Whole genome sequencing identified a 16 kilobase deletion on ECA13 associated with distichiasis in Friesian horses

BACKGROUND

Distichiasis, an ocular disorder in which aberrant cilia (eyelashes) grow from the opening of the Meibomian glands of the eyelid, has been reported in Friesian horses. These misplaced cilia can cause discomfort, chronic keratitis, and corneal ulceration, potentially impacting vision due to corneal fibrosis, or, if secondary infection occurs, may lead to loss of the eye. Friesian horses represent the vast majority of reported cases of equine distichiasis, and as the breed is known to be affected with inherited monogenic disorders, this condition was hypothesized to be a simply inherited Mendelian trait.

RESULTS

A genome wide association study (GWAS) was performed using the Axiom 670 k Equine Genotyping array (MNEc670k) utilizing 14 cases and 38 controls phenotyped for distichiasis. An additive single locus mixed linear model (EMMAX) approach identified a 1.83 Mb locus on ECA5 and a 1.34 Mb locus on ECA13 that reached genome-wide significance (pcorrected = 0.016 and 0.032, respectively). Only the locus on ECA13 withstood replication testing (p = 1.6 × 10- 5, cases: n = 5 and controls: n = 37). A 371 kb run of homozygosity (ROH) on ECA13 was found in 13 of the 14 cases, providing evidence for a recessive mode of inheritance. Haplotype analysis (hapQTL) narrowed the region of association on ECA13 to 163 kb. Whole-genome sequencing data from 3 cases and 2 controls identified a 16 kb deletion within the ECA13 associated haplotype (ECA13:g.178714_195130del). Functional annotation data supports a tissue-specific regulatory role of this locus. This deletion was associated with distichiasis, as 18 of the 19 cases were homozygous (p = 4.8 × 10- 13). Genotyping the deletion in 955 horses from 54 different breeds identified the deletion in only 11 non-Friesians, all of which were carriers, suggesting that this could be causal for this Friesian disorder.

CONCLUSIONS

This study identified a 16 kb deletion on ECA13 in an intergenic region that was associated with distichiasis in Friesian horses. Further functional analysis in relevant tissues from cases and controls will help to clarify the precise role of this deletion in normal and abnormal eyelash development and investigate the hypothesis of incomplete penetrance.

A missense mutation in damage-specific DNA binding protein 2 is a genetic risk factor for ocular squamous cell carcinoma in Belgian horses

BACKGROUND

Belgian horses are commonly affected with ocular squamous cell carcinoma (SCC), the most common cancer of the equine eye. A missense mutation in damage-specific DNA binding protein 2 (DDB2 c.1013C>T, p.Thr338Met) has been established as a recessive genetic risk factor for ocular SCC in the Haflinger breed. A sample of Belgian horses with unknown SCC phenotype was shown to possess this variant at a similar frequency to the Haflinger breed. Retrospective studies indicate that chestnut coat colour may predispose to the development of SCC.

OBJECTIVES

To determine if DDB2 c.1013C>T is a risk factor for ocular SCC in a strictly phenotyped sample of Belgian horses. To investigate associations between coat colour loci genotypes and ocular SCC.

STUDY DESIGN

Retrospective and prospective case identification, genetic investigation.

METHODS

Genomic DNA was isolated from blood, hair or formalin-fixed paraffin-embedded tissue from 25 Belgian horses with histologically confirmed ocular SCC and 18 unaffected Belgian horses. Association testing of 34 single nucleotide variants from 11 genomic loci and genotyping for DDB2 c.1013C>T and coat colour alleles were performed. Exons of DDB2 were sequenced in four cases and two controls. Associations were analysed by Chi-square or Fisher's exact tests and relative risk was calculated.

RESULTS

Homozygosity for DDB2 c.1013C>T was significantly associated with ocular SCC (P = 7.4 × 10^-7 ). Seventy-six per cent of affected horses were homozygous for the variant. Relative risk for homozygous horses developing SCC was 4.0 (P = 1.0 × 10^-4 ). Sequencing DDB2 did not identify a variant more concordant with disease phenotype. An association between disease and coat colour loci was not identified.

MAIN LIMITATIONS

Phenotyping was determined at a single timepoint. Each included horse genotyped as chestnut, so association with this MC1R variant could not be investigated.

CONCLUSIONS

A missense variant, DDB2 c.1013C>T, p.Thr338Met, is a risk factor for ocular SCC in Belgian horses. A genetic risk test is commercially available.

Genetic risk for squamous cell carcinoma of the nictitating membrane parallels that of the limbus in Haflinger horses

Squamous cell carcinoma (SCC) is the most common cancer affecting the equine eye, with a higher incidence documented in Haflinger horses. Recently, a missense variant in the gene damage specific DNA binding protein 2 (DDB2, p.Thr338Met) on ECA12 was identified as a risk factor for the development of limbal SCC in Haflinger horses. SCC also occurs on the nictitating membrane; therefore, investigating the role of this missense variant in nictitating membrane SCC is warranted. In this study, a common ancestor was identified among Haflinger horses affected with limbal SCC or with nictitating membrane SCC, thus supporting a recessive risk factor for the development of cancer at both ocular locations. Analysis of genotype data from Haflinger horses with and without nictitating membrane SCC revealed that the same region on ECA12 associated with limbal SCC was also associated with nictitating membrane SCC (P < 2.04 × 10^-5 ). Fine mapping of this locus using 25 cases and 49 controls supported the hypothesis that DDB2:c.1013C>T, p.Thr338Met, is a risk factor for nictitating membrane SCC, as 88% of our cases were homozygous for this variant and no other polymorphism was more strongly associated (P = 4.13 × 10^-14 ). These data indicate that the genetic risk is the same for the development of both limbal and nictitating membrane SCC in Haflinger horses and validates utilization of genetic testing of the DDB2 variant for both clinical management and the guidance of mating decisions.