Special barley B-amylase allele in a Finnish landrace line HA52 with high grain enzyme activity

He edi as 134: 91-95 (2001) ~~~ B ie epo Special ba ley P-amylase allele in a Finnish land ace line HA52 wi h high g ain enzyme ac i i y MARIA J. ERKKILA and HANNU AHOKAS Plan P oduc ion Resea ch, Ag icul u al Resea ch Cen e, Mylly ie 10, FIN-31 600 Jokioinen, Finland E-mail: [email p o ec ed] (Recei ed Feb ua y 19, 200 1. Accep ed May 2 1, 200 1) Ba ley (Ho deum uulga e L.) g ain mainly consis s o s a ch, which p o ides ene gy du ing ge mina ion and seedling g ow h ( o e iew, see MACGREGOR and FINCHER 1993). S a ch deg ada ion equi es con- ce ed ac ion o limi dex inase, P-amylase, a-glu- cosidase (SUN and HENSON 1990) and a-amylase. Libe a ion o mal ose and limi dex ins om he non educing ends o s a ch is ca alysed by P-amylase (1,Ca-D-glucan mal ohyd olase, EC 3.2.1.2) (ROBYT and WHELAN 1968; SOPANEN and LAURIERE 1989). Being syn hesised du ing g ain de elopmen (KREIS e al. 1987), P-amylase is one o he majo p o eins ound in he s a chy endospe m (HEJGAARD and BOISEN 1980). The endospe m P-amylase gene (P- amyl) is loca ed in ch omosome 4H (KREIS e al. 1987). Ano he P-amylase gene (P-amy2), called ubiq- ui ous, is loca ed in ch omosome 2H and he p o ein is ound in lea es and oo s (KREIS e al. 1988; SHARP e al. 1988). Mode n plan b eeding has educed he gene ic a iabili y in domes ica ed ba ley (THOMPSON e al. 1990; FORSTER e al. 1991) and declined land aces in Finland p io o abou 1950 (AHOKAS 2000). La ge gene ic di e si y is ound in wild ba ley H. ulga e ssp. spon aneum (K. Koch) A. & G ., abb e ia ed as H. spon aneum (AHOKAS 1982; ZHANG e al. 1993; SAGHAI MAROOF e al. 1995). High P-amylase ai was inhe i ed in he backc ossed p ogeny o H. spon- aneum and domes ica ed ba ley (AHOKAS and ERKKILA 1992). Howe e , H. spon aneum was no widely used in b eeding o se e al easons such as sha e ing o spikele s, ad e se g ow h y hm a high la i udes, and appa en low g ain yield. Because land aces a e be e adap ed o local en i- onmen s and a e mo phologically close o desi ed domes ica ed ypes han wild ba ley, hey a e a mo e sui able sou ce o gene ic a ia ion o ba ley b eed- ing. Compa ed wi h he a ia ion in wild ba ley g own in Finland (AHOKAS and NASKALI 1990), land aces had signi ican ly highe mean ac i i ies o a- and P-amylase and o P-glucanase (AHOKAS and POUKKULA 1999). The line HA52 was selec ed om Finnish land aces (AHOKAS 1977), a di e se gene ic esou ce (AHOKAS and MANNINEN 2001). La e , HA52 was ound o ha e a high P-amylase ac i i y (AHOKAS e al. 1996) and also a high he mos abili y (AHOKAS and MANNINEN 2000). The esul p esen ed he e is an ex ension p e ious s udies o P-amyl alleles ound in ba ley (ERKKILA e al. 1998; ERKKILA 1999). YOSHIGI e al. (1995) de- sc ibed an allele in c . Ha una Nijo (EMBL-Gen- Bank da abase accession numbe D49999). We ha e p e iously ound wo P-amyl alleles, one in c . Ado a and ano he in H. spon aneum s ain PI 296897 (EMBL-GenBank da abase accession num- be s AF061203 and AF061204, ERKKILA e al. 1998). All he alleles di e om each o he h ough hei nucleo ide sequences especially in in on 111, bu no signi ican ly in he open eading ame (ERKKILA e al. 1998). Fu he s udies showed ha se e al ba ley cul i a s, lines and wild s ains had ei he a c . Ado a-like, c . Ha una Nijo-like o H. spon aneum PI 296897-like P-amyl allele (ERKKILA 1999). To s udy he P-amyl locus in he Finnish land ace ba ley line ha ing high enzyme ac i i y, he gene o he line HA52 was sequenced and compa ed wi h p e iously ound alleles. The P-amyl sequence in HA52 - The locus o 0-amyl in he Finnish land ace line HA52 was PCR ampli ied om genomic DNA using p ime s based on he sequence o he c . Ha una Nijo (YOSHIGI e al. 1995). DNA sequencing was pe o med on ALF DNA Sequence (Pha macia-LKB) and sequences we e analysed wi h he PC/Gene so wa e package (In elliGene ics). The sequence o he HA52 P-amyl gene ob ained was 4951 bp in leng h and included a p omo e , se en exons and six in ons, which is con- sis en wi h he schema ic s uc u e o all he alleles sequenced be o e (Fig. 1, YOSHIGI e al. 1995; ERKKILA e al. 1998). The P-amyl sequence o HA52 has an EMBL accession numbe AJ301645. T an- sc ip ion ini ia ion si e is si ua ed a posi ion 1205 bp om he beginning o he clone. Some p omo e elemen s, such as TATA and CCAAT boxes, a e common o many genes an- sc ibed by polyme ase 11. In he HA52 P-amyl a 92 M. J. E kkila and H. Ahokas He edi as 134 (2001) Fig. 1. Schema ic p esen a ion o P-amy I s uc u al gene. Se en exons a e ma ked wi h black boxes, and six in ons wi h whi e boxes (nume als I-VI). G ay box, ATG and TAG indica e p omo e , ansla ional s a and s op codons, espec i ely. The HASZspeci ic dele ions and inse - ions a e ma ked in in ons I1 and 111. A segmen o p omo e is enla ged showing he ela i e posi ions o ATG ansla ion s a si e, ansc ip ion ini ia ion si e, TATA box, I-box, wo GGTTT mo i es and HA52-speci ic 92-bp dele ion. TATA box is a - 31 bp, as can be expec ed, since in plan s, i is no mally be ween - 29 bp and - 33 bp (MESSING e al. 1983). T ansla ion s a s a posi ion + 55 bp (Fig. 1) and has he same mo i e as common consensus sequence in plan s, CCACCATG (KOZACK 1984). In ons and inse ions - Compa ed wi h he alleles o c . Ha una Nijo, c . Ado a, and H. spon aneum PI 296897 (YOSHIGI e al. 1995; ERKKILA e al. 1998), he HA52 P-amy l allele had se e al subs i u ions, and addi ionally, bo h single-base and sho mul iple-base dele ions and inse ions (Table 1). Majo sequence di e ences be ween he ou P-amy 1 alleles we e in he p omo e egion, and in he in ons I1 and I11 (Fig. 1). The e we e sho inse ions o 3 bp and o 10 bp close o each o he in he in on I1 o HA52 (Fig. 2B). The 10-bp inse ion con ains a epea , 5'- ATATTTA-3', which is u he mo e ound once o wice in he in ons I11 o he ba leys s udied so a . The in on I1 inse ion in he HA52 P-amyl is 10 bp long and con ains he ATATTTA epea (Fig. 2B) and is also ound wice in i s in on 111, hence sug- ges ing a con e sion a eplica ion. GNIADKOWSKI e al. (1 996) sugges ed ha U- ich sequences s imula e splicing in RNA, ega dless o hei posi ion wi hin an in on. This holds in dico s, bu he monoco splicing machine y is less dependen on UA composi- ion (GOODALL and FILIPOWICZ 1991). Thus he sho AT- ich inse ion in HA52, a monoco , in on I1 has p obably no e ec on ansc ip ion. Dele ions - The e is a dele ion o 21 bp in he in on I11 o HA52 (Fig. 2C). In c . Ado a, c . Ha una Nijo and H. spon aneum his agmen con ains a epea , GGTGGG, which is ound ou imes a he end o he ORF in all P-amyl alleles, also in HA52 allele. The longe agmen in he in on I11 may se e as a binding si e o a nega i e ansc ip ion ac o being a eason o he highe p-amylase ac i - i y in HA52 han he ba leys ha ing he G- ich epea . The 92-bp dele ion in he p omo e egion o HA52 (Fig. 2A) is posi ioned 451 bp ups eam om he TATA box (Fig. 1). OKADA e al. (2000) ound wo di ec epea s in he same egion in c . Ha una Nijo. The o he segmen s o he epea s a e ela i ely close o he CCAAT box a - 194 bp. Only 15 bp up- s eam om he CCAAT HA52 has an I-box, de ined as GATAA by TERZAGHI and CASHMORE (1995). Rela ed mo i es o GATAA a e ound in many plan p omo e s, some o which a e ligh egula ed ( e- iewed by TERZAGHI and CASHMORE 1995). The I-box in P-amyl is 183 bp ups eam om he TATA Table 1. Numbe o he bases in ol ed in he unique al e a ions o he P-amyl alleles in ba ley Ba ley To al numbe o bases In dele ions In inse ions In subs i u ions ~~ Ado a 0 HA52 125 H. spon aneum 39 Ha una Nijo 25 Numbe o bases 132 28 1 0 8 40 5 7 In he longes dele ion In he longes inse ion Ado a 0 HA52 92 H. spon aneum 38 Ha una Nijo 25 126 10 0 0 He edi as 134 (2001) B ie epo 93 A Ado a HA5 2 H. spon aneum Ha una Nijo Ado a HA5 2 H. spon aneum Ha una Ni j o Ado a HA5 2 H. spon aneum Ha una Nijo B Ado a HA5 2 H. spon aneum Ha una Nijo C Ado a HA5 2 H. spon aneum Ha una Nijo -616 -524 -615 -615 -567 -566 -566 -517 -482 -516 -516 TTTTTTTGGCCCCC-GAAGCATATTCTTCCGGGAGCCAAATTGACATTCC TTTTTTTGGTCCCTGGAAGCATATTCTCCCTTGAGCCAAATT-------- TTTTTTTGGCCCCC-GAAGCATATTCTTCCGGGAGCCAAATTGACATTCC TTTTTTTGGCCCCC-GAAGCATATTCTTCCGGGAGCCAAATTGACATTCC GGTCATGATGTGWTTGGATC-GTTAGTTATACAGATAAGGATATAT mTACCTCAACCGAATCTAGGTTACAACAAGCTTAACACTCATGCATTAG .................................. AACATTCATGCATTAG CPTACCTCAACCGAATCTAGGTTACAACAAGCTTAACACTCATGCATTAG CPTACCTCAACCGAATCTAGGTTACAACAAGCTTAACACTCATGCATTAG 739 CTAGTTCTCTGATGCATAT-T---TATA---------- TAGAAGTTCAAG 736 CTAGTTCTCTGATGCATATATAGATATACATATTTAGATAGAAGTTCAAG 739 CTAGTTCTCTGATGCATAT-T---TATA----------TAGAAGTTCAAG 739 CTAGTTCTCTGATGCATAT-T---TATA---------- TAGAAGTTCAAG 1982 1871 1835 1877 TGCTTATGGAGAAAGGTQTATGCATTTATACTTCAACAATAAGAATA TGCTTATGGA--------------------- TACTTCAACAATAAGAATA TGCTTATGGGGAAAGGTQTATGCATTTATACTTCAACAATAAAAATA TGCTTATGGGGAAAWTGGGCTATGCATTTATACTTCAACAATAAAAATA Fig. 2A-C. Sec ions o nucleo ide sequence alignmen s o P-a nyl alleles o c . Ado a, HA52, H. spon aneu n PI 296897 and c . Ha una Nijo. A A dele ion o 92 bp in he p omo e egion o HA52. Palind omic sequences a e in i alics and in e ed epea s a e in bold ace. B Inse ions in in on I1 o HA52 wi h he AT-mo i e in bold ace. C A dele ion o 21 bp in in on I11 o HA52 wi h he bold GT-mo i e. The sequence in bold ace in B also appea s in in on I11 wice, and he bold sequence in C appea s ou imes in owa ds he end o he se en h exon o he P-amyl gene. box (Fig. 1). This is in ag eemen wi h many ligh - egula ed ibulose- 1,5-bisphospha e ca boxylase genes ha ing a single I-box 100-300 bp ups eam om he TATA box (BORELLO e al. 1993). The dele ed agmen o 92 bp in he p omo e o HA52 is ele an in he o he 0-amyl alleles. Th ee in e ed epea s, ou palind omic sequences, and se en hai pin loops we e ound om he 92 bp agmen in he p omo e egion in P-amyl o c . Ado a, c . Ha una Nijo and H. spon aneum PI 296897 (Fig. 2A and da a no shown). This highly epe i i e sequence in p omo e egion 482 bp up- s eam o he ansc ip ion ini ia ion si e (Fig. 1) is a pu a i e binding si e o a nega i e ansc ip ion ac- o . I s absence may con ibu e o he high 0-amylase ac i i y o g ain mass in HA52, being 2.5 imes ha o c . Ha una Nijo and 2.9 imes ha o c . Ado a (AHOKAS and MANNINEN 2000). GGTTT mo i e and subs i u ions - Two GGTTT mo i es we e ound a posi ion -421 bp and -436 bp om ansc ip ion ini ia ion si e (Fig. 1). The GGTTT mo i e and addi ionally a GCCGC mo i e a e c i ical o exp ession in endospe m and emb yo in he maize Adhl (alcohol dehyd ogenase 1) gene p omo e exp essed in ansgenic ice (KYOZUKA e al. 1994). Because hese mo i es a e also equi ed o exp ession in o he issues, KYOZUKA e al. (1994) assumed ha he e migh be issue-speci ic pos - ans- la ional modi ica ions o binding p o eins o possibly addi ional p omo e elemen s. These mo i es may also hold ue o he Adhl -like genes in o he mono- co s bu no necessa ily o o he genes. Because no GCCGC mo i e was ound in 0-amyl gene, u he in es iga ions a e needed o explo e i he GGTTT mo i e (Fig. 1) alone is adequa e o speci ying he gene exp ession in endospe m. In 0-amyl p omo e , 94 M. J. E kkilu and H. Ahokas He edi as 134 (2001) OKADA e al. (2000) did no ind any speci ic se- quence simila o he endospe m box, which is known as a common sequence in he p omo e egion o p olamin genes and which seems o be associa ed wi h seed speci ic exp ession (HAMMOND-KOSACK e al. 1993). Fou single-base subs i u ions we e ound in he open eading ame o HA52. All o hese led o amino acid subs i u ions. The amino acid subs i u- ions in P-amyl o HA52 a e unique compa ed wi h known amino acid sequences o c . Ado a, c . Ha una Nijo, and H. spon aneum PI 296897 namely A g-115 + Cys, Asp-164 -- Glu, Phe-246 + Leu, and Val-430 -- Ala. None o hese subs i u ions is ound in conse ed egions o in he icini y o ac i e si es. The amino acids a he posi ions 115, 164 and 430 a e iden ical in c . Ha ing on (c . KANEKO e al. 2000) and in HA52. Because he e is no P-amylase ac i i y da a om same ial, we can only specula e on he ole o he wo amino acid di e ences, Phe-246 + Leu and Th -520 -- Ala, be ween c . Ha ing on and HA52. The p omo e egion and he in ons p obably ha e mo e e ec on he P-amylase ac i i y han amino acid subs i u ions, al hough hese may a ec seconda y modi ica ions o a ini ies o he p o ein. The Canadian c . Ha ing on has ances o s among Swedish and No wegian land aces. The e is a 25% chance ha c . Ha ing on inhe i ed i s P-amy 1 allele om c . Bjmneby, p o ided ha he e was no in a- genic ecombina ion. C . Bj~ neby o igina ed om he a ea o T ysil, No way (HAUGUM 1940). A ound 1600 Finns, especially om Sa onia, a sou h-eas e n p o ince o Finland, immig a ed o his a ea whe e hey used slash-and-bu n cul i a ion (HANSEN 1904; HAMALAINEN 1947). The Finns b ough ce eal seeds wi h hem (NORDMANN 1888; HAMALAINEN 1947). The e o e, he c . Bja neby and he line HA52 may ep esen a common gene pool om SE Finland. O he simila i ies o hei P-amyl genes would be wo h compa ing. The p omo e egion lacking om he HA52 P- amyl allele p o ides g ounds o pos ula e ha his egion se es as binding si e o a nega i e egula ion ac o . The HA52-speci ic p omo e and amino acid sequence may also cumula e in he P-amylase ac i i y. Compa ison be ween he di e en P-amy 1 alleles and P-amylase ac i i ies sugges ha he p omo e and he in on egions exe mo e in luence on P-amyl gene exp ession han he obse ed amino acid changes. Fu he s udies o he p omo e egion a e needed o con i m p esen pos ula ions and o de e mine spe- ci ic egions o enhancing, ep essing and endospe m- speci ic exp ession. ACKNOWLEDGEMENTS We hank Ms A Vi a o he ALF ope a ions. This esea ch was unded by he Raisio Yh yman Tu kimussaa- i6 Founda ion and Finnish Cul u al Founda ion. REFERENCES Ahokas H, (1977). Inc ease in p o ein con en by pa ial e ili y. Ba ley Gene . Newsl. 7: 6-8. Ahokas H, (1982). Cy oplasmic male s e ili y in ba ley. XI. The msm2 cy oplasm. Gene ics 102: 285-295. Ahokas H, (2000). Impac s on ag icul u al de elopmen by Cons an in Boije, a missiona y and he i s plan b eede in Finland. Yliopis opaino, Helsinki. Ahokas H and E kkila MJ, (1992). Ba ley P-amylase and P-glucanase ac i i ies a ge mina ion in ulga e- ype lines om backc osses o wild, spon aneum s ains wi h c . Ado a. Ag ic. Sci. Finl. 1: 339-350. g ain P-amylase and P-glucanase in Finnish land ace ba leys and hei pu a i e pas adap edness. He edi as 132: 111-118. Ahokas H and Manninen M-L, (2001). Polymo phisms o phospha e acquisi ion pa ame e s in ba ley (Ho deum ulga e) land aces: sec e ed acid phopha ase and milieu acidi ica ion o oo s a e ge mina ion in i o. Biol. Ag ic. Ho ic. 18: 385-399. Ahokas H and Naskali L, (1990). Va ia ion o @-amylase, P-amylase, P-glucanase, pullulanase, p o einase and chi inase ac i i y in ge mina ed samples o he wild p ogeni o o ba ley. J. Ins . B ew. 96: 27-31. Ahokas H and Poukkula M, (1999). Mal ing enzyme ac i - i ies, g ain p o ein a ia ion and yield po en ials in he displaced gene ic esou ces o ba ley land aces o Fin- land. Gene . Resou . C op E ol. 46: 251-260. Ahokas H, Uu ela P, E kkila MJ and Vahamiko S, (1996). Ano he sou ce o genes wi h high be a-amylase ac i i y in ba ley g ain: Finnish land aces. Ba ley Gene . Newsl. 25: 36-40. Bo e110 U, Cecca elli E and Giuliano G, (1993). Cons i u- i e, ligh - esponsi e and ci ca dian clock- esponsi e ac o s compe e o he di e en I box elemen s in plan ligh - egula ed p omo e s. Plan J. 4: 61 1-619. E kkila MJ, (1999). In on 111-speci ic ma ke s o sc een- ing o P-amylase alleles in ba ley cul i a s. Plan Mol. Biol. Rep. 17: 139-147. E kkila MJ, Leah R, Ahokas H and Came on-Mills V, (1998). Allele-dependen ba ley g ain P-amylase ac i i y. Plan Physiol. 117: 679-685. Fo s e BP, Thompson DM, Wa e s J and Powell W, (1991). Wa e -soluble p o eins o ma u e ba ley en- dospe m: gene ic con ol, polymo phism, and linkage wi h P-amylase and sp ing/win e habi . Theo . Appl. Gene . 81: 787-792. Gniadkowski M, Hemmings-Mieszczak M, Klah e U, Liu H-X and Filipowicz W, (1996). Cha ac e iza ion o in- onic u idine- ich sequence elemen s ac ing as possible a ge s o nuclea p o eins du ing p e-mRNA splicing in Nico iana plumbagini olia. Nucleic Acids Res. 24: 6 19-627” Goodall GJ and Filipowicz W, (1991). Di e en e ec s o in on nucleo ide composi ion and seconda y s uc u e on p e-mRNA splicing in monoco and dico plan s. EMBO J. 10: 2635-2644. Ahokas H and Manninen M-L, (2000). The mos ab He edi as 134 (2001) B ie epo 95 Hamalainen A, (1947). Keski-Skandina ian suomalaise . We ne Sode s om Oy, Po oo, Helsinki. Hammond-Kosack MCU, Holdswo h MJ and Be an MW, (1993). In i o oo p in ing o a low molecula weigh glu enin gene (LMWG-IDl) in whea en- dospe m. EMBO J. 12: 545-554. Hansen AM, (1904). LandnHm i No ge. WC Fab i ius & Smne A/S, K is iania. Haugum 0, (1940). VH e ko nso e . C Dahls Bok- & Kuns ykke i A/S, Oslo. Hejgaa d J and Boisen S, (1980). High-lysine p o eins in Hip oly ba ley b eeding: iden i ica ion, nu i ional sig- ni icance and new sc eening me hods. He edi as 93: 311-320. Kaneko T, Kiha a M and I o K, (2000). Gene ic analysis o P-amylase he mos abili y o de elop a DNA ma ke o mal ennen abili y imp o emen in ba ley, Ho deum ulga e. Plan B eeding 119: 197-201. Kozack M, (1984). Compila ion and analysis o sequences ups eam om he ansla ional s a si e in euca yo ic mRNAs. Nucleic Acids Res. 12: 857-872. K eis M, Williamson M, Bux on B, Pywell J, Hejgaa d J and S endsen I, (1987). P ima y s uc u e and di e en- ial exp ession o P-amylase in no mal and mu an ba - leys. Eu . J. Biochem. 169: 517-525. K eis M, Williamson MS, Shew y PR, Sha p P and Gale M, (1988). Iden i ica ion o a second locus encoding P-amylase on ch omosome 2 o ba ley. Gene . Res. 51: 13-16. Kyozuka J, Oli e M, Peacock WJ, Dennis ES and Shi- mamo o K, (1994). P omo e elemen s equi ed o de elopmen al exp ession o he maize Adhl gene in ansgenic ice. Plan Cell 6: 799-810. MacG ego AW and Finche GB, (1993). Ca bohyd a es o he ba ley g ain. In: Ba ley: Chemis y and Technology. (eds AW MacG ego and RS Bha y), Ame ican Associ- a ion o Ce eal Chemis s, S . Paul, MN, p. 73-130. Messing J, Ge agh y D, Heidecke G, Hu N-T, K idl J and Rubens ein I, (1983). Plan gene s uc u e. In: Gene ic enginee ing o plan s. (eds T Kosuge, CP Me edi h and A Hollaende ), Plenum Publishing Co p, p. 21 1-227. No dmann P, (1888). Finna ne i melle s a S e ige. Tidn- ings- & T ycke i- ak iebolage , Helsing o s. Okada Y, Kiha a M, Ku oda H, Yoshigi N and I o K, (2000). Cloning and sequencing o he p omo e egion o he seed speci ic P-amylase gene om ba ley. J. Plan Physiol. 156: 762-767. Roby JF and Whelan WJ, (1968). The P-amylases. In: S a ch and i s de i a i es (ed J Radley). Chapman and Hall, London, p. 430-476. Saghai Ma oo MA, Zhang Q and Biyashe R, (1995). Compa ison o es ic ion agmen leng h polymo - phisms in wild and cul i a ed ba ley. Genome 38: 298- 306. Sha p PJ, K eis M, Shew y PR and Gale MD, (1988). Loca ion o P-amylase sequences in whea and i s ela- i es. Theo . Appl. Gene . 75: 286-290. Sopanen T and Lau ie e C, (1989). Release and ac i i y o bound P-amylase in a ge mina ing ba ley g ain. Plan Physiol. 89: 244-249. Sun Z and Henson C, (1990). Deg ada ion o na i e s a ch g anules by ba ley cl-glucosidases. Plan Physiol. 94: 320-327. Te zaghi WB and Cashmo e AR, (1995). Ligh - egula ed ansc ip ion. Annu. Re . Plan Physiol. Plan Mol. Biol. 46: 445-474. Thompson DM, Powell W and Fo s e BP, (1990). Use o isoelec o ocusing in ba ley a ie al iden i ica ion. Ann. Appl. Biol. 117: 625-631. Yoshigi N, Okada Y, Saha a H and Tamaki T, (1995). A s uc u al gene encoding P-amylase o ba ley. Biosci. Bio ech. Biochem. 59: 1991- 1993. Zhang Q, Saghai Ma oo MA and Kleinho s A, (1993). Compa a i e di e si y analysis o RFLPs and isozymes wi hin and among popula ions o Ho deum ulga e ssp. spon aneum. Gene ics 134: 909-916.