RAPGEF5 (EFIH)

What it does

Crossing two of them

Its gene carrot

Gene key

horsegenetics.rapgef5

Priority

89

Alleles

efih N

Outcomes

wild, rapgef5-carrier, efih

Coat effect

none — every outcome is a wild type

Default allele

N

Wild population

1.4% of founders carry one copy; none carries two

Founder draws

1 × nextFloat()

Catalogue entries

1 — it never changes what a horse looks like

The combination table

The most severe and the rarest

It takes the largest bite out of the foal’s max health, so it dies fastest, and it has the lowest carrier rate of the four. That pairing is the point: the disorders a player is most likely to run into are the ones nearest to survivable, and the ones that kill outright are the ones that have been bred down hardest.

At a glance

As with PLOD1, the genomic letters (G>T) and the coding letters (C>A) differ because the gene is read from the reverse strand. Same event, two reference frames.

The gene, and a tail that turns out to matter

RAPGEF5 encodes a guanine-nucleotide exchange factor in Rap-family small-GTPase signalling. In the work behind EFIH it was connected to regulation of nuclear β-catenin and Wnt signalling — pathways central to embryonic tissue patterning and organ development. Rocha et al., PMC7544121

The mutation turns a serine codon into a stop, and removes only the extreme C-terminal tail of the predicted protein — which is why its importance was not obvious at first. Two things settled it. The site is extraordinarily conserved: the affected residue and six of the seven remaining C-terminal amino acids were completely conserved across 100 vertebrate species examined. And functional work in frog embryos showed the truncated equine protein had substantially reduced activity against the wild-type equine protein.

What EFIH does to a foal

Formerly called idiopathic hypocalcaemia of Thoroughbred foals, EFIH is a severe inherited primary hypoparathyroid condition. Affected foals show markedly low blood calcium including low ionised calcium, often raised phosphorus and low magnesium, and parathyroid hormone that is low or nominally normal at a moment when it should be strongly elevated. Clinically that is muscle stiffness, tetany, weakness, recumbency, inability to stand, sometimes heavy sweating, and seizures.

The finding that gives the gene its mechanism: at necropsy, investigators could not identify normal parathyroid tissue in affected foals. That points at abnormal parathyroid development rather than at a normal gland secreting badly — which is what makes a developmental-signalling gene a plausible cause in the first place.

Onset is early but variable: reported affected foals presented from about 4 to 35 days, and in the genetically characterised cases from hours after birth to around 30 days. Milk calcium intake and supplementation can temporarily mask the disorder. Calcium supplementation prevents the seizures, which supports the reading that the seizures are secondary to hypocalcaemia and not evidence of a primary seizure disorder — but calcium homeostasis remains extremely hard to maintain, and the condition is described as invariably fatal or ending in euthanasia.

Inheritance

All four affected foals in the original study were homozygous; their tested dams, and the one available tested sire, were heterozygous, and the pedigree structure supported recessive inheritance. A carrier pairing gives the ordinary 25% / 50% / 25%.

None of that makes a carrier weak, small, slow or poor-jumping. A carrier is expected to be phenotypically normal with respect to EFIH; the risk is specifically the production of an affected foal when the other parent also carries. The management answer is to test breeding stock and avoid carrier × carrier — not to remove every carrier, which needlessly shrinks diversity in a closed studbook.

Coat, build, size, speed, jumping

Coat. No published evidence of any effect on base colour, dilution, white patterning, grey, roan, appaloosa patterning, mane and tail colour, skin colour or eye colour. Two reasons, and the second is the stronger: homozygotes do not survive to an age where an adult appearance phenotype could be assessed, and the discovery and prevalence studies were built around hypocalcaemia and the lethal foal phenotype and reported no coat association.

Build and size. No demonstrated association between carrier status and mature height, growth trajectory, frame, bone, muscling, topline, limb proportions, head type, birth size or any conformation trait. Severe loss of parathyroid function in an affected foal will obviously disrupt development — that is not the same claim as the variant producing a build, and there is no evidence that one allele in a carrier changes anything.

Speed. No equine study has linked the allele to racing speed, stride traits, exercise physiology, muscle-fibre composition, aerobic capacity, adult injury risk or earnings. The absence matters more here than elsewhere, because the allele lives in Thoroughbreds and it is tempting to read racing aptitude out of pedigree patterns. The allele was found through a rare lethal foal disorder, not through a performance GWAS.

Jumping. Nothing on scope, technique, carefulness, bascule, take-off power, hindlimb mechanics, adult soundness or competitive outcome. The variant was also reported as Thoroughbred-exclusive in the original screen, so there is no basis for treating it as a marker in Warmblood jumping programmes at all.

Breeds and frequency

The best population estimate tested 1,789 US Thoroughbreds born 1988–2019.

No homozygotes appeared in that registration sample, which is what early lethality predicts: an affected foal would not usually survive to the time of registration and parentage sampling. The oldest confirmed carrier sample was from a horse born in 1992, so this is not a newly arisen mutation, and there was no significant change in allele frequency between the older and newer cohorts. PMC10073348

An earlier, much smaller set of 82 unaffected Thoroughbreds found 3 carriers — an estimated 1.8% allele frequency. The 1,789-horse study is the better estimate, being far larger and structured across birth years and regions.

Under Hardy-Weinberg, an allele frequency of 0.8% predicts affected homozygotes at roughly 0.008² — about 1 in 15,600 conceptions. That is a population expectation only; real incidence moves with non-random mating, family lines, selection, embryonic and neonatal loss, and whether affected foals are diagnosed and tested at all.

The discovery work screened public sequence from 123 horses across 12 non-Thoroughbred breed groups — Akhal-Teke, Friesian dwarf, German and Swiss Warmbloods, Haflinger, Icelandic, Dutch Warmblood, Quarter Horse, Shetland Pony, Standardbred, Yakutian and Franches-Montagnes — and did not see the variant. Read that carefully: it supports Thoroughbred-specificity in the data available then. It does not prove the allele can never appear in a crossbred with Thoroughbred ancestry, and it does not establish a zero-frequency estimate for any other breed, because those sample sizes were small and were not formal population surveys.

What is settled, and what is not

Strongly supported. c.2624C>A / p.Ser875* is strongly associated with EFIH in Thoroughbred foals; EFIH is autosomal recessive; affected foals have severe hypocalcaemia and a hypoparathyroid phenotype; normal parathyroid tissue was not identified in the examined foals; the allele is functionally impaired in the Xenopus overexpression model; and the gene is highly expressed in human parathyroid tissue, expressed in normal equine parathyroid tissue, and substantially expressed in equine brain and spinal cord.

Plausible, not resolved. That RAPGEF5 is required for embryonic development of the parathyroid glands via Wnt/β-catenin-related signalling — the missing tissue is compatible, but the developmental step and cell type are unproven. That variation in milk consumption or milk calcium explains why some foals crash within days and one presented near 30 days — offered explicitly as a hypothesis, not a confirmed modifier. That the gene has roles beyond the parathyroid, given its neural expression — though the affected foals’ seizures responded to calcium correction and no primary neurological syndrome was found. And that the equine finding makes RAPGEF5 a candidate for unexplained human hypoparathyroidism, which is a research implication rather than an established human disease gene.

Not supported. That EFIH determines coat colour or markings; that it makes horses taller, shorter, heavier, finer, stockier or more athletic; that it helps or harms speed or jumping in healthy carriers; that it causes adult-onset disease in carriers; or that it is in any sense a “Thoroughbred performance gene”. A carrier result is a reproductive-risk result and nothing else.