Skip to main content

Screening of whole genome sequences identified high-impact variants for stallion fertility



Stallion fertility is an economically important trait due to the increase of artificial insemination in horses. The availability of whole genome sequence data facilitates identification of rare high-impact variants contributing to stallion fertility. The aim of our study was to genotype rare high-impact variants retrieved from next-generation sequencing (NGS)-data of 11 horses in order to unravel harmful genetic variants in large samples of stallions.


Gene ontology (GO) terms and search results from public databases were used to obtain a comprehensive list of human und mice genes predicted to participate in the regulation of male reproduction. The corresponding equine orthologous genes were searched in whole genome sequence data of seven stallions and four mares and filtered for high-impact genetic variants using SnpEFF, SIFT and Polyphen 2 software. All genetic variants with the missing homozygous mutant genotype were genotyped on 337 fertile stallions of 19 breeds using KASP genotyping assays or PCR-RFLP. Mixed linear model analysis was employed for an association analysis with de-regressed estimated breeding values of the paternal component of the pregnancy rate per estrus (EBV-PAT).


We screened next generation sequenced data of whole genomes from 11 horses for equine genetic variants in 1194 human and mice genes involved in male fertility and linked through common gene ontology (GO) with male reproductive processes. Variants were filtered for high-impact on protein structure and validated through SIFT and Polyphen 2. Only those genetic variants were followed up when the homozygote mutant genotype was missing in the detection sample comprising 11 horses. After this filtering process, 17 single nucleotide polymorphism (SNPs) were left. These SNPs were genotyped in 337 fertile stallions of 19 breeds using KASP genotyping assays or PCR-RFLP. An association analysis in 216 Hanoverian stallions revealed a significant association of the splice-site disruption variant g.37455302G>A in NOTCH1 with the de-regressed estimated breeding values of the paternal component of the pregnancy rate per estrus (EBV-PAT). For 9 high-impact variants within the genes CFTR, OVGP1, FBXO43, TSSK6, PKD1, FOXP1, TCP11, SPATA31E1 and NOTCH1 (g.37453246G>C) absence of the homozygous mutant genotype in the validation sample of all 337 fertile stallions was obvious. Therefore, these variants were considered as potentially deleterious factors for stallion fertility.


In conclusion, this study revealed 17 genetic variants with a predicted high damaging effect on protein structure and missing homozygous mutant genotype. The g.37455302G>A NOTCH1 variant was identified as a significant stallion fertility locus in Hanoverian stallions and further 9 candidate fertility loci with missing homozygous mutant genotypes were validated in a panel including 19 horse breeds. To our knowledge this is the first study in horses using next generation sequencing data to uncover strong candidate factors for stallion fertility.


A major challenge in current equine genomics is to understand the genetic basis of mutations influencing stallion fertility. In Hanoverian stallions, several studies implicated loci that have been shown to play a significant role in stallion fertility. Mutations within CRISP3 [1, 2], SPATA1, INHBA, ACE, SP17, FSHB, PRLR [35], PLCz1 [6] and FKBP6 [7] were significantly associated with stallion fertility. Abnormalities in number of X- and/or Y chromosomes have been shown as causes for abnormal sexual development [812]. Until now, only few mutations were identified to be associated with sub- or infertility in stallions. FKBP6 genotyping was recommended for stallions affected by impaired acrosome reaction (IAR). In human, only 20 % of male factor infertility can be diagnosed using genetic testing [13]. Whole-genome sequencing approaches facilitate a comprehensive identification of high-impact variants with potentially deleterious effects on traits analysed [14]. Rare genetic variants are more likely to be functionally deleterious and to cause loss-of function of coding proteins [14, 15]. Recently, whole-genome sequencing was used to uncover the roles of rare genetic variants associated with spermatogenic failure in men [16]. To date, no whole-genome sequence screens were performed for stallion fertility. We employed next-generation sequence (NGS) data to screen male fertility associated genes in order to unravel the role of genetic variants for which damaging effects on the protein structure were predicted. Herein, we employed bioinformatic analysis to identify human and mice genes linked through common gene ontology (GO) male reproductive processes. In order to focus the search for rare high-impact variants within the gene list, we filtered the whole genomes sequence data for variants which were only present as wild type homozygotes or heterozygotes in fertile horses. We assumed that in fertile stallions, variants with deleterious effects on fertility should be restricted to heterozygous genotypes for the recessive alleles or the homozygous wild type genotype for the dominant allele, and the presence of one of these variants may reduce stallion fertility. The objective of the present study was to genotype rare high-impact variants retrieved from NGS-data of 11 horses in order to unravel harmful genetic variants in a large sample of stallions.


Search for genes involved in stallion reproduction

Gene ontology (GO) terms for male reproductive processes and Ensembl and NCBI databases were used to search for genes with an effect on male fertility in human and mice. We retrieved a list of 1256 genes identities (Additional file 1). Using g:Profiler [17, 18] we could convert 1194 genes to known equine orthologs and had to exclude 62 genes due to unknown equine annotation or unknown chromosomal position. Using PANTHER classification system [19, 20], the 1194 male fertility related genes were assigned to 67 pathways, 85 molecular functions, 82 biological processes and 16 cellular components.

Mutation detection in 1194 equine genes

We searched the whole genome sequences of all 11 horses for genetic variants within the 1194 male reproduction related genes. The first filter sorted out all genetic variants for which both homozygous genotypes and the heterozygous genotypes were present. This filter left 1259 genetic variants with influence on the coding sequence and contained only variants for which the mutant homozygous genotype was not present. The second filtering for high impact variants retrieved a total of 19 SNPs within 18 genes and 38 indels within 31 genes (Additional file 2). Further evaluation was restricted to SNP variants only. Of the 19 SNPs, 16 SNPs were located in 15/1194 genes for stallion reproduction and three SNPs (ENSECAG00000020135, ENSECAG00000021286 and ENSECAG00000018118) were not obviously related to male reproduction, but located within uncharacterized genes or pseudogenes on the complementary DNA strand (Table 1, Additional file 3). The most frequently predicted high-impact effects were donor splice site disruptions (n = 7) and stop codon gains (n = 6) (Table 2). We could assign these 15 genes with possibly rare high-impact variants to four pathways, 14 molecular functions, 14 biological processes and one cellular component. Using GeneMANIA, we constructed a gene-gene functional network for 13/15 genes based on GO annotation (Additional file 4).

Table 1 Comparison of SNPs with high effect on protein structure in genes for stallion fertility among breeds using whole-genome sequences of fertile stallions
Table 2 Identification of variants with high effect on protein structure in genes for stallion fertility

Validation of high-impact SNPs in fertile stallions

In total, we genotyped 19 SNPs whereof two SNPs, one in FNDC3A and one SNP on the complementary DNA strand (ENSECAG00000018118), proved monomorphic in a sample of 96 stallions. Out of the 17 remaining SNPs genotyped, 15 SNPs were related to male reproduction. Two SNPs (ENSECAG00000020135, ENSECAG00000021286) were located within uncharacterized genes/pseudogenes on the complementary DNA strand whereas these SNPs had no high impact within the candidate genes on protein structure. Genotyping for the 17 polymorphic SNPs was performed in 337 stallions from 19 different breeds using PCR-RFLP and competitive allele specific PCR (KASP) genotyping technique (Additional file 5). Three SNPs within NEURL1, GHRL and BTNL2 deviated from Hardy Weinberg equilibrium (HWE) (Table 3). For all SNPs, the minor allele was represented by the mutant allele with a mean minor allele frequency (MAF) of 0.07 (range = 0.002–0.306). For 6/17 variants all three genotypes were found, of which the mutant homozygous genotypes were underrepresented with a mean genotype frequency of 0.06 (range = 0.003–0.223). For 9/17 genetic variants, no stallion exhibited the mutant homozygous genotype (Table 4). The expected number of mutant homozygous genotypes for stallions, calculated as q2 (mutant allele frequency squared) x number of stallions (n) stallions, revealed two variants within FOXP1 and OVGP1 with expected number of genotypes of 0.92 and 1.04 (Table 5). The variants within ENSECAG00000020135 and ENSECAG00000021286 were homozygous wild type in all stallions (Table 4).

Table 3 Validation of 17 high-impact SNPs in 337 fertile stallions
Table 4 Minor allele (MA), minor allele frequencies (MAF), minor genotype frequencies (MAG) for high-impact variants within candidate genes for stallion fertility validated in 337 fertile stallions
Table 5 Evaluation of high-impact variants for stallion fertility

We grouped the genetic variants genotyped into four groups (I-IV) according to their minor allele frequency (MAF). In group I, the mutant allele did not occur (MAF = 0), in group II, the mutant allele occurred in a very low frequency, which were classified as rare variants (MAF < 0.01) [14]. In group III, the mutant allele occurred in a low frequency (MAF > 0.01–0.1), and in group IV the mutant allele had a moderate frequency (MAF > 0.1). Variants in non-male reproduction related genes ENSECAG00000020135 and ENSECAG00000021286 were sorted into group I. Variants in CFTR, PKD1 and SPATA31E1 were classified as rare variants (group II). Eight variants were sorted into group III and four variants into group IV (Table 5). The distribution of the variants genotyped within discovery breeds and validation cohorts should allow distinguish between private SNPs and SNPs shared by several breeds. With the exception of two private Icelandic SNPs within ENSECAG00000020135 and ENSECAG00000021286, all high-impact SNPs were shared by 2–12 breeds (Additional file 6).

Association analysis in Hanoverian stallions

In 216 Hanoverian stallions, a nominally significant association with de-regressed EBV-PAT was found for the NOTCH1 variant g.37455302G>A (P = 0.00003) using a model fitting the genotypes. Accounting for multiple testing of 15 SNPs using Bonferroni correction, resulted in a significant association for the NOTCH1 variant g.37455302G>A (P = 0.00045) explaining 11 % of the variation of de-regressed EBV-PAT. A mixed model analysis accounting for genomic relationships among stallions gave a P-value of 0.00052 and after Bonferroni correction a P-value of 0.00828. Hanoverian stallions exhibiting the wild-type for the NOTCH1-associated SNP had a de-regressed EBV-PAT of 101 onto the scale 100 ± 20 indicating stallion fertility closely at the population average. The differences among the NOTCH1-homozygotes were 41 points of the de-regressed EBV-PAT (P-value < 0.0001). Stallions heterozygous for the mutant NOTCH1-variant had a mean of 100 for the de-regressed EBV-PAT. Other variants failed the nominal significance thresholds (Additional file 7). Stallions heterozygous or homozygous for the KDR-variant showed a higher mean of de-regressed EBV-PAT than homozygous wild type stallion. Due to high standard errors and low genotype frequencies, no significant associations, but a tendency among the homozygous and heterozygous genotype means for de-regressed EBV-PAT were found for six SNPs within NEURL1, OVGP1, FOXP1, NOTCH1 (g.37453246G>C), SPATA31E1 and FBXO43 (Additional file 7).

Classification of high-impact variants

Based on validation and association results, we distinguished three classes among the high-impact variants (Table 6). The first class (class I) regards loci with a significant association with EBV-PAT and a low to moderate allele frequency. Only the NOTCH1 variant g.37455302G>A fulfills these requirements for class I loci. The second class (class II) should contain loci which may be considered as putative indicators for fertility. Herein, we classified 10 variants as class II loci. The third class (class III) comprises high-impact variants with no obvious significant relation to stallion reproduction. As class III loci we sorted in two private SNPs, only present in Icelandic stallions and variants with all three genotypes but not significantly associated with de-regressed EBV-PAT.

Table 6 Most promising high-impact variants in genes contributing to stallion fertility


Screening of whole genome sequences for high-impact genetic variants within gene networks for male fertility in stallions appeared useful to identify variants that may highly influence stallion fertility. All high-impact variants we classified due to stop codon gains, start losses, donor or acceptor splice site disruptions can be classified as loss- of-function (LOF) mutations [14, 21]. A major effect on fertility by loss-of-function mutations was demonstrated in previous studies [2225]. Data from this study indicate a high-impact variant as a stallion fertility-associated locus (class I) in NOTCH1 and 9 high-impact variants as putative stallion fertility-associated loci (class II). Genes putatively associated with stallion fertility included CFTR, OVGP1, FBXO43, TSSK6, PKD1, FOXP1, TCP11, SPATA31E1 and NOTCH1 (g.37453246G>C). These genes are connected with each other via functional interacting networks based on GO male fertility processes.

In mice, NOTCH1 gain-of-function resulted in reduced male fertility due to failure of spermatogenesis [26], while immotile spermatozoa and sterility was detected in NEURL1 null mice [27]. The fertile stallions used for validation were frequently uses in artificial insemination, thus sperm motility and morphology are in the normal range in these stallions. Fertile stallions that exhibited unfavorable genotypes in NOTCH1-high-impact variants had de-regressed EBV-PAT as well EBV-PAT below the population mean suggesting reduced fertility due to other unknown damaging mechanism on sperm function. For the KDR high-impact variant g.77472655G>C in KDR, heterozygous or homozygous carriers showed putatively higher de-regressed EBV-PAT and EBV-PAT implicating a beneficial effect of this variant. The latter is supported by the ‘less-is-more’ hypothesis [28] and an advantageous effect of deleterious variants due to positive selection [29]. An advantageous effect on reproduction of deleterious variants alleles due to positive selection was recently demonstrated in buffalo [30]. NOTCH1, NEURL1 and KDR are key factors for male fertility, making their application for stallion fertility highly desirable. NOTCH1 is part of the Notch signaling network and regulates interactions between physically adjacent cells [31]. NEURL1 is required for axonemal integrity in spermatozoa [27]. It functionally belongs to spermatogenesis (GO:0007283) and is associated to spermatid differentiation (GO:0048515), sperm motility (GO:0030317) and spermatid differentiation (GO:0048515). KDR is present in spermatozoa [32] and was identified as a regulator of germ cell survival during establishment of spermatogenesis in cattle [33].

Based on our hypothesis of missing mutant genotypes in fertile stallions, we detected 9 putative loci for stallion fertility defined as class II variants. These non-private and high-impact variants were only present in very low to or low frequency. The expected frequency of mutant homozygous genotypes for stallions E(q2 stallion) indicated a low-probability of their occurrence in fertile stallions. As an explanation for the missing homozygous mutant genotypes in fertile stallions one can assume that homozygous mutant genotypes are limited to infertile stallions. This may be further tested in an unselected sample of male foals under the assumption that these mutations are still segregating in very few males with the homozygous mutant genotype may be infertile.

Class II variants within CFTR and TSSK6 indicated high-impact on male fertility, confirmed by several studies in human and mice. In infertile men, high-effect variants at acceptor splice site in CFTR were found in non-obstructive azoospermia, oligospermia, asthenospermia and teratospermia men [34, 35]. Infertility due to spermatogenic impairment was reported in male TSSK6 knockout mice [36] and in Chinese men exhibiting a triallelic SNP in TSSK6 [37]. PKD1 is required for male reproductive tract development. PKD1-null mice develop multiple organ defects in the male reproductive tract and infertility [38]. TCP-11 modulates sperm fertilizing ability. It stimulates capacitation and inhibits spontaneous acrosome loss [39, 40]. Subfertility in human could result from alterations in the structure of TCP-11 [39].

Validation of class II high-impact variants within SPATA31E1, OVGP1, FOXP1, FBXO43 indicate further putative loci for stallion fertility, however the consequences of mutations on male infertility in human and mice are not yet known. Nevertheless, SPATA31E1 functionally belongs to spermatogenesis (GO:0007283). OVGP1 is linked to single fertilization (GO:0007338) via binding to oocyte zona pellucida [41]. It enhances sperm viability, motility [42] and capacitation and shows beneficial effects on fertilization and early embryo development [43, 44]. FOXP1 is an androgen-responsive transcription factor that modulates androgen receptor signaling [45]. FBXO43 place a role in sperm-induced meiotic exit [46] and adjustment of the number of round spermatids [47].


Out of 1194 equine genes involved in male fertility processes we identified the g.37455302G>A NOTCH1 variant as a high-impact SNP with a significant effect on Hanoverian stallion fertility. As this detected high-impact-variant only occurs in a rare frequency in fertile stallions of many different breeds we recommend selection against this unfavourable genotype to improve stallion fertility. Next, we identified 9 high-impact SNPs in CFTR, OVGP1, FBXO43, TSSK6, PKD1, FOXP1, TCP11, SPATA31E1 and NOTCH1 (g.37453246G>C) absent in fertile stallions with putative deleterious effects in fertility of stallions. As a consequence, an early identification of carriers in unselected young stallions should be a beneficial to select only high-quality stallions for future breeding.


Ethics statement

All animal work has been conducted according to the national and international guidelines for animal welfare. The Lower Saxony state veterinary office at the Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit, Oldenburg, Germany, was the responsible Institutional Animal Care and Use Committee (IACUC) for this specific study. The EDTA-blood sampling for the present study had been approved by the IACUC of Lower Saxony, the state veterinary office Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit, Oldenburg, Germany (registration number 33.42502-05-07A482).

Search for genes effecting stallion reproduction

Gene ontology (GO) terms related to male reproduction were used to obtain a comprehensive list of known human und mice genes predicted to participate in the regulation of male reproduction. We used the go.obo download file which contains core GO ontology terms provided by the gene ontology consortium (, accessed 02.08.2014). In addition, a keyword search for annotated genes in the genome browsers Ensembl and NCBI for ‘male reproduction’, ‘male infertility’, ‘male gonad development’,spermatogenesis’, ‘acrosome reaction', was conducted. We then searched NCBI PubMed with keywords for mutations within genes affecting male reproduction in human and livestock animals (including horses) from articles published before November 1st 2014. We employed Ensembl BioMart [48, 49] ( to retrieve all human genes (Ensembl gene ids) associated with a GO term and converted human genes into equine orthologs using g.Profiler [17, 18] ( If an equine orthologous gene was not available in Ensembl we searched for uncharacterized novel equine orthologs based on human annotations from the Ensemble genome browser ( or searched for annotations matching the NCBI genome database ( For equine annotations, chromosomes (ECA) and gene start and stop positions in base pairs (bp) were extracted using Ensembl BioMart. Genes were mapped to GO annotations (molecular function biological process, cellular component) using PANTHER (Protein analysis through evolutionary relationships, version 8.0) database [19, 20] ( The database GeneMania [50] ( was used to visualise co-expression and common pathways for networks of male fertility related genes.

Detection of deleterious mutations

We screened publicly available whole genome sequences of seven stallions including each one Icelandic, Standardbred and Przewalski stallion [51], two Hanoverian stallions and each one Arabian and Sorraia stallion [52], and four mares including each one Arabian, Norwegian fjord [51], Dülmen Horse [52], and the re-sequenced reference genome of the Thoroughbred mare Twilight [51] for mutations within selected candidate genes for stallion fertility.

Variants within candidate genes were extracted according to the equine chromosomal positions obtained from the Ensembl annotation. We excluded all genetic variants with hits in the genome databases dbSNP (, download 19.09.2013), Broad Institute (*.xls, download 19.09.2013) and Ensembl (, download 25.07.2014).

Functional analysis of variants obtained from re-sequencing data was performed using SnpEFF software [53]. The SnpEFF impact category HIGH was used to filter SNPs for putative high-impact effects (large chromosome deletion, exon deletion, insertion/deletion frame shift, donor splice site disruptions, acceptor splice site disruptions, stop codon gains, stop codon losses, start losses). Validation was done using SIFT and Polyphen 2. High-impact effect variants were considered as potentially deleterious or expected to potentially correlate with complete loss of function (LoF) of the affected transcripts [14, 54]. In order to identify rare variants influencing stallion reproduction we filtered high-impact variants within candidate genes for stallion fertility with missing homozygous mutant genotypes in fertile stallions. Under this scenario, we expected that homozygosity for those variants is less likley in fertile stallions. Analyses were performed using SAS/Genetics, version 9.4 (Statistical Analysis System, Cary, NC, 2015).

Validation population

In total, we analysed 337 fertile stallions of 19 breeds (Additional file 8). The largest proportion of the fertile horses were Hanoverian stallions (n = 226), used for artificial insemination at the National State Stud Celle of Lower Saxony in Celle. For 216 Hanoverian stallions breeding values of the paternal component of the pregnancy rate per estrus cycle (EBV-PAT) and de-regressed EBV-PAT were available. The mean reliabilities of the EBVs were at 0.7. For fertile stallions other than Hanoverian, at least 10 viable offspring were reported, targeting those stallions as fertile.

Validation of high-impact variants

We assumed that the mutant allele with high impact on protein structure should be completely absent or in a very low frequency in fertile stallions. We genotyped high-impact SNPs in fertile stallions using polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) and competitive allele specific PCR (KASP) genotyping technique [55] (LGC Genomics GmbH, Berlin, Germany) (Additional file 5). PCR-RFLP primers were designed using Primer3 [56, 57] ( DNA was amplified according to standard protocols on PTC 100 thermal cyclers, MJ Research, Watertown, MA, US. For enzymatic digestion, we used 15 μl reaction volume containing 1.5 μl buffer, 1.5 U endonucleases and 5 μl PCR amplicons. The KASP genotyping reaction was run in 5 μl KASP Master mix 2x (FAM and HEX dye-labeled FRET cassette, Taq polymerase, buffer), 0.14 μl KASP Assay mix (KASP-by-design: two allele-specific primers, one common primer) and 5 μl template DNA. Allelic discrimination was carried out using the Applied Biosystems 7300 Real-Time PCR System (Life Technologies).

Statistical analysis

The ALLELE procedure of SAS/Genetics was used to calculate allele and genotype frequencies, minor allele frequencies (MAF), observed heterozygosity (HET) and χ2-tests for Hardy-Weinberg equilibrium for the SNPs genotyped.

Calculation of EBV-PAT in Hanoverians and the model used for estimation are described elsewhere [58]. In brief, data and model employed are given in the following. Fertility data included the breeding seasons 1997 to 2005 of the National State stud Celle of Lower Saxony with a total of 19,897 broadmares, 246 stallions, 199,000 artificial insemination records and 96,114 estrus cycles. Trait analysed was the pregnancy rate per estrus. The trait values were encoded 1 when the artificial insemination (AI) was successful in an estrus cycle otherwise the trait value was 0. The success rate of AIs was verified through the reports of the breeders on the pregnancy of the mare and/or a foaling and/or an abortion. All estrus periods with consecutive AIs in the same breeding season were treated as unsuccessful. In addition, estrus periods reported by the breeder as unsuccessful (missing pregnancy of the mare) were also encoded with a trait value 0. We predicted breeding values (EBVs) for the paternal and embryonic component for each stallion. The animal threshold model for prediction of EBVs included the fixed environmental effects of insemination centre, age of stallion, breeding season, period within breeding season, the number of coverings within breeding season (covering number), time interval between coverings within a estrus (insemination regime), breeding history of mares (previous breeding achievement of mares), the random permanent environmental effects of the mare and stallion and the random additive genetic effects of the stallion and the embryo. The analyses were performed using MTGSAM and Gibbs sampling [59, 60] for estimation of random effects and variance components. Heritability for the paternal component of the pregnancy rate per estrus was 1.1 %. All EBVs were standardized onto a mean of 100 and a standard deviation of 20. The random additive genetic effect of the stallion was defined as the paternal component of the pregnancy rate per estrus (EBV-PAT). EBVs >100 mean higher conception rates of the stallions than the population average. An association analysis for each SNP genotype was performed using the procedure GLM of SAS, version 9.4 (Statistical Analysis System, SAS Institute, Cary, NC, USA). The proportion of variance explained for (de-regressed) EBV-PAT was estimated using the GLM procedure of SAS. In order to account for the population structure a mixed linear model (MLM) was employed using the respective genotype as fixed effect and a random animal effect through an identity-by-state-kinship (IBS) matrix. The IBS matrix reflects the genomic relationship matrix among all individuals genotyped and captures the relatedness among animals as well as the cryptic family structure. We used genotypings from the Equine SNP50 Beadchip (Illumina, San Diego, CA, USA) including 54,602 SNPs. Quality criteria were minor allele frequency (MAF) >0.05, genotyping rate per SNP and animal >0.90 and HWE (P < 0.00001). After filtering for quality criteria, 46,074 SNPs remained for building the IBS-matrix [6]. The analysis was run using TASSEL, version 3.0.146 [61]. The Bonferroni correction was calculated using the MULTIPLE TEST procedure of SAS, version 9.4 to determine the threshold for experiment-wide significance.

Accession numbers

Newly discovered polymorphisms were submitted to NCBI dbSNP database (ss1457622626 -ss1457622648). The whole genome sequences of four stallions, one Arabian, two Hanoverian and one Sorraia are available at the NCBI Sequence Read Archive (SRA) (, accession number SRP033361. Corresponding VCF files can be downloaded at Intrepid Bioinformatics at The whole genome sequences of three stallions, one Icelandic, Standardbred and Przewalski and the re-sequenced reference genome of the Thoroughbred mare ‘Twilight’ can be found at SRA under accession number SRA082086. Corresponding BAM and VCF files can be downloaded at





Black Forest Horses


cystic fibrosis transmembrane conductance regulator


Dülmen Horse


estimated breeding values for the paternal component of the pregnancy rate per estrus


F-box only protein 43


forkhead box protein P


German Riding Pony


appetite-regulating hormone


gene ontology






Hardy Weinberg equilibrium




competitive allele specific PCR




minor allele frequency


Mecklenburg Cold Blood


neuralized E3 ubiquitin protein ligase 1


next-generation sequences


Norwegian fjord


neurogenic locus notch homolog protein 1




oviduct-specific glycoprotein


polymerase chain reaction






restriction fragment length polymorphism


Rhenish Cold Blood






Schleswig Cold Blood


Southern German Cold Blood


single nucleotide polymorphism




spermatogenesis-associated protein 31E1


Saxon Thuringia Cold Blood




t-complex protein 11 homolog




testis-specific serine/threonine-protein kinase 6




  1. 1.

    Hamann H, Jude R, Sieme H, Mertens U, Töpfer-Petersen E, Distl O, Leeb T. A polymorphism within the equine CRISP3 gene is associated with stallion fertility in Hanoverian warmblood horses. Anim Genet. 2007;38(3):259–64.

    CAS  Article  PubMed  Google Scholar 

  2. 2.

    Giese A, Jude R, Kuiper H, Piumi F, Schambony A, Guerin G, Distl O, Topfer-Petersen E, Leeb T. Molecular characterization of the equine AEG1 locus. Gene. 2002;292(1-2):65–72.

    CAS  Article  PubMed  Google Scholar 

  3. 3.

    Giesecke K, Hamann H, Stock KF, Wöhlke A, Sieme H, Distl O. Evaluation of SPATA1-associated markers for stallion fertility. Anim Genet. 2009;40(4):359–65.

    CAS  Article  PubMed  Google Scholar 

  4. 4.

    Giesecke K, Hamann H, Stock KF, Klewitz J, Martinsson G, Distl O, Sieme H. Evaluation of ACE, SP17, and FSHB as candidates for stallion fertility in Hanoverian warmblood horses. Anim Reprod Sci. 2011;126(3-4):200–6.

    CAS  Article  PubMed  Google Scholar 

  5. 5.

    Giesecke K, Hamann H, Sieme H, Distl O. Evaluation of prolactin receptor (PRLR) as candidate gene for male fertility in Hanoverian warmblood horses. Reprod Domest Anim. 2010;45(5):e124–130.

    CAS  PubMed  Google Scholar 

  6. 6.

    Schrimpf R, Dierks C, Martinsson G, Sieme H, Distl O. Genome-wide association study identifies phospholipase C zeta 1 (PLCz1) as a stallion fertility locus in Hanoverian warmblood horses. PLoS One. 2014;9(10):e109675.

    Article  PubMed  PubMed Central  Google Scholar 

  7. 7.

    Schrimpf R, Metzger J, Martinsson G, Sieme H, Distl O. Implication of FKBP6 for male fertility in horses. Reprod Domest Anim. 2015;50(2):195–9.

    CAS  Article  PubMed  Google Scholar 

  8. 8.

    Demyda-Peyras S, Anaya G, Bugno-Poniewierska M, Pawlina K, Membrillo A, Valera M, Moreno-Millan M. The use of a novel combination of diagnostic molecular and cytogenetic approaches in horses with sexual karyotype abnormalities: a rare case with an abnormal cellular chimerism. Theriogenology. 2014;81(8):1116–22.

    CAS  Article  PubMed  Google Scholar 

  9. 9.

    Holl HM, Lear TL, Nolen-Walston RD, Slack J, Brooks SA. Detection of two equine trisomies using SNP-CGH. Mamm Genome. 2013;24(5-6):252–6.

    Article  PubMed  Google Scholar 

  10. 10.

    Pujar S, Meyers-Wallen VN. Sequence variations in equine candidate genes For XX and XY inherited disorders of sexual development. Reprod Domest Anim. 2012;47(5):827–34.

    CAS  Article  PubMed  Google Scholar 

  11. 11.

    Kuiper H, Distl O. Intersexuality in horses. Dtsch Tierarztl Wochenschr. 2007;114(2):50–6.

    CAS  PubMed  Google Scholar 

  12. 12.

    Kuiper H, Blum N, Distl O. An XY agonadal Oldenburg warmblood horse exhibiting a male phenotype. Sex Dev. 2010;4(6):348–51.

    CAS  Article  PubMed  Google Scholar 

  13. 13.

    Hotaling J, Carrell DT. Clinical genetic testing for male factor infertility: current applications and future directions. Andrology. 2014;2(3):339–50.

    CAS  Article  PubMed  Google Scholar 

  14. 14.

    MacArthur DG, Manolio TA, Dimmock DP, Rehm HL, Shendure J, Abecasis GR, Adams DR, Altman RB, Antonarakis SE, Ashley EA, et al. Guidelines for investigating causality of sequence variants in human disease. Nature. 2014;508(7497):469–76.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  15. 15.

    Lee S, Abecasis GR, Boehnke M, Lin X. Rare-variant association analysis: study designs and statistical tests. Am J Hum Genet. 2014;95(1):5–23.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  16. 16.

    Lopes AM, Aston KI, Thompson E, Carvalho F, Goncalves J, Huang N, Matthiesen R, Noordam MJ, Quintela I, Ramu A, et al. Human spermatogenic failure purges deleterious mutation load from the autosomes and both sex chromosomes, including the gene DMRT1. PLoS Genet. 2013;9(3):e1003349.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  17. 17.

    Reimand J, Arak T, Vilo J. g:Profiler--a web server for functional interpretation of gene lists (2011 update). Nucleic Acids Res. 2011;39(Web Server issue):W307–315.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  18. 18.

    Reimand J, Kull M, Peterson H, Hansen J, Vilo J. g:Profiler--a web-based toolset for functional profiling of gene lists from large-scale experiments. Nucleic Acids Res. 2007;35(Web Server issue):W193–200.

    Article  PubMed  PubMed Central  Google Scholar 

  19. 19.

    Mi H, Muruganujan A, Casagrande JT, Thomas PD. Large-scale gene function analysis with the PANTHER classification system. Nat Protoc. 2013;8(8):1551–66.

    Article  PubMed  Google Scholar 

  20. 20.

    Mi H, Muruganujan A, Thomas PD. PANTHER in 2013: modeling the evolution of gene function, and other gene attributes, in the context of phylogenetic trees. Nucleic Acids Res. 2013;41(Database issue):D377–386.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  21. 21.

    Rausell A, Mohammadi P, McLaren PJ, Bartha I, Xenarios I, Fellay J, Telenti A. Analysis of stop-gain and frameshift variants in human innate immunity genes. PLoS Comput Biol. 2014;10(7):e1003757.

    Article  PubMed  PubMed Central  Google Scholar 

  22. 22.

    Fritz S, Capitan A, Djari A, Rodriguez SC, Barbat A, Baur A, Grohs C, Weiss B, Boussaha M, Esquerre D, et al. Detection of haplotypes associated with prenatal death in dairy cattle and identification of deleterious mutations in GART, SHBG and SLC37A2. PLoS One. 2013;8(6):e65550.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  23. 23.

    Sonstegard TS, Cole JB, VanRaden PM, Van Tassell CP, Null DJ, Schroeder SG, Bickhart D, McClure MC. Identification of a nonsense mutation in CWC15 associated with decreased reproductive efficiency in Jersey cattle. PLoS One. 2013;8(1):e54872.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  24. 24.

    Fujiwara Y, Matsumoto H, Akiyama K, Srivastava A, Chikushi M, Ann Handel M, Kunieda T. An ENU-induced mutation in the mouse Rnf212 gene is associated with male meiotic failure and infertility. Reproduction. 2015;149(1):67–74.

    Article  PubMed  Google Scholar 

  25. 25.

    Pausch H, Kolle S, Wurmser C, Schwarzenbacher H, Emmerling R, Jansen S, Trottmann M, Fuerst C, Gotz KU, Fries R. A nonsense mutation in TMEM95 encoding a nondescript transmembrane protein causes idiopathic male subfertility in cattle. PLoS Genet. 2014;10(1):e1004044.

    Article  PubMed  PubMed Central  Google Scholar 

  26. 26.

    Huang Z, Rivas B, Agoulnik AI. NOTCH1 gain of function in germ cells causes failure of spermatogenesis in male mice. PLoS One. 2013;8(7):e71213.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  27. 27.

    Vollrath B, Pudney J, Asa S, Leder P, Fitzgerald K. Isolation of a murine homologue of the Drosophila neuralized gene, a gene required for axonemal integrity in spermatozoa and terminal maturation of the mammary gland. Mol Cell Biol. 2001;21(21):7481–94.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  28. 28.

    Olson MV. When less is more: gene loss as an engine of evolutionary change. Am J Hum Genet. 1999;64(1):18–23.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  29. 29.

    Sabeti PC, Schaffner SF, Fry B, Lohmueller J, Varilly P, Shamovsky O, Palma A, Mikkelsen TS, Altshuler D, Lander ES. Positive natural selection in the human lineage. Science. 2006;312(5780):1614–20.

    CAS  Article  PubMed  Google Scholar 

  30. 30.

    van Hooft P, Greyling BJ, Getz WM, van Helden PD, Zwaan BJ, Bastos AD. Positive selection of deleterious alleles through interaction with a sex-ratio suppressor gene in African buffalo: a plausible new mechanism for a high frequency anomaly. PLoS One. 2014;9(11):e111778.

    Article  PubMed  PubMed Central  Google Scholar 

  31. 31.

    Artavanis-Tsakonas S, Matsuno K, Fortini ME. Notch signaling. Science. 1995;268(5208):225–32.

    CAS  Article  PubMed  Google Scholar 

  32. 32.

    Obermair A, Obruca A, Pohl M, Kaider A, Vales A, Leodolter S, Wojta J, Feichtinger W. Vascular endothelial growth factor and its receptors in male fertility. Fertil Steril. 1999;72(2):269–75.

    CAS  Article  PubMed  Google Scholar 

  33. 33.

    Caires KC, de Avila J, McLean DJ. Vascular endothelial growth factor regulates germ cell survival during establishment of spermatogenesis in the bovine testis. Reproduction. 2009;138(4):667–77.

    CAS  Article  PubMed  Google Scholar 

  34. 34.

    Chen H, Ruan YC, Xu WM, Chen J, Chan HC. Regulation of male fertility by CFTR and implications in male infertility. Hum Reprod Update. 2012;18(6):703–13.

    CAS  Article  PubMed  Google Scholar 

  35. 35.

    Tamburino L, Guglielmino A, Venti E, Chamayou S. Molecular analysis of mutations and polymorphisms in the CFTR gene in male infertility. Reprod Biomed Online. 2008;17(1):27–35.

    CAS  Article  PubMed  Google Scholar 

  36. 36.

    Spiridonov NA, Wong L, Zerfas PM, Starost MF, Pack SD, Paweletz CP, Johnson GR. Identification and characterization of SSTK, a serine/threonine protein kinase essential for male fertility. Mol Cell Biol. 2005;25(10):4250–61.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  37. 37.

    Su D, Zhang W, Yang Y, Zhang H, Liu YQ, Bai G, Ma YX, Peng Y, Zhang SZ. c.822+126T>G/C: a novel triallelic polymorphism of the TSSK6 gene associated with spermatogenic impairment in a Chinese population. Asian J Androl. 2010;12(2):234–9.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  38. 38.

    Nie X, Arend LJ. Pkd1 is required for male reproductive tract development. Mech Dev. 2013;130(11-12):567–76.

    CAS  Article  PubMed  Google Scholar 

  39. 39.

    Fraser LR. The modulation of sperm function by fertilization promoting peptide. Hum Reprod. 1998;13 Suppl 4:1–10.

    CAS  Article  PubMed  Google Scholar 

  40. 40.

    Ma Y, Zhang S, Xia Q, Zhang G, Huang X, Huang M, Xiao C, Pan A, Sun Y, Lebo R, et al. Molecular characterization of the TCP11 gene which is the human homologue of the mouse gene encoding the receptor of fertilization promoting peptide. Mol Hum Reprod. 2002;8(1):24–31.

    CAS  Article  PubMed  Google Scholar 

  41. 41.

    Lapensee L, Paquette Y, Bleau G. Allelic polymorphism and chromosomal localization of the human oviductin gene (MUC9). Fertil Steril. 1997;68(4):702–8.

    CAS  Article  PubMed  Google Scholar 

  42. 42.

    Abe H, Sendai Y, Satoh T, Hoshi H. Bovine oviduct-specific glycoprotein: a potent factor for maintenance of viability and motility of bovine spermatozoa in vitro. Mol Reprod Dev. 1995;42(2):226–32.

    CAS  Article  PubMed  Google Scholar 

  43. 43.

    King RS, Anderson SH, Killian GJ. Effect of bovine oviductal estrus-associated protein on the ability of sperm to capacitate and fertilize oocytes. J Androl. 1994;15(5):468–78.

    CAS  PubMed  Google Scholar 

  44. 44.

    Mondejar I, Martinez-Martinez I, Aviles M, Coy P. Identification of potential oviductal factors responsible for zona pellucida hardening and monospermy during fertilization in mammals. Biol Reprod. 2013;89(3):67.

    Article  PubMed  Google Scholar 

  45. 45.

    Takayama K, Horie-Inoue K, Ikeda K, Urano T, Murakami K, Hayashizaki Y, Ouchi Y, Inoue S. FOXP1 is an androgen-responsive transcription factor that negatively regulates androgen receptor signaling in prostate cancer cells. Biochem Biophys Res Commun. 2008;374(2):388–93.

    CAS  Article  PubMed  Google Scholar 

  46. 46.

    Shoji S, Yoshida N, Amanai M, Ohgishi M, Fukui T, Fujimoto S, Nakano Y, Kajikawa E, Perry AC. Mammalian Emi2 mediates cytostatic arrest and transduces the signal for meiotic exit via Cdc20. EMBO J. 2006;25(4):834–45.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  47. 47.

    Perrard MH, Chassaing E, Montillet G, Sabido O, Durand P. Cytostatic factor proteins are present in male meiotic cells and beta-nerve growth factor increases mos levels in rat late spermatocytes. PLoS One. 2009;4(10):e7237.

    Article  PubMed  PubMed Central  Google Scholar 

  48. 48.

    Triplet T, Butler G. A review of genomic data warehousing systems. Brief Bioinform. 2014;15(4):471-83.

  49. 49.

    Kasprzyk A. BioMart: driving a paradigm change in biological data management. Database. 2011;2011:bar049.

    Article  PubMed  PubMed Central  Google Scholar 

  50. 50.

    Warde-Farley D, Donaldson SL, Comes O, Zuberi K, Badrawi R, Chao P, Franz M, Grouios C, Kazi F, Lopes CT, et al. The GeneMANIA prediction server: biological network integration for gene prioritization and predicting gene function. Nucleic Acids Res. 2010;38(Web Server issue):W214–220.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  51. 51.

    Orlando L, Ginolhac A, Zhang G, Froese D, Albrechtsen A, Stiller M, Schubert M, Cappellini E, Petersen B, Moltke I, et al. Recalibrating Equus evolution using the genome sequence of an early Middle Pleistocene horse. Nature. 2013;499(7456):74–8.

    CAS  Article  PubMed  Google Scholar 

  52. 52.

    Metzger J, Tonda R, Beltran S, Agueda L, Gut M, Distl O. Next generation sequencing gives an insight into the characteristics of highly selected breeds versus non-breed horses in the course of domestication. BMC Genomics. 2014;15(1):562.

    Article  PubMed  PubMed Central  Google Scholar 

  53. 53.

    Cingolani P, Platts A, le Wang L, Coon M, Nguyen T, Wang L, Land SJ, Lu X, Ruden DM. A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly (Austin). 2012;6(2):80–92.

    CAS  Article  Google Scholar 

  54. 54.

    MacArthur DG, Tyler-Smith C. Loss-of-function variants in the genomes of healthy humans. Hum Mol Genet. 2010;19(R2):R125–130.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  55. 55.

    He C, Holme J, Anthony J. SNP genotyping: the KASP assay. Methods Mol Biol. 2014;1145:75–86.

    CAS  Article  PubMed  Google Scholar 

  56. 56.

    Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC, Remm M, Rozen SG. Primer3--new capabilities and interfaces. Nucleic Acids Res. 2012;40(15):e115.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  57. 57.

    Koressaar T, Remm M. Enhancements and modifications of primer design program Primer3. Bioinformatics. 2007;23(10):1289–91.

    CAS  Article  PubMed  Google Scholar 

  58. 58.

    Hamann H, Sieme H, Distl O. Genetic analysis of the fertility in Hanoverian Warmblood horses. Anim Reprod Sci. 2005;89(1-4):201–3.

    CAS  PubMed  Google Scholar 

  59. 59.

    Van Tassel CP, Van Vleck LD. A Manual for Use of MTGSAM: A Set of Fortran Programs to Apply Gibbs Sampling to Animal Models for Variance Component Estimation. U.S. Department of Agriculture, Agricultural Research Service, Clay Center, NE, 1995.

  60. 60.

    Van Tassel CP, Van Vleck LD, Gregory KE. Bayesian analysis of twinning and ovulation rates using a multiple-trait threshold model and Gibbs sampling. J Anim Sci. 1998;76:2048–61.

    Google Scholar 

  61. 61.

    Bradbury PJ, Zhang Z, Kroon DE, Casstevens TM, Ramdoss Y, Buckler ES. TASSEL: software for association mapping of complex traits in diverse samples. Bioinformatics. 2007;23:2633–5.

    CAS  Article  PubMed  Google Scholar 

Download references


This study was supported by the Mehl-Mülhens Stiftung, Köln (DI-MM/2-1). We acknowledge the support of the National State Stud Celle of Lower Saxony, Celle, the Hanoverian breeding association for providing data and samples of Hanoverian warmblood stallions. We thank Heike Klippert Hasberg, Mogens Kilian Drabert, Stefan Neander and Jörn Wrede for expert technical support.

Author information



Corresponding author

Correspondence to Ottmar Distl.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

RS, JM, HS and OD conceived and designed the study. RS, MG, JM and OD performed the experiments. OD, RS and MG analysed the data. GM, HS, JM and OD contributed samples, insemination data and analysis tools. RS and OD drafted and finalized the manuscript. All authors read and approved the final manuscript.

Additional files

Additional file 1:

Gene list of human (gene ID human) and corresponding equine gene identities (mapped gene ID horse) involved in male reproduction processes. Start and end of equine genes in base pairs (bp), chromosomal position of the equine gene (ECA), gene symbol, gene name, gene ontology (GO) term, GO molecular function, GO biological process, GO cellular component and GO pathway are given. (XLSX 5632 kb)

Additional file 2:

Whole genome sequences of horses filtered for high-impact variants in 11 horses. Type of mutation, single nucleotide polymorphism (SNPs) and Insertions/Deletions (Indels) and type of high-impact effect are given. (XLSX 20 kb)

Additional file 3:

Equine gene models for 14 genes related with stallion fertility and high-impact variants. Gene models were built based on the Ensembl annotation. Translated exons are shown as solid black boxes, untranslated exons are shown as open boxes. Numbers above the boxes indicate the exon number. Continuous lines represent introns, numbers below the boxes and lines indicate the respective sizes of exons and introns in base pairs. The predicted translation start and stop codon is indicated. The variant with high effect on protein structure, identified by screening whole genome sequences of horses, is given above the exon. (DOCX 734 kb)

Additional file 4:

Network diagram showing protein and genetic interactions, pathways, co-expression, co-localization and protein domain similarity of 15 genes screened for fertility-related high-impact variants. The genes related with stallion fertility are given in black. (DOCX 503 kb)

Additional file 5:

Techniques and primer sequences used for genotyping high-impact variants. Competitive allele specific PCR (KASP) genotyping technique was used for genotyping 16/17 variants. Specific FAM and HEX dye labeled primer (Primer_Allele FAM, Primer_Allele HEX) with common primers and annealing temperature (AT) are given. Polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP) was used for 1/17 variants. Primer sequences and position, product size, annealing temperature (AT) and the restriction enzyme used are shown. (DOCX 16 kb)

Additional file 6:

Distribution of high-impact variants contributing to stallion fertility in discovery breeds compared to validation breeds. The discovery breeds include horses of the breeds Arabian (AR), Hanoverian (HA), Standardbred (SB), Icelandic (IS), Norwegian fjord (NO), and the non-breed horses Dülmen Horse (DU) and Sorraia (SO), and the wild horse Przewalski (PR). The validation breeds include stallions of the pure breeds Arabian (AR), and Thoroughbred (TH), German sport horses including Hanoverian (HA), Holstein (HO), Oldenburg (OL), Westphalian (WE) Rhinelander (RH) and German Riding Pony (GE). Next, cold blood breeds, represented by Black Forest Horses (BF), Mecklenburger Cold Blood (MC), Rhenish German Cold Blood (RG), Saxon Thuringia Cold Blood (ST), Schleswig Cold Blood (SC) and Southern German Cold Blood (SG), further Dülmen Horse (DU) and Sorraia (SO). Further, robust horses and non-breed horses were represented by Norwegian fjord (NO), Konik (KO) and Tarpan (TA). (DOCX 30 kb)

Additional file 7:

Number of stallions genotyped (n) and genotypic means with their standard errors (SE) and P-values (P) of the de-regressed estimated relative breeding values of the paternal component for the pregnancy rate per estrus (EBV-PAT) in 216 Hanoverian stallions for high-impact variants, their nominal (P) and for multiple tests Bonferroni-corrected P-values (Pmultiple) for differences among the genotypes. The de-regressed EBV-PAT values were standardized onto a scale of 100 ± 20. (DOCX 20 kb)

Additional file 8:

Distribution of fertile stallions genotyped for high-impact variants. Fertile stallions have a minimum number of 10 progeny. (DOCX 14 kb)

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Schrimpf, R., Gottschalk, M., Metzger, J. et al. Screening of whole genome sequences identified high-impact variants for stallion fertility. BMC Genomics 17, 288 (2016).

Download citation


  • Whole genome screening
  • High-impact variants
  • Stallion fertility
  • NOTCH1