SHORT COMMUNICATION doi:10.1111/j.1365-2052.2004.01156.x Ovine alpha- : isolation, linkage mapping and association analysis with milk traits

J. H. Calvo*, S. Marcos*, A. E. Beattie†, C. Gonzalez*, J. J. Jurado* and M. Serrano* *Departamento de Mejora ´ tica Animal, INIA, 28040 Madrid, Spain. †AgResearch Molecular Biology Unit, Department of Biochemistry and Centre for Gene Research, University of Ottago, Dunedin, New Zealand

Summary On the basis of comparisons between cattle and sheep genome mapping information the ovine a-amylase gene was examined as a possible genetic marker for milk traits in sheep. The objective of the present study was to isolate, map and determine whether this gene is a candidate gene for milk traits. DNA fragments (832 and 2360 bp) corresponding to two different AMY genes were isolated, and one SNP in intron 3 and one GTG deletion in exon 3 of the 2360 bp DNA fragment were found. The 2360 bp ovine AMY DNA fragment was located on 1 by linkage mapping using the International Mapping Flock. No association was found between estimated breeding values for milk yield, and fat contents and AMY genotypes in a daughter design comprising 13 Manchega families with an average of 29 daughters (12–62) per sire.

Keywords amylase gene, association, linkage mapping, sheep.

a-Amylase belongs to a family of that hydrolyses in Table 1. Genomic DNA (100 ng) was amplified in a final the a1-4 bonds in glucose polymers. In humans, a- volume of 25 ll containing 5 pmol of each primer, 200 lM are encoded by a multigene family, clustered within a dNTPs, 2.0 mM MgCl2,50mM KCl, 10 mM Tris-HCl, 0.1% 250 kb region, containing five active genes, including two Triton X-100 and 0.75 U Taq polymerase (EcoTaq Plus, pancreatic genes (AMY2A and AMY2B) and three salivary Ecogen, Madrid, Spain). After denaturation at 94 C for genes (AMY1A, AMY1B and AMY1C) (Gumucio et al. 2 min, 30 amplification cycles were performed comprising 1988). a-Amylase genes have been mapped to bovine denaturation at 94 C for 1 min, annealing temperature chromosome 3 (Laurent et al. 1999), human chromosome according with the primer pair for 1 min, extension at 1 (Dracopoli & Meisler 1990), and caprine chromosome 13 72 C for the time considering the length of the expected (Schibler et al. 1998). fragment (1 min for every kb), followed by a further 5 min Quantitative trait loci (QTL) for some milk traits have extension at 72 C. The PCR products were cloned in been reported on bovine chromosome 3 (BTA3) close to the pMOSBlue (Amersham Pharmacia, Little Chalfont, UK) and AMY cluster (Lipkin et al. 1998; Zhang et al. 1998). sequenced using an ABI PRISM 3700 (Applied Biosystem, Therefore, we examined the ovine AMY genes as potential Madrid, Spain) and standard protocols. genetic markers for milk traits in sheep. Specific sheep primers were designed for each DNA frag- A comparative walking strategy using DNA sequence ment (Table 1, PCR 4, 5), and direct sequencing of PCR from bovine (EMBL: AF054834) and human genomic DNA products from six different animals (two animals of (EMBL: NT_030567) was used to obtain the partial genomic Manchega, one animal from each of the Awassi, Assaf and DNA sequences of 832 bp (Table 1, PCR 2, 3) and 2360 bp Rasa Aragonesa sheep breeds and one Mouflon) was per- (Table 1, PCR 1) fragments corresponding to two different formed using an ABI PRISM 3700 automatic sequencer ovine AMY genes. Primer sequences, the annealing tem- (Applied Biosystem) and standard protocols. peratures and the size amplified of the fragments are shown The exons were identified in the 832 and 2360 bp sheep AMY DNA fragments by homology with known AMY Address for correspondence sequences of cattle, swine, rat and humans, using BLAST

Unidad de Tecnologia en Produccion Animal, Centro de Investigacion y (National Center for Biotechnology Information: http:// Tecnologia Agroalimentaria de Aragon, Apartado 727, 50080 www.ncbi.nlm.nih.gov/BLAST/), DNASIS (Hitachi Software Zaragoza, Spain. Engineering, Ltd, 1991), CLUSTALW (http://www.ebi.ac.uk/ E-mail: [email protected] clustalw/) and GENEVIEWER (http://www.itba.mi.cnr.it/ Accepted for publication 3 May 2004 webgene/) software. The isolated 832 bp AMY DNA

2004 International Society for Animal Genetics, Animal Genetics, 35, 329–332 329 330 Calvo et al.

Table 1 Primer sequences and GenBank reference information.

PCR Primer sequence (5¢fi3¢) Site GenBank Annealing temperature (C) Size (kb)

1 GATATTGCTCTTGAATGTG Exon 2 NT_030567 (human) 50 GGTTTACTTCCTGAAGGAA Exon 5 AF054834 (cattle) 2.399 2 GTTCGTATTTATGTGGAT Exon 4 AF054834 (cattle) 50 TCCTGGTAAATGAAAGGTTTAC Exon 5 AF054834 (cattle) 0.809 3 TCTACCTCTCTGTAAGCC Intron 3 AY396045 (sheep) 53 GCAAGATCAAGAAGACTA Exon 5 AF054834 (cattle) 0.659 4 GTCCGCATTTATGTGGATGC Exon 4 AY396044 (sheep) 52 TCCTGGTAAATGAAAGGTTTAC Exon 5 AF054834 (cattle) 0.809 5 CTTAGCCCCAAGGATTTGG Intron 2 AY396045 (sheep) 51 CCAGCTTGTGTTTAGATTATG Exon 5 AY396045 (sheep) 2.360 6 TCAGTGATATAAAGTAGGCC Intron 2 AY396045 (sheep) 50 CTGCTTAAAAGGAACAAGAG Intron 3 AY396045 (sheep) 0.282

fragment contains exons 4 (198 bp), 5 (231 bp) and intron denaturation step were used. The SSCP analysis was per- 3 (22 bp), intron 4 (381 bp). The 2360 bp AMY DNA formed in a 7% polyacrylamide gel [acrylamide-bisacryla- fragment contains exons 3 (147 bp), 4 (198 bp), 5 (195 bp) mide (49:1), and 8.7% glycerol]. Constant running and introns 2 (361 bp), 3 (1070 bp) and 4 (389 bp). The conditions were as follows: 8 W, 20 C for 12 h. The DNA sequence data have been submitted to GenBank (EMBL: bands were visualized by silver staining. The allele with the AY396044 and AY396045). GTG deletion was called B. The two ovine AMY DNA sequences were aligned, Genotype distributions within breeds were tested for showing that they correspond to two different AMY genes. Hardy–Weinberg equilibrium using GENEPOP version 3.1 Both sequences were compared against the databases, and (Raymond & Rousset 1995). Estimates of heterozygosity the human contig, which contains the AMY gene cluster were derived using the BIOSYS-1 computer program (Swof- (NT_030567). The exon sizes were similar to those of ford & Selander 1989). The GTG deletion was present in all human. Maximum identity was found for bovine AMY of the domestic breeds studied, the level of polymorphism sequence (832 bp AMY DNA sequence), and human was similar among the breeds (data not shown). Further- AMY2B (2360 bp AMY DNA sequence), but it cannot be more, in the Manchega varieties, the genotype frequencies concluded which human amylase is homologous to the are in disequilibrium, which may be due to selection for ovine AMY sequences described here. Sequence analysis of milk production in Manchega white variety (P < 0.01). In the introns using REPEATMASKER software (http://ftp.genome. Manchega black variety (P < 0.05), this situation may be washington.edu/cgi-bin/RepeatMasker) revealed a repetit- due to a small number of founders. ive element in intron 3 (SINEbov elements) of the 2360 bp The deletion in exon 3 of the 2360 bp AMY DNA frag- AMY DNA sequence. ment was also used for linkage mapping and association Direct sequencing of PCR products amplified from DNA studies. Linkage mapping was used to map the 2360 bp samples of Manchega, Awassi, Assaf, Rasa aragonesa sheep AMY DNA fragment against the markers on the latest and Mouflon, for the 832, and 2360 bp AMY DNA sequences version of the sheep framework map (Maddox et al. 2001). revealed one SNP (T/C) at position 633 in intron 3, and one Multipoint linkage analysis of the International Mapping GTG deletion at position 386 in exon 3 in the 2360 bp AMY Flock (IMF) pedigrees (Crawford et al. 1995) using CRI-MAP DNA fragment. The polymorphism in exon 3 results in an (Lander & Green 1987) localized the 2360 bp AMY DNA amino acid deletion in the predicted protein. The 832 bp fragment to OAR1 (support for multipoint linkage to the AMY DNA fragment did not show any polymorphism. group is LOD of 4.1) (Fig. 1). Sheep AMY was located on The deletion in exon 3 of the 2360 bp AMY DNA frag- by 2-point linkage mapping using the IMF ment was analysed using single-strand conformation poly- pedigrees, with linkage being found to AE57 (h ¼ 0.16, morphism (SSCP) technique in four domestic sheep breeds LOD ¼ 3.61). This map assignment is consistent with (Ovis aries) and in the European Mouflon (O. smelini musi- comparative mapping information on bovine and goat. mon). Forty-seven, 144, 18, 44, 44 and 16 animals were A population of Manchega individuals structured as a analysed for Manchega black variety, Manchega white daughter design comprising 13 families, with an average of variety, Rasa Aragonesa, Assaf, Awassi, and Mouflon, 29 daughters (12–62) per sire, was evaluated for associa- respectively. The set of primers used for PCR amplification tions between AMY genotypes and performance. All the was chosen from the 2360 bp AMY DNA fragment animals belonged to the selection nucleus of the Spanish (Table 1, PCR 6). The PCR conditions were as before except Manchega sheep-breeding scheme and were sampled from that only 50 ng of DNA, 0.2 U Taq polymerase and 30 s flocks connected by artificial insemination. Estimated

2004 International Society for Animal Genetics, Animal Genetics, 35, 329–332 Ovine alpha-amylase genes 331

Table 2 Association analysis between estimated breeding values (EBVs) of milk traits and genotypes for the GTG deletion of the 2360 bp ovine AMY DNA fragment, within and across five families of Manchega sheep breed. INRA130 S29 Average EBVs Statistics Trait AA BB F statistic P-value INRA003 AMY Family (n ¼ 15) Milk yield 13.09 (5.57)1 2.42 (8.23) 7.44 0.0172* BM4129 Fat percentage 0.08 (0.12) 0.008 (0.12) 0.98 0.3402 Protein percentage 0.21 (0.09) 0.209 (0.11) 0.01 0.9262 MCM58 Across families (n ¼ 89) BL41 Milk yield 19.04 (9.44) 20.79 (11.68) 0.61 0.4354 10 cM AE57 Fat percentage 0.04 (0.17) 0.085 (0.21) 1.14 0.2886 Protein percentage 0.02 (0.18) 0.048 (0.15) 0.58 0.4465

*Difference of mean values significant at P < 0.05. 1Standard error values in brackets. BMS482

homozygotes). Further studies are needed to evaluate this possible association more effectively, as these data may be Sheep chromosome 1 too preliminary to allow us to distinguish between the possible explanations: false positive associations, causal Figure 1 Linkage map position of the 2360 bp ovine AMY DNA effects of AMY and effects of linked genes. fragment on sheep chromosome 1. The 2360 bp ovine AMY DNA fragment maps between BM4129 and INRA003, at 18.4 cM of AE57. Acknowledgements breeding values (EBVs) of milk yield and fat and protein This work has been supported by INIA project SC00-053. percentages were used in single/locus association analysis. Authors thank CENSYRA-Valdepen˜as, CSIC-Leo´n, SIA- The EBVs of traits come from the 2003 BLUP evaluation of Arago´n, and INIA for kindly providing the Manchega, animals of this breed. Awassi, Assaf, Rasa Aragonesa and Mouflon breed animals. Statistical analysis of associations between genotypes Also thank Dr Ken G. Dodds for suggestions and English for the GTG deletion and genetic values was performed correction of the manuscript. across and within half-sib families in which the sire was heterozygous, using the GLM-procedure of SAS (1998; version 6.12, SAS Institute Inc., Cary, NC, USA). The References comparison was restricted to homozygous daughters (AA Crawford A.M., Dodds K.G., Ede A.J. et al. (1995) An autosomal vs. BB). AB genotypes were not considered as we could genetic linkage map of the sheep genome. Genetics 14, 703– not determine which allele had been received from the 24. sire. Dracopoli N.C. & Meisler M.H. (1990) Mapping the human amylase The model applied was: gene cluster on the proximal short arm of chromosome 1 using a highly informative (CA)n repeat. Genomics 7, 97–102. Yij ¼ l þ Fi þ Gj þðF GÞij þ eij Gumucio D.L., Wiebauer K., Caldwell R.M., Samuelson L.C. & where, Yij ¼ estimated breeding value of animal j within Meisler M.H. (1988) Concerted evolution of human amylase family i, l ¼ population mean, F ¼ family I, G ¼ genotype i i genes. Molecular Cell Biology 8, 1197–205. of animal j (AA, AB, BB)+(F· G) ¼ interaction effect ij Lander E.S. & Green P. (1987) Construction of multilocus genetic between family and genotype, and eij ¼ residual error. The linkage maps in humans. Proceedings of the National Academic effect of family was removed from the analysis when the Sciences of the United States of America 84, 2363–7. within family study was done. Laurent P., Hayes H., Eggen A., Bishop M. & Leveziel H. (1999) Assignment of the amylase gene cluster to bovine chromosome 3 The GTG deletion was tested in the five half-sib families bands 3q23 fi q24 by in situ hybridization. Cytogenetics and Cell with heterozygous sires. No association was found between Genetics 85, 315–6. the EBVs and the AA and BB genotypes in the whole data. Lipkin E., Mosig M.O., Darvasi A., Ezra E., Shalom A., Friedmann A. In within-family analysis, one family showed a significant & Soller M. (1998) Quantitative trait locus mapping in dairy effect (P < 0.05) for milk yield (Table 2). This association cattle by means of selective milk DNA pooling using dinucleotide may be spurious as it relies upon an unbalanced and small microsatellite markers: analysis of milk protein percentage. number of individuals (12 AA homozygotes vs. 3 BB Genetics 149, 1557–67.

2004 International Society for Animal Genetics, Animal Genetics, 35, 329–332 332 Calvo et al.

Maddox J.F., Davies K.P., Crawford A.M. et al. (2001) An enhanced Swofford D.L. & Selander R.B. (1989) BIOSYS-1. A Computer Pro- linkage map of the sheep genome comprising more than 1000 gram for the Analysis of Allelic Variation in Population Genetics and loci. Genome Research 11, 1275–89. Biochemical Systematics (Release 1.7). University of Illinois, Urb- Raymond M. & Rousset F. (1995) GENEPOP version 3.1b. Popu- ana, Champaign. lation genetics software for exact tests and ecumenicism. Journal Zhang Q., Boichard D., Hoeschele I. et al. (1998) Mapping quanti- of Heredity 86, 248–9. tative trait loci for milk production and health of dairy cattle in a Schibler L., Vaiman D., Oustry A., Giraud-Delville C. & Cribiu E.P. large outbred pedigree. Genetics 149, 1959–73. (1998) Comparative gene mapping: a fine-scale survey of chro- mosome rearrangements between ruminant and humans. Gen- ome Research 8, 901–15.

2004 International Society for Animal Genetics, Animal Genetics, 35, 329–332