Int. J. Biol. Sci. 2011, 7 659 Ivyspring International Publisher International Journal of Biological Sciences 2011; 7(5):659-663 Letter The Heparan and Heparin Metabolism Pathway is Involved in Regulation of Fatty Acid Composition Zhihua Jiang1,, Jennifer J. Michal1, Xiao-Lin Wu2, Zengxiang Pan1 and Michael D. MacNeil3 1. Department of Animal Sciences, Washington State University, Pullman, WA 99164-6351, USA; 2. Department of Dairy Science, University of Wisconsin-Madison, Madison, WI 53706-1284, USA; 3. USDA-ARS, Fort Keogh Livestock and Range Research Laboratory, Miles City, MT 59301, USA Corresponding author: Dr. Zhihua Jiang, Phone: 509-335-8761; Fax: 509-335-4246; Email: [email protected] © Ivyspring International Publisher. This is an open-access article distributed under the terms of the Creative Commons License (http://creativecommons.org/ licenses/by-nc-nd/3.0/). Reproduction is permitted for personal, noncommercial use, provided that the article is in whole, unmodified, and properly cited. Received: 2011.03.04; Accepted: 2011.05.16; Published: 2011.05.21 Abstract Six genes involved in the heparan sulfate and heparin metabolism pathway, DSEL (dermatan sulfate epimerase-like), EXTL1 (exostoses (multiple)-like 1), HS6ST1 (heparan sulfate 6-O-sulfotransferase 1), HS6ST3 (heparan sulfate 6-O-sulfotransferase 3), NDST3 (N-deacetylase/N-sulfotransferase (heparan glucosaminyl) 3), and SULT1A1 (sul- fotransferase family, cytosolic, 1A, phenol-preferring, member 1), were investigated for their associations with muscle lipid composition using cattle as a model organism. Nineteen single nucleotide polymorphisms (SNPs)/multiple nucleotide length poly- morphisms (MNLPs) were identified in five of these six genes. Six of these mutations were then genotyped on 246 Wagyu x Limousin F2 animals, which were measured for 5 carcass, 6 eating quality and 8 fatty acid composition traits. Association analysis revealed that DSEL, EXTL1 and HS6ST1 significantly affected two stearoyl-CoA desaturase activ- ity indices, the amount of conjugated linoleic acid (CLA), and the relative amount of saturated fatty acids (SFA) and monounsaturated fatty acids (MUFA) in skeletal muscle (P<0.05). In particular, HS6ST1 joined our previously reported SCD1 and UQCRC1 genes to form a three gene network for one of the stearoyl-CoA desaturase activity indices. These results provide evidence that genes involved in heparan sulfate and heparin me- tabolism are also involved in regulation of lipid metabolism in bovine muscle. Whether the SNPs affected heparan sulfate proteoglycan structure is unknown and warrants fur- ther investigation. Key words: Heparan sulfate and heparin metabolism pathway, muscle fatty acid composition, as- sociations, genetic networks. Research has shown that the enzymes and pro- lism [1]. Both heparan sulfate and heparin are mem- teins encoded by DSEL (dermatan sulfate epi- bers of the glycosaminoglycan family of carbohy- merase-like), EXTL1 (exostoses (multiple)-like 1), drates that are very closely related in structure. As HS6ST1 (heparan sulfate 6-O-sulfotransferase 1), reviewed by Kolset and Salmivirta [2], cell surface HS6ST3 (heparan sulfate 6-O-sulfotransferase 3), heparan sulfate proteoglycans play biological roles in NDST3 (N-deacetylase/N-sulfotransferase (heparan several aspects of lipoprotein metabolism. For exam- glucosaminyl) 3), and SULT1A1 (sulfotransferase ple, the binding of lipoproteins to heparan sulfate family, cytosolic, 1A, phenol-preferring, member 1), presents an important process for the cellular uptake are involved in heparan sulfate and heparin metabo- and turnover of lipoproteins. Heparan sulfate also http://www.biolsci.org Int. J. Biol. Sci. 2011, 7 660 serves as a primary interaction site for lipoprotein Table 1. Primers designed for mutation detection in six lipase and hepatic lipase on cell surfaces and trans- bovine genes ports lipoprotein lipase from extravascular cells to the Region Primer sequences (5’-3’) Size Tm luminal surface of the endothelia. Furthermore, Wilsie DSEL and colleagues [3] found that heparan sulfate proteo- Promoter F-GGAAGCAAGACGCTCTTCATTTGT 551 bp 60°C glycans facilitate fatty acid transport across the plas- ma membrane of adipocytes, thus contributing to in- R-AAAGAGGAGCCCAGATGCAAAGAT tracellular lipid accumulation in the cell. On the other 3’UTR (I) F-CAGCACAGTTTTTGGTGATTTGGT 553 bp 60°C hand, heparin has been reported to decrease the deg- R-TTTTCTGCCAACATGAAGGGAAAT radation rate of lipoprotein lipase in adipocytes [4] 3’UTR (II) F-TTCCAAACCTTAGCCGGGTATCTT 572 bp 60°C and promote adipocyte differentiation [5]. In the pre- R-AGCTGAAATCATGGGACTGCATTT sent study, we tested the hypothesis that genes in- EXTL1 volved in heparan sulfate and heparin metabolism are Exon 1 F- CTCACAGACAGGAGCCAATCAGAG 584 bp 60°C also involved in regulation of lipid metabolism in R- CCTGACCTTAGCCTTGAGGGAGAG bovine muscle. Cattle were used as a model organism in the 3’UTR (I) F- CGTAAGAAGTATCGCAGCCTGGAG 571 bp 60°C present study. The bovine DSEL, EXTL1, HS6ST1, R- CCAGGCACGTAGCTGATGTCTATC HS6ST3, NDST3 and SULT1A1 genes were annotated 3’UTR (II) F- GCCTAATGAACTCCACGCCTACAC 589 bp 60°C using a protocol described previously [6]. In brief, a R- GCTACCACTCAGCCCACCTAGAAA cDNA sequence for each of these genes was retrieved HS6ST1 from the GenBank database and then extended to a 3’UTR F- AGTCCCTAGACTGAGGGGAGCTGT 588 bp 60°C full length cDNA sequence using electronic rapid R- AGCGTTTGCAATGGACTGAACAT amplification of cDNA ends (e-RACE) [7]. Next, the HS6ST3 full-length cDNA sequence was used to search for 3’UTR (I) F- GCTTGGATGTTCTGCTGAAACTGA 571 bp 60°C genomic DNA contigs against the 7.15X bovine ge- nome sequence database (see the Bovine Genome R- AAGAGGCCTGCTCCAAATAGGAAA Resources at NCBI). A total of 15 primer pairs were 3’UTR (II) F- AAGGAGCTGAAGGCAAAATGAGTG 537 bp 60°C designed to amplify various targets located in 6 genes R- TCTGGACAATAACGGGTGGTTTCT (Table 1). Approximately 50 ng of genomic DNA from 3’UTR F- TCTCCCTTCCTGATGATTTGTTCC 550 bp 60°C (III) each of six Wagyu x Limousin F1 bulls were amplified in a final volume of 10 μl that contained 12.5 ng of R- GGAGAGGACAAGTGTGTTGCTTCA NDST3 each primer, 150 μM dNTPs, 1.5 mM MgCl2, 50 mM KCl, 20 mM Tris-HCl and 0.25 U of AmpliTaq Gold Exon 2 F- CATTCTCCATTGCTTCACATGACC 520 bp 60°C polymerase (Applied Biosystems, Branchburg, NJ). R- ATGGCAGACAACTCATCCCAGTTT PCR conditions were as follows: 95ºC for 10 minutes, Exon 14 F- CTTGTATCTCCTCCTCCCACCTCA 538 bp 60°C 35 cycles of 94ºC for 30 sec, 60ºC for 30 sec, and 72ºC R- CAGGCAAACAGCAGCCTAAAAGTC for 30 sec, and an extension step at 72ºC for 10 min. SULT1A1 PCR amplicons were sequenced on a capillary se- Promoter F- AGGCAAGAATACTGGAGTGGGTTG 601 bp 60°C quencer by High-Throughput Sequencing Solutions (Seattle, WA). A total of 19 mutations were identified R- AGATGCCAAGAGTTCAGGTGGAAG in five of these six genes, including 2 single nucleotide Exon 8 F- AGAGGACCACAGTCAAGGAACAGG 576 bp 60°C polymorphisms (SNP) in DSEL, 1 multiple nucleotide R- ATATGCCTCCAGAGGACCACTCAC length polymorphism (MNLP) in EXTL1, 8 SNPs in 3’UTR F- CTGTTGGGAGCAAAGAACAAACCT 507 bp 60°C HS6ST1, 3 SNPs and 1 MNLP in HS6ST3 and 4 SNPs R- GACTGCGTTCACACATCTCCACTT in NDST3, respectively (Figure S1). Based on the ini- tial linkage disequilibrium of these mutations ob- served among six Wagyu x Limousin F bulls and 1 As described previously [8-9], 19 phenotypes their compatibility in forming multiplexes for geno- were measured on these F animals, including 5 car- typing with the Sequenom iPLEX assay design, only 2 cass, 6 eating quality and 8 fatty acid composition six mutations (see Figure S1) were genotyped on 246 traits. Associations between genotypes and pheno- F animals by the Genomics Center at the University 2 types were evaluated using linear models described of Minnesota. previously by Daniels et al. [6]. Systematic factors in the linear models included the effects of harvest year http://www.biolsci.org Int. J. Biol. Sci. 2011, 7 661 (i=1,2,3), sex (j=1,2), sire (k=1,2,3,4,5,6), and age in cantly affected MUFA in an additive QTM. The p days at harvest (as a covariate). The effects of markers values of genetic modes were obtained from Monte were estimated either individually in the model or Carlo simulation with 10,000 replicates. For example, jointly in a multiple regression. In single-marker over-dominance genetic mode was evaluated as Hα: { analyses, ANOVA was conducted by testing the ˆQq max ˆ qq , ˆ QQ or ˆQq min ˆ qq , ˆ QQ } vs. model with the presence of marker effects 3 H0: {Otherwise}, where ˆ X was the estimated effect ( y* x b e ) vs. the model assuming null i ij j i of genotype X. Note that, when two or more markers j1 (genes) significantly affected a trait, the effects esti- marker effects ( y *=µ+e ), where y * is the phenotypic i i i mated by single marker analyses were theoretically value of the i-th individual which has been adjusted biased. So, the three markers with their QTMs on dif- for the effects of harvest years, sexes, sires, and age in ferent phenotypes were then merged with other days at harvest using a full model. Equivalently, this markers previously reported by Jiang et al. [9] and yields the null hypothesis H : µ = µ = µ = µ vs. the 0 1 2 3 combined into a multiple regression analysis for each alternative H : {Otherwise), where µ = µ+b is the α i i trait in attempt to identify their roles in the genetic mean of the i-th genotypes. The resulting p values regulation of fatty acid composition. The AIC-based were adjusted using the Bonferroni correction [10]. model selection suggested that the addition of Briefly, let the significance level for the whole family HS6ST1 with SCD1 and UQCRC1 formed a three-gene of tests be (at most) α, then the Bonferroni correction network for R3 (Figure 1B), because it had a smaller evaluate each of the individual association tests at a AIC value (which was 1499.11) than the model fea- significance level of α/n, where n = 6 is the number of turing a two-gene (UQCRC1 and SCD1) network independent tests (i.e., association tests per trait under (which was 1521.87).