Q&A: Professor Emmeline Hill tells us more about her inbreeding study

New research led by Professor Emmeline Hill has warned that the rise of inbreeding in the international thoroughbred population could compromise the future of the breed, likening the problem and the lack of willingness to tackle it to global warming.

Full details of the report can be read here.

Professor Hill, UCD professor in equine genomics and chief science officer at Plusvital, gave us more details of the study in a Q&A.

RP Can you explain more how inbreeding can compromise overall population fertility and health?

EH There is a very strong correlation between inbreeding and the presence of deleterious (or harmful) mutations in the genome. Inbreeding increases the exposure of deleterious mutations that in single copy form (i.e., inherited from one parent) are masked by a ‘normal’ copy from the other parent.

Inbreeding creates large chromosomal regions that are inherited as identical segments from both parents causing the effects of the mutations to be exposed.

Generally, harmful mutations are rare, but mating close relatives that may possess a copy of the same mutation increases the probability that two copies of a mutation will be passed on to progeny.

If a mutation has a seriously deleterious effect on a physical trait, for instance if it causes a major developmental abnormality or a disease that affects a foal early in life, then it will disappear quickly from the population because it will not be passed on to the next generation.

However, when deleterious mutations are not lethal and inbreeding is recent, selection will not have had enough time to remove them from the population.

Many mutations do not have singular, large, obvious effects. However, in inbred populations multiple mutations with small effects accumulate. Over time, the individual effects of these mutations may combine to have a large impact on a trait. Generally, these mutations affect development and fertility traits.

Not all inbreeding is ‘bad’ inbreeding. The thoroughbred was developed on the basis of inbreeding to successful ancestors. Such purposeful inbreeding attempts to duplicate (i.e. create two copies of the same) favourable mutations to reinforce traits that are beneficial to performance.

Chromosome regions inherited from ancestors in the pedigree that have been maintained through generations of selection are likely to contain beneficial gene variants. These chromosome regions are old and have been passed through the population for generations. However, this beneficial inbreeding must be balanced in the context of the negative consequences of inbreeding.

Genomics tools can now be used to analyse an individual horse’s DNA by generating information from thousands of variable DNA markers evenly spaced across the horse genome.

From this DNA information the proportion of the DNA that has been influenced by positive inbreeding and the proportion of DNA influenced by inbreeding that is more likely to harbour deleterious mutations can be quantified. Application of these tools in practice could be used to manage inbreeding in individuals, within a breeding herd and within the entire population.

RP Can you give us more specific information about the 10,000 horses who were included in the study?

EH The number of horses studied from each geographic region was: Australia and New Zealand 2,733; Europe 4,099; Japan 192; South Africa 1,102; US 1,992; total 10,118.

Horses used in the study were not a ‘random’ sample and there may be some bias considering sample origins. Horses used in the study had been submitted for genetic testing by Plusvital and were approved for use in research.

RP Can you give any more details of which sires or broodmares the thoroughbred population is heavily inbred to?

EH Identifying individual horses that have disproportionately contributed to the genetics of the population was not the intended purpose of the study. Since the study did not focus on specific individuals, drawing any conclusions about individual horses other than the main influences on genetic diversity in the population would not be appropriate.

However, we did look at stallions representing the Byerley Turk sire line, given concerns about the possible demise of this lineage. We found that the five stallions that trace back to the Byerley Turk on the male line were not genetically distinct from other horses and inbreeding levels were comparable to those of other horses.

This is consistent with recent analysis of the male specific Y chromosome that established that there are likely more true Byerley Turk lineages in the current population than previously thought; horses descended from Galopin, that on pedigree are attributed to the Darley Arabian male lineage, are in fact genetic descendants of the Byerley Turk.

RP Many breeders might think they could infer a level of inbreeding in a horse from looking at its tabulated pedigree but you say this is not accurate; can you explain why?

EH Genomic inbreeding is a measure of the proportion of similarity of the DNA inherited from the sire and dam. Traditionally, breeders have used pedigree information to estimate the inbreeding co-efficient for an individual or for progeny arising from a hypothetical mating.

Thoroughbreds are now all very closely related to one another and pedigree information does not provide the same resolution to understand inheritance patterns as it may have done in the past when more diverse bloodlines existed.

Pedigree has no way of showing what genes are actually inherited from generation to generation. Genetic evaluation of inbreeding assesses the true state of shared genetics inherited from both sire and dam lines.

There are two reasons why pedigree and genetic levels of inbreeding may not agree - firstly due to the randomness of halving of the ancestral genetic material at each generation and secondly due to the unpredictable amount of sharing of genetics across the whole population.

Each horse inherits half of its DNA from its dam and half from its sire. In the creation of the gametes (the dam’s egg cells and the sire’s sperm cells) the DNA from the maternal and paternal grandparents is ‘shuffled’. This shuffling of the genes produces a different genetic combination in each gamete (egg/sperm) and what that combination is cannot be predicted from the pedigree.

This ensures that no two progeny from a mating between the same mare and sire will inherit identical DNA and introduces randomness to what is passed on from one generation to the next.

Intuitively, we might expect that each horse should inherit 25 per cent of its DNA from each grandparent, but because of the random nature of the ‘shuffling’ this can actually vary between 20 to 30 per cent. That means that a horse may inherit more or less than expected of a particular grandparent’s DNA (or great-grandparent, great-great-grandparent etc).

Our new genetic tools can assess and quantify the amount of DNA inherited from grandparents in a pedigree. We can now look beyond the pedigree to the reality of inheritance, just as is now routinely done in human ancestry testing.

Relatedness and inbreeding are two sides of the same coin. We have observed that pedigree is informative for first and second degree relatives, but the relationship between pedigree and actual genetic relatedness breaks down quickly after that. Our observations indicate that a horse with a high pedigree inbreeding co-efficient may in fact have a low measure of genomic inbreeding and vice versa.

In other words, horses that may appear to be ‘inbred’ on pedigree, may not in fact have high true levels of inbreeding. The only way to accurately assess the amount of the genome that has been identically inherited from the sire and dam is by measuring genomic inbreeding through systematic examination of thousands of DNA markers.

RP The global warming metaphor seems apt, as it is hard to believe that breeders will put concern for the long-term health of the breed ahead of what they perceive will make them a profit in the short term. Do you really think people will sit up and take notice?

EH Addressing the increasing inbreeding trend in the population must be tackled by industry bodies by monitoring, assessing and providing solutions.

The breeders that we have spoken to appear to be keenly aware of the problem, but are conflicted by the requirement to produce and sell horses with market appeal.

We believe this is still achievable by harnessing the tools that we have developed, which allow breeders to assess the true genetic relationship between two individuals and predict the level of inbreeding likely to arise from a particular mating. This is now possible using computer-based algorithms when DNA information of a mare and stallion are both available.

Of course, it will be crucial to maintain genetically diverse stallions that are attractive in the market to enable breeders to have opportunities for their mares to select suitable outcrosses.

Sometimes there can be surprises and a stallion that a breeder may have thought was ‘too close’ on pedigree, may in fact be genetically less similar than he appears to be and therefore could in fact be a good choice.

These new tools are less likely to limit choice than to provide alternative breeding options that may not have been otherwise considered.

Knowing the genetic make-up of mares will likely bring short-term benefits to breeders, since higher inbreeding in other species is known to negatively affect fertility.

Though it remains to be scientifically tested, we hypothesise that choosing more genetically distant stallions for mares may be economically beneficial by increasing the chances of maintaining pregnancies and producing robust, more healthy foals. It will also help towards maintaining genetic diversity in the population.

The underlying mutations for certain conditions are rapidly being discovered and catalogued through the efforts of the international horse genome mapping workshop community of scientists.

Therefore, the genome-wide DNA marker data generated for assessment of inbreeding could be used to identify carriers and avoid matings that might produce affected offspring. Also, once an animal has had its genome “profiled”, it will be relatively straightforward to predict carrier status for additional genetic disorders, as they are identified and catalogued over the coming years.

These profiles will also allow breeders to recognise positive gene variants in their animals that have been selected for generations for performance attributes.

See our stallion videos:

Meet Coolmore's new stallions Calyx, Magna Grecia and Ten Sovereigns

Kirsten Rausing talks us through the Lanwades Stud roster

Highclere Stud's Jake Warren gives us the lowdown on Land Force