Text of Gene Transcription in Saccharomyces cerevisiae
Vol. 173, No. 1
DAL82, a Second Gene Required for Induction of Allantoin SystemGene Transcription in Saccharomyces cerevisiaeMICHAEL G. OLIVE, JON R. DAUGHERTY, AND TERRANCE G. COOPER*
Department of Microbiology and Immunology, University of Tennessee,Memphis, Tennessee 38163
Received 3 July 1990/Accepted 9 October 1990
Several highly inducible enzyme activities are required for the degradation of allantoin in Saccharomycescerevisiae. Induction of these pathway enzymes has been shown to be regulated at transcription, and responseto inducer is lost in dal81 and dal82/durM mutants. The similar phenotypes generated by da)81 and dal82mutations prompted the question of whether they were allelic. We demonstrated that the DAL81 and DAL82loci are distinct, unlinked genes situated on chromosomes IX and XIV. DAL82 gene expression did not respondto induction by the allantoin pathway inducer or to nitrogen catabolite repression. Expression was also notsignificantly affected by mutation of the daI80 locus. From the nucleotide sequence of the DAL82 gene, wededuced that it encodes a protein with a mass of 29,079 Da that may possess the structural motifs expected ofa regulatory protein. This protein was shown to be required for the function mediated by the cis-actingupstream induction sequence situated in the 5'-flanking regions of the inducible allantoin pathway genes.
Expression of several allantoin pathway genes (DAL4,DAL7, and DUR1,2) in Saccharomyces cerevisiae increases10- to 20-fold when compounds that can be degraded to theinducer, allophanate, are added to the culture medium (4, 10,36, 39). A similar response is caused by the gratuitous induceroxalurate (30). In contrast, when cells are provided with areadily used nitrogen source, such as asparagine or glutamine,mRNA levels drop precipitously whether or not inducer ispresent, a phenomenon termed nitrogen catabolite repression(4, 10, 25, 36, 39). cis-acting elements required for basal-levelgene expression, response to inducer, and nitrogen cataboliterepression have been shown to be situated upstream of theinducible DAL7 transcribed sequences, suggesting that regula-tion of the gene occurs at transcription (38).The DAL7 cis-acting elements were found to be of three
types (38). The first type of element, a nitrogen-regulatedupstream activation sequence (UASNTR), was originally re-ported to be responsible for DALS gene expression (26). Thiselement, when present in two or more copies, supportshigh-level, inducer-independent, nitrogen catabolite repres-sion-sensitive gene expression (6, 26, 38). A functional GLN3gene has been shown to be required for UASNTR function (5).Saturation mutagenesis of the element was completed re-cently (2b), and the results suggest that its core sequence,5'-GATAA-3', is very similar in structure to chicken Eryf-1,mammalian GF-1 binding sites, a cis-acting site upstream ofthe Neurospora crassa qa-JF gene, and a footprinted regionupstream of the rat growth hormone gene (1, 9, 18, 24, 33, 35).The second element was shown to possess the properties of anegatively acting upstream repression sequence (URS). Whena DNA fragment containing the URS element was placedadjacent to a DAL UASNTR-containing fragment, it mediatedcomplete inhibition of transcriptional activation (38). Theheterologous CYCJ UAS was affected by the DAL7 URS inthe same way (38). The URS-mediated inhibition of transcrip-tional activation was independent of inducer. A third elementwas shown to be required for response to inducer, theupstream induction sequence (UIS). When this element was
* Corresponding author.
placed adjacent to the UAS and URS elements, the construc-tion (UIS-UAS-URS) supported normal oxalurate-mediatedinduction of P-galactosidase synthesis (38).Three mutant classes with apparent defects in the induc-
tion process have been reported. We isolated a set ofrecessive mutations (dal8J) (34) that resulted in completeloss of response to induction at both the enzyme andsteady-state mRNA levels (10, 34, 39). The dal8J locus waslocated on the right arm of chromosome IX by conventionalgenetic mapping methods (33). The DAL81 gene has beencloned and sequenced, and its expression was shown not torespond to induction by oxalurate or to nitrogen cataboliterepression (2, 2a). Mutations at the dal80 locus, which resultin 10- to 20-fold overproduction of allantoin pathway en-zymes, were also shown not to significantly affect DAL81expression. A dal81 null mutant, generated by deleting thegene, possessed the same physiological characteristics as thepoint mutant originally characterized (2). Mutation of asecond locus, durMIdal82, was reported to affect urea ami-dolyase activity and DURl,2-specific mRNA levels in amanner similar to dal81 (15). The gene was cloned but notsequenced or characterized (16). The third class of mutants,durL, also exhibited a phenotype similar to dal8J mutantsbut have not been characterized (15).The similarity of phenotypes generated by the dal81 and
da182 mutations raises the possibility that they are allelic. Ifthe two loci are distinct, the lack of information aboutexpression of the DAL82 gene, its potential regulation byinduction, by nitrogen catabolite repression, or in responseto mutation at the dal80 locus, and the identity of thecis-acting elements upstream of the DAL7 gene which re-quire a wild-type DAL82 protein to function make it difficultto elucidate its role in the induction process. Therefore, wetested the potential allelism of the dal8J and da182 muta-tions. Finding that they were not allelic, we isolated andsequenced the DAL82 gene, characterized its expression,and identified the cis-acting elements upstream of the DAL7gene that require its product for function.(A preliminary report of this work has already appeared
FIG. 1. Restriction map of plasmids used to test complementa-tion of the da182-1 mutation. The circled B designates a destroyedBamHI site. Hatched areas represent vector sequences.
MATERIALS AND METHODS
Strains and media. The strains used in this work are listedin Table 1. Strain M02 was made by selecting for a ura3mutant allele of strain 13H9b following treatment with 5-flu-oroorotic acid. Strain M1689-13d was derived from a cross ofstrains 13H9b and M1-2b. Genetic manipulation and transfor-mation were accomplished by standard procedures (12-14).
Isolation of plasmids able to complement mutations inDAL82. Mutations at the da182 (durM) locus were originallyisolated in an 11278b genetic background (strain 13H9b),which is not easily transformed. Therefore, strain 13H9bwas crossed to strain M1-2b, which transformed at highfrequency, to generate strain M1689-13d. This strain wastransformed with 10 ,ug of DNA derived from strain S288Ccarried in shuttle vector YCp5O (28). Ura+ transformantswere selected. Top agar containing these transformants wasscraped from the plates and macerated, and the cell suspen-sion was filtered. The cells were recovered, washed withsterile minimal medium devoid of a nitrogen source, andplated on YNB-allantoin plates containing tryptophan but nouracil. After 3 days of incubation at 30°C, the colonies thatappeared were selected for further analysis.DNA sequence analysis. DNA sequence analysis was con-
ducted by the dideoxy chain termination method reportedearlier (2b). A combination of synthetic oligonucleotideprimers and mung bean exonuclease III-generated nested
TABLE 2. Complementation of the da182-1 mutation by variousplasmid-borne DNA fragmentsa
a Plasmids YCp5O, YEp24, and pPB14 have been described previously (2).The plasmid indicated was used to transform mutant strain M02, which itselfwas isogenic to strain Y.1278b except at the URA3 locus. Transformants weregrown to a cell density of 30 Klett units in glucose-proline (Pro) minimalmedium. At that time, each culture was split into two portions. One portionreceived oxalurate (OXLU, 0.25 mM final concentration), while the other didnot. Following growth to a cell density of 60 Klett units, each culture wassampled in duplicate and assayed for the urea amidolyase (UALase) activityit contained by previously described procedures (36). UALase activities areexpressed as picomoles per minute per milliliter of culture.
deletions were used in the sequencing strategy. Both strandswere completely sequenced, and all sites were crossed.Northern (RNA) blot analysis. Strains used for Northern
blot analysis were grown to mid-log phase in Wickerham'smedium containing the indicated nitrogen source at a con-centration of 0.1% (37). Glucose (0.6%) was provided as acarbon source. Samples (10 jLg) of polyadenylated[poly(A)+] RNA were prepared (3) and resolved on a 1.4%agarose-6% formaldehyde gel by the procedures of Sumradaand Cooper (31). Following electrophoresis, the separatedspecies were transferred to a nylon membrane. The blot washybridized with the 1-kb BglII-BamHI fragment from plas-mid pMO8 that was radioactively labeled by random-primedlabeling (Boehringer Mannheim Biochemical Inc.). The blotwas then stripped and hybridized a second time to DNAfrom plasmid pTCM3.2, radioactively labeled by random-primed labeling, and the 2.5-kb HindlIl fragment fromplasmid pHY3 carrying the DAL7 gene.
TABLE 1. Strains usedStrain Relevant genotype
S. cerevisiae11278B .............MATox13H9b ............. MATa da182-1M02 ............. MATa ura3 da182-1M1-2b ............. MATax ura3-52 trpl-289M1689-13d .........MATa ura3-52 trpl-289 da182-1M1682-19b .........MATa ura3-52 trpl-289PB200. MATa ura3-52 trpl-289 /dal81::hisGM970 . MA Tot lys5M 9 0 ................
MATa Iys2M1081 . MATot Iys2 dal80-1M 1...............
MATa lysS dal80-1AM100...MATa his4-42 lys-23 +
MATot + + metl3-25
AM2O1 . MATa his442 lys23 gln3-1 +MA Tot + + gln3-1 met13-25
Plasmids containing sequences that complement dal82 mu-tations. The procedures described in Materials and Methodswere used to isolate two plasmids (pMO8 and pMO9) whichwere able to complement a dal82 mutation at high frequencyand contained common DNA fragments following digestionwith endonucleases BamHI and EcoRI. Plasmids pMO8 andpMO9 contained 10- and 11-kb inserts, respectively. PlasmidpMO8 was selected for further analysis. To test whetherplasmid pMO8 sequences complemented the biochemicalphenotype of the dal82 mutation (loss of urea amidolyaseinducibility), transformants of strain M02 were grown in theabsence or presence of oxalurate and assayed for ureaamidolyase activity (36). These transformants containednearly four times the activity exhibited by a wild-type strainof the same genetic background (11278b) and far moreactivity than the recipient strain (MO2) transformed withonly the cloning vector (YCp5O) (Table 2).
Structure of the putative DAL82-containing plasmid. Thestructure of plasmid pMO8 was determined (Fig. 1). Sub-cloning experiments localized the dal82-complementing(growth phenotype) sequences to a 2.9-kb BglII-EagI frag-ment (plasmid pMOll, Fig. 1). To test whether plasmidpMOll was able to complement the biochemical phenotype
ALLANTOIN SYSTEM INDUCTION IN S. CEREVISIAE 257
of the da182 mutation, we assayed transformants containingplasmid pMOll in the presence and absence of oxalurate.The subclone was able to restore urea amidolyase inducibil-ity (compare strain M02 containing plasmids pMOll andYEp24, Table 2). However, the fully induced level of activ-ity observed was only half that observed with plasmidpMO8, suggesting that a portion of the complementingsequences may have been removed.
Test for allelism between the dai8) and dal82 loci. Theobservation that dal8J and dal82 mutants had the samephenotype raised the possibility that mutations carried in thetwo strains were allelic. To test this possibility, we trans-formed a da182 mutant strain (MO2) with plasmid pPB14,which contained the wild-type DAL81 gene, and assayed thetransformants for urea amidolyase (plasmids YEp24 andpPB14 in Table 2). No complementation of the biochemicaldefect of strain M02 was observed. A more positive indica-tion that the two loci were distinct was obtained by identi-fying the chromosomal location of the dal82-complementingsequences carried on plasmid pMO8. This was accomplishedby hybridizing radioactive DNA derived from plasmidpMO12 against yeast chromosomes that had been resolvedby pulsed-field electrophoresis. DNA carried on plasmidpMO12 hybridized only to one chromosome which wasidentified, by comparison with a standard, as chromosomeXIV (data not shown). The DAL81 gene has been previouslyshown to be situated on the right arm of chromosome IX.These data suggested that the dal8I and da182 loci wereindeed distinct even though mutations in them generatedindistinguishable phenotypes.
Expression of the DAL82 gene and its regulation. Regulatedexpression of the DAL82 gene could play a central role incontrol of the allantoin system structural genes. Thisprompted us to assess the levels of DAL82-specific mRNAderived from wild-type, dal8J, and dal80 mutant strainsgrown under various physiological conditions. Cultures ofwild-type strain M970 grown in nonrepressing medium (glu-cose and proline) either without or with inducer or inglucose-asparagine medium (a condition of high nitrogencatabolite repression) exhibited similar levels of DAL82-specific mRNA (Fig. 2, lanes A to C).DAL7- and TCMI-specific probes were used as controls to
visualize the expected behavior of an allantoin pathway gene(DAL7) and one that had been demonstrated previously notto respond to oxalurate-mediated induction or nitrogen ca-tabolite repression (TCMI). DAL82-specific mRNA levelsincreased approximately threefold (quantitation obtained bydensitometric scanning of the X-ray film; data not shown) ina dal80 mutant strain compared with the wild type (Fig. 2,lanes A and E). There was also threefold less DAL82-specificmRNA in dal80 cultures grown in glucose-asparagine me-dium than in dal80 cultures grown in glucose-proline medium(Fig. 2, lanes D and E). We observed a 1.5- to 2-fold increasein the levels of DAL82-specific mRNA in a dal8J mutantcompared with the wild type (Fig. 2, lanes F to I). Finally,mutation of the gln3 locus resulted in a 30% decrease inDAL82-specific mRNA (Fig. 2, lanes J and K). These effects,though measurable, are not nearly as marked as normallyobserved with the allantoin pathway genes.
Nucleotide sequence analysis of the DAL82 gene. We deter-mined the nucleotide sequence of the DAL82 gene as de-scribed in Materials and Methods. The sequence of the geneand its flanking regions (3,012 bp) is shown in Fig. 3.Computer analysis (Genetics Computer Group sequenceanalysis package) resulted in identification of a single openreading frame encoding a 255-amino-acid protein with a
W.T. dal 80I _. I I _ _--o0o z z oE ic C
ft1 "i * 1-I
A B C D E
W.T. da8l.x : cmm-- 1-
F G H I J K
FIG. 2. Expression of the DAL82 gene in wild-type and mutantcultures grown under various physiological conditions. Lanes A toE, Poly(A)+ RNA prepared from wild-type (W.T., M970) and daI80mutant (M1081) strains grown in minimal glucose-proline (PRO) orglucose-asparagine (ASN) medium. Oxalurate was provided at afinal concentration of 0.25 mM where indicated (+). The RNAspecies were resolved on a 1.4% agarose-formaldehyde gel andtransferred to a nylon membrane. They were then hybridized to the960-bp BamHI-BglII fragment of pMO10 labeled by the random-primed DNA-synthetic method. After generation of the autoradio-graphs, the blot was stripped and hybridized a second time withsimilarly labeled DNA containing TCMJ (plasmid TCM3.2) or DAL7(2.5-kb HindlII fragment of plasmid pHY3 ). Lanes F to K,Poly(A)+ RNA prepared from wild-type (W.T., M1682-19b, lanes Fand G; AM100, lane J), daI81 deletion (pB200, lanes H and I), andgln3 mutant (AM201, lane K) strains grown as described above. TheNorthern blots generated from these RNA preparations were ana-lyzed as described for lanes A to E above. Sample size in each lanewas 10 ,ug.
calculated mass of 29,079 Da and a predicted isoelectricpoint of 7.7. This information suggested that the insert inplasmid pMOll contained the entire coding sequence of theDAL82 gene but lacked all of the sequences thereafter,presumably including the transcriptional termination andpoly(A) addition sequences. This information may explainthe decreased level of urea amidolyase induction observedwhen plasmid pMOll was used to complement the da182mutation (Table 2).Comparison of the DAL82 protein with those in the data
bases did not reveal the presence of a zinc finger, cyclicAMP-dependent phosphorylation sites, a nucleotide-bindingsite, or obvious helix-turn-helix or helix-loop-helix motif (8,11, 17, 22, 23). The initial 100 amino acids were found tohave a net charge of + 10, while residues 12 to 61 had a netcharge of +14. Residues 51 to 60, RTLKTKFRRL, exhib-ited some similarity to a nuclear localization signal (20, 21,27). A serine-rich (31%) region was observed between resi-dues 100 and 150, with a string of serine resdues (9 of 12) atpositions 100 to 112. Two potential glycosylation sites wereobserved at positions 98 and 130, respectively. There wasalso a region at the carboxy terminus of the protein (residues211 to 246) that could potentially be folded into an amphi-pathic helix resembling the coiled-coil motif reported bySorger and Nelson (29).
cis-acting site at which the DAL82 gene product functions.Previous studies by Yoo and Cooper (38) identified threefunctionally distinct cis-acting elements in the 5'-flankingregion of the inducible DAL7 gene. An important issue in ourunderstanding of the biochemical events that regulate allan-toin pathway gene expression and the role of the DAL82gene product in this control mechanism was identification ofthe cis-acting element(s) whose function requires the DAL82protein. To address this question, we transformed wild-type
,". 4mso 40m*4 **04D40I
VOL. 173, 1991
258 OLIVE ET AL.
-1910* -1890* -1870*ACG GCC GTT GAT GAC CGA ATG CAT TGG TTC ACT GAT TCA GAT GTG GAA GAA CAA TCG CAL
-50* -30* -10*CAT TGA TAA CAL CAT GGT TTA ACG ATA CTT TGA AAG GTT TAG GCG GAG CCA ACC ATA ATG
FIG. 3. Nucleotide sequence of the DAL82 gene.
(M1682-19b) and da182 mutant (MO2) strains with plasmidsthat contained the DAL7 UAS (plasmid pHY129), the UASand URS (plasmid pHY135), or all three elements (plasmidpHY174). ,B-Galactosidase production was then measured inthese transformants grown in the absence or presence ofinducer. The detailed structure of the plasmids used in thisexperiment has been described previously (38).High levels of P-galactosidase production were supported
by plasmid pHY129 in a wild-type strain (Fig. 4). The level ofactivity observed in the da182 mutant was lower by 35 to 50%(Fig. 4). Plasmid pHY135 did not support 3-galactosidaseproduction in either strain. This was the expected result,because plasmid pHY135 was used to determine whether theURS element functions and the da182 mutant did not exhibitthe phenotype predicted for loss of a negatively actingfactor. A quite different result was observed when plasmidpHY174 was used in this experiment. Plasmid pHY174supported 20- to 25-fold-greater ,B-galactosidase productionin the wild-type strain grown in the presence of oxaluratethan in its absence. In the da182 mutant, no induction wasobserved. Moreover, the uninduced level of lacZ expressionwas 10-fold depressed compared with that in the wild type(Fig. 4).The observations just described are consistent with the
suggestion that the DAL82 gene product is required foroperation of the DAL7 gene UIS element. However, it ispossible that a similar phenotype would be generated ifinducer was excluded from the cell. To assess this possibil-ity, we repeated the above experiment and used arginine asan inducer in place of oxalurate. Arginine is a more repres-sive nitrogen source than proline in this strain (unpublishedobservations), which accounts for the 3.5-fold-lower level of3-galactosidase production supported by plasmid pHY129 in
a wild-type strain grown in glucose-arginine medium than inglucose-proline (Fig. 4). A similar response was observed forthe da182 mutant, i.e., the mutant and wild-type strains didnot respond significantly differently. When plasmid pHY174was used in this experiment, induction was again completelylost. In this instance, inducer (arginine degraded to allopha-nate within the cell) exclusion cannot be suggested as anexplanation for loss of response in the da182 mutant strain.
The results of this work suggest that increased transcrip-tion observed in response to addition of inducer, a centralcharacteristic of allantoin system gene regulation in S.cerevisiae, requires a wild-type DAL82 gene product as well
VOL. 173, 1991 ALLANTOIN SYSTEM INDUCTION IN S. CEREVISIAE 259
+10* +30* +50*GAT GAA TCG GTG GAT CCT GTG GAG CTG CTT CTA CGA CTA CTG ATA CGG CAC AAA CCT CATAsp Glu Ser Val Asp Pro Val Glu Leu Lou Lou Arg Lou Lou Ile Arg His Lys Pro His
+70* +90* +110*CTG AAA CCA TAT GCC TAC AGA CAA GAT AGC TGG CAA AGG GTG CTC GAT GAG TAC AAC AGALou Lys Pro Tyr Ala Tyr Arg Gln Asp Ser Trp Gln Arg Val Lou Asp Glu Tyr Asn Arg
+130* +150* +170*CAG ACT GGG TCA AGA TAT AGA CAA TCA AGG ACG TTA AAA ACC AAA TTT CGT CGA CTG AAGGln Thr Gly Ser Arg Tyr Arg Gln Ser Arg Thr Lou Lys Thr Lys Ph. Arg Arg Lou Lys
+190* +210* +230*GAC CTC TTC AGC GCA GAT CGA GCC CAA TTC TCT CCT TCC CAG TTG AAG CTG ATG GGA GCAAsp Lou Ph. Ser Ala Asp Arg Ala Gln Ph. Ser Pro Ser Gln Lou Lys Lou Met Gly 'Ala
+250* +270* +290*CTC TTG GAC GAA GCA CCA GAA CAT CCA AGA CCA AGA ACT AAA TTC GGA AAT GAA TCA TCTLou Lou Asp Glu Ala Pro Glu His Pro Arg Pro Arg Thr Lys Ph. Gly Asn Glu Ser Ser
+310* +330* +350*TCA TCC TTA TCA TCA TCT TCT TTC ATT AAA AGT CAT CCG GGG CCT GAT CCG TTT CAA CAASer Ser Lou Ser Ser Ser Ser Ph. Ile Lys Ser His Pro Gly Pro Asp Pro Ph. Gln Gln
+370* +390* +410*TTA TCA TCC GCT GAA CAT CCG AAT AAC CAC AGC TCC GAC GAT GAG CAT TCA GGC TCA CAALou Ser Ser Ala Glu His Pro Asn Asn His Ser Ser Asp Asp Glu His Ser Gly Ser Gln
+430* +450* +470*CCG CTG CCC CTG GAT TCA ATA ACG ATT GGA ATT CCG CCT ACT CTT CAC ACA ATC CCC ATGPro Lou Pro Lou Asp Ser Ile Thr Ile Gly Ile Pro Pro Thr Lou His Thr Ile Pro Met
+490* +510* +530*ATT CTG TCT AAG GAT AAC GAC GTC GGG AAA GTC ATC AAA AGC CCT AAG ATA AAC AAG GGTIle Lou Ser Lys Asp Asn Asp Val Gly Lys Va1 Ile Lys Ser Pro Lys Ile Asn Lys Gly
+550* +570* +590*ACA AAT AGG TTC AGC GAG ACA GTA CTG CCT CCA CAA ATG GCT GCT GAG CAA TCG TGG TCGThr Asn Arg Ph. Ser Glu Thr Val Lou Pro Pro Gln Met Ala Ala Glu Gln Ser Trp Ser
+610* +630* +650*GAC TCT AAT ATG GAA TTG GAA ATA TGT CTA GAT TAT CTT CAC AAC GAA CTC GAG GTG ATAAsp Ser Asn Met Glu Lou Glu Ile Cys Lou Asp Tyr Lou His Asn Glu Lou Glu Val Ile
+670* +690* +710*AAG AAA AGG CAA GAA GAT TTT GAG TGT AAA GTT TTA AAC AAG CTC AAC ATA ATT GAG GCTLys Lys Arg Gln Glu Asp Ph. Glu Cys Lys Val Lou Asn Lys Lou Asn Ile Ile Glu Ala
+730* +750* +770*CTC CTT TCA CAG ATG AGA CCA CCC AGC CAA GGA GAT AAA ATA TAA AAA CTT CTA TTA GATLou Lou Ser Gln Met Arg Pro Pro Ser Gln Gly Asp Lys Ile End
+790* +810* +830*ATG CTT GAT TCA TGT CCA TAT GTA TCT ATT TAT ACA AAC ATT ACG TAA TAT ACA CAG ATT
+850* +870* +890*ATA CAT GAA AGG TGC TCT TCA TAT TTG GCT TCT TGC TTC CAC TAG CCT ATC CCA ACC AGG
+910* +930* +950*AGA GTC ATC CCA TCT GCC TTT GCC TTT ACT CAT TAG AGC TTT GAT AGT AGA TCT ACA TTT
as the DAL81 product documented previously (2, 33). Both extent, the property that they do not greatly (twofold or less)of these proteins appear to participate in the regulatory affect transcriptional activation mediated by the DAL systemfunction mediated by the DAL system UIS element. The UASNTR element.dal81 and da182 mutants also share, to a greater or lesser Although dal81 and da182 mutants are quite similar phe-
FIG. 4. Effect of a da182 mutation on the function of DAL7 cis-acting elements. Plasmids containing one or more of the DAL7 cis-actingelements were transformed into either wild-type (W.T.) strain M1682-19b or da182 mutant strain M02. Following growth in minimal medium(Difco yeast nitrogen base), glucose-proline medium (PRO), glucose-proline medium containing 0.25 mM oxalurate (PRO + OXLU), orglucose-arginine medium (ARG), the cells were harvested and assayed for ,-galactosidase activity (38). The plasmids used in this experimenthave been described in detail earlier (38). Coordinates indicated in the figures are occupied by these sequences in the upstream region of theDAL7 gene. Solid, checkered, and wavy-lined areas indicate regions of the UASNTR, URS, and UIS elements, respectively.
260 OLIVE ET AL.
notypically, they exhibit several distinct properties suggest-ing that their cognate products carry out different functionsin the induction process. The first distinguishing character-istic was observed in the complementation experimentsperformed during cloning. Complementation of daI81 muta-tions by a plasmid-borne gene was always found to be poorregardless of the plasmid copy number (2). In no case wasthe level of induced urea amidolyase activity in a dal81mutant strain carrying a complementing plasmid ever greaterthan the wild-type level (2). This was true even when a dal81deletion was used in the experiment to avoid potentialcomplications caused by negative complementation (2a). Incontrast, complementation of a da182 mutation by a plasmid-borne gene resulted in induced levels of urea amidolyase thatwere nearly fourfold greater than observed in a wild-typestrain (Table 2).The second distinct difference in the characteristics of the
dal81 and da182 mutations was their effect on basal-levelactivity observed in transformants containing plasmidpHY174. This plasmid contains all three cis-acting elementsidentified in the DAL7 upstream region. The uninduced levelof activity supported by this plasmid is not significantlyaffected by deletion of the DAL81 gene (Fig. 8 in reference2). In contrast, uninduced levels were decreased 10- to15-fold in a da182 mutant. The molecular basis for thesedifferences is not known at present, but they provide aframework within which to study the functions performed bythe proteins encoded by these genes.The data presented in this work and that published earlier
concerning the DAL81 gene suggest that the DAL80, DAL81,and DAL82 genes are not organized in a regulatory cascadesimilar to that suggested for other metabolic systems. Al-though the two- to fourfold effects of daI80, dal8J, and gln3mutations on expression of DAL82 cannot be categoricallydismissed as insignificant, they are not of the magnitude thatwould be expected of a cascade regulatory circuit. There-fore, our current working hypothesis is that all four geneproducts regulate allantoin gene expression but functionprimarily by modulating expression of pathway structuralgenes rather than the regulatory genes. The suggestion thatthe DAL81 and DAL82 gene products function in parallelrather than as a cascade in no way precludes the possibilitythat they may interact to mediate the induction process,especially since both gene products have been shown to berequired for the UIS-mediated function (Fig. 4) (also see Fig.8 in reference 2).
We thank Francine Messenguy and Evelyn Dubois for providingstrain 13H9b, on which this work so heavily depended. We thankThomas Cunningham, who developed the pulsed-field electrophore-sis methods in our laboratory and assisted in the experimentpresented in this work. We also appreciated the efforts of theUniversity of Tennessee yeast group, who read the manuscript andoffered suggestions for its improvement.The oligonucleotides used in this work were provided by the
University of Tennessee Molecular Resource Center. This work wassupported by Public Health Service grant GM-35642.
REFERENCES1. Baum, J. A., R. Geever, and N. Giles. 1987. Expression of qa-IF
activator protein: identification of upstream binding sites in theqa gene cluster and localization of the DNA-binding domain.Mol. Cell. Biol. 7:1256-1266.
2. Bricmont, P. A., and T. G. Cooper. 1989. A gene product neededfor induction of allantoin system genes in Saccharomycescerevisiae but not for their transcriptional activation. Mol. Cell.
Biol. 9:3869-3877.2a.Bricmont, P., J. R. Daugherty, and T. G. Cooper. Mol. Cell.
Biol., in press.2b.Bysani, N., J. Daugherty, R. Rai, and T. G. Cooper. Submitted
for publication.3. Carlson, M., and D. Botstein. 1982. Two differentially regulatedmRNAs with different 5' ends encode secreted and intracellularforms of yeast invertase. Cell 28:145-154.
4. Cooper, T. G., V. T. Chisholm, H. J. Cho, and H. S. Yoo. 1987.Allantoin transport in Saccharomyces cerevisiae is regulated bytwo induction systems. J. Bacteriol. 169:4660-4667.
5. Cooper, T. G., D. Ferguson, R. Rai, and N. Bysani. 1990. TheGLN3 gene product is required for transcriptional activation ofallantoin system gene expression in Saccharomyces cerevisiae.J. Bacteriol. 172:1014-1018.
6. Cooper, T. G., R. Rai, and H. S. Yoo. 1989. Requirement ofupstream activation sequences for nitrogen catabolite repres-sion of the allantoin system genes in Saccharomyces cerevisiae.Mol. Cell. Biol. 9:5440-5444.
7. Courchesne, W. E., and B. Magasanik. 1988. Regulation ofnitrogen assimilation in Saccharomyces cerevisiae: roles of theURE2 and GLN3 genes. J. Bacteriol. 170:708-713.
7a.Daugherty, J., M. Olive, and T. G. Cooper. 1990. 15th Int. Conf.Yeast Genet. Mol. Biol., 21-26 July 1990, The Hague, TheNetherlands.
8. Edelman, A. M., D. K. Blumenthal, and E. G. Krebs. 1987.Protein serine/threonine kinases. Annu. Rev. 56:576-613.
9. Evans, T., and G. Felsenfeld. 1989. The erythroid-specific tran-scription factor Eryfl: a new finger protein. Cell 58:877-885.
10. Genbauffe, F. S., and T. G. Cooper. 1986. Induction andrepression of the urea amidolyase gene in Saccharomycescerevisiae. Mol. Cell. Biol. 6:3954-3964.
11. Higgens, C. F., I. D. Hiles, G. P. C. Salmond, D. R. Gill, J. A.Downie, I. J. Evans, I. B. Holland, L. Gray, S. D. Buckel, A. W.Bell, and M. A. Hermodson. 1986. A family of related ATP-binding subunits coupled to many distinct biological processesin bacteria. Nature (London) 323:448-450.
12. Hinnen, A. H., J. B. Hicks, and G. R. Fink. 1978. Transforma-tion of yeast. Proc. Natl. Acad. Sci. USA 75:1929-1933.
13. Hoffman, C. S., and F. Winston. 1987. A ten-minute DNApreparation from yeast efficiently releases autonomous plasmidsfor transformation of Escherichia coli. Gene 57:267-272.
14. Ito, H., Y. Fukada, K. Murata, and A. Kimura. 1983. Transfor-mation of intact yeast cells treated with alkali cations. J.Bacteriol. 153:163-168.
15. Jacobs, E., E. Dubois, C. Hennaut, and J. M. Wiame. 1981.Positive regulatory elements involved in urea amidolyase andurea uptake induction in Saccharomyces cerevisiae. Curr.Genet. 4:13-18.
16. Jacobs, E., E. Dubois, and J. M. Wiame. 1985. Regulation ofurea amidolyase synthesis in Saccharomyces cerevisiae: RNAanalysis and cloning of the positive regulatory gene DURM.Curr. Genet. 9:333-339.
17. Johnston, M. 1987. A model fungal regulatory mechanism: theGAL genes of Saccharomyces cerevisiae. Microbiol. Rev. 51:458-476.
18. Kemper, B., P. D. Jackson, and G. Felsenfeld. 1987. Protein-binding sites within the 5' DNase I-hypersensitive region of thechicken acd_globin gene. Mol. Cell. Biol. 7:2059-2069.
19. Mitchell, A. P., and B. Magasanik. 1984. Regulation of glu-tamine-repressible products by the gln3 function in Saccharo-myces cerevisiae. Mol. Cell. Biol. 4:2758-2766.
20. Moreland, R. B., G. L. Langevin, R. H. Singer, R. L. Gareea,and L. M. Hereford. 1987. Amino acid sequences that determinethe nuclear localization of yeast histone 2B. Mol. Cell. Biol.7:4048-4057.
21. Moreland, R. B., H. G. Nam, L. M. Hereford, and H. M. Fried.1985. Identification of a nuclear localization signal of a yeastribosomal protein. Proc. Natl. Acad. Sci. USA 82:6561-6565.
22. Murre, C., P. S. McCaw, and D. Baltimore. 1989. A new DNAbinding and dimerization motif in immunoglobulin enhancerbinding, daughter-less, MyoD and Myc proteins. Cell 56:777-783.
ALLANTOIN SYSTEM INDUCTION IN S. CEREVISIAE 261
23. Pabo, C. O., and R. T. Sauer. 1984. Protein-DNA recognition.Annu. Rev. Biochem. 53:293-321.
24. Plumb, M., J. Frampton, J. Wainwright, M. Walker, K. Mac-leod, G. Goodwin, and P. Harrison. 1989. GATAAG: a cis-control region binding an erythroid-specific nuclear factor witha role in globin and non-globin gene expression. Nucleic AcidsRes. 17:73-91.
25. Rai, R., F. S. Genbauffe, and T. G. Cooper. 1987. Transcrip-tional regulation of the DAL5 gene in Saccharomyces cerevi-siae. J. Bacteriol. 169:3521-3524.
26. Rai, R., F. S. Genbauffe, R. A. Sumrada, and T. G. Cooper.1989. Identification of sequences responsible for transcriptionalregulation of the allantoate permease gene in Saccharomycescerevisiae. Mol. Cell. Biol. 9:602-608.
27. Rhee, S. K., T. Icho, and R. B. Wickner. 1989. Structure andnuclear localization signal of the SKI3 antiviral protein of S.cerevisiae. Yeast 5:149-158.
28. Rose, M. D., P. Novick, J. H. Thomas, D. Botstein, and G. R.Fink. 1987. A Saccharomyces cerevisiae genomic plasmid bankbased on a centromere-containing shuttle vector. Gene 60:237-243.
29. Sorger, P. K., and H. C. M. Nelson. 1989. Trimerization of ayeast transcriptional activator via a coiled-coil motif. Cell59:807-813.
30. Sumrada, R. A., and T. G. Cooper. 1974. Oxaluric acid: anon-metabolizable inducer of the allantoin-degradative enzymesin Saccharomyces cerevisiae. J. Bacteriol. 117:1240-1247.
31. Sumrada, R. S., and T. G. Cooper. 1982. Isolation of the CAR]gene from Saccharomyces cerevisiae and analysis of its expres-
sion. Mol. Cell. Biol. 2:1514-1523.32. Tsai, S.-F., D. I. K. Martin, L. I. Zon, A. D. D'Andrea, G. G.
Wong, and S. H. Orkin. 1989. Cloning of cDNA for the majorDNA binding protein of the errythroid lineage through expres-sion in mammalian cells. Nature (London) 339:446-451.
33. Turoscy, V., G. Chisholm, and T. G. Cooper. 1984. Location ofthe genes that control induction of the allantoin-degradingenzymes in Saccharomyces cerevisiae. Genetics 108:827-831.
34. Turoscy, V., and T. G. Cooper. 1982. Pleiotropic control of fiveeucaryotic genes by multiple regulatory elements. J. Bacteriol.151:1237-1246.
35. West, B. L., D. F. Catanzaro, S. H. Mellon, P. A. Cattini, J. D.Baxter, and T. L. Reudelhuber. 1987. Interaction of a tissue-specific factor with an essential rat growth hormone genepromoter element. Mol. Cell. Biol. 7:1193-1197.
36. Whitney, P. A., T. G. Cooper, and B. Magasanik. 1973. Theinduction of urea carboxylase and allophanate hydrolase inSaccharomyces cerevisiae. J. Biol. Chem. 248:6203-6209.
37. Wickerham, L. J. 1946. A critical evaluation of the nitrogenassimilation tests commonly used in the classification of yeast.J. Bacteriol. 52:293-301.
38. Yoo, H. S., and T. G. Cooper. 1989. The DAL7 promoterconsists of multiple elements that cooperatively mediate regu-lation of the gene's expression. Mol. Cell. Biol. 9:3231-3243.
39. Yoo, H. S., F. S. Genbauffe, and T. G. Cooper. 1985. Identifi-cation of the ureidoglycolate hydrolase gene in the DAL genecluster of Saccharomyces cerevisiae. Mol. Cell. Biol. 5:2279-2288.