Wild horses: bay, wildtype bay, seal brown, brown?

I repeatedly wrote on my blog that western wild horses had, according to genetic studies, several colours: black, black dun, bay, bay dun and leopard spotted. 

The basis for the presence of bay in wild horses is that genetic tests in predomestic DNA samples have identified the presence of both the allele a and A on the Agouti locus in wild horses [1]. But it is, unfortunately, not as simple as it seems at first. 

A bay European wild horse. All rights reserved

There are two types of bay colour: “normal” bay and wildtype bay (or wild bay). The wildtype bay is similar to bay but lighter in colour, and only the toes are dark-coloured while in bays the whole legs are dark. Wildtype bay is hypothesized to be caused by the hypothetical allele A⁺ [2]. Personally, I do not believe that wildtype bay is actually the wildtype version of bay. The dark areas on the legs in Przewalski’s horses cover almost the entire leg and not only the toes, as in horses of a normal bay colour, what suggests that they share the allele which would then very likely be the wildtype allele. So, we have bay and a form of bay which is called wildtype but is not necessarily wildtype. 

And then there is seal brown, a colour found for example in Exmoor ponies. The genetic background of seal brown seems to be unclear. It is speculated that it is caused by the hypothetical allele A^t on the Agouti locus. There was a genetic test for this allele, which is now considered inadequate, and the sequence of the allele has not been published [2]. 

Druml et al have a different hypothesis on the genetic background of seal brown. They postulate that this phenotype is caused by the genotype A/a, therefore being the result of heterozygosity of the allele causing black in homozygous individuals (a) and the allele causing bay in homozygous individuals (A). What is striking is that they consider bay itself to be the result of a heterozygous genotype, namely A/A E/e [3]. The e allele is the domestic allele on the Extensionlocus that causes a chestnut colour in homozygous individuals. Therefore, bay would be a domestic colour, and the wildtype E/E A/A colour would be brown, which, according to them, is caused by that genotype. I have to say that I doubt the purported genotypes for these colours, for once because they differ from what is usually considered to be the genotype for those colours, and because 1) if seal brown was the result of A/a, there would be not only bay and seal brown Exmoor ponies but also black Exmoor ponies, which is not the case in the current population. Also, if bay was the result of the genotype E/e, there would be chestnut (e/e) Exmoor ponies, which is not the case and 2) if bay was the result of a heterozygous state, it would be impossible to breed a breed that is exclusively bay. However, there is at least one such a breed, the Cleveland Bay horse [2]. Therefore, bay cannot be the result of a heterozygous state. Brown, is, as defined by Sponenberg and Bellone (2017), sooty + bay. The genetic background of sooty is unclear [2].

One could assume that since, according to the literature, both the alleles A and a but not A⁺ and A^t have been found in European wild horses the case is clear that seal brown and wildtype bay are not wildtype colours. But the problem is that the common genetic test for the alleles A and a can only confirm the presence of a, and the presence of A is deduced by the absence of a [2]. Therefore, the test cannot discriminate between A, A⁺ and A^t (if the latter two exist at all). If the authors of the papers proposing the presence of A in wild horses used that test, the presence of the hypothetical alleles A⁺ and A^t in wild horses cannot be ruled out currently. 

To further complicate the subject, Sponenberg and Bellone (2017) state that seal brown can also be caused by the pangare allele Pa⁺ diluting a black phenotype [2]. This would have consequences for the phenotype of wild horses. If that is correct, it would mean that this form of seal brown is a wildtype colour and that the non-pangare allele Pa^npwould have to have been present in wild horses besides the pangare allele, otherwise black and other non-pangare phenotypes would be domestic colours. The pangare allele has been identified [2], therefore I hope that it will be tested for extinct wild horse samples one day. 

Another factor that plays in the game is the Dun locus. Since 2015 we know that there are three alleles on this locus: dun (a wildtype colour identified in all living wild equines and Pleistocene wild horse samples), non-dun 1 d1 which is wildtype as well, and non-dun 2 d2 which is domestic [2,4]. Wildtype non-dun and domestic non-dun look different; the former have a clearly visible dorsal stripe and the surrounding colour is lighter than in the latter, the latter are darker all over the body [4]. Wildtype non-dun bay horses, such as some Gotland ponies, almost look like bay dun horses, only slightly darker in shade. 

 

So, which colour did the wild horses have that were neither leopard spotted, black, black dun or any other form of dun? I think that this cannot be said with a 100% certainty with the current knowledge on horse colour genetics and the current genetic tests we have. Very likely the horses had the A allele since it is the basal allele of wild equines, but the presence of other alleles on the Agouti locus cannot be ruled out yet. The genetic background of sooty and seal brown would have to be identified and tested for wild horses. Also, it would be interesting to know if wild horses had the non-pangare allele. The colour of Exmoor ponies, however, is domestic in any case because so far this breed has been found to exclusively carry the domestic non-dun 2 allele d2. It would be interesting to know what seal brown combined with a d1/d1 phenotype would look like. 

 

Literature

 

[1] Pruvost et al.: Genotypes of predomestic horses match phenotypes painted in Paleolithic works of cave art. 2011. 

[2] Sponenberg & Bellone: Equine color genetics. 2017.  

[3] Druml et al.: Discriminant analysis of colour measurements reveal allele dosage effect of ASIP/MC1R in bay horses. 2018. 

[4] Imsland et al.: Regulatory mutations in TBX3 disrupt asymmetric hair pigmentation that underlies Dun camouflage color in horses. 2015.