Examination of Testicular Gene Expression Patterns in Yorkshire Pigs with High and Low Levels of Boar Taint

Animal Biotechnology, 21: 77–87, 2010 Copyright # Taylor & Francis Group, LLC ISSN: 1049-5398 print=1532-2378 online DOI: 10.1080/10495390903500607

EXAMINATION OF TESTICULAR GENE EXPRESSION PATTERNS IN YORKSHIRE PIGS WITH HIGH AND LOW LEVELS OF BOAR TAINT

Maxwell C. K. Leung, Kiera-Lynne Bowley, and E. James Squires Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario, Canada

Boar taint refers to the objectionable odor and flavor in meat of some uncastrated male pigs, which is primarily due to high levels of androstenone, a steroid produced in the testis, and 3-methylindole (skatole) which is produced by bacterial degradation of tryptophan in the intestinal tract. We determined testicular gene expression patterns of Yorkshire pigs with high and low levels of boar taint using swine DNA microarrays with two-color hybridization. The microarrays contained 19486 annotated probes; the expressions of 8719 genes were detected. Fifty-three genes were significantly up-regulated in the high boar taint group and four were significantly down-regulated (p < 0.05; fold change > 1.55). Gene ontology (GO) analysis short-listed 11 significant GO terms (p < 0.05), most of which are associated with steroid metabolism and mitochondrial components. Comparing the results of this study with published work on Duroc and Norwegian Landrace boars,1 eleven genes (HSB17B4, FDX1, CYP11A1, DHRS4, PRDX1, CYB5, CYP17A1, FTL, IDI1, SULT2A1, and RDH12) were over-expressed in all three breeds with a high androstenone level. The current findings confirmed a number of candidate genes identified in previous functional studies and suggest several new genes differentially expressed with different levels of boar taint.

Keywords: Boar taint; Candidate genes; Testis

Boar taint refers to the objectionable odor and flavor in meat of some intact male pigs (boars).2 Since a significant percentage of intact male pigs can possess this undesirable meat quality trait,3 most pigs in North America are castrated shortly after birth to remove the potential for taint.4 Castration, however, results in a lower feed efficiency, a lower lean yield, as well as animal welfare concerns.3,5 Selective breeding of intact pigs would be a more preferable method to prevent boar taint, so a better understanding of the genes involved in boar taint is warranted in order to develop genetic markers for boar taint. Boar taint is primarily caused by high levels of androstenone and=or skatole in carcasses.6 Androstenone is a steroid produced by the testis near sexual maturity,7,8

The authors thank Yanping Lou for technical assistance. This work was supported by a NSERC Discovery grant and funding from the Ontario Ministry of Agriculture and Food. Address correspondence to E. James Squires, Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario, Canada, N1G 2W1. E-mail: [email protected]

77 78 M. C. K. LEUNG ET AL. whereas skatole is produced by bacteria in the hindgut of the pig and then absorbed into the blood stream.9 The production of androstenone is mainly affected by genetic factors rather than diet and environmental factors.4 Several functional studies, furthermore, have identified a number of genetic markers associated with increased production of androstenone.10–15 A recent study was published on differences in gene expression profiles in Duroc and Norwegian Landrace boars with high and low fat androstenone,1 and a total of 563 and 160 genes were differentially expressed between the two groups, respectively. However, only 53 genes were differentially expressed in both breeds, suggesting that there were both breed specific changes in gene expression as well as general changes affecting boar taint. Moreover, the degree of sexual maturity and steroidogenic potential of the boars was not reported by Moe et al.1 Previous studies by Stewart et al.16 comparing gene expression profiles between boars with high and low levels of plasma estrone sulfate using a smaller (1,700 gene) human microarray identified a number of the same genes as the study by Moe et al.1 comparing low and high androstenone boars. Thus, in order to identify those genes that are specific to boar taint, it is important to differ- entiate between boars with high and low levels of steroidogenesis and boar with high and low levels of androstenone. This allows the separation of desirable genes for ‘‘maleness,’’ from genes that are important for boar taint, which we want to eliminate. The objective of the current experiment was to examine the differential gene expressions associated with high and low levels of boar taint in Yorkshire boars using a swine protein-annotated oligonucleotide microarray.17 The results were further analyzed using gene ontology and KEGG pathway analyses. The current results are also compared with the previous findings from Duroc and Norwegian Landrace boars. This will provide useful information for future studies identifying breed specific and general genetic markers associated with boar taint.

MATERIALS AND METHODS Tissue Samples and Biochemical Analysis Ten Yorkshire boars (144 40 kg) were selected from a larger pool obtained from local breeders in Ontario. Testis tissue samples were taken at time of slaughter, frozen in liquid nitrogen, and stored at 70C until use. Plasma and back fat samples were collected and stored at 20C. Sexual maturity of the boars was confirmed 18 by measuring the length of bulbourethral glands, as well as the plasma concentration of estrone sulfate (E1S) by radioimmunoassay.19 Fat and plasma androstenone levels were assayed using an ELISA method modified from Claus et al.20 as described by Squires and Lundstro¨m.21 Pigs having a fat androstenone level of less than 0.5 mg=g and greater than 1.0 mg=g were defined as low and high boar taint samples, respectively. Concentrations of skatole in plasma and fat were determined by a HPLC method modified after Claus et al.22 and Denhard et al.23 as described by Lanthier et al.24

Design of Microarray Experiment Five testis samples from boars with high levels of boar taint were paired with ﬁve testis samples from low taint boars for a competitive two-color microarray TESTICULAR GENE EXPRESSION AND BOAR TAINT 79 analysis. Relative gene expression of each pair of samples was assessed by a microarray with a replicate of dye swap for each pair for a total of ten microarrays used. The porcine 70-mer oligonucleotide-microarrays used in the current study were provided by the laboratory of David W. Galbraith at the University of Arizona. The oligonucleotides printed on the microarrays were supplied in part by the contri- bution of the US Pig Genome Coordinator, Max Rothschild, Iowa State University, and the contributions of the swine genome array coordination committee, Scott Fahrenkrug, University of Minnesota, Chair. Each slide contains 19486 annotated probes (18244 protein-annotated probes, 198 unrepresented porcine RefSeq genes, and 1044 TIGR porcine consensus sequences), 914 controls (mismatch and scramble sequences), and blanks. A full description of the porcine array and a complete list of EST annotation can be found at http://www.pigoligoarray.org/.

Preparation of RNA Samples and Dye-Labeled cDNA Total RNA was extracted from 200 mg of frozen testis sample using RNeasy Midi kit (Qiagen, Valencia, CA) following the manufacturer’s protocol. The quality and quantity of RNA was assessed using an Agilent 2100 Bioanalyzer (Santa Clara, CA). cDNA was prepared using 20 mg of total RNA, 8 ml of 5x first-strand buffer, 100 pmol of oligo(dT) primer, 20 mM dNTP-dTTP (6.67 mM each of dATP, dGTP, and dCTP), 2 mM of 2-aminoallyl-dUTP (AA-dUTP, Sigma-Aldrich Inc., St. Louis, MO), 2 mM dTTP, and 20 mM DTT in a reaction volume of 40 ml. The mixture was incubated at 65C for 5 minutes and then 42C for 5 minutes. Then, 4 ml of Super- script II Reverse Transcriptase (Invitrogen Corp., Carlsbad, CA) was added and the reactions were incubated for 4 hours at 42C. The reaction was then heated to 95C for 5 minutes to inactivate the enzyme. RNA was hydrolyzed by adding 8 ml of 1 M NaOH followed by a 15 minute incubation at 65C. Then 8 mlof1MHCl and 4 ml of 1 M Tris-Cl (pH 7.5) were added to neutralize the reaction. The PureLink PCR purification kit (Invitrogen, Carlsbad, CA) was used according to the manufacturer’s instructions to purify the cDNA in a final volume of 6 ml. Cyanine 3 (Cy3) and cyanine 5 (Cy5) monofunctional reactive dyes (Amersham Pharmacia Ltd., Piscataway, NJ) were each dissolved in 72 ml of dimethylsulphoxide and each dye was then aliquoted into 4.5 ml of dye per tube and stored at 70C until use. The dye aliquot and 3 ml of 0.3 M sodium bicarbonate was added to each sample and incubated for 2 hours in the dark at room temperature to allow for coupling of the dyes. The Qiaquick nucleotide removal kit (Qiagen) was used according to the manufacturer’s instructions to remove unincorporated dye. Following purification, the samples labeled with two different dyes were combined and 3.0 ml glycogen, 30 ml of 3 M sodium acetate, and 333 ml of isopropanol were added. The samples were incubated for 2 hours at 70C to precipitate the labeled cDNA. Following precipi- tation, the samples were reconstituted in 15 ml of water.

Hybridization of Microarray and Data Acquisition The labeled cDNA was added to 2.5 ml of yeast tRNA (Invitrogen; 10 mg=ml) and 2.5 ml of calf thymus DNA (Sigma; 10 mg=ml) and 50 ml DIG Easy Hyb solution (Roche, Basel, Switzerland). The mixture was incubated at 65C for 2 minutes, 80 M. C. K. LEUNG ET AL. cooled to room temperature, and transferred to a 24 60 mm cover slip for placement on the microarray. Hybridizations were conducted for 18 hours in a con- ventional hybridization oven at 42C in the dark. Following the hybridization, the cover slip was removed in 1 saline sodium citrate (SSC) after the hybridization. The microarray was then washed in 1 SSC, 0.1% sodium dodecyl sulfate (SDS), and 0.1 mM dithiothreitol (DTT) at 50C for 10 minutes for three times, in 1 SSC at 25C for two minutes for two times, and then in 0.1 SSC at 25C for one minute for one time. The microarray was then dried by centrifugation. The microarray was scanned using an Axon GenePix 4000A scanner and the spots were quantiﬁed with Gene-Pix Pro 3.0 analysis software (Axon Instruments, Inc., Foster City, CA). The PMT settings of the scanner were adjusted to give a glo- bal image ratio of 635 nm (Cy5)=532 nm (Cy3) of 1.00, and the median foreground Cy5 and Cy3 intensity values were obtained to create a table of raw spot intensities.

Data Processing and Analyses The raw spot intensities were analyzed using GeneSpring GX (Agilent). Spots showing a pixel intensity of less than 70 were excluded from further analysis. The intensities were normalized with a locally weighted linear regression (LOESS). Dif- ferences in gene expression between the two levels of boar taint were examined using Student’s t-test. Those having P < .05 and a fold change of 55% were shortlisted and corrected using the Benjamini and Hochberg procedure.25 The differentially expressed genes with corrected P < .05 were further examined with gene ontology analysis using DAVID Bioinformatics.26 Gene ontology terms with a corrected P < .05 after Benjamini Hochberg procedure were considered signiﬁcant.

RESULTS Biochemical Analysis The biochemical measurements of the boar samples are presented in Table 1. The bulbourethral gland length of each sampled boar was greater than 11.0 cm and the plasma E1S concentration was greater than 18 ng=mL. Plasma levels of estrone sulfate and bulbourethral gland lengths were not signiﬁcantly different (P > .05) between the high and low androstenone groups. The boars were, therefore, considered to be of sufﬁcient sexual maturity for the current study 18 and there was no difference in sexual maturity indicators between the two groups. Each 200 mg of

Table 1 Biochemical measurements of boar samples (mean SE)

Measurements Low boar taint group (N ¼ 5) High boar taint group (N ¼ 5)

Plasma androstenone (ng=mL) 17.8 6.5 52.6 18.6 Fat androstenone (mg=g) 0.411 0.046 5.611 3.117 Plasma skatole (ng=mL) 1.8 0.3 8.8 3.3 Fat skatole (ng=g) 16.0 4.2 150.5 61.4 Plasma estrone sulfate (ng=mL) 21.2 5.9 36.3 13.8 Bulbourethral gland length (cm) 13.8 0.6 13.3 1.3 TESTICULAR GENE EXPRESSION AND BOAR TAINT 81

Table 2 Differentially expressed genes associated with high boar taint levels in Yorkshire boars

Gene Fold Corrected Gene ID symbol Associated protein change P value

NM_001007026 ATN1 Atrophin-1 7.35 0.039 NM_022779 DDX31 ATP-dependent RNA helicase 6.88 0.045 NM_005376 MYCL1 L-myc-1 proto-oncogene protein 2.17 0.038 SLC35E4 solute carrier family 35, member E4 2.17 0.048 NM_006387 CHERP Calcium homeostasis endoplasmic reticulum þ1.57 0.032 protein NP_008640 COX3 Cytochrome c oxidase subunit 3 þ1.57 0.006 NM_024665 TBL1XR1 Transducin beta-like 1X-related protein 1 þ1.57 0.025 NM_214306 HSD17B4 17-beta-estradiol dehydrogenase 4 þ1.58 0.010 NP_002285 LAMP-2 Lysosome-associated membrane glycoprotein þ1.59 0.039 2 precursor NP_000206 JAK-3 Tyrosine-protein kinase JAK3 þ1.60 0.031 NP_008646 CYTB Cytochrome b þ1.61 0.016 NM_214065 FDX1 Adrenodoxin, mitochondrial precursor þ1.61 0.012 NM_001852 COL9A2 Collagen alpha-2(IX) chain precursor þ1.63 0.044 NM_001927 DES Desmin þ1.63 0.010 MT-ND1 NADH-ubiquinone oxidoreductase chain 4 þ1.63 0.028 NM_004046 ATP5A1 ATP synthase alpha chain þ1.65 0.022 MT-ND1 NADH-ubiquinone oxidoreductase chain 2 þ1.66 0.030 LONP2 Peroxisomal LON protease-like þ1.69 0.030 NM_001064 TKT Transketolase þ1.71 0.018 MT-ND5 NADH dehydrogenase 5, mitochondrial þ1.73 0.048 NM_214427 CYP11A1 Cytochrome P450 11A1, mitochondrial þ1.76 0.020 precursor NM_000146 FTL Ferritin, light polypeptide þ1.79 0.004 NM_003380 VIM Vimentin þ1.79 0.006 SMAP-1 Smooth muscle cell associated protein-1 þ1.80 0.030 isoform 2 NM_002574 PRDX1 Peroxiredoxin-1 þ1.82 0.005 NM_002018 FLII Protein ﬂightless-1homolog þ1.83 0.027 NM_005989 AKR1D1 3-oxo-5-beta-steroid 4-dehydrogenase þ1.84 0.033 NM_205845 AKR1C2 3-alpha-hydroxysteroid dehydrogenase þ1.86 0.014 type III NM_214355 EPHX1 Epoxide hydrolase 1 þ1.87 0.008 MT-ND1 NADH-ubiquinone oxidoreductase chain 1 þ1.87 0.005 ACT Alpha-1-antichymotrypsin precursor þ1.88 0.038 NM_012175 FBXO3 F-box only protein 3 þ1.89 0.046 NM_181738 PRDX2 Peroxiredoxin-2 þ1.89 0.016 NM_000558 HBA1 Hemoglobin alpha subunit þ1.91 0.007 NM_214300 MGST1 Microsomal glutathione S-transferase 1 þ1.94 0.003 NM_214073 CBR1 Carbonyl reductase 1 þ1.95 0.003 NM_001914 CYB5 Cytochrome b5 (2) þ1.95 0.024 NM_017742 ZCCHC2 Zinc ﬁnger CCHC domain-containing þ1.97 0.045 protein 2 NM_019111 HLA-DRA Major histocompatibility complex, class II, þ1.98 0.044 DR alpha precursor NM_181840 KCNK18 Potassium channel, subfamily K, member 18 þ2.01 0.002 NM_214019 DHRS4 Carbonyl reductase=NADP-retinol þ2.03 0.006 dehydrogenase

(Continued ) 82 M. C. K. LEUNG ET AL.

Table 2 Continued

Gene Fold Corrected Gene ID symbol Associated protein change P value

NM_014098 PRDX3 Peroxiredoxin-3 þ2.04 0.007 NP_002961 SAT1 Diamine acetyltransferase 1 þ2.05 0.002 NM_182920 ADAMTS9 ADAMTS-9 precursor þ2.11 0.008 NM_054012 ASS Argininosuccinate synthase þ2.11 0.044 NM_004508 IDI1 Isopentenyl-diphosphate delta-isomerase 1 þ2.19 0.015 NM_214428 CYP17A1 Cytochrome P450 17A1 þ2.21 0.002 NM_001037150 SULT2A1 DHEA preferring sulfotransferase þ2.21 0.004 NM_173552 C3orf58 Protein C3orf58 precursor þ2.22 0.022 NM_022051 EGLN1 Egl nine homolog 1 þ2.22 0.005 NM_213752 SC4MOL C-4 methylsterol oxidase þ2.26 0.025 NM_152443 RDH12 Retinol dehydrogenase 12 þ2.35 0.002 NM_000029 AGT Angiotensinogen precursor þ2.37 0.005 NM_003739 AKR1C3 Aldo-keto reductase family 1 member C3 þ2.39 0.002 NM_214429 CYP19A1 Cytochrome P450 19A1 þ3.06 0.011 NM_213755 STAR Steroidogenic acute regulatory protein, þ3.49 0.023 mitochondrial precursor NM_000519 HBD Hemoglobin delta subunit þ4.07 0.004

Genes that are also differentially expressed in Norwegian Landrace and Duroc boars1 are shown in bold. testis sample yielded an average of 109 32 mg of total RNA. All RNA samples showed sharp bands corresponding to 28S and 18S ribosomal subunits, thereby providing no evidence of RNA degradation.

Table 3 Signiﬁcant gene ontology terms associated with different taint levels

Number of Percentage of signiﬁcant signiﬁcant Corrected Category Terma genesb genes (%)c P valued

Biological process Steroid metabolism 10 17.2 4.80 106 Steroid biosynthesis 7 12.1 2.30 104 Cellular lipid metabolism 12 20.7 3.20 104 Lipid metabolism 13 22.4 3.30 104 Cellular component Cytoplasm 31 53.4 4.70 106 Mitochondria 13 22.4 3.30 104 Molecular function Oxidoreductase activity 20 34.5 1.50 109 Iron ion binding 12 20.7 7.50 107 Electron transporter activity 11 19.0 1.20 105 Oxygen binding 6 10.3 5.70 105 Transporter activity 18 31.0 2.50 103

aDifferent gene ontology terms may have overlapping genes; bolded terms are elaborated in Table 4. bNumber of the signiﬁcant genes (P < 0.05) found in the gene list associated with the gene ontology term. cPercentage of the signiﬁcant genes (P < 0.05) in the gene list associated with the gene ontology term. dThe P value of each gene ontology term was corrected using the Benjamini Hochberg procedure.25 TESTICULAR GENE EXPRESSION AND BOAR TAINT 83

Table 4 Differential gene expressions associated with selected gene ontology term

Terms Differential gene expression involved

Lipid metabolism SULT2A1,STAR,AKR1C3,SC4MOL,SERPINA3,CYP19A1,AKR1D1,CYP17A1, CYP11A1,IDI1,AKR1C2,HSD17B4,FDX1 Mitochondria ATP6,CYP11A1,ND1,CYTB,,PRDX3,MGST1,STAR,DHRS4,FDX1,ND2, ND4,FDX1 Oxidoreductase AKR1CL1,CYTB,AKR1C3,ND2,SC4MOL,CYP19A1,CYP17A1,AKR1D1, activity CYP11A1,ND1,AKR1C2,CYB5A,PRDX3,EGLN1,RDH12,CBR1,DHRS4, PRDX2,HSD17B4,ND4 Iron ion binding CYP11A1,HBD,CYTB,FTL,CYB5A,EGLN1,FDX1,HBB,HBA1,SC4MOL, CYP19A1,CYP17A1,PRDX3,EGLN1,RDH12,CBR1,DHRS4,PRDX2, HSD17B4,ND4 Transporter CYTB, STAR, AKR1C3, ND2, CYP19A1, AKR1D1, CYP11A1, ND1, AKR1C2, activity HBD, CYB5A, DHRS4, PRDX2, HSD17B4, HBB, FDX1, HBA1, ND4

Gene Expression Analysis The 8719 genes were expressed at an average pixel intensity greater than 70. The expressions of 3706 genes were up-regulated and 5013 were down-regulated in samples with a high level of boar taint compared to those with a low level of boar taint. The expressions of 108 genes showed a fold change greater than 1.55 with P < 0.05. Fifty-seven genes have a corrected P < 0.05 and, therefore, were deemed signiﬁcant after the Benjamini and Hochberg procedure (Table 2). Fifty three genes were signiﬁcantly over-expressed and four genes were under-expressed in the samples showing a high level of boar taint. Eleven of the over-expressed genes (HSB17B4, FDX1, CYP11A1, DHRS4, PRDX1, CYB5, CYP17A1, FTL, IDI1, SULT2A1, STAR, and RDH12) were also found in the Duroc and Norwegian Landrace pigs with a high androstenone level. In addition, while CYP19A1 and AKR1C4 were expressed at higher levels in high androstenone Duroc and Norwegian Landrace,1 we found similar genes (CYP19A2 and AKR1C2=3) at increased expression levels in high androstenone Yorkshire boars.

Gene Ontology Analysis Gene ontology analysis using the DAVID database revealed 4, 2, and 5 signiﬁ- cant terms of biological process, cellular component, and molecular function, respectively (Table 3). The differential gene expressions associated with ‘‘lipid metabolism,’’ ‘‘mitochondria,’’ ‘‘oxidoreductase activity,’’ ‘‘iron ion binding,’’ and ‘‘transporter activity’’ are given in Table 4.

DISCUSSION Our findings of differential gene expression support the current understanding of androstenone metabolism as well as the previous findings in functional studies. For instance, steroidogenic acute regulatory protein (STAR) is responsible for the 84 M. C. K. LEUNG ET AL. transport of cholesterol into mitochondria, while ferredoxin (FDX1) and cytochrome P45011A1 (CYP11A1) are involved in the conversion of cholesterol into pregnenolone. These two reactions are important regulatory steps in steroid synthesis in testis, and precede the synthesis of androgens and androst-16-ene steroids. Cytochrome b5 (CYB5) and cytochrome P45017A1 (CYP17A1), furthermore, are involved in the production of androst-16-ene steroids as well as the androgen dehy- droepiandrosterone from pregnenolone. The over-expression of STAR, CYP11A1, and CYP17A1 would result in an overall increase in steroid synthesis,27 while the expression of CYB5 has been shown to be correlated with increased synthesis of the first of the androst-16-ene steroids.28 The expression of aromatase (CYP19A2) was previously shown to be upregu- lated in Norwegian Landrace and Duroc boars with high boar taint,1 and this might also be an indicator of an increased level of sexual maturity in these boars. The increased expression of CYP19A1 in high boar taint Yorkshire boars in the present study is somewhat unexpected, since we found no significant difference in plasma levels of estrone sulfate between the high and low boar taint pigs in our study. The expression of isopentyl-diphosphate delta isomerase (IDI1) was increased in high boar taint pigs in all three breeds. This enzyme catalyzes the interconversion of the isoprenes, 3-isopentyl pyrophosphate and dimethylallyl pyrophosphate, which then undergoes successive condensation reactions to form cholesterol. However, since cholesterol is the precursor for the biosynthesis of the sex steroids (testosterone and estrogen) as well as androstenone, this gene may not be useful as a selective marker for androstenone biosynthesis.1 We found increased expression of hydroxysteroid sulfotransferase SULT2A1 in high boar taint Yorkshire boars and this has also been previously shown in both Duroc and Norwegian Landrace boars.1 We have previously reported that SULT2A1 is involved in the sulfation of androstenone and found a negative corre- lation between the expression of SULT2A1 and androstenone levels in fat;29 therefore, the relationship between sulfotransferase expression and boar taint requires further study. Ferritin (FTL), an iron storage protein, was over-expressed in the high boar taint testis samples in the current experiment as well as in Norwegian Landrace and Duroc boars.1 However, the role of ferritin in androstenone synthesis has not been investigated. Moe et al.1 proposed that the FTL over-expression was associated with the up-regulation of CYB5=CYP450 electron transfer. Steroidogenesis has been associated with elevated oxidative stress in testis.30 The over-expression of FTL may be a protective mechanism against oxidative stress by preventing the Fenton reaction.31 PRDX1, a member of the peroxiredoxin family of antioxidant enzymes, which reduce hydrogen peroxide and alkyl hydroperoxides, may also play an antioxidant protective role in cells. These genes, however, are unlikely to be specific markers for boar taint. A number of dehydrogenase enzymes, including retinol dehydrogenase 12 (RDH12), carbonyl reductase=NADP-retinol dehydrogenase (DHRS4), aldo-keto reductase 1 (AKRC1), and 17b-dehydrogenase (HSD17B4) were expressed at higher levels in high androstenone boars of all three breeds and these may be involved in the biosynthesis of androstenone. Human RDH12 was shown to be involved in steroid TESTICULAR GENE EXPRESSION AND BOAR TAINT 85 metabolism, reducing dihydrotestosterone to androstanediol in vitro.32 Murine RDH12, in contrast, did not show any steroid-metabolizing activities. In another study, RDH12 was found to be differentially expressed in response to 17a- methyldihydrotestosterone exposure in zebra fish.33 These findings suggested that RDH12 may be involved in the metabolism of androstenone, although its specific function requires further investigation. Carbonyl reductase=NADP-retinol dehydrogenase (DHRS4) from pig has also been shown to reduce 3-ketosteroids to 3-hydroxy steroids.34 17b-dehydrogenase (HSD17B4) is involved in the conversion of the 17-keto group of androstenedione and estrone to the 17-hydroxy group in testosterone and estradiol respectively and may not play a role in the metabolism of androstenone. Aldo-keto reductase 1 (AKRC1) is part of a group of enzymes that can act as 3a and 3b hydroxysteroid dehydrogenases and convert androstenone to either 3a-androstenol or 3b-androstenol.35,36 In addition, 3b-hydroxysteroid dehydrogenase is involved in the conversion of androst-5,16,dien-3-ol to androst- 4,16,dien-3-one in the biosynthesis of androstenone,35 and this might explain its increased expression in high androstenone boars. The microarrays used in the present study with Yorkshire pigs were different that those used by Moe et al.1 in the Duroc and Norwegian Landrace experiments. Annotation or oligo design issues of closely related genes may explain the different variants found for the different populations. There have been a few studies designed to identify QTLs for androstenone and skatole accumulation. Two studies involved the same experimental cross between Large White and Meishan pig breeds.11,37 Quintanilla et al.37 found significant QTLs for androstenone on chromosomes (SSC) 3 at region 38–44 cM, SSC 6 at 186 cM, SSC 7 at 43–66 cM, SSC 9 at 156 cM and SSC 14 at 47 cM. The QTL on SSC 7, close to the major histocompatibility complex of the pig, showed the largest effects; it was the only QTL that was consistently found between the two studies. Lee et al.37 also found a QTL for skatole on SSC 14 at 61 cM and a QTL for androstenone and boar flavor on SSC 6 at 90 cM and SSC 14 at 100 cM. A third study11 was carried out with a Landrace outbred population; these authors failed to detect any QTLs for androstenone in any of the candidate chromosome regions and, instead, detected a QTL for skatole on SSC6 at 93 cM. Despite the variations in the results from the different studies, it is useful to determine if the chromosomal locations of differentially expressed genes from microarray studies map to these QTL regions. The HBA1 gene maps to 3:35.25 m which is close to the QTL location on SSC 3. The PRDX1 gene maps to 6:118.77 m which compares to QTL regions on SSC 6 near 92 cm. For SSC 7, the STAR gene at 7:56.24 m and the DHRS4 gene at 7:81.58 m are potential candidates. Six of the differentially expressed genes are located on SSC 14: KCNK18 at 14: 13.32 m, PRDX3 at 14:13.51 m, EGLN1 at 14:61.45 m, AGT at 14:62.04 m, CYP17A1 at 14:118.88 m and SLC35E4 at 14:48.32 m. This compares to the QTL regions on SSC 14 at 47 cM and 100 cM. Further fine mapping of the QTL regions is necessary to confirm the involvement of these genes in the QTL effects. In summary, we have extended the results of differential gene expression patterns in high and low androstenone boars from Duroc and Norwegian Landrace breeds to include Yorkshire boars. Our results confirm the differential expression of a number of genes that are common among all three breeds and identify others 86 M. C. K. LEUNG ET AL. that may be unique to the Yorkshire breed. Future studies using these candidate genes to identify genetic markers for boar taint are warranted.

REFERENCES

1. Moe M, Meuwissen T, Lien S, Bendixen C, Wang X, Conley LN, Berget I, Tajet H, Grindflek E. Gene expression profiles in testis of pigs with extreme high and low levels of androstenone. BMC Genomics. 2007;8:405. 2. Bonneau M. Boar taint in entire male pigs-a review. Pig News Inf. 1997;18:15–18. 3. Thun R, Gajewski Z, Janett F. Castration in male pigs: techniques and animal welfare issues. J Physiol Pharmacol. 2006;57(Suppl. 8):189–194. 4. Zamaratskaia G, Squires EJ. Biochemical, nutritional and genetic effects on boar taint in entire male pigs. Animal Dec. 2008;1508–1521 5. Babol J, Squires EJ. Quality of meat from entire male pigs. Food Res Int. 1995;28:201–212. 6. Claus R, Weiler U, Herzog A. Physiological aspects of androstenone and skatole forma- tion in the boar—A review with experimental data. Meat Sci. 1994;38:289–305. 7. Gower DB, Harrison FA, Heap RB, Saat YA. Studies of the in-vivo biosynthesis and the excretion of C 19 16-unsaturated steroids in the boar. J Endocrinol. 1972;52:3–4. 8. Gower DB. 16-Unsaturated C19 steroids-A review of their chemistry, biochemistry and possible physiological role. J. Steroid Biochem. 1972;3:45. 9. Jensen MT, Cox RP, Jensen BB. 3-Methylindole (skatole) and indole production by mixed populations of pig fecal bacteria. Appl Environ Microbiol. 1995;61:3180–3184. 10. Diaz GJ, Squires EJ. Metabolism of 3-methylindole by porcine liver microsomes: responsible cytochrome P450 enzymes. Toxicol Sci. 2000;55:284–292. 11. Varona L, Vidal O, Quintanilla R, Gil M, Sańchez A, Folch JM, Hortos M, Rius MA, Amills M, Noguera JL. Bayesian analysis of quantitative trait loci for boar taint in a Landrace outbred population. J Anim Sci. 2005;83:301–307. 12. Lee GJ, Archibald AL, Law AS, Lloyd S, Wood J, Haley CS. Detection of quantitative trait loci for androstenone, skatole and boar taint in a cross between Large White and Meishan pigs. Anim Genet. 2004;36:14–22. 13. Skinner TM, Doran E, McGivan JD, Haley CS, Archibald AL. Cloning and mapping of the porcine cytochrome-P450 2E1 gene and its association with skatole levels in the domestic pig. Anim Genet. 2005;36:417–422. 14. Lin Z, Lou Y, Squires EJ. Functional polymorphism in porcine CYP2E1 gene: Its association with skatole levels. J Steroid Biochem Mol Biol. 2006;99:231–237. 15. Peacock J, Lou Y, Lundstro¨m K, Squires EJ. The effect of a c.-8G > T polymorphism on the expression of cytochrome b5A and boar taint in pigs. Anim Genet. 2008;39:15–21. 16. Stewart JD, Lou Y, Squires EJ, Coussens PM. Using human microarrays to identify differentially expressed genes associated with increased steroidogenesis in boars. Anim Biotechnol. 2005;16:139–151. 17. Fahrenkrug SC, Venkatraman A, Dickens CM, Murtaugh MP, Hering BJ, Archibald A, Cassady JP, Ernst CW, Lunney J, Reecy J, Rothschild MF, Elsik CG. 2007. A phylogen- etically annotated porcine whole genome 70-mer oligonucleotide microarray. In: Plant and Animal Genome XIV Conference, January 13–17, 2007, Town and Country Conven- tion Center, San Diego, California. Abstract P757. E-document: http://www.intl-pag. org/pag/15/abstracts/PAG15_P07a_757.html. Last accessed 16 February, 2010. 18. Sinclair PA, Squires EJ, Raeside JI. Early postnatal plasma concentrations of testicular steroid hormones, pubertal development, and carcass leanness as potential indicators of boar taint in market weight intact male pigs. J Anim Sci. 2001;79:1868–1876. TESTICULAR GENE EXPRESSION AND BOAR TAINT 87

19. Schwarzenberger F, Toole GS, Christie HL, Raeside JI. Plasma levels of several androgens and estrogens from birth to puberty in male domestic pigs. Acta Endocrinol (Copenh). 1993;128:173–177. 20. Claus R, Mahler, G, Muunster,} E. Determination of boar taint steroid 5a-androst-16-en-3- one in adipose tissue of pigs with a rapid microtiter plate enzyme-immunoassay (MTE). Archiv Lebenmittelhyg. 1988;4:87–90. 21. Squires EJ, Lundstro¨m K. Relationship between cytochrome P450IIE1 in liver and levels of skatole and its metabolites in intact male pigs. J Anim Sci. 1997;75:2506–2511. 22. Claus R, Dehnhard M, Herzog A, Bernal-Barragan H, Gimeńez, T. Parallel measurements of indole and skatole (3-methylindole) in feces and blood plasma of pigs by HPLC. Livest Prod Sci. 1993;34:115–126. 23. Denhard M, Claus R, Hillenbrand M, Herzog A. High performance liquid chromato- graphic method for the determination of 3-methylindole (skatole) and indole in adipose tissue of pigs. J Chromatogr. 1993;616:205–209. 24. Lanthier F, Lou Y, Squires, EJ. Skatole metabolism in the intact pre-pubescent male pig: The relationship between hepatic enzyme activity and skatole concentrations in plasma and fat. Livest Sci. 2007;106:145–153. 25. Benjamini Y, Hochberg Y. Controlling the false discovery rate—a practical and powerful approach to multiple testing. J R Stat Soc Ser B. 1995;57:289–300. 26. Dennis G Jr, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA. DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 2003;4:P3. 27. Blomberg LA, Zuelke KA. Expression analysis of the steroidogenic acute regulatory protein (STAR) gene in developing porcine conceptuses. Mol Reprod Dev. 2005;72:419–429. 28. Lin Z, Lou Y, Peacock J, Squires EJ. A novel polymorphism in the 5’ untranslated region of the porcine cytochrome b5 (CYB5) gene is associated with decreased fat androstenone level. Mamm Genome. 2005;16:367–373. 29. Sinclair PA, Gilmore WJ, Lin Z, Lou Y, Squires EJ. Molecular cloning and regulation of porcine SULT2A1: Relationship between SULT2A1 expression and sulfoconjugation of androstenone. J Mol Endocrinol. 2006;36:301–311. 30. Peltola V, Huhtaniemi I, Metsa-Ketela T, Ahotupa M. Induction of lipid peroxidation during steroidogenesis in the rat testis. Endocrinology. 1996;137:105–112. 31. Orino K, Watanabe K. Molecular, physiological and clinical aspects of the iron storage protein ferritin. Vet J. 2008;178:191–201. 32. Keller B, Adamski J. RDH12, a retinol dehydrogenase causing Leber’s congenital amaurosis, is also involved in steroid metabolism. J Steroid Biochem Mol Biol. 2007;104:190–194. 33. Hoffmann JL, Thomason RG, Lee DM, Brill JL, Price BB, Carr GJ, Versteeg DJ. Hepatic gene expression profiling using GeneChips in zebrafish exposed to 17alpha- methyldihydrotestosterone. Aquat Toxicol. 2008;87:69–80. 34. Endo S, Maeda S, Matsunaga T, Dhagat U, El-Kabbani O, Tanaka N, Nakamura KT, Tajima K. Molecular determinants for the stereospecific reduction of 3-ketosteroids and reactivity towards all-trans-retinal of a short-chain dehydrogenase=reductase (DHRS4). Arch Biochem Biophys. 2009;481:183–190. 35. Gower DB. 16-Unsaturated steroids. A review of their chemistry, biochemistry and possible physiological role. J Steroid Biochem. 1972;3:45–103. 36. Doran E, Whittington FM, Wood JD, MacGivan JD. Characterization of androstenone metabolism in pig liver microsomes. Chem-Biol Interact. 2004;147:141–149. 37. Quintanilla R, Demeure O, Bidanel JP, Milan D, Iannuccelli N, Amigues Y, Gruand J, Renard C, Chevalet C, Bonneau M. Detection of quantitative trait loci for fat androstenone levels in pigs. J Anim Sci. 2003;81:385–394. Copyright of Animal Biotechnology is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.