ГЕНЕТИКА, 2015, том 51, № 10, с. 1163-1170
SUSCEPTIBILITY LOCI FOR UMBILICAL HERNIA IN SWINE DETECTED BY GENOME-WIDE ASSOCIATION
© 2015 X. J. Liaoa, b, *, L. Lia, *, Z. Y. Zhanga, Y. Longa, B. Yanga, G. R. Ruana, c, Y. Sua, H. S. Aia, W. C. Zhanga, W. Y. Denga, S. J. Xiaoa, J. Rena, N. S. Dinga, and L. S. Huanga
aState Key Laboratory for Pig Genetic Improvement and Production Technology, Jiangxi Agricultural University, NanChang, 330045 China e-mail: firstname.lastname@example.org bCollege of Life Science of Jinggangshan University, JiAn, 343009 China cFuJian Vocational College of Agriculture, FuZhou, 360119 China Received December 20, 2014
Umbilical hernia (UH) is a complex disorder caused by both genetic and environmental factors. UH brings animal welfare problems and severe economic loss to the pig industry. Until now, the genetic basis of UH is poorly understood. The high-density 60K porcine SNP array enables the rapid application of genome-wide association study (GWAS) to identify genetic loci for phenotypic traits at genome wide scale in pigs. The objective of this research was to identify susceptibility loci for swine umbilical hernia using the GWAS approach. We genotyped 478 piglets from 142 families representing three Western commercial breeds with the Illumina PorcineSNP60 BeadChip. Then significant SNPs were detected by GWAS using ROADTRIPS (Robust Association-Detection Test for Related Individuals with Population Substructure) software base on a Bonferro-ni corrected threshold (P = 1.67E-06) or suggestive threshold (P = 3.34E-05) and false discovery rate (FDR = 0.05). After quality control, 29,924 qualified SNPs and 472 piglets were used for GWAS. Two suggestive loci predisposing to pig UH were identified at 44.25MB on SSC2 (rs81358018, P = 3.34E-06, FDR= 0.049933) and at 45.90MB on SSC17 (rs81479278, P = 3.30E-06, FDR = 0.049933) in Duroc population, respectively. And no SNP was detected to be associated with pig UH at significant level in neither Landrace nor Large White population. Furthermore, we carried out a meta-analysis in the combined pure-breed population containing all the 472 piglets. rs81479278 (P = 1.16E-06, FDR = 0.022475) was identified to associate with pig UH at genome-wide significant level. SRC was characterized as plausible candidate gene for susceptibility to pig UH according to its genomic position and biological functions. To our knowledge, this study gives the first description of GWAS identifying susceptibility loci for umbilical hernia in pigs. Our findings provide deeper insights to the genetic architecture of umbilical hernia in pigs.
Keywords: association analysis, candidate gene, genotype, piglets, susceptibility loci, umbilical hernia. DOI: 10.7868/S0016675815100100
Umbilical hernia (UH) is one of the most common duce the prcdurtion efficiency and benefit T^refc^
congenital and developmental undesirable defects in researches for the pg UH haw important welfare and
many mammalian species, such as human, dog, horse, economic ^lKatMnss.
swine and cattle. This birth defect is usually detectable The aetiology of UH are likely to be multifactorial
with visible veterinary clinics, and is well respected by and include genetic and non-genetic components.
protrusion of abdominal contents into the overlying The latter mainly consist of physical injury, obesity, ex-
subcuits, resulting from failure of the normal closure cessive pressure °f the abdommal ^ty, mu^uki-
of the umbilical ring. The prevalence of pig UH ranges weakness around the umMkal stump ot navd area,
from 0.13 to 2.25% in different breeds and farms [1- mappropmrte remold of the umM^al c°rcl, _over-
3]. UH brings a lot ofadverse effects, for example, dis- crowding, and ^fj mfecti°ns [3, 4] In a^iti^ to
the environmental factors, clusters of UH in a family
torted physical appearance, impaired growth, reduced
. . . ^ , , , . ,. ' n • . and the different prevalence of this defect in different
feed conversi°nrateand reducedp°rk quahty. Special breeds suggest that genetic susceptibility plays an im-treatments to affected ammak, such as herniorrhaphy, portant role in the occurrence of the disease [2, 5, 6]. have been commonly used in pig production. Additional expenditure of labors, medicine and barn re- Until now the genetic basis of UH is p°°rly understood. Identification of the causal mutation(s) for UH
' Both authors contribute equally to this study and should be con- susceptibility would benefit for the pig industry. Sever-sidered co-first authors. al research groups have devoted to deciphering the ge-
Tablel. Distribution of umbilical hernia phenotypes in 3 pig breeds
Breed N Case Control
Duroc 175 63 112
Landrace 67 19 48
Large White 236 78 158
Total 478 160 318
netic architecture of UH. Ron et al.  identified a locus for UH at the centromeric end of chromosome 8 in cattle. Ding et al.  reported several susceptibility loci for pig UH on 12 chromosomes. In humans, interstitial deletions on chromosomal 6q are reported to cause congenital UH [9, 10]. Recently, Radhakrishna et al.  found a 7l0-kb duplication on chromosome 1p31.3 that is responsible for autosomal dominant omphalocele. In human, epimutations of imprinted genes in the human genome may play a role in the occurrence of UH [12, 13]. Further endeavors are required to identify additional susceptibility loci and causative genes for pig UH.
Genome-wide association study (GWAS) has emerged as a powerful tool for genetic analyses in both human and other species . Since 2009, the high-density 60K porcine SNP array  enables the rapid application of the GWAS to identify genetic loci for phenotypic traits at genome wide scale in pigs [16— 18]. To further clarify the genetic basis of swine UH susceptibility, we herein performed GWAS to identify candidate susceptibility loci for pig UH in 3 international commercial breeds.
MATERIALS AND METHODS
Animals and phenotypes. A total of 478 purebred pigs including 175 Duroc (DU), 67 Landrace (LR) and 236 Large White (LW) from 142 families were collected from 15 farms in 8 provinces of China during the period of September 2006 to October 2011 (Table 1). These animals consist of 160 affected individuals (cases) and 318 unaffected sib counterparts (controls). In each family, one affected individual, one unaffected sib and their parents were sampled. In 68 families, only one affected animal and one sib and/or one parent were collected. As umbilical hernia and umbilical ab-
scess are prone to be promiscuous, especially in cattle and swine, the affected pigs were diagnosed carefully by veterinarians and collected for real genetic UH at the age of 21 to 72 days. For each pig, we recorded breed, sex, name of farm and pedigree, and took picture for affected pigs.
Genomic DNA was extracted from pig ear tissue with a standard protocol by primary digestion with proteinase K, followed by phenol/chloroform extraction and ethanol precipitation. All DNA samples were quantified and diluted to a final concentration of 50 ng/^L in 96-well plates.
Genotyping and quality control. Genotyping was performed with the Illumina PorcineSNP60 Bead-Chip (San Diego, CA, USA) containing 62,163 SNPs. The genotypes were recorded by BeadStudio (Version 3.2.2, Illumina) and a custom cluster file developed from the 478 samples. Quality control was carried out for the genotype date with Plink Version 1.07 (http:// pngu.mgh.harvard.edu/purcell/plink/). First, all 478 samples with >90% call rate were included for the subsequent analysis. Second, 19,128, 3,531, and 8,745 SNPs were discarded according to their Hardy— Weinberg equilibrium test (p < 10E-5), SNP call rates (<90%) and minor allele frequencies (MAF < 0.05), respectively. One Landrace family including 1 case, 2 sibs and their parent were removed for Mendelian errors. In addition, one Landrace sample was excluded for sex error. A final set of 29,924 SNPs (Table 2) and 472 pigs were retained for the subsequent analysis.
Genome-wide association study. GWAS was conducted by ROADTRIPS software (Version 1.2) . The population structure and kinship stratification were corrected by an individual-paired genome-wide similarity covariance matrix under the linear framework. Three statistics were output in ROADTRIPS named as RM, RW, and Rx, respectively. The RM test was adopted to detect loci associated with UH in this paper, because the RM test is powerful for two-allele disease model under the linear framework in natural populations, and it has higher power than the other two test when sampled individuals were related and partial or complete pedigree information was available. The P-values for the RM statistic were used to indicate the association between the disease and the marker.
Table 2. Distribution of qualified SNPs on chromosomes and the average distances between adjacent SNPs
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 X 0 b Total
No. SNPs 3099 2068 1725 1943 1390 1744 1794 1690 2042 1057 1052 933 2314 2187 1467 946 974 778 721 4232 34156
Distance (kb)a 101.6 78.5 83.7 73.8 80.1 90.2 75.0 87.3 75.2 73.6 83.4 68.1 94.3 70.3 107.3 91.9 71.2 78.2 199.6 NA
a The average distances between adjacent SNPs were defined according to the porcine genome sequence assembly (Sscrofa10.2). b These SNPs are not assigned to any chromosomes; NA: not available.
1 2 3 4 5 6 7 8 9 10 12 13 14 15 17 X
0 12 3 4
Fig. 1. Manhattan plots of genome-wide association for susceptibility to UH in Duroc. a — Negative lg10-transformed P values of all qualified SNPs are plotted against their genomic position. Different chromosomes are represented by different color. The X chromosome is indicated by chromosome 19 (as for Figs. 2—4). The dotted line represents suggestive significant threshold (4.48). b - QQ-plots (as for Figs. 2, 3).
In GWA meta-analysis, data from the three br
Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.