tandemly repeated 147 bp elements cause structural and functional variation in divergentmal...

. 13: 1135–1144 (1997)

Tandemly Repeated 147 bp Elements Cause Structuraland Functional Variation in Divergent MAL Promotersof Saccharomyces cerevisiae

PHILIP J. L. BELL1,2,3*, VINCENT J. HIGGINS1,2, IAN W. DAWES2,3 AND PETER H. BISSINGER1,2

1Burns Philp Research and Development, 67 Epping Road, North Ryde, New South Wales 2113, Australia2CRC for Food Industry Innovation, Sydney, New South Wales 2052, Australia3University of New South Wales, Department of Biochemistry and Molecular Genetics, Sydney,New South Wales 2052, Australia

Received 1 November 1996; accepted 5 March 1997

We have studied four novel MAL promoters isolated from a single strain of bakers’ yeast. Within these promoterswe have identified up to five tandem 147 bp repeats located between the MAL UAS region and the MALT TATAbox. These repeats strongly reduce MALT (maltose permease) gene expression but only weakly reduce MALS(maltase) gene expression. Insertion of the 147 bp elements into the heterologous CYC1 promoter reducedexpression when located between the CYC1 UAS and the TATA box, but not when located upstream of the UAS.We propose that these naturally occurring repeats have evolved as a mechanism to lower the level of MALTexpression relative to MALS expression, thus avoiding possible toxic effects associated with over-expression frommultiple copies of the permease gene. Accession numbers are: WIG1, U86359; WIG3, U86360; WIG4, U86361;WIG5, U86362. ? 1997 John Wiley & Sons, Ltd.

Yeast 13: 1135–1144, 1997.No. of Figures: 7. No of Tables: 0. No. of References: 31.

— yeast; MAL6; divergent promoter; repeats; nucleosomes

INTRODUCTION

In laboratory strains of yeast, five different MALloci have been identified (MAL1, MAL2, MAL3,MAL4 and MAL6), each of which shares a com-mon structure, and any one functional locus allowsyeast to utilize maltose (Barnett, 1976). TheMAL6locus derived from laboratory strain CB11 hasplayed a central role in the development of thestandard model for MAL gene structure andfunction (Needleman, 1991). At this locus twogenes, MAL6T encoding a maltose permease andMAL6S encoding a maltase, are divergently tran-scribed from a single 887 bp promoter region(Needleman et al., 1984). A third gene,MAL6R, islocated upstream of MAL6T and encodes a posi-tive trans-acting regulatory protein that in thepresence of maltose coordinately induces theexpression of the MAL6T and MAL6S genes

(Chang et al., 1988; Dubin et al., 1986; Goldenthal,et al., 1987).The MAL divergent promoter contains a UAS

region (Hong and Marmur, 1987; Levine et al.,1992; Sirenko et al., 1995), two symmetricallyoriented TATA boxes (Hong and Marmur, 1986),and two functional consensus sequences for thebinding of Mig1 transcription factor (Hu et al.,1995; Figure 1). Induction by maltose is controlledby the bidirectional UAS elements (Levine et al.,1992; Yao et al., 1994) and repression by glucose iscontrolled partly through the action of the twofunctional Mig1p binding sites and partlythrough MAL6R autoregulation (Hu et al., 1995).Under induced conditions expression levels ofthe MAL6T and MAL6S genes are comparableand relatively high (Levine et al., 1992) whileunder repressed conditions expression levels fromboth genes are reduced to very low levels andunder non-inducing/non-repressing conditions the*Correspondence to: Philip J. L. Bell.

CCC 0749–503X/97/121135–10 $17.50? 1997 John Wiley & Sons, Ltd.

.
1136 . . .
-s’ellAn9daldds6y

aloe-defdh

d.

expression of the MAL6S gene is considerablyhigher than that of theMAL6T gene (Levine et al.,1992).Bakers’ yeast (mostly strains of Saccharomyces

cerevisiae) has been used in the manufacture ofbread for several millennia. During this periodthe yeast have been required to ferment maltosepresent in the doughs rapidly and efficiently. It ispossible that this selection for rapid utilization ofmaltose by industrial yeast strains has evolvedvariants of the MAL loci that have not beendescribed in laboratory strains. One structuralvariant of the MAL divergent promoter has beenidentified in an industrial strain of bakers’ yeast(Kopetski et al., 1989).In this paper, we describe the cloning of four

closely related divergent MAL promoters contain-ing 147 bp tandem repeats located between theMAL UAS region and the MALT TATA box. Byusing a two-reporter gene system (Bell et al., 1995),we have demonstrated that the repeats stronglyreduce expression of the MALT gene relative totheMALS gene. The repeats thus provide a mech-anism to control the relative levels of expressionfrom the divergently arranged MAL promoters.To our knowledge, this is the first time thatnaturally occurring tandem repeats have beenimplicated in the down-regulation of a yeastpromoter.

MATERIALS AND METHODS

Strains and plasmidsSaccharomyces cerevisiae strain N1, a com

mercially produced industrial strain of bakeryeast (Mauri foods, Sydney) was the source of thMAL promoters. Assays of expression from aconstructs were performed in strain 1403-7(MATaMAL4-CMAL1g MAL3g trp1 ura3; Kahand Eaton, 1971). Escherichia coli strains JM10and XL-blue (Stratagene) were used for all plasmimanipulations. Plasmid pBIDB is an episomversion of the pIBID integrating vector describepreviously (Bell et al., 1995). It is a YEp-baseURA3 vector containing lacZ and MEL1 genetranscribed under the control of the MALpromoter. Plasmid pLG669-Z is described bGuarente and Ptashne (1981).

Media growth conditionsYeast strain 1403-7A was grown on minim

medium (0·67% yeast nitrogen base without aminacids, supplemented with all auxotrophic requirments except uracil) to maintain the pBIDB-basevectors. Fermentable substrates (sugars) weradded to a concentration of 2% (w/v). A mixture oethanol (2% v/v) and glycerol (3% v/v) was addeto provide a non-fermentable carbon source. Ricmedium (1% yeast extract, 2% peptone (w/v), 2%

Figure 1. Structure of the MAL6 locus. Three genes are located at the MAL6 locus: aregulatory gene (MAL6R), a maltose permease gene (MAL6T) and a maltase gene (MAL6S).The MAL6T and MAL6S genes are divergently arranged about a single promoter region887 bp in length. The promoter contains a number of known transcriptional control elementsincluding two TATA elements (Hong and Marmur, 1986), two functional Mig1p bindingsites (Hu et al., 1995) and a UAS region containing three MALR binding sites (Hong andMarmur, 1987; Levine et al., 1992; Sirenko and Needleman, 1995).

? 1997 John Wiley & Sons, Lt . 13: 1135–1144

t.edors

a-alnn-I,do-er-dI/eI/e

-eh-es-s,tsdddL-e

seds),I

7

7)

(w/v) glucose) was used to culture cells when noselective pressure was required.

TransformationsE. coli transformations were performed using

commercially prepared competent cells (Stratagenestrain JM109). Yeast transformations were per-formed using a kit from Bio101 according to themanufacturer’s instructions.

Production of cell-free extracts and determinationof protein concentrationOvernight cultures were harvested at an OD640

of 0·3 to 0·4 for assessment of expression from thepBIDB-based vectors, and of 0·9 to 1·1 for assess-ment of expression from the CYC1 promoter.Harvested cells were resuspended in breakagebuffer (0·1 -sodium citrate buffer, 0·1 -EDTA,1 m-DTT, 0·17 mg/ml PMSF, 0·7 ì-pepstatin),and homogenized for 60 s in a Braun homogenizer(setting 1, cooled by CO2). The extracts werecentrifuged for 1 min at 13 000 g at 4)C, and thesupernatant was used as a cell-free extract. Proteinconcentration was determined by the method ofBradford (1976).

Enzyme assaysâ-Galactosidase activity was determined using

cell-free extracts. The extracts were assayedusing o-nitrophenyl-glucospyranoside as substrateas described by Miller (1972), and specific activitiesdetermined based on the protein concentration ofthe extracts. Specific activity was measured asnmol o-nitrophenol released per min per mg pro-tein at 28)C and pH 7. All assays were performedin triplicate from duplicate experiments. Similarly,á-galactosidase activity was determined usingcell-free extracts. Cell-free extract and assaybuffer were added to a total volume of 400 ìl,followed by the addition of 100 ìl p-nitrophenyl-galactopyranoside (15 mg/ml), and activity wasnormalized using the protein concentration of theextracts. The specific activity of cell-free extractwas defined as nmol p-nitrophenol released permin per mg of protein at 28)C at pH 4. All assayswere in triplicate from duplicate experiments.

DNA sequence determinationAutomated sequencing was carried out accord-

ing to the dideoxy chain-termination method ofSanger et al. (1977) using the ABI PRISM Ready

Reaction Dye Terminator Sequencing KiAppropriate restriction fragments, created by onor two different restriction enzymes, were prepareby agarose gel electrophoresis and cloned intplasmid pUC18. Universal sequencing primewere used to prime the sequencing reactions.

Cloning of divergent promoters from strain N1using a nested mini-library strategyA 1·1 kb EcoRI/BglII probe (labelled using

DIG labelling kit, Boehringer Mannheim) containing part of the MAL6T and MAL6S structurgenes, and the entire MAL6 promoter regiowas used to identify MAL sequences by Southerhybridization (Sambrook et al., 1989). Chromosomal DNA (2·4 mg) was digested with HindIIthe products were separated by electrophoresis anfour bands were purified which corresponded tfour of the five HindIII polymorphic fragments identified by the MAL probe. ThHindIII-digested DNA from each band was futher digested with EcoRI/KpnI, separated anbands corresponding to the length of four EcoRKpnI polymorphic species were purified. Thdouble purified DNA was subcloned into EcoRKpnI-digested pUC18 and screened using th1·1 kb EcoRI/BglII probe.

Construction of pBIDB-based vectors for analysisof divergent expressionThe wild-type intergenic promoters were intro

duced into the pBIDB vector by digesting both thvector and pWIG (see Results) plasmids witEcoRI/BfrI. Purification of the appropriate promoter and vector fragments allowed ligation of thintergenic regions into the two-reporter gene sytem. To delete all but one of the tandem repeatpUC-based vectors with multiple tandem repeawere digested with StyI (a restriction site locateonce in each repeat), and religated. The religatevectors were screened for clones that containeonly one copy of the tandem repeat in the MApromoter and suitable promoters were subcloned into the pBIDB vector as described for thwild-type promoters.

Construction of CYC1+ tandem repeat promoterThe tandem 147 bp repeats were cut out of th

MAL promoters using HaeII/HinFI enzymes anXhoI linkers added. The pLG669-Z vector wapartially digested in XhoI (which cuts at two sitestreated with Klenow enzyme (to destroy the Xho

113

? 1997 John Wiley & Sons, Ltd. . 13: 1135–1144 (199

f

eeissdsdtsee)d-ledfLfs-Tly6dses

netsyeete

.

d.

sites) and religated to circularize the plasmids. Therecircularized plasmids were transformed intoE. coli strain XL2-Blue and screened for thedesired forms. Isolates of plasmids containingsingle XhoI sites were identified (representingsingle XhoI site deletions) and used to subclone themodified XhoI-linkered repeats into the CYC1promoter region.

RESULTS

DNA sequence of four highly homologous MALpromoters identifies 147 bp tandemly repeatedregionsThe MAL loci of strain N1 were examined by

Southern hybridization. Five restriction lengthpolymorphisms (9·4, 8·1, 7·6, 7·4 and 7·0 kb) wereidentified when chromosomal DNA was digestedwith HindIII and probed with the MAL6 inter-genic region and part of the MAL6S structuralgene. Further mapping showed that four ofthese polymorphic species could be cloned asEcoRI/KpnI fragments (1·8, 1·65, 1·5 and 1·2 kb).Using a nested mini-library technique (described inMaterials and Methods), 18 clones with homologyto the MAL promoter were identified, includingrepresentatives of each of the four EcoRI/KpnIpolymorphisms. Restriction mapping of one rep-resentative of each of the polymorphic speciesindicated that the four types of clones were highlyhomologous, except for the presence of a variablenumber of repeated regions containing StyI sites.One isolate of each of the species was chosen forfurther study. The clones were designated pWIG1(Wild-type Intergenic Region 1 StyI site), pWIG3,

pWIG4 and pWIG5 according to the number oStyI sites present within the MAL promoters.The largest of the cloned promoters (in clon

pWIG5) was sequenced and found to contain fivtandemly repeated copies of a 147 bp region thatfound only once in the MAL6 promoter. ClonepWIG1, pWIG3 and pWIG4 were also sequenceand found to contain one, three and four copierespectively of the same repeated region founin pWIG5. In each clone, the 147 bp elemenwere located at an identical position within thpromoter. Alignment of the repeats with thequivalent MAL6 region (Levine et al., 1992revealed that only eight positions were altereamong the 14 copies examined (Figure 2). Comparison of the 147 bp region with the availabyeast DNA database (Entrez Release 21) identifiefive sequences with significant homology. Othese five sequences, four were located in MApromoters, and one was located in the promoter othe AGT1 gene (a general alpha-glucoside tranporter which is highly homologous to the MAL6gene; Han et al., 1995). Apart from the tandemrepeated regions, nine differences from the MALpromoter were common to each of the clonepromoters from strain N1. These nine differenceincluded single base pair changes in both thMALT TATA box and two of theMAL UAS site(Figure 3).The tandem repeat elements are located betwee

the UAS region and the Mig1p site proximal to thMALT gene. The location of the tandem repeacauses two significant structural changes: first, theincrease the distance between the UAS and thMALT TATA box; second, they increase thdistance between the UAS and the Mig1p si

Figure 2. Comparison between tandem 147 bp repeated elements. Each of the tandemly repeated sequences arealigned with the equivalentMAL6 region. Repeats are numbered according to their position relative to theMALTpromoter. Thus WIG3 R1 represents the repeat most proximal to theMALT structural gene found in clone pWIG3.Differences between the cloned tandem repeats and the MAL6 region are shown.


1138 . . .

o

adL

9

7)

proximal to the MALT gene. Thus in the MAL6and WIG1 promoters the distance between theUAS and the MALT TATA box is approxi-mately 360 bp, while in the WIG3 (three repeats)promoter the distance is increased to 2660 bp, inthe WIG4 (four repeats) promoter it is increased to2810 bp and in the WIG5 (five repeats) promoterit is increased to 2960 bp. The tandem repeats donot increase the distances between the UAS regionand the MALS TATA box nor the distance

between the UAS and the Mig1p site proximal tthe MALS gene.

The tandem repeats strongly reduce expressionfrom the MALT promoter but have minor effectson expression from the MALS promoterEach of the promoters was introduced into

two reporter gene vector (Bell et al., 1995) andivergent expression was examined in the MA

Figure 3. Sequence comparison between WIG promoters and the MAL6 promoter. For clarity, thesequences are shown with the number of repeated regions reduced to one. The boxes show the position of thetwo functional Mig1p binding sites (Hu et al., 1995). The boxes with circular edges show the location of threeMALR binding sites (Sirenko and Needleman, 1995). The shaded boxes show the position of the TATA boxsequences. The underlined region shows the position of the 147 bp region that is repeated in clones pWIG3,pWIG4 and pWIG5. Start codons for theMALT andMALS genes are in bold type. The StyI and BfrI sitesare overlined.


113

c-lrestsesad

1nLs1d,dd,steedone

.

d.

constitutive strain 1403-7A. On maltose-containing substrate, expression strength in theMALT (lacZ reporter gene) direction was lower inpromoters containing multiple tandem repeatsthan in promoters containing only one copy of therepeat (Figure 4). Thus there was an approxi-mately 17-fold difference between expression fromthe strongest MALT promoter and the weakest ofthe MALT promoters. By contrast, expressionstrength in the MALS (MEL1 reporter gene) di-rection was similar between the WIG1 promoterwith a single repeat or WIG5 with five repeats.When expression was examined in a well-regulatedstrain, MALT (lacZ reporter gene) promoterstrength was again found to be lower in promoterscontaining multiple tandem repeats than in pro-moters containing only one copy of the repeat.Expression in the MALS direction was againrelatively similar between clones (data not shown).To determine whether the 147 bp repeats were

reducing expression from theMALT promoter, weexamined the effect of reducing the number ofrepeats to one (see Materials and Methods). Theconstructs were assayed in strain 1403-7A underinducing (maltose), repressing (glucose), and non-inducing/non-repressing (ethanol/glycerol) con-ditions. In each case, deletion of repeats increasedexpression from the MALT promoter (up to 22-fold) with minor effects on the MALS promoterstrength (1·5-fold; Figure 5). On both glucoseand maltose, the degree to which expression wasreduced in theMALT direction correlated with the

number of repeats, whereas in the MALS diretion it did not (Figure 6). On ethanol/glyceromedia (which corresponds to neither induction norepression) expression was reduced by the presencof the repeats, although the effect was not aclearly correlated with the number of repea(Figure 6). The high expression levels on glucosand ethanol/glycerol occurred because the assaywere performed in strain 1403-7A which containsglucose-insensitive MALR gene (Charron anMichels, 1987; Kahn and Eaton, 1971).

Tandem repeats block expression from the CYC1promoterThe tandem repeats were inserted into the CYC

promoter to examine the effects of the repeats oexpression from promoters other than the MAdivergent promoter. The CYC1 promoter containtwo XhoI sites; one is located between the CYCUAS and the TATA box, and the other is locateupstream of the CYC1 UAS (McNeil and Smith1986). To facilitate insertion of the repeateregions into the CYC1 promoter, two modifieversions of the CYC1 promoter were constructedone in which the XhoI site proximal to CYC1 waremoved, and the other in which the XhoI sidistal to the CYC1 promoter was removed (seMaterials and Methods). Insertion of the repeateregions into the proximal XhoI site was found treduce CYC1 promoter strength, whereas insertiointo the distal XhoI site did not (Figure 7). Th

Figure 4. Expression from the WIG promoters and the MAL6 promoter instrain 1403-7A under inducing conditions. Cultures were grown in 2% maltoseminimal medium and harvested at OD 0·3–0·4. Activities are expressed asnmol/min per mg protein. All values are averages of duplicate experimentsperformed in triplicate assays.


1140 . . .

enrs

1

7)

insertion of a single copy of the repeat (2150 bp)between the UAS and the TATA box reducedexpression approximately 200-fold and insertion ofadditional copies of the repeats reduced expressiona further five- to six-fold.

DISCUSSION

The results obtained in this study illustrate somof the complexity of MAL gene organization iindustrial yeast. Within the four MAL promote

Figure 6. Effect of the 147 bp elements on expression in the MALT andMALS directions. Specific enzyme activities were used to calculate the ratio ofpromoter strength from the unaltered and modified promoters. Each columnrepresents the ratio of the expression from the modified promoter (with oneremaining 147 bp element) and the corresponding native WIG promoter withmultiple 147 bp elements.


Figure 5. Effect of the 147 bp elements on expression from the WIG5, WIG4 and WIG3 promoters.In each case, the number of repeats was reduced to one and compared to the expression from theunaltered promoter. â-Galactosidase activity corresponds to expression from the MALT promoter,and á-galactosidase activity corresponds to expression from the MALS promoter. All activities areexpressed as nmol/min per mg protein. Cultures were assayed during growth under inducing(maltose), repressing (glucose) and non-inducing/non-repressing (ethanol/glycerol) conditions. Allvalues are averages of duplicate experiments performed in triplicate assays.

114

gnein

tetat

pc-etser.gyeer,s-nS

.

d.

that were isolated from strain N1, tandemlyrepeated 147 bp elements were identified thatreduce transcription from the MALT promoterbut have only a minimal effect on the MALSpromoter. By reducing the expression of only onegene from a bidirectionally activated promoter,these tandem repeats provide a mechanism tomodulate the relative levels of gene expression ofthe two genes.The sequence comparison between the four

MAL promoters identified in this study andthe MAL6 promoter revealed several mutationsincluding single base substitutions and tandemlyrepeated regions. Excluding the tandemly repeatedregions, 17 single base substitutions occur (Figure3). Of these base substitutions, nine are common toeach of the industrial MAL promoters, indicatingthat the MAL promoters isolated from strain N1are more closely related to each other than they areto theMAL6 promoter. Three of the substitutionsin common occur in known regulatory elements;two in the MALR binding sites and one in theMALT TATA box. The effect of these mutationson expression was not determined; however, eachof the novel promoters expressed more stronglyin the MALS direction (approximately two-fold

stronger) than did theMAL6 promoter, suggestinthe presence of mutations that increase expressiolevels. The promoter WIG3 was identical to thpreviously reported GLUCP1 clone (Kopetsket al., 1989), except for a single base pair transitio(T]C) at a position 129 bp from theMALT ATGcodon (Figure 3). Using the 147 bp repeat elemenas a probe in Southern hybridizations, we havfound that these 147 bp tandem repeats are noonly found in strain N1 but are also found inrange of other bakers’ yeast strains (data noshown).Our studies demonstrate that the tandem 147 b

repeats reduce expression in the MALT diretion more than in the MALS direction. Thus thquestion arises as to how the 147 bp elemencause a selective reduction in expression of ongene from a bidirectionally activated promoteOne possibility is that they function by alterinthe structural relationship between regulatorelements within the bidirectional promoter. In thMALT direction, where the 147 bp repeats arlocated between the UAS and the basal promotethe repeats caused a significant decrease in expresion. This decrease in expression correlates with aincrease in the distance between the MAL UA

Figure 7. Effect of the 147 bp elements on CYC1 promoter strength.Two versions of the CYC1 promoter found on plasmid pLG669-Z(Guarente and Ptashne, 1981) were constructed, each of which containedonly one XhoI site. The single and multiple 147 bp elements were insertedinto both constructs as shown in the schematic diagram. XhoI sites aremarked ‘X’. Specific â-galactosidase (â-gal.) activity produced by theCYC1/lacZ fusion protein was measured in cultures grown on ethanol/glycerol minimal medium (derepressed conditions). All activities areexpressed in nmol/min per mg protein and are averages of three exper-iments performed in triplicate.


1142 . . .

egeirocil-ofanepaled

t.

s,eec-

rinl.

dees

i-ee.

ddin6,

s-no

rD

ds-t.

3

7)

and the MALT TATA box. Conversely, in theMALS direction, where the 147 bp repeats arelocated upstream of the UAS and thus do notincrease the distance between the UAS and theTATA box, the repeats caused only a minordecrease in expression.When inserted between the CYC1 UAS and

the basal promoter, the 147 bp elements dra-matically reduced promoter strength. Since earlystudies reported that insertion of random DNAfragments of similar length from E. coli causedonly a moderate decrease in expression (Guarenteand Hoar, 1984), the reduced CYC1 expressioncould suggest that the 147 bp element containsspecific repressor sites. However, a recent studyhas shown that random yeast DNA of approxi-mately 150 bp length inserted into the sameposition blocks expression in a similar fashion tothe 147 bp element (Vincent and Gancedo, 1995).In that study, a number of fragments of theFBP1 structural gene were inserted into the XhoIsite of the CYC1 promoter, where they werefound to reduce CYC1 expression by up to 100-fold. The observation that yeast DNA of differ-ent origins (our 147 bp element, and the randomFBP1 gene fragments) block expression from theCYC1 promoter indicates that the reduction isdue to a structural effect of yeast DNA ratherthan a sequence-specific effect. When the repeatswere inserted upstream of the CYC1 UASthere was no reduction in expression. Again,the repeats appeared to function only whenlocated between the UAS and the basal promoteralthough in this case the absence of an effect maybe related to the independent regulation of theCYC1 and ANB1 genes (Lowry et al., 1983;Zitomer and Lowry, 1992).It is noteworthy that the length of the 147 bp

repeated regions almost exactly coincides with the146 bp length of DNA that wraps around the coreparticle of a nucleosome (McGhee and Felsenfeld,1980). Since nucleosome positioning is increasinglyimplicated in controlling the access of regulatoryfactors to promoter regions, a possible relationshipmay exist between the length of the repeats andnucleosome positioning. Alternatively, the corre-lation in length may be coincidental, and thelength of the repeats determined by the presence ofalmost perfect (8/10) direct repeats at either end ofthe 147 bp element. These small direct repeatscould allow production of the tandem 147 bpelements via unequal exchanges or replicationslippages.

We speculate that the biological origin of th147 bp repeats may be related to two opposinselective pressures operating on yeast. On the onhand yeast have been selected for centuries on theability to utilize maltose rapidly. This appears thave been achieved by duplication of theMAL loresulting in a greater capacity to hydrolyse matose. On the other hand, yeast strains are subject taccelerated death caused by excessive uptake omaltose (Entian and Loureiro-Dias, 1990; Postmet al., 1990). By selectively reducing the expressioof some of these additional maltose permeasgenes relative to the maltose gene, the 147 brepeats may allow the yeast to overcome potentitoxic effects associated with the presence of thmultiple functional MAL loci required for rapimaltose utilization.

REFERENCES

Barnett, J. A. (1976). The utilisation of sugars by yeasAdv. Carbohydr. Chem. Biochem. 32, 125–134.

Bell, P. J. L., Bissinger, P. H., Evans, R. J. and DaweI. W. (1995). A two-reporter gene system for thanalysis of bi-directional transcription from thdivergent MAL6T-MAL6S promoter in Sacharomyces cerevisiae. Curr. Genet. 28, 441–446.

Bradford, M. (1976). A rapid and sensitive method fothe quantification of microgram quantities of proteutilising the principle of protein-dye binding. AnaBiochem. 72, 248–254.

Chang, Y. S., Dubin, R. A., Perkins, E., Forrest, D. anMichels, C. A. (1988). MAL63 codes for a positivregulator of maltose fermentation in Saccharomyccerevisiae. Curr. Genet. 14, 201–209.

Charon, M. J. and Michels, C. A. (1987). The consttutive, glucose-repression-insensitive mutation of thyeastMAL4 locus is an alteration of theMAL43 genGenetics 116, 23–31.

Dubin, R. A., Perkins, E. L., Needleman, R. B. anMichels, C. A. (1986). Identification of a secontrans-acting gene controlling maltose fermentationSaccharomyces carlsbergensis. Mol. Cell. Biol.2757–2765.

Entian, K. D. and Loureiro-Dias, M. C. (1990). Miregulation of maltose uptake in a glucose repressiodefective mutant of Saccharomyces cerevisiae leads tglucose poisoning. J. Gen. Microbiol. 136, 855–860.

Entrez database. Release 21 (1996). National centre fobiotechnology information, NIH, Bethesda, M20894, USA.

Goldenthal, M. J., Vanoni, M., Buchferer, B. anMarmur, J. (1987). Regulation of MAL gene expresion in yeast: gene dosage effects. Mol. Gen. Gene209, 508–517.

114


e

n1-

s.g

in

l.sa-

.der

).ldr,

Ac.

).p7,

astx.

r,din

nes

.

Guarente, L. and Hoar, E. (1984). Upstream activationsites of the CYC1 gene of Saccharomyces cerevisiaeare active when inverted but not when placed down-stream of the TATA box. Proc. Natl. Acad. Sci. USA81, 7860–7864.

Guarente, L. and Ptashne, M. (1981). Fusion ofEscherichia coli lacZ to the cytochrome c gene ofSaccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA78, 2199–2203.

Han, E. K., Cotty, F., Sottas, C., Jiang, H. and Michels,C. A. (1995). Characterisation of AGT1 encodinga general alpha-glucoside transporter from Sac-charomyces. Mol. Microbiol. 17, 1093–1107.

Hong, S. H. and Marmur, J. (1986). Primary structureof the maltase gene of the MAL6 locus of Sac-charomyces carlsbergensis. Gene 41, 75–84.

Hong, S. H. and Marmur, J. (1987). Upstreamregulatory regions controlling the expression of theyeast maltase gene. Mol. Cell. Biol. 7, 2477–2483.

Hu, Z., Nehlin, O. J., Ronne, H. and Michels, C. A.(1995). MIG1-dependent and MIG1-independentglucose regulation of MAL gene expression in Sac-charomyces cerevisiae. Curr. Genet. 28, 258–266.

Kahn, N. A. and Eaton, N. R. (1971). Genetic control ofmaltose formation in yeast. I. Strains producing highand low basal levels of enzymes.Mol. Gen. Genet. 112,317–322.

Kopetski, E., Buckel, P. and Schumacher, G. (1989).Cloning and characterisation of bakers’ yeastá-glucosidase: Over-expression in a yeast straindevoid of vacuolar proteinases. Yeast 5, 11–24.

Levine, J., Tanouye, L. and Michels, C. A. (1992). TheUASMAL is a bidirectional promoter element requiredfor the expression of both the MAL61 and MAL62genes of the Saccharomyces MAL6 locus. Curr. Genet.22, 181–189.

Lowry, C. V., Weiss, J. L., Walthall, D. A. and Zitomer,R. S. (1983). Modulator sequences mediate theoxygen regulation of CYC1 and a neighboring gene inyeast. Proc. Natl. Acad. Sci. USA 80, 151–155.

McGhee, J. D. and Felsenfeld, G. (1980). Nucleosomstructure. Annu. Rev. Biochem. 49, 1115–1156.

McNeil, J. B. and Smith, M. (1986). Transcriptioinitiation of the Saccharomyces cerevisiae iso-cytochrome c gene. J. Mol. Biol. 187, 363–378.

Miller, J. H. (1972). Experiments in Molecular GeneticCold Spring Harbor Laboratory Press, Cold SprinHarbor, New York.

Needleman, R. (1991). Control of maltase synthesisyeast. Mol. Microbiol. 5, 2079–2084.

Needleman, R. B., Kaback, D. B., Dubin, R. A., et a(1984). MAL6 of Saccharomyces: a complex locucontaining three genes required for maltose fermenttion. Proc. Natl. Acad. Sci. USA 81, 2811–2815.

Postma, E., Verduyn, C., Kuiper, A., Scheffers, W. Aand Van Dijken, J. P. (1990). Substrate-acceleratedeath of Saccharomyces cerevisiae CBS 8066 undmaltose stress. Yeast 6, 149–158.

Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989Molecular Cloning: A Laboratory Manual. CoSpring Harbor Laboratory, Cold Spring HarboNew York.

Sanger, F., Nicklen, S. and Coulson, A. R. (1977). DNsequencing with chain terminating inhibitors. ProNatl. Acad. Sci. USA 74, 5463–5467.

Sirenko, O. I., Ni, B. and Needleman, R. B. (1995Purification and binding properties of the Mal63activator of Saccharomyces cerevisiae. Curr. Genet. 2509–516.

Vincent, O. and Gancedo, J. M. (1995). Expression ofyeast gene can be blocked by insertion of short yeaDNA fragments between a UAS and the TATA boCurr. Genet. 27, 387–389.

Yao, B., Sollitti, P., Zhang, X. L. and MarmuJ. (1994). Shared control of maltose induction ancatabolite repression of the MAL structural genesSaccharomyces. Mol. Gen. Genet. 243, 622–630.

Zitomer, R. S. and Lowry, C. V. (1992). Regulatioof gene expression by oxygen in Saccharomyccerevisiae. Microbiol. Rev. 56, 1–11.

1144 . . .

? 1997 John Wiley & Sons, Ltd. . 13: 1135–1144

tandemly repeated 147 bp elements cause structural and functional variation in divergentmal...

Documents