusted to a year 2012 basis using EPD from the most recent national cattle evaluations. The 2012 basis year is chosen because yearling records for weight and carcass traits should have been accounted for in EPDs for progeny born in 2012 in the Spring 2014 EPD national genetic evaluations. Factors to adjust Spring 2014 EPD of

134 18 breeds to a common base were calculated and are of origin. Due to lack of pedigree and different selec- reported in Tables 1-3 for birth weight (BWT), wean- tion histories, dams mated to the AI sires and natural

ing weight (WWT), and yearling weight (YWT) and service bulls mated to F1 females were also assigned in Table 4 for the maternal milk (MILK) component of to separate genetic groups (i.e., Hereford dams were maternal weaning weight (MWWT). Tables 5-7 sum- assigned to different genetic groups than Hereford AI marize the factors for marbling score (MAR), ribeye sires). Cows from Hereford selection lines (Koch et area (REA), and fat thickness (FAT). al., 1994) were used in Cycle IV of GPE and assigned into their own genetic groups. Through Cycle VIII, The across-breed table adjustments apply only most dams were from Hereford, Angus, or MARCIII to EPD for most recent (spring, 2014) national cat- (1/4 Angus, 1/4 Hereford, 1/4 Pinzgauer, 1/4 Red Poll) tle evaluations. Serious errors can occur if the table composite lines. In order to be considered in the anal- adjustments are used with earlier or later EPD which ysis, sires had to have an EPD for the trait of interest. may have been calculated with a different with- All AI sires were considered unrelated for the analysis in-breed base. in order to adjust resulting genetic group effects by the The following describes the changes that have average EPD of the sires. occurred since the update released in 2013 (Kuehn and Fixed effects in the models for BWT, WWT Thallman, 2013): (205-d), and YWT (365-d) included breed (fit as New samplings of sires in the USMARC GPE genetic groups) and maternal breed (WWT only), program continued to increase progeny records for year and season of birth by GPE cycle by age of dam all of the breeds. The GPE program has entered a (2, 3, 4, 5-9, >10 yr) combination (255), sex (heifer, new phase in which more progeny are produced from bull, steer; steers were combined with bulls for BWT), breeds with higher numbers of registrations. Breeds a covariate for heterosis, and a covariate for day of with large increases in progeny numbers as a per- year at birth of calf. Models for WWT also included a centage of total progeny included South Devon and fixed covariate for maternal heterosis. Random effects Tarentaise (especially for yearling weight) and Santa included animal and residual error except for the anal- Gertrudis and Chiangus (especially for maternal milk). ysis of WWT which also included a random maternal However, all of the breeds will continue to produce genetic effect and a random permanent environmental progeny in the project and sires continue to be sam- effect. pled on a continuous basis for each of the 18 breeds in For the carcass traits (MAR, REA, and FAT), the across-breed EPD program. These additional prog- breed (fit as genetic groups), sex (heifer, steer) and eny improve the accuracy of breed differences estimat- slaughter date (265) were included in the model as ed at USMARC (column 3 in Tables 1-7) particularly fixed effects. Fixed covariates included slaughter age for breeds with less data in previous GPE cycles (e.g., and heterosis. Random effects were animal and resid- South Devon, Tarentaise, Santa Gertrudis, Chiangus). ual error. To be included, breeds had to report carcass Materials and Methods EPD on a carcass basis using age-adjusted endpoints, All calculations were as outlined in the as suggested in the 2010 BIF Guidelines. 2010 BIF Guidelines. The basic steps were given The covariates for heterosis were calculated by Notter and Cundiff (1991) with refinements by as the expected breed heterozygosity for each animal Núñez-Dominguez et al. (1993), Cundiff (1993, 1994), based on the percentage of each breed of that animal’s Barkhouse et al. (1994, 1995), Van Vleck and Cundiff parents. In other words, it is the probability that, at any (1997–2006), Kuehn et al. (2007-2011), and Kuehn location in the genome, the animal’s two alleles origi- and Thallman (2012, 2013). Estimates of variance nated from two different breeds. Heterosis is assumed components, regression coefficients, and breed effects to be proportional to breed heterozygosity. For the were obtained using the MTDFREML package (Bold- purpose of heterosis calculation, AI and dam breeds man et al., 1995). All breed solutions are reported as were assumed to be the same breed and Red Angus differences from Angus. The table values of adjust- was assumed the same breed as Angus. For purposes ment factors to add to within-breed EPD are relative to of heterosis calculation, composite breeds were con- Angus. sidered according to nominal breed composition. For Models for Analysis of USMARC Records example, Brangus (3/8 Brahman, 5/8 Angus) Angus An animal model with breed effects represent- ´ is expected to have 3/8 as much heterosis as Brangus ed as genetic groups was fitted to the GPE data set ´ Hereford. (Arnold et al., 1992; Westell et al., 1988). In the anal- Variance components were estimated with a ysis, all AI sires (sires used via artificial insemination) derivative-free REML algorithm with genetic group were assigned a genetic group according to their breed 135 solutions obtained at convergence. Differences be- EPD from the breed associations. The basic calcula- tween resulting genetic group solutions for AI sire tions for all traits are as follows: breeds were divided by two to represent the USMARC USMARC breed of sire solution (1/2 breed breed of sire effects in Tables 1-7. Resulting breed solution) for breed i (USMARC (i)) converted to an differences were adjusted to current breed EPD levels industry scale (divided by b) and adjusted for genetic by accounting for the average EPD of the AI sires of trend (as if breed average bulls born in the base year progeny/grandprogeny, etc. with records. Average AI had been used rather than the bulls actually sampled): sire EPD were calculated as a weighted average AI sire EPD from the most recent within breed genetic Mi = USMARC (i)/b + [EPD(i)YY - EPD(i)USMARC]. evaluation. The weighting factor was the sum of rela- Breed Table Factor (Ai) to add to the EPD for a bull of tionship coefficients between an individual sire and all breed i: progeny with performance data for the trait of interest Ai = (Mi - Mx) - (EPD(i)YY - EPD(x)YY). relative to all other sires in that breed. where, For all traits, regression coefficients of progeny USMARC(i) is solution for effect of sire breed i performance on EPD of sire for each trait were calcu- from analysis of USMARC data, lated using an animal model with EPD sires excluded EPD(i)YY is the average within-breed 2014 EPD from the pedigree. Genetic groups were assigned in for breed i for animals born in the base year (YY, place of sires in their progeny pedigree records. Each which is two years before the update; e.g., YY = sire EPD was ‘dropped’ down the pedigree and re- 2012 for the 2014 update), duced by ½ depending on the number of generations EPD(i)USMARC is the weighted (by total relationship each calf was removed from an EPD sire. In addition of descendants with records at USMARC) average to regression coefficients for the EPDs of AI sires, of 2014 EPD of bulls of breed i having descen- models included the same fixed effects described dants with records at USMARC, previously. Pooled regression coefficients, and re- b is the pooled coefficient of regression of prog- gression coefficients by sire breed were obtained. eny performance at USMARC on EPD of sire These regression coefficients are monitored as accu- (for 2014: 1.16, 0.84, 1.05, and 1.11 BWT, WWT, racy checks and for possible genetic by environment YWT, and MILK, respectively; 1.00 was applied interactions. In addition, the regression coefficients by to MAR, REA, and FAT data), sire breed may reflect differences in genetic trends for i denotes sire breed i, and different breeds. The pooled regression coefficients x denotes the base breed, which is Angus in this were used as described in the next section to adjust for report. differences in management at USMARC as compared Results to seedstock production (e.g., YWT of males at US- Heterosis MARC are primarily on a slaughter steer basis, while Heterosis was included in the statistical model in seedstock field data they are primarily on a breeding as a covariate for all traits. Maternal heterosis was also bull basis). For carcass traits, MAR, REA, and FAT, fit as a covariate in the analysis of weaning weight. regressions were considered too variable and too far Resulting estimates were 1.41 lb, 13.83 lb, 20.51 lb, removed from 1.00. Therefore, the regressions were -0.04 marbling score units (i.e. 4.00 = Sl00, 5.00 = assumed to be 1.00 until more data is added to reduce Sm00), 0.26 in2, and 0.035 in for BWT, WWT, YWT, the impact of sampling errors on prediction of these MAR, REA, and FAT respectively. These estimates are regressions. However, the resulting regressions are interpreted as the amount by which the performance of still summarized. an F1 is expected to exceed that of its parental breeds. Records from the USMARC GPE Project are The estimate of maternal heterosis for WWT was 9.78 not used in calculation of within-breed EPD by the lb. breed associations. This is critical to maintain the Across-breed adjustment factors integrity of the regression coefficient. If USMARC Tables 1, 2, and 3 (for BWT, WWT, and YWT) records were included in the EPD calculations, the summarize the data from, and results of, USMARC regressions would be biased upward. analyses to estimate breed of sire differences on a Adjustment of USMARC Solutions 2012 birth year basis. The column labeled 6 of each ta- The calculations of across-breed adjustment ble corresponds to the Across-breed EPD Adjustment factors rely on breed solutions from analysis of re- Factor for that trait. Table 4 summarizes the analysis cords at USMARC and on averages of within-breed of MILK. Tables 5, 6, and 7 summarize data from the carcass traits (MAR, REA, FAT). Because of the accu- 136 racy of sire carcass EPDs and the greatest percentage The relatively heavy birth weights of Brahman-sired of data being added to carcass traits, sire effects and progeny would be expected to be offset by favorable adjustment factors are more likely to change for car- maternal effects reducing birth weight if progeny were cass traits in the future. from Brahman or Brahman cross dams which would Column 5 of each table represents the best esti- be an important consideration in crossbreeding pro- mates of sire breed differences for calves born in 2012 grams involving Brahman cross females. Changes in on an industry scale. These breed difference estimates breed of sire effects were generally small, less than 1.5 are reported as progeny differences, e.g., they repre- lb for all breeds relative to last year’s update (Kuehn sent the expected difference in progeny performance and Thallman, 2013). of calves sired by average bulls (born in 2012) of two Weaning Weight different breeds and out of dams of a third, unrelated All of the 17 breed differences (Table 2, breed. Thus, they represent half the difference expect- column 5) were within 6 lb of the values reported by ed between purebreds of the respective breeds. Kuehn and Thallman. (2013). Changes in breed effects In each table, breed of sire differences were for all 18 breeds seem to be stabilizing since continu- added to the raw mean of Angus-sired progeny born ous sampling started in 2007. 2009 through 2013 at USMARC (Column 4) to make Yearling Weight these differences more interpretable to producers on Breed of sire effects for yearling weight were scales they are accustomed to. also similar to Kuehn and Thallman (2013) in gener- Figures 1-4 illustrate the relative genetic al. South Devon and Tarentaise had the first yearling trends of most of the breeds involved (if they sub- weight records recorded in the GPE program; their mitted trends) adjusted to a constant base using the breed differences relative to Angus were smaller adjustment factors in column 6 of Tables 1-7. These than estimated from previous sampling in the 1970s. figures demonstrate the effect of selection over time Angus continued to have the greatest rate of genetic on breed differences; breeders within each breed apply change for yearling weight, causing most breed of variable levels of selection toward each trait resulting sire differences relative to Angus to decrease at least in reranking of breeds for each trait over time. These slightly. figures and Column 5 of Tables 1-7 can be used to Maternal Milk identify breeds with potential for complementarity in Changes to the maternal milk breed of sire mating programs. differences (Table 4, column 5) were generally small. Across-breed EPD Adjustment Factor Example All changes were less than 6 lb difference from those Adjustment factors can be applied to compare reported in 2013. However, the breed solution esti- the genetic potential of sires from different breeds. mates (Table 4, column 3) are expected to change the Suppose the EPD for yearling weight for a Red An- most in future updates as GPE heifers from each of the gus bull is +85.0 (which is above the birth year 2012 18 breeds being continuously sampled are developed average of 83 for Red Angus) and for a Charolais bull and bred. Chiangus and Santa Gertrudis estimates is +37.0 (which is below the birth year 2012 average and factors for maternal milk are presented here for of 45.7 for Charolais). The across-breed adjustment the first time. No females from newly sampled South factors in the last column of Table 3 are -29.9 for Red Devon or Tarentaise sires have weaned progeny as of Angus and 40.9 for Charolais. Then the adjusted EPD yet. We would expect their solutions to change the for the Red Angus bull is 85.0 + (-29.9) = 55.1 and for most in future reports. the Charolais bull is 37.0 + (40.9) = 77.9. The expect- Marbling, Ribeye Area, and Fat Thickness ed yearling weight difference when both are mated to Most changes to breed of sire differences were another breed of cow, e.g., Hereford, would be 55.1 – minor for each of these carcass traits. South Devon 77.9 = -22.8 lb. The differences in true breeding value was predicted to have less marbling relative to Angus between two bulls with similar within-breed EPDs are in comparison to Kuehn and Thallman (2013), likely primarily due to differences in the genetic base from due to new progeny carcass records from sires sam- which those within-breed EPDs are deviated. pled in 2011. Adjustment factors for Brahman are Birth Weight reported for the first time in this update. The range in estimated breed of sire differenc- Accuracies and Variance Components es for BWT (Table 1, column 5) ranged from 1.1 lb Table 8 summarizes the average Beef Improve- for Red Angus to 7.5 lb for Charolais and 10.9 lb for ment Federation (BIF) accuracy for bulls with progeny Brahman. Angus continued to have the lowest esti- at USMARC weighted appropriately by average rela- mated sire effect for birth weight (Table 1, column 5). tionship to animals with phenotypic records. The sires 137 sampled recently in the GPE program have generally for MAR, REA, and FAT are shown in Table 11. been higher accuracy sires, so the average accura- Each of these coefficients has a theoretical expected cies should continue to increase over the next several value of 1.00. Compared to growth trait regression years. coefficients, the standard errors even on the pooled Table 9 reports the estimates of variance estimates are higher, though they have decreased from components from the animal models that were used to the previous year. While REA and FAT are both close obtain breed of sire and breed of MGS solutions. Her- to the theoretical estimate of 1.00, we continued to itability estimates for BWT, WWT, YWT, and MILK use the theoretical estimate of 1.00 to derive breed of were 0.57, 0.17, 0.44, and 0.15, respectively. Herita- sire differences and EPD adjustment factors. Pooled bility estimates for MAR, REA, and FAT were 0.50, regression estimates for these two traits may be used 0.48, and 0.43, respectively. in future updates. Regression Coefficients Prediction Error Variance of Across-Breed EPD Table 10 updates the coefficients of regres- Prediction error variances were not included sion of records of USMARC progeny on sire EPD for in the report due to a larger number of tables included BWT, WWT, and YWT which have theoretical expect- with the addition of carcass traits. These tables were ed values of 1.00. The standard errors of the specific last reported in Kuehn et al. (2007; available online at breed regression coefficients are large relative to the http://www.beefimprovement.org/proceedings.html). An updat- regression coefficients. Large differences from the ed set of tables is available on request (Larry.Kuehn@ theoretical regressions, however, may indicate prob- ars.usda.gov). lems with genetic evaluations, identification, or sam- Implications pling. The pooled (overall) regression coefficients of Bulls of different breeds can be compared on a 1.16 for BWT, 0.84 for WWT, and 1.05 for YWT were common EPD scale by adding the appropriate across- used to adjust breed of sire solutions to the base year breed adjustment factor to EPD produced in the most of 2012. These regression coefficients are reasonably recent genetic evaluations for each of the 18 breeds. close to expected values of 1.0. Deviations from 1.00 The across-breed EPD are most useful to commercial are believed to be due to scaling differences between producers purchasing bulls of two or more breeds to performance of progeny in the USMARC herd and of use in systematic crossbreeding programs. Uniformity progeny in herds contributing to the national genetic in across-breed EPD should be emphasized for rota- evaluations of the 18 breeds. Breed differences calcu- tional crossing. Divergence in across-breed EPD for lated from the USMARC data are divided by these re- direct weaning weight and yearling weight should be gression coefficients to put them on an industry scale. emphasized in selection of bulls for terminal cross- A regression greater than one suggests that variation at ing. Divergence favoring lighter birth weight may be USMARC is greater than the industry average, while a helpful in selection of bulls for use on first calf heif- regression less than one suggests that variation at US- ers. Accuracy of across-breed EPD depends primarily MARC is less than the industry average. Reasons for upon the accuracy of the within-breed EPD of individ- differences in scale can be rationalized. For instance, ual bulls being compared. cattle at USMARC, especially steers and market heifers, are fed at higher energy rations than some seedstock animals in the industry. Also, in several recent years, calves have been weaned earlier than 205 d at USMARC, likely reducing the variation in weaning weight of USMARC calves relative to the industry. The coefficients of regression for MILK are also shown in Table 10. Several sire (MGS) breeds have regression coefficients considerably different from the theoretical expected value of 1.00 for MILK. Standard errors, however, for the regression coefficients by breed are large except for Angus and Her- eford. The pooled regression coefficient of 1.11 for MILK is reasonably close to the expected regression coefficient of 1.00. Regression coefficients derived from regression of USMARC steer progeny records on sire EPD

138 0.0 2.7 4.1 6.2 3.3 6.4 4.4 7.0 2.3 8.8 2.2 3.4 3.8 4.9 2.2 3.4 1.9 (6) 11.0 Factor to Factor To Angus To adjust EPD a (5) 0.0 4.4 1.1 6.6 4.1 4.9 3.4 5.4 3.3 7.5 4.1 2.4 3.7 4.8 2.0 3.8 2.0 10.9 BY 2012 BY Sire Breed Sire Difference (4) 87.0 91.3 88.0 93.5 91.1 91.8 97.9 90.3 92.3 90.2 94.4 91.1 89.3 90.6 91.8 88.9 90.8 88.9 Average BY 2012 BY Sire Breed Sire 0.0 3.7 0.2 6.7 4.2 6.3 3.9 6.6 5.7 8.3 4.5 4.2 3.4 6.7 3.2 5.7 2.5 (3) 11.2 (vs Ang) (vs Breed Soln Breed at USMARC 1.8 2.3 1.5 2.1 0.9 0.5 0.8 0.6 4.5 0.2 3.5 2.1 1.0 2.7 2.4 3.4 2.1 (2) -2.1 Bulls USMARC 1.8 3.5 2.2 2.6 0.3 1.7 0.8 0.2 2.8 0.5 3.7 0.8 1.7 1.7 1.6 2.2 1.9 Ave. Base EPD Ave. -1.2 (1) 2012 Breed 682 508 195 465 682 477 276 454 288 506 459 245 1936 2318 1126 1038 1104 1107 Direct Progeny Number 49 55 25 53 56 53 21 30 24 79 67 48 50 85 17 AI 152 149 107 Sires Breed = (3) / b + [(1) – (2)] + (Recent Raw Angus Mean: 87.0 lb) with b = 1.16 Raw = (3) / b + [(1) – (2)] (Recent Angus) = (4) – (4, Angus) – [(1) (1, Angus)] = (5) – (5, The breed difference estimates represent half the The breed difference Angus Table 1. Breed of sire solutions from USMARC, mean breed and USMARC EPD used to adjust for genetic trend to the year 2012 the year to genetic trend adjust for EPD used to and USMARC mean breed USMARC, solutions from of sire 1. Breed Table WEIGHT (lb) – BIRTH an Angus equivalent EPD to adjust within breed to base and factors a differences that would be expected between purebreds differences Hereford Red Angus Red Shorthorn South Devon Beefmaster Brahman Brangus Santa Gertrudis Braunvieh Charolais Chiangus Gelbvieh Limousin Maine Anjou Salers Simmental Tarentaise Calculations: (4) (5) (6)

139 9.9 0.0 (6) -6.4 -0.8 -5.1 -4.2 -5.2 30.7 37.9 40.6 37.2 44.8 15.4 -19.5 -19.4 -19.0 -23.4 -22.1 Factor to Factor To Angus To adjust EPD a 9.8 0.0 (5) -1.3 -2.9 -2.9 -3.9 -0.8 -8.3 -5.7 15.5 12.8 -29.1 -28.2 -12.1 -32.1 -16.1 -22.9 -10.2 BY 2012 BY Sire Breed Sire Difference (4) 573.0 584.1 589.8 545.2 571.4 571.4 546.1 562.1 570.4 542.1 587.0 565.9 573.4 558.1 551.3 564.1 568.6 574.3 Average BY 2012 BY Sire Breed Sire 0.9 8.9 1.5 1.3 8.3 0.0 (3) -5.0 -6.3 -2.8 -1.1 -1.9 -5.2 -3.1 19.9 21.2 15.5 19.9 19.3 (vs Ang) (vs Breed Soln Breed at USMARC 5.0 6.1 (2) -2.6 40.7 57.2 29.7 38.6 33.8 57.2 47.3 14.5 21.7 48.0 15.0 26.1 12.9 27.7 27.1 Bulls USMARC Ave. Base EPD Ave. 3.5 (1) 16.0 38.4 64.5 45.9 38.8 41.0 64.2 39.3 25.6 16.0 24.3 54.0 15.2 43.0 10.0 46.5 48.0 2012 Breed 237 256 974 470 436 422 591 456 263 656 478 176 442 Direct 1015 1009 1022 2144 1784 Progeny Number 17 24 79 66 48 50 84 30 56 53 21 49 55 25 53 AI 106 147 152 Sires Breed = (3) / b + [(1) – (2)] + (Raw Angus Mean: 553.4 lb) with b = 0.84 = (3) / b + [(1) – (2)] (Raw Angus) = (4) – (4, Angus) – [(1) (1, Angus)] = (5) – (5, The breed difference estimates represent half the differences that would be expected between purebreds of the estimates represent half the differences The breed difference Tarentaise Calculations: (4) (5) (6) Chiangus Gelbvieh Limousin Maine Anjou Salers Simmental Braunvieh Charolais Brahman Brangus Santa Gertrudis Red Angus Red Shorthorn South Devon Beefmaster Hereford Angus of the two breeds. 2012 the year to genetic trend adjust for EPD used to and USMARC mean breed USMARC, solutions from of sire 2. Breed Table WEIGHT (lb) – WEANING an Angus equivalent EPD to adjust within breed to base and factors a two breeds.

140 5.2 0.0 (6) 10.3 40.9 43.5 10.1 33.3 27.8 -24.6 -13.6 -41.5 -38.7 -24.9 -45.6 -47.7 -24.4 -29.9 -23.6 Factor to Factor To Angus To adjust EPD a 0.6 0.0 (5) -6.4 -30.6 -47.1 -49.7 -41.4 -17.7 -61.0 -71.8 -37.3 -37.3 -50.9 -38.7 -30.4 -33.3 -32.9 -34.1 BY 2012 BY Sire Breed Sire Difference (4) 990.4 979.6 1044.9 1020.8 1004.2 1001.7 1009.9 1033.6 1052.0 1014.0 1014.0 1000.4 1012.6 1020.9 1018.0 1018.4 1051.3 1017.2 Average BY 2012 BY Sire Breed Sire 0.6 5.0 4.3 3.3 0.0 (3) -8.7 -2.9 -8.7 22.2 21.9 -38.7 -11.8 -29.1 -24.0 -23.5 -28.7 -18.4 -26.7 (vs Ang) (vs Breed Soln Breed at USMARC 1.1 9.5 (2) Bulls 83.0 64.6 78.5 59.2 73.7 28.2 71.1 73.6 40.4 10.9 19.1 55.1 23.7 69.9 48.3 46.4 USMARC Ave. Base EPD Ave. 5.2 (1) 93.2 28.6 80.0 77.8 83.3 93.2 45.7 70.7 61.9 43.5 25.0 14.0 80.0 24.9 83.0 86.0 75.5 2012 Breed 891 234 404 429 954 920 930 222 237 399 333 534 337 175 429 595 Direct 1598 1997 Progeny Number 78 17 50 44 64 75 24 21 30 48 56 49 25 52 47 AI 101 135 140 Sires Breed = (3) / b + [(1) – (2)] + (Raw Angus Mean: 1013.6 lb) with b = 1.05 = (3) / b + [(1) – (2)] (Raw Angus) = (4) – (4, Angus) – [(1) (1, Angus)] = (5) – (5, The breed difference estimates represent half the differences that would be expected between purebreds of the estimates represent half the differences The breed difference Simmental Tarentaise Calculations: (4) (5) (6) Salers Maine Anjou Limousin Gelbvieh Charolais Chiangus Santa Gertrudis Braunvieh Brangus Brahman Beefmaster South Devon Shorthorn Angus Hereford Angus Red Table 3. Breed of sire solutions from USMARC, mean breed and USMARC EPD used to adjust for genetic trend to the year 2012 the year to genetic trend adjust for EPD used to and USMARC mean breed USMARC, solutions from of sire 3. Breed Table WEIGHT (lb) – YEARLING an Angus equivalent EPD to adjust within breed to base and factors a two breeds.

141 1.9 6.7 1.0 3.2 0.0 1.5 6.4 3.6 0.5 2.1 1.3 (6) -7.0 -7.1 21.7 23.9 25.1 13.0 -17.7 Factor to Factor To Angus To adjust EPD a 7.2 0.0 5.9 0.2 1.7 1.3 (5) -9.6 -4.5 -0.1 -1.4 -8.4 10.9 -22.9 -10.8 -12.8 -15.6 -10.9 -10.8 BY 2012 BY Sire Breed Sire Difference (4) 540.4 553.7 570.5 563.3 558.8 563.2 569.2 561.9 563.5 565.0 552.5 554.9 550.4 574.2 564.6 547.7 552.3 552.5 Average BY 2012 BY Sire Breed Sire 0.0 2.1 7.0 (3) -2.1 -3.7 -2.4 -8.7 -8.7 -0.6 -7.0 21.9 12.8 18.6 10.4 15.0 18.1 23.6 -24.2 (vs Ang) (vs Breed Soln Breed at USMARC 5.5 4.0 6.9 5.3 5.3 5.7 (2) -2.2 -0.1 14.1 10.0 14.5 30.6 19.0 19.1 27.2 20.8 19.9 33.5 Bulls USMARC Ave. Base EPD Ave. 7.7 2.2 6.0 0.6 0.2 2.0 (1) 24.0 18.8 18.0 28.0 22.6 10.2 24.0 23.7 20.2 19.0 11.1 33.0 2012 Breed 82 69 78 80 89 659 841 233 385 359 394 156 387 241 161 172 158 101 Direct Progeny 855 161 409 347 807 610 504 341 313 163 637 336 Gpr 2915 3593 1561 1509 1709 1663 Direct Number 6 40 87 21 69 60 41 14 53 36 45 65 35 21 26 34 AI 127 129 Sires Breed = (3) / b + [(1) – (2)] + (Raw Angus Mean: 553.4 lb) with b = 1.11 = (3) / b + [(1) – (2)] (Raw Angus) = (4) – (4, Angus) – [(1) (1, Angus)] = (5) – (5, The breed difference estimates represent half the differences that would be expected between purebreds of the estimates represent half the differences The breed difference Red Angus Red Charolais Chiangus Gelbvieh Limousin Table 4. Breed of maternal grandsire solutions from USMARC, mean breed and USMARC EPD used to adjust for genetic trend to to genetic trend adjust for EPD used to and USMARC mean breed USMARC, solutions from grandsire of maternal 4. Breed Table – MILK (lb) an Angus equivalent EPD to adjust within breed to 2012 base and factors the year a two breeds. Angus Hereford Shorthorn South Devon Brahman Maine Anjou Salers Simmental Tarentaise Brangus Santa Gertrudis Calculations: (4) (5) (6) Braunvieh Beefmaster 142 (6) 0.00 -0.72 -0.10 -0.41 -0.43 -0.35 -0.71 -0.11 -0.85 -0.67 -0.43 -0.19 -0.34 -0.31 Factor to Factor To Angus To adjust EPD b (5) 0.00 -1.02 -0.40 -0.78 -0.71 -0.84 -1.22 -0.21 -1.35 -1.17 -0.91 -0.66 -0.43 -0.76 ) BY 2012 BY a Sire Breed Sire Difference (4) 5.09 5.71 5.32 5.39 5.27 4.88 5.89 4.76 4.93 5.20 5.44 5.67 6.10 5.34 Average BY 2012 BY Sire Breed Sire (3) 0.00 -0.67 -0.63 -0.78 -0.77 -0.97 -0.42 -1.04 -0.87 -0.65 -0.37 -0.36 -0.07 -0.51 (vs Ang) (vs Breed Soln Breed at USMARC (2) 0.19 0.12 0.20 0.02 0.46 Bulls -0.39 -0.03 -0.24 -0.07 -0.01 -0.02 -0.03 -0.07 -0.01 USMARC Ave. Base EPD Ave. (1) 0.50 0.20 0.13 0.01 0.20 0.00 0.02 0.22 0.40 0.03 0.05 0.41 2012 -0.01 -0.01 Breed 68 714 193 423 400 383 220 222 113 239 107 228 928 220 Direct Progeny Number 46 74 71 59 44 54 21 46 24 22 51 46 AI 118 137 Sires 00 , 5.00 = Sm 00 Breed = (3) / b + [(1) – (2)] + (Raw Angus Mean: 5.79) with b = 1.00 = (3) / b + [(1) – (2)] (Raw Angus) = (4) – (4, Angus) – [(1) (1, Angus)] = (5) – (5, The breed difference estimates represent half the differences that would be expected between purebreds purebreds between be expected that would half the differences represent estimates difference The breed 4.00 = Sl of the two breeds. of the two Table 5. Breed of sire solutions from USMARC, mean breed and USMARC EPD used to adjust for genetic trend to the year 2012 the year to genetic trend adjust for EPD used to and USMARC mean breed USMARC, solutions from of sire 5. Breed Table units score – MARBLING (marbling an Angus equivalent EPD to adjust within breed to base and factors a b Angus Salers Simmental Calculations: (4) (5) (6) Gelbvieh Limousin Maine Anjou Brahman Santa Gertrudis Charolais Chiangus South Devon Shorthorn Hereford Angus Red 143 (6) (6) 0.46 0.82 0.93 1.08 0.67 0.46 1.04 0.23 0.23 0.00 0.000 -0.08 -0.09 -0.08 -0.02 -0.224 -0.206 -0.149 -0.145 -0.131 -0.213 -0.103 -0.135 -0.150 -0.051 -0.027 -0.135 Factor to Factor To Angus To Factor to Factor To Angus To adjust EPD adjust EPD a a (5) (5) 0.74 0.37 0.62 1.15 0.61 0.06 0.77 0.00 0.000 -0.02 -0.48 -0.52 -0.27 -0.28 -0.36 -0.237 -0.216 -0.219 -0.144 -0.191 -0.222 -0.111 -0.135 -0.150 -0.059 -0.040 -0.154 BY 2012 BY BY 2012 BY Sire Breed Sire Difference Sire Breed Sire Difference (4) (4) 0.401 0.422 0.420 0.494 0.447 0.416 0.527 0.503 0.489 0.639 0.580 0.598 0.485 13.93 13.56 13.81 14.35 13.81 13.25 13.97 13.18 12.72 12.68 12.92 12.92 12.83 13.19 Average BY 2012 BY Average BY 2012 BY Sire Breed Sire Sire Breed Sire ) 2 (3) (3) 0.91 0.78 1.00 1.31 0.92 0.42 1.03 0.37 0.16 0.00 0.000 -0.12 -0.16 -0.20 -0.23 -0.215 -0.201 -0.225 -0.211 -0.136 -0.213 -0.098 -0.154 -0.128 -0.056 -0.037 -0.144 (vs Ang) (vs (vs Ang) (vs Breed Soln Breed at USMARC Breed Soln Breed at USMARC (2) (2) 0.52 0.03 0.14 0.31 0.33 0.04 0.04 0.01 0.07 0.21 0.01 0.08 0.001 0.000 0.011 0.002 0.006 0.008 Bulls Bulls -0.04 -0.13 -0.007 -0.051 -0.078 -0.003 -0.004 -0.008 -0.008 USMARC USMARC ) with b = 1.00 2 Ave. Base EPD Ave. Ave. Base EPD Ave. (1) (1) 0.76 0.03 0.17 0.55 0.42 0.08 0.08 0.05 0.21 0.23 0.28 0.14 0.48 0.010 0.000 0.011 0.001 0.010 0.002 0.010 0.002 2012 -0.02 2012 -0.060 -0.050 -0.003 -0.003 -0.009 Breed Breed 68 68 715 194 424 400 220 107 239 227 114 927 219 228 424 194 220 384 402 108 227 114 240 928 220 228 715 Direct Direct Progeny Progeny Number Number 46 74 70 44 24 46 54 21 46 51 22 74 46 44 59 71 24 54 21 46 22 46 51 AI AI 118 137 137 118 Sires Sires Breed Breed = (3) / b + [(1) – (2)] + (Raw Angus Mean: 0. 630 in) with b = 1.00 = (3) / b + [(1) – (2)] (Raw Angus) = (4) – (4, Angus) – [(1) (1, Angus)] = (5) – (5, = (3) / b + [(1) – (2)] + (Raw Angus Mean: 12.79 in = (3) / b + [(1) – (2)] (Raw = (4) – (4, Angus) = (4) – (4, Angus) – [(1) (1, Angus)] = (5) – (5, The breed difference estimates represent half the differences that would be expected between purebreds purebreds between be expected that would half the differences represent estimates difference The breed The breed difference estimates represent half the differences that would be expected between purebreds of the estimates represent half the differences The breed difference Table 7. Breed of sire solutions from USMARC, mean breed and USMARC EPD used to adjust for genetic trend to the year 2012 the year to genetic trend adjust for EPD used to and USMARC mean breed USMARC, solutions from of sire 7. Breed Table THICKNESS (in) – FAT an Angus equivalent EPD to adjust within breed to base and factors a of the two breeds. of the two Angus Salers Simmental Calculations: (4) (5) (6) Gelbvieh Maine Anjou Chiangus Charolais Brahman Santa Gertrudis Hereford Angus Red Shorthorn South Devon Calculations: (4) Simmental Salers Maine Anjou Limousin Gelbvieh Chiangus Brahman Santa Gertrudis Charolais South Devon Hereford Angus Red Shorthorn Table 6. Breed of sire solutions from USMARC, mean breed and USMARC EPD used to adjust for genetic trend to the year 2012 the year to genetic trend adjust for EPD used to and USMARC mean breed USMARC, solutions from of sire 6. Breed Table (in – RIBEYE AREA an Angus equivalent EPD to adjust within breed to base and factors a two breeds. (5) (6) Angus

144 (6) 0.000 -0.224 -0.206 -0.149 -0.145 -0.131 -0.213 -0.103 -0.135 -0.150 -0.051 -0.027 -0.135 Factor to Factor To Angus To adjust EPD a (5) 0.000 -0.237 -0.216 -0.219 -0.144 -0.191 -0.222 -0.111 -0.135 -0.150 -0.059 -0.040 -0.154 BY 2012 BY Sire Breed Sire Difference (4) 0.401 0.422 0.420 0.494 0.447 0.416 0.527 0.503 0.489 0.639 0.580 0.598 0.485 Average BY 2012 BY Sire Breed Sire (3) 0.000 -0.215 -0.201 -0.225 -0.211 -0.136 -0.213 -0.098 -0.154 -0.128 -0.056 -0.037 -0.144 (vs Ang) (vs Breed Soln Breed at USMARC (2) 0.001 0.000 0.011 0.002 0.006 0.008 Bulls -0.007 -0.051 -0.078 -0.003 -0.004 -0.008 -0.008 USMARC Ave. Base EPD Ave. (1) 0.010 0.000 0.011 0.001 0.010 0.002 0.010 0.002 2012 -0.060 -0.050 -0.003 -0.003 -0.009 Breed 68 715 194 424 400 220 107 239 227 114 927 219 228 Direct Progeny Number 46 74 70 44 24 46 54 21 46 51 22 AI 118 137 Sires Breed = (3) / b + [(1) – (2)] + (Raw Angus Mean: 0. 630 in) with b = 1.00 = (3) / b + [(1) – (2)] (Raw Angus) = (4) – (4, Angus) – [(1) (1, Angus)] = (5) – (5, The breed difference estimates represent half the differences that would be expected between purebreds purebreds between be expected that would half the differences represent estimates difference The breed Table 7. Breed of sire solutions from USMARC, mean breed and USMARC EPD used to adjust for genetic trend to the year 2012 the year to genetic trend adjust for EPD used to and USMARC mean breed USMARC, solutions from of sire 7. Breed Table THICKNESS (in) – FAT an Angus equivalent EPD to adjust within breed to base and factors a of the two breeds. of the two Angus Salers Simmental Calculations: (4) (5) (6) Gelbvieh Maine Anjou Chiangus Charolais Brahman Santa Gertrudis Hereford Angus Red Shorthorn South Devon

145 Table 8. Mean weighteda accuracies for birth weight (BWT), weaning weight (WWT), yearling weight (YWT), maternal weaning weight (MWWT), milk (MILK), marbling (MAR), ribeye area (REA), and fat thickness (FAT) for bulls used at USMARC

Breed BWT WWT YWT MILK MAR REA FAT

Angus 0.80 0.78 0.72 0.73 0.53 0.52 0.51 Hereford 0.66 0.62 0.62 0.59 0.28 0.41 0.32 Red Angus 0.92 0.91 0.91 0.89 0.68 0.66 0.68 Shorthorn 0.82 0.80 0.74 0.79 0.61 0.59 0.54 South Devon 0.41 0.45 0.41 0.44 0.06 0.08 0.09 Beefmaster 0.87 0.89 0.86 0.74 Brahman 0.51 0.48 0.43 0.32 0.08 0.11 0.08 Brangus 0.87 0.81 0.79 0.68 Santa Gertrudis 0.73 0.66 0.59 0.53 0.20 0.44 0.28 Braunvieh 0.57 0.50 0.40 0.42 Charolais 0.80 0.75 0.67 0.68 0.48 0.51 0.46 Chiangus 0.82 0.79 0.79 0.74 0.24 0.22 0.33 Gelbvieh 0.82 0.81 0.81 0.79 0.54 0.53 0.54 Limousin 0.93 0.90 0.83 0.84 0.76 0.76 Maine Anjou 0.80 0.79 0.78 0.78 0.30 0.29 0.33 Salers 0.83 0.82 0.76 0.80 0.25 0.29 0.33 Simmental 0.94 0.94 0.94 0.93 0.73 0.71 0.71 Tarentaise 0.94 0.93 0.91 0.94

aWeighted by relationship to phenotyped animals at USMARC for BWT, WWT, YWT, MAR, REA, and FAT and by relationship to daughters with phenotyped progeny MILK.

146 Table 9. Estimates of variance components (lb2) for birth weight (BWT), weaning weight (WWT), yearling weight (YWT), and maternal weaning weight (MWWT) and for marbling (MAR; marbling score units2), ribeye area (REA; in4), and fat thickness (FAT; in2) from mixed model analyses

Direct

Analysis BWT WWTa YWT Direct Animal within breed (19 breeds) 70.18 479.78 3560.76 Maternal genetic within breed (19 breeds) 435.18 Maternal permanent environment 723.89 Residual 53.70 1256.00 4533.89

Carcass Direct MAR REA FAT Animal within breed (13-14 breeds) 0.280 0.674 0.0105 Residual 0.278 0.737 0.0141 aDirect maternal covariance for weaning weight was -61.96 lb2

205 days (WWT), and 365 days (YWT) of F1 progeny and for calf weights (205 d) of F1 dams Table(MILK) 10. on Pooled sire expected and within-breed progeny difference regression and coefficients by sire breed (lb/lb) for weights at birth (BWT),

BWT WWT YWT MILK Pooled 1.16 ± 0.04 0.84 ± 0.03 1.05 ± 0.04 1.11 ± 0.07 Sire breed Angus 1.05 ± 0.09 0.86 ± 0.07 1.23 ± 0.07 1.05 ± 0.15 Hereford 1.18 ± 0.07 0.72 ± 0.05 1.01 ± 0.06 1.05 ± 0.15 Red Angus 1.06 ± 0.14 0.82 ± 0.14 0.60 ± 0.16 1.42 ± 0.27 Shorthorn 0.66 ± 0.21 0.58 ± 0.20 0.52 ± 0.26 1.16 ± 0.71 South Devon -0.31 ± 0.53 0.67 ± 0.31 0.50 ± 0.32 0.18 ± 1.57 Beefmaster 2.03 ± 0.33 1.00 ± 0.22 0.66 ± 0.34 3.31 ± 0.70 Brahman 1.91 ± 0.21 1.04 ± 0.18 1.35 ± 0.21 -0.05 ± 0.66 Brangus 1.69 ± 0.23 0.94 ± 0.20 1.35 ± 0.28 0.06 ± 0.56 Santa Gertrudis 3.63 ± 0.71 1.04 ± 0.23 1.10 ± 0.29 0.26 ± 0.89 Braunvieh 0.68 ± 0.26 0.59 ± 0.24 0.59 ± 0.38 0.52 ± 0.54 Charolais 1.13 ± 0.12 0.95 ± 0.11 0.85 ± 0.12 1.16 ± 0.24 Chiangus 1.46 ± 0.30 0.17 ± 0.25 0.56 ± 0.29 0.18 ± 0.44 Gelbvieh 1.11 ± 0.14 0.85 ± 0.11 1.13 ± 0.12 0.82 ± 0.25 Limousin 0.99 ± 0.11 1.01 ± 0.09 1.15 ± 0.12 1.81 ± 0.25 Maine Anjou 1.44 ± 0.18 0.92 ± 0.19 0.76 ± 0.25 2.02 ± 0.41 Salers 1.26 ± 0.23 0.80 ± 0.26 0.47 ± 0.25 1.70 ± 0.40 Simmental 1.10 ± 0.14 1.47 ± 0.13 1.33 ± 0.13 0.89 ± 0.31 Tarentaise 0.85 ± 0.59 1.06 ± 0.24 1.48 ± 0.34 1.13 ± 0.93

147 2 2 ribeye area (REA; in /in ), and fat thickness (FAT; in/in) of F1 progeny on sire expected Tableprogeny 11. difference Pooled and and within-breed by sire breed regression coefficients marbling (MAR; score/score),

MAR REA FAT Pooled 0.60 ± 0.04 0.83 ± 0.06 0.94 ± 0.08 Sire breed Angus 0.90 ± 0.08 0.75 ± 0.13 1.11 ± 0.15 Hereford 0.66 ± 0.15 0.62 ± 0.13 0.97 ± 0.18 Red Angus 0.76 ± 0.15 1.07 ± 0.20 0.51 ± 0.40 Shorthorn 1.68 ± 0.30 1.66 ± 0.50 1.87 ± 0.48 South Devon -0.15 ± 0.23 1.64 ± 2.25 5.59 ± 2.65 Brahman 2.57 ± 1.01 1.22 ± 0.36 1.50 ± 0.60 Santa Gertrudis 0.83 ± 0.62 1.12 ± 0.44 0.74 ± 0.46 Charolais 1.29 ± 0.25 1.07 ± 0.27 1.45 ± 0.44 Chiangus 0.57 ± 0.22 0.20 ± 0.43 0.35 ± 0.45 Gelbvieh 1.21 ± 0.20 1.30 ± 0.16 1.70 ± 0.27 Limousin 1.20 ± 0.37 1.22 ± 0.17 Maine Anjou 0.77 ± 0.68 -0.91 ± 0.48 -1.19 ± 0.73 Salers 0.07 ± 0.07 1.63 ± 0.60 1.29 ± 0.59 Simmental 0.84 ± 0.17 0.69 ± 0.15 0.11 ± 0.31

148 Figure 1. Relative genetic trends for birth weight (lb) of the seven most highly used beef breeds (1a) and all breeds that submitted 2014 trends (1b) adjusted for birth year 2012 using the 2014 across-breed EPD adjustment factors. 1a.

1b.

149 Figure 2. Relative genetic trends for weaning weight (lb) of the seven most highly used beef breeds (2a) and all breeds that submitted 2014 trends (2b) adjusted for birth year 2012 using the 2014 across-breed EPD adjustment factors. 2a.

2b.

150 Figure 3. Relative genetic trends for yearling weight (lb) of the seven most highly used beef breeds (3a) and all breeds that submitted 2014 trends (3b) adjusted for birth year 2012 using the 2014 across-breed EPD adjustment factors. 3a.

3b.

151 Figure 4. Relative genetic trends for maternal milk (lb) of the seven most highly used beef breeds (4a) and all breeds that submitted 2014 trends (4b) adjusted for birth year 2012 using the 2014 across-breed EPD adjustment factors. 4a.

4b.

152 Literature Cited Kuehn, L. A., L. D. Van Vleck, R. M. Thallman, and L. V. Cundiff. 2008. Across-breed EPD tables for Arnold, J. W., J. K. Bertrand, and L. L. Benyshek. the year 2008 adjusted to breed differences 1992. Animal model for genetic evaluation of for birth year of 2006. Proc. Beef Improvement multibreed data. J. Anim. Sci. 70:3322-3332. Federation 40th Annual Research Symposium and Annual Meeting, Calgary, AB. June 30-July 3, Barkhouse, K. L., L. D. Van Vleck, and L. V. 2008. pp 53-74. Cundiff. 1994. Breed comparisons for growth and maternal traits adjusted to a 1992 base. Proc. Kuehn, L. A., L. D. Van Vleck, R. M. Thallman, and L. Beef Improvement Federation 26th Research V. Cundiff. 2009. Across-breed EPD tables for the Symposium and Annual Meeting, Des Moines, IA, year 2009 adjusted to breed differences for birth May, 1994. pp 197-209. year of 2007. Proc. Beef Improvement Federation 41th Annual Research Symposium and Annual Barkhouse, K. L., L. D. Van Vleck, and L. V. Cundiff. Meeting, Sacramento, CA. April 30-May 3, 2009. 1995. Mixed model methods to estimate breed pp 160-183. comparisons for growth and maternal traits adjusted to a 1993 base. Proc. Beef Improvement Kuehn L. A., L. D. Van Vleck, R. M. Thallman, and L. Federation 27th Research Symposium and Annual V. Cundiff. 2010. Across-breed EPD tables for Meeting, Sheridan, WY. May 31-June 3, 1995. pp the year 2010 adjusted to breed differences 218-239. for birth year of 2008. Proc. Beef Improvement Federation 42nd Annual Research Symposium Boldman, K. G., L. A. Kriese, L. D. Van Vleck, and S. D. and Annual Meeting, Columbia, MO. June 28-July Kachman. 1993. A Manual for Use of MTDFREML 1, 2010. pp. 71-92. (DRAFT). A set of programs to obtain estimates of variances and covariances. USDA-ARS, Roman Kuehn L. A., L. D. Van Vleck, R. M. Thallman, and L. L. Hruska U.S. Meat Animal Research Center, Clay V. Cundiff. 2011. Across-breed EPD tables for the Center, NE. (120 pp). year 2011 adjusted to breed differences for birth year of 2009. Proc. Beef Improvement Federation Cundiff, L. V. 1993. Breed comparisons adjusted 43rd Annual Research Symposium and Annual to a 1991 basis using current EPD’s. Proc. Beef Meeting, Bozeman, MT. June 1-4, 2011. pp. 92- Improvement Federation Research Symposium 111. and Annual Meeting, Asheville, NC. May 26-29, 1993. pp 114-123. Kuehn L. A., and R. M. Thallman. 2012. Across-breed EPD tables for the year 2012 adjusted to breed Cundiff, L. V. 1994. Procedures for across breed EPD’s. differences for birth year of 2010. Proc. Beef Proc. Fourth Genetic Prediction Workshop, Beef Improvement Federation 44th Annual Research Improvement Federation, Kansas City, MO. Jan. Symposium and Annual Meeting, Houston, TX. 1994. April 18-21, 2012. pp. 152-177.

Koch, R. M., L. V. Cundiff, and K. E. Gregory. 1994. Kuehn L. A., and R. M. Thallman. 2013. Across-breed Cumulative selection and genetic change for EPD tables for the year 2013 adjusted to breed weaning or yearling weight or for yearling differences for birth year of 2011. Proc. Beef weight plus muscle score in Hereford cattle. J. Improvement Federation 45th Annual Research Anim. Sci. 72:864-885. Symposium and Annual Meeting, Oklahoma City, OK. June 12-15, 2013. pp. 114-141. Kuehn, L. A., L. D. Van Vleck, R. M. Thallman, and L. V. Cundiff. 2007. Across-breed EPD tables for Notter, D. R., and L. V. Cundiff. 1991. Across-breed the year 2007 adjusted to breed differences expected progeny differences: Use of within- for birth year of 2005. Proc. Beef Improvement breed expected progeny differences to adjust Federation 39th Annual Research Symposium breed evaluations for sire sampling and genetic and Annual Meeting, Fort Collins, CO. June 6-9, trend. J. Anim. Sci. 69:4763-4776. 2007. pp 74-92.

153 Núñez-Dominguez, R., L. D. Van Vleck, and L. V. Van Vleck, L. D., and L. V. Cundiff. 2000. Across-breed Cundiff. 1993. Breed comparisons for growth EPD tables for 2000 adjusted to a 1998 base. traits adjusted for within-breed genetic trend Proc. Beef Improvement Federation 32th Annual using expected progeny differences. J. Anim. Sci. Research Symposium and Annual Meeting, 71:1419-1428. Wichita, KS. July 12-15, 2000. pp 98-116.

Van Vleck, L. D. 1994. Prediction error variances Van Vleck, L. D., and L. V. Cundiff. 2001. Across-breed for inter-breed EPD’s. Proc. Fourth Genetic EPD tables for 2001 adjusted to breed differences Predication Workshop, Beef Improvement for birth year 1999. Proc. Beef Improvement Federation, Kansas City, MO. Jan. 1994. Federation 33th Annual Research Symposium and Annual Meeting, San Antonio, TX. July 11-14, Van Vleck, L. D., and L. V. Cundiff. 1994. Prediction 2001. pp 44-63. error variances for inter-breed genetic evaluations. J. Anim. Sci. 71:1971-1977. Van Vleck, L. D., and L. V. Cundiff. 2002. Across- breed EPD tables for 2002 adjusted to breed Van Vleck, L. D., and L. V. Cundiff. 1995. Assignment differences for birth year of 2000. Proc. Beef of risk to across-breed EPDs with tables of Improvement Federation 34th Annual Research variances of estimates of breed differences. Proc. Symposium and Annual Meeting, Omaha, NE. Beef Improvement Federation 27th Research July 10-13, 2002. pp 139-159. Symposium and Annual Meeting, Sheridan, WY. May 31-June 3, 1995. pp 240-245. Van Vleck, L. D., and L. V. Cundiff. 2003. Across-breed EPD tables for the year 2003 adjusted to breed Van Vleck, L. D., and L. V. Cundiff. 1997. Differences differences for birth year of 2001. Proc. Beef in breed of sire differences for weights of male Improvement Federation 35th Annual Research and female calves. Proc. Beef Improvement Symposium and Annual Meeting, Lexington, KY. Federation Research Symposium and Annual May 28-31, 2003. pp 55-63. Meeting, Dickinson, ND. May 14-17, 1997. pp 131-137. Van Vleck, L. D., and L. V. Cundiff. 2004. Across-breed EPD tables for the year 2004 adjusted to breed Van Vleck, L. D., and L. V. Cundiff. 1997. The across- differences for birth year of 2002. Proc. Beef breed EPD tables adjusted to a 1995 base. Improvement Federation 36th Annual Research Proc. Beef Improvement Federation Research Symposium and Annual Meeting, Sioux Falls, SD. Symposium and Annual Meeting, Dickinson, ND. May 25-28, 2004. pp 46-61. May 14-17, 1997. pp 102-117. Van Vleck, L. D., and L. V. Cundiff. 2005. Across-breed Van Vleck, L. D., and L. V. Cundiff. 1998. Across-breed EPD tables for the year 2005 adjusted to breed EPD tables for 1998 adjusted to a 1996 base. differences for birth year of 2003. Proc. Beef Proc. Beef Improvement Federation Research Improvement Federation 37th Annual Research Symposium and Annual Meeting, Calgary, Symposium and Annual Meeting, Billings, MT. Alberta, Canada. July 2, 1998. pp 196-212. July 6-9, 2005. pp 126-142.

Van Vleck, L. D., and L. V. Cundiff. 2006. Across-breed of breed of dam on across-breed adjustment EPD tables for the year 2006 adjusted to breed Vanfactors. Vleck, L.Midwestern D., and L. Section V. Cundiff. ASAS 1998. and InfluenceMidwest differences for birth year of 2004. Proc. Beef Branch ADSA 1998 Meeting, Des Moines, IA. Improvement Federation 39th Annual Research Abstract # 10. p 31. Symposium and Annual Meeting, Choctaw, MS. April 18-21, 2006. Available online at: http://www. Van Vleck, L. D., and L. V. Cundiff. 1999. Across-breed beefimprovement.org/proceedings/06proceedings/2006-bif- EPD tables for 1999 adjusted to a 1997 base. vanvleck-cundiff.pdf. Proc. Beef Improvement Federation 31th Annual Research Symposium and Annual Meeting, Roanoke, VA. June 15-19, 1999. pp 155-171.

154 Van Vleck, L. D., L. V. Cundiff, T. L. Wheeler, S. D. MEAN EPDs REPORTED BY DIFFERENT Shackelford, and M. Koohmaraie. 2007. Across- BREEDS breed adjustment factors for expected progeny Larry A. Kuehn1 and R. Mark Thallman1 differences for carcass traits. J. Anim. Sci. 85:1369-1376. 1Roman L. Hruska U.S. Meat Animal Research Center, USDA-ARS, Clay Center, NE 68933 Westell, R. A., R. L. Quaas, and L. D. Van Vleck. 1988. Expected progeny differences (EPDs) have Genetic groups in an animal model. J. Dairy Sci. been the primary tool for genetic improvement of beef 71:1310-1318. cattle for over 40 years beginning with evaluations of growth traits. Since that time, EPDs have been added for several other production traits such as calving ease, stayability, carcass merit and conformation. Most recently, several breed associations have derived economic indices from their EPDs to increase profit under different management and breeding systems. It is useful for producers to compare the EPDs of potential breeding animals with their breed average. The current EPDs from the most recent genetic evaluations of 24 breeds are presented in this report. Mean EPDs for growth traits are shown in Table 1 (24 breeds), for other production traits in Table 2 (18 breeds), and for carcass and composition traits in Ta- ble 3 (20 breeds). Several breeds also have EPDs and indices that are unique to their breed; these EPDs are presented in Table 4. Average EPDs should only be used to determine the genetic merit of an animal relative to its breed average. To compare animals of different breeds, across breed adjustment factors should be added to animals’ EPDs for their respective breeds (see Across-breed EPD Tables reported by Kuehn and Thallman in these proceedings). This list is likely incomplete; evaluations for some breeds are not widely reported. If you see a breed missing and would like to report the average EPDs for that breed, please contact Larry (Larry. [email protected]) or Mark ([email protected]. gov).

155 Table 1. Birth year 2012 average EPDs from 2014 evaluations for growth traits Birth Weaning Yearling Maternal Total Breed Weight (lb) Weight (lb) Weight (lb) Milk (lb) Maternal (lb)

Angus 1.8 48 86 24 Hereford 3.5 46.5 75.5 18.8 42.1 Murray Grey 3.8 22 35 4 15 Red Angus -1.2 54 83 18 45 Red Poll 1.6 15 23 7 Shorthorn 2.2 15.2 24.9 2.2 9.8 South Devon 2.6 43 80 24 45

Beefmaster 0.3 10 14 2 Braford 1.1 1.2 17 3 9 Brahman 1.7 16 25 6 Brangus 0.8 24.3 43.5 11.1 23.2 Red Brangus 1.5 12.7 20.0 5.3 11.7 Santa Gertrudis 0.2 3.5 6.2 0.2 Senepol 0.8 8.3 9.0 4.0 4.0 Simbrah 3.9 62.5 84.4 22.5 53.7

Braunvieh 2.8 39.3 61.9 33.0 52.7 Charolais 0.5 25.6 45.7 7.7 20.5 Chianina 3.7 38.4 70.7 10.2 29.3 Gelbvieh 0.8 64.5 93.2 27.9 60.2 Limousin 1.7 45.9 83.3 22.6 Maine-Anjou 1.7 38.8 77.8 20.2 39.5 Salers 1.6 41 80 19 Simmental 2.2 64.2 93.2 23.7 55.8 Tarentaise 1.9 16 28.6 0.6

156 Table 2. Birth year 2012 average EPDs from 2014 evaluations for other production traits Calving Calving Ease Ease Mature Heifer Direct Maternal Scrotal Docility Weight Pregnancy Stayability Breed (%) (%) Circ (cm) Score (lb) (%) (%)

Angus 5 8 0.77 12 35 9.2 Hereford 0.8 1.1 0.8 85 Murray Grey -0.6 -0.2 0.2 52 Red Angus 4 5 10 11 Shorthorn -1.3 -1.4 South Devon 0.1

Beefmaster 0.2 Brangus 5.1 7.1 0.55 Simbrah 2.7 6.7 7.9

Braunvieh -0.2 -0.6 Charolais 3.0 3.7 0.66 Chianina 5.5 -2.2 Gelbvieh 9.7 6.8 5.2 Limousin 9.0 4.5 0.46 20.7 20.0 Maine Anjou 9.2 3.5 Salers 0.3 0.4 0.3 9 23 Simmental 9.3 10.6 10.3 19.6 Tarentaise -1.2 0.6

157 0.04 (lb) -0.31 -0.02 WBSF a a a a (in) 0.00 0.01 -0.008 -0.057 Rump fat fat Rump a a a a (in) 0.00 0.01 0.01 0.001 0.011 0.01 0.002 0.00 0.00 0.000 0.011 0.002 -0.06 -0.003 -0.05 -0.009 -0.060 -0.003 -0.43 Fat Thickness Fat a a a a ) 2 (in 0.76 0.03 0.55 0.17 0.23 0.42 0.08 0.43 0.14 0.21 0.08 0.48 0.28 0.10 0.04 0.31 0.58 0.06 0.05 Carcass -0.02 Ribeye Area Area Ribeye a a a a 0.13 0.2 0.20 0.00 0.4 0.02 0.01 0.0 0.00 0.41 0.22 0.02 0.50 0.05 0.01 Score Score -0.01 -0.24 -0.03 -0.09 -0.01 Marbling Marbling Yield Grade -0.31 -0.07 -0.18 -0.02 -0.22 (%) 0.0 0.33 0.8 0.4 -0.01 -0.08 Retail Product Product 1.1 8.3 6 5.1 -0.5 28.4 22 25.1 27 12.9 18.2 25.9 30 12 25.3 18 15.1 28 Wt (lb) Wt Carcass Birth year 2012 average EPDs from 2014 Simmental Salers Maine-Anjou Beefmaster Limousin Breed South Devon Brangus Braunvieh Shorthorn Chianina Gelbvieh Hereford Grey Murray Brahman Simbrah Red Angus Red Braford Santa Gertrudis Charolais Angus Derived using ultrasound measures and reported on Table 3. Table evaluations for carcass and composition traits a an ultrasound scale (IMF% instead of marbling score)

158 Table 4. Birth year 2012 average EPDs from 2014 evaluations for other traits unique to individual breeds Residual Cow Weaned Average Daily Mature Yearling Energy Calf Feedlot Grid Beef Angus Gain (lb) Height (in) Height (in) Value ($) Value ($) Value ($) Value ($) Value ($) 0.16 0.4 0.5 -3.78 29.89 27.89 30.29 69.76

Baldy Calving Ease Maternal Index Index ($) Beef Index ($) Index ($) Hereford ($) Brahman Influence Certified Hereford 16.98 15.19 21.42 14.99

Mature Cow Maintenance Red Angus (Mcal/mo) 0

Feedlot Carcass Gelbvieh Merit ($) Value ($) 31.61 17.24

Mainstream Limousin Terminal Index ($) 44.5

All Purpose Terminal All Purpose Terminal Simmental Index ($) Index ($) Simbrah Index ($) Index ($) 119.5 68.3 69.10 51.60

$ British Maternal Shorthorn $ Calving Ease $ Feedlot Index 18.20 16.12 20.06

Murray Gestational Days to Grey 600-d wt (lb) length (d) calving (d) 51 -0.2 -0.8 159 SPONSORS PATRON PLATINUM

GeneSeek ARS, Meat Animal Research Center IANR Merial Zoetis, Inc. Angus Journal Beef Magazine

160 PLATINUM

Nebraska Cattlemen Foundation National Association of Animal Breeders (Accelerated Genetics, ABS Global Inc., Genex, ORIGEN and Select Sires)

161 SPONSORS GOLD SILVER

National Program for Genetic Improvement of Feed Efficiency in Beef Cattle American Angus Association American Hereford Association Boehringer Ingelheim GrowSafe Systems Red Angus Association of America Illumina Progressive Cattleman Nebraska Cattlemen

162 SILVER

American Gelbvich Association TH E Home American Shorthorn Association Agency American Simmental Association thehomeagency.com Capple Sales, Inc. Hunt Limousin Ranch Rishel Angus The CUP Lab The Home Agency Wagonhammer Ranches Livestock Gentec Allflex Trans Ova Genetics Nebraska Beef Council

P O Box 1555 North Platte, Nebraska 69103 308.534.5305

163 SPONSORS BRONZE FRIEND

American Brahman Breeders Association A & B Cattle American Maine Anjou Association American-International Charolais Association Flying H Genetics Cross Diamond Cattle Company Gallagher Grace Mayer International Brangus Breeder Association Lincoln County Feedyard M&P Gelbvieh Moly MFG., Inc. Schuler Red Angus/Schuler-Olsen Ranches Inc. Affymetrix Agrigenomics Cow Sense Software Nebraska Corn Board

MOLY MFG., INC.

RREETT GG UURR AATTEE TM ®

T T TM

164 FRIEND

American Akaushi Association Canaday Ranch Frenzen Angus & Polled Herefords

165