Identification and characterization of a cDNA encoding mouse CAP: a

Journal of Cell Science 105, 777-785 (1993) Printed in Great Britain © The Company of Biologists Limited 1993

777

Identification and characterization of a cDNA encoding mouse CAP: a homolog of the yeast adenylyl cyclase associated protein Anne B. Vojtek* and Jonathan A. Cooper Fred Hutchinson Cancer Research Center, Seattle, WA 98104, USA *Author for correspondence

SUMMARY CAP, an adenylyl cyclase associated protein, is present in Saccharomyces cerevisiae and Schizosaccharomyces pombe. In both organisms, CAP is bifunctional: the Nterminal domain binds to adenylyl cyclase, thereby enabling adenylyl cyclase to respond appropriately to upstream regulatory signals, such as RAS in S. cere visiae; the C-terminal domain is required for cellular morphogenesis. Here, we describe the isolation of a cDNA encoding a CAP homolog from a higher eukaryote. The mouse CAP cDNA contains an open reading frame capable of encoding a 474 amino acid protein. The protein encoded by the mouse CAP cDNA shows extensive homology to the yeast CAP proteins, particularly in the central poly-proline rich region and in the

C-terminal domain. By northern analysis, the CAP message appears to be ubiquitous, but not uniform. By indirect immunofluorescence, ectopically expressed mouse CAP protein is found in the cytoplasm of fibroblasts and, in migrating cells, at the leading edge. Expression of the mouse CAP cDNA in S. cerevisiae complements defects associated with loss of the yeast CAP carboxyterminal domain. Hence, the function of the CAP carboxy-terminal domain has been conserved from yeast to mouse.

INTRODUCTION

In S. pombe, the CAP/adenylyl cyclase complex functions in the sexual differentiation pathway. Loss of aminoterminal CAP function results in conjugation and sporulation under inappropriate environmental conditions (Kawamukai et al., 1992). In contrast to S. cerevisiae, RAS does not regulate CAP/adenylyl cyclase activity in S. pombe (Fukai et al., 1986). The identity of the regulatory protein that lies upstream of the CAP/adenylyl cyclase complex is currently unknown. Deletion of the CAP gene in S. cerevisiae and S. pombe is also accompanied by morphological and nutritional deficiencies. In S. cerevisiae, loss of CAP function results in an inability to grow in rich medium, in synthetic medium containing valine, and in synthetic medium at elevated temperatures (Field et al., 1990; Gerst et al., 1991). In addition, the cap cells are grossly enlarged, bud randomly, and contain disorganized actin cytoskeletons (Field et al., 1990; Vojtek et al., 1991). Expression of the carboxy-terminal domain of CAP is sufficient to suppress these defects (Gerst et al., 1991; Vojtek et al., 1991). In S. pombe, deletion of the carboxy-terminal domain of CAP results in slow growth on synthetic medium and an inability to grow at elevated temperatures; the S. pombe cap cells are also altered in cell shape and size (Kawamukai et al., 1992). Moreover, expression of the CAP protein from either S. cerevisiae or S. pombe will complement the morphological and nutri-

The adenylyl cyclase associated protein CAP has been identified in two distantly related organisms, Saccharomyces cerevisiae and Schizosaccharomyces pombe (Field et al., 1990; Fedor-Chaiken et al., 1990; Kawamukai et al., 1992). In both organisms, CAP appears to have at least two functions, which reside in separate domains of the protein. The amino-terminal domain of CAP associates tightly with adenylyl cyclase. This CAP/adenylyl cyclase interaction is necessary for adenylyl cyclase to respond appropriately to its regulatory proteins in vivo. The carboxy-terminal domain of CAP appears to play a regulatory or structural role in cellular morphogenesis. In S. cerevisiae, the CAP/adenylyl cyclase complex responds to signals from RAS. Cells with an activated RAS allele, such as RAS2val19, are sensitive to environmental stresses; for example, RAS2val19 cells do not accumulate the storage carbohydrates glycogen and trehelose and are unable to recover from a 55°C heat shock (Toda et al., 1985). RAS2val19 cap cells do accumulate glycogen and trehelose and are resistant to the 55°C heat shock (Field et al., 1990). Therefore, in S. cerevisiae, CAP is required for cellular responsiveness to RAS. Expression of the amino-terminal domain of CAP in RAS2val19 cap cells is sufficient to restore RAS responsiveness (Gerst et al., 1991).

Key words: actin cytoskeleton, adenylyl cyclase associated protein, signal transduction

778

A. B. Vojtek and J. A. Cooper

tional defects due to loss of carboxy-terminal CAP function in the other organism (Kawamukai et al., 1992). This result suggests that the function of the CAP carboxy-terminal domain has been conserved in these two distantly related organisms. Overexpression of components of the RAS/cAMP pathway in S. cerevisiae has no effect on the morphological and nutritional deficiencies associated with loss of the carboxyterminal domain of CAP (Field et al., 1990). TPK1, a catalytic subunit of the cAMP-dependent protein kinase, positively modulates the RAS/cAMP pathway but, when overexpressed, does not alleviate the nutritional or morphological defects of cap cells. Overexpression of genes that negatively impact the RAS/cAMP pathway such as BCY1, which encodes the regulatory subunit of the cAMPdependent protein kinase, or PDE1, which encodes a cAMP phosphodiesterase, also do not alleviate the nutritional or morphological defects of cap cells. These studies suggest that CAP has an additional role outside of the RAS/cAMP pathway. The defects associated with loss of the carboxy-terminal domain of CAP are suppressed by overexpression of profilin (Vojtek et al., 1991). Profilin is a well characterized actin (Pollard and Cooper, 1986) and phospholipid binding (Lassing and Lindberg, 1985, 1988) protein that is thought to play a key role in coordinating the reorganization of the actin cytoskeleton in response to growth factor derived signals, such as platelet derived growth factor or epidermal growth factor (Goldschmidt-Clermont et al., 1990, 1991). The available data suggest that, in order to suppress the CAP carboxy-terminal domain defects, profilin interacts with acidic phospholipids, such as phosphatidylinositol 4,5bisphosphate (PIP2) (Vojtek et al., 1991); whether actin binding is also required has not yet been examined. Therefore, it has been suggested that profilin suppresses the defects of a cap strain by binding to and sequestering PIP2 and, correspondingly, that CAP participates in the regulation of acidic phospholipids. Alternatively, the binding of profilin to PIP 2 may be necessary to localize an actin binding protein to the membrane in order to compensate for the absence of the CAP carboxy-terminal domain. A second high copy suppressor of the defects associated with the loss of the carboxy-terminal domain of CAP, SNC1, has also been identified (Gerst et al., 1992). SNC1 is a homolog of the synaptic vesicle-associated membrane proteins (VAMPS), also called synaptobrevins (Elferink et al., 1989; Archer et al., 1990). The mechanism by which SNC1 suppresses the cap defects is unclear; perhaps SNC1 also interacts with or alters the distribution of acidic phospholipids. Curiously, in contrast to profilin, SNC1 is capable of suppressing the defects associated with loss of the carboxy-terminal domain of CAP only in the presence of an activated RAS allele. CAP has been identified from two distantly related yeasts. In both organisms, CAP is bifunctional, and both functions are conserved. The presence of two functional domains in one protein in two divergent organisms suggests the intriguing possibility that the two CAP functions are connected. CAP may serve to integrate adenylyl cyclase regulation with cytoskeletal function. Although both CAP functions appear to be conserved, the regulation of the

CAP/adenylyl cyclase complex appears to have diverged. RAS signals through the CAP/adenylyl cyclase complex in S. cerevisiae, but not in S. pombe. To determine to what extent the CAP protein has been conserved in evolution, and whether its regulation would reflect that of S. cerevisiae or S. pombe, we sought to identify a CAP gene from a higher eukaryote. We report here the isolation and initial characterization of a CAP homolog from the mouse.

MATERIALS AND METHODS Strains, media, and genetic methods The genotype of Saccharomyces cerevisiae strain SKN32 is Mata leu2 ura3 trp1 ade8 can1 cap::HIS3; the genotype of S. cerevisiae strain SKN37 is Mata leu2 ura3 trp1 ade8 can1 RAS2val19 cap::HIS3. Both strains have been described previously (Field et al., 1990). Yeast strains were grown on synthetic medium with appropriate auxotrophic supplements (SC), SC with 6.5 mM valine (SC+VAL), or rich medium (YPD) (Rose et al., 1990). Strains lacking CAP were grown at room temperature. Replica plating methods were used to assay sensitivity to heat shock, temperature, rich medium, and valine (Field et al., 1990; Gerst et al., 1991). The 10T1/2 mouse fibroblast cell line was generously provided by Dr Steve Tapscott. Cell lines were grown in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% bovine calf serum. DNA was introduced by calcium phosphate mediated transfection (Sambrook et al., 1989).

Cloning of mouse CAP A 586 nucleotide fragment of the mouse CAP gene was isolated by reverse transcription (RT) of randomly primed 10T1/2 RNA followed by 48 cycles of the polymerase chain reaction (PCR). The RT reaction contained the following components per 10 µl reaction: 1 µg of 10T1/2 RNA, 100 pmoles random hexamers (Pharmacia), 1 µl of 10 mM dNTPs (USB), 1 unit of rRNasin (Promega), 200 units of Moloney murine leukemia virus reverse transcriptase (BRL), and 1 µl 10× PCR-HB (500 mM KCl, 200 mM Tris-HCl, pH 8.4, 30 mM MgCl 2, 1 mg/ml BSA). The RT reaction was incubated for 45 min at 37°C followed by a 5 min, 95°C incubation. The PCR reaction contained the following components per 50 µl reaction: 10 µl RT reaction, 4 µl 10× PCR-LG (500 mM KCl, 200 mM Tris-HCl, pH 8.4, 11.3 mM MgCl2, 1 mg/ml gelatin), 1 unit of Taq polymerase (Perkin Elmer Cetus), and 10 µg of each degenerate oligonucleotide primer. Degenerate oligonucleotide primers were designed to match amino acid sequences in the carboxy-terminal domains of the S. cerevisiae and S. pombe proteins and to encode all possible codons for each amino acid. The sequence of the forward oligonucleotide primer was: 5′ CGTTCTAGAACNCA(C/T)AA(A/G)AA(C/T)CC 3 ′, and encodes the amino acids THKNP (residues 361 to 365 of S. pombe; residues 333 to 337 of S. cerevisiae). The sequence of the reverse oligonucleotide primer was: 5′ CGTGAATTC(G/A)TG(T/C)TTNAN(T/C)TGNTC 3 ′, and is complementary to the strand encoding the amino acids (D/E)Q(F/L/I/M/V)KH (residues 531 to 535 of S. pombe; residues 506 to 510 of S. cerevisiae). The PCR reaction conditions were as follows: (1) three cycles of denaturation at 94°C for 40 s, annealing at 48°C for 1 min followed by a 1 min ramp from 48°C to 60°C, extension at 72°C for 1 min; (2) 45 cycles of denaturation at 94°C for 40 s, annealing at 60°C for 40 s, extension at 72°C for 1 min; and (3) extension at 72°C for 10 min.

Characterization of a mouse CAP cDNA The PCR reaction was electrophoresed through a 4% NuSieve (FMC BioProducts) agarose gel. The 586 nucleotide band was excised from the gel, and the DNA was recovered by treatment of the gel slice with β-agarase (New England Biolabs). The DNA was then cloned directly into pT7Blue (Novagen). The presence of a CAP mouse homologue was confirmed by sequence analysis of clones using Sequenase Version 1.0 or 2.0 (USB), following the manufacturer’s directions. A positive mouse clone was used to screen a λZap cDNA library derived from a mouse erythroleukemia cell line, which was generously provided by Dr Robert Benezra (Benezra et al., 1990). The largest clone obtained was approximately 1.9 kb in length. Sequence analysis revealed that this clone, pCAP 3-1, contained 0.97 kb of CAP coding sequence and approximately 0.9 kb of 3′ untranslated region. To obtain the 5′ end of the CAP gene, the RACE (Rapid Cloning of cDNA Ends) PCR strategy was employed (Frohman, 1990). The oligonucleotide primer for the RACE RT reaction was: 5′ AGAATGGTTTGGGGCCACTC 3 ′. The RT reaction was performed as described (Frohman, 1990), except Superscript RT (BRL) and 10× PCR-HB were substituted for their respective counterparts in the reaction. The oligonucleotide primers for the RACE PCR are listed below and are named using the nomenclature of Frohman: (1) dT15-Ri-Ro: 5′ CGACGGCCAGTGAATTGTAATACGACTCACTATAGGCGCTCGAGT15 3′; (2) Gene-specific primer 1 (GSP1): 5′ AACTCCTTGATGTAAGC 3 ′; (3) R o: 5 ′ CCAGTGAATTGTAATAC 3′; (4) GSP2: 5′ CATTCTAGAATGTAAGCCTGCAGCTC 3 ′; (5) R i: 5′ AATACGACTCACTATAG 3 ′. The RACE PCR reaction was electrophoresed through a 4% NuSieve agarose gel. Two abundant DNA fragments of approximately 0.7 and 0.53 kb were recovered from the gel, digested with XhoI and XbaI, and cloned into XhoI-XbaI digested pBluescript II (Stratagene). Sequence analysis revealed that clones of both sizes were CAP derived sequences. The larger of the two clones contained 586 nucleotides of CAP coding sequence, 64 nucleotides of 5′ untranslated region, and approximately 35 Ts (generated from the tailing reaction of the RACE protocol).

Plasmids pB1, which contains the CAP coding sequence from amino acid 1 to 198 in pBluescript II (Stratagene), was constructed by reverse transcription of 10T1/2 RNA (as above) followed by 30 cycles of PCR, using Vent polymerase (New England Biolabs). Vent polymerase was chosen due to its relatively low error rate. The sequence of the forward oligonucleotide primer was: 5′ CGTGGATCCACCATGGCTGACATGCAAAATC 3 ′. The sequence of the reverse oligonucleotide primer was: 5′ CATTCTAGAATGTAAGCCTGCAGCTC 3 ′; the reverse oligonucleotide primer was also used to prime the RT reaction. Plasmid pR2 was constructed by ligation of the following three fragments: (1) the 0.6 kb BamHI-PstI fragment derived from pB1, which contains the coding sequence for the first 193 amino acids of CAP; (2) the 0.9 kb PstI-XbaI fragment derived from the λZap cDNA clone pCAP 3-1, which contains the CAP coding sequence from amino acids 194 to 474 and 74 nucleotides of 3′ untranslated region; and (3) BamHI-XbaI digested pBluescript II (Stratagene). Plasmid pADMCAP was constructed by inserting the SalI and

779

XbaI (Klenow treated) fragment derived from pR2 into the S. cere visiae expression vector pAD4∆ (Ballester et al., 1989) at the SalISmaI sites. pADMCAP contains the entire coding sequence of mouse CAP under control of the yeast ADH1 promoter. Plasmid pCSMCAP expresses the coding region of the CAP cDNA under control of the simian cytomegalovirus IE94 promoter of pCS+ (D. L. Turner, R. A. W. Rupp, and H. Weintraub, unpublished). The CAP gene in pCSMCAP contains six copies of the thirteen amino acid Myc epitope (Roth et al., 1991) inserted inframe and 5′ to the CAP ATG. The mouse CAP gene was derived by reverse transcription of 10T1/2 RNA followed by 30 cycles of PCR using Vent polymerase and the following two oligonucleotide primers: (1) 5′ CGTGGATCCACCATGGCTGACATGCAAAATC 3 ′; (2) 5′ CGTAGATCTAGAGCTTATCCAGCGATTTCTG 3 ′. The RT reaction was primed with the latter oligonucleotide primer.

Sequence analysis The mouse CAP coding region, 5′ untranslated region, and 3′ untranslated region between the stop codon and the XbaI site at nucleotide 1506 were sequenced on both strands by a combination of the following methods: (1) Sequenase (USB), according to manufacturer’s directions; (2) Taq dye primer cycle sequencing (ABI); and (3) Taq dye terminator cycle sequencing (ABI). The dye primer and dye terminator sequencing reactions were analyzed on an Applied Biosystems DNA sequencing system. The sequence of CAP from nucleotide 454 to 1506 was derived from the CAP cDNA clone. The sequence of CAP from nucleotide +1 to 587 was derived from two independent PCR clones to avoid errors arising from polymerase misincorporation during the PCR.

Northern analysis RNA was prepared from mouse tissues by extraction with acid guanidium thiocyanate-phenol-chloroform as described by Chomczynski and Sacchi (1987). Total RNA was electrophoresed through a 1% formaldehyde agarose gel and transferred overnight to Hybond (Amersham) as described (Sambrook et al., 1989). RNA concentrations were quantitated by spectrophotometric readings and confirmed by ethidium staining of a formaldehyde agarose gel and by methylene blue staining of the Hybond filter after hybridization. The probe was prepared by PCR. The PCR reaction contained the following components per 8 µl reaction: 40 µCi dCTP (NEN); 2 µg reverse oligonucleotide primer; 1 µl template DNA; 2 µl dNTP/PCR buffer (2 µl 0.2 mM dNTPs, excluding dCTP, 2 µl 10× PCR buffer, 0.5 µl Taq polymerase per 5 µl). The template for the reaction was the 1.06 kb CAP fragment from nucleotide 262 to 1326. The reaction conditions consisted of 30 cycles of 30 s at 94°C, 30 s at 55°C, followed by 30 s at 72°C. Unincorporated nucleotides were removed prior to hybridization by Sephadex G50 chromatography. Hybridization of the filter was performed following the manufacturer’s instructions (Amersham).

Immunofluorescence methods Twenty-four hours after transfection, 10T1/2 cells were transferred from 60 mm dishes to polylysine treated glass cover slips. Twentyfour or fourty-eight hours later, the cells were fixed in 3% paraformaldehyde in phosphate buffered saline (PBS) for 10 min, permeabilized with 0.25% Triton in PBS for 5 min, and then incubated with the 9E10 monoclonal hybridoma supernatant for 1.5 hours at room temperature. This was followed by a 60 min incubation with a 1:300 dilution of FITC-conjugated goat anti-mouse IgG in PBS (Jackson ImmunoResearch). Cells were viewed on a Bio-Rad MRC600 confocal microscope.

780


RESULTS Cloning and sequence of the mouse CAP gene A 586 nucleotide fragment with significant homology at the amino acid level to the carboxy-terminal domains of the S. cerevisiae and S. pombe genes was isolated by a coupled reverse transcription/polymerase chain reaction strategy. In the first step, RNA, primed with random hexamers, was transcribed with reverse transcriptase. The complementary DNA, or cDNA, generated in this first step then served as template for the polymerase chain reaction (PCR), which was primed with degenerate oligonucleotide primers. Primer sequences were based on several stretches of five consecutive amino acids located in the carboxy-terminal domains of the S. cerevisiae and S. pombe CAP proteins. Three primer combinations were tested on cDNA samples from a variety of cell lines and organisms. One primer pair generated a fragment of the expected size when used in GGAAG

TGGAG

AGCGG

CTGAT

CGCAG

TCCGG

AGGTG

combination with cDNA from 10T1/2 mouse fibroblasts or mouse embryonic stem cells. Sequence analysis of the fragment revealed significant homology at the amino acid level to the carboxy-terminal domain of the yeast CAP proteins. The PCR fragment was then used to screen a lambda phage cDNA library. The largest CAP clone isolated contained a 1.9 kb insert, but was not full length; the 5′ end of the CAP gene was isolated by the rapid amplification of cDNA ends (RACE) PCR technique (Frohman, 1990). The nucleotide and predicted amino acid sequence of the mouse CAP gene is shown in Fig. 1. The longest open reading frame encodes 474 amino acids. Stop codons are present in all three reading frames 5′ to the first methionine of the open reading frame. The presence of the stop codons and homology to the yeast CAP proteins in this region suggests that the first methionine of the open reading frame is the initiating methionine. A notable feature of the protein is the stretch of prolines from AGGCG

GAACT

CTTAA

GCAGG

TGGAA CGTT

ATG GCT GAC ATG CAA AAT CTT GTA GAA AGA TTG GAG AGG GCA GTG GGC CGC CTG GAG GCA GTG M A D M Q N L V E R L E R A V G R L E A V

63 21

TCA CAT ACT TCA GAC ATG CAC TGT GGA TAT GGA GAC AGC CCT TCA AAA GGA GCA GTT CCA TAT GTG S H T S D M H C G Y G D S P S K G A V P Y V

129 43

CAA GCA TTT GAC TCG CTG CTT GCC AAT CCC GTG GCA GAG TAC TTG AAG ATG AGT AAG GAG ATC GGG Q A F D S L L A N P V A E Y L K M S K E I G

195 65

GGA GAT GTG CAG AAA CAC GCG GAG ATG GTC CAC ACA GGC CTG AAG TTG GAG CGA GCT CTC CTG GCT G D V Q K H A E M V H T G L K L E R A L L A

261 87

ACA GCT TCT CAG TGC CAG CAG CCA GCT GGT AAT AAA CTT TCT GAT TTG TTG GCA CCT ATC TCG GAG T A S Q C Q Q P A G N K L S D L L A P I S E

327 109

CAG ATC CAA GAA GTT ATA ACC TTC CGG GAG AAG AAC CGA GGC AGC AAG TTT TTT AAT CAT CTA TCT Q I Q E V I T F R E K N R G S K F F N H L S

396 131

GCT GTC AGT GAA AGC ATC CAG GCC TTG GGC TGG GTG GCT CTG GCT GCG AAA CCT GGC CCC TTT GTG A V S E S I Q A L G W V A L A A K P G P F V

459 153

AAA GAG ATG AAT GAC GCG GCC ATG TTT TAC ACA AAT CGT GTC CTC AAG GAG TAC AGA GAT GTG GAT K E M N D A A M F Y T N R V L K E Y R D V D

525 175

AAG AAG CAT GTG GAC TGG GTC AGA GCT TAC TTG AGT ATA TGG ACG GAG CTG CAG GCT TAC ATC AAG K K H V D W V R A Y L S I W T E L Q A Y I K

591 197

GAG TTT CAT ACT ACT GGC CTG GCC TGG AGC AAG ACG GGG CCT GTG GCA AAA GAA CTG AGT GGA TTG E F H T T G L A W S K T G P V A K E L S G L

657 219

CCA TCT GGA CCC TCT GTG GGA TCA GGC CCA CCT CCT CCC CCA CCG GGC CCG CCT CCT CCC CCA ATC P S G P S V G S G P P P P P P G P P P P P I

723 241

CCT ACC AGT TCT GGT TCT GAC GAC TCT GCA TCA CGC TCA GCA CTG TTT GCA CAG ATT AAT CAG GGG P T S S G S D D S A S R S A L F A Q I N Q G

789 263

GAA AGC ATC ACA CAT GCC CTG AAA CAT GTA TCT GAT GAC ATG AAG ACT CAC AAG AAC CCT GCC CTG E S I T H A L K H V S D D M K T H K N P A L

855 285

AAA GCT CAG AGC GGT CCA GTT CGG AGT GGC CCC AAA CCA TTC TCT GCA CCT AAA CCC CAA ACT AGC K A Q S G P V R S G P K P F S A P K P Q T S

921 307

CCC TCC CCC AAA CCA GCC ACA AAG AAG GAA CCA GCT CTG CTG GAA CTG GAA GGC AAG AAA TGG AGA P S P K P A T K K E P A L L E L E G K K W R

987 329

GTG GAA AAC CAG GAG AAT GTT TCT AAC CTG GTG ATT GAT GAC ACT GAG CTG AAG CAG GTG GCT TAC V E N Q E N V S N L V I D D T E L K Q V A Y

1053 351

ATC TAC AAG TGT GTC AAC ACA ACA TTG CAA ATC AAG GGC AAA ATT AAC TCC ATT ACA GTA GAT AAC I Y K C V N T T L Q I K G K I N S I T V D N

1119 373

TGT AAG AAG CTT GGC CTG GTG TTT GAT GAC GTG GTG GGC ATT GTG GAG ATA ATC AAT AGT AGG GAT C K K L G L V F D D V V G I V E I I N S R D

1185 395

GTC AAA GTT CAG GTG ATG GGA AAA GTG CCA ACC ATT TCC ATT AAC AAA ACA GAT GGC TGC CAT GCT V K V Q V M G K V P T I S I N K T D G C H A

1251 417

TAC CTG AGC AAG AAC TCC CTG GAC TGT GAG ATA GTC AGT GCC AAA TCT TCT GAG ATG AAT GTC CTC Y L S K N S L D C E I V S A K S S E M N V L

1317 439

ATT CCT ACC GAA GGC GGT GAT TTT AAC GAG TTC CCA GTC CCC GAG CAG TTC AAG ACC CTG TGG AAC I P T E G G D F N E F P V P E Q F K T L W N

1383 461

GGA CAG AAG TTG GTC ACC ACA GTG ACA GAA ATC GCT GGA TAA GCAAA TACAT GGGTC CTGTG CCCTC T G Q K L V T T V T E I A G *

1458 474

CCCT TCACA CCATG GGATA CATCT GTATG GAGAC GGTTC TTTTC TAGA

Fig. 1. The DNA and predicted amino acid sequences of the mouse CAP cDNA. The predicted amino acid sequence of the CAP open reading frame begins with the ATG at nucleotide +1 and extends to the termination codon, indicated by the asterisk, at nucleotide 1425.

Characterization of a mouse CAP cDNA SpCAP MSDMINIRETGYNFTTILKRLEAATSRLEDLVESGHKPLPNMHRPSRDSNSQTHNISFNIGTPTAPTVSTGSPAVASLHDQVAAAISPRNRSLTS | || | :: ||| | ||| : : ||: MCAP MADMQN----------LVERLERAVGRLEAVSHTSDMHCGYGDSPSKGAVP-------------------------------------------| | |: ||| | :||| |: : || | ScCAP MPDSKYTMQ-GYNLVKLLKRLEEATARLEDVTIYQEGYIQNKLEASKNNKPS---------------------------DSGADANTTNEPSAEN

95

SpCAP TSAVEAVPASISAYDEFCSKYLSKYMELSKKIGGLIAEQSEHVEKAFNLLRQVLSVALKAQKPDMDSPELLEFLKPIQSELLTITNIRDEHRTAP : |:| : |: :|| ||| : | | : | | :| | | | : | || : : |: | MCAP ---------YVQAFDSLLANPVAEYLKMSKEIGGDVQKHAEMVHTGLKLERALLATASQCQQPAG--NKLSDLLAPISEQIQEVITFREKNRGSK : || | :: : : :| | | :: | | | | | | || |:| : :| || || ScCAP APEVEQDPKCITAFQSYIGENIDPLVELSGKIDTVVLDALQLLKGGFQSQLTFLRAAVRSRKPDYSSQTFADSLRPINENIIKLGQLKESNRQSK

190

41 67

125 162

SpCAP EFNQLSTVMSGISILGWVTVEPTPLSFMSEMKDSSQFYANRVMKEFKGKDDLQIEWVRSYLTLLTELITYVKTHFKTGLTWS--TK------QDA || || | | |||| : | |: || | || ||::||:: | ::||| ||:: ||| |:| ||| || | MCAP FFNHLSAVSESIQALGWVALAAKPGPFVKEMNDAAMFYTNRVLKEYRDVDKKHVDWVRAYLSIWTELQAYIKEFHTTGLAWS---KTG------:| |||:|| |||: | | : ||| |:|||:|||||: | |:||: :| | |||||:||||: | | | ScCAP YFAYLSALSEGAPLFSWVAVD-TPVSMVTDFKDAAQFWTNRILKEYRESDPNAVEWVKKFLASFDNLKAYIKEYHTTGVSWK---KDGMDFADAM

277

SpCAP VPLKTALANLSASKTQAPSSGDSANGGLPPPPPPPPP--SNDFWKDSNEPAPADNKGDMGAVFAEINKGEGITSGLRKVDKSEMTHKNPNLR-KT |: | | || | | | ||||| ||| | : |:|| || || || :|| | ||||| |: : MCAP -PVAKELSGLPS----GPSVG-S--GP-PPPPPGPPPPPI-------PTSSGSDDSASRSALFAQINQGESITHALKHVSDDMKTHKNPALKAQS | : | || : :| ||||| || : | :| : |:|| :|||| || :|| | ||||| || || ScCAP AQSTKNTGATSS-----PSPASATAAPAPPPPPPAPPASVFEISNDTPATSSDANKGGIGAVFAELNQGENITKGLKKVDKSQQTHKNPELR-QS

369

SpCAP GPT-PGPKP--KIKSSA-PSKPAETAPVKPPRIELENTKWFVENQVDNHSIVLDSVELNHSVQIFGCSNCTIIIKGKLNTVSMSNCKRTSVVVDT || |||| | | | :||| || |||| :|:| || |: | | |: ||||:|:::: |||: :| | MCAP GPVRSGPKPFSAPKPQTSPS-PKPCTKKEPALLELEGKKWRVENQENVSNLVIDDTELKQVAYIYKCVNTTLQIKGKINSITVDNCKKLGLVFDD | | | | | | | | || | || :|| || : :: | :| || : :|||||:| |:: :| | ScCAP STV-SSTG--SKSGPPPRPKKPSTLKTKRPPRKELVGNKWFIENYENETESLVIDANKDESIFIGKCSQVLVQIKGKVNAISLSETESCSVVLDS

460

SpCAP LVAAFDIAKCSNFGCQVMNHVPMIVIDQCDGGSIYLSKSSLSSEVVTSKSTSLNINVP-NEEGDYAERAVPEQIKHKVNEKGELVSEIVRHE |: :| ||| || | | || |||| || |:| ||: :|: :| | ||: | |||| | : ||: : VVGIVEIINSRDVKVQVMGKVPTISINKTDGCHAYLSKNSLDCEIVSAKSSEMNVLIP-TEGGDFNEFPVPEQFK-TLWNGQKLVTTVTEIAG : :::| | :|| :| ||| |:|| |||| || || : |: :|| :| | |: |||:||| | : | | : | | || ScCAP SISGMDVIKSNKFGIQVNHSLPQISIDKSDGGNIYLSKESLNTEIYTSCSTAINVNLPIGEDDDYVEFPIPEQMKHSFADG-KFKSAVFEHAG

MCAP

781

210 253

289 342

383 434 551 474 526

Fig. 2. Comparative sequence analysis of CAP proteins. The yeast CAP proteins were aligned to the mouse CAP protein with the IBI, Inc. MacVector program. To maximize homologies, minor adjustments to the computer alignment were made by eye. Vertical bars (|) indicate identities; colons (:) indicate similarities. Similar amino acids, as specified by the Jimenez-Montano and Zamora-Cortina alphabet, are grouped as follows: V, L, I, M; F, Y, W; K, R; E, D; Q, N; S, T; and A, G. SpCAP: S. pombe CAP; MCAP: mouse CAP; ScCAP: S. cerevisiae CAP.

amino acids 229 to 240, that separates the amino and carboxy-terminal domains. This polyproline rich region is conserved among all CAP proteins identified to date, although its function has not yet been delineated. The mouse CAP protein shows significant homology to the CAP proteins from the yeasts S. cerevisiae and S. pombe (38% identity). An alignment of the mouse CAP protein to the yeast proteins is shown in Fig. 2. Of the two CAP functional domains, the carboxy-terminal region (amino acids 286 to 526 of S. cerevisiae, 315 to 551 of S. pombe, and 241 to 474 of mouse CAP) is the most highly conserved among the three proteins. The carboxy-terminal domain of mouse CAP shows 38% and 39% identity to the S. cere visiae and S. pombe proteins, respectively; the yeast proteins are 39% identical over this region. The amino-terminal domains of the three proteins are less well conserved. The amino-terminal domain of mouse CAP shows 26% and 27% identity to the S. cerevisiae and S. pombe CAP proteins, respectively. The mouse CAP protein is as related to the yeast proteins in this region as they are to each other (24% identity). One noteworthy structural feature conserved in the amino-terminal domains of the three proteins is the four-three hydrophobic repeat located at amino acids 17 to 31 in S. pombe CAP, 7 to 21 in mouse CAP, and 16 to 30 in S. cerevisiae CAP. This region has the potential to form an amphipathic helix in which the charged residues are positioned on one face of the helix and the hydrophobic

residues on the other face. Such helices have been implicated in protein-protein interactions (Busch and SassoneCorsi, 1990; Cohen and Parry, 1986). Expression of CAP RNA The tissue distribution of mouse CAP RNA was investigated by northern blot analysis. A single CAP message was found in all tissues examined, but the message level varied (Fig. 3). The CAP message was most prevalent in brain, kidney, spleen, and uterus. The message was moderately expressed in heart and liver. Skeletal muscle exhibited a low but detectable level of message. In addition, CAP RNA

Fig. 3. Northern blot of total RNA from mouse tissues and cell lines, probed with the 1.06 kb CAP fragment (nucleotides 2621326). Lane 1, brain; lane 2, heart; lane 3, kidney; lane 4, liver; lane 5, skeletal muscle; lane 6, spleen; lane 7, uterus; lane 8, embryo (day 10 of gestation); lane 9, 10T1/2 mouse fibroblast cell line. Lanes 1 to 7 contain 40 µg of total RNA; lanes 8 and 9 contain 10 µg of total RNA.

782


Fig. 4. Immunolocalization of MCAP. Mouse fibroblasts were transiently transfected with a vector expressing the mouse CAP protein fused to six copies of the Myc epitope at its N-terminus (MT-MCAP), and then processed for indirect immunofluorescence as described in Materials and Methods. Pronounced cytoplasmic staining of MT-MCAP is observed (A,B). In addition, increased staining of MTMCAP is found in the leading edges (see arrows) of migrating cells (C,D). Bars, 25 µM.

was highly expressed in mouse fibroblasts and in mouse embryos at day 10 of gestation.

cell in the center of the field is surrounded by five negative cells.

Immunolocalization of CAP The intracellular distribution of an epitope-tagged mouse CAP protein was determined by indirect immunofluorescence. Mouse fibroblasts were transiently transfected with the vector pCSMCAP, which expresses the mouse CAP protein fused at its N terminus to six copies of the Myc epitope tag (MT-MCAP). Two or three days after transfection, cells expressing MT-MCAP were detected by staining with the 9E10 monoclonal antibody, which recognizes the Myc epitope tag (Evan et al., 1985). Pronounced staining of MT-MCAP was observed in the cytoplasm of mouse fibroblasts (Fig. 4A,B). In addition, MT-MCAP also appeared to localize to the leading edge of migrating cells (Fig. 4C,D). The 9E10 monoclonal antibody specifically recognizes the Myc epitope-tagged CAP protein: when mouse fibroblasts were transiently cotransfected with plasmids expressing the epitope-tagged MCAP or β-galactosidase at a 3:1 ratio, all β-galactosidase-positive cells showed staining with the 9E10 antibody; in addition, cells transfected with the control vector pCS showed no staining with the antibody. This specificity of staining is illustrated in Fig. 4D, where the intensely stained MT-MCAP-positive

Expression of mouse CAP in S. cerevisiae Deletion of the CAP gene in S. cerevisiae leads to an inability to grow in rich medium (YPD), in synthetic medium containing 6.5 mM valine, and in synthetic medium at 37°C (Field et al., 1990; Gerst et al., 1991). Expression of the carboxy-terminal domain of CAP is sufficient to restore growth under these conditions and, therefore, these growth defects have been designated CAP carboxy-terminal defects (Gerst et al., 1991). To determine if expression of the mouse CAP gene could complement defects associated with loss of the carboxy-terminal domain of CAP in S. cerevisiae, the mouse CAP cDNA was placed under control of the yeast ADH1 promoter in the yeast expression vector pAD4∆. This expression plasmid, pADH-MCAP, was then introduced into SKN32 (cap::HIS3). SKN32 cells expressing pADH-MCAP grew in YPD, in synthetic medium containing 6.5 mM valine, and in synthetic medium at 37°C (Fig. 5). Growth of SKN32 cells transformed with pADH-MCAP was similar to that of SKN32 cells transformed with pADH-CAP, the S. cerevisiae CAP expression plasmid. SKN32 cells transformed with pAD4∆, the vector control, were unable to

Characterization of a mouse CAP cDNA

783

Fig. 6. Effect of MCAP on RAS2val19 cap cells. The RAS2val19 cap strain SKN37 was transformed with the indicated plasmids and plated onto SC-Leu plates. Independent transformants were patched onto SC-Leu plates, incubated at room temperature, and then replica plated to SC-Leu plates, which had been prewarmed at 55°C. These plates were then further incubated at 55°C for 0 or 5 min. After heat shock treatment, the plates were incubated at room temperature for 3 days. Eight independent transformants behaved identically. Fig. 5. Complementation of the S. cerevisiaecap strain by overexpression of MCAP. The cap strain SKN32 was transformed with the indicated plasmids and plated onto SC-Leu plates. Independent transformants were patched onto SC-Leu plates, incubated at room temperature, and then replica plated to SC-Leu with valine (SC+VAL), YPD, and SC-Leu (SC), and incubated at room temperature or, as indicated, at 37°C. Eight independent transformants behaved identically.

grow under these conditions. Thus, the mouse CAP cDNA complements defects associated with loss of the carboxyterminal domain of yeast CAP. A yeast strain with an activated RAS allele, RAS2val19, is exquisitely sensitive to a 55°C heat shock (Toda et al., 1985). Deletion of the CAP gene in such a strain leads to heat shock resistance (Field et al., 1990). Expression of the amino-terminal domain of CAP is sufficient to restore heat shock sensitivity and, therefore, this defect has been designated a CAP amino-terminal defect (Gerst et al., 1991). To determine if expression of pADH-MCAP could complement defects associated with loss of the amino-terminal domain of CAP, pADH-MCAP was introduced into SKN37 (RAS2val19 cap::HIS3). SKN37 cells expressing pADHMCAP were resistant to a 55°C heat shock, as were SKN37 cells transformed with pAD4∆, the vector control (Fig. 6). SKN37 cells transformed with pADH-CAP, the yeast CAP expression plasmid, were sensitive to the 55°C heat shock. Therefore, mouse CAP is unable to complement defects associated with loss of the amino-terminal domain of yeast CAP. In addition, transformation of R1′ (RAS2val19) with pADH-MCAP did not alter the response of R1′ to a 55°C heat shock treatment nor affect its growth in rich medium (data not shown). This result indicates that overexpression of mouse CAP does not appreciably interfere with normal S. cerevisiae CAP function. DISCUSSION The CAP protein has been identified in the budding yeast

Saccharomyces cerevisiae and in the fission yeast Schizosaccharomyces pombe (Field et al., 1990; FedorChaiken et al., 1990; Kawamukai et al., 1992). In both organisms, CAP has at least two functions, which require separate domains of the protein. The amino-terminal domain of CAP associates tightly with adenylyl cyclase. The carboxy-terminal domain of CAP appears to play a role in the regulation and/or organization of cellular morphology. Each of these domains is capable of functioning independently of the other. But, since the two domains are present in the same protein in two quite divergent organisms, it seems likely that the two functions are coordinated in some fashion. We report here that CAP is not restricted to single cell eukaryotes, but is present in mouse, where it is ubiquitously expressed. We have also isolated the region encoding the carboxy-terminal domain of the CAP gene from a human glioblastoma cell line library (data not shown). During the preparation of this manuscript, the sequence of a human CAP cDNA clone from this library was reported (Matviw et al., 1992). Mouse and human CAP are 95% identical at the amino acid level. The mouse CAP protein is highly related to the yeast CAP proteins. An alignment of the three CAP proteins revealed greatest homology in residues 169 to 367, a region that appears functionally silent in S. cerevisiae (Gerst et al., 1991). Conceivably, this region serves to coordinate the functions of the amino and carboxy-terminal domains. The region of next highest similarity is the carboxy-terminal domain. In addition to conservation of primary sequence, the function of the CAP carboxy-terminal domain is conserved in these disparate organisms. The S. pombe CAP protein complements defects associated with loss of the S. cerevisiae CAP carboxy-terminal domain (Kawamukai et al., 1992); the S. cerevisiae CAP protein complements defects associated with the loss of the S. pombe CAP carboxy-terminal domain (Kawamukai et al., 1992); and we have shown here that the mouse CAP protein complements

784


defects associated with loss of the S. cerevisiae CAP carboxy-terminal domain. The sequences of the amino-terminal domains of the CAP proteins are less well conserved. Consistent with this lower degree of conservation is the observation that the amino-terminal domains are unable to function in heterologous systems. Expression of mouse CAP is not sufficient to complement defects associated with loss of amino-terminal CAP function in S. cerevisiae. Expression of S. pombe CAP is also not sufficient to complement defects associated with loss of amino-terminal CAP function in S. cerevisiae (Kawamukai et al., 1992). Yet both S. cerevisiae and S. pombe CAP tightly associate with their cognate adenylyl cyclases through their amino-terminal domains (Field et al., 1990; Gerst et al., 1991; Kawamukai et al., 1992). Likewise, mouse CAP may also associate with an adenylyl cyclase activity. One intriguing feature of the CAP amino-terminal domain is a potential amphipathic helix located at amino acids 7 to 21 of mouse CAP, and similarly located in the yeast CAP proteins. Such helices have been implicated in protein-protein interactions (Busch and Sassone-Corsi, 1990; Cohen and Parry, 1986). This region may facilitate an interaction between CAP and adenylyl cyclase. Consistent with this idea is the observation that deletion of amino acids 2 to 31 is sufficient to render the S. cerevisiae CAP/adenylyl cyclase complex unresponsive to RAS (Gerst et al., 1991). Loss of S. cerevisiae carboxy-terminal CAP function is associated with a number of seemingly unrelated morphological and nutritional defects. Cells that lack carboxy-terminal CAP function are altered in cell shape and size, bud randomly, and contain disorganized actin cytoskeletons (Field et al., 1990; Vojtek et al., 1991). In addition, cap cells do not grow at elevated temperatures, in rich medium, or in synthetic medium containing excess valine (Field et al., 1990; Gerst et al., 1991). All of these defects are suppressed by overexpression of profilin (Vojtek et al., 1991), an actin and phospholipid binding protein (Pollard and Cooper, 1986; Lassing and Lindberg, 1985, 1988). Recently, an actin sequestering protein ASP56 was purified from pig platelets (Gieselman and Mann, 1992). Tryptic peptides of this protein show homology to yeast CAP and extensive homology to mouse CAP. On the basis of this homology, we speculate that mouse CAP also interacts with actin. Fractionation of mouse fibroblasts that expressed the Myc epitope-tagged mouse CAP protein indicated that the majority of CAP was not associated with the detergent insoluble cytoskeletal matrix (data not shown). This is the expected result if CAP is not a major component of the actin cytoskeletal matrix, but is instead an actin monomer sequestering protein, as is ASP56. To suppress cap defects in S. cerevisiae, some function of profilin other than, or in addition to, actin binding is required (Vojtek et al., 1991). This other function is most likely PIP2 binding. Whether CAP also binds PIP2, as does profilin, has not yet been investigated. In S. cerevisiae, CAP is a likely candidate to integrate nutritional signals through the RAS/cAMP pathway to the spatial organization of cell growth. Our immunolocalization experiments show that the mouse CAP protein is appropriately poised to play a similar role in fibroblasts: although

predominantly cytoplasmic, CAP is found associated with the leading edge of migrating cells, and hence may coordinate growth signals to the cytoskeletal reorganization that enables cell movement. Intriguingly, profilin is enriched in highly dynamic lamellopodia of rat fibroblasts (Buss et al., 1992) and, in activated platelets, an increase in the localization of profilin to the plasma membrane has been reported (Hartwig et al., 1989). Therefore, it would not be surprising if profilin and CAP shape similar events in mammalian cells, as they appear to do in the yeast S. cerevisiae. Reorganization of the actin cytoskeleton is a common response to growth signals. In S. cerevisiae, CAP is required for transduction of growth signals through the RAS/cAMP pathway and for spatial organization of the actin cytoskeleton. Likewise, mouse CAP may participate in the regulation and/or organization of the actin cytoskeleton in higher eukaryotes. The identity of the signal transduction pathway linked by mouse CAP to the rearrangement of the actin cytoskeletal matrix in higher eukaryotes, and whether such a pathway involves an adenylyl cyclase activity, as in yeast, will be the focus of future studies. We thank Dave Turner for insightful discussions and for critically reading the manuscript. We also thank Mike Wigler for providing the S. pombe CAP sequence prior to publication, Andrew Waskiewicz and Kathy Neary for valuable comments on the manuscript, and Mary Kay Dolejsi for assistance with the automated sequencing. This work was supported by a grant from the National Institutes of Health (J.A.C.). The mouse CAP cDNA sequence is available from the EMBL/GenBank and DDBJ databases under accession number L12367.

REFERENCES Archer, B., Ozcelik, T., Jahn, R., Francke, U. and Sudhof, T. (1990). Structures and chromosomal localizations of two human genes encoding synaptobrevins 1 and 2.J. Biol. Chem. 265, 17267-17273. Ballester, R., Michaeli, T., Ferguson, K., Xu, H., McCormick, F. and Wigler, M. (1989). Genetic analysis of mammalian GAP expressed in yeast. Cell 59, 681-686. Benezra, R., Davis, R., Lockson, D., Turner, D. and Weintraub, H. (1990). The protein Id: a negative regulator of helix-loop-helix DNA binding proteins. Cell 61, 49-59. Busch, S. and Sassone-Corsi, P. (1990). Dimers, leucine zippers and DNAbinding domains. Trends Genet. 6, 36-40. Buss, F., Temm-Grove, C., Henning, S. and Jockusch, B. (1992). Distribution of profilin in fibroblasts correlates with the presence of highly dynamic actin filaments.Cell Motil. Cytoskel. 22, 51-61. Chomczynski, P. and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156-159. Cohen, C. and Parry, D. (1986). Alpha-helical coiled-coils: a widespread motif in proteins. Trends Biochem. Sci. 11, 245-248. Elferink, L., Trimble, W. and Scheller, R. (1989). Two vesicle-associated membrane protein genes are differentially expressed in the rat central nervous system. J. Biol. Chem. 264, 11061-11064. Evan, G., Lewis, G., Ramsay, G. and Bishop, J. (1985). Isolation of monoclonal antibodies specific for human c-myc proto-oncogene product. Mol. Cell. Biol. 5, 3610-3616. Fedor-Chaiken, M., Deschenes, R. and Broach, J. (1990). SRV2, a gene required for RAS activation of adenylate cyclase in yeast. Cell 61, 329340. Field, J., Vojtek, A., Ballester, R., Bolger, G., Colicelli, J., Ferguson, K., Gerst, J., Kataoka, T., Michaeli, T., Powers, S., Riggs, M., Rodgers, L., Wieland, I., Wheland, B. and Wigler, M. (1990). Cloning and characterization of CAP, the S. cerevisiae gene encoding the 70 kd adenylyl cyclase associated protein. Cell 61, 319-327.

Characterization of a mouse CAP cDNA Frohman, M. (1990). Rapid amplification of cDNA ends (RACE): user friendly cDNA cloning. Amplifications: A Forum For PCR Users 5, 1115. Fukai, Y., Kozasa, T., Kaziro, Y., Takeda, T. and Yamamoto, M. (1986). Role of a ras homolog in the life cycle of Schizosaccharomyces pombe. Cell 44, 329-336. Gerst, J., Ferguson, K., Vojtek, A., Wigler, M. and Field, J. (1991). CAP is a bifunctional component of the Saccharomyces cerevisiae adenylyl cyclase complex. Mol. Cell. Biol. 11, 1248-1257. Gerst, J., Rodgers, L., Riggs, M. and Wigler, M. (1992). SNC1, a yeast homolog of the synaptic vesicle associated membrane protein/synaptobrevin gene family: genetic interactions with RAS and CAP genes. Proc. Nat. Acad. Sci. USA 89, 4338-4342. Gieselman, R. and Mann, K. (1992). ASP-56, a new actin sequestering protein from pig platelets with homology to CAP, an adenylate cyclase associated protein from yeast. FEBS Lett. 298, 149-153. Goldschmidt-Clermont, P., Kim, J., Machesky, L., Rhee, S. and Pollard, T. (1991). Regulation of phospholipase C by profilin and tyrosine phosphorylation. Science 251, 1231-1233. Goldschmidt-Clermont, P., Machesky, L., Baldassare, J. and Pollard, T. (1990). The actin binding protein profilin binds to PIP2 and inhibits its hydrolysis by phospholipase C. Science 247, 1575-1578. Hartwig, H., Chambers, K., Hopcia, K. and Kwiatkowski, D. (1989). Association of profilin with filament-free regions of human leukocyte and platelet membranes and reversible membrane binding during platelet activation. J. Cell Biol. 109, 1571-1579. Kawamukai, M., Gerst, J., Field, J., Riggs, M., Rodgers, L., Wigler, M. and Young, D. (1992). Genetic and biochemical analysis of the adenylyl cyclase associated protein, cap, in S. pombe. Mol. Biol. Cell 3, 167-180. Lassing, I. and Lindberg, U. (1985). Specific interaction between

785

phosphatidylinositol-4,5-bisphosphate and profilactin. Nature 314, 472474. Lassing, I. and Lindberg, U. (1988). Specificity of the interaction between phosphatidylinositol 4,5-bisphosphate and the profilin:actin complex. J. Cell. Biochem. 37, 255-267. Matviw, H., Yu, G. and Young, D. (1992). Identification of a human cDNA encoding a protein that is structurally and functionally related to the yeast adenylyl cyclase associated CAP proteins. Mol. Cell. Biol. 12, 50335040. Pollard, T. and Cooper, J. (1986). Actin and actin binding proteins. A critical evaluation of mechanisms and functions. Annu. Rev. Biochem.55, 987-1035. Rose, M., Winston, F. and Hieter, P. (1990). Methods in Yeast Genetics: a Laboratory Course Manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Roth, M., Zahler, A. and Stolk, J. (1991). A conserved family of nuclear phosphoproteins localized to sites of polymerase II transcription. J. Cell Biol. 115, 587-596. Sambrook, J., Fritch, E. F. and Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Toda, T., Uno, I., Ishikawa, T., Powers, S., Kataoka, T., Broek, D., Cameron, S., Broach, J., Matsumoto, K. and Wigler, M. (1985). In yeast, RAS proteins are controlling elements of adenylate cyclase. Cell 40, 27-36. Vojtek, A., Haarer, B., Field, J., Gerst, J., Pollard, T., Brown, S. and Wigler, M. (1991). Evidence for a functional link between profilin and CAP is the yeast S. cerevisiae. Cell 66, 497-505. (Received 29 January 1993 - Accepted 2 April 1993)