Bovine Exome Sequence Analysis and Targeted SNP Genotyping of Recessive Fertility Defects BH1, HH2, and HH3 Reveal a Putative Causative Mutation in SMC2 for HH3

C Bovine Exome Sequence Analysis and Targeted SNP Genotyping of Recessive Fertility Defects BH1, HH2, and HH3 Reveal a Putative Causative Mutation in SMC2 for HH3 Matthew C. McClure 0 1 Derek Bickhart 0 1 Dan Null 0 1 Paul VanRaden 0 1 Lingyang Xu 0 1 George Wiggans 0 1 George Liu 0 1 Steve Schroeder 0 1 Jarret Glasscock 0 1 Jon Armstrong 0 1 John B. Cole 0 1 Curtis P. Van Tassell 0 1 Tad S. Sonstegard 0 1 Stephen Moore, University of Queensland, Australia 0 1 United States Department of Agriculture, Agriculture Research Service, Bovine Functional Genomics Laboratory, Beltsville, Maryland, United States of America, 2 United States Department of Agriculture, Agriculture Research Service, Animal Improvement Programs Laboratory, Beltsville, Maryland, United States of America , 3 Cofactor Genomics, St. Louis, Missouri , United States of America 1 Mutation in SMC2 Affecting Fertility in Holsteins The recent discovery of bovine haplotypes with negative effects on fertility in the Brown Swiss, Holstein, and Jersey breeds has allowed producers to identify carrier animals using commercial single nucleotide polymorphism (SNP) genotyping assays. This study was devised to identify the causative mutations underlying defective bovine embryo development contained within three of these haplotypes (Brown Swiss haplotype 1 and Holstein haplotypes 2 and 3) by combining exome capture with next generation sequencing. Of the 68,476,640 sequence variations (SV) identified, only 1,311 genomewide SNP were concordant with the haplotype status of 21 sequenced carriers. Validation genotyping of 36 candidate SNP identified only 1 variant that was concordant to Holstein haplotype 3 (HH3), while no variants located within the refined intervals for HH2 or BH1 were concordant. The variant strictly associated with HH3 is a non-synonymous SNP (T/C) within exon 24 of the Structural Maintenance of Chromosomes 2 (SMC2) on Chromosome 8 at position 95,410,507 (UMD3.1). This polymorphism changes amino acid 1135 from phenylalanine to serine and causes a non-neutral, non-tolerated, and evolutionarily unlikely substitution within the NTPase domain of the encoded protein. Because only exome capture sequencing was used, we could not rule out the possibility that the true causative mutation for HH3 might lie in a nonexonic genomic location. Given the essential role of SMC2 in DNA repair, chromosome condensation and segregation during cell division, our findings strongly support the non-synonymous SNP (T/C) in SMC2 as the likely causative mutation. The absence of concordant variations for HH2 or BH1 suggests either the underlying causative mutations lie within a nonexomic region or in exome regions not covered by the capture array. - Funding: The contribution by scientists in the Animal Improvement Programs and Bovine Functional Genomics Laboratories are supported by appropriated projects 1245-31000-101-00 and 1265-31000-104-00, respectively. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. While a number of recessive disease loci and a few causative disease mutations have been discovered in cattle [1,2], few individual loci that reduce embryo viability or early fetus development in cattle were known until recently. The genotyping of hundreds of thousands of animals within popular U.S. dairy breeds using commercial, genome-wide single nucleotide polymorphism (SNP) panels has provided the opportunity to find regions of the genome with unbalanced Mendelian inheritance. When first investigated, the absence of expected homozygous haplotypes for five genomic regions (5 Mbp haplotypes) could be associated with effects on fertility in Brown Swiss, Holstein, and Jersey cattle [1]. Breed associations and artificial insemination companies have officially reported the inheritance of these haplotypes for genotyped animals since August 2011. While disease diagnosis by SNP haplotype analysis has been very successful, it is not 100% accurate. Intuitively, the next step in creating an effective diagnostic marker is to leverage next generation sequencing approaches to rapidly identify putative causative polymorphisms, and then test those mutations contained within the haplotype that might disrupt normal gene function for cellular processes important for embryo development. Causative mutations for 2 of the initial 5 bovine haplotypes, Holstein haplotype 1 (HH1) and Jersey haplotype 1 (JH1), were recently discovered by re-sequencing only non-carrier and carrier animals, because homozygous HH1 or JH1 animals do not exist. Both mutations were single nucleotide substitutions creating pre-mature stop codons in the coding regions of the APAF1 gene for HH1 [3] and the CWC15 gene for JH1 [4]. Similarly, a fertility haplotype associated with Brachyspina, a defect causing early abortion or births of calves with congenital defects, was found to be a 3,329 bp deletion in the FANCI gene [5]. Fertility effects for the remaining three haplotypes, known as Brown Swiss haplotype 1 (BH1), Holstein haplotype 2 (HH2), and HH3; were confirmed significant by comparing normal conception rates (27% for Brown Swiss and 31% for Holsteins) to rates when mating heterozygous males to daughters of heterozygous males [1]. Conception rates were lower by 3.461.5 for BH1, 3.060.8 for HH2, and 3.260.4 for HH3 compared to the average. Numbers of normal matings were 68,000 for Brown Swiss and 14 million for Holsteins, compared to 936 heterozygote matings for BH1, 3,252 for HH2, and 14,114 for HH3. Most conception losses occurred before 100 days of gestation for HH2 and BH1and before 60 days for HH3 [6]. Stillbirth rates for HH2 and HH3 were slightly, but not significantly, higher than normal. The founding alleles for these haplotypes were traced by pedigree to the sires: 163153 West Lawn Stretch Improver for BH1, 334489 Willowholme Mark Anthony for HH2, and 1556373 Glendell Arlinda Chief and 1244845 Gray View Skyliner for HH3. Subsequently, Schwarzenbacher and colleagues [7] used an independent data set to identify harmful recessive haplotypes and confirmed a lack of homozygotes within the BH1 haplotype region for Brown Swiss. This effect was detected for heifer fertility, but not for cow fertility. Hayes and colleagues announced a probable causative mutation for HH3 by analyzing SNP data derived from the 1,000 Bull Genomes Project (http://www. 1000bullgenomes.com/), but this finding was not supported by SNP testing to validate co-segregation in designated HH3 carriers [8]. More recently, Fritz and colleagues confirmed the lethal fertility effect of HH3 in French Holsteins [9]. Based on these previous findings, the main goal of this present study was to find and validate causative mutations for BH1, HH2, and HH3; and thereby, allow development of single marker diagnostic SNP tests that could be added to commercial SNP chips already in use by the dairy industry for generating genomic (...truncated)