Biosci. Biotechnol. Biochem., 75 (3), 451–458, 2011 Differences in Hepatic Gene Expression as a Major Distinguishing Factor between Korean Native Pig and Yorkshire Seung-Soo KIM,1 So-Ra KIM,1 Jung-Rok KIM,1 Jin-Kyoo MOON,1 Bong-Hwan CHOI,2 y Jae-Won LEE,3 Kwan-Suk KIM,4 Tae-Hun KIM,2 Hyun-Jung KIM,5 and Cheol-Koo LEE1; 1College of Life Sciences and Biotechnology, Korea University, Seoul 136-701, Korea 2Division of Animal Genomics and Bioinformatics, National Institute of Animal Science, Rural Development Administration, Omokchun-dong 564, Kwonsun-gu 441-706, Korea 3Department of Statistics, College of Political Science and Economics, Korea University, Seoul 136-701, Korea 4Department of Animal Science, Chungbuk National University, Cheongju 361-763, Korea 5College of Pharmacy, Chung-Ang University, Seoul 156-756, Korea Received August 26, 2010; Accepted November 25, 2010; Online Publication, March 7, 2011 [doi:10.1271/bbb.100625] There are phenotypic differences between Korean physiology to humans.7) There are diverse pig breeds native pig (KNP) and Yorkshire (YS) breeds due to worldwide generated by the demarcation of continents different interests in selection. YS has been selected and regions, including Korean pigs (KNPs from Jeju for industrial interests such as a growth rate and lean island and Seonghwan, on the mainland), Chinese pigs meat production, while KNP has been maintained as a (for example Min pig, Xiang pig, and Wuzhishan pig) regional breed with local interests such as disease and European pigs (for example Berkshire, Duroc, resistance and fat content in and between muscle. A Landrace, and YS). These pig breeds have diverse comparison of gene expression profiles from liver tissue genetic features and phenotypic traits that provide rich reflected overall long-term effects of artificial selection resources for animal genetics. for these two pig breeds. Based on minimum positive DNA microarrays are used as a tool in genetic false discovery rate (less than 10%) and fold change research for gene expression and genetic variation. The (jFCj > 1:5), 73 differentially expressed genes (DEGs) porcine genome sequence consortium has completely were identified. Functional analysis of these DEGs sequenced the whole genome, and the first draft is about indicated clear distinctions in signaling capacity related to be released. Currently (August 2010), the NCBI map to epidermal growth factor (EGF), extracellular struc- viewer provides ordered sequence information for nine ture, protein metabolism, and detoxification. Hepatic of 18 autosomes, including 1, 4, 5, 7, 11, 13, 14, 15, and DEGs demonstrated the importance of hormonal and 17, and the X chromosome. The release of the porcine metabolic capabilities to differences between these two genome sequence should accelerate new genetic dis- pig breeds. coveries and the usability of single nucleotide poly- morphisms (SNPs). In particular, for aspects of func- Key words: gene expression profile; Korean native pig; tional genomics, the transcriptome database should be Yorkshire; liver; differentially expressed important and useful in annotating the pig genome.8) gene We have demonstrated the usefulness of transcrip- tome data by searching for DEGs that characterize the The liver is the largest organ in a mammalian system. differences among pig breeds in adipose tissue and It plays an important role in regulating overall body skeletal muscle.9,10) Functional analysis of DEGs in fat growth by secreting insulin-like growth factor 1 (IGF1), tissue from KNP and YS suggested that there was a produced in response to growth hormone secreted from difference in xenobiotic metabolism.10) DEGs in this the pituitary gland. It is also the center for energy functional category were significantly down-regulated in metabolism of major nutrients, including carbohydrates, KNP, and suggests that the difference plays an important lipids, and proteins,1) and for the elimination of toxic role in overall meat flavor. In addition, we found a chemicals from the body.2) The liver also can regenerate significantly higher content of arachidonic acid and a after hepatic damage by inducing active cell prolifer- lower expression level for cytochrome P450 genes in the ation.3,4) Because of these diverse and active functions, adipose tissue of KNP as compared to YS.11) In the hepatocytes have large number of mitochondria and free skeletal muscle, substantial numbers of DEGs were ribosomes.5) related to proliferation and differentiation. We did not Pigs (Sus scrofa) are used as a research model for find any clear direction of changes in these functions, animal genetics because of their large number of but most proliferation-related DEGs were directly or offspring, relatively short generation interval, and cost- indirectly associated with p53.9) To add to these results, effective population maintenance and disease control.6) we investigated DEGs in the liver tissue. The overall They are also used as a biomedical research model gene expression profiles from KNP and YS showed a because of their similarity in organ size, structure, and clear difference between the two breeds. y To whom correspondence should be addressed. Tel: +82-2-3290-3008; Fax: +82-2-921-1715; E-mail: [email protected] 452 S.-S. KIM et al. Materials and Methods Statistical analysis and identification of DEGs. To obtain statisti- cally significant DEGs between the two breeds, the signals were Animals and the collection of animal tissues. The KNP and YS transformed into a log2 scale to stabilize data variation. To reduce boars used in this experiment were raised under identical conditions of array variation further, the signals were processed by quantile 12) feeding and management in the National Livestock Research Institute, normalization. The signals between the two groups were compared Jeju island, South Korea. When the animals reached market weight by t-test, and the p values were corrected by the minimum positive 13) (89–100 kg for 7 month-old KNPs and 100–110 kg for 6 month-old YS false discovery rate (minimum pFDR ¼ q value). All statistical pigs), eight pigs (four of each breed) were slaughtered on the same day. analyses were done using the Partek Genomics Suite (Partek, USA). To minimize unnecessary variation, we stopped feeding for 24 h before When the transcripts were tested to within less than 10% of the q value slaughter and transported them using the same vehicle. In addition, the error rate and more than 1.5 fold of the difference, they were defined as liver tissue was taken from the end of the middle lobe of all eight pigs. DEGs. In the gene annotation and functional studies, we used the Tissue samples were immediately frozen in liquid nitrogen and stored information from Affymetrix Porcine Annotation Revision 5 (http: 14) at À80 C until preparation of total RNA. All the animals were fed //www4.ncsu.edu/~stsai2/annotation/) for the gene names, and used identically depending on their developmental status. All the diets were GeneCards and PubMed to search for the functions of the genes. purchased from the Nonghyup Feed Inc. (Seoul, Korea). We fed newly born piglets TrueMilk (brand name) for 3 weeks. Then we changed Quantitative real-time polymerase chain reaction (qPCR). To the diet to TrueMill #1 and TrueMill #2 in sequence for 4 weeks. estimate the real differences in DEGs from the microarrays, we After that, we fed the pigs with MyungPum Plus Jutdon for 1.5 months, independently performed real time PCR with a rotary-type real-time then changed to MyungPum Plus Yukseongdon until slaughter. analyzer, Rotor-Gene 6000 (Corbett Research, Australia). The genes TrueMill #1, #2, and MyungPum Plus Jutdon contained antibiotic we tested included all 15 DEGs in a structural category with a mixtures for pathogens. However, the final diet, MyungPum Plus reference gene. All primer sets were designed using sequence Yukseongdon did not contain any antibiotics. The company-made diets information from the NetAffx Analysis Center (http://www.affymetrix. were based on the National Research Council (NRC) standard of the com/analysis/index.affx) using the Primerquest program (http://www. United States. idtdna.com/Scitools/Applications/Primerquest/). Briefly, to obtain the cDNA for qPCR, 25 mg of total RNA was reverse-transcribed. For Ribonucleic acid extraction, target preparation, and DNA micro- cDNA synthesis, total RNA was mixed with 12.5 mg anchored array hybridization. Total RNA was extracted from 1 mg of liver tissue Oligo(dT)20 primer (Invitrogen, USA) and 5 mLof10mM dNTP mix using 2–3 mL of TRIzolÒ Reagent (Invitrogen, USA) by homogeni- (Invitrogen, USA), then incubated at 65 C for 5 min. After the zation with a Power Gen 125 S1 (Fisher Scientific, USA). RNA purity reaction, the tubes were placed on ice. While still on ice, 20 mLof5X M was measured about 1.5 as an A260/A280 ratio by UV spectropho- First-Strand Buffer (Invitrogen, USA) and 10 mL of 0.1 DTT tometer (Eppendorff, Germany). The integrity of the total RNA was (Invitrogen, USA) were added, and the reaction was incubated at Ò checked by electrophoresis on a 1% agarose gel containing form- 42 C for 2 min. Inside the PCR machine, 1,000 units of SuperScript aldehyde. Ten mg of total RNA after DNase I (Invitrogen, USA) II Reverse Transcriptase (Invitrogen, USA) was added, and the mixture treatment was converted into double-stranded cDNAs using a was further incubated at 42 C for 50 min. Then the reaction was GeneChipÒ Expression 30-Amplification One Cycle cDNA Synthesis inactivated at 70 C for 15 min. Using this cDNA solution, qPCR was Kit (Affymetrix, USA), and cDNA was purified with the GeneChipÒ done in a PCR tube at 25 mL total volume containing 5 mM forward and Sample Cleanup Module (Affymetrix, USA). Biotin-labeled antisense reverse primers, and 12.5 mL of 2X SensiMixPlus SYBR (2X mix cRNA was synthesized from the purified cDNA through in vitro containing reaction buffer, heat-activated DNA polymerase, dNTPs, Ò transcription (IVT) using GeneChipÒ Expression 30-Amplification 6mM MgCI2, internal reference dye, stabilizers, and SYBR Green I) Reagents for IVT Labeling Kit (Affymetrix, USA).