In other species: taurine cattle , indicine cattle (zebu) , goat , domestic yak , bighorn sheep , water buffalo , kouprey

Categories: Craniofacial phene

Links to possible relevant human trait(s) and/or gene(s) in OMIM: 110100 (trait)

Mendelian trait/disorder: yes

Disease-related: no

Key variant known: yes

Year key variant first reported: 2011

Cross-species summary: Horns are paired appendages with a bony core that is attached to the skull and a keratin outer sheath. There is substantial variation in the extent of horn growth, making classification difficult. However, in general, the presence or absence of horns can be attributed to the action of two alleles at an autosomal locus. The type of gene action varies considerably between and among species. Absence of horns is called 'polled'. See also entries for 'scurs'.

Species-specific symbol: Ho

Species-specific description: A fascinating evolutionary story concerning the polled/horns locus in Soay sheep was uncovered by Johnston et al. (2013), who showed that the polymorphism at the polled/horns locus in Soay sheep "is maintained by a trade-off between natural and sexual selection . . . We found that an allele conferring larger horns, Ho+, is associated with higher reproductive success, whereas a smaller horn allele, HoP, confers increased survival, resulting in a net effect of overdominance (that is, heterozygote advantage) for fitness at RXFP2." Cheng et al. (2023) analysed “whole-genome sequencing individuals of 1,098 domestic sheep from 154 breeds, and 69 wild sheep from seven Ovis species. … several introgressed haplotypes contributed to the morphological differentiations across sheep breeds, such as a RXFP2 haplotype from Iranian mouflon conferring the spiral horn trait … .”

History: A comprehensive review of the inheritance of horns in sheep was provided by Dolling (1970), while Tibau i Font et al. (1982) compiled a bibliography.

Inheritance: In his series of papers in 1960 and 1961 on the inheritance of polled/horns in Australian sheep, Dolling reported that polled is dominant to horns in Merino, with the effect of the horns allele being greater in males than in females; that the Dorset Horn breed is homozygous for the dominant polled allele; and that polled is dominant to horns in the Border Leicester breed, with dominance being "complete in the female and partially so in the male". In their report on mapping the polled/horns locus in their New Zealand Romney x Merino population, Pickering et al. (2009) cited the Soay mode of inheritance for polled/horns, namely that polled is recessive, as described by Clutton-Brock and Pemberton (2004). The most recent summary of the type of gene action in Soay sheep is by Johnston et al. (2011): in females, the two alleles are additive (Ho+/Ho+ = horned; Ho+/HoP = scurred; HoP/HoP = polled); and in males, the horned allele is dominant (Ho+/Ho+ and Ho+/HoP = horned, and HoP/HoP are either horned or scurred with roughly equal frequency). It is, of course, quite possible for the type of gene action to vary between breeds.

Mapping: Vaiman et al. (1996) discovered that the horns locus in goats is located in the distal region of goat chromosome 1. Since this chromosome of goats is homologous to one of the arms of sheep chromosome 1, it seemed possible that the horns gene in sheep is located on sheep chromosome 1. Interestingly, the horns gene in cattle is located on cattle chromosome 1, which is homologous to goat chromosome 1. However, the locations of the bovine and caprine horns genes are not homologous (Vaiman et al., 1996). The debate was settled by Montgomery et al. (1996; next paragraph) who showed that the horns locus in sheep is actually located on chromosome 10. Thus the horns loci of sheep, cattle and goats all appear to be different. Montgomery et al. (1996) mapped this locus to the proximal end of OAR10 in a Romney-Merino cross flock, showing no recombinants with microsatellite AGLA226, whose location is OAR10q13. Beraldi et al. (2006) confirmed this location, in Soay sheep. Johnston et al. (2010) narrowed the region to 7.4cM in Soay sheep. Pickering et al. (2009) narrowed the region in domestic (New Zealand Romney x Merino backcross) sheep to 200kb. The OAR10 location of horns/polled in sheep bears no homology to the BTA1 location of horns/polled in cattle, suggesting that the same trait in the two species is determined by different genes. In a genome-wide association study (GWAS) on 486 Soay sheep, each genotyped with the Ovine SNP50 BeadChip (yielding 35,831 informative SNPs) and phenotyped into normal, scurred and polled, Johnston et al. (2011) narrowed down the location of the horns/polled locus in Soay sheep to a 250kb region at around 29.4Mb on chromosome OVA10. A GWAS analysis on EBV for horn length on 160 of the above sheep confirmed this location. In Australian Merino sheep, Dominik et al. (2012) genotyped each of 10 phenotyped rams and 918 of their phenotyped offspring with the Illumina ovine 50k SNP chip, yielding 48,640 informative SNPs. A linkage-disequilibrium analysis mapped the trait to "a SNP located at Mb position 29.38 on sheep chromosome [OVA]10", the nearest gene being RXFP2. Also in 2012, when Kijas et al. genotyped each of 2,819 sheep from 74 different breeds with the Illumina Ovine 50K SNP chip and searched for signatures of selection by calculating FST for each of the 49,034 informative SNPs, they discovered that "the strongest selection signal was identified immediately adjacent to RXFP2 [at Mb position 29.54 on chromosome OVA10] . . . Strong evidence supports that RXFP2 was targeted by breeding for the removal of horns, likely to be one of the oldest morphological modifications that accompanied domestication".

Molecular basis: The gene nearest to the SNPs with the strongest association in the candidate region on OVA10 is the gene for Relaxin-like receptor 2 (RXFP2). Johnston et al. (2011) provided strong evidence for this gene's primary involvement in horn development. Specifically, they designated the two alleles of a C>T SNP in the 3' UTR of RXFP2 as alleles Ho+ (wild type) and HoP (polled or scurred), respectively. Gene expression studies of RXFP2 by Allais-Bonnet et al. (2013) showed differential expression in horn-bud tissues from polled and horned cattle, thereby establishing a link between polledness in sheep and cattle. By sequencing the candidate RXFP2 region in seven Swiss sheep breeds, Wiedemar and Drögemüller (2015) discovered a "1833-bp genomic insertion located in the 3'-UTR region of RXFP2 present in polled animals only". Interestingly, they provided "evidence that the polled-associated insertion adds a potential antisense RNA sequence of EEF1A1 to the 3'-end of RXFP2 transcripts". He et al. (2016) reported that "a PCR analysis for the detection of the 1.8-kb insertion associated with polled sheep in other breeds failed to verify the association with polledness in the three Chinese breeds" Following a large-scale investigation, Lühken et al. (2016) reported that "Multiplex PCR genotyping of 489 sheep from 34 breeds and some crosses between sheep breeds showed a nearly perfect segregation of the insertion polymorphism with horn status in sheep breeds of Central and Western European origin. In these breeds and their crossings, heterozygous males were horned and heterozygous females were polled." However, "in breeds with sex-dependent and/or variable horn status, especially in sheep that originated from even more southern European regions and from Africa . . . we observed almost all possible combinations of genotype, sex and horn status phenotype". Thus, the insertion appears to be causal except in breeds "with sex-dependent and/or variable horn status".

Associated gene: