Gene, 109 (1991) 143-147 © 1991 Elsevier Science Publishers B.V. All rights reserved. 0378-1119/91/$03.50
The Saccharomyces cerevisiae ADEI gene: structure, overexpression and possible regulation by general amino acid control (Yeast; gene cloning; nucleotide sequencing; gene expression; acid phosphatase promoter; purine biosynthesis; recombinant
DNA) Andrey N. Myasnikov', Kestutis V. Sasnauskasb, Arvidas A. Janulaitisb and Mikhail N. Smirnov~ " Biological Institute, Leningrad University, Leningrad 198904 (U.S.S.R.) and b Institute of Applied Enzymology, Vilnius 2028 (Lithuanian) Tel. (0122)641889 Received by J. Marmur: 17 September 1989 Revised/Accepted: 15 April/21 April 1991 Received at publishers: 27 May 1991
The ADEI gene of the yeast Saccharomyces cerevisiae has been cloned by complementation of the adel mutation. The nucleotide sequence has been determined for the 918-bp coding region, 240-bp 5'-noncoding region and 292-bp 3'-noncoding region. The sequenced region includes a single large open reading frame coding for a protein of 306 amino acid (aa) residues. The promoter of the ADEI gene contains a copy of the 5'-TGACTC hexanucleotide, a feature characteristic of promoters under general aa control. Subsequent search of other published purine biosynthesis gene sequences revealed that all of them also contain general aa control signals in their promoter regions. An expression plasmid containing the ADEI coding region under control of the PH05 promoter produced N-succinyl-5-aminoimidazole-4-carboxamide ribotide (SAICAR) synthetase in yeast cells at a level of 40% of total cellular protein. One-step purification resulted in an almost homogeneous preparation of SAICAR synthetase.
The ADEI gene of the yeast S. cerevisiae codes for SAICAR synthetase (EC 22.214.171.124; 7.22), an enzyme of the purine biosynthesis pathway catalyzing the reaction of CAIR and aspartic acid with the formation of SAICAR. Mutations in this gene usually lead to red pigmentation of yeast cells (Fischer, 1969). This pigment is a polymerization Correspondence to: Dr. A.N. Myasnikov, Biological Institute, Leningrad University, Oranienbaumskoe shosse 2, Stary Peterhoff, Leningrad 198904 (U.S.S.R.) Tel. (812)257-5788; Fax (812)218-2822. Abbreviations: aa, amino acid(s); ADEI, yeast gene encoding Nosuccinyl5-aminoimidazole-4-carboxamide ribotide synthetase; Ap, ampicillin: bp, base pair(s); CAIR, 4-carboxy-5-aminoimidazole ribotide; kb, kilo-
product of aminoimidazole ribotide, the metabolic precursor of CAIR and SAICAR. Yeast vectors, containing the gene have the advantage of selecting transformants on the basis ofcolor. Previously, ADEI has been cloned by several groups (Dimock et al., 1984; Crowley et al., 1984). The aim of the present study was to determine the nt sequence of the ADEI gene and to overproduce the SAICAR synthetase through the use of the inducible PH05 promoter. Parts of base(s) or 1000 bp; nt, nucleotide(s); ORF, open reading frame; PAGE, polyacrylamide-gel electrophoresis; PEP, 2% peptone/2% glucose; P~, inorganic phosphate; S., Saccharomyces; SAICAR, N-succinyi-5-aminoimidazole-4-carboxamide ribotide; SC, 0.67% yeast nitrogen base (Difco)/2% glucose/appropriate aa at 100 #g/ml; SDS, sodium dodecyl sulfate; UAS, upstream activating sequence; wt, wild type; YEPD, 2% peptone/l% yeast extract/2% glucose; , designates plasmid-carrier state.
144 this work have been published in Russian (Myasnikov et al., 1986; Myasnikov and Smirnov, 1986; Sasnauskas et al., 1986).
EXPERIMENTAL AND DISCUSSION
(a) Cloning of the ADE1 gene The ADE1 gene has been cloned by complementation of the adel mutation. A gene library was constructed from the DNA of S. cerevisiae strain FH4C partially digested with Sau3A (the strain was provided by Dr. D. Botstein, Stanford University, Stanford, CA). The Sau3A fragments of 6-12 kb were inserted into the BamHI site of the pYEI vector (Sasnauskas et al., 1986). After transformation of the S. cerevisiae strain p63-DC5 (MATa, leu2-3,112, his311,15, adel; courtesy of Dr. D.A. Gordenin, Leningrad University, Leningrad, U.S.S.R.) several adenine-independent clones have been recovered. Plasmid DNA was isolated from these clones and used to transform Escherichia coil to Ap resistance. All plasmids recovered were able to transform the yeast strain p63-DC5 to Ade + phenotype. After analysis by restriction mapping, a 4-kb HindII! DNA fragment common to most of these plasmids was checked for its ability to complement the adel mutation and subcloned for further study. Fig. 1 illustrates the restriction map of the 4-kb DNA fragment that correlates well with the restriction maps of ADEI published previously (Dimock et al., 1984; Crowley and Kaback, 1984). The location of ADEI within the 4-kb insert was established after we observed that insertion of the IS/element at a site shown by the arrow in Fig. 1 permits the plasmid to complement a corresponding mutation (purC) of E. coli (obviously, by supplying a bacterial promoter). Similarly, insertion of IS2 was shown to create a promoter function (Pilacinski et al., 1977). (b) Sequencing of the AD£1 gene The 0.9-kb Hindlll-EcoRl and 1.1-kb EcoRl fragments of the 4-kb insert were subcloned in pUCI9 and then sequenced separately. The small DNA fragments for shotgun IS1 ADEI
Fig. I. Restriction map of a 4-kb yeast chromosome I fragment containing the ADEI gene. The positions are shown for B a m H l (B), EcoRI (R), Hindlll (H), Sail (S) and Xhol (X) sites. The position at which an ISl insertion leads to expression of the ADEI gene in E. coli is marked by an arrow. The shaded box corresponds to the sequenced region. The thin horizontal line represents the position of the ADEI coding region.
sequencing were generated by 4-bp-recognizing restriction endonucleases, BAL 31 exonuclease or ultrasonic treatment (Deininger, 1983), cloned in appropriate M 13-based vectors, and analysed by dideoxy technology (Sanger et al., 1977; 1981). Most of the sequence was established from six to eight independent clones, covering both strands. The junction of the two fragments at the EcoRl site has been verified by isolating overlapping Taql clones from the original 4-kb insert. The nt sequence of 1.5 kb covers an ORF of 306 aa, 235 bp of 5' and 320 bp of 3' noncoding region (Fig. 2).
(c) Analysis of the sequence There are two Met codons near beginning of the ADEI ORF. We conclude that the first AUG initiates translation as based on the following: (i) good agreement between the calculated and the experimentally determined Mr of SAICAR synthetase (Ostanin et al., 1989); (ii) the nt context around the first Met codon that is preceded by an A-rich sequence, typical of yeast genes; (iii)the second of the two candidate Met codons is closely preceded by an ATG that is not in frame. The conclusion about the position of the translation start site in ADEI was subsequently supported by experiments with expression of this gene under control of the PH05 promoter (see section e). The noncoding regions of ADEI contain severa; sequence elements that may have specific functions. A prominent feature ofthe 5'-flanking region is a 20 nt long stretch of alternating A and T at nt -133 to -154 relative to ATG. It has been proposed that such sequences may serve both as constitutive promoters and as 'stoppers' for the propagation of a signal from enhancer-like UAS elements (Struhl, 1984; 1985). The best candidate for the role of TATA box is probably the TATATATT sequence (nt -120 to -127). This element is located at a distance from the ATG codon that is typical for TATA elements of yeast genes but the close proximity of the oligo(dAT) sequence makes this assignment somewhat arbitrary. A sequence element 5'-TTTTATCTTTT (nt -86 to -76) matches the consensus for a 'capping box' (Corden et al., 1980). In the 3'-noncoding region sequence fragments 5 ' TATGT...TIT (66 and 98 bp downstream from the TAA codon) may serve as transcription termination signals (Zaret and Sherman, 1982), while two elements, TATAA -78 bp and AACAA -93 bp from the stop codon, resemble the sequence AATAA that is believed to be a polyadenylation signal (Benoist et al., 1980). The codon composition of the ADEI gene reflects a moderate expression level of the gene. The codon bias index calculated by the algorithm of Bennetzen and Hall (1982) is 0.4. This value is intermediate between a value of nearly 1.0 for highly expressed genes (PGK) and close to 0 for several transcription-activating factors that are expressed at a very low level.
-201 -101 -1 75 25 150 50
~ ARG ~ lIT
~ ASP ~ GLD
~ & 2.T ~ ZLE~.TER ~ ~ RSII S E R
CCT GAK AAG GGG KTC CTA TI'G &CC ltAK G'I'G TCK GAG 2TC TGG ~ ~ ~ ~ ~ = PRO GL~/LY8 GLY l"Ug l , ~ LEO TIIR LY8 Z,L'IIJ 8 n GLO , H g ~ PHZ LY8 , m e Z,L'O 5 1 ~ ~
~ GLU ~ ZLE ~ ~ ~ ~ ~ ASP V'AL ARG &SII RZ8
TI'G GTC VAL CitA ¢TA LEO
~ PRO ~ V&L
~ L~8 ~ &RG
T~ &SP ~ GLY
308 10 375 125
ATG TCA lint S l ~ GTJL GAC YAK ASP
KTT ZLE GCT ALA
GItC ]LSP GitA GLO
ACG TER GGT 6LY
AAG ACT LI[8 T E R JkCG ~ TER Llm
A2C ~ XLI~ GiiC CGC ASP MRG
CCA PRO TCT ~
GItA ¢ T G ~ LE0 C'I'G T'I'T LI~ PIIE
GI~ ~"T AT& A S P nv-2f ZLE G'JL'C GCT Jl~G VAL ALK Tlm
GGT AAG ACT GL2[ L~[8 TIIR CTA TI'G GTT LL'U ~ V&I[,
ATT TI'C Z v-~- ~B]~ CAC IUUL RZ8 L , 8
TTG ~ Git~ ASP
CCA PRO CGT ARG
GILT ~ItC &SP TER CAT ]UtA HZ8 LY8
~ GTG ~ ~L J~L'C ~ ILE
GCC ALK ~ ALK
~-~-~'~-~ ~ ~I.~ ~ ~ T Y R RSP
CTA CC~ GC& AA& LI~ AT,& L~8 ¢TA ATT ~ ~ LL'O ZLg ,RO LEO
~ ~ LY8 ~ ~ V A L ZLE
~ ~ ~ LI~ 81~ ~ ~A ~ GZ,O VAL z]r~
N=C GGI~ TCT GC'~ TGG AitA GAG T&C G~A AitA &r.A GGT &CT GTG C i ~ GG~ TI,G NU~ ~ ~ TIIR GLX ~ &LK TRIP Z,,5 GLU T~R VAIL, LY8 THR G L , ~ V&L RX8 G L , Z.L'0 L , 8 GLH
~ ~ L~[S T= ~ T'/R XLE
~ ~ ~ ~ ~ GLII G L , L l m L , 8
TCT r.AG GAG TI'~ CCK GAA CC& ATC ~'1'C ACC CCK TCG &CC IUtG GCT GA& CA& GGT GA& ~ ~ ~ ~ ~ ~ 81~ G]r,lf G][,D , m e ]PRO G ~ I , R O ZLg ,nl~ ~ ,gO 8]~ ~ L~8 AL& GLU GIN GL'Z GLU RZ8 ,liSP GLO AS~ z]rJ~ 81~
CC'Z GCC ClkG GCC ~]~/k GIIG C'I'G G'2'G GG'J~ ~GGi~xG ~ 2'~G ~ A C ~ AGA G'I'G ~ ~ ~ ~ PR0 ALI~ GLII &L& GLO Z.g0 V&L GL~ liSP LILIJ 8 g R MRG ARG VAL ~ J t . GLIJ LL'0 ALIt. E~'ORZ J U ~ T G C AKIk C ~ T TAT G e T A I ~ C J ~ ] U ~ ~ A T e A T e & T ~ G C K ~JkC A ~ T ~ ~ ~ ~ LY8 C'/8 LY8 &.qP ~ ~ LY8 GL0 L~[8 ZZ.E XLE ZLg ItlLA liSP ~ LY8 ~HI~ GL0 Prig JmaZ
~ ~ ~ T~ ~ VAL L~[8 L g 0 TMR
~ ~ ~ ~ ~ GL~ ZLg A8P GLU LY8
~,u a n ma u p AA~ GGC 6 T ~ Air% C~Z ~ T V A L v.28 lint . o AS~ GC~ ~ ARA ~ IP3T ~ GLY 8gR LY8 ~ 8ER iiZ8
a n Tx,~ a a p ..--v'~slmap,m z~m ~u~ aap m CJUk ~ o.. ~P
~ ~ ZZ~
um ~m ~
a u v.z8 ram ~um m,~ mm
A G G R E A A G G GO[: A A K T A T A T A 6 U ~ ~ ' T TAT G R A JtDK T T G ARG ~fl~ AP.G ~ L,8 ~ ZLg GLU T2R GMJ TER LL'O
TAK O ~ ' ( ] ~ . T I ~ T A P . K T A T A G ~ T A ~ % % G ~ G ~ I ' ~ ~ T ~ T I ' C A T A C K T A O ~ C G T A ' I ~ T A **~ 3 O 6
188 922 1092 1192
Fig. 2. Nucleotide sequence of the $. cerevisiae ADE! gene. The deduced aa sequence of SAICAR synthetase is also shown, and both the nt and aa are numbered from the first ATG codon of the large ORF. Asterisks denote the stop codon. The d(A,T),, run and TATA box are underlined. The nt sequence of the ADE! gene has been deposited with GenBank under accession No. 61209.
(d) General aa control signals in purine biosynthetic genes Besides the features mentioned above, the promoter region of ADEI contains a consensus sequence (TGACTC) recognized by the general aa control regulatory factor, the GCN4 protein (Hinnebush and Fink, 1983; Thireos et al., 1984). This finding prompted us to look for this element located at other genes in the adenine biosynthesis pathway. It was found that promoter regions of all the genes of this pathway sequenced so far - ADEI, ADE2 (Kraev et al., 1984), ADE3 (Staben and Rabinowitz, 1986), ADE4 (Mantsala and Zalkin, 1984) and ADES,7 (Henikoff, 1986) do indeed contain one or two copies of this consensus sequence (Fig. 3). In all cases examined the position of this sequence in the promoter (nt -130 to -350 upstream from the start codon) is compatible with the hypothesis that it may be involved in regulation of expression of these genes. As opposed to typical genes regulated by general aa control, which contain multiple copies of the 5'-TGACTC sequence, most purine biosynthesis genes have only one copy
ADEI ADE2 ADE2 ADE3 ADE4 ADES,7 ADES,7
&CGiS.GTCAGTC - 2 1 9 ~ , C T C T ' I ' G C G A G ~ A T G A~C~,~., - 192r~..CT C r l ~ _ . ~ ~ TATATTTGCCA AATCGATGTAT TTTTTTTCAGT CAAGTGCCGAC
- 213TGACTC - 357TGACTC - 217TGACTC - 183TGACTC
CTCCCAGTGACA TTCCTGACCGAA GCCCCGTCGGTA GTGTCCTGGTAA
Fig. 3. Representation of the GCN4-recognition signals in the promoter region of purine biosynthetic genes. The fragments of nt sequences of these promoters that contain the TGACTC signals are shown. The position of the general control consensus sequences is given relative to the ATG start codon.
of the sequence (but two copies in ADE2 and in ADES,7). This implies that aa starvation may have only small effect on the transcription levels of purine biosynthesis genes. Nevertheless, this finding suggests very strongly the involvement of a general aa control system in the regulation of purine biosynthesis. Also, it should be noted that ADE3 codes for C,-tetrahydrofolate synthetase, an enzyme involved not only in purine biosynthesis but in a number of biosynthetic pathways. Therefore, it is quite possible that the general aa control is in fact an even more general regulatory system that coordinates (or modulates) the expression of many anabolic genes. (e) Overexpression of the A D E ! gene and purification of SAICAR synthetase A multicopy yeast plasmid, pJDB (MSADIR), was constructed through a series of standard genetic engineering steps (not shown). The plasmid contains an expression cassette consisting of the PH05 promoter and terminator fragments from pMS46 (Ostanin et al., 1988) flanking the promoterless ADEI gene (with a deletion in the 5'-noncoding region starting at nt -40 upstream from the presumed ATG codon). This expression cassette was inserted into the high-copy-number shuttle vector, pJDB207 (Beggs, 1981). S. cerevisiae strain GRFI8-MA To~, leu2-3,112, his3-11,15, can R (kindly provided by Dr. A. Hinnen, Ciba-Geigy, Basel) was used as the host for transformation (Hinnen etal., 1978). When strain GRF18 [pJDB(MSADIR)] is grown in low Pi medium it accumulates SAICAR synthetase at a level of about 40% of total
kDa 67 -->
(Ostanin et al., i989). No N-terminal residue could be detected by the dansyl chloride method (Narita et al., 1975) using both native and SDS-denatured protein. A probable explanation of this fact is that the N terminus of SAICAR synthetase is acetylated.
Fig. 4. Electrophoretic analysis of SAICAR-synthetase-producing strain GRFI8[pJDB(MSADIRI and purified enzyme. Yeast cells grown in PEP medium were collected by low-speed centrifugation, washed with water and broken by shaking with glass beads in 2 vols. 2% SDS, containing 2 mM phenylmethanesulfonyl fluoride. After 15 min centrifugation at 15000 rpm in a Beckman Minifuge 11 table top centrifuge the supernatant was used for 0.1% SDS/12% PAGE (Laemmli, 1970). The gels, stained with Coomassie blue R-250, were scanned on Beckman DUB spectrophotometer and the scans were used for estimating SAICAR-synthetase production level. The purified SAICAR synthetase was obtained in two steps. First, the protein of the crude cellular extract was precipitated by ammonium sulfate (45-60~ of saturation); second, the precipitate was dialyzed in 20 mM Tris. HCI pH 8.0, batch adsorbed on DEAESephadex and eluted with the same buffer containing 0.2 M NaCI. Lanes: I, cellular extract of a control yeast strain GRFI8[pJDB207]; 2, cellular extract ofGRF! 8[pJDB(MSADIR)]; 3, purified preparation of SAICAR synthetase.
soluble protein (Fig. 4). SAICAR synthetase can easily be purified to homogeneity from the cellular extract of GRF 18[pJ DB(M SADI R)] by standard biochemical techniques. It should be noted that SAICAR synthetase is quite ubiquitous and it is an essential enzyme in lnany organisms. Nevertheless, no rich sources of this enzyme have been detected previously. The recombinant-plasmid-carrying strain GRFI8[pJDB(MSADIR)] provides a good source of SAICAR synthetase, thus making possible an extensive study of its enzymological properties. SAICAR synthetase crystallizes readily from an ammonium sulfate solution. The three-dimensional structure of this enzyme is now being investigated in the Institute of Crystallography (Moscow). SAICAR-synthetase activity was measured by a spectrophotometric method using CAIR as substrate (Ostanin et al., 1981). The purified SAICAR synthetase has a specific activity of about 90 units/mg, which is slightly higher than the previously reported value for the homogeneous enzyme
We appreciate the help of Drs. A. Hinnen, D. Botstein and D.A. Gordenin who provided us with the yeast strains. We thank Dr. V.D. Domkin who provided us with a unique preparation of chemically synthesized CAIR and also V.D. Domkin and Dr. K.V. Ostanin for helpful discussions on the properties of SAICAR synthetase. We acknowledge assistance by Dr. A.V. Sorokin who helped us with sequencing technology. Special thanks are due to Yu.A. Plavnik, V.S. Yakovleva and Yu.l. Kovaleva for skillful assistance in the experimental work.
REFERENCES Beggs, J.D.: Multiple-copy yeast plasmid vectors. In: yon Wettstein, D., Friis, J., Kielland-Brandt, M. and Standerup, A. (Eds.), Molecular Genetics in Yeast. Alfred Benson Symposium No. 16, Munksgaard, Copenhagen, 198 I, pp. 383-395. Bennetzen, J.L. and Hail, B.D.: Codon selection in yeast. J. Biol. Chem. 257 (1982) 3026-303 I. Benoist, C., O'Hare, K., Breathnach, R. and Chambon, P.: The ovalburain gene: sequence of putative control regions. Nucleic Acids Res. 8 (1980) 127-142. Corden, J., Waslyk, B., Buckwalder, A., Sassone-Corsi, P., Kedinger, C. and Chambon, P.: Promoter sequences ofeucaryotic protein coding regions. Science 209 (1980) 1406-1414. Crowley, J.C. and Kaback, D.B.: Molecular cloning of chromosome ! DNA from Saccharomyces cerevisiae: isolation of the ADEi gene. J. Bacteriol. 159 (1984) 413-417. Deininger, P.: Random subcloning of sonicated DNA: application to shotgun DNA sequence analysis. Anal. Biochem. 129 (1983) 216-223. Dimock, K., James, A.P. and Seligny, V.C.: Molecular cloning of the ADEi gene of Saccharomyces cerevisiae and stability of the transformants. Gene 29 (1984) 233-237. Fischer, C.R.: Enzymology of the pigmented adenine-requiring mutants of Saccharomyces and Schizosaccha,omyces. Biochem. Biophys. Res. Commun. 34 (1969) 306-310. Gibson, T.J.: Studies on Epstein-Barr Virus Genome. Ph.D.Thesis, University of Cambridge, U.K., 1984. Henikoff, S.: The Saccharomyces cerevisiaeADES,7 protein is homologous to overlapping Drosophila melanogaster Gart polypeptides. J. Mol. Biol. 190 (1986) 519-528. Hinnebush, A.G. and Fink, G.R.: Repeated DNA sequences from HISI also occur at several other co-regulated genes in Saccharomyces cerevisiae. J. Biol. Chem. 258 (1983) 5238-5247. Hinnen, A., Hicks, J.B. and Fink, G.R.: Transformation of yeast. Proc. Natl. Acad. Sci. USA 75 (1978) 1929-1933. Kraev, A.S., Mironova, M.V. and Sasnauskas, K.V.: Yeast ADE2 gene: primary structure and recombination mapping. In: Abstracts of the 16-th FEBS Conference. Moscow, 1984, p. 382.
147 Laemmli, U.K.: Cleavage of structural proteins during the assembly of the head of bacteriophage "1"4.Nature 227 (1970) 680-685. Mantsala, P. and Zalkin, H.: Nucleotide sequence of Saccharomyces cerevisiae ADE4 gone encoding the glutamine-phosphoribosylpyrophosphate amidotransferase. J. Biol. Chem. 259 (1984) 8478-8484. Myasnikov, A.N. and Smirnov, M.N.: Purine biosynthesis genes contain the signals of general amino acid control. Biopolimery i Kletka 2 (1986) 31-32 (in Russian). Myasnikov, A.N., Plavnik, Yu., A., Sasnauskas, K.V., Gedminene, G.K., Janulaitis, A.A., Smirnov, M.N.: Nucleotide sequence of Saccharomyces cerevisiae ADEI gone. Bioorgan. Khim. 12 (1986) 555-558 (in Russian). Narita, K., Hisayuki, M. and Nakajima, T.: End group determination. In: Needleman, S.B. (Ed.), Protein Sequence Determination. Springer Verlag, Berlin, 1975, pp. 30-103. Ostanin, K.V., Domkin, V.D. and Smirnov, M.N.: Spectrophotometric method for assaying the activity of SAICAR-synthetase (EC 126.96.36.199) and properties of the yeast enzyme. Chem. Abstr. 98 (1981) 1342s (Deposited doc. ViNITI). Ostanin, K.V., Smirnova, T.M., Myasnikov, A.N., Gedvilaite, D.A., Sasnauskas, K.V. and Smirnov, M.N.: Expression vectors based on promoters of PH03 and PHO5 genes: construction, comparison of effectiveness, use for superexpression of phosphoribosylaminoimidazolo-carhoxylase. Biopolimery i Kletka 4 (1988) 259-266 (in Russian). Ostanin, K.V., Alenin, V.V., Domkin, V.D. and Smirnov, M.N.: Isolation and properties of phosphoribosylamino-imidazole succinocarboxamid¢ synthetase of yeast Saccharomyces cerevisiae. Biokhimyia 54 (1989) 1265-1271 (in Russian).
Pilacinski, W., Mosharrafa, E., Edmundson, R., Zissler, J., Fiandt, M. and Szybalski, W.: Insertion sequence IS2 associated with int-constitutive mutants of bacteriophage lambda. Gone 2 (1977) 61-74. Sanger, F., Nicklen, S. and Coulson, A.R.: DNA sequencing with chainterminating inhibitors. Prec. Natl. Acad. Sci. USA. 74 (1977) 546-567. Sanger, F., Coulson, A.R., Barrel, C.T., Smith, A.J.H. and Roe, B.A.: Cloning in single stranded bacteriophages as an aid to rapid DNA sequencing. J. Mol. Biol. 143 (1981) 161-178. Sasnauskas, K.V., Gedminene, G.K., Markevicius, A.I., Naktinid, V.J. and Janulaitis, A.A.: Cloning and physical mapping of ADEl gone of Saccharomyces cerevbiae. Genetika 22 (1986) 549-556 (in Russian). Smith, H.O.: Recovery of DNA from gels. Methods Enzymol. 65 (1980) 371-379. Staben, C. and Rabinowitz, J.C.: Nucleotide sequence of the Saccharomyces cerevisiae ADE3 gone encoding C~-tetra-hydrofolate synthase. J. Biol. Chem. 261 (1986) 4629-4637. Struhl, K.: Genetic properties and chromatin structure of the yeast GAL regulatory element. Proc. Natl. Acad. Sci. USA. 81 (1984)7865-7869. Struhl, K.: Constitutive and regulatory promoter: evidence for two distinct mechanisms. In: Hicks, J. (Ed.), Molecular Biology of Yeast. Cold Spring Harbor Laboratory, New York, 1985, pp. 361-366. Thireos, G., Penn, M.D. and Greer, H.: 5'-sequences are required for the translational control of a yeast regulatory gene. Proc. Natl. Acad. Sci. USA 81 (1984) 5096-5100. Zaret, K.S. and Sherman, F.: DNA sequence required for efficient transcription termination in yeast. Cell 18 (1982) 563-573.