Gene, 90 (1990) 87-92 Elsevier
87
GENE 03532
I s o l a t i o n o f a novel p r o t e i n k i n a s e - e n e o d i n g g e n e f r o m y e a s t by o l i g o d e o x y r i b o n u c l e o t i d e probing (Recombinant DNA; serine-threonine kinase; Saccharomyces cerevisiae)
David G.L. Jones* and John Rosamond
Department of Biochemistry and Molecular Biology. Universityof Manchester. Manchester Mi 3 9PT (U.K.) Received by 5.(3. Oliver: 25 August 1989 Revised: 8 October 1989 Accepted: 15 February 1990
SUMMARY
We have identified a novel protein kinas~encoding gene, KIN3, in the genome of the budding yeast Sacckaromyces cerevisiae. The gene was isolated from a library of cloned genomic fragments by probing with an oligodeoxyribonucleotide mixture corresponding to part of a highly-conserved region in the catalytic domain of protein serine-threonine kinases. KIN3 is unique in the yeast genome, maps to chromosome VI and is actively expressed in mitotically dividing cells to produce a 1400 nucleotide (nt) message. The nt sequence of KIN3 predicts a protein product of 43.4 kDa which contains all of the conserved elements found in known protein serine-threonine kinases, although the organisation of these elements in the KIN3 gene product differs significantly from the consensus. The function of the KIN3-encoded protein kinase is unclear although it appears not to be essential for growth, conjugation or sporulation.
INTRODUCFION
Protein kinases (KINs) play important roles in regulating a variety of cellular processes (reviewed in Hunter and Cooper, 1985). Nearly 100 such enzymes have now been identified as members of the KIN family, usually on the basis of the presence of certain key residues within the
Correspondence to: Dr. J. Rosamond. Dept. of Biochemistry and Molecular Biology, University of Manchester, Oxford Rd.. Manchester M 13 9PT (U.K.) Tel. 061-275-5100 or 061-275-5103; Fax 061-275-5082. * Present address: Microbiological Institute, Swiss Federal Institute of Technology, CH-8092 Zurich (Switzerland) Tel. 4 I- 1-2562211.
Abbreviations: aa, amino acid(s); bp, base pair(s); BSA, bovine serum albumin; eDNA, DNA complementary to RNA; kb. kilobase(s) or 1000bp; KIN, gene encoding KIN; KIN, protein kinase; nt, nucleotide(s); OFAGE, orthogonal field agarose-gel electrophoresis; oligo, oligodeoxyribonucleotide; ORF, open reading frame; on'. origin of DNA replication; PA, polyacrylamide; S., Saccharomyces; Sc., Schizosaccharo. myces; SDS, sodium dodecyl sulfate; SSC, 0.15M NaCI/0.015M Ha3' citrate pH 7.6; TMAC, tetramethylammonium chloride; X, any amino acid. 0378-1119J90/$03.50 © 1990Elsevier Science Publishers B.V. (Biomedical Division)
catalytic domain where several short stretches of aa are highly conserved in these enzymes (Groffen et ai, 1983; Hanks et ai., 1988; Kamps et al., 1984). In the budding yeast Saccharomyces cerevbiae, KINs ~tre involved in metabolic regulation (Celenza and Carlson, 1986; Roussou et al., 1988), antibiotic resistance (Boguslawski and Polazzi, 1987), cell-type specialisation (Teague et al., 1986) and cell-cycle control (Reed et al., 1985; Patterson et al., 1986). In this last case, the characterisation of the CDC28 and CDC7 gene products as KINs suggests that progress through G 1 phase to S phase may be achieved or regulated by sequential activation of KINs and phosphorylation of key target proteins (Patterson et al., 1986; Mendenhall et al., 1987). Since the CDC7 gene product differs from all other known KINs in the organisation of the conserved aa within its catalytic domain (Patterson et al., 1986; Hanks et al., 1988), temporal and functional coordination of this component of the mitotic cell cycle might then be achieved as a consequence of structural variations in the catalytic domains of the various KINs. To examine this possibility, we have sought in the first instance to identify other KIN genes that might encode
88
components of this putative sequential phosphorylation pathway. Our approach to this was similar to that adopted to screen for Ser-Thr KIN genes in a HeLa cell cDNA library (Hanks, 1987) and putative Tyr KIN genes in yeast (Levin et al., 1987). In both cases, degenerate oligo probes were used to recognise target sequences that code for short, highly conserved regions in the KIN catalytic domain. The aim of the present study was to use an oligo mixture to isolate and characterise a novel yeast KIN gene whose product has an unusual catalytic domain structure.
o)
ealHl-a
Oao).,1
v a n n !- • p s
(xal).,z
i, a n n L K P q . . ( 1 6 9 )
..ttsa)
(alm)..Y
v n n n L • p A ..(zso)
(laS).,v
!, n n n v g P t
(130)..V
V H R D L • 1~ I1 . . ( 3 3 8 )
Cmme~t~lUJ
tO
..(le6)
- - H R D L • P -
1
2 MATERIALS AND METHODS
kb
----4.4
Escherichia coil HW87 was used for the routine maintenance and propagation of plasmids (Patterson et ai., 1986); strain JM103 was used for the preparation of MI3 clones for DNA sequencing (Messing et al., 1981). Restriction analysis, cloning in E. coli and transformation of S. cerevisiae strain SB303 (Mata[Mat~ his/his ura3-52/ ura3-52 trpl-289/trp1-289) were as described previously (Patterson et al., 1986).
RESULTS AND DISCUSSION "---
1.1
¸
0.5--
Fig. I. Design and characterisation of the degenerate olign probes. (a) Comparison of a conserved subdomain of five Ser-Thr KINs and the derived degenerate probe sequence for the consensus. The sequences shown in standard IUPAC single-letter en code are SNFI and CDC28 from $, cerevbiae (Celenza and Carlson, 1986; Lorincz and Reed, 1984), weel *, from ~chizosaccharomyces oombe (Russell and Nurse, 1987), CDC2Hs, the human homologne ofSc.pombe cdc2 + (Lee and Nurse, 1987) and CaMiI.% the ~-subunit of a calcium/calmodulin-dependent KIN from rat (Henley et al., 1987). The numbers in parentheses to the left and right oftbe en sequence are the number ofresidues to the N and C terminus, respectively. The olign probe mixture was synthesised by the phosphoramidite method using an ABI380 synthesiser, and purified by electrnphoresis through 25% PA. (b)Hybridisation of the degenerate oligo probe to yeast genomic DNA. An EcoRV digest of yeast genomic DNA was resolved by electrophoresis through 1% agarose, transferred to nitrocellulose (Southern, 1975) and probed using oligns that had been labelled with J2p by polynucleotide kinase and [~.32PlATP (Grunstein and Hogness, 1975; Maxam end Gilbert, 1977). Filters were prehybridised for 6-8h at 37°C in a solution containing 6 × SSC/0.1% Ficoll/0.2% polyvinylpyrrolidone/0.2~o BSA/0.5% SDS/15% formamide, then hybridised for 18-24 h at 20°C in a similar solution from which SDS was omitted and to which labelled olign probe was added.
(a) Design and eharaeterisation of oligo probes The design of the oligo probes was based on a short aa sequence within the catalytic domain of five Ser-Thr KINs with diverse cellular functions (Fig. la). The consensus sequence for this region is His-Arg-Asp,Leu.Lys-Pro (HRDLKP in single-letter code), which is highly conserved in these proteins with the Asp residue thought to interact with the phosphate groups of ATP via salt bridges (Brenner, 1987). This subdomain also differs significantly from the equivalent sequence in mammalian Tyr KINs where Lys is usually replaced by Ala or Arg, and Pro is replaced by Ala (Hanks et al., 1988). To optimise the potential hybridisation frequency of probes based on this sequence, a mixture of oligos containing 17 nt was synthesised, corresponding to 32 of the possible codons that could give rise to the HRDLKP sequence. Codon selection was based on utilisation frequencies in yeast protein-coding sequences. The HRDLKP-encoding oligo mixture was used initially to screen a blot of yeast genomic DNA that had been digested with EcoRV, and the filter washed in TMAC at
4 After hybridisation, filters were washed four times for 5 rain each in ice-cold 6 × SSC, then twice for 45 min in 3 M TMAC/S0 mM Tris. HC! pH 7,8/2 mM EDTA/0.1% SDS at 46°C (lane I) or 53°C (lane 2). These conditions select specifically for hybridising sequences containing long contiguous matches (Wood et al., 1985). After washing, filters were autoradiographed overnight at -80 ° C using two X-Omat intensifying screens.
89
46°C. TMAC eliminates the preferential melting of A-T versus G-C pairs, allowing the stringency of hybridisation to be controlled as a function of probe length. The conditions used here are equivalent to a reduced stringency wash requiring a 14-bp contiguous match between probe and target (Wood et ai., 1985). On this basis, sequence comparisons indicate that the HRDLKP probe should hybridise to five of the known yeast KIN genes, including CDC28 which was used in the design of the probe (Fig. la). In practice, the probe hybridised to about ten discrete genomic fragments that range in size from 1.1-10.5 kb (Fig. lb, lane 1) suggesting that the olign probe mixture could be identifying a limited number of previously unidentified KINs. Under conditions of maximum stringency imposed by washing with TMAC at 53°C, the probe hybridised only to two genomic fragments of 1.1 kb and 4.4 kb (Fig. lb, lane 2). (b) Isolation of KIN3 The HRDLKP=encoding oligo pool was used to screen a library of :yeast genomic DNA fragments in the vector YRpI2 (Barker and Johnson, 1983) for which the filters were washed in TMAC at 46°C. Positive sign,s were detected from nine of about 5000 colonies (approximately 3-4 genome equivalents ofS. cerevisiae) and plasmid DNA was prepared from each of these nine clones. Preliminary analysis revealed that one of these plasmids carried the CDC7 gene, while two others carlied the oligo-hybridising
12 kb
g'.t
345
6
•
4.a
•
Q
2"0 I
Q 8
0.S
Fig. 2. Isolation and characterisation of a clone hybridising to the HRDLgP.encoding olign. A yeast genomic library in the vector YRpl2 (Barker and Johnson, 1983) was screened by colony hybridisation using labelled oligns as described in the legend to Fig. lb. Plasmid DNA which hybridised to the olign was isolated from one clone and designated pC3. pC3 plasmid DNA was digested with (lanes): Accl (1), BamHl (2), EcoRV (3), Pstl (4), Sail (5), and TaqI (6), and the fragments resolved through a 1% agarose gel. The DNA was transferred to nitrocellulose and probed with the HRDLKP-encoding pool as described in the legend to Fig. lb.
sequence in different putative intergenic regions that were closed in all reading frames (data not shown). Two other clones have not been characterised further. Surprisingly, the same yeast genomic f r ~ e n t was present in four of the remaining clones, one of which, pC3, was characterised initially by digesting the plasmid DNA with a variety of restriction enzymes and probing the fragments with the HRDLKP-encoding oligo pool (Fig. 2). Of the DNA fragments hybridising to the probe, those produced by EcoRV (1.2 kb), Accl (1.1 kb) ~Jad Taql (0.6 kb) were small enough for further characterisation. In addition, by comparing the size ofthe pC3 DNA fragments that hybridised to the HRDLKP-encoding probe with restriction maps derived from nt sequence data for known yeast KIN, it was deduced that the genomic fragment of pC3 did not correspond to any of the previously identified KIN genes. The 600-bp Taql fragment of pC3 carrying the sequences responsible for hybridisation was subeloned into the Accl site of Ml3mpl8 and sequenced. Within the nt sequence obtained for this fragment was the element, 5'-CATCGTGATCTGAAACC-3', which is a perfect match with one member of the oligo probe pool. In addition, the sequence contained a single incomplete ORF that was capable of encoding a peptide carrying the HRDLKP motif. This ORF was termed KIN3. (e) KIN3 encodes a protein kinase The complete nt sequence of the KIN30RF was determined together with 500 bp of upstream and 370 bp of downstream sequence, using the deduced restriction map of this region ofthe pC3 clone (Fig. 3a). This ORF extends for 1107 bp and is capable of coding for a protein of 43.4 kDa whose deduced aa sequence is shown in Fig. 3b. Comparison of the aa sequence deduced from the KIN3 nt sequence with three yeast KINs involved in cell-cycle control (CDC28), cell-type specialisation (STET) and carbon catabolite repression (SNF1) revealed that, in addition to the HRDLKP element, the KIN3 protein contains all of the conserved elements found within the catalytic domains of KINs (Fig. 4). These include the sequence Gly-X-Gly-XX-Giy that is thought to interact with the ribose and terminal pyrophosphate moieties of ATP (Sternberg and Taylor, 1984), and which lies 15-25 aa residues N terminal of an invariant Lys which appears to be directly involved in the phosphotransfer reaction (Kamps and Sefton, 1986); the triplet Asp-Phe-Gly which represents the most highly conserved sequence within the catalytic domain; and the consensus elements Ala-Pro-Glu and Asp-X-Trp-Ser-X-Gly that are frequently used as indicators of KIN activity. While KIN3 contains all of the conserved elements characteristic of the KIN catalytic domain, the organisation of these elements differs markedly from all other KINs. This is seen most noticeably in the spacing between the
90 a
AAAAACCGCAAATGTGACACCGTCCCTCAGTATTACTCTAGAGGGCTTAATGCCATMTA LysAs.GlyLysC¥:;AspThrVa I I'roGluTyrTyrSerArGGI y L e u A s . k l a I | e l l e
1471{ 320
CA'i'TCAATGATACATGTGA~AAGAACCAGC~CCAuI I ~11GAATTACTGCAA('.AT III sSe rMe L I I eAsl)Va | A$-LeuArGThrArfal'roSe rTh¢l'heG I uLeut.euG h~As p
1538 340
ATTCAGATACr-,AACTGCAAGAAAGTCGTTGCAATTAGAGAGATTTCAAAGGAGGTTACTC I IcG I n I I ©ArrjThrA I aArl;Ly5 S©|'Le uG I nLe uG I uArfjPheC I uArcArlILeuUe u
I -~98 360
GACTATGAAAATGAGCTC,ACAAACAIT[GAARAAATCCTCG AGAAGCAAGCTATTGAATA AspTyrG I uAsnC I uLeuThrAsn I le>>> 369
I fi58
98
CGAAAGAGAACTGAGTCAGTTGAAGGAACAATTTACC,CAGGCAGTGGAGGAGCGACCCAG
1718
GAGAT~CCA*~t, CAGTCCCCCGGCCACGCCCTGCCACCATACCCACGCCGAAACAAGC
|58
GGAAGTAATTAGCGGTAAGAAAGTTGGTAAGGTTCCAGAATCTATAAACGGATATTATGC
|778
CCTCATGAGCCCGAAGTGGCGAGCCCGATCTr¢CCCATCCGTGATGTCGGCGATATAGGC
218
TA AAAAATTTGCCAAACCTGCATACCACTGGCAAAC,ru~GATATCCATAAAATTTGCTCTG
1838
CCCAGAACCGACCTGTCGCGCCGGTGATGCCGGCCACf,AGGCGTCCGGCGTCGAGGATCA
278
AGAACATTTATACKi'rAATACATACGGCTATGTATGAATCATATTCAGTGACGAGTTACA
| 898
AT/U~TCr CJU~ATACGATATCTGACATGATTCCAGTGGCACAGTCTTCOGATACGTTACA
338
CTGGCGACTTGTAGTATATGTAAATCACGTrAGTGAGACCATTTAGKCCCTGTCAATGTT
195S
C C C ~ C u | t | I i u I i I I~AGG'I-i'rCTGAAI I I I | iCATTACGTTTCCTAA'i'rAGGTTAAA
398
T CATAAGCCrCTATATATTTGGCCCTTCTCCTGTCGAC
CGTA/~TA&AAGCGA~.t a"f"tCAAAAGAATATCAAATCATATAACATGAAAAGAGCAAGAT
458
TCACAATACTAATTTCAATGAGAGTCCTA'I'IWGATA ATTGAATCAATT CTCTGTGTCAGT
$18
ATGCATAGACGACAt, I I i I tCCAAGAATACCGTAGTCCTCAGCACCACCAGGGACACCCA M©t I I I s A rGAr~,G I nl'hci'heG I .G I uTyrA r~Se r Pt'oG |nG I .C I .G I .G I yll i ~ I'ro
5?8 20
CCAAGGTCAGAATACCAAGTTCTCGAAGAAATTCCCAGAGGTrCATTTCGGTCTGTACG& I'roArgSc rG I uTyrG I nVa I Le uG I uC 1u I I eC I yAr/~G I ySe rl'heG lySc rV;, I At/;
638 40
Fig. 4. Comparison of the KIN3 catalytic domain with the equivalent regions in three other yeast KINs, CDC28 (Lorincz and Reed, 1984), SNFI (Celenza and Carlson, 1986) and STE7 (Teague et ai., 1986). The extent of noneonserved aa residues is indicated by numbers in parentheses. The numbers above the KIN3 sequence correspond to the location of the aa in the KIN3 protein (of. Fig. 3b).
A I
V '
T V T II I
AE Ill
TX II
T '
VC ¥
A I
2sobp I
I
b
GTCGACi;CTCTCCCTTATCCGACTCCTGCATTAGCAAG 38
CAGCCCAGTAGTAGGTTGAC,GCCGTTGAGCACCGCCGCCGCAAGCAATGGTGCATGCA/'.G
•
A
•
•
•
•
AAAGTCATCCATATACCTACCAAGAAACTI3"TGGTTAGAAACGATATCAAATATGGCCAT L.vsVa I I l e l l i s l lelaroThrLysLy,..LeuLeuV.'l IAr~LysAspl leLysTyrGlyll| ~.
GgR fi0
ATGAAT^GCAAAGAGAGACAACAGCTCATCGCTGAATGTAGCATTCTATCGCAGTTGAAG Me tAsnS©rLy~.GI uArr.G htGhlLe.I leAlaC| uCysSer I leLcu$crGI nLeuLys
758 80
CATGA~TATTGTAGAATTYrATAACI'GGGACTTCGATGAACAAAAAC&ACTGTTATAC Ill ~C I uAsni I eva IGI ul'heTyrA:mTrp&Hd'heA~pG hiG hd.ysG I ut.euL©uTyr
SI S
CTTTATATGGAATATTGTTCCAUGGGTGA'rTTATCCCAGATGATTAAGCACrACAAACAG LeuT~rMe t GI uTyrCysSe rAt'f,G lyA~pl.euSerG I rime t I I e L y s l i | ~ T y r l . y s G h t
HTS 120
CAGCATAAATATATACCAGAAAAAATTATATGGCCTATCCTGGCCCAAT'|'ATI'G^CTGCG GI n l l l i h ~ l * T y r | l e l ' r e C l u L p I I'el l e T r p G | y l It~LeuAluGh~Lq:uLeuThrAI.
93R 140
CTCTATAAATGTCATTATCGTG'rfGAM'TGCCAACFTTG^CCACAATATATGACCGG^TG LeuTyr L~'"Cysll I "Ty rGI yV~d G blue uProThr LcttTh r T h r I leTyrAsp^ri~let
; ~9~; , If~0
AAACCACCGGTAAAAGGCAAA^ACATCGTTATCCA'rCGTGATCTGAAACCAGGAAATATA Ly~l'rol'roVall.ysGlyl.y~A,..I leVai I lelli,-Arl;A.%~Le.bys h'nG yA~, 'd le
10$R 180
k
•
A'&
1996
100
A ......
B
Vii XIII,
•
TTCTTAAGCTATGATGATAGCG&TTACA^TATIA,~TGAACAAGTAGACGGTCACGAGGAA IqmLeuSe rTyrAspAspSc rA~'I)T~'rA:-~I I eA~t~G IuG htV~t I A~,pGI :~lll ~,GhiG I u
I I 18 200
GTG&ATAGTAM'TA'I"rACAGAGACC^TAGAGTC,AATTCA~C,GAA^AGGG(;G^G~ATG V" I AS flSe rA{t nT]/r'i'yrArlIAtqdll sArl;Va I A:mSe I,GI yL¥~Ar~G | yea rl,|,oMu t
I I 7t4 220
GACTATAGTCAAGTTG',i'GGTAAAGTTACGTGATTTTGCGrTGACCC^AATCTCTGAAACT A~'l)T~r $0 rG h~Vtl I Vtl I Vu I I,ysLo tic I)'hs plqieG I .vl.etiSe rG h d I.eSe rGI uThr
1238 240
AGTATI'CAATTTGCCACAACATACGTCGGTACACCATATTACATGTCGCCf(;AAGTCTTG Set'l leGInlq~eAlaThrThrTyrVnlGlyThrl~roT:~rTyrHet$©rlh.oGhNalLeu
129H 2G0
ATGUACCAACCATAC;TCCCCACTATCCGACATCTGGTCACTAGCTTG.rGTTATTTTTGAG Me t ASlIGI nl*l'oTyr-~erltroLe"$e r a s p I I cYl'|)$elt'LauG I ~Cy~Vtl I I I el'hr~G I tt • • A • • • ATGTGTTCGTTCCACCCCCCATTTCAGGCA^^AAATT&TCTC(;AGCTACAAACTAAGATT Me t CysSe I'Leull I s I:roProl,heG h~A I aLy,*.A~nT:er Le ~G h,Le uG h~.rhr L¥~ i | e
13,~S |SO
14 | S 300
Fig, 3. Restriction analysis and nt sequence of|he pC3 clone hybrid|sing to the HRDLKP-encoding oligo. (n)Partial restriction map of the HRDLKP region of pC3. The very thick line represents the KIN30RF. A, Accl; (7., ClaI; E, EcoRl; T, Taql; V, Ec~RV; X, Xbal. (b)The nt sequence of KIN3 and deduced aa sequence. The nt sequence was deter. mined by the dideoxy chain-termination method (Sanger et aL, 1977) for which DNA fragments to be sequenced were cloned into Ml3mpl0, mpl8 or mpl9 (Yaniseh-Perron et ai., 1985). Reaction products were resolved through 67e PA gels under denaturing conditions and detected by autoradiography for 1-4 h at room temperature. The nt sequence corresponding to the ollgo probe is overlined. Conserved aa in the KIN cataiytie domain are indicated by triangles (see RESULTS AND DISCUSSION, section ¢), while the > > > symbol indicates the presumptive stop codon.
Fig. S. Mapping KIN3 by hybrid|sat|on to a yeast chromosome blot. Yeast chromosomes (numbered I-XYi) were prepared in agaros~ microbeads and resolved through 1% agarose by OFAGE as described by Schwartz and Cantor (1984). &~er electropboresis, chromosomes were detected by ethidinm bromide staining (panel A). The chromosomal DNA was then transferred to nitrocellulose and probed with the 600-bp Taql fragment from the KIN3 gene that had been 3=P-labelled by oligo priming (Feinberg and Vogelstein, 1983). ARer hybrid|sat|on (as described in the legend to Fig. I), filters were washed in 2 × SSC at room temperature for 30 rain, then at 65"C for 2h in 0.l x SSC/0.5% SDS with one change of wash solution. Autoradiography was at -80"C for 12-18 h with two X-Croat intensifying screens (panel B).
91
HRDLKP and DFG subdomains. In 38 known or putative Ser-Thr KINs (Hanks et al., 1988), this interval is highly conserved at 12-15 aa, whereas in KIN3 the conserved elements are separated by 51 aa (Fig. 4). One explanation for this could be the presence ofintrons in KIN3. However, the £1N3 sequence lacks any of the consensus exon-intron and 3' splice signal sequences (Teem et al., 1984) suggesting that these extra aa are present in the KIN3 protein. Thc effect that such a variation might have on KIN activity is '~known.
Four of the resulting URA + transformants were analysed by genomic blotting and were found to have acquired a disrupted KIN3 Iocu~ (data not shown). Al~er sporulation, haploid cells carrying the KIN3 disruption were found to be both viable and conjugation-proficient, while diploids in which both copies of KIN3 were inactivated underwent a normal meiosis. Therefore, the KIN3 gene product does not seem to play an essential role in mitosis, conjugation or meiosis. (e) Comelusions
(d) Characterisation of the KIN3 gene To determine whether the KIN3 gene was unique in the yeast genome or was a member of a family of closely related enzymes, the 600-bp TaqI fragment containing most of the KIN3 catalytic domain was used to probe an EcoRV digest of yeast genomic DNA under stringent and relaxed conditions. In both cases, a single hybridisation signal was obtained corresponding to a 1.1-kb EcoRV fragment, indicating that KIN3 is apparently unique in the yeast genome and is not closely related to any other gene at the level of nt sequence (data not shown). The 1. l-kb EcoRV fragment detected with the 600-bp Taql probe probably corresponds to the major l.l-kb DNA fragment detected by the HRDLKP-encoding oligo in genomic blots (Fig. lb, lane 2). To map KIN.~ within the yeast genome, the 600-bp Taql fragment was oligo primed and used to probe a blot of yeast chromosomes separated by pulsed-field electrophoresis; the blot was kindly provided by Dr. Paul Dickinson, Department of Pathology, University of Oxford, A single ~i~.al .~,~~ obtained from hybridisation of the KIN3 probe to chromosome VI (Fir, 5A,B); none of the previously characterised yeast KIN genes :x,:f~psto this chromosome. To determine whether KIN3 is expressed in ~,ct~x,clyproliferating cells, total RNA was isolated from a logarithmically growing culture of $. cerevisiae. RNA was electrophoresed under denaturing conditions, transferred to a nylon membrane and probed with the 600-bp Taql fragment. A single hybridisation signal was obtained corresponding to an RNA species ofabout 1400 nt (not shown), which is consistent with the size ofthe KIN30RF determmcd from the nt sequencing. In order to examine the cellular function of KIN3 and to determine if the product was essentia' for cell viability, the ~;00-bp TaqI fragment of KIN3 was cloned into the integ/~.ting vector YIp5 and the resulting plasmid used to transform S. cerevisiae strain SB303 to uracil prototrophy. Since Ylp5 lacks a functional ori, stable transformants only arise after integration of the plasmid DNA into the yeast genome, a process that occurs by homologous recombination (Hicks et al., 1979; Scherer and Davis, 1979). In this instance, recombination between the cloned Taql fragment and one of the chromosomal KIN3 loci results in a disruption of this locus.
(1) We have used an oligo pool to isolate a nova KIN gene from a library of yeast genomic sequences. The nt sequence of this gene, KIN3, predicts a protein product of 369 aa and a n M r of 43 435. This size is in good agreement with the size ofthe KIN3 gene transcript, while the ORF is flanked by sequences typically found adjacent to yeast ORFs. (2) The product of KIN3 is likely to be a KIN on the basis of the presence of short aa stretches that are characteristic of Ser-Thr KIN. However the deduced KIN3 aa sequence is unusual in that the interval between subdomains VI and Vll of the KIN catalytic domain is 51 aa in KIN3 but only 12-15 in all other known Ser-Thr KINs. Such modifications though do not necessarily prevent KIN activity (e.g., CDC7 KIN; Bahman et al., 1988). (3) The KIN3 protein lacks any of the distinguishing features ofTyr KINs, nor do regions of the protein outside the catalytic domain have homology to other KINs. (4) Cells carrying disrupted KIN3 genes undergo normal mitosis, meiosis and conjugation, suggesting that KIN3 is not essential for any of these processes. This behaviour is unlikely to be due to complementation by a closely related KIN since KLan3probes hybridise only to KIN3 sequences in yeast genomic blots even und,~r conditions of reduced stringency. Thus, the cellular role of KIN3 is |imit.ed to a nonessential or conditional function.
ACKNOWLEDGEMENTS We thank the rest of the group for discussions, helpful discussions and liquid refreshment; Tony Hunter for providing data prior to publication; Lee Johnston and Paul Dickinson for material assistance; and Margaret Barber for typing the manuscript. This work was supported by grants from S.E.R.C. and the University of Manchester Research Support Fund.
REFERENCES Bahman, A.M., Buck, V.M., White, A. and Rosamond, J.: Characterisation of the CDC7 gene product of SaccAarom~.s ceremiae as
92
a protein kinasc needed for the initiation of miotic DNA synthesis. Biochim. Biophys. Acts 951 (1988) 335-343. Barker, D.G. and Johnson, L.H.: Saccharomyces cerevisiae CDC9, a structural ~,~enefor DNA figase which complements $chizosaccharomyces pombe cJci?. European J. Biochem. 134 (1983) 315-319. Boguslawski, G. and Polazzi, J.O.: Complete nucieotide sequence of a gane conferring polymyxin B resistance on yeast: similarity of the predict ,d polypeptideto protein kinases. Proc. Nail. Acad. SCI.USA 84 (1987) 58~8-5852. Brenner, S.: Phosphotransferase sequence homology. Nature 329 (1987) 21. Celenza, J.L. and Carlson, M.: A yeast gene that is essential for release from 81ucose repression encodes a p~otein kinase. Science 233 (1986) 1175-1180. .,.~ Feinberg, A.P. and Vngelstein, B.: A te~mique for radiolabelling DNA restriction endonuclease fragment:, to high speei~c activity. Anal. Binchem. 132 (1983) 6-13. Groffen, J., Heisterkamp, J., Reynolds, F.H. and Stephenson, J.R.: Homology between phosphotyrosine acceptur site ofhnman c-abl and viral oncngene products. Nature 304 (1983) 167-169. Granstein, M. and Hoguess, D.S.: Colony hybridization: a method for the isolation of cloned DNAs that contain a specific gene. Proc. Natl. Acad. Sci. USA 72 (1975) 3961-3965. Hanks, $.K." Homology probing: identification of eDNA clones encoding members of the protein-serine kinase family, prnc. Natl. Acad. Sci. USA 84 (1987) 388-392. Hanks, S.K., Quinn, A.M. and Hunter, T.: The protein kinase family: conserved features and deducted phylogeny of the catalytic domain. Science 241 (1988) 42-52. Henley, R.M., Means, A.R., Ono, T., Kemp, B.E., Burgin, K.E., Waxham, N. and Kelly, P.T.: Functional analysis of a complementary DNA for the 50.kilodalton subunit ofcalmoduiin kinase 11. Science 237 (1987) 293-297. Hicks, J.B., Hinnen, A. and Fink, G.R.: Properties of yeast transformation. Cold Spring Harbor Syrup. Quant. Biol. 43 (1979) 1305-1314. Hunter, T. and Cooper, J.A.: Protein tyrosine kinases. Annu. Rev. Biochem. 54 (I 985 ) 897-930. Kemps, M.P. and Sefion, B.M.: Neither nrginine nor histidine can carry out the function oflyaine.295 in the ATP-binding site of p60sre. Mol, Cell. Biol, 6 (1986) 751-757. Kamps, M.P., Taylor, S.S. and Set,on, B,M." Direct evidence that onco8enic tyrosine kinases and cyclic AMP.dependent protein kinase have homologous ATP-binding sites. Nature 310 (1984) 589-592. Lee, M.G. and Nurse, P.: Complementation used to clone a human homologue ofthe fission yeast cell cycle control gene cdc2. Nature 327 (1987) 31-35. Levin, D.E., Hammond, C.I., Ralston, R.O. and Bishop, J.M.: Two yeast genes that encode unusual protein kinases. Proc. Natl, Aead. Sei. 84 (1987) 6035-6G.~9. Lorincz, A.T. and Reed, S.!.: Prime~2,structure homology bet~,,een the
product of yeast cell division control gene CDC28 and vertebrate oncngenes. Nature 307 (1984) 183-185. Maxam, A.M. and Gilbert, W.: A new method for sequencing DNA. Proc. Natl. Acad. Sci. 74 (1977) 560-564. Mendenhall, M,D., Jones, C.A. and Reed, S.I.: Dual regulation of the yeast CDC28-p40 protein kinase complex: cell cycle, pheromone and nutrient limitation effects. Cell 50 (1987) 927-935. Messing, J., Crea, R. and Seeburg, P.: A system for.shntgun DNA sequencing. Nucleic Acids Res. 9 (1981) 30~-321. Patterson, M., Sclafani, R.A., Fangman, W.L. and Rosamond, J.: Molecular characterisatinn of cell cycle gene CDC7 from Saccharomyces cerevisiae. Mol. Cell. Biol. 6 (1986) 1590-1598. Reed, S., Hadwiger, J. and Lorincz, A.T.: Protein kinase activity associated with the product of the yeast cell division cycle gene CDC28. Proc. Natl. Acad. Sci. USA 82 (1985) 4055-4059. Roussou, I., Thireos, G. and Hange, B.M.: Transcriptional-translational regulatory circuit in Saccharomyces cerevisiae which involves the GCN4 transcriptional activator and the GCN2 protein kinase. Moi. Cell. Biol. 8 (1988) 2132-2139. Russell, P. and Nurse, P.: Negative regulation of mitosis by weal *, a gene encoding a protein kinase homologue. Cell 49 (1987) 559-567. Sanger, F., Nickle~, S. and Coalson, A.R.: DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. SCI. USA 74 (1977) 5463-5467. Scherer, S. and Davis, R.W.: Replacement ofchromosome segments with altered DNA sequences constructed in vitro. Proc. Natl. Aead. Sci. USA 76 (1979) 4951-4955. Schwartz, D.C. and Cantor, C.R.: Separation ofyeast chromosome-sized DNAs by pulsed field gradient gel electrophnresis. Cell 37 (1984) 67-75. Southern, E.M.: Detection of specific sequences among DNA fragments separated by gel ele©trophoresis. J. Mol. Biol. 98 (1975) 503-517. Sternb~r8, MJ.E. and Taylor, W.R.: Modelling the ATP-bindin8 site of oncogene products, the epidermal growth factor receptor and related proteins. FEBS Lett. 175 (1984) 387-392. Teem, J., Abovich, N., Kanfer, N., Schwindioger, W., Warner, J., Levy, A., Woolford, J., Leer, R., Van Raamsdonk-Duin, M., Maser, W., Plants, R.~ Schultz, L., Friesen, J., Fried, H. and Rosbach, M.: A comparison ofyeast ribosomal protein 8ene DNA sequences. Nucieie Acids Res. 12 (1984) 8295-8312. Teague, M.A., Chaleff, D.T. and Errede, B.: Nucleotide sequence of the yeast regulatory gene ST£? predicts a protein homologous to protein kinases. Proc. Natl. Aead. Sci. USA 83 (1986) 7371-7375. Wood, W.L, Gitschier, J., Lasky, L.A. and Lawn, R,M.: Base composition-dependent hybridization in tetramethylammonium chloride: a method for oligunucleotide screening of highly complex 8ene libraries. Proe. Natl, Acad. Sci, USA 82 (1985) 1585-1588. Yanisch-Perron, C., Vieira, J. and Messing, J.: Improved MI3 phage cloning vectors and host strains: nucleotide sequences of the Ml3mpl8 and pUCI9 vectors. Gene 33 (1985) 103-119.
87
GENE 03532
I s o l a t i o n o f a novel p r o t e i n k i n a s e - e n e o d i n g g e n e f r o m y e a s t by o l i g o d e o x y r i b o n u c l e o t i d e probing (Recombinant DNA; serine-threonine kinase; Saccharomyces cerevisiae)
David G.L. Jones* and John Rosamond
Department of Biochemistry and Molecular Biology. Universityof Manchester. Manchester Mi 3 9PT (U.K.) Received by 5.(3. Oliver: 25 August 1989 Revised: 8 October 1989 Accepted: 15 February 1990
SUMMARY
We have identified a novel protein kinas~encoding gene, KIN3, in the genome of the budding yeast Sacckaromyces cerevisiae. The gene was isolated from a library of cloned genomic fragments by probing with an oligodeoxyribonucleotide mixture corresponding to part of a highly-conserved region in the catalytic domain of protein serine-threonine kinases. KIN3 is unique in the yeast genome, maps to chromosome VI and is actively expressed in mitotically dividing cells to produce a 1400 nucleotide (nt) message. The nt sequence of KIN3 predicts a protein product of 43.4 kDa which contains all of the conserved elements found in known protein serine-threonine kinases, although the organisation of these elements in the KIN3 gene product differs significantly from the consensus. The function of the KIN3-encoded protein kinase is unclear although it appears not to be essential for growth, conjugation or sporulation.
INTRODUCFION
Protein kinases (KINs) play important roles in regulating a variety of cellular processes (reviewed in Hunter and Cooper, 1985). Nearly 100 such enzymes have now been identified as members of the KIN family, usually on the basis of the presence of certain key residues within the
Correspondence to: Dr. J. Rosamond. Dept. of Biochemistry and Molecular Biology, University of Manchester, Oxford Rd.. Manchester M 13 9PT (U.K.) Tel. 061-275-5100 or 061-275-5103; Fax 061-275-5082. * Present address: Microbiological Institute, Swiss Federal Institute of Technology, CH-8092 Zurich (Switzerland) Tel. 4 I- 1-2562211.
Abbreviations: aa, amino acid(s); bp, base pair(s); BSA, bovine serum albumin; eDNA, DNA complementary to RNA; kb. kilobase(s) or 1000bp; KIN, gene encoding KIN; KIN, protein kinase; nt, nucleotide(s); OFAGE, orthogonal field agarose-gel electrophoresis; oligo, oligodeoxyribonucleotide; ORF, open reading frame; on'. origin of DNA replication; PA, polyacrylamide; S., Saccharomyces; Sc., Schizosaccharo. myces; SDS, sodium dodecyl sulfate; SSC, 0.15M NaCI/0.015M Ha3' citrate pH 7.6; TMAC, tetramethylammonium chloride; X, any amino acid. 0378-1119J90/$03.50 © 1990Elsevier Science Publishers B.V. (Biomedical Division)
catalytic domain where several short stretches of aa are highly conserved in these enzymes (Groffen et ai, 1983; Hanks et ai., 1988; Kamps et al., 1984). In the budding yeast Saccharomyces cerevbiae, KINs ~tre involved in metabolic regulation (Celenza and Carlson, 1986; Roussou et al., 1988), antibiotic resistance (Boguslawski and Polazzi, 1987), cell-type specialisation (Teague et al., 1986) and cell-cycle control (Reed et al., 1985; Patterson et al., 1986). In this last case, the characterisation of the CDC28 and CDC7 gene products as KINs suggests that progress through G 1 phase to S phase may be achieved or regulated by sequential activation of KINs and phosphorylation of key target proteins (Patterson et al., 1986; Mendenhall et al., 1987). Since the CDC7 gene product differs from all other known KINs in the organisation of the conserved aa within its catalytic domain (Patterson et al., 1986; Hanks et al., 1988), temporal and functional coordination of this component of the mitotic cell cycle might then be achieved as a consequence of structural variations in the catalytic domains of the various KINs. To examine this possibility, we have sought in the first instance to identify other KIN genes that might encode
88
components of this putative sequential phosphorylation pathway. Our approach to this was similar to that adopted to screen for Ser-Thr KIN genes in a HeLa cell cDNA library (Hanks, 1987) and putative Tyr KIN genes in yeast (Levin et al., 1987). In both cases, degenerate oligo probes were used to recognise target sequences that code for short, highly conserved regions in the KIN catalytic domain. The aim of the present study was to use an oligo mixture to isolate and characterise a novel yeast KIN gene whose product has an unusual catalytic domain structure.
o)
ealHl-a
Oao).,1
v a n n !- • p s
(xal).,z
i, a n n L K P q . . ( 1 6 9 )
..ttsa)
(alm)..Y
v n n n L • p A ..(zso)
(laS).,v
!, n n n v g P t
(130)..V
V H R D L • 1~ I1 . . ( 3 3 8 )
Cmme~t~lUJ
tO
..(le6)
- - H R D L • P -
1
2 MATERIALS AND METHODS
kb
----4.4
Escherichia coil HW87 was used for the routine maintenance and propagation of plasmids (Patterson et ai., 1986); strain JM103 was used for the preparation of MI3 clones for DNA sequencing (Messing et al., 1981). Restriction analysis, cloning in E. coli and transformation of S. cerevisiae strain SB303 (Mata[Mat~ his/his ura3-52/ ura3-52 trpl-289/trp1-289) were as described previously (Patterson et al., 1986).
RESULTS AND DISCUSSION "---
1.1
¸
0.5--
Fig. I. Design and characterisation of the degenerate olign probes. (a) Comparison of a conserved subdomain of five Ser-Thr KINs and the derived degenerate probe sequence for the consensus. The sequences shown in standard IUPAC single-letter en code are SNFI and CDC28 from $, cerevbiae (Celenza and Carlson, 1986; Lorincz and Reed, 1984), weel *, from ~chizosaccharomyces oombe (Russell and Nurse, 1987), CDC2Hs, the human homologne ofSc.pombe cdc2 + (Lee and Nurse, 1987) and CaMiI.% the ~-subunit of a calcium/calmodulin-dependent KIN from rat (Henley et al., 1987). The numbers in parentheses to the left and right oftbe en sequence are the number ofresidues to the N and C terminus, respectively. The olign probe mixture was synthesised by the phosphoramidite method using an ABI380 synthesiser, and purified by electrnphoresis through 25% PA. (b)Hybridisation of the degenerate oligo probe to yeast genomic DNA. An EcoRV digest of yeast genomic DNA was resolved by electrophoresis through 1% agarose, transferred to nitrocellulose (Southern, 1975) and probed using oligns that had been labelled with J2p by polynucleotide kinase and [~.32PlATP (Grunstein and Hogness, 1975; Maxam end Gilbert, 1977). Filters were prehybridised for 6-8h at 37°C in a solution containing 6 × SSC/0.1% Ficoll/0.2% polyvinylpyrrolidone/0.2~o BSA/0.5% SDS/15% formamide, then hybridised for 18-24 h at 20°C in a similar solution from which SDS was omitted and to which labelled olign probe was added.
(a) Design and eharaeterisation of oligo probes The design of the oligo probes was based on a short aa sequence within the catalytic domain of five Ser-Thr KINs with diverse cellular functions (Fig. la). The consensus sequence for this region is His-Arg-Asp,Leu.Lys-Pro (HRDLKP in single-letter code), which is highly conserved in these proteins with the Asp residue thought to interact with the phosphate groups of ATP via salt bridges (Brenner, 1987). This subdomain also differs significantly from the equivalent sequence in mammalian Tyr KINs where Lys is usually replaced by Ala or Arg, and Pro is replaced by Ala (Hanks et al., 1988). To optimise the potential hybridisation frequency of probes based on this sequence, a mixture of oligos containing 17 nt was synthesised, corresponding to 32 of the possible codons that could give rise to the HRDLKP sequence. Codon selection was based on utilisation frequencies in yeast protein-coding sequences. The HRDLKP-encoding oligo mixture was used initially to screen a blot of yeast genomic DNA that had been digested with EcoRV, and the filter washed in TMAC at
4 After hybridisation, filters were washed four times for 5 rain each in ice-cold 6 × SSC, then twice for 45 min in 3 M TMAC/S0 mM Tris. HC! pH 7,8/2 mM EDTA/0.1% SDS at 46°C (lane I) or 53°C (lane 2). These conditions select specifically for hybridising sequences containing long contiguous matches (Wood et al., 1985). After washing, filters were autoradiographed overnight at -80 ° C using two X-Omat intensifying screens.
89
46°C. TMAC eliminates the preferential melting of A-T versus G-C pairs, allowing the stringency of hybridisation to be controlled as a function of probe length. The conditions used here are equivalent to a reduced stringency wash requiring a 14-bp contiguous match between probe and target (Wood et ai., 1985). On this basis, sequence comparisons indicate that the HRDLKP probe should hybridise to five of the known yeast KIN genes, including CDC28 which was used in the design of the probe (Fig. la). In practice, the probe hybridised to about ten discrete genomic fragments that range in size from 1.1-10.5 kb (Fig. lb, lane 1) suggesting that the olign probe mixture could be identifying a limited number of previously unidentified KINs. Under conditions of maximum stringency imposed by washing with TMAC at 53°C, the probe hybridised only to two genomic fragments of 1.1 kb and 4.4 kb (Fig. lb, lane 2). (b) Isolation of KIN3 The HRDLKP=encoding oligo pool was used to screen a library of :yeast genomic DNA fragments in the vector YRpI2 (Barker and Johnson, 1983) for which the filters were washed in TMAC at 46°C. Positive sign,s were detected from nine of about 5000 colonies (approximately 3-4 genome equivalents ofS. cerevisiae) and plasmid DNA was prepared from each of these nine clones. Preliminary analysis revealed that one of these plasmids carried the CDC7 gene, while two others carlied the oligo-hybridising
12 kb
g'.t
345
6
•
4.a
•
Q
2"0 I
Q 8
0.S
Fig. 2. Isolation and characterisation of a clone hybridising to the HRDLgP.encoding olign. A yeast genomic library in the vector YRpl2 (Barker and Johnson, 1983) was screened by colony hybridisation using labelled oligns as described in the legend to Fig. lb. Plasmid DNA which hybridised to the olign was isolated from one clone and designated pC3. pC3 plasmid DNA was digested with (lanes): Accl (1), BamHl (2), EcoRV (3), Pstl (4), Sail (5), and TaqI (6), and the fragments resolved through a 1% agarose gel. The DNA was transferred to nitrocellulose and probed with the HRDLKP-encoding pool as described in the legend to Fig. lb.
sequence in different putative intergenic regions that were closed in all reading frames (data not shown). Two other clones have not been characterised further. Surprisingly, the same yeast genomic f r ~ e n t was present in four of the remaining clones, one of which, pC3, was characterised initially by digesting the plasmid DNA with a variety of restriction enzymes and probing the fragments with the HRDLKP-encoding oligo pool (Fig. 2). Of the DNA fragments hybridising to the probe, those produced by EcoRV (1.2 kb), Accl (1.1 kb) ~Jad Taql (0.6 kb) were small enough for further characterisation. In addition, by comparing the size ofthe pC3 DNA fragments that hybridised to the HRDLKP-encoding probe with restriction maps derived from nt sequence data for known yeast KIN, it was deduced that the genomic fragment of pC3 did not correspond to any of the previously identified KIN genes. The 600-bp Taql fragment of pC3 carrying the sequences responsible for hybridisation was subeloned into the Accl site of Ml3mpl8 and sequenced. Within the nt sequence obtained for this fragment was the element, 5'-CATCGTGATCTGAAACC-3', which is a perfect match with one member of the oligo probe pool. In addition, the sequence contained a single incomplete ORF that was capable of encoding a peptide carrying the HRDLKP motif. This ORF was termed KIN3. (e) KIN3 encodes a protein kinase The complete nt sequence of the KIN30RF was determined together with 500 bp of upstream and 370 bp of downstream sequence, using the deduced restriction map of this region ofthe pC3 clone (Fig. 3a). This ORF extends for 1107 bp and is capable of coding for a protein of 43.4 kDa whose deduced aa sequence is shown in Fig. 3b. Comparison of the aa sequence deduced from the KIN3 nt sequence with three yeast KINs involved in cell-cycle control (CDC28), cell-type specialisation (STET) and carbon catabolite repression (SNF1) revealed that, in addition to the HRDLKP element, the KIN3 protein contains all of the conserved elements found within the catalytic domains of KINs (Fig. 4). These include the sequence Gly-X-Gly-XX-Giy that is thought to interact with the ribose and terminal pyrophosphate moieties of ATP (Sternberg and Taylor, 1984), and which lies 15-25 aa residues N terminal of an invariant Lys which appears to be directly involved in the phosphotransfer reaction (Kamps and Sefton, 1986); the triplet Asp-Phe-Gly which represents the most highly conserved sequence within the catalytic domain; and the consensus elements Ala-Pro-Glu and Asp-X-Trp-Ser-X-Gly that are frequently used as indicators of KIN activity. While KIN3 contains all of the conserved elements characteristic of the KIN catalytic domain, the organisation of these elements differs markedly from all other KINs. This is seen most noticeably in the spacing between the
90 a
AAAAACCGCAAATGTGACACCGTCCCTCAGTATTACTCTAGAGGGCTTAATGCCATMTA LysAs.GlyLysC¥:;AspThrVa I I'roGluTyrTyrSerArGGI y L e u A s . k l a I | e l l e
1471{ 320
CA'i'TCAATGATACATGTGA~AAGAACCAGC~CCAuI I ~11GAATTACTGCAA('.AT III sSe rMe L I I eAsl)Va | A$-LeuArGThrArfal'roSe rTh¢l'heG I uLeut.euG h~As p
1538 340
ATTCAGATACr-,AACTGCAAGAAAGTCGTTGCAATTAGAGAGATTTCAAAGGAGGTTACTC I IcG I n I I ©ArrjThrA I aArl;Ly5 S©|'Le uG I nLe uG I uArfjPheC I uArcArlILeuUe u
I -~98 360
GACTATGAAAATGAGCTC,ACAAACAIT[GAARAAATCCTCG AGAAGCAAGCTATTGAATA AspTyrG I uAsnC I uLeuThrAsn I le>>> 369
I fi58
98
CGAAAGAGAACTGAGTCAGTTGAAGGAACAATTTACC,CAGGCAGTGGAGGAGCGACCCAG
1718
GAGAT~CCA*~t, CAGTCCCCCGGCCACGCCCTGCCACCATACCCACGCCGAAACAAGC
|58
GGAAGTAATTAGCGGTAAGAAAGTTGGTAAGGTTCCAGAATCTATAAACGGATATTATGC
|778
CCTCATGAGCCCGAAGTGGCGAGCCCGATCTr¢CCCATCCGTGATGTCGGCGATATAGGC
218
TA AAAAATTTGCCAAACCTGCATACCACTGGCAAAC,ru~GATATCCATAAAATTTGCTCTG
1838
CCCAGAACCGACCTGTCGCGCCGGTGATGCCGGCCACf,AGGCGTCCGGCGTCGAGGATCA
278
AGAACATTTATACKi'rAATACATACGGCTATGTATGAATCATATTCAGTGACGAGTTACA
| 898
AT/U~TCr CJU~ATACGATATCTGACATGATTCCAGTGGCACAGTCTTCOGATACGTTACA
338
CTGGCGACTTGTAGTATATGTAAATCACGTrAGTGAGACCATTTAGKCCCTGTCAATGTT
195S
C C C ~ C u | t | I i u I i I I~AGG'I-i'rCTGAAI I I I | iCATTACGTTTCCTAA'i'rAGGTTAAA
398
T CATAAGCCrCTATATATTTGGCCCTTCTCCTGTCGAC
CGTA/~TA&AAGCGA~.t a"f"tCAAAAGAATATCAAATCATATAACATGAAAAGAGCAAGAT
458
TCACAATACTAATTTCAATGAGAGTCCTA'I'IWGATA ATTGAATCAATT CTCTGTGTCAGT
$18
ATGCATAGACGACAt, I I i I tCCAAGAATACCGTAGTCCTCAGCACCACCAGGGACACCCA M©t I I I s A rGAr~,G I nl'hci'heG I .G I uTyrA r~Se r Pt'oG |nG I .C I .G I .G I yll i ~ I'ro
5?8 20
CCAAGGTCAGAATACCAAGTTCTCGAAGAAATTCCCAGAGGTrCATTTCGGTCTGTACG& I'roArgSc rG I uTyrG I nVa I Le uG I uC 1u I I eC I yAr/~G I ySe rl'heG lySc rV;, I At/;
638 40
Fig. 4. Comparison of the KIN3 catalytic domain with the equivalent regions in three other yeast KINs, CDC28 (Lorincz and Reed, 1984), SNFI (Celenza and Carlson, 1986) and STE7 (Teague et ai., 1986). The extent of noneonserved aa residues is indicated by numbers in parentheses. The numbers above the KIN3 sequence correspond to the location of the aa in the KIN3 protein (of. Fig. 3b).
A I
V '
T V T II I
AE Ill
TX II
T '
VC ¥
A I
2sobp I
I
b
GTCGACi;CTCTCCCTTATCCGACTCCTGCATTAGCAAG 38
CAGCCCAGTAGTAGGTTGAC,GCCGTTGAGCACCGCCGCCGCAAGCAATGGTGCATGCA/'.G
•
A
•
•
•
•
AAAGTCATCCATATACCTACCAAGAAACTI3"TGGTTAGAAACGATATCAAATATGGCCAT L.vsVa I I l e l l i s l lelaroThrLysLy,..LeuLeuV.'l IAr~LysAspl leLysTyrGlyll| ~.
GgR fi0
ATGAAT^GCAAAGAGAGACAACAGCTCATCGCTGAATGTAGCATTCTATCGCAGTTGAAG Me tAsnS©rLy~.GI uArr.G htGhlLe.I leAlaC| uCysSer I leLcu$crGI nLeuLys
758 80
CATGA~TATTGTAGAATTYrATAACI'GGGACTTCGATGAACAAAAAC&ACTGTTATAC Ill ~C I uAsni I eva IGI ul'heTyrA:mTrp&Hd'heA~pG hiG hd.ysG I ut.euL©uTyr
SI S
CTTTATATGGAATATTGTTCCAUGGGTGA'rTTATCCCAGATGATTAAGCACrACAAACAG LeuT~rMe t GI uTyrCysSe rAt'f,G lyA~pl.euSerG I rime t I I e L y s l i | ~ T y r l . y s G h t
HTS 120
CAGCATAAATATATACCAGAAAAAATTATATGGCCTATCCTGGCCCAAT'|'ATI'G^CTGCG GI n l l l i h ~ l * T y r | l e l ' r e C l u L p I I'el l e T r p G | y l It~LeuAluGh~Lq:uLeuThrAI.
93R 140
CTCTATAAATGTCATTATCGTG'rfGAM'TGCCAACFTTG^CCACAATATATGACCGG^TG LeuTyr L~'"Cysll I "Ty rGI yV~d G blue uProThr LcttTh r T h r I leTyrAsp^ri~let
; ~9~; , If~0
AAACCACCGGTAAAAGGCAAA^ACATCGTTATCCA'rCGTGATCTGAAACCAGGAAATATA Ly~l'rol'roVall.ysGlyl.y~A,..I leVai I lelli,-Arl;A.%~Le.bys h'nG yA~, 'd le
10$R 180
k
•
A'&
1996
100
A ......
B
Vii XIII,
•
TTCTTAAGCTATGATGATAGCG&TTACA^TATIA,~TGAACAAGTAGACGGTCACGAGGAA IqmLeuSe rTyrAspAspSc rA~'I)T~'rA:-~I I eA~t~G IuG htV~t I A~,pGI :~lll ~,GhiG I u
I I 18 200
GTG&ATAGTAM'TA'I"rACAGAGACC^TAGAGTC,AATTCA~C,GAA^AGGG(;G^G~ATG V" I AS flSe rA{t nT]/r'i'yrArlIAtqdll sArl;Va I A:mSe I,GI yL¥~Ar~G | yea rl,|,oMu t
I I 7t4 220
GACTATAGTCAAGTTG',i'GGTAAAGTTACGTGATTTTGCGrTGACCC^AATCTCTGAAACT A~'l)T~r $0 rG h~Vtl I Vtl I Vu I I,ysLo tic I)'hs plqieG I .vl.etiSe rG h d I.eSe rGI uThr
1238 240
AGTATI'CAATTTGCCACAACATACGTCGGTACACCATATTACATGTCGCCf(;AAGTCTTG Set'l leGInlq~eAlaThrThrTyrVnlGlyThrl~roT:~rTyrHet$©rlh.oGhNalLeu
129H 2G0
ATGUACCAACCATAC;TCCCCACTATCCGACATCTGGTCACTAGCTTG.rGTTATTTTTGAG Me t ASlIGI nl*l'oTyr-~erltroLe"$e r a s p I I cYl'|)$elt'LauG I ~Cy~Vtl I I I el'hr~G I tt • • A • • • ATGTGTTCGTTCCACCCCCCATTTCAGGCA^^AAATT&TCTC(;AGCTACAAACTAAGATT Me t CysSe I'Leull I s I:roProl,heG h~A I aLy,*.A~nT:er Le ~G h,Le uG h~.rhr L¥~ i | e
13,~S |SO
14 | S 300
Fig, 3. Restriction analysis and nt sequence of|he pC3 clone hybrid|sing to the HRDLKP-encoding oligo. (n)Partial restriction map of the HRDLKP region of pC3. The very thick line represents the KIN30RF. A, Accl; (7., ClaI; E, EcoRl; T, Taql; V, Ec~RV; X, Xbal. (b)The nt sequence of KIN3 and deduced aa sequence. The nt sequence was deter. mined by the dideoxy chain-termination method (Sanger et aL, 1977) for which DNA fragments to be sequenced were cloned into Ml3mpl0, mpl8 or mpl9 (Yaniseh-Perron et ai., 1985). Reaction products were resolved through 67e PA gels under denaturing conditions and detected by autoradiography for 1-4 h at room temperature. The nt sequence corresponding to the ollgo probe is overlined. Conserved aa in the KIN cataiytie domain are indicated by triangles (see RESULTS AND DISCUSSION, section ¢), while the > > > symbol indicates the presumptive stop codon.
Fig. S. Mapping KIN3 by hybrid|sat|on to a yeast chromosome blot. Yeast chromosomes (numbered I-XYi) were prepared in agaros~ microbeads and resolved through 1% agarose by OFAGE as described by Schwartz and Cantor (1984). &~er electropboresis, chromosomes were detected by ethidinm bromide staining (panel A). The chromosomal DNA was then transferred to nitrocellulose and probed with the 600-bp Taql fragment from the KIN3 gene that had been 3=P-labelled by oligo priming (Feinberg and Vogelstein, 1983). ARer hybrid|sat|on (as described in the legend to Fig. I), filters were washed in 2 × SSC at room temperature for 30 rain, then at 65"C for 2h in 0.l x SSC/0.5% SDS with one change of wash solution. Autoradiography was at -80"C for 12-18 h with two X-Croat intensifying screens (panel B).
91
HRDLKP and DFG subdomains. In 38 known or putative Ser-Thr KINs (Hanks et al., 1988), this interval is highly conserved at 12-15 aa, whereas in KIN3 the conserved elements are separated by 51 aa (Fig. 4). One explanation for this could be the presence ofintrons in KIN3. However, the £1N3 sequence lacks any of the consensus exon-intron and 3' splice signal sequences (Teem et al., 1984) suggesting that these extra aa are present in the KIN3 protein. Thc effect that such a variation might have on KIN activity is '~known.
Four of the resulting URA + transformants were analysed by genomic blotting and were found to have acquired a disrupted KIN3 Iocu~ (data not shown). Al~er sporulation, haploid cells carrying the KIN3 disruption were found to be both viable and conjugation-proficient, while diploids in which both copies of KIN3 were inactivated underwent a normal meiosis. Therefore, the KIN3 gene product does not seem to play an essential role in mitosis, conjugation or meiosis. (e) Comelusions
(d) Characterisation of the KIN3 gene To determine whether the KIN3 gene was unique in the yeast genome or was a member of a family of closely related enzymes, the 600-bp TaqI fragment containing most of the KIN3 catalytic domain was used to probe an EcoRV digest of yeast genomic DNA under stringent and relaxed conditions. In both cases, a single hybridisation signal was obtained corresponding to a 1.1-kb EcoRV fragment, indicating that KIN3 is apparently unique in the yeast genome and is not closely related to any other gene at the level of nt sequence (data not shown). The 1. l-kb EcoRV fragment detected with the 600-bp Taql probe probably corresponds to the major l.l-kb DNA fragment detected by the HRDLKP-encoding oligo in genomic blots (Fig. lb, lane 2). To map KIN.~ within the yeast genome, the 600-bp Taql fragment was oligo primed and used to probe a blot of yeast chromosomes separated by pulsed-field electrophoresis; the blot was kindly provided by Dr. Paul Dickinson, Department of Pathology, University of Oxford, A single ~i~.al .~,~~ obtained from hybridisation of the KIN3 probe to chromosome VI (Fir, 5A,B); none of the previously characterised yeast KIN genes :x,:f~psto this chromosome. To determine whether KIN3 is expressed in ~,ct~x,clyproliferating cells, total RNA was isolated from a logarithmically growing culture of $. cerevisiae. RNA was electrophoresed under denaturing conditions, transferred to a nylon membrane and probed with the 600-bp Taql fragment. A single hybridisation signal was obtained corresponding to an RNA species ofabout 1400 nt (not shown), which is consistent with the size ofthe KIN30RF determmcd from the nt sequencing. In order to examine the cellular function of KIN3 and to determine if the product was essentia' for cell viability, the ~;00-bp TaqI fragment of KIN3 was cloned into the integ/~.ting vector YIp5 and the resulting plasmid used to transform S. cerevisiae strain SB303 to uracil prototrophy. Since Ylp5 lacks a functional ori, stable transformants only arise after integration of the plasmid DNA into the yeast genome, a process that occurs by homologous recombination (Hicks et al., 1979; Scherer and Davis, 1979). In this instance, recombination between the cloned Taql fragment and one of the chromosomal KIN3 loci results in a disruption of this locus.
(1) We have used an oligo pool to isolate a nova KIN gene from a library of yeast genomic sequences. The nt sequence of this gene, KIN3, predicts a protein product of 369 aa and a n M r of 43 435. This size is in good agreement with the size ofthe KIN3 gene transcript, while the ORF is flanked by sequences typically found adjacent to yeast ORFs. (2) The product of KIN3 is likely to be a KIN on the basis of the presence of short aa stretches that are characteristic of Ser-Thr KIN. However the deduced KIN3 aa sequence is unusual in that the interval between subdomains VI and Vll of the KIN catalytic domain is 51 aa in KIN3 but only 12-15 in all other known Ser-Thr KINs. Such modifications though do not necessarily prevent KIN activity (e.g., CDC7 KIN; Bahman et al., 1988). (3) The KIN3 protein lacks any of the distinguishing features ofTyr KINs, nor do regions of the protein outside the catalytic domain have homology to other KINs. (4) Cells carrying disrupted KIN3 genes undergo normal mitosis, meiosis and conjugation, suggesting that KIN3 is not essential for any of these processes. This behaviour is unlikely to be due to complementation by a closely related KIN since KLan3probes hybridise only to KIN3 sequences in yeast genomic blots even und,~r conditions of reduced stringency. Thus, the cellular role of KIN3 is |imit.ed to a nonessential or conditional function.
ACKNOWLEDGEMENTS We thank the rest of the group for discussions, helpful discussions and liquid refreshment; Tony Hunter for providing data prior to publication; Lee Johnston and Paul Dickinson for material assistance; and Margaret Barber for typing the manuscript. This work was supported by grants from S.E.R.C. and the University of Manchester Research Support Fund.
REFERENCES Bahman, A.M., Buck, V.M., White, A. and Rosamond, J.: Characterisation of the CDC7 gene product of SaccAarom~.s ceremiae as
92
a protein kinasc needed for the initiation of miotic DNA synthesis. Biochim. Biophys. Acts 951 (1988) 335-343. Barker, D.G. and Johnson, L.H.: Saccharomyces cerevisiae CDC9, a structural ~,~enefor DNA figase which complements $chizosaccharomyces pombe cJci?. European J. Biochem. 134 (1983) 315-319. Boguslawski, G. and Polazzi, J.O.: Complete nucieotide sequence of a gane conferring polymyxin B resistance on yeast: similarity of the predict ,d polypeptideto protein kinases. Proc. Nail. Acad. SCI.USA 84 (1987) 58~8-5852. Brenner, S.: Phosphotransferase sequence homology. Nature 329 (1987) 21. Celenza, J.L. and Carlson, M.: A yeast gene that is essential for release from 81ucose repression encodes a p~otein kinase. Science 233 (1986) 1175-1180. .,.~ Feinberg, A.P. and Vngelstein, B.: A te~mique for radiolabelling DNA restriction endonuclease fragment:, to high speei~c activity. Anal. Binchem. 132 (1983) 6-13. Groffen, J., Heisterkamp, J., Reynolds, F.H. and Stephenson, J.R.: Homology between phosphotyrosine acceptur site ofhnman c-abl and viral oncngene products. Nature 304 (1983) 167-169. Granstein, M. and Hoguess, D.S.: Colony hybridization: a method for the isolation of cloned DNAs that contain a specific gene. Proc. Natl. Acad. Sci. USA 72 (1975) 3961-3965. Hanks, $.K." Homology probing: identification of eDNA clones encoding members of the protein-serine kinase family, prnc. Natl. Acad. Sci. USA 84 (1987) 388-392. Hanks, S.K., Quinn, A.M. and Hunter, T.: The protein kinase family: conserved features and deducted phylogeny of the catalytic domain. Science 241 (1988) 42-52. Henley, R.M., Means, A.R., Ono, T., Kemp, B.E., Burgin, K.E., Waxham, N. and Kelly, P.T.: Functional analysis of a complementary DNA for the 50.kilodalton subunit ofcalmoduiin kinase 11. Science 237 (1987) 293-297. Hicks, J.B., Hinnen, A. and Fink, G.R.: Properties of yeast transformation. Cold Spring Harbor Syrup. Quant. Biol. 43 (1979) 1305-1314. Hunter, T. and Cooper, J.A.: Protein tyrosine kinases. Annu. Rev. Biochem. 54 (I 985 ) 897-930. Kemps, M.P. and Sefion, B.M.: Neither nrginine nor histidine can carry out the function oflyaine.295 in the ATP-binding site of p60sre. Mol, Cell. Biol, 6 (1986) 751-757. Kamps, M.P., Taylor, S.S. and Set,on, B,M." Direct evidence that onco8enic tyrosine kinases and cyclic AMP.dependent protein kinase have homologous ATP-binding sites. Nature 310 (1984) 589-592. Lee, M.G. and Nurse, P.: Complementation used to clone a human homologue ofthe fission yeast cell cycle control gene cdc2. Nature 327 (1987) 31-35. Levin, D.E., Hammond, C.I., Ralston, R.O. and Bishop, J.M.: Two yeast genes that encode unusual protein kinases. Proc. Natl, Aead. Sei. 84 (1987) 6035-6G.~9. Lorincz, A.T. and Reed, S.!.: Prime~2,structure homology bet~,,een the
product of yeast cell division control gene CDC28 and vertebrate oncngenes. Nature 307 (1984) 183-185. Maxam, A.M. and Gilbert, W.: A new method for sequencing DNA. Proc. Natl. Acad. Sci. 74 (1977) 560-564. Mendenhall, M,D., Jones, C.A. and Reed, S.I.: Dual regulation of the yeast CDC28-p40 protein kinase complex: cell cycle, pheromone and nutrient limitation effects. Cell 50 (1987) 927-935. Messing, J., Crea, R. and Seeburg, P.: A system for.shntgun DNA sequencing. Nucleic Acids Res. 9 (1981) 30~-321. Patterson, M., Sclafani, R.A., Fangman, W.L. and Rosamond, J.: Molecular characterisatinn of cell cycle gene CDC7 from Saccharomyces cerevisiae. Mol. Cell. Biol. 6 (1986) 1590-1598. Reed, S., Hadwiger, J. and Lorincz, A.T.: Protein kinase activity associated with the product of the yeast cell division cycle gene CDC28. Proc. Natl. Acad. Sci. USA 82 (1985) 4055-4059. Roussou, I., Thireos, G. and Hange, B.M.: Transcriptional-translational regulatory circuit in Saccharomyces cerevisiae which involves the GCN4 transcriptional activator and the GCN2 protein kinase. Moi. Cell. Biol. 8 (1988) 2132-2139. Russell, P. and Nurse, P.: Negative regulation of mitosis by weal *, a gene encoding a protein kinase homologue. Cell 49 (1987) 559-567. Sanger, F., Nickle~, S. and Coalson, A.R.: DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. SCI. USA 74 (1977) 5463-5467. Scherer, S. and Davis, R.W.: Replacement ofchromosome segments with altered DNA sequences constructed in vitro. Proc. Natl. Aead. Sci. USA 76 (1979) 4951-4955. Schwartz, D.C. and Cantor, C.R.: Separation ofyeast chromosome-sized DNAs by pulsed field gradient gel electrophnresis. Cell 37 (1984) 67-75. Southern, E.M.: Detection of specific sequences among DNA fragments separated by gel ele©trophoresis. J. Mol. Biol. 98 (1975) 503-517. Sternb~r8, MJ.E. and Taylor, W.R.: Modelling the ATP-bindin8 site of oncogene products, the epidermal growth factor receptor and related proteins. FEBS Lett. 175 (1984) 387-392. Teem, J., Abovich, N., Kanfer, N., Schwindioger, W., Warner, J., Levy, A., Woolford, J., Leer, R., Van Raamsdonk-Duin, M., Maser, W., Plants, R.~ Schultz, L., Friesen, J., Fried, H. and Rosbach, M.: A comparison ofyeast ribosomal protein 8ene DNA sequences. Nucieie Acids Res. 12 (1984) 8295-8312. Teague, M.A., Chaleff, D.T. and Errede, B.: Nucleotide sequence of the yeast regulatory gene ST£? predicts a protein homologous to protein kinases. Proc. Natl. Aead. Sci. USA 83 (1986) 7371-7375. Wood, W.L, Gitschier, J., Lasky, L.A. and Lawn, R,M.: Base composition-dependent hybridization in tetramethylammonium chloride: a method for oligunucleotide screening of highly complex 8ene libraries. Proe. Natl, Acad. Sci, USA 82 (1985) 1585-1588. Yanisch-Perron, C., Vieira, J. and Messing, J.: Improved MI3 phage cloning vectors and host strains: nucleotide sequences of the Ml3mpl8 and pUCI9 vectors. Gene 33 (1985) 103-119.
