Department of Animal Science, Cornell University, Ithaca, NY 14853.
Department of Animal Science, University of Minnesota, St. Paul 55108.
J Dairy Sci. 2020 Oct;103(10):9177-9194. doi: 10.3168/jds.2020-18290. Epub 2020 Jul 23.
The objectives of our study were to expand phenotypic characterization of digital cushion thickness (DCT) to the Jersey breed and include mature bulls and to identify breed-specific quantitative trait loci (QTL) for DCT within the Jersey or Holstein breeds and common QTL across breeds and sexes to better understand biological regulation and guide future use of marker-assisted selection. In a cohort of 698 cows and 85 bulls (Holstein and Jersey) from 8 farms in New York State, data were collected on DCT, body condition score (BCS), withers height, and sacral height. All animals underwent sonographic examination of the digital cushion evaluated at the sole ulcer site for the right front and hind feet. Linear mixed models were conducted on DCT separately for cows and bulls with fixed effects of time point, breed, age, digit, and BCS group. The models included random effects to control for the random subset of animals per farm, repeated measures, and multiple measurements from each animal. The phenotypic results indicated that DCT varied by time point, breed, age, digit, and BCS group for cows and by breed, age, digit, and BCS group for bulls. For the genotypic study, 616 cow DNA samples were genotyped on the Illumina BovineHD 777K BeadChip (Illumina Inc., San Diego, CA), whereas 76 bull DNA samples were genotyped on different platforms ranging from 5K to 150K. Multiple genome-wide association studies were conducted to highlight pertinent phenotyping parameters and genetic markers for genomic selection. Data were separated into 8 data sets based on different combinations of breed and sex. Each data set was assessed for quality of markers and samples before conducting genome-wide association studies for DCT, testing the inheritance models and genetic variation of digit, foot, and average thickness. Ten markers passed the Bonferroni correction threshold and 9 passed false discovery rate from 10 genome-wide association studies using a combination of the covariates breed, sex, genotyping batch plate, age, BCS, withers height, and sacral height. Of the 43 candidate genes, 8 novel biologically plausible genes were identified on Bos taurus autosomes 3, 4, 7, and 9: SFRS18 and LRRFIP1 function in fat deposition, whereas AHR, BZW2, EFNA5, USP45, and VAV3 effect bone growth, and SOSTDC1 is related to epidermal keratinocyte function. The genetic markers associated with DCT in this study were explored for variation between cows and bulls within and across breeds for their potential use in marker-assisted selection.
我们的研究目的是扩展数字垫厚度（DCT）的表型特征描述，包括成熟公牛，并确定泽西品种或荷斯坦品种特有的 DCT 数量性状基因座（QTL），以及跨品种和性别共有的 QTL，以更好地了解生物学调节，并指导未来使用标记辅助选择。在纽约州 8 个农场的 698 头奶牛和 85 头公牛（荷斯坦和泽西）中，我们收集了 DCT、体况评分（BCS）、肩高和荐骨高的数据。所有动物都接受了右前肢和后肢足底溃疡部位的数字垫超声检查。我们分别对奶牛和公牛的 DCT 进行线性混合模型分析，固定效应包括时间点、品种、年龄、数字和 BCS 组。模型还包括随机效应，以控制每个农场的动物随机子集、重复测量和每个动物的多次测量。表型结果表明，DCT 随时间点、品种、年龄、数字和 BCS 组而变化，奶牛的结果如此，公牛的结果则随品种、年龄、数字和 BCS 组而变化。在基因研究中，616 头奶牛的 DNA 样本在 Illumina BovineHD 777K BeadChip（Illumina Inc.，圣地亚哥，CA）上进行了基因分型，而 76 头公牛的 DNA 样本则在不同的平台上进行了基因分型，范围从 5K 到 150K。我们进行了多次全基因组关联研究，以突出相关表型参数和基因组选择的遗传标记。根据品种和性别不同的组合，数据分为 8 个数据集。在对 DCT 进行全基因组关联研究之前，每个数据集都要对标记和样本的质量进行评估，测试数字、脚和平均厚度的遗传模型和遗传变异。在 10 个全基因组关联研究中，有 10 个标记通过了 Bonferroni 校正阈值，9 个标记通过了假发现率，这些标记使用了协变量品种、性别、基因分型板、年龄、BCS、肩高和荐骨高的组合。在 43 个候选基因中，有 8 个新的具有生物学意义的基因被确定位于牛染色体 3、4、7 和 9 上：SFRS18 和 LRRFIP1 参与脂肪沉积，而 AHR、BZW2、EFNA5、USP45 和 VAV3 影响骨骼生长，SOSTDC1 与表皮角质形成细胞功能有关。本研究中与 DCT 相关的遗传标记在牛和公牛之间进行了品种内和品种间的变异探索，以评估其在标记辅助选择中的潜在用途。
