Cytogenet Genome Res 102:211–216 (2003) DOI: 10.1159/000075751

Molecular characterization and mutational screening of the PRKAG3 in the horse

H.B. Park,a S. Marklund,a J.T. Jeon,a J.R. Mickelson,b S.J. Valberg,b K. Sandberga and L. Anderssona,c a Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences, Uppsala (Sweden); b University of Minnesota, Veterinary Pathobiology and Clinical and Population Sciences, St. Paul MN (USA); c Department of Medical Biochemistry and Microbiology, Uppsala University, Uppsala (Sweden)

Abstract. The PRKAG3 gene encodes a muscle-specific iso- of the PRKAG3 coding sequence in a small case/control mate- form of the regulatory Á subunit of AMP-activated protein rial of horses affected with polysaccharide storage myopathy (AMPK). A major part of the coding PRKAG3 sequence did not reveal any mutation that was exclusively associated was isolated from horse muscle cDNA using reverse-transcrip- with this muscle storage disease. The breed comparison re- tase (RT)-PCR analysis. Horse-specific primers were used to vealed several potentially interesting SNPs. One of these amplify genomic fragments containing 12 exons. Comparative (Pro258Leu) occurs at a residue that is highly conserved among sequence analysis of horse, pig, mouse, human, Fugu, and AMPK Á . In an SNP screening, the variant allele was only zebrafish was performed to establish the exon/intron organiza- found in horse breeds that can be classified as heavy (Belgian) tion of horse PRKAG3 and to study the homology among dif- or moderately heavy (North Swedish Trotter, Fjord, and Swed- ferent isoforms of AMPK Á genes in vertebrates. The results ish Warmblood) but not in light horse breeds selected for speed showed conclusively that the three different isoforms (Á1, Á2, or racing performance (Standardbred, Thoroughbred, and and Á3) were established already in bony fishes. Seven single Quarter horse) or in ponies (Icelandic horses and Shetland nucleotide polymorphisms (SNPs), five causing amino acid pony). The results will facilitate future studies of the possible substitutions, were identified in a screening across horse breeds functional significance of PRKAG3 polymorphisms in horses. with widely different phenotypes as regards muscle develop- Copyright © 2003 S. Karger AG, Basel ment and intended performance. The screening of a major part

The AMP-activated protein kinase (AMPK) is a metabolic a regulatory Á subunit. In mammals, seven genes encode differ- stress-sensing protein kinase that plays an important role in the ent isoforms of the different subunits (·1, ·2, ß1, ß2, Á1, Á2, regulation of energy homeostasis within the eukaryotic cell and Á3) and these may form 12 different heterotrimeric combi- (Hardie et al., 1998; Kemp et al., 1999). The active is a nations. AMPK is allosterically activated by the increased heterotrimer composed of a catalytic · subunit, a ß subunit and AMP to ATP ratio that occurs during metabolic stress such as nutrient starvation and exercise (Hardie and Carling, 1997; Nielsen et al., 2003). Once activated, AMPK turns on ATP- producing pathways such as oxidation and inhibits Supported by the Swedish Research Council for Environment, Agricultural Sciences, ATP-consuming pathways such as fatty acid synthesis, thereby and Spatial Planning. S.M. was supported by the AgriFunGen program at the Swedish University of Agricultural Sciences. conserving energy homeostasis (Muoio et al., 1999). AMPK is also activated by muscle contraction and this leads to transloca- Received 31 May 2003; manuscript accepted 2 September 2003. tion of glucose transporter 4 (GLUT4) to the plasma mem- Request reprints from Leif Andersson Department of Medical Biochemistry and Microbiology brane, increased glucose uptake, and increased glycogen syn- Uppsala University, Uppsala Biomedical Centre thesis (Holmes et al., 1999). Box 597, SE–751 24 Uppsala (Sweden) Recently, several mutations with phenotype effects in skele- telephone: +46 18 471 4904; fax: +46 18 471 4833 e-mail: [email protected] tal muscle or heart have been identified in the mammalian Á Present address of J.T.J.: Division of Animal Science and Technology PRKAG2 and PRKAG3 genes encoding respectively the 2 Gyengsang National University, Chinju, Gyengnam 660-701 (Korea) and Á3 isoform of the regulatory subunit of AMPK. In mam-

Fax + 41 61 306 12 34 © 2003 S. Karger AG, Basel Accessible online at: ABC E-mail [email protected] 0301–0171/03/1024–0211$19.50/0 www.karger.com/cgr www.karger.com Fig. 1. Schematic diagram of the equine PRKAG3 gene. The exon-intron structure is shown together with the PCR amplicons used for sequencing (F0–F5). Open boxes indicate exons, and lines connecting the boxes indicate introns. Numbers below the open boxes represent the size of exons. Large introns are depicted with a broken line. Black boxes within the depicted PCR amplicons (F0–F5) indicate the corresponding exons. Primer locations are indicated by arrows. Arrows in bold represent sequencing primers. *1 and *2 indicate that the F1 forward and F5 reverse degenerated primers were used to amplify the PRKAG3 cDNA from horse muscle.

Table 1. Primers used for PCR amplification and sequencing of the causes the Wolff-Parkinson-White cardiomyopathy in hu- equine PRKAG3 gene mans. This investigation of the horse (Equus caballus) was ini- Fragmentsa Primer sequencesb (5'→3') Product size (kb) tiated for two reasons. Firstly, we wanted to investigate the pos- sibility that equine polysaccharide storage myopathy (PSSM) is F0 F, CACCATGGAGCCCGAGCTGGAGCA 1.1 caused by a PRKAG3 mutation. PSSM is an inherited myopa- R, CCTGCTGCCCCTGCTCCCATCTC F1 F, AGCATCAAGAGATGAGCTTCCTAGAGCAAG 2.5 thy in Quarter horses that resembles the porcine RN phenotype R, CCCACGAAGCTCTGCTTCTT in that horses have a dramatically high muscle glycogen con- F2 F, CTTCTTTGCCCTGGTGGCCA 1.0 centration (Valberg et al., 1992). All glycogenolytic and glyco- R, GAGACCACAGGCTTGAAGCA F3 F, AGAGGAAGCAGGGGAAGGGTG 0.7 lytic enzyme activities in PSSM muscle are similar to healthy R, TGACCACAGGCAGCGCAGAG horses (Valberg et al., 1998). The increased muscle glycogen in F4 F, CTTCCTTTCCCGCACCATCC 1.2 PSSM horses is associated with enhanced sensitivity of skeletal R, AAGCGAGAGTAGAGGCCCACGA F5 F, GGTGGTGGTGGAGGTGAAAGAG 1.4 muscle to insulin (De La Corte et al., 1999). The mode of inher- R, CCAGCAGGGCTGAGCACCAGTGCCTGAAGG itance for PSSM has not been firmly established but pedigree studies and limited breeding trials indicate a founder stallion a Indicates PCR fragments depicted in Fig. 1. b Primer sequences used for both PCR and sequencing are in bold. and transmission of PSSM to offspring consistent with a reces- sive inheritance (Valberg et al., 1996; De La Corte et al., 2002). A second goal of this study was to determine if the strong selec- tion in draught horses and racing horses may have influenced the population frequency of PRKAG3 alleles in some breeds. mals, the Á3isoform has been found to be primarily expressed The large increase in skeletal muscle glycogen content in the in white (fast-twitch, type IIb) skeletal muscle fibers, in which it RN pig suggests that PRKAG3 mutations may affect muscle is the predominant Á isoform, suggesting a key role for strength and endurance as it is very well established that glyco- PRKAG3 in this tissue (Milan et al., 2000; Mahlapuu et al., gen content is an important factor for resistance to muscle 2003). The Á2 isoform is predominantly expressed in the fatigue. Here we report the sequencing and characterization of human heart but has a more broad tissue distribution in the equine PRKAG3 gene and a screening for functionally rodents (Milan et al., 2000; Mahlapuu et al., 2003). In the pig, a important PRKAG3 mutations among horse breeds with strik- missense mutation Arg200Gln in PRKAG3 produces the dom- ingly different muscle phenotypes. inant RN phenotype characterized by markedly increased gly- cogen storage in skeletal muscle and highly significant effects on meat quality (Milan et al., 2000). A second missense muta- Materials and methods tion in pig PRKAG3, Val224Ile, has been reported to have an Animals and DNA isolation opposite effect on meat quality compared with the Arg200Gln The study included three Quarter horses affected by polysaccharide stor- mutation (Ciobanu et al., 2001). Several mutations in the age myopathy (PSSM) and two clinically healthy controls from the same human PRKAG2 gene have been found to cause cardiomyopa- breed. These samples were obtained from horses diagnosed at the University of Minnesota with PSSM on the basis of abnormal polysaccharide in skeletal thies (Gollob et al., 2001; Hamilton et al., 2001). Interestingly a muscle biopsies stained with periodic acid Schiff (PAS). Samples from eight missense mutation (Arg302Gln), occurring at the correspond- unrelated Standardbred stallions with outstanding racing performance and ing position as the Arg200Gln mutation in pig PRKAG3, five unrelated Standardbred horses with average racing performance were

212 Cytogenet Genome Res 102:211–216 (2003) Table 2. PRKAG3 exon/intron organization a a in horse, human, pig, mouse, and zebrafish Exon Length (bp) Intron Length (bp) Horse Pig Human Mouse Zebrafish Horse Pig Human Mouse Zebrafish

1 n.a 108 33 33 n.a. 1 n.a. 302 362 306 n.a. 2 40b 40 40 40 n.a. 2 >700 478 434 447 n.a. 3 156 156 156 153 n.a. 3 318 360 361 313 n.a. 4 404 404 404 407 108 4 >1500 890 1377 1200 216 5 82 82 82 82 82 5 >400 460 456 114 3225 6 59 59 59 59 59 6 122 101 125 100 2577 7 46 46 46 46 46 7 183 216 203 179 87 8 55 55 55 55 55 8 177 201 201 468 307 9 127 127 127 127 127 9 140 132 154 99 4342 10 166 166 166 166 166 10 >1500 1127 2349 ~1900 97 11 38 38 38 38 38 11 >100 175 170 168 1247 12 147 147 147 147 147 12 355 356 341 330 96 13 n.a. 117 117 117 108

a n.a.: Not available. b Estimated based on the size of the corresponding exon in other mammals.

analysed. Furthermore, genomic DNA samples from horses previously sub- PRKAG3 genes and apparently included a slightly longer sequence corre- jected to paternity testing at the Blood Typing Laboratory in Uppsala, Swe- sponding to exon 12. However, an analysis of the genomic sequence showed den, were used for a breed comparison. The sample included heavily muscled that exon 11 was present and well-conserved in the Fugu genome. We there- draught horses (Belgian), moderately muscled breeds (North Swedish Trot- fore constructed a Fugu PRKAG3 transcript containing the exon 11 ter, Fjord, and Swedish Warmblood), breeds strongly selected for racing per- sequence to facilitate the phylogenetic comparison with the mammalian formance (Thoroughbred and Standardbred) and ponies (Icelandic horse and homologues. Multiple sequence alignment was done with ClustalW (Thomp- Shetland pony). son et al., 1994). A Neighbor-Joining phylogenetic tree, based on genetic dis- tances calculated with Kimura’s two-parameter method, was constructed RT-PCR cloning and genomic sequencing using MEGA 2.1 (Kumar et al., 2001). Needle muscle biopsies were obtained from the gluteus medius muscles of three Quarter Horses affected with PSSM and two healthy controls. The SNP screening tissue was frozen in liquid nitrogen immediately after biopsy. mRNA was The single nucleotide polymorphism (SNP) at codon 258 (Pro258Leu) isolated from the muscle biopsies using the Invitrogen Micro-FastTrack 2.0 was genotyped by pyrosequencing, after PCR amplification of an 182-bp kit. cDNA was prepared using the Invitrogen Superscript II RT kit with ran- genomic fragment containing exon 8 using a biotinylated forward primer (5)- dom hexamers, followed by PCR with degenerated primers designed on the GAGGTGGGACAGTCTGGGGGCT) and a reverse primer (5)-ACTGA- basis of human, mouse, and pig sequences to amplify an equine PRKAG3 AGGGCTGGGGAAGGGACT). The pyrosequencing reaction was carried gene fragment covering 11 exons (Fig. 1 and Table 1). Amplification was out with an internal sequencing primer (5)-GGAGAGATGGAGACCAGA) conducted in 20-Ìl reactions each containing 30 ng cDNA, 0.2 mM dNTPs, according to the manufacturer’s recommendation (Pyrosequencing AB, 1.5 mM MgCl2, 5 pmol of each primer, AmpliTaq Gold DNA polymerase, Uppsala, Sweden). and reaction buffer (PE Applied Biosystems, Foster City, USA). The cycling conditions included an initial incubation at 94 ° C for 5 min followed by 32 cycles comprising 1 min at 94 °C, 1 min at 55°C, and 1 min at 72 °C. PCR Results products were purified using the QIAquick PCR purification kit (Qiagen, Hilden, Germany) and directly sequenced using BigDye Terminator chemis- try (PE Applied Biosystems). Inter-exon PCR amplifications were performed Characterization of the horse PRKAG3 gene and using genomic DNA to determine the PRKAG3 genomic organization as phylogenetic analysis indicated in Fig. 1. Annealing temperatures were in the range 50–65 °C. The One specific 1,281-bp RT-PCR product with an open read- PCR products were sequenced directly, and the exon-intron boundaries were ing frame encoding 427 amino acids was obtained using degen- established by comparing genomic and cDNA sequences. The numbering of erate primers. BLAST searches revealed that horse PRKAG3 codons follows the one reported for pig PRKAG3 (Milan et al., 2000). The sequence data reported in this paper have been deposited in GenBank under has an 87:83%, 88:83%, and 90:88% nucleotide:amino acid accession numbers AY376689 and AY42371–AY423273. identity with the corresponding human, mouse, and pig PRKAG3 sequences, respectively. Four genomic fragments Bioinformatic characterization (F1–F4) containing 11 exons were amplified. A forward primer DNA and predicted protein sequences were analyzed using the Se- quencher program (version 3.0, Gene Codes Corp). The BLAST family of designed from the mouse PRKAG3 cDNA sequence was used programs was used for database searches on the NCBI servers at http:// to obtain an additional genomic fragment (F0) containing the www.ncbi.nlm.nih.go/blast. Cross-species comparison between pig and horse putative translation start codon. The exon/intron organization genomic sequences was done with the Alfresco program (Jareborg and Dur- was determined by comparing genomic and cDNA sequences bin, 2000). The results were confirmed by a comparative genomic sequence ) analysis of horse, pig, mouse, human, and zebrafish sequences. For the phylo- (Fig. 1; Table 2). All splice sites corresponded to the 5 donor genetic analysis, AMPK Á nucleotide sequences corresponding to exons 4–13 (GT) and 3) acceptor (AG) consensus sequences (Breathnach in the Á3 gene, were retrieved from GenBank or the Ensembl databases. and Chambon, 1981). However, the zebrafish AMPK Á2 sequence was retrieved from the TIGR PRKAG3 homologues have previously been characterized zebrafish Gene Index database (http://www.tigr.org/tdb/tgi/zgi/). The Fugu in several mammalian species. A bioinformatic search revealed AMPK Á3 sequence was retrieved from the IMCB database (http://scrap- py.fugu-sg.org/Fugu_rubripes/) and only included eight exons (exons 6–13). clear evidence for the presence of PRKAG3 homologues also in The Fugu cDNA sequence lacked 38 bp encoded by exon 11 in mammalian two fish species, the zebrafish (ENSDART00000000381) and

Cytogenet Genome Res 102:211–216 (2003) 213 the pufferfish (Fugu; SINFRUT00000067047). The overall ge- No obvious association between PRKAG3 polymorphism nomic organization was well conserved among mammalian and polysaccharide storage myopathy (PSSM) and fish species (Table 2). A multiple alignment including our The complete PRKAG3 coding sequence, except the first 13 partial horse PRKAG3 sequence and AMPK Á nucleotide and the last 24 codons, was determined by RT-PCR analysis of sequences from other species was used to construct a Neighbor- muscle samples from three Quarter horses affected by PSSM Joining phylogenetic tree (Fig. 2). The AMPK Á1, Á2, and Á3 and two non-affected controls from the same breed (Table 3). isoforms formed distinct clusters and the data provided conclu- We detected four SNPs, two of which were non-synonymous sive evidence that the horse PRKAG3 homologue has been substitutions, but none showed a complete association with identified in this study. PSSM. Since we have screened almost the entire coding sequence (428 out of 465 codons) and since the affected horses did not share any specific PRKAG3 haplotype we can conclude that it is highly unlikely that PSSM is caused by a single muta- tion affecting the PRKAG3 coding sequence. However, we can- not exclude the possibility that this disease is caused by a PRKAG3 mutation occurring in an untranslated region or in a regulatory element.

Genetic variation at the PRKAG3 locus across divergent horse breeds To test the possibility that the strong directional selection for muscle strength or racing performance in certain breeds has influenced the PRKAG3 allele frequency distribution across breeds we decided to determine almost the entire coding Fig. 2. Phylogenetic tree constructed for AMPK Á nucleotide sequences corresponding to exon 6–13 in the mammalian PRKAG3 gene. Numbers at sequence from a sample of horses representing different breeds. the nodes represent the bootstrap support values derived from 1,000 repli- Initially we sequenced eight Standardbred stallions with out- cates. The scale indicates the genetic distance. The accession numbers for the standing racing performance and five controls with average rac- sequences used are as follows: human (Hu) PRKAG1, NM_002733; mouse ing performance. Six SNPs were detected, four non-synony- (Mu) PRKAG1, NM_016781; human PRKAG2, AF087875; mouse PRKAG2, NM_145401; human PRKAG3, AF214519; mouse PRKAG3, mous and two synonymous substitutions (Table 4). There was NM_153744; pig PRKAG3, AF214520; horse PRKAG3, AY376689; an indication of an allele frequency difference between stal- zebrafish (Da) PRKAG2, TC143275; zebrafish PRKAG3, ENS- lions and controls for the SNPs at codons 26 and 51, but the DART00000000381; Fugu (Fu) PRKAG1, SINFRUT00000162627; difference was not significant in this limited sample. The Fugu PRKAG2, SINFRUT00000165144; Fugu PRKAG3, SINFRUT Asn362His mutation that was found in a single breeding stal- 00000067047; Drosophila (Dro) PRKAG, AF094764. All sequences are from GenBank except the fish sequences that were obtained from the lion is also potentially interesting, since Asn362 is highly con- ENSEMBL and TIGR Gene Index databases. served among mammalian PRKAG3 sequences.

Table 3. Comparison of PRKAG3 cDNA sequences of horses affected Table 4. PRKAG3 nucleotide substitutions identified among 22 horses with polysaccharide storage myopathy (PSSM) and non-affected controls. A representing six breedsa and observed allele frequencies among breeding dash indicates identity to the master sequencea stallions (n = 8) and controls (n = 5) of Standardbred horses

Horse Codon (Exon) Exon/nucleotide Nucleic acid Amino acid Allele frequencyc positionb change substitution 26(3) 49(3) 271(9) 291(9) Breeding Controls stallions GGA CCG GCC GGC G P A G 3 / 76 GA G26R 0.31 0.10 3 / 146 CT P49L 0 0.10 Affected horses 3 / 151 GA E51K 0.25 0.10 1 R-- -Y------8 / 773 CT P258L 0 0 R/G P/L - - 9 / 813 - 0.44 0.60 2 --- -Y------CA - P/L - - 9 / 873 TC - 0.12 0.10 3 ------A --T 10 / 1084 AC N362H 0.06 0

- - - - a Non-affected horses Standardbred, Thoroughbred, North Swedish Trotter, Belgian, Shetland pony, 4 ------and Quarter horse. b Nucleotide numbers counted from the translation start codon located in exon 3. - - - - c 5 R------The allele frequencies refer to the allele indicated as the variant allele. R/G - - -

a Nucleotide R=A/G, Y=C/T.

214 Cytogenet Genome Res 102:211–216 (2003) We then sequenced the PRKAG3 gene from one horse from KIAA0173 gene that is located in the near vicinity of PRKAG3 each of the following breeds: Belgian, North Swedish Trotter, in both mammals and the two fish species. Shetland pony, and Thoroughbred. The seven SNPs detected We identified seven equine PRKAG3 SNPs, five of which across breeds are given in Table 4. All SNPs were confirmed by cause amino acid substitutions. The limited number of horses sequencing both strands across the SNP and all SNPs except screened in this study did not indicate a complete association the one at exon 10, nt1084 were found in more than one indi- between any SNP and equine polysaccharide storage myopathy vidual. The non-synonymous substitution at codon 258 that occurs in Quarter horses. Further studies are required to (Pro258Leu) found in a Belgian and a North Swedish Trotter reveal the genetic basis for this disease. A major step forward was particularly interesting. This substitution occurs at residue would be to establish a map localization to exclude the majority 5 in the second cystathionine ß-synthase domain (CBS2) of the of the large number of potential candidate genes, in addition to AMPK Á3 chain. Proline at this residue is conserved among all PRKAG3, that may influence glycogen content in skeletal mus- mammalian AMPK Á isoforms and also in a Drosophila homo- cle. logue. We therefore decided to further investigate the allele fre- The comparative sequence analysis across nine different quency distribution of this SNP by genotyping 111 horses horse breeds revealed several potentially interesting mutations. representing nine different breeds (Table 5). The SNP screen- An SNP screening comprising 111 horses revealed the presence ing revealed a putative association between the presence of the of the mutant allele (Leu258) in breeds that can be classified as Leu258 allele and muscle development since it was only found heavily muscled (Belgian) or moderately heavy (North Swedish in the heavy (Belgian) and moderately heavy (North-Swedish Trotter, Fjord, and Swedish Warmblood) but it was not found Trotter, Swedish Warmblood, and Fjord) horses but not among in breeds selected for speed or racing performance (Stan- 29 horses representing three light breeds strongly selected for dardbred, Thoroughbred, or Quarter horse) or in ponies (Ice- speed and racing performance (Standardbred, Thoroughbred, landic horse and Shetland pony). The fact that Pro258 is evolu- and Quarter horse) nor among the pony breeds included in this tionarily very well conserved implies that this may be a func- study (Icelandic horse and Shetland pony). tional SNP. This study will facilitate future studies of possible associations between PRKAG3 polymorphism and muscle de- velopment/function in the horse. It will be of particular interest Discussion to measure glycogen contents in horses with different genotypes since all functionally important AMPK mutations detected so In this study, we have reported the cDNA and correspond- far are associated with altered glycogen content in skeletal mus- ing genomic sequence for the equine PRKAG3 gene. The RT- cle (PRKAG3 mutations) or in heart (PRKAG2 mutations). PCR analysis documented the expression of PRKAG3 in horse skeletal muscle as expected from previous studies in pig and human (Milan et al., 2000). A phylogenetic analysis including Acknowledgements horse, pig, mouse, human, pufferfish, and zebrafish sequences Á We thank Valérie Amarger and Erik Bongcam-Rudloff for valuable sug- showed conclusively that the horse AMPK sequence reported gestions on the genetic and bioinformatic analysis. in this study is a Á3 homologue. The result suggested that the genes for the three different mammalian isoforms of the AMPK Á chains evolved by gene duplications from a common ances- tral gene subsequent to the divergence from an invertebrate ancestor but before the divergence of bony fishes and mammals (Fig. 2). The PRKAG3 homologues in zebrafish and in puffer- fish identified in this study have not yet been correctly anno- tated but our interpretation of homology is strongly supported by the presence of conserved synteny involving for instance the

Table 5. Frequency of the PRKAG3 Leu258 allele among nine horse breeds

Breed n Leu258

Belgian 21 0.14 North Swedish Trotter 20 0.27 Fjord 10 0.20 Swedish Warmblood 10 0.10 Icelandic horse 10 0 Shetland pony 11 0 Thoroughbred 11 0 Standardbred 13 0 Quarter horse 5 0

Cytogenet Genome Res 102:211–216 (2003) 215 References

Breathnach R, Chambon P: Organization and expres- Hardie DG, Carling D, Carlson M: The AMP-acti- Muoio DM, Seefeld K, Witters LA, Coleman RA: sion of eucaryotic split genes coding for proteins. A vated/SNF1 protein kinase subfamily: metabolic AMP-activated kinase reciprocally regulates tria- Rev Biochem 50:349–383 (1981). sensors of the eukaryotic cell? A Rev Biochem cylglycerol synthesis and fatty acid oxidation in liv- Ciobanu D, Bastiaansen J, Malek M, Helm J, Woollard 67:821–855 (1998). er and muscle: evidence that sn-glycerol-3-phos- J, Plastow G, Rothschild M: Evidence for new Holmes BF, Kurth-Kraczek EJ, Winder WW: Chronic phate acyltransferase is a novel target. Biochem J alleles in the protein kinase adenosine monophos- activation of 5)-AMP-activated protein kinase in- 338:783–791 (1999). phate-activated gamma(3)-subunit gene associated creases GLUT-4, hexokinase, and glycogen in mus- Nielsen JN, Mustard KJ, Graham DA, Yu H, MacDon- with low glycogen content in pig skeletal muscle cle. J Appl Physiol 87:1990–1995 (1999). ald CS, Pilegaard H, Goodyear LJ, Hardie DG, and improved meat quality. Genetics 159:1151– Jareborg N, Durbin R: Alfresco – a workbench for com- Richter EA, Wojtaszewski JF: 5)-AMP-activated 1162 (2001). parative genomic sequence analysis. Genome Res protein kinase activity and subunit expression in De La Corte FD, Valberg SJ, Williamson S, MacLeay 10:1148–1157 (2000). exercise-trained human skeletal muscle. J Appl JM, Mickelson JR: Glucose uptake in horses with Kemp BE, Mitchelhill KI, Stapleton D, Michell BJ, Physiol 94:631–641 (2003). polysaccharide storage myopathy (PSSM). Am J Chen ZP, Witters LA: Dealing with energy de- Thompson JD, Higgins DG, Gibson TJ: CLUSTAL W: Vet Res 60:458–462 (1999). mand: the AMP-activated protein kinase. Trends Improving the sensitivity of progressive multiple De La Corte FD, Valberg SJ, MacLeay JM, Mickelson Biochem Sci 24:22–25 (1999). sequence alignment through sequence weighting, JR: Developmental onset of polysaccharide storage Kumar S, Tamura K, Jakobsen IB, Nei M: MEGA2 – position-specific gap penalties and weight matrix myopathy in 4 Quarter Horse foals. J Vet Int Med Molecular Evolutionary Genetics Analysis soft- choice. Nucl Acids Res 22:4673–4680 (1994). 16:581–587 (2002). ware, Arizona State University, Tempe, Arizona, Valberg S, Cardinet III, GH, Carlson GP, DiMauro S: Gollob MH, Green MS, Tang AS, Gollob T, Karibe A, USA (2001). Polysaccharide storage myopathy associated with Ali Hassan AS, Ahmad F, Lozado R, Shah G, Mahlapuu M, Johansson C, Lindgren K, Hjalm G, exertional rhabdomyolysis in the horse. Neuro- Fananapazir L, Bachinski LL, Roberts R, Hassan Barnes B, Krook A, Zierath J, Andersson L, Mark- musc Disorders 2:351–359 (1992). AS: Identification of a gene responsible for familial lund S: Expression profiles representing the Á-sub- Valberg SJ, Geyer C, Sorum SA, Cardinet III GH: Wolff-Parkinson-White syndrome. New Engl J unit isoforms of AMP-activated protein kinase sug- Familial basis for exertional rhabdomyolysis in Med 344:1823–1831 (2001). gest a major role for Á3 in white skeletal muscle Quarter horse-related breeds. Am J Vet Res Hamilton SR, Stapleton D, O’Donnell JB Jr, Kung JT, fibers of mammals. Am J Physiol Endocrin Metab, 57:286–290 (1996). Dalal SR, Kemp BE, Witters LA: An activating in press (2003). Valberg SJ, Townsend D, Mickelson JR: Skeletal mus- mutation in the gamma1 subunit of the AMP-acti- Milan D, Jeon JT, Looft C, Amarger V, Robic A, The- cle glycolytic capacity and phosphofructokinase vated protein kinase. FEBS Lett 500:163–168 lander M, Rogel-Gaillard C, Paul S, Iannuccelli N, regulation in horses with polysaccharide storage (2001). Rask L, Ronne H, Lundstrom K, Reinsch N, Gel- myopathy. Am J Vet Res 59:782–785 (1998). Hardie DG, Carling D: The AMP-activated protein lin J, Kalm E, Roy PL, Chardon P, Andersson L: A kinase-fuel gauge of the mammalian cell? Eur J mutation in PRKAG3 associated with excess gly- Biochem 246:259–273 (1997). cogen content in pig skeletal muscle. Science 288:1248–1251 (2000).

216 Cytogenet Genome Res 102:211–216 (2003)