A genome-wide association study for recurrent laryngeal neuropathy in the Thoroughbred horse identifies a candidate gene that regulates myelin structure

BackgroundEquine recurrent laryngeal neuropathy (RLN) is an economically important upper respiratory tract (URT) disease with a genetic contribution to risk, but genetic variants independent of height have not been identified for Thoroughbreds. The method of clinical assessment for RLN is critical to accurately phenotype groups for genetic studies.ObjectivesTo identify genetic risk loci for RLN in Thoroughbreds in a genome-wide association study (GWAS) following high-resolution phenotyping.Study designCase-control.MethodsThoroughbred horses were characterised as RLN cases and controls using resting and exercising URT endoscopic examinations and laryngeal ultrasonography, with the case-cohort supplemented using a questionnaire. Genotypes for 43 831 autosomal single-nucleotide polymorphisms (SNPs) from n = 235 horses (n = 110 cases; n = 125 controls) were used to estimate trait heritability and identify significantly associated SNPs in a GWAS. Haplotypes were examined in cases and controls and risk allele frequencies were examined in a population cohort (n = 3126).ResultsHeritability was h2 = 0.30 including sex and 5PCs as covariates. A SNP on ECA20 located between candidate genes, DAAM2 and LRFN2, was significantly associated with RLN. Six index SNPs with allelic effect sizes OR = 1.5-2.9 were identified on ECA1, ECA14, and ECA20 close to candidate genes ATPA10, KCNN2, and TFAP2A. Eleven ECA20 SNPs defined seven haplotypes with homozygous H2/H2 horses having a 3.1x higher risk of RLN. Risk alleles segregate in the population, and stallions are carriers.Main limitationsThe main study population was young. Horses in the control group had no evidence of RLN as 2- or 3-year olds but may have developed RLN later.ConclusionsGenetic markers for RLN were identified which may be useful for the development of a polygenic risk score. Candidate genes with functions in neuropathies may further the understanding of RLN pathobiology. HistorialLa neuropat & iacute;a lar & iacute;ngea recurrente (RNL) en el equino es una enfermedad del tracto respiratorio superior (URT) econ & oacute;micamente importante con una contribuci & oacute;n gen & eacute;tica al riesgo, cuyas variantes gen & eacute;ticas independientes a la altura no han sido identificadas en el Fina Sangre de Carrera. El m & eacute;todo de diagn & oacute;stico cl & iacute;nico de RNL es cr & iacute;tico para determinar grupos fenot & iacute;picos para estudios gen & eacute;ticos.ObjetivosIdentificar loci de riesgo gen & eacute;tico para RLN en Fina Sangre de Carrera en un estudio de asociaci & oacute;n de genoma generalizado (GWAS) despu & eacute;s de fenotipificaci & oacute;n de alta resoluci & oacute;n.Dise & ntilde;o del EstudioCaso-control.M & eacute;todosCaballos Fina Sangre de Carrera fueron caracterizados como casos RNL y controles usando examinaciones en reposo y durante ejercicio por endoscopia del URT y ultrasonido de la laringe, con la cohorte de casos suplementados usando un cuestionario. Lo genotipos de 43 831 polimorfismos autosomales de nucle & oacute;tido singular (SNPs) de n = 235 caballos (n = 110 casos; n = 125 controles) fueron usados para estimar la heredabilidad del rasgo e identificar SNPs asociados en forma significativa en un GWAS. Haplotipos fueron examinados en los casos y controles, y las frecuencias de los alelos de riesgo fueron examinadas en una cohorte de la poblaci & oacute;n (n = 3126).ResultadosLa heredabilidad fue de h2 = 0.30 incluyendo sexo y 5PCs como covariantes. Un SNP en ECA20 localizado entre los genes candidatos DAAM2 y LRFN2, fue asociado en forma significativa con la RLN. Seis SNPs & iacute;ndices con efecto al & eacute;lico de tama & ntilde;o OR = 1.5-2.9 fueron identificados en ECA1, ECA14 y ECA20 cerca de los genes candidatos ATPA10, KCNN2, y TFAP2A. Once SNPs en ECA20 definieron siete haplotipos en caballos homocigotos H2/H2 de tener un riesgo 3.1x mayor de RLN. Los alelos de riesgo son segregados en la poblaci & oacute;n y los potros son portadores.Limitaciones PrincipalesLa poblaci & oacute;n principal era joven. Los caballos en el grupo control no ten & iacute;a evidencia de RLN a los dos o tres a & ntilde;os, pero pueden haber desarrollado RLN m & aacute;s tarde.ConclusionesSe identificaron marcadores gen & eacute;ticos para RLN los cuales pueden ser & uacute;tiles para desarrollar un puntaje de riesgo polig & eacute;nico. Los genes candidatos con funciones en neuropat & iacute;as podr & iacute;an aumentar el conocimiento de la patobiolog & iacute;a de RLN.