white 3 (sgg) segment polarity gene of Drosophila. This regulatory protein is functionally homologous to glyco- gen synthase kinase-3 in mammals (GSK-3), ...
Mol Gen Genet (1994) 242:337-345 © Springer-Verlag 1994
Arnbidopsis homologs of the shnggy and GSK-3 protein kinases: molecular cloning and functional expression in Eset erichia coli Michele W. Bianchi ~, Dominique Guivarc'h ~, Martine Thomas ~, James R. Woodgett ~'*, Martin Kreis ~ ~ Biologie du D~veloppement des Plantes, Centre de Recherches sur les Plantes URA 1128, Universit6 de Paris-Sud, Brit. 430, F-91405 Orsay Cedex, France ~ Ludwig Institute for Cancer Research, 91 Riding House Street, London WlP 8BT, UK Received: 3 May 1993 / Accepted: 2 August 1993
Abstract. The conservation in evolution of fundamental signal transduction modules offers a means of isolating genes likely to be involved in plant development. We have amplified by PCR Arabidopsis cDNA and genomic sequences related to the product of the shaggy/zestewhite 3 (sgg) segment polarity gene of Drosophila. This regulatory protein is functionally homologous to glycogen synthase kinase-3 in mammals (GSK-3), which regulates, among others, the DNA-binding activity of the cjun/AP1 transcription factor. Analysis of PCR products led to the identification of five genes; for two of which, corresponding full-length cDNAs, ASK-~ and y (for Arabidopsis shaggy-related protein kinase), were characterized. The encoded proteins were 70% identical to GSK-3 and sgg over the protein kinase catalytic domain and, after production in Escherichia coli, autophosphorylated mainly on threonine and serine residues, but phosphotyrosine was also detected. ASK-~ and ASK-y also phosphorylated phosphatase inhibitor-2 and myelin basic protein, on threonine and serine, respectively. The high conservation of the protein kinases of GSK-3 family, and their action at the transcriptional level, suggest that the ASK proteins have important functions in higher plants. Key words: A S K - zeste-white 3 - MCK1 - PCR - Glutathione-S-transferase
Introduction Although relatively few plant protein kinases have been characterized at the molecular level (Lawton et at. 1989; reviewed in Lawton 1990; Hanks 1991), protein phosphorylation is now well established as a regulatory mechanism in higher plants, as a consequence of both bioCommunicated by E. Meyerowitz * Present address: Ontario Cancer Institute, 500 Sherbourne Street, Toronto, Ontario, Canada M4X 1K9 Co~'espondence to: M.W. Bianchi
chemical (reviewed in Ranjeva and Boudet 1987; Budde and Chollet 1988; Budde and Randall 1990; Trewavas and Gilroy 1991) and genetic studies (Stein et al. 1991; Goring and Rothstein 1992; Kieber et al. 1993). The identification of plant homologs of evolutionary conserved protein kinases likely to be involved in important regulatory pathways, should usefully complement genetic approaches for the molecular dissection of development and differentiation in these organisms. At the present time, such analyses have been reported in the case of the p34 c~c2kinase, which controls the cell cycle (Feiler and Jacobs 1990; Ferreira et al. 1991; Hirt et al. 1991) and of the SNF1 kinase in Saceharomyces cerevisiae, involved in carbon catabolite repression (Alderson et al. 1991; Le Guen et al. 1992). A member of the mitogenactivated protein kinase family was also recently isolated from alfalfa (Duerr et al. 1993). We have identified, in Arabidopsis thaliana, a gene family that encodes proteins highly related to three recently identified protein kinases: the product of the sha99y/zeste-white 3 (s99) segment polarity gene of Drosophila (Bourouis et al. 1990; Siegfried et al. 1990), glycogen synthase kinase-3 (GSK-3) in m a m m a l s (Woodgett 1990), and MCK1 (first termed YPK1) in S. cerevisiae (Dailey et al. 1990; Neigeborn and Mitchell 1991; Shero and Hieter 199t). These proteins possess a catalytic domain characteristic of serine/threonine kinases and have been collectively termed the GSK-3 family, due to their remarkable structural homology (Hanks 1991). sha99y mutants show multiple defects throughout fruitfly development, including abnormal segmentation and cell fate changes (Simpson et al. 1988; Bourouis et al. 1989; Perrimon and Smouse 1989). Genetic evidence demonstrated an action of the sgg protein kinase at the transcriptional level, in response to positional information (Siegfried et al. 1992). ~Similarly, GSK-3, initially identified for its phosphorylation of glycogen synthase (Woodgett and Cohen 1984), was recently shown to block the DNA-binding activity of c-jun/AP-1 (Boyle et al. 1991a; Plyte et al. I992) and to phosphorylate the products of the c-myb (Boyle et al. 1991a) and L-myc
338 (Saksela et al. 1992) protooncogenes. All characterized GSK-3 substrates are phosphorylated on serine or threonine residues (reviewed in Plyte et al. 1992). The catalytic domains ofsgg and GSK-3 are 85% identical and the two kinases were recently shown to be functional homologs in vitro and in vivo (Plyte et al. 1992; Siegfried et al. 1992; Ruel et al. 1993a). Both proteins are also related to MCK1, which is necessary for transcriptional activation of early meiotic genes (Neigeborn and Mitchell 1991) and involved in centromere function (Shero and Hieter 1991). The Arabidopsis protein kinases, after production in Escherichia coli, showed an in vitro specificity similar to those of sgg and GSK-3 and autophosphorylated on threonine, serine and, albeit to a much lesser extent, tyrosine residues. The information stemming from the genetic and biochemical analysis of protein kinases of the GSK-3 family, together with their striking evolutionary conservation, suggests the involvement of its plant homologs in equally important transduction pathways.
from at least two independent amplifications. This made it possible to correct for TaqI polymerase errors and to rule out any artifacts that could be expected from the co-amplification of related sequences. Only one clone for c11413 was isolated, but its sequence was free of polymerase errors, since it was identical to the coding regions of a corresponding genomic PCR product (G284, see Fig. 2B).
Materials and methods
Bacterial production of ASK-~ and ASK-`/. An AhalIXbaI fragment from pBs6.1.1 was inserted, after tilting-in of both fragment and plasmid with Klenow enzyme, in the EcoRI site of the bacterial expression vector pGEX3X (Smith and Johnson 1988), to yield, after selection of the correct orientation, pGEX-3X:ASKm The first ASK-ct residue present in this fusion is serine 3. Similarly, an SstII (blunted with T4 polymerase)-EeoRI (3' polylinker) fragment was excised from a pUC18 BamHI-KpnI subclone of the ASK-3' cDNA and cloned into pGEX-2T digested with SmaI-EcoRI, to yield pGEX-2T:ASK,/. The first ASK-`/ amino acid in this fusion protein is valine 12. To produce the glutathione-S-transferase (GST, 26 kDa):ASK fusion proteins, 2.5 ml of an overnight culture of XL1 Blue carrying either pGEX-3X:ASK~ or pGEX-2T:ASK3', were inoculated into 50 ml of LB medimn containing 0.1 mg/ml ampicillin and grown at 37°C with shaking for 1 h (0.3-0.4OD600). IPTG (0.4 mM final) was then added, and induction was carried out for 4 h at 27 ° C, before harvesting by centrifugation. Bacteria were resuspended in 0.7 mt ice-cold PBSTE (150raM NaC1, 16mM Na2HPO4, 4 m M NaH2PO4, 2 mM EDTA, 0.1% v/v Triton X-100, pH 7.3) and quickly lysed by mild sonication; debris was removed by centrifugation at 12000 × 9, 4° C for 20 rain. After incubation of the supernatant with 50 gl of a 50% v/v slurry of glutathione (GSH)-agarose beads (Sigma) in PBSTE for 10 rain at 4 ° C with gentle rocking, beads were recovered by centrifugation, washed 4 times with 1 ml PBSTE, and stored as a 50% slurry on ice. Typical yields were 0.4~0.6 gg fusion protein/ml of culture, as determined by calibration with protein standards on SDS-polyacrylamide gel electrophoresis.
Construction of a cDNA library from Arabidopsis shoots. A. thaliana plants (ecotype Columbia, inbred strain GH 50) were grown in soil under a 12-h photoperiod. RNA was purified as described (Harpster and Taylor 1986) from the whole aerial parts of young shoots at the 2"a to 4t~ non-cotyledonary leaf stage. A )~ZAPII cDNA library was constructed following the manufacturer's instructions (Stratagene cDNA Synthesis kit) to yield, before amplification, 3.7 x 10~ pfu from 100 ng double-stranded cDNA. PCR amplification of ASK sequences. For PCR amplification, 20 ng of purified recombinant phage DNA from the Arabidopsis cDNA library were added to 100 gl PCR mixtures composed of: 0.1 mM each of dATP, dCTP, dGTP, and dTTP; 2 mM MgC12; 2.5 units of TaqI polymerase (Promega); PCR buffer (50mM KC1, 10raM TRIS-HC1 pH 9 at 25 ° C, 0.1% v/v Triton X-100); 1 gM each of the CHRDI and (Y/F)YRAPE oligonucleotide pools (oligos VIA, VIB and VIII, details in Fig. 2A). PCR was carried out in a water-bath thermocycler (Braun) as follows: 5 cycles composed of 1.5 rain denaturation at 94 ° C, 1 rain annealing at 40 ° C, a 2 rain ramp to 72 ° C, and a 30 s extension at 72 ° C, followed by 35 cycles of 30 s at 93 ° C, 1 rain at 50 ° C, 30 s at 60 ° C, and 30 + 1 s extension/cycle at 72 ° C. For the amplification of genomic sequences, 10 ng of Arabidopsis total DNA (Dellaporta et al. 1983) were subjected to PCR as described above, with the following modifications, to obtain an efficient selection of specific amplification products: oligonucleotides VIA and VIB were employed separately, and increasing concentrations of formamide were included in the reactions. Analysis of PCR products. Each type of genomic (G202, G284, G297 and G312) or cDNA (cl 14c~, `/ and g) sequence was determined from several clones originating
Isolation of full-size cDNA clones. Approximately 6 × 106 recombinants of the Arabidopsis shoots cDNA library were screened with a 32p-labeled probe (1.6 × 109 cpm/ gg) generated by amplifying PCR product c114a in the presence of 40 ~tCi of c~-[32P]dCTP (Amersham, 3000 Ci/ raM) and 0.1 mM each of dATP, dGTP and dTTP. Two classes of cDNAs were identified containing, respectively, the exact sequences of c114et and cl t43,. The longest inserts of each class, those of clones pBs6.1.1 (ASK-c0 and pBs8.4.2 (ASK-`/), were completely sequenced on both strands, after generation of nested deletions of the appropriate subclones with exonuclease III.
In situ autophosphorylation of immobilized A SKproteins. Five micrograms of GST :ASK~ and GST :ASK,/fusion proteins (73 kDa) on GStt-agarose beads were mixed with an equal volume of 2 × Laemmli sample buffer
339 (Laemmli 1970) and directly electrophoresed in a 8% SDS-polyacrylamide gel. After transfer to poly(vinylidene difluoride) membranes (Immobilon-P, Millipore), proteins were subjected to a denaturation-renaturation cycle as described by Ferrell and Martin (1991). The membrane was then submerged in 15 ml of 30 mM TRIS-HC1 pH 7.4, 10 mM MgC12, 2 mM MnCI2 and autophosphorylation was allowed to proceed in the presence of 50 gCi of y-[32p]ATP (Amersham, 5000 Ci/ raM) for 30 rain at room temperature. After removal of unincorporated radioactivity as described (Ferrell and Martin 1991), the membrane was exposed to X-ray film with two intensifying screens for 4 h at - 70 ° C.
Phosphorylation of inhibitor-2 and myelin basic protein, and phosphoamino acid analysis. Phosphorylation reactions were performed in a final volume of 50 gl kinase buffer (20 mM HEPES pH 7.5, 10 mM MgC12, 1 mM MnC12, 0.1% v/v 2-mercaptoethanol, 50 mM ATP and 5 gCi y-[32P]ATP) containing 5 gg of GST :ASK fusion proteins (as a 50% v/v slurry of GSH-agarose beads equilibrated in kinase buffer minus ATP), 1 gg inhibitor2 of phosphatase-1 (I-2; a generous gift of Prof. P. Cohen), or 5 gg of myelin basic protein (MBP, Sigma). After a 30rain incubation at 30 ° C, reactions were stopped by addition of 15 gl of 5 x Laemmli sample buffer, fractionated by SDS-PAGE, and proteins were transferred to Immobilon-P (Millipore). Following autoradiography, phosphorylated proteins on the excised membrane were subjected to partial acid hydrolysis, and phosphorylated amino acids were separated by bidimensional thin-layer electrophoresis as described (Boyle et al. 1991b).
genomic DNA
cDNA
[
[ 1
2
3
4
5
6
7
8
Fig. 1. PCR amplification of cDNA and genomic DNA of Arabidopsis thaliana. Amplification products were electrophoresedin a 2.4% low melting point agarose gel and stained with ethidium bromide. Lanes 1, 2 and 3, amplification of genomic DNA (10 rig) using oligonucleotidesVIA and VIII, in the presence of 0%, 2.5% and 3.5% v/v formamide, respectively. Lanes 4, 5, 6, amplification of genomic DNA (10 ng) with oligonucleotides VIB and VIII (see Fig. 2A), in the presence of 0%, 2.5% and 3.5% formamide, respectively. Lanes 7 and 8, two independent amplifications of 2 ng recombinant LZAPII DNA with oligonucleotides VIA, VIB and VIII. bp*, amplified products subsequentlyidentified as ASK sequences are indicated by the length of inserts, oligonucleotides(49 bases in total) excluded, x, positions of two formamide-sensitive products, determined to be the outcome of non-specific priming. Primer dimers, visible in all reactions, are indicated by an open arrow
Results
Identification by PCR of Arabidopsis 9enes related to shaggy, GSK-3 and MCK1 An alignment of the protein sequences of the sgg, GSK-3 and MCK1 protein kinases (see Fig. 4) highlighted several conserved motifs, among which C H R D I and (Y/F)YRAPE, in catalytic subdomains VIb and VIII, respectively, were selected for the design of degenerate oligonucleotides (oligonucleotides VIA, VIB and VIII, Fig. 2A). These motifs have not been found together in any of the 96 sequences of protein serine/threonine kinases surveyed (Hanks 1991) and occupy positions in the catalytic domain that appeared to be invariant among the functional homologs of several evolutionarily conserved proteins, such as p34 c~c2, protein kinase C or the cAMP-dependent protein kinase (Hanks 1991). Oligonucleotides VI and VIII were therefore used to amplify cDNA and total genomic D N A of A. thaliana. The products of cDNA amplification resolved as a single band (band 114 in Fig. 1) of a size compatible with an insert encoding the 38 amino acid region encompassed by the C H R D I a n d (Y/F)YRAPE motifs. However, the amplification of genomic D N A yielded several bands ranging from 200 to 300 bp (oligonucleotides excluded;
Fig. 1). Each of the amplification products visible in lanes 2, 3, 5, 6, 7 and 8 of Fig. 1 was individually cloned and characterized. This led to the identification of four classes of cDNA products, c114~, 13, ~' and 5, as well as four different genomic clones, G202, 284, 297 and 312 (Fig. 2B). Sequence comparisons showed that the differences in size between the genomic PCR products were due to the presence of introns. These were bordered by canonical splice sites and did not share significant homology, while all exonic sequences differed almost exclusively in codon usage (data not shown). The c114~, 13and ~ cDNAs are identical to the exons of three corresponding genomic PCR products, while c1146 and G312 identify two additional genes (Fig. 2B). An alignment of the amino acid sequences deduced from the different classes of PCR products with the corresponding regions of sgg, GSK-3 and MCK1, revealed significant and specific stretches of identity such as the CDFGSAK and the SYICSR motifs at subdomains VII and VIII, respectively (Fig. 2C). These motifs have not been found in other protein kinases and are invariant among all members of the GSK-3 family. Therefore, these results identify five Arabidopsis genes encoding proteins related to the catalytic domains
340
of the protein kinases of the GSK-3 family. We have named these genes ASK-a, 13, 7, 6 and ~, for Arabidopsis shaggy-related protein kinase.
A
KpnI VIA VIB vm
Characterization of two full-length A S K cDNAs
The A S K genes encode putative plant homologs of the shaggy/GSK-3 protein kinase ASK-a and ASK-y are 68-71% identical to an embryonic form of sgg (Bourouis et al. 1990; Ruel et al. 1993a) and to the ]3 form of GSK-3 (Woodgett 1990) over the catalytic domain (Fig. 4). As will be discussed, such a high and extended degree of identity is strongly indicative of functional homology. The identity between sgg and GSK-3[~ is 85%, a higher value which probably reflects the phylogenetic relationships of the respective organisms. The catalytic domains of ASK-a, ASK-y, GSK-313 and sgg are only from 36% to 39% identical to that of the MCK1 kinase, but all proteins share sequence motifs in the core of the catalytic domain which are not found in any other protein kinase. In particular, the tyrosine resi-
H
R
D
I 3'
- 5' GA CGTACC TG(CT) CA(CT) AG(AG) GA(CT) AT(ACT) A
3'
Y/F
Y
R/S
- 3' A(AT) (AG) AT(AG)
B A c D N A library from Arabidopsis shoots was screened at reduced stringency with PCR product c l l 4 a . Two classes of cDNAs were identified, each containing the exact sequences of el 14a or c114?. The complete nucleotide sequences of the longest c D N A clone of each class, ASK-a (1.5 kb) and ASK-? (1.6 kb), were determined in both directions (Fig. 3). The ASK-a and ? cDNAs contained open reading frames o f 1215 bp (405 amino acids) and 1227 bp (409 amino acids), respectively. In both cases the presumed initiation codons, preceded by stop codons in all three reading frames, were situated in contexts which conformed closely to the consensus observed for plant genes (Joshi 1987). The encoded polypeptides, of predicted molecular weights of 45 994 Da for A S K - a and 46 557 Da for ASK-?, presented all the characteristic hallmarks of serine/threonine protein kinases (Hanks et al. 1988; Hanks 1991). The two proteins were 98 % identical over the 285 residues-long catalytic domain and 84% identical over the regions at the N- and C-termini. However, comparison at the nucleotide sequence level clearly showed that these cDNAs derived from two different genes, as already deduced from the analysis of PCRgenerated clones. Transcripts hybridizing to the ASK-a and ? cDNAs were detected at similar levels in roots, young shoots and fully expanded leaves (data not shown). The estimated size of both ASK-a and ? m R N A s was 1.6 kb, indicating that the ASK-y c D N A represents a virtually full-size transcript, while up to 100 bases could be missing from the 5' untranslated leader of ASK-a. The ASK-a and ASK-y cDNAs possess a 29 bp stretch of identity in the' 3' untranslated region (Fig. 3), possibly reflecting sequences involved in post-transcriptional regulation. The ASK-? sequence also contains motifs similar to the nos response element, which mediates translational repression in Drosophila (Wharton and Struhl 1991 ; Dalby and Glover 1993; see Fig. 3).
C
- 5' GA GGTACC TG(CT) CA(CT) CGI GA(CT) AT(ACT) A
A
P
1 0114~
genornic DNA G312 ~
fT-21"K-4
2 c1147
EeoRI
(GT) CI CGI GGI CT(CT) CTTAAC AG 5'
eDNA
Freq. 82 cl14c~
E
81
G297 ~
96
G284L4
a5
G202 F/I/'/]
~
117 L/'~#'~,4 87
[',',4
8s
88
[f~
I*,d
r/I
2 cl146 C CHRDI.--I~.
F ~---YYRAPE
KPQNLLVNnvTHeVKICDFGSAKmLipGEPNISYICSR KPQNLLVNPHTHQVKLCDFGSAKVLVKGEPNISYICSR
U~T& E(G312)
KPQNLLVNPHTHQLKLCDFGSAKVLVKGEPNISYICSR
6
KPQNLLLDPdTAVLKLCDFGSAKQLVrGEPNVSYICSR
GSK-3
KPQNLLLDPETAVLKLCDFGSAKQLIHGEPNVSYICSR
Sg~
~sNvLVDPETgVLKICDFGSAKkLeHnqPsISYICSR ,,o
MCKI
I - - subdomain ~ZZb- - ] - -
~23Z--.I
~
I
Fig. 2A-C. PCR-generated genomic and eDNA sequences of ASK genes. A Oligonucleotides used to amplify sog-related sequences. VIA and VIB, sense oligonucleotides differing in codon usage at the R codon in the CHRDI motif of catalytic subdomain VIb of protein kinases of the GSK3 family. The first adenine of the immediately following lysine codon was included at the Y end. VIII, antisense oligonucleotide encoding the (Y/F)YRAPE motif of subdomain VIII. Two serine codons were allowed at the R position, to include all of the six arginine codons in a single oligonucleotide. Degenerate positions are reported in parenthesis and the restriction sites added to facilitate cloning are indicated. I, deoxyinosine. B Schematic representation of the different eDNA and genomic sequences identified, PCR products are named according to the length of the amplified inserts, oligonueleotides (49 bp in total) excluded. Shaded areas indicate protein coding sequences, while open boxes represent introns of indicated lengths. Corresponding cDNA and genomic amplifiats are side by side, with perfect nucleotide identity being represented by the same shading patterns. PCR products c114aG297, c114[3-G284, c114y-G202, c1146 and G312 originated from five different genes: ASK-a, [~,y, 6 and s, respectively. The observed frequencies of each class of eDNA products are reported. C Alignment of amino acid sequences encoded by the amplified fragments (oligonueleotides excluded) with the corresponding regions of sgg, GSK-3 and MCK1. The Arabidopsissequences are indicated by the names of the fivepostulated ASK genes. The same protein sequence was encoded by products c114ct,y and their corresponding genomic amplifiats, as well as by G312. Residues conserved in two or more entries are in upper case letters, while residues conserved in all entries are in bold. Dots below the alignment indicate amino acids which are invariant or nearly invariant among all serine/threonine protein kinases. The sequences targeted by oligonucleotides VI and VIII are reported, while the tyrosine residue subjected to regulatory phosphorylation in GSK-3 and sgg (22) is labeled with an asterisk due found to be phosphorylated in GSK-3 and sgg (Hughes et al. 1993) is conserved in the plant proteins. However, ASK-a and ASK-y, as MCK1, do not share the same N-terminal sequences as sgg and GSK-313 (see Discussion).
341
A
B
ASK-
ASK~
ATTCATCATCATCAATCAATCCTTCATTTTATGGATCTACTCATATCTTGATTGATTCTT 60 CCTTCTATCTCTCCCTTTTCTTCCATCTCTTTTTCTCTGGGTTTCCCCGATTTGAAGAGC 120
CTCAGATCGATGAAGAGAAGAATTAGGATTTTTACGTTTTCATCGGCTTGAAAGTTTGAA
60
M A S GTGACAAAGGAAGAATCTTTTATTAAAACAAATTCTTCTGTTTTAATCTTGGGATGGCCT
3
180
M A S V G I A P N P G A R D 14 GAGTTTTGTAGCCTGAAAAATGGCGTCAGTGGGTATAGCTCCTAATCCTGGAGCAAGAGA 120 AhaII S T G V D K L P E E M N D M K I R D D K 34 CTCTACTGGTGTTGATAAATTGCCTGAAGAAATGAATGACATGAAAATTCGTGACGATAA 180
V G I Z P S A A V R Z S T G N V T D A D 2 3 CGGTGGGCATAGAGCCTAGTGCCGCGGTTAGAGAATCTACTGGAAACGTTACTGATGCTG 240
E M E A T V V D G N G T E T G H I I V T 54 AGAAATGGAAGCGACAGTGGTAGATGGAAATGGAACAGAGACTGGACATATCATTGTGAC 240 < T I G G R N G Q P K Q T I S Y M A E R V 74 TACTATTGGTGGTAGAAATGGCCAACCAAAACAGACAATTAGCTACATGGCTGAGCGTGT 300
I V N G N V T E T G H I I V T T I G G R 6 3 CGATTGTTAATGGCAATGTGACTGAGACTGGCCATATAATAGTAACTACTATAGGAGGAA 360 < N G Q P K Q T I S Y M A E R V V G H G S 8 3 GAAATGGCCAGCCAAAACAGACAATCAGTTACATGGCGGAGCGAGTTGTTGGACATGGCT 420
V G H G S F G V V F Q A K C L E T G E T TGTTGGTCACGGATCTTTTGGTGTTGTGTTCCAAGCGAAATGTCTTGAGACAGGAGAAAC
94 360
F G V V F Q A K C L Z T G E T V A I K K I 0 3 CCTTTGGTGTTGTGTTTCAGGCCAAATGTTTAGAAACAGGAGAAACTGTTGCTATAAAGA 480
V A I K K V L Q D R R Y K N R E L Q T M TGTTGCGATAAAGAAAGTTTTACAAGATAGGAGGTACAAGAACCGTGAGCTTCAAACCAT
114 420
V L Q D R R Y K N R E L Q T M R L L D H I 2 3 AAGTTCTACAAGATCGGAGGTACAAGAATCGTGAGCTTCAAACAATGAGGCTACTTGACC 540
R L L D H P N V V S L K H C F F S T T E I 3 4 GAGGCTACTTGACCATCCTAATGTTGTGTCTCTGAAACATTGTTTCTTCTCAACCACTGA 480
P N V V S L K H C F F S T T Z K D E L Y I 4 3 ATCCAAATGTTGTGTCTTTGAAACATTGTTTCTTCTCTACAACCGAAAAAGATGAGCTTT 600
K D E L Y L N L V L E Y V P E T V H R V I 5 4 AAAAGATGAGCTTTACCTCAATCTTGTTCTTGAGTACGTTCCAGAAACTGTTCATCGTGT
540
L N L V L E Y V P E T V H R V I K H Y N I 6 3 ATCTCAACTTGGYTCTGGAATACGTTCCGGAAACTGTGCACCGCGTCATCAAACACTACA 660
I K H Y N K L N Q R M P L I Y V K L Y T TATCAAACACTACAACAAACTGAATCAGAGAATGCCTCTTATATACGTCAAACTTTACAC
174 600
K L N Q R M P L V Y V K L Y T Y Q I F R I 8 3 ACAAACTTAACCAACGAATGCCTCTCGTTTACGTCAAACTTTACACTTATCAGATTTTTA 720
Y Q I F R A L S Y I H R C I G V C H R D 194 TTATCAGATTTTTAGAGCCTTATCTTACATTCACCGATGCATTGGTGTGTGTCATCGTGA 660
S L S Y I H R C I G V C H R D I K P Q N 2 0 3 GGTCCTTATCCTACATTCACCGATGYATCGGCGTATGTCATCGAGACATCAAACCTCAAA 780
I K P Q N L L V N P H T H Q V K L C D F 2 1 4 ~AT~AAACCTCA~ACTTGTTGGTAAATCCGCACACTCATCAAGTAAAGCTATGTGATTT
720
L L V N P H T H Q V K L C D F G S A K V 2 2 3 ACTTGTTGGTAAATCCACACACTCATCAAGTGAAACTATGCGATTTTGGAAGTGCGAAAG 840
G S A K V L V K G E P N I S Y I C S R Y 2 3 4 TGGAAGTGCAAAAGTATTGGTAAAAGGAGAACCAAACATTTCCTACATCTGCTCGAGGTA 780
L V K G Z P N I S Y I C S R Y Y R A P E 2 4 3 TATTGGTTAAAGGAGAGCCAAACATTTCATACATTTGCTCGAGGTATTACAGAGCACCTG 900
Y R A P E L I F G A T E Y T T A I D V W 2 5 4 TTACAGAGCACCTGAACTTATTTTTGGAGCAACCGAGTATACGACAGCCATTGATGTCTG 840
L I F G A T E Y T T A I D V W S A G C V 2 6 3 AGCTCATTTTTGGAGCCACCGAGTATACTACAGCCATTGATGTCTGGTCTGCAGGATGTG 960
S A G C V L A E L L L G Q P L F P G E S 2 7 4 GTCTGCAGGATGTGTTCTAGCTGAACTATTGCTTGGACAGCCCTTGTTCCCTGGTGAGAG 900
L A E L L L G Q P L F P G E S G V D Q L 2 8 3 TTCTCGCCGAGCTTCTTCTCGGGCAGCCATTGTTCCCGGGTGAGAGCGGTGTTGATCAAC 1020
G V D Q L V H I I K V L G T P T R E Z I 2 9 4 CGGTGTTGATCAACTTGTACACATTATCAAGGTCTTGGGAACGCCTACTAGAGAAGAAAT 960
V E I I K V L G T P Y R E E I K C M N P 3 0 3 TTGTAGAGATTATAAAGGTTTTGGGAACACCAACAAGGGAAGAAATCAAATGCATGAACC 1080
K C M N P N Y T E F K F P Q I R A H P W 3 1 4 CAAGTGCATGAACCCAAACTACACGGAATTCAAATTCCCTCAGATTAAAGCTCATCCATG 1020
N Y T E F K F P Q I K A H P W H K I F H 3 2 3 CGAATYACACAGAGTTCAAATTTCCTCAGATTAAAGCTCATCCATGGCATAAGATTTTCC 1140
H K I F H K R M P P E A V D L V S R L L 3 3 4 GCACAAGATTTTCCACAAACGCATGCCTCCAGAAGCTGTTGATTTGGTCTCAAGACTTCT 1080 > Q Y S P N L R S A A L D T L V H P F F D 3 5 4 TCAATACTCTCCTAATCTACGAAGTGCCGCTCTCGACACATTAGTCCACCCATTCTTTGA 1140
K R M P P E A V D L V S R L L Q Y S P N 3 4 3 ACAAGAGAATGCCTCCAGAAGCTGTTGATTTGGTCTCAAGGCTTCTTCAATACTCTCCCA 1200 > L R C A A L D S L V H P F F D E L R D P 3 6 3 ATCTCCGTTGTGCTGCTCTTGATTCATTGGTCCACCCATTC77TGACGAGCTAAGAGATC 1260
E L R D P N A R L P N G R F L P P A F H 3 7 4 TGAGTTAAGAGACCCAAACGCACGTCTACCTAATGGACGTTTCCTYCCACCGGCTTTTCA 1200
N A R L P N G R F L P P L F N F K P H E 3 8 3 CGAATGCGCGATTACCCAACGGACGTTTCCTTCCACCGCTCTTTAACTTTAAGCCTCATG 1320
F K P H E L K G V P L E M V A K L V P E 394 CTTCAAGCCTCACGAGCTGAAAGGTGTACCATTGGAGATGGTAGCTAAGTTAGYACCTGA 1260
L K G V P V E M V A K L V P E H A R K Q 4 0 3 AACTTAAAGGTGTGCCTGTGGAGATGGTGGCGAAGTTAGTTCCAGAACATGCGAGGAAGC 1380
H A R K Q C P W L G L e n d 405 GCATGCAAGGAAGCAGTGTCCTTGGCTCGGTTTGTGATTTCCTCTTAATGTAGCATGAAC 1320
C P W L S L end 409 AATGTCCGTGGCTCAGTTTATGATTTGTTCTCACCTGCAAACACGAAAACTAGAGCAAAG 1440
ACAACAAACACTTCTTATAAATTACCTCTCTATGTATCAATATGTCACAAACTGATATGC 1380
CAGTCGAGATATTCATCYCTTCTCTTCTCTCTCCTTCCTCTGTATTAATATTATTATAAT 1500
AgC~TTTGTT~TG!~T.qAGT.6GAG~AAAAAGAGTTATTACTATGGTTGGTTGGTTCA 1440
GATCATAYCTCAATCTGATGATTTAGTAACCCTTTGTTTGTTGTATG6GTAGAGAAAGAG 1560
TAATGTAAAAGCCCACCAAGATTTTTTATCTAGATAAAGAGTTTGCTAAAAAAAAA ~aI
TGAATCATTTGTGGGGGTTATGATATTGTATAAGCCAACAAAGATTATTTTTTAAAGAGA 1620
1496
SacII
R L P E E M K D M K I Q D D K E M E A T 4 3 ATAGATTACCCGAGGAGATGAAGGACATGAAAATTCAAGATGATAAAGAAATGGAAGCTA 300
GT:TCGTGTTTTCTGTeTcAAAAAAAAAAAAAAAA~AA
Fig. 3A, B, Nucleotide and deduced protein sequences of the two ASK c D N A clones. The complete nucleotide sequence of the longest ASK-a and ASK-y c D N A s are shown in (A) and (B), respectively, with the predicted amino acid residues indicated above, in single letter code. The stretches identical to PCR products c114c~ and G297 (introns excluded) for ASK-c~, and to c1147 and G202 (intron excluded) for ASK-y, are underlined. Asterisks indicate the guanine bases bordering the introns present in the corresponding genomic PCR products. Amino acids are numbered starting with
16S8
the first methionine as + l and nucleotide positions are reported. The protein kinase catalytic domains as defined by Hanks (1991) are delimited by < and > . The first amino acids present in the GST :ASK ~ s i o n proteins are Nghlighted by ~ t s . Relevant restriction sites are indicated. The 29 bp identity region in the 3' untranslated region of the ASK-a and ASK-y cDNAs is un~rlmed, while two moti~ in the ASK-y sequence similar to the nos response elements are ~ u b @ un~rlined
342
