Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-5504(Print)
ISSN : 2383-8272(Online)
Journal of Agriculture & Life Science Vol.54 No.2 pp.1-8
DOI : https://doi.org/10.14397/jals.2020.54.2.1

Genome-wide Association Study for Ethological Traits of Purebred Landrace and Yorkshire Populations

Tae-Jeong Choi1†, Ho-Chan Kang2†, Jae-Bong Lee3, Chae-Kyoung Yoo4, Eun-Ho Kim5, Shin-Jae Rhim6, Hyun-Tae Lim2,4,5*
1Swine Science Division, National Institute of Animal Science, RDA, Cheonan, 31000, Korea
2Department of Animal Science, Gyeongsang National University, Jinju, 52828, Korea
3Korea Zoonosis Research Institute(KoZRI), Chonbuk National University, Iksan, 54531, Korea
4Institute of Agriculture and Life Science, Gyeongsang National University, Jinju, 52828, Korea
5Division of Applied Life Science, Gyeongsang National University, BK21 Plus, Jinju 52828, Korea
6School of Bioresource and Bioscience, Chung-Ang University, Ansung, 17546, Korea

These authors contributed equally to this work.


*Corresponding author: Hyun-Tae Lim Tel: +82-55-772-1945 Fax: +82-55-772-1949 E-mail: s_htim@gnu.ac.kr
December 30, 2019 April 2, 2020 April 8, 2020

Abstract


Some behaviors of pigs that are not expressed in the wild state or are observed in a small minority of individuals after groups of pigs are mixed have been reported to indicate poor welfare. A GWAS analysis was performed by measuring the frequency and duration of the four ethological traits and using the mlma command provided by the genome-wide complex trait analysis (GCTA). The positional candidate genes on significantly identified single nucleotide polymorphism (SNP) markers were identified by using the dbSNP provided by the National Center for Biotechnology Information (NCBI). When the GWAS analysis was applied the 43,565 (of the purebred Landrace population) and 41,700 (of the purebred Yorkshire population) SNP markers, 1, 2, and 1 significant SNP markers were identified for the traits of feeding frequency (LOC110262254), locomotion time (LOC110260361), and locomotion frequency (LOC110260361) of the purebred Landrace population, respectively. Meanwhile, 1 and 7 significant SNP markers were identified for the traits of drinking time (LOC110260090) and feeding frequency (MAP3K19; LOC110257013; ACMSD; TMEM163; RAB3GAP1) of the purebred Yorkshire population, respectively. The results of this study may suggest that the GWAS analysis of the ethological traits of purebred Landrace and Yorkshire populations could be used to perform a GWAS analysis on non-economic traits, and the results can thus be provided as basic data for GWAS analyses of other non-economic traits in the future.



초록


    Rural Development Administration
    PJ009971032017

    INTRODUCTION

    The modern meaning of animal welfare includes securing the stability of livestock products by preventing risks of various diseases, as well as humanitarian purposes (Ahn et al., 2014). Animal welfare includes such factors as proper laval density in rearing, transportation methods before slaughter, temperature and density of the rearing facility, and behaviors of individual animals. It has been reported that some behaviors that occurs after mixing of groups, although not expressed in the wild or a few of observations, has been shown to indivated poor welfare (Turner et al., 2010). The need for behavior research to resolve poor welfare can be two-fold to understand the physiology of individuals to improve productivity and to collect information and establish customized strategies to improve welfare based on the research results (McGlone, 1994).

    The genetic effects of the Landrace population’s sociality as a non-economic trait were studied recently using genomewide association study (GWAS) (Hong et al., 2018), and the results indicate that an analysis of GWAS can serve as a new method of determining the relation between non-economic traits and genes. An analysis of GWAS can be effective in identifying genetic variation and genes associated with traits because the method can detect variations in many natural allelic traits using a variety of groups to analyze the association between phenotypic variation and nucleotide polymorphisms (Kolbehdari et al., 2009, Yano et al., 2016). An analysis of GWAS, whose statistical power is greater than an association analysis in terms of studies of the complex traits of many genes with low effects (Tabor et al., 2002), has been used to detect variation-related indexes in genome-wide genetic information within a group by using single nucleotide polymorphism (SNP) markers (Fan et al., 2011, Jung et al., 2014, Horodyska et al., 2017).

    This study was conducted to detect for candidate genes related to ethological traits to improve animal welfare through feed intake and stress reduction.

    MATERIALS AND METHODS

    1. Animal data

    This research was approved by animal ethical committee of National Institute of Animal Science, RDA, Cheonan, Republic of Korea. The purebred Landrace and Yorkshire populations used in this study were tested at the National Institute of Animal Science from 2012 to 2017. This population constituted the purebred Landrace population of 115 animals and the purebred Yorkshire population of 118 animals. Their genomic DNA (gDNA) was extracted from their ear skin tissues.

    2. Analysis of genotype and quality control

    For the high-purity of gDNA extracted from their ear skin tissues, the genotypes of the 61,565 SNP markers were collected from a Porcine SNP 60K Ver.2 BeadChip (Illumina, USA). Quality control for a GWAS analysis was performed under the conditions of minor allele frequency (MAF) < 0.05, genotyping error > 0.1, and hardy-weinberg equilibrium (p value ≤0.000001). Based on the quality control results, 42,262 (from the purebred Landrace population) and 41,700 (from the purebred Yorkshire population) SNP markers were used in the GWAS analysis.

    3. Ethology of description

    To determine the phenotypic traits of pigs’ behaviors, images of individual behaviors were filmed with a Sony HDR-AS20 at the National Institute of Animal Science from 2012 to 2017, and images of behaviors filmed with the Sony Vegas 13 program (USA) were observed. The observed traits included drinking, feeding (eating), inactive, and locomotion time and frequency of each traits. For each behavior, duration time and frequency per hour were measured.

    Definitions of each behavior are shown in Table 1 (Statham et al., 2011, Hwang et al., 2016). When a behavior continued for 10 seconds, the behavior was considered present and included in the observation. To calibrate the observed phenotypes, the duration time and frequency of the observed behaviors from individuals were divided by the total frequency of the measurements.

    The observed traits were tested for normality with the Ryan-Joiner method using a MINITAB program (Minitab inc., USA) determine whether they followed a normal distribution (RJ score ≥ 0.95), and outliers of the phenotypes were removed. The phenotypic traits that did not follow a normal distribution were added to a natural logarithm formula to confirm whether they followed a normal distribution (RJ score ≥ 0.95) using the Ryan-Joiner test for normality, and outliers of the phenotypes were removed. The reason for using natural log is that it is used when phenotype cannot be verified for normality, and values such as p value and the phenotype are identical even if natural log values are applied.

    4. Genome-wide association study

    The GWAS analysis was performed using the mlma command provided by the genome-wide complex trait analysis (GCTA) (Yang et al., 2011).

    y = X b + Z 1 a + Z 2 u + e

    Here, y = the phenotypic vector of a ethological traits, b = fixed effect (sex), a = the fixed effect of a SNP marker, u = the random additive effect vector, and the mean and variation of u are u~N(0, G σ a 2 ). G is a genomic relationship matrix, σ a 2 is an additive genetic variance, e is a random residual vector, and its mean and variation are e~N(0, 1 σ e 2 ). I is the identity matrix, and σ e 2 is the residual variance. Z1 is the frequency vector on a, and X, Z2 are frequency matrixes on b and u, respectively.

    The p value of the SNP markers from the results of the GWAS analysis were calculated with –log (p value) using the R program, and the Manhattan plot and the quantile-quantile (QQ) plot were shown.

    5. Significant SNP marker and candidate gene search

    Significant SNP markers were identified by the Bonferroniadjusted genome-wide suggestive level (Bonferroni-adjusted threshold: 1/ the number of autosomal SNP markers). Results of quality control, the SNP markers were marked on the Manhattan plot based on p value = 2.366 × 10-5 for the purebred Landrace population and p value = 2.938 × 10-5 for the purebred Yorkshire population. To search the candidate genes of significant SNP markers based on the GWAS analysis results, using the dbSNP (https://www.ncbi.nlm.nih.gov/snp) map ver. 11.1 provided by the National Center for Biotechnology Information (NCBI). We identified the genes as positional candidate genes that were closest adjacent to the locations of the markers or were included in the exon.

    RESULTS AND DISCUSSION

    The reason for searching candidate genes for behavioral traits is to contribute to improving econmic efficiency through traits such as energy meabolism and movement of pigs as industrial animals. Therefore, the results of this study using behavioral traits from industrial animals, pigs, were conducted in order to explore the possibility of application rather than detecting a significant SNP marker for application in the domestic pig industry. This study was conducted in a small population, it is considered to be meaningful as a preliminary study of non-economic trait analysis of GWAS, where economic trait analysis was dominant.

    1. Normal distribution verification of phenotype

    For the purebred Landrace population, the natural logarithm was substituted for the traits of drinking time, inactive frequency, and locomotion time, For the purebred Yorkshire population, the natural logarithm was substituted for the traits of drinking time, drinking frequency, feeding frequency, inactive frequency, locomotion time, and locomotion frequency. After the traits were confirmed to follow a normal distribution, outliers were removed. The descriptive statistics of the phenotypic traits whose outliers were removed are shown in Table 2.

    2. Genome-wide association study

    The significant SNP markers that satisfy the Bonferroni-adjusted threshold were identified of both population. In the pure-bred Landrace population, significant SNPs were detected in feeding frequency, locomotion time and locomotion frequency traits (genome-wide suggestive level = 1/42,262). In the purebred Yorkshire population, significant SNPs were detected in drinking frequency and feeding frequency (genome-wide suggestive level = 1/41,700). They are shown in the Manhattan plot and QQ plot of the SNP markers of each variety (Fig. 1-5). The information of the significant SNP markers they are shown in Table 3 and Table 4. Different results in the same trait by population may include seasons measured for ethological traits, and it is estimated that this is due to gender (male and female), individual growth differences, and rank within the population.

    1, 2, and 1 significant SNP markers were identified in the traits of feeding frequency, locomotion time and locomotion frequency of the purebred Landrace population, respectively. For the feeding frequency trait, the ALGA0054798 marker was significant on Sus Scrofa Chromosome (SSC) 9 and the LOC 110262254 gene was detected as the closest adjacent positional candidate gene. For the locomotion time trait, 2 significant SNP markers (INRA0016936 and DRGA0005192) were identified on SSC 4, and the LOC110260361 gene was detected as the closest adjacent positional candidate gene for the 2 markers. For the locomotion frequency trait, 1 significant SNP marker was identified on SSC 4, and the LOC110260361 gene, as with the locomotion time trait, was detected as the positional candidate gene. For the locomotion frequency trait, 1 significant SNP marker was identified on SSC 4, and the LOC110260361 gene, as with the locomotion time trait, was detected as the positional candidate gene. The function of the two positional candidate genes for the three traits is unknown.

    Meanwhile, 1 and 7 significant SNP markers were identified for the traits of drinking time and feeding frequency of the purebred Yorkshire population, respectively. For the drinking frequency, the closest adjacent positional candidate gene of the ASGA0092767 marker was the LOC110260090 gene on SSC 3, but its genetic functions are unknown. When the functions of other adjacent genes, as well as those of the closest adjacent one were determined, however, the METTL22 gene on SSC 3 was reported to affect systemic sclerosis (Gorlova et al., 2018). The adjacent RBFOX1 gene has been reported to function in the development and maturity of neurons (Stambolian et al., 2013), indicating that it may affect behaviors related to nerve development. For the feeding frequency of the purebred Yorkshire population, 7 SNP markers were identified on SSC 15 (ALGA 0084119, DRGA0014981, H3GA0043848, H3GA0043849, M1GA 0020193, ASGA0068760 and ASGA0068758). The closest adjacent positional candidate gene of the ALGA0084119 marker was MAP3K19, a gene reported to affect the human lung injury and pulmonary fibrosis (Huelsmann et al., 2019). The closest adjacent positional candidate gene of the DRGA0014981 marker was LOC110257013, whose functions are unknown. The adjacent positional candidate gene of the H3GA0043848 and H3GA 0043849 markers was ACMSD, and the positional candidate gene of the ASGA0068760 and ASGA0068758 markers was RAB3GAP1. The ACMSD and RAB3GAP1 genes have been reportedly associated with the red blood cell percentage in blood metabolism (Shin et al., 2014). Especially, RAB3GAP1 is associated with Warburg micro syndrome 1(Handley et al., 2013), it is known to affect eyes and nervous system. The closest adjacent positional candidate gene of the M1GA0020193 marker was TMEM163, a gene that is reportedly associated with human diseases, such as diabetes (Tabassum et al., 2013), and that may affect traits of feeding and drinking related to metabolic activities. The genes affecting metabolism, RAB3GAP1, ACMSD and TMEM163, are not only disease but also behavioral constraints are linked to the supply of nutruents in the body and growth of individuals, resulting in econmic loss due to productivity decline. Therefore, it is thought that genes adjacent to the positional candidate gene with significant results in this study affect the behavior of the pig.

    This study performed an analysis of genome-wide association using ethological phenotypes for the first time in Korea, and the results may serve as preceding, basic data in studying ethological traits using genome-wide association analysis in the future. For the traits whose significant SNP markers were identified, further studies may be needed to determine the functions of the candidate genes as causative genes. The selection of individuals with the inactive trait may reduce hours of activities and consequentially reduce hours of negative ethology, such as biting tails, contributing to improvements in animal welfare related to individual stress and in productivity. The results of this study’s analysis of animal ethological traits may be helpful in improving animal welfare to enhance productivity and to resolve disbelief by animal welfare among consumers concern-ing product stability. In addition, the results could be used for selecting individuals useful for the livestock industry.

    Acknowledgement

    This research was supported by the BK21Plus project of the Ministry of Education, Science and Technology, Republic of Korea. This work was also supported by the Project No. PJ009971032017 of RDA (Rural Development Administration), Republic of Korea.

    Figure

    JALS-54-2-1_F1.gif

    These are the Manhattan (p value) and Quantile- quantile (QQ) plots based on the GWAS results for the feeding frequency of the purebred Landrace. (A) The Manhattan plot shows the association between 42,262 SNP markers in 18 pig autosomes and the feeding frequency. The X-axis is the position of the SNP marker for the Sus Scrofa Chromosome (SSC), and the Y-axis it the –log (p value) for each SNP marker. (B) QQ plot about the p value of SNP markers.

    JALS-54-2-1_F2.gif

    These are the Manhattan (p value) and Quantile- quantile (QQ) plots based on the GWAS results for the locomotion of the purebred Landrace. (A) The Manhattan plot shows the association between 42,262 SNP markers in 18 pig autosomes and the locomotion. The X-axis is the position of the SNP marker for the Sus Scrofa Chromosome (SSC), and the Y-axis it the –log (p value) for each SNP marker. (B) QQ plot about the p value of SNP markers.

    JALS-54-2-1_F3.gif

    These are the Manhattan (p value) and Quantile- quantile (QQ) plots based on the GWAS results for the locomotion frequency of the purebred Landrace. (A) The Manhattan plot shows the association between 42,262 SNP markers in 18 pig autosomes and the locomotion frequency. The X-axis is the position of the SNP marker for the Sus Scrofa Chromosome (SSC), and the Y-axis it the –log (p value) for each SNP marker. (B) QQ plot about the p value of SNP markers.

    JALS-54-2-1_F4.gif

    These are the Manhattan (p value) and Quantile- quantile (QQ) plots based on the GWAS results for the drinking frequency of the purebred Yorkshire. (A) The Manhattan plot shows the association between 41,700 SNP markers in 18 pig autosomes and the drinking frequency. The X-axis is the position of the SNP marker for the Sus Scrofa Chromosome (SSC), and the Y-axis it the –log (p value) for each SNP marker. (B) QQ plot about the p value of SNP markers.

    JALS-54-2-1_F5.gif

    These are the Manhattan (p value) and Qiantile-quantile (QQ) plots based on the GWAS results for the feeding frequency of the purebred Yorkshire. (A) The Manhattan plot shows the association between 41,700 SNP markers in 18 pig autosomes and the feeding frequency. The X-axis is the position of the SNP marker for the Sus Scrofa Chromosome (SSC), and the Y-axis it the –log (p value) for each SNP marker. (B) QQ plot about the p value of SNP markers.

    Table

    Definition and meaning of ethological traits

    Descriptive statistics of ethological traits in purebred Landrace and Yorkshire populations

    Information of SNP markers significant as a result of GWAS analysis of purebred Landrace population

    Information of SNP markers significant as a result of GWAS analysis of purebred Yorkshire population

    Reference

    1. Ahn GC , Song YH and Park KK. 2014. Global trends and settlement of certification of animal welfare for livestocks in South Korea(overview). Ann. Anim. Resour. Sci. 25: 157-171.
    2. Fan B , Onteru SK , Du ZQ , Garrick DJ , Stalder KJ and Rothschild MF. 2011. Genome-wide association study identifies loci for body composition and structural soundness traits in pigs. PLoS ONE 6: e14726.
    3. Gorlova OY , Li Y , Gorlov I , Ying J , Chen WV , Assassi S , Reveille JD , Arnett FC , Zhou X , Bossini-Castillo L , Lopez-Isac E , Acosta-Herrera M , Gregersen PH , Lee AT , Steen VD , Fessler BJ , Khanna D , Schiopu E , Silver RM , Molitor JA , Furst DE , Kafaja S , Simms RW , Lafyatis RA , Carreira P , Simeon CP , Castellvi I , Beltran E , Ortego N , Amos CI , Martin J and Mayes MD. 2018. Gene-level association analysis of systemic sclerosis: A comparison of African-Americans and White populations. PLoS One 13: e0189498.
    4. Handley MT , Morris-Rosendahl DJ , Brown S , Macdonald F , Hardy C , Bem D , Carpanini SM , Borck G , Martorell L , Izzi C , Faravelli F , Accorsi P and 23 others.2013. Mutation spectrum in RAB3GAP1, RAB3GAP2, and RAB18 and genotype-phenotype correlations in Warburg Micro syndrome and Martsolf syndrome. Hum. Mutat. 34: 686-696.
    5. Hong JK , Jeong YD , Cho ES , Choi TJ , Kim YM , Cho KH , Lee JB , Lim HT and Lee DH. 2018. A genome-wide association study of social genetic effects in Landrace pigs. Asian-Australas J. Anim. Sci. 31: 784-790.
    6. Horodyska J , Hamill RM , Varley PF , Reyer H and Wimmers K. 2017. Genome-wide association analysis and functional annotation of positional candidate genes for feed conversion efficiency and growth rate in pigs. PloS One 12: e0173482.
    7. Huelsmann M , Hecker N , Springer MS , Gatesy H , Sharma V and Hiller M. 2019. Genes lost during the transition from land to water in cetaceans highlight genomic changes associated with aquatic adaptations. Sci. Adv. 5: eaaw6671.
    8. Hwang HS , Lee JK , Eom TK , Son SH , Hong JK , Kim KH and Rhim SJ. 2016. Behavioral characteristics of weaned piglets mixed in different groups. Asian-Australas J. Anim. Sci. 29: 1060-1064.
    9. Jung EJ , Park HB , Lee JB , Yoo CK , Kim BM , Kim HI , Kim BW and Lim HT. 2014. Genome-wide association analysis identifies quantitative trait loci for growth in a Landrace purebred population. Anim. Genet. 45: 442-444.
    10. Kolbehdari D , Wang Z , Grant JR , Murdoch B , Prasad A , Xiu Z , Marques E , Stothard P and Moore SS. 2009. A whole genome scan to map QTL for milk production traits and somatic cell score in Canadian Holstein bulls. J. Anim. Breed. Genet. 126: 216-227.
    11. McGlone JJ. 1994. Pig behaviour research in the United States. Pig New Info. 15: 71.
    12. Shin SY , Fauman EB , Petersen AK , Krumsiek J , Santos R , Huang J , Arnold M , Erte I , Forgetta V , Yang TP , Walter K , Menni C , Chen L , Vasquez L , Valdes AM , Hyde CL , Wang V , Ziemek D , Roberts P , Xi L and Grundberg E , Multiple Tissue Human Expression Resource (MuTHER) Consortium. Waldenberger M , Richards JB , Mohney RP , Milburn MV , John SL , Trimmer J , Theis FJ , Overington JP , Suhre K , Brosnan MJ , Gieger C , Kastenmüller G , Spector TD and Soranzo N. 2014. An atlas of genetic influences on human blood metabolites. Nat. Genet. 46: 543-550.
    13. Stambolian D , Wojciechowski R , Oexle K , Pirastu M , Li X , Raffel LJ , Cotch MF , Chew EY , Klein B , Klein R , Wong TY , Simpson CL , Klaver CC , Van Duijn CM , Verhoeven VJ , Baird PN , Vitart V , Paterson AD , Mitchell P , Saw SM , Fossarello M , Kazmierkiewicz K , Murgia F , Portas L , Schache M , Richardson A , Xie J , Wang JJ , Rochtchina E , DCCT/EDIC Research Group, Viswanathan AC , Hayward C , Wright AF , Polasek O , Campbell H , Rudan I , Oostra BA , Uitterlinden AG , Hofman A , Rivadeneira F , Amin N , Karssen LC , Vingerling JR , Hosseini SM , Döring A , Bettecken T , Vatavuk Z , Gieger C , Wichmann HE , Wilson JF , Fleck B , Foster PJ , Topouzis F , McGuffin P , Sim X , Inouye M , Holliday EG , Attia J , Scott RJ , Rotter JI , Meitinger T and Bailey-Wilson JE. 2013. Metaanalysis of genome-wide association studies in five cohorts reveals common variants in RBFOX1, a regulator of tissue-specific splicing, associated with refractive error. Hum. Mol. Genet. 22: 2754-2764.
    14. Statham P , Green L , Bichard M and Mendl M. 2011. A longitudinal study of the effects of providing straw at different stage of life on tail-biting and other behavior in commercially housed pigs. Applied Anim. Behav. Sci. 134: 100-104.
    15. Tabassum R , Chauhan G , Dwivedi OP , Mahajan A , Jaiswal A , Kaur I , Bandesh K , Singh T , Mathai BJ , Pandey Y , Chidambaram M , Sharma A , Chavali S , Sengupta S , Ramakrishnan L , Venkatesh P , Aggarwal SK , Ghosh S , Prabhakaran D , Srinath RK , Saxena M , Banerjee M , Mathur S , Bhansali A , Shah VN , Madhu SV , Marwaha RK , Basu A , Scaria V , McCarthy MI , DIAGRAM, INDICO, Venkatesan R , Mohan V , Tandon N and Bharadwaj D. Genome-wide association study for type 2 diabetes in Indians identifies a new susceptibility locus at 2q21. Diabetes 62: 977-986.
    16. Tabor MP , Brakenhoff RH , Ruijter-Schippers HJ , Van der Wal JE , Snow GB , Rene´ Leemans C and Braakhuis BJM.2002. Multiple head and neck tumors frequently originate from a single preneoplastic lesion. Am. J. Pathol. 161: 1051-1060.
    17. Turner SP , D'Eath RB , Roehe R and Lawrence AB. 2010. Selection against aggressiveness in pigs at re-grouping: Practical application and implications for long-term behavioural patterns. Anim. Welfare 19: 123-132.
    18. Yang J , Lee SH , Goddard ME and Visscher PM. 2011. GCTA: A tool for genome-wide complex trait analysis. Am. J. Hum. Genet. 88: 76-82.
    19. Yano K , Yamamoto E , Aya K , Takeuchi H , Lo PC , Hu L , Yamasaki M , Yoshida S , Kitano H , Hirano K and Matsuoka M. 2016. Genome-wide association study using wholegenome sequencing rapidly identifies new genes influencing agronomic traits in Rice. Nat. Genet. 48: 927-934.
    오늘하루 팝업창 안보기 닫기