Genetically Speaking

Ireland is not a big country. It covers only 32,599 square miles, making it about the same size as Indiana. But Ireland plays a large and important role in the Thoroughbred industry. Some of the best runners in the world are raised on its limestone-enriched pastures. Two Irish-bred standouts-High Chaparral and Domedriver-won races at the 2002 Breeders' Cup World Thoroughbred Championships, and another, Rock of Gibraltar, was Europe's 2002 Horse of the Year.

Ireland is also a hotbed of genetic research involving the Thoroughbred. Patrick Cunningham, PhD, and his team of scientists at Trinity College in Dublin have been looking at the role heredity plays in the development of the breed for more than 20 years.

"The first study we did was on inbreeding in Thoroughbreds," Cunningham said. "As you know, Thoroughbreds are notoriously infertile. It takes about three mares to produce two foals each year if you look at statistics, and that's very low. If you look at populations of single offspring species like cattle and sheep-or horses in the wild-they'll normally produce a foal crop of about 85% in natural conditions. We decided to look at infertility in the Thoroughbred, thinking it might well be due to inbreeding because in all species, as you increase inbreeding, fertility is one of the things that declines.

"What we showed is that inbreeding is about 13-14% in Thoroughbreds going back to the base of the population," he continued. "But the variation in inbreeding was not related to variation in fertility of mares. How do you explain that? Our explanation essentially was that this inbreeding has been accumulating very slowly over 30 generations and fertility is subject to natural selection. We calculated the amount of natural selection that would offset any expected decline from inbreeding, and we concluded that if you eliminated about 8% of mares in any one generation for reasons of fertility, that would be enough to sustain fertility. So at any rate, we concluded that inbreeding was not something we had to worry about in the Thoroughbred population at that time because the population was quite big. We still don't think we need to worry about it today."

Looking at the situation further, the scientists identified another factor that was more important than inbreeding in infertility.

"We studied the pattern of ovulation in mares that were brought to slaughterhouses in Australia and Ireland," Cunningham said. "Of course, there is a seasonal pattern, and mares tend to go quiet; they don't ovulate in midwinter. But the breeding season is tied to the fact that in the Northern Hemisphere, a horse is declared a year older on Jan. 1, so everybody wants to have their foals born early in the year. They start breeding their horses on the 15th of February, and they stop breeding on the 15th of July. And if you match that breeding season with the mare's natural ovulation pattern, you find that you are actually including some quite infertile months and excluding some of the fertile ones. So, we think that is a major contributor to infertility. We reckon the administrative structure (of the breeding season) is costing the industry about 10 foals per 100 mares per year."

Cunningham and his fellow researchers also studied the heritability of track performance, trying to determine whether the Thoroughbred was improving as an athlete or not. They graphed the winning times of classic races in Europe and America.

"What we were able to show was that for races of a mile-and-a-half or more, the winning times had plateaued from the early part of the 20th century," Cunningham said. "So as far as winning times were concerned, it seemed that there wasn't much progress. Records were not being broken. It was more pronounced in longer races than in shorter races. But by a different kind of analysis, we were able to show that, in fact, the population of Thoroughbreds as a whole is improving at the rate of about 1% a year. So you could say, in a way, the losers are getting better. If you look at Quarter Horses, which run in very short races, you find that records continue to be broken. In the sprints, they are still pushing the limits. So our interpretation of that is there is a physiological limit on performance, which is lactic acid clearance from the blood. And that really only becomes an issue when the race is a mile-and-a-quarter or longer.

"But," he added, "there certainly is genetic variation for performance. When we measure it using Timeform ratings, it is about 30% heritable. In other words, of all the variations, 30% are due to genes and the remaining 70% are due to training, injuries, and so on."

In more recent work, the Trinity College scientists studied the genetic origins of the Thoroughbred and confirmed that they are very narrow-perhaps more narrow than most people thought. Looking at horses registered in the British and Irish General Stud Book, they found there were 10 "founder" females that account for 72% of maternal lineages in the modern Thoroughbred population, and one founder stallion, the Darley Arabian, is responsible for 95% of the paternal lineages. The other founder stallions are the Byerley Turk and the Godolphin Arabian.

"We were surprised to find such an imbalance between the three stallions," Cunningham said. "Our conclusion is that it's a bit like winner takes all. It happens over a long period of time. At the early stage of the breed, there were many contenders for prominence as stallions. If you look at the pattern of breeding in the early generations, you can see why some of the stallions were used more, and the breeders at the time have written about why they preferred one over another. But after a number of generations, you find that they settled down to using relatively few stallions, and from there, the one that had the most presence in the population was the one that was most likely to increase its presence. It's a little bit like Hertz and Avis. Hertz got to be biggest first, and it's hard for Avis to catch up."

That study was published in Animal Genetics in 2001, with Cunningham the article's lead writer. Earlier this year, the results of more research from the Trinity College team were published in the same magazine. They revealed inaccuracies in the official pedigree records contained in the General Stud Book, which is compiled by Weatherbys. Cunningham also worked on that study, but Emmeline Hill, PhD, was the number one researcher.

Hill and her fellow scientists looked at the samples of mitochondrial DNA taken from 100 Thoroughbreds representing 19 of the most common matrilineal families. These were determined using a numerical system that was developed by Bruce Lowe in the 1800s to classify foundation mares based on the success of their descendants on the racetrack and as producers.

Mitochondrial DNA is passed only through females and does not change from generation to generation. According to the researchers, they found nearly half the families they studied had at least one individual whose DNA sequences did not match those of the rest of the family members.

"This is historically interesting, but it isn't surprising," Hill said, "because in the early days of the stud book (the 1790s), it was a huge task for its original compilers to put all the information together. What we have been able to show is that some of the founder females have contributed more than we thought and some of them, in fact, may not have contributed distinct lines at all. Now some breeders will say, 'That's OK because we only go back a few generations when we plan our matings.' But it may have some implications for breeders who want to take into account those old lines."

Said Cunningham, "It is possible to correct these errors. And there is a small amount of interest in doing so from the industry here (in Europe), but it would be quite a big job. However, it is easy to do for a particular case, when a breeder has a mare that he paid a lot of money for, and he wants to be absolutely sure that she does descend from the original founder that the records show."

Looking Ahead Looking to the future, Cunningham believes there is great potential for Thoroughbred breeding to benefit from genetic research. The completion of the working draft of the entire human genome was announced last year. And equine scientists can use that information to help them identify and locate genes in horses. There also is an international effort to develop a gene map for the horse, but the work is moving much slower than on the human project. Still, according to Cunningham, equine scientists already are making advances in determining the role of genetics in exercise-induced pulmonary hemorrhage (EIPH) and stallion fertility. An interesting area of recent research involves the ACE (angiotensin converting enzyme) gene. People with a certain variation of this gene are more likely to have superior endurance and be better athletes. Researchers in Australia are looking closely at the ACE gene and how it relates to the horse.

"They are working with the idea of developing a laboratory test that could tell you whether a young horse is likely to have a capacity for stamina or whether the offspring of a pair of prospective parents will have a capacity for stamina," Cunningham said. "I think there will be a usable test for this in the next three years."

At Trinity College, the focus in the future will be on using genetics to breed Thoroughbreds in a more precise way to get desired results in performance.

"With Thoroughbreds, you have breeding advisers or consultants who put their own and other available information together in a way that is rather like a witch's brew," Cunningham said. "They use selective data-some based on their own experiences-to come up with recommendations about why certain mares should be crossed with certain stallions. Many of these people do a very good job of assessing the information that they essentially hold in their heads. But their evaluations are often subjective, with a lot of smoke and mirrors involved. In other species, such as cattle and salmon, things are done on a more systematic basis. You first identify exactly what it is you're trying to improve, and then you find out how heritable it is. The Thoroughbred industry has not been as receptive as other industries to this scientific type of approach."

According to Cunningham, Trinity College would like to use the results of genetics research to develop scientific breeding procedures that would be marketed to Thoroughbred horsemen.

"I think we will probably offer services to breeders on an individual, confidential basis," he said. "You might see papers published about our methodology as we go along, but we'd like to set up a commercial operation. But for now, we are just testing the water because we have no way of knowing who will be receptive to it or on what scale. One of the services we would like to develop would give an individual breeder information on the inbreeding of any particular horse at the touch of a button. It would also give the breeder information on the relationship of their horse to any other horse in the Thoroughbred population."

About the Author

Deirdre Biles

Deirdre Biles is the Bloodstock Sales Editor for The Blood-Horse magazine.

Stay on top of the most recent Horse Health news with FREE weekly newsletters from Learn More