So the two talks by Dr Brooks and Dr Cosgrove at the Al Khamsa spring meeting were mind-blowing for me, because, well, I simply was not aware of how far technology has come and what is now available to us for the management of our horses’ health. I am still recovering, in particular, from learning that testing of any individual horse’s genome is now commercially available, and also from the figure shared above, which is incredibly exciting.
The figure comes from the paper by Cosgrove et al. 2020 (open access link) on the genomes of the Arabian. To explain what you are seeing there, every single one of those 312 vertical bars is the genome of a horse. Each different colour represents a different founder population, so the more colours in any given vertical bar/genome, the more genetic diversity is present in that individual horse’s genome.
On the far right, the mostly blue horses are Thoroughbreds, while the solid mustard-coloured block next to it are Icelandic horses, both used as outgroups, i.e. reference groups not part of the groups being analysed, as are the Turkemen horses next to the Icelandics.
So here we can see that the Icelandic is pretty homogenous, which makes sense, as it has been isolated longer than any other breed that I can think of. The Thoroughbred is also very homogenous, thanks to intense selection pressure for speed and registry requirements regarding ancestry. Some Thoroughbreds have slightly more diverse genomes - there’s one horse, towards the left of the Thoroughbred cluster, which shares most of its genes with the general Thoroughbred population, but also has a small patch of mustard colour in its column, as well as brown and some red. So this particular individual has inherited DNA from four founder populations.
The Turkemen horses, by contrast, have more diversity than either the Thoroughbred or the Icelandic, with individuals expressing DNA from up to, I think, nine founder populations.
So, with the outgroups for comparison out of the way, now we can get down to the nitty-gritty and the fun stuff with the Arabians. To the surprise of absolutely no-one, I am going to start with the desertbred groups.
Immediately, we can see that the horses from the cradle countries (plus Tunisia) are genetically diverse. The horses from Saudi Arabia have the least diversity, while the Syrian horses have the most. Unfortunately, there are very few samples for these horses - you can see there are only two vertical columns, thus two horses, in the Syrian group. However, this diversity agrees with the findings of Khanshour et al. 2013 and Almarzook et al. 2017. It also makes sense, as these are the horses from the homelands of the Arabian, while the Arabian populations outside the desert have been limited to an extent by the founder effect, i.e. the reduction in variation that results from a smaller sub-set of the wider population creating new satellite colonies. For instance, the Davenport Arabians trace back to the twenty-seven horses that Homer Davenport imported to the US in 1906; not all of those twenty-seven bred on, and several of them were closely related, being siblings or parents and offspring. So on paper the Davenport horses have twenty-seven founder genomes at most to play with, but in practice they have rather fewer.
The diversity we see among the desertbred horses is also indicative of the Arabian in its home being almost more of a landrace than a breed with a set type. Increased genetic diversity leads to increased heterozygosity, which means that it is harder to fix a specific physical type or standard. So there are big Arabians and small Arabians in their homelands, chunky Arabians and more refined Arabians, Arabians that are faster over shorter distances and Arabians that are faster over longer distances, and so on and so forth. The greater diversity also tells us that these are the original populations, and that the horses still in the desert today bear a close genetic resemblance to their ancestors.
The next group I want to look at are the Polish horses.
Unsurprisingly, the Polish horses are less diverse, in general, than the desertbreds, with the pink population from the desert horses dominating. Some Polish horses are more diverse than others - the ones towards right of the centre of the Polish block have more colourful genomes, so more populations. In fact, the horse to immediately to the right of the centre of the block has eight colours in its genome, so is genetically very valuable. None of the Polish horses sampled show any of the Thoroughbred blue, despite their historical selection for racing performance. The prevalence of pink in the Polish horses may indicate founder effect (in this case probably due to the effects of the two World Wars), or may indicate selection for specific traits considered desirable in these horses, or be a mixture of both.
Of the Arabian breeding groups studied by Cosgrove et al., the one with the least diversity is the Straight Egyptian.
While some of these horses still have quite a few different blocks of colour in their genomes, these blocks are incredibly minute, and with each new generation, they become smaller, or are simply not handed down, by the chances of genetic recombination. The genomes of these Straight Egyptians show intense selection for the teal population, and several of the horses to the right of this group have absolutely no diversity at all. They are entirely teal. The DNA from other ancestral groups has been obliterated.
This has two effects. Both are the results of the increased homozygosity. For breeders, homozygosity helps to fix type and make the outcome of any match more predictable. When you have fewer genes to play with, there are fewer ways in which they can recombine, so fewer genotypes available.
On the other hand, homozygosity increases the incidence of recessive diseases in a population. Lavender Foal Syndrome is a known Straight Egyptian issue: this is why. Other Arabian diseases are found frequently in Straight Egyptian bloodlines, again because of their increased homozygosity. Without diversity, not only are the horses vulnerable to disease, they are also vulnerable to inbreeding depression, i.e. fertility problems. The Straight Egyptian is going the way of the Friesian, with its multitude of health issues and reproductive problems, or, to take a human analogy, the way of the Spanish Hapsburgs.
There is so much I want to talk about here - the great big middle section of multi-origin Arabians, mostly US-based, for instance, showing the contributions of the variety of founder populations, most strikingly the pink and teal associated with Polish and Straight Egyptians, as well as salmon and baby blue, which have not been linked to a specific founder population yet; one current guess is that those colours may be associated with Crabbet/Old English horses. Then there’s the section with the purple bar and asterisk beneath it, showing noticeable amounts of Thoroughbred blue: those are all racing “Arabians”, surprise surprise.
But really, the big takeaways for me would be that the Straight Egyptians are more vulnerable than I realised, and that Arabian breeders outside the cradle countries need to make more use of AI to incorporate genetics from the desertbred horses. Maintaining the breed’s unique and marvellous levels of diversity is important for the long-term health and survival of the breed. Also that we can actually have our horses’ genomes tested now, which is incredibly exciting, and such a powerful tool for breeding, for welfare, for management.
Almarzook et al.. 2017. ‘Genetic diversity of Syrian Arabian horses’. Animal Genetics 48:486-489
Cosgrove et al.. 2020. ‘Genome Diversity and the Origin of the Arabian Horse’. Scientific Reports 10:9702
Khanshour et al.. 2013. ‘Microsatellite Analysis of Genetic Diversity and Population Structure of Arabian Horse Populations’. Journal of Heredity 104:386–398