homologous uniqueg caenorhabditis elegans

CELL REGULATION, Vol. 2, 135-154, February 1991

Homologous and unique G protein alpha subunitsin the nematode Caenorhabditis elegans

Michael A. Lochrie,*t Jane E. Mendel,*$Paul W. Sternberg,** and Melvin 1. Simon*§tHoward Hughes Medical Institute*Division of BiologyCalifornia Institute of TechnologyPasadena, California 91125

A cDNA corresponding to a known G protein alphasubunit, the alpha subunit of G. (Goa), was isolatedand sequenced. The predicted amino acid sequenceof C. elegans G,a is 80-87% identical to other Goasequences. An mRNA that hybridizes to the C. ele-gans Goa cDNA can be detected on Northern blots.A C. elegans protein that crossreacts with an anti-bovine Goa antibody can be detected on immuno-blots. A cosmid clone containing the C. elegans G.agene (goa-1) was isolated and mapped to chromo-some 1. The genomic fragments of three other C.elegans G protein alpha subunit genes (gpa-1, gpa-2, and gpa-3) have been isolated using the poly-merase chain reaction. The corresponding cosmidclones were isolated and mapped to disperse lo-cations on chromosome V. The sequences of twoof the genes, gpa-1 and gpa-3, were determined.The predicted amino acid sequences of gpa-1 andgpa-3 are only 48% identical to each other. There-fore, they are likely to have distinct functions. Inaddition they are not homologous enough to G pro-tein alpha subunits in other organisms to be clas-sified. Thus C. elegans has G proteins that areidentifiable homologues of mammalian G proteinsas well as G proteins that appear to be unique toC. elegans. Study of identifiable G proteins in C.elegans may result in a further understanding oftheir function in other organisms, whereas studyof the novel G proteins may provide an understand-ing of unique aspects of nematode physiology.

IntroductionG proteins mediate transmembrane signaltransduction by physically coupling cell surface

t Current address: Department of Microbiology and Im-munology, University of California, San Francisco, CA 94143.§ Corresponding author.

receptors to effector proteins that influence in-tracellular metabolism (Gilman, 1987). A largefamily of G protein-linked receptors has beenidentified (Caron, 1989). All of these receptorshave seven putative transmembrane segmentsand bind small organic molecules or peptidehormones. The effectors regulated by G pro-teins that have been identified include adenylatecyclase, cGMP phosphodiesterase, phospholi-pases, and various ion channels.G proteins are composed of three subunits:

a, ,B, and y. The a subunit undergoes a guaninenucleotide exchange and hydrolysis cycle. Inmost cells it is the a subunit that directly inter-acts with effector proteins. G protein a subunitsare substrates for several covalent modifica-tions. They are ADP-ribosylated by pertussistoxin, cholera toxin, or both (Moss and Vaughan,1988). Some G protein a subunits are also my-ristoylated (Buss et al., 1987). Myristoylationoccurs on glycine-2 and is thought to be im-portant for localization of G protein a subunitsto the inner face of the cytoplasmic membrane(Jones et al., 1990; Mumby et al., 1990). G pro-tein a subunits are also phosphorylated (Gun-derson and Devreotes, 1990; reviewed by Sagi-Eisenberg, 1989).G proteins are members of a diverse family

of proteins with a variety of functions and tissuedistributions. G., the "other" G protein (Neeret al., 1984; Sternweis and Robishaw, 1984), isabundant in brain and is found only in organismswith nervous systems. Go has been found inDrosophila (deSousa et al., 1989; Schmidt et al.,1989; Thambi et al., 1989; Yoon et al., 1989),rat (Jones and Reed, 1987), cow (Van Meurs etal., 1987), hamster (Hsu et al., 1990), human(Lavu et al., 1988), and Xenopus (Olate et al.,1989) but not in yeast or the slime mold Dictyo-stelium. In vertebrates, G. is concentrated inneuropil (Gabrion et aL, 1989; Worley et al.,1986) and growth cones (Strittmatter et al.,1990). In Drosophila, G. has been found in brainand ovaries (deSousa et al., 1989; Schmidt etal., 1989; Thambi etal., 1989; Yoon etaL, 1989).Go may function in several signal transductionpathways (Serventi et al., 1990). In many cases

© 1991 by The American Society for Cell Biology 135

M.A. Lochrie et al.

it appears to be able to interact with the samereceptors as the closely related Gi proteins, al-though there are some exceptions (Senogles etal., 1990; Ueda et al., 1990). However, G. mayregulate a set of effectors distinct from thoseregulated by the Gi proteins. The Gi proteinsregulate adenylate cyclase and atrial potassiumchannels, whereas there is evidence that Goregulates calcium channels (Hescheler et al.,1987; Harris-Warrick et al., 1988; McFadzeanet al., 1989), neuronal potassium channels (VanDongen et al., 1988), and phospholipase C(Moriarty et al., 1989). G. has no direct effecton adenylate cyclase activity.To understand the role of G proteins in de-

velopment and the function of the nervous sys-tem and to study G proteins in an organismwhere an extensive genetic analysis could alsobe undertaken, we sought to characterize Gprotein genes in C. elegans. C. elegans is amodel organism for studying development andthe function of the nervous system (Kenyon,1988; Wood, 1988). It is the only multicellularorganism for which a complete cell lineage isknown (Sulston, 1988; Sulston et al., 1988). Thehermaphrodite contains 959 somatic nuclei; ofthese, 302 are neuronal. The entire neuroanat-omy and neuronal wiring diagram is known(White et al., 1986, 1988; Chalfie and White,1988). C. elegans exhibits many behaviors (re-viewed by Chalfie and White, 1988). It has manyof the neurotransmitters and signaling systemsfound in coelomates (Chalfie and White, 1988).Numerous mutants that affect developmentaland neuronal processes have been isolated(Hodgkin et al., 1988). Finally, the molecular ge-netics of C. elegans has advanced to the stagewhere gene transformation is facile (Fire, 1986;C. Mello, V. Ambros, J. Kramer, and D. Stinch-comb, personal communication) and an orderedset of cosmid clones covering most of the ge-nome is available to facilitate gene cloning andmapping (Coulson et al., 1986, 1988).Our first goal was to determine how many G

protein genes exist in C. elegans. The charac-terization of a ,B subunit gene (gpb-1; Van derVoorn et al., 1990) and an a subunit gene (gpa-2; Fino Silva and Plasterk, 1990) has been re-ported. This paper reports the molecular cloningand sequencing of a C. elegans Goa cDNAand the genes encoding two other apparentlynovel C. elegans G protein a subunits (gpa-1and gpa-3). These cloned genes can pro-vide tools to study their role in C. elegansbiology.

Results

Polymerase chain reactionsTo isolate C. elegans G protein a subunit genes,we used short, degenerate oligonucleotideprimers in polymerase chain reactions on ge-nomic DNA. The primers correspond to threeclosely spaced amino acid sequences that arehighly conserved in G protein a subunits. Theyrepresent three of the six regions that arethought to be involved in guanine nucleotidebinding and, therefore, would be expected to bepresent in most G protein alpha subunit se-quences. The distances between the oMP19,oMP20/oMP21 and the oMP19, Ta29 primerpairs in a cDNA would be -200 and 375 basepairs, respectively. Because genomic DNA wasamplified in this study, the PCR products couldbe larger if they contain introns. PCR amplifica-tion of C. elegans genomic DNA with the oMP19and oMP20 primer pair produced PCR productsof 270, 525, and 1000 base pairs. Amplificationusing oMP19 and oMP21 produced PCR prod-ucts of 260, 300, and 550 base pairs. Amplifi-cation using oMP19 and Ta29 produced PCRproducts of 500, 600, and 800 base pairs. The550 base pair PCR product from the oMP19 andoMP21 reaction, and the 500,600, and 800 basepair PCR products from the oMP19 and Ta29reaction, were subcloned. The sequences ofseveral subclones of each PCR product were an-alyzed. The sequences fell into three classesdesignated as gpa- 1, gpa-2, and gpa-3. Thesesequences contained amino acids that are foundin all G protein a subunits (Lochrie and Simon,1988). The 550 base pair PCR product corre-sponds to gpa-2, the 500 base pair PCR productto gpa-3, and the 600 base pair PCR product togpa-1. The 800 base pair PCR product derivedby amplification with the oMP19 and Ta29 prim-ers is an artifact of the PCR. Most of the sub-clones of this PCR product had truncated gpa-1or gpa-3 sequences fused to unknown se-quences. Two were gpa-3/gpa- 1 or gpa-2/gpa- 1hybrid sequences. Out of a total 85 subclonesthat were sequenced, 86% were gpa-1, gpa-2,or gpa-3 sequences. All three sequences con-tained introns. The positions of these intronswere identical to those of the sixth and seventhintrons found in the mammalian Gia, Goa, andTra genes (Kaziro et al., 1988; Raport et al., 1989).

G.a cDNA characterizationA mixture of the three PCR-derived gene frag-ments was used as a probe to screen a C. ele-

CELL REGULATION136

G< in C. elegans

gans cDNA library at low stringency. Fifty-twoof 45 000 plaques hybridized to the PCR-derivedprobes. Seven of these clones tested positiveby PCR analysis for the presence of G proteina subunit related sequences using the primerpairs oMP19, oMP20/21 or oMP19, Ta29. Theinsert of one of these clones (XCe6-2) was sub-cloned. The restriction map and sequencingstrategy for this cDNA are shown in Figure 1and the DNA sequence is shown in Figure 2.The open reading frame found in this sequenceencodes a protein that is clearly homologous toGoa sequences identified in other organisms.The predicted amino acid sequence of C. e/e-gans Goa is 870/o identical to Drosophila Goa,80% identical to Xenopus Goa, and 81-82%identical to human, rat, mouse, hamster, or bo-vine Goa (Figure 3). Two different forms of Goa,called G.a-A and G.a-B, have been describedfrom hamster (Hsu et al., 1990) and mouse(Strathmann et al., 1990). They differ in theirsixth and seventh exons as a result of alternativesplicing patterns. However, the C. elegans Goacorresponds to neither form of Goa because itis as different from G.a-A as it is from G.a-B.All of the motifs involved in GTP binding andhydrolysis (Lochrie and Simon, 1988) are highlyconserved in the C. elegans Goa sequence. Asite for myristoylation on rat Goa, Glycine-2(Mumby et al., 1990), is present. Amino acidsthat are substrates for cholera toxin (Arg-1 79)and pertussis toxin (Cys-351) are also found inC. elegans Goa, as they are in other Goa proteins.The region of C. elegans Goa that is most dif-ferent from other Goa sequences is in the regionof amino acids 90-140 and 290-320. This di-vergence would be expected because this re-

gion is also the most variable between any otherpair of G protein a subunits.No sequences resembling the SLi or SL2

spliced leader sequences, which are splicedposttranscriptionally onto certain C. elegansmRNAs (Krause and Hirsh, 1987; Huang andHirsh, 1989), were found at the 5' end of theGoa cDNA. However, the Goa cDNA sequencein Figure 2 is probably not full length, becausethe size of the mRNA detected on a Northernblot is -2.2 kilobase pairs (Figure 4). Thus, ex-cluding the poly-Al tail, the sequence may bemissing -400 base pairs. The results of South-ern blot analysis indicate that the C. elegansGoa gene is present in a single copy in the C.elegans genome (data not shown).To identify a candidate protein encoded by

the C. elegans G.a cDNA, we performed an im-munoblot analysis with a heterologous antibody.A protein of 40 kDa can be detected on immu-noblots of total C. elegans extracts with an af-finity-purified antibody to bovine G.a (Figure 5).The intensity of the signal is the same in extractsfrom the wild-type N2 hermaphrodite strain asin extracts from a him-5 strain that produces-30% males. Thus Goa is probably not male-specific. A protein of the same molecular weightcan also be detected with an antipeptide anti-body (OC1; McFadzean et al., 1989) that wasmade against the sequence Ala-Asn-Asn-Leu-Arg-Gly-Cys-Gly-Leu-Phe (data not shown). Thissequence is found as the last 10 amino acidsof all Goa proteins except Xenopus Goa and ro-dent G.a-B.Isolation and mapping of cosmid clonesThe PCR-derived fragments of gpa- 1, gpa-2, andgpa-3 and the Goa cDNA insert were used as

E Nc S T

I I I IT Bc Cl H2

I I I I

Ns Ba E

I I II L- I IIII*- * *-I

N C

4 4- 4- 44- 4 4- 4-

300 bpFigure 1. Restriction map and DNA sequencing strategy of C. elegans Goa cDNA. The C. elegans Goa cDNA in XCe6-2 isrepresented as a horizontal line. The Goa coding region is represented as a thicker horizontal line. The arrows indicate thedirection and extent of DNA sequences determined. Abbreviations for restriction enzyme recognition sites are Ba, BamHl;Bc, Bcl l; Cl, Cla l; E, EcoRI; H2, Hincil; Nc, Nco l; Ns, Nsi l; S, Sph l; T, Taq 1. Other abbreviations used are: N, amino terminus;C, carboxyl terminus.

Vol. 2, February 1991

I

137

M.A. Lochrie et al.

CGAGCTGCACCACATTACAGTGAGTGAGTAGAGGATATCAAGTGGAGACCGCTACCGGGGCATAGGTCCACCGTT 75CATCAACTCTAGGTGCCATGGGTTGTACCATGTCACAGGAAGAGCGTGCCGCTCTTGAAAGATCACGAATGATTG 150

MetGlyCysThrMetSerGlnGluGluArgAlaAlaLeuGluArgSerArgMetIleG 20

AGAAAAATCTTAAAGAAGACGGCATGCAAGCGGCAAAAGATATCAAACTGCTGCTACTTGGTGCAGGAGAATCAG 225luLysAsnLeuLysGluAspGlyMetGlnAlaAlaLysAspIleLysLeuLeuLeuLeuGlyAlaGlyGluSerG 45

GAAAATCGACTATTGTAAAACAGATGAAAAT TCACGAATCAGGATCACAGCAGAAGACTACAAACAGTACA 300lyLysSerThrIleValLysGlnMetLysIleIleHisGluSerGlyPheThrAlaGluAspTyrLysGlnTyrL 70

AGCCGGTTGTCTACAGTAACACGGTTCAATCATTGGTCGCTATTTTGCGCGCCATGAGCAACTTAGGCGTTTCAT 375ysProValValTyrSerAsnThrValGlnSerLeuValAlaIleLeuArgAlaMetSerAsnLeuGlyValSerP 95

TTGGTTCGGCT AAAATTAGTGTGGATGTGGTGGCACGAATGGAGGACACAGAGCCAT 450heGlySerAlaAspArgGluValAspAlaLysLeuValMetAspValValAlaArgMetGluAspThrGluProP 120

TCTCAGAAGAATTGCTCAGTTCAATGAAACGGTTGTGGGGAGACGCAGGTGTACAGGATTGTTTCTCAAGGAGTA 525heSerGluGluLeuLeuSerSerMetLysArgLeuTrpGlyAspAlaGlyValGlnAspCysPheSerArgSerA 145

ACGAGTATC ATTGAATGATTCAGCCAAATATTTCCTTGACGACCTGGAAAGGTTAGGAGAGGCAATATATCAAC 600snGluTyrGlnLeuAsnAspSerAlaLysTyrPheLeuAspAspLeuGluArgLeuGlyGluAlaIleTyrGlnP 170

CAACTGAGCAACATATTCTCCGAACTCGTGTCAAAACAACTGGTATTGTTGAAGTTCACTTCACATTCAAAAATC 675roThrGluGlnHisIleLeuArgThrArgValLysThrThrGlyIleValGluValHisPheThrPheLysAsnL 195

TCAATTTCAAATTGTTCGATGTGGGAGGTCAAAGATCAGAAAGGAAGAAGTGGATTCATTGTTTCGAAGATGTTA 750euAsnPheLysLeuPheAspValGlyGlyGlnArgSerGluArgLysLysTrpIleHisCysPheGluAspvalT 220

CTGCTATTATTTTCTGTGTTGCCATGTCAGAGTATGATCAACTGTTGCACGAAGATGAGACAACAAACCGAATGC 825hrAlaIleIlePheCysValAlaMetSerGluTyrAspGlnLeuLeuHisGluAspGluThrThrAsnArgMetH 245

ACGAATCGCTGAAGCTGTTCGATTCGATCTGTAATAACAAATGGTTCACAGATACATCGATTATTCTGTTCCTGA 900isGluSerLeuLysLeuPheAspSerIleCysAsnAsnLysTrpPheThrAspThrSerIleIleLeuPheLeuA 270

ACAAGAAGGATCTGTTTGAAGAGAAAATCAAGAAAGCCCGTTAACGATCTGCTTCCCAGAATATTCAGGACGAC 975snLysLysAspLeuPheGluGluLysIleLysLysSerProLeuThrIleCysPheProGluTyrSerGlyArgG 295

AAGACTACCACGAGGCATCTGCGTATATTCAAGCACAATTTGAGGCTAAAAACAAATCAGCGAATAAAGAAATCT 1050lnAspTyrHisGluAlaSerAlaTyrIleGlnAlaGlnPheGluAlaLysAsnLysSerAlaAsnLysGluIleT 320

ATTGCCACATGACATGTGCCACAGACACAACTAACATTCAATTTGTGTTTGACGCTGTCACCGATGTGATTATTG 1125yrCysHisMetThrCysAlaThrAspThrThrAsnIleGlnPheValPheAspAlaValThrAspValIleIleA 345

CCAATAATCTTCGTGGATGCGGCTTGTATTAAGCTGTCGTTTCGCCGCCCTCTTCTACCATTCTGTGTGTATCT 1200laAsnAsnLeuArgGlyCysGlyLeuTyrEnd 354

TTGCTTACTTTTCCCAAATTTTTAAAGATTCGTTATTTTTCTCATCCCGCGAGTATGCATCTATAATTTGAGAGC 1275TTTACATGTACATTATGGTTGAACTGTTTTTATTTTTTGGAAAAAGTTGCAATTTCAGATGTATATGGGCTTTTT 1350TTTAACCTCTGATTCATTCATAATAGTCTCACCGCCCTTCTTTTTCAACTAGGATCCCTAATTTTTCTGTCAC 1425AAAAATAACACATAAAACTCACAATTATTTATTTTTTCCTTTTTTCTTATTTATTATATATTGTCATTATTCACG 1500CCACATCCCCCGCCCTTCTCACTCCCGTGTAGCCTTATCGCTATTTGAAAACAAGAAAACTTTAAAAATCTAAAT 1575TCAGTTG 1582

Figure 2. Sequence of the C. elegans G0a cDNA. Amino acid abbreviations used are Ala, alanine; Arg, arginine; Asn, asparagine;Asp, aspartic acid; Cys, cysteine; Gin, glutamine; Glu, glutamic acid; Gly, glycine; His, histidine; lie, isoleucine; Leu, leucine;Lys, lysine; Met, methionine; Phe, phenylalanine; Pro, proline; Ser, serine; Thr, threonine; Trp, tryptophan; Tyr, tyrosine; Val,valine.

probes to isolate the corresponding cosmid three, one, three, and one, respectively. Theclones. The number of cosmids that hybridized cosmids that contain the gpa- 1, gpa-2, gpa-3,to the gpa-1, gpa-2, gpa-3, and Goa probes was and Goa (goa- 1) genes mapped to unique po-

CELL REGULATION138

Ga in C. elegans

28S rRNA

-- 18S rRNA

Figure 3. Hybridization of C. elegans total mRNA to C.elegans Ga cDNA. Total mRNA (12.5 qg) from mixed pop-ulations of the N2 wild-type hermaphrodite strain was frac-tionated on a 1% formaldehyde gel and hybridized to theC. elegans G.a cDNA. The positions of the 18S (1750 bp)and 28S (3500 bp) rRNAs are indicated.

sitions on the C. elegans contig map. The gpa-1, gpa-2, and gpa-3 genes map to chromosomeV. The gpa- 1 gene maps between the actin (act-1, act-2, and act-3) gene cluster and myo-3 andis -100 kilobases from the actin gene cluster.The gpa-2 gene maps between mec-1 and HIS-2, in agreement with Fino Silva and Plasterk(1990). The gpa-3 gene maps between hsp-16and her- 1. The goa- 1 gene maps to chromosomeI between unc-13 and lin-10 and is -50-100kilobase pairs from unc-13. Each gene mapsnear a number of mutations. Although the po-sitions of many of these mutations on the ge-netic map are ambiguous, some of these mu-tations result in phenotypes that suggest theycould be in G protein alpha subunit genes.

Gene sequences of gpa-1 and gpa-3The restriction map, intron/exon organization,and DNA sequencing strategy of the gpa-1 andgpa-3 genes are shown in Figures 6 and 8. TheDNA sequences of the gpa-1 and gpa-3 genesare shown in Figures 7 and 9. The gpa- 1 codingregion is divided by seven introns and the gpa-3 coding region is divided by five introns. Intronboundaries were established by several criteria:homology to the amino acid sequences ofknown G protein a subunit proteins, the spacingbetween amino acids found in all G protein asubunits, the presence of C. elegans consensussplice junction sequences (Emmons, 1988), theabsence of stop codons in open reading frames,the absence of any sequence homology to Gprotein a subunits in putative introns, and con-servation of intron position relative to mam-malian G protein a subunit genes. In almost ev-ery case the 5' GTAAGTT and 3' 1TTCAGconsensus C. elegans mRNA splice sites (Em-mons, 1988) or closely related variants arefound at each intron/exon junction. The se-quence GTAAG is found at all of the putative 5'splice junctions in the gpa-1 gene and three outof five 5' junctions in the gpa-3 gene. The se-quence TTXCAG is found at most of the 3' junc-tions. An A at position -16 to -17 from the 3'splice junction border that is conserved-andis believed to be the site where lariat formationoccurs during mRNA splicing-is also found inmost of the putative introns in the gpa-1 andgpa-3 genes, with the exception of some of thesmaller ones. Most of the introns are also AT-rich, as observed for other C. elegans introns.All of the introns in the gpa-1 and gpa-3 genesare >650/o AT except for introns 1, 5, and 7 inthe gpa- 1 gene. More introns may exist 5' or 3'to the coding regions, especially in the 5' regionof the gpa-3 gene, where several 3' consensussplice junctions are found (see below).The sizes and positions of the introns in the

gpa- 1, gpa-2, and gpa-3 genes are shown in Fig-ure 10. The positions of four introns in the gpa-1, gpa-2, and gpa-3 genes are conserved relativeto the mammalian Gia, Goa, and Tra genes (Itohet al., 1988; Kaziro et al., 1988; Weinstein et al.,1988; Raport et al., 1989). Those introns in thegpa- 1, gpa-2, and gpa-3 genes that occur at thesame position as an intron in a mammalian genenot only occur at the same position within theamino acid sequence but also at the same po-sition within the codon. In addition, these in-trons occur near amino acid sequences that arewell conserved (i.e., those involved in guanine

Vol. 2, February 1991 139

M.A. Lochrie et al.

C: MG E ED:MGC ER I

R: ERC EH

X :

C: KQMKIIHD: IKQMKIIH GjTR: IKQMKIIHE

X:KQKI

R: v T

C: DS ~A4.D : IDSAK~4

R: DSAK44L4 c rX: DSAK] L c r

C :

D:R:X.

DVGGQRSERKDVGGQRSERKDVGGQRSERKDVGGQRSERK

C: FDSICNND: IFDSICNNFR: IFDSICNNBTFX: FDSICNN

KWIHCFEKWIHCFEKWIHCFEKWIHCFE

C :

D:R:X.

ILLLG AGESGKSTIV 50.LLLG AGESGKSTIV.LLLG AGESGKSTIV.LLLG AGESGKSTIV

N GVSFGS 100wT |SIQYSNNtDT VEYGDKDTUIEYGDK

QL 50C FaRmYQLN|;C FRtEYQLN|

Fw KN L 200TEtF|EKN LF

tEHF| TKN4w

laAIIFCVP YDILHEDET TN RS 250VT AIIFCV* |YD V|HEDET TN ISLKDVT AIIFCV YDHEDET TN SVT AIIFCV7YLIHDTTN

KKD K PLTICFPE QDYH 300KKDIF1EKIR KqPLTICFPE 4QEYGKKD K KBPLTICFPE 41SNTYE

S PLTICFPE NSFT

cRM rBIT DAVT II LR 350cqMICATDN NI DAVT II LRqcpjCATDN NI DAVT II LRICATDQ I DAVT

C: CGLF 354D: ICGLFIR: ICGLFIX: CGL[

Figure 4. Comparison of G4a amino acid sequences from four species. G.a protein sequences from C, C. elegans (thisstudy), D, Drosophila melanogaster (deSousa et al., 1989), R, rat (Jones and Reed, 1987), and X, Xenopus laevis (Olate et al.,1989) were aligned. Amino acids that are found in all four sequences at the same position are boxed. The mouse (Strathmannet al., 1990), human (Lavu et al., 1988), hamster (Hsu et al., 1990), and bovine (Van Meurs et al., 1987) G.a amino acidsequences are not shown because they are >900/o identical to the rat G,a sequence. Similarly, the locust G.a sequence (Hsuet al., 1990) is 930/o identical to the Drosophila sequence. Amino acid abbreviations used are A, alanine; C, cysteine; D,aspartic acid; E, glutamic acid; F, phenylalanine; G, glycine; H, histidine; I, isoleucine; K, lysine; L, leucine; M, methionine; N,asparagine; P, proline; Q, glutamine; R, arginine; S, serine; T, threonine; V, valine; W, tryptophan; Y, tyrosine.

nucleotide binding). The introns in the gpa-1,gpa-2, and gpa-3 genes that are extra or missingrelative to the mammalian Gia, G.a, andTra genes occur within the region that is

most variable between G protein alpha sub-units.A remarkable feature of the intron/exon or-

ganization of both genes is that the intron at

CELL REGULATION

YQP TEQDILRTRV KTTGIIVYQP TEQDILRTRV KTTGIVYQP TEQDILRTRV KTTGIVYQP TEQDILRTRV KTTGIV

j TSIILFLNDTSIILFLNC TSIILFLNIDTSIILFLN

140

G,. in C. elegans

kDa3

40

1 2 3Figure 5. Immunoblot analysis of C. elegans Goa protein.An affinity-purified rabbit anti-bovine Goa antisera was usedat a dilution of 1 :1000 to detect GOa proteins as describedin Methods. Lane 1, 1 gg purified bovine G,a; Lane 2, 50pg total protein from C. elegans N2; Lane 3, 50 Ag totalprotein from C. elegans him-5 (e 1490).

the first position within the coding region is un-usually large. The first intron of the gpa-1 geneis 718 base pairs and the first intron of the gpa-3 gene is 745 base pairs. The first introns of thegpb- 1 gene (1447 base pairs) and the gpa-2 gene(1020 base pairs) are also large (Fino Silva andPlasterk, 1990; Van der Voorn et a., 1990). Thisis unusual because in C. elegans most of theintrons are -50 base pairs in length (Emmons,1988). In fact, the average length of the otherintrons in the gpa-1, gpa-2, gpa-3, and gpb-1genes is -130 base pairs. The only extendedregion of sequence homology that has been de-tected between any pair of these four large in-trons is between the first introns of gpa-2 andgpa-3. In this case a stretch of 85 base pairswith 63% nucleotide sequence identity wasfound. No open reading frames resembling Gprotein a subunit sequences have been de-tected within them, indicating that they do notcontain alternatively spliced exons. At the otherextreme, the third intron in the gpa-1 gene,which is 43 base pairs long, is among the small-est introns known.

Another interesting feature of the gpa- 1 geneis the presence of an almost perfect invertedrepeat (see Figure 6C). One repeat element islocated at positions 583-918 within the first in-tron and the other is at positions 2012-2350within the fifth intron. The elements of the re-peat are -950/o identical. The repeats are -330base pairs long and are - 1 100 base pairs apart.This corresponds well to the average dimen-sions of repeats in C. elegans DNA reported byEmmons et al. (1980). Palindromic sequencesare found at the boundaries of the repeats.The nucleotide sequences 5' to the initiation

codons of the gpa-1 and gpa-3 genes were ex-amined for the presence of consensus CAAT-box, TATA-box, and transcription initiation sitesequences (Breathnach and Chambon, 1981).The gpa- 1 gene has sequences at positions 50-55 (GCTAAT), 86-92 (TATATA), and 113 thatconform well to the spacing constraints andconsensus sequences for the CAAT-box, TATA-box, and transcription initiation site. In addition,a consensus 3' splice junction acceptor site(TTTCAG) is found immediately preceeding theinitiation codon. This site may be where a trans-splicing reaction occurs. A very similar patternof consensus CAAT-box, TATA-box, transcrip-tion initiation site, and 3' splice junction se-quence is found in the 5' region of the gpa-2gene (Fino Silva and Plasterk, 1990).The 760 base pairs of sequence that is avail-

able 5' to the gpa-3 coding region has only oneTATA sequence located at position 721-724,which is 35 base pairs from the translation ini-tiation codon. No sequences resembling theCAAT-box sequence are readily found. How-ever, the region from positions 6 to 52 has 680/onucleotide sequence identity to a region in thegpa-2 gene that is between its putative CAAT-box and TATA-box. Based on this homology,positions 3-7 (GCATT) and 38-43 (TTTATG)may be the CAAT-box and TATA-box sequencesof the gpa-3 gene, respectively. Several potential3' splice junctions are found at positions 83-90 (T1TrCCAG), 253-257 (TTTAG), 320-326(I I I I CAG), 694-700 (I I I I IAG), and 730-736(TTACAG). However, no sequences resemblingthe consensus 5' splice junction sequence arefound within this region.The sequences 5' to the initiation codon were

compared with each other to detect sequencehomologies that might, for example, be indic-ative of similarities in gene regulation. Otherthan the homology between gpa-2 and gpa-3mentioned above, pairwise sequence compar-isons of available sequences from the 5' regions


M.A. Lochrie et al.

H3 X P XhXIP PNs Bg K Sa Xh K11 1 1 11 I 11 II l I

Sa Bg H3

I II

lOOb1000 bp

Xh X Ns P Nc P Ns Bg

I IIIl Ii III_ --* * *-*+- -* --

-~~~- ~~~. -.4- -p - 4- 4-

300 bp

NI .......... III, "'iI.

- U - - - -

300 bp

Figure 6. Restriction map,DNA sequencing strategy,and gene organization of C.elegans gpa-1 gene. (A) Re-striction map of 10 kbp Hindlilfragment that contains the C.elegans gpa-1 gene. The ar-rows indicate the directionand extent of DNA sequencesdetermined. (B) Restrictionmap of 3.1 kbp Xho I/Bgl 11subclone from 1 0 kbp HindlIlfragment. The arrows indicatethe direction and extent ofDNA sequences determined.Abbreviations for restrictionenzyme recognition sites areBg, Bgl Il; H3, Hindlil; Nc, Ncol; Ns, Nsi l; K, Kpn l; P, Pst l;Sa, Sal l; X, Xba l; Xh, Xho I.(C) Intron/exon organization ofC. elegans gpa- 1 gene. Exons

C are represented as horizontalI lines. Introns are represented

as boxes. Striped arrows in-dicate the location of the in-verted repeats. Abbreviationsused are N, amino terminus;C, carboxyl terminus.

of the gpa-1, gpa-2, gpa-3, and gpb-1 genes didnot reveal any striking homologies. Each 5' re-gion has several repeated sequences, but noneof these are common to any pair of sequences.The open reading frames on both strands thatare found 5' and 3' to the coding regions of thegpa- 1 and gpa-3 genes were compared with theGenBank database, but no significant homolo-gies were detected.The available sequences 3' to the termination

codons of the gpa-1 and gpa-3 genes were ex-amined for an AATAAA consensus polyadenyl-ation signal sequence (Proudfoot and Brownlee,1979; Wickens and Stephenson, 1984). This se-quence was found in the 3' region of the gpa-1gene at positions 3340-3345. It was not foundin the gpa-3 3' region, although minor variantsthat might be functional are found at posi-tions 2825-2830 (AATACA) and 2891-2896(AATATA).The predicted amino acid sequences of the

gpa-1, gpa-2, and gpa-3 proteins are comparedin Figure 11. The percent sequence identity ob-served between them is gpa-1 versus gpa-2:410/0, gpa-1 versus gpa-3: 48%, and gpa-2 ver-sus gpa-3: 580/%. The gpa-1, gpa-2, and gpa-3

proteins are 490/o, 460/o, and 530/o identical tothe C. elegans G.a sequence, respectively. Allthree gpa proteins have the six regions impli-cated in binding guanine nucleotides (Lochrieand Simon, 1988). Also, all three have the con-sensus myristoylation signal Met-Gly-X-X-X-Serat the amino terminus (Towler et al., 1988).Moreover, all three have the arginine (Arg-179in gpa- 1, Arg-181 in gpa-2, and Arg-179 in gpa-3) that is the substrate amino acid for choleratoxin in other G protein a subunits. The gpa-1and gpa-3 proteins have the cysteine (Cys-354in gpa-1 and Cys-351 in gpa-3) four amino acidsfrom the carboxyl terminus that is the substrateamino acid for pertussis toxin in other G proteina subunits. However, the gpa-2 protein has aserine at this position (Fino Silva and Plasterk,1990). The presence of cholera and pertussistoxin substrates in C. elegans has been reported(Van der Voorn et al., 1990) but the effects ofthese toxins on C. elegans physiology has notbeen well studied.

DiscussionWe have determined the sequence of a cDNAencoding a C. elegans G.a protein that is very

CELL REGULATION

Ns Nc xA)

B)

C)

142

G,, in C. elegans

similar to that of Goa proteins found in otherspecies. Therefore, it is likely that C. elegansG.a is similar in function to other Goa proteins.In vertebrates and Drosophila, Goa is found pre-dominantly in neurons. However, its role in sig-nal transduction pathways is not well delin-eated. Thus, the presence of a Goa homologuein C. elegans may provide new approaches forunderstanding its function in other organisms.The results of Southern blot analysis indicate

the C. elegans goa-1 gene is single copy, andonly one size class of mRNA is observed on aNorthern blot using the C. elegans Goa cDNAas a probe. It is, therefore, likely that the proteindetected on the immunoblot is the same as thatencoded by the Goa cDNA described here andnot some other closely related G protein a sub-unit. However, the possibility that there may bemultiple Goa proteins in C. elegans cannot beexcluded until the genomic sequence of goa-1has been determined. Multiple species of Goa,which arise as a result of alternative mRNAsplicing patterns, have been found in Drosophila(deSousa et aL, 1989; Thambi et al., 1989; Yoonet al., 1989) and rodents (Hsu et al., 1990;Strathmann et al., 1990).

In contrast to C. elegans G.a, the gpa- 1, gpa-2, and gpa-3 proteins cannot be classified ac-cording to their relatedness to other G proteina subunits. They are 40-500/o identical to anyother non-C. elegans G protein a subunit se-quence. G protein a subunits that are >800/oidentical are generally considered to be in thesame class (Lochrie and Simon, 1988) and canactivate the same effector proteins (Gillespieand Beavo, 1988; Yatani et al., 1988). G proteina subunits that are as little as 60-700/o identical(e.g., Goa, Gia, and Tra) can interact with someof the same receptors, although heterologousinteractions are somewhat less efficient thanhomologous ones (Kahano et aL, 1984; Cerioneet al., 1986). However, G proteins in this classhave not been found to regulate the same ef-fector proteins. G protein a subunits that are40-600/o identical have not been observed tointeract with the same receptor or effector andare generally considered to be distinct in func-tion. Therefore, gpa- 1, gpa-2, and gpa-3 are likelyto have separate functions. This divergencedoes not exclude the possibility that they mayoperate in signaling systems similar to those inmammals, because the receptors and effectorsthat gpa-1, gpa-2, and gpa-3 interact with mayalso be different in sequence from their mam-malian counterparts.

Although the sequences of the gpa-1, gpa-2,and gpa-3 proteins provide no clues about theirfunction, several approaches are available in C.elegans for investigating their function. One ap-proach initiated in this study is the mapping ofgenes near known mutations with visible phe-notypes, followed by a determination of whichof these mutations, if any, correspond to mu-tations in G protein alpha subunit genes. Thisdetermination can be done by transformationof mutant strains with the cosmid clones iso-lated in the work described here. However, it ispossible that none of the existing mutations willbe in a G protein gene, because the genetic mapis not saturated with mutations. If so, there area number of strategies to identify mutant phe-notypes for these genes. For example, recessivemutations might be isolated by a PCR-basedscreen for transposon insertion mutations, us-ing methods similar to those developed in Dro-sophila (Ballinger and Benzer, 1989; Kaiser andGoodwin, 1990). Dominant mutations can begenerated by transformation with mutants ofthe gpa genes constructed in vitro. Based onstudies of mutant G protein a subunits found inhuman tumor cells (Landis et al., 1989) and thebiochemical properties of G protein a subunitssynthesized in E. cofi (Freissmuth and Gilman,1989; Graziano and Gilman, 1989) and mam-malian cells (Masters et al., 1989), certain aminoacid changes in the C. elegans G protein alphasubunit sequences might be expected to resultin a dominant phenotype. A determination ofgpa gene expression patterns could also provideinformation about the function of the gpa pro-teins. However, our initial attempts to detectgpa gene expression by Northern blotting wereunsuccessful. In addition, we were unable toisolate cDNA clones for gpa- 1, gpa-2, or gpa-3.gpa-2 and gpa-3 are the only two known G

protein a subunits that have a cysteine insteadof a serine in the highly conserved sequenceGAGESGK found at the amino terminus (Re-sults; Fino Silva and Plasterk, 1990). The se-quence of G,a also differs in this region fromother G protein a subunits. It has the sequenceGTSNSGK. These differences found in Gza arethought to be responsible for its very lowGTPase activity (Casey et al., 1990). This regionis close to the phosphates of GDP in the crystalstructure of the analogous ras proteins (Milburnet al., 1990) and mutations in this region of Gsareduce its GTPase activity (Graziano and Gil-man, 1989). Thus, gpa-2 and gpa-3 may have aGTPase activity that differs from gpa-1. Such


M.A. Lochrie et al.

CTCGAGTTGTTTATGCGGCGAAAAAAAATCTGTAGAGGTATACGAAAAAGCTAATTAGGT 60GGITTGCCGTCTAGTAATACTCGATTATATATTTCTTGTCGTCATCGTTACCACTGCTCT 120CCATTTTCTCTCCTCACTCACAAAAACTTGAATATTCTTCGCTTCCTGTTGTGATCAGTG 180AAAATTCGTCTGATTTCTTCAATATTATTCAATGTCATTTATTCGAAGACTTCAGATGGG 240

MetGl

ACTGCGAATCGA ACTTGTAGCTCALGCCACA G&TAAAATALTCA CACGGh 300yAsnCy3GluSerArgGluLeuValAlaGlnAlaLysGlnAsnLysIleIleAsnThrGl

ACTT TGCTGTTAGGTAAGTT 360uLeuAspLy3AlaLysLysThrA3pGluAsnIleIleLysLeuLeuLeuLeuG

AGATTCTGAACCAAAAACAGCTTTCTGCAATACTTGTTGGTACATTCAAATCGCGCTCAT 420TTTTCATTGAAAATTAAACGCGGCGAACATTTTTCTTACAAATAGGTGGTGACAACCTCA 480TTTCATTCTGGTATCGGCACGAAAACATGTTTCTAGAAAGGTCTAGTTACCCAACCAAAA 540AGTTGAAAGCACCACAGCTGTAGCGAAACCCAAAATAGAAGGAATTAAAATCTACCCACT 600TGAGTTCACTAACACCCGCAAATGCATTAAAGAAACGTTTAAAAAACATCAGGTTGAAAT 660CTTGTTCAATTATATCCGCTCAGAAGAGCAAGCCACTACGGCTCCAAAATTCATCGGCTC 720CGTGCGGTCATCAAGTTCCGAAGAACTGTTTCACGCTCGCGGCTCGCCCGTCTCCCACTC 780ATCGCGGATGGCATCTCTACCAATTTGATGGGAATTACGAGATACATACTGCAGAAATGA 840TTCTGTATAGTATGGTCTCGATGTAAGCAGAGTGCTGACGGGGTTAGTGTGTACACGACA 900GCCGACACCTCGCGGGTTTAGCCTGCCACCCTGTTTTGTGTGTCTGCCTGACCACCTGCC 960GCCGATTGCATTGATAGCATTGATGCTCCATGGAAATGCGATTACTAATCAAACATCATT 1020CGAATGTTTGACTTTTCTATCGTTTAAACATCTGAAATTAACTATTTCCAGGAGCTGGAG 1080

lyAlaGlyG

AGAGTGGAAAAAGTACCGTGCTGAAACAAATGAAGTAAGTTCTTAAAGATGAGTTCGAAA 1140luSerGlyLysSerThrValLeuLysGlnMetLy

AAGAGTGTTTGCTATAAAATGCTCGGAATAATGCAATATGGGGCAAATGACGGATTAATC 1200TTTCAGCTTTTATTTTAAGAGGAAGTTCAAAAATTTACAATTATTATTAGATTTTTTTTC 1260AAAATTTCTTTTTTGTGGCATTTCATCATATTGGAATTTTCCAAGGTCTTTAAAGCAAAA 1320GCTAAGTTAACCGAACTCTACTTTCAATCAAATGAAGAAATGTACCAATAAAATGATTTA 1380CAGAATCATCCATAACAQTGGATTCTCTCAAGAAGAAATATCAAATAAACGAAATGTTGT 1440

sIleIleHisAsnSerGlyPheSerGlnGluGluIleSerAsnLysArgAsnValVa

CTGTGCGAATACTGTGCAAGCAATGGGAGCATTGTTAGATGGAATGAAACAkCTTCAATT 1500lCysAlaAsnThrValGlnAlaMetGlyAlaLeuLeuAspGlyMetLysGlnLeuGlnPh

CGATTTTTCAACCCGAGTTTGTAATGTAAGAAATGTTAATTTTTGTGTTACATAGTACTT 1560eAspPheSerThrArgValCysAsn

ATTTATAGGCACATGAGAAGTTGATACGTGAAACATTGA&TGATAAAGCTGAGTATGGAC 1620AlaHisGluLysLeuIleArgGluThrLeuAsnAspLysAlaGluTyrGlyP

CATTCAGTGATGCAATGTTCAAGTAAGAACCTTATGTCATTTTGTGAATTTATTCAAAAA 1680roPheSerAspAlaMetPheAs

ATATTTTGATAGCAATTCTCCGATTTGACAAAACTTTCATAACAATGGAAGTTCACAATA 1740AAATTTTAATAAATCTAAACTATATTCAGCGCACTTACTG&GTTGTGGGCGGAC AAAGG 1800

nAlaLeuThrGluLeuTrpAlaAspLysGly

GTTCAGTGTGCATACGATAAGCGAGAGTTTTTTTACCTTCATGATTCTGCAAAATAGTAA 1860ValGlnCysAlaTyrAspLysArgGluPhePheTyrLeuHisAspSerAlaLysTy

Figure 7. DNA sequence of C. elegans gp-1 gene. The predicted amino acid sequence of the gpa- I protein is shown belowthe DNA sequence. The standard three-letter amino acid abbreviations are given in the legend to Figure 2. Coding sequencesare shown in boldface lettering. The primers used in PCR reactions are underiined. oMP19 corresponds to nucleotide positions2564-2581, oMP20/oMP21 to positions 2791-2808, and Ta29 to positions 3140-3157.

CELL REGULATION144

G<, in C. elegans

GTTAAACTTTTTTTTAAATATCAGAACTTTGCGATATTTTAAGTATTTAAAAACATAGAG 1920CGTTAGAGGCTCAATATTAGTTCAACAAAACAGGTTATTCCATTTCCTAATTGCAATCAA 1980GAGGCGTGTCTGTGCCTACACTTAGGATGCTGAATGACTGAACCCGCGAGGTGTCGGCTG 2040TCGTGTACACACTAGCCCCGCCAGCACTCTGCGTACATCGAGACCATACTATACAGAATC 2100ATTTCTGCAGTATGTATCTCGTAATTCCCATCAAATTGGTAGAGATGCCATCCGCGATGA 2160GTGGGAGACGGGCGAGCCGCGAGCGTGAAACAGTTCTTCGGAACTTGATGACCCGCACGG 2220AGCCGATGAATTTTGGAGCCGTAGTGGCTTGCTCTTCTGAGATCGGATATAATCGAACAA 2280GGTTTCAATCTGATGTTTTTTTAAACGTTTTTTTAATGCATTTGCGGGTGTCTAGTGCGA 2340TTTTATAATTGAATTACATTAAAAATTGCGAGTAGGTAAATAAGATTTAATTTTCAGTTT 2400

rPh

TCTTGATCGAATTGCAAGGGTTCACACTcCCAATTACGTACCAACTGAAAATGATATTCT 2460eLeuAspArgIleAlaArgValHisThrProAsnTyrValProThrGluAsnAspIleLe

GCATACACGGGTTCCA CAATGGGTGTTATAG&&GTTAATTTCACAATAAAGGGCAAATT 2520uHisThrArgValProThrMetGlyValIleGluValAsnPheThrIleLysGlyLysPh

TTTTCGAGTATTTGATGTGGGCGGACAGCGAT CZIILTGGATTCACTGTTT 2580ePheArgValPheAspValGlyGlyGlnArgSerGlnArgLysLysTrp leHisCysPh

TGATGATGCAAAAGCTATGATATATGTTGCTTCGTTGTCAGAGTATGATCAAGTTTTATT 2640sAspAspAlaLysAlaMetIleTyrValAlaSerLeuSerGluTyrAspGlnValLeuLe

*GAGAoCAATACTACTGTAAGTTTTGATATTTTCAAAAACAAATAATTCAATAGTAATTT 2700uGluAspAsnThrThr

CTTGTTTTTCAGAATCGAATGCACGAATCAATACAGTTATTTAAGCAAGTAATCAACAAT 2760AsnArgMetHisGluSerIleGlnLeuPheLysGlnValIleAsnAsn

AAATATTTTGTAARCACTTCAGTTATATTATTCCTGAACAARATTGATTTATTCGAGGAA 2820LysTyrPheValAsnThrSerValIleLeuPheLeuAsnLysTleAspLeuPheGluGlu

AAAATTGTTACCAAAAAACGAAGTCTTGGTATTGCATTTGAGTCATTTAGTGGTAAGTGC 2880LysIleValThrLysLysArgSerLeuGlyIleAlaPheGluSerPheSerG

AATTTTTGAAGGCGAATGTGGAGGCACTTGAACCCCGTACACCTGTCCGCCGTTCTGGCC 2940GAGTGGTCTAAGGCGCTGCGTTCAGGTCGCAGTCCTCTCCGGAGGGCGCAGGTTCGAATC 3000CTGCGGACGGCATTTCTTTTTGTGCATTCTCTTTTTTTTTGCAGGACCGAGCCAAGATCT 3060

lyProSerGlnAspLe

CAATGCAGCTGTAGCGTTCGTTGAAAAAAGTAT 3120uAsnAlaAlaValAlaPheValGluLysLysTyrArgSerMetAlaGluAsnLysGluLy

GAACATTTATTGTCATCACACTTGTGCTACAG&C&CACACAGGTGCAGTACGTACTCGA 3180sAsnIleTyrCysHisHisThrCysAlaThrAspThrGlnGlnValGlnTyrValLeuAs

TGCGGTTCTAGATACGATACTATCTACCAAACTGAAGGGATGTGGATTGTATTGAGAGAT 3240pAlaValLeuAspThrIleLeuSerThrLysLeuLysGlyCysGlyLeuTyrEnd

TTTAAATTATTGTTACCTTTGGTACAAACCTTGTAAACCTGAGT6ATAATTATCGGGTAC 3300CGTTCGGTATTATCTTGATTGGTGTAATATATAGAATTTAATAAATAATGAATCAAATTT 3360

Figure 7. (Continued)

differences could have a profound effect on their homologue of the mammalian Gsa gene (gsa-1)biological properties. has been isolated (S. Ohshima, S. Tashiro, J.-H.There may be G protein genes in C. elegans Park, T. Tani, and Y. Ohshima, personal com-

other than those reported to date. A C. elegans munication). This gene could not have been iso-


M.A. Lochrie et al.

A)H3 P X X H2 Ns

I I1I

400 bp

B)

L41

C)

Ns P H3

-l. 11.-_ -* -. 4

44-4.4-. 4 .4

00 bp

N

I -

H3 P X X H2 Ns

-. -

-__-* -

.4- 4- 4-4*-.- .4- .4-4*-

C

400 bp

H3

Figure 8. Restriction map,DNA sequencing strategy,and gene organization of C.elegans gpa-3 gene. (A) Re-striction map of 3 kbp Hindlilfragment from C. elegans gpa-3 gene. The arrows indicatethe direction and extent ofDNA sequences determined.(B) Restriction map of 2.7 kbpNsi I fragment from C. elegansgpa-3 gene. The arrows indi-cate the direction and extentof DNA sequences deter-mined. Abbreviations for re-striction enzyme recognitionsites are Ns, Nsi l; H2, Hincil;H3, Hindlil; P, Pst l; X, Xba 1.(C) Gene organization of C.elegans gpa-3 gene. Exons arerepresented as horizontallines. Introns are representedas boxes. Abbreviations usedare N, amino terminus; C,carboxyl terminus.

lated by the methods used in this paper becauseG,Ca is too divergent from other G proteins. Fur-thermore, the PCR analysis reported here sug-gests that other G protein alpha subunit genesmay exist, because not all of the PCR productscan be accounted for by the clones that areavailable. It is also possible that a given sizeclass of PCR product may contain more thanone G protein a subunit sequence, as observedby Strathmann et al., (1989). C. elegans mayhave other G proteins that have so far beenfound only in mammals, such as Gza and Gal1 -16 (M. Strathmann, T. Wilke, and T. Amatruda,personal communication), or G proteins thathave been found in both Drosophila and mam-mals, such as Gia (Provost et al., 1988) and Gqa(Strathmann and Simon, 1990). Conversely,other organisms may have homologues of theC. elegans gpa-1, gpa-2, and gpa-3 genes. Fi-nally, because G proteins are always found asheterotrimers, it is likely that C. elegans alsohas G protein y subunit genes.

In conclusion, C. elegans has at least five Gproteins. Two of these (G. and Gj) are identifi-able homologues of mammalian G proteins,whereas three (gpa- 1, gpa-2, and gpa-3) appearto be unique to C. elegans. All multicellular or-ganisms that have been examined have Go andG,. However, unicellular organisms, such asyeast and Dictyostelium, do not have G. and Gr.

Instead they have G proteins with novel se-quences. Therefore, it may be that G proteinssuch as G. and G6 arose during the developmentof multicellular organisms and have been highlyconserved across phylogeny because they havefundamental roles in intercellular signaling. Onthe other hand, the G proteins such as gpa-1,gpa-2, and gpa-3 that are not well conservedmay have functions specific to a particular or-ganism.

Methods

NematodesThe wild-type hermaphrodite N2 (Bristol) strain is fromBrenner (1974) and him-5 (e1490) is from Hodgkin et al.(1979).

Polymerase chain reactions (PCRs)Twenty-five nanograms of C. elegans genomic DNA fromthe hermaphrodite wild-type N2 strain were amplified in a10-il reaction containing 50 mM KCI, 1.5 mM MgCI2, 10mM tris(hydroxymethyl)aminomethane (Tris)-HCI, pH 8.3,0.01% gelatin, 200 Mm deoxyribonucleotides (dATP, dCTP,dGTP, and dTTP), 2.5 U Taq polymerase (Perkin Elmer-Ce-tus, Norwalk, CT), and 0.3 Mg of each primer. The primerpairs that were used are oMP19, oMP20/oMP21 andoMP19, Ta29. The primers oMP19, oMP20, and oMP21have been described (Strathmann et al., 1989). The primeroMP19 corresponds to the sense strand that encodes theamino acid sequence Lys-Trp-lle-His-Cys-Phe/Leu. Theprimers oMP20 and oMP21 correspond to the antisense

CELL REGULATION

._

146

Ga, in C. elegans

strand that encodes the amino acid sequence Phe-Leu-Asn-Lys-Lys-Asp. The sequence of Ta29 is GAATTC(GATC)-GT(AG)TC(GATC)GT(GATC)GC(AG)CA(GATC)GT. This se-quence contains an EcoRI restriction site at one end to fa-cilitate subcloning and corresponds to the antisense strandthat encodes the amino acid sequence Thr-Cys-Ala-Thr-Asp-Thr. Thirty amplification cycles consisting of 1 min at 920C,30 s at 370C, and 1 min at 72°C were performed. At theend of the amplification cycles, the reactions were incubatedanother 10 min at 720C to complete partially extendedchains. The products of the reaction were analyzed on 3%NuSieve (FMC, Rockland, ME) agarose gels. The PCR prod-ucts were excised from the gel, reamplified in a 1 00-al re-action using the same primers, made blunt with Klenow,phosphorylated with T4 polynucleotide kinase, ligated tolinkers, and subcloned (Maniatis et al., 1982) into pBluescriptKS/- (Stratagene, Burlingame, CA). The DNA sequences ofthe subclones were determined as described below.

Isolation of G.a cDNAA C. elegans cDNA library containing -250 000 clones wasconstructed in Xgtl0 and kindly supplied by Stuart Kim(Stanford Univ.). The mRNA used to construct this libraryrepresented all developmental stages of the hermaphroditewild-type N2 strain. The library was plated on 15-cm LBagar plates and replicated to nitrocellulose filters (Maniatiset al., 1982). The filters were prehybridized in 6x SSC (1 xSSC is 0.15 M NaCI, 0.015 M sodium citrate, pH 7.0), 0.1%sodium dodecyl sulfate (SDS), 30%o formamide, 100 ,g/mlsonicated salmon sperm DNA, 5x Denhardt's solution (1 xDenhardt's solution is 0.02% Ficoll 400,0.02% bovine serumalbumin (BSA), 0.02% polyvinylpyrrolidone) at 370C for 2 h.Probes were added to the prehybridization solution and hy-bridized under the same conditions for 24 h. The probesconsisted of the PCR-derived genomic fragments of the gpa-1, gpa-2, and gpa-3 genes. These were isotopically labeledwith a Multiprime kit (Amersham, Arlington Heights, IL) usinga-32P-dATP. The probes were denatured by boiling for 5 minbefore use, and 1 x 1 o6 cpm/ml of each probe was addedto the hybridization solution. After hybridization, the filterswere washed with 0.1X SSC, 0.1% SDS at 230C and ex-posed to X-ray film for 48 h.

Northern blot analysisTotal RNA was isolated by the guanidine isothiocyanateprocedure described by Maniatis et al. (1982) from the N2strain. The RNA was denatured, fractionated on a 1% form-aldehyde gel, and blotted to a nitrocellulose filter. The insertfrom the G.a cDNA was isotopically labeled using a-3P-dCTP by the method of Feinberg and Vogelstein (1983). Thefilter was hybridized to the probe in 50Yo formamide, 5xSSC, 5x Denhardt's solution, 25 mM NaPO4, 0.1% SDS,and 0.25 mg/ml salmon sperm DNA at 420C. The filter waswashed in 0.2x SSC, 0.1% SDS at 600C and exposed to X-ray film.

Immunoblot analysisMixed populations of N2 nematodes at various develop-mental stages were grown on 10-cm NGM plates (Sulstonand Hodgkin, 1988) and washed off with M9 buffer. Thenematodes were pelleted and 5x concentrated gel samplebuffer (Laemmli, 1970) was added to a final concentrationof 1 x. The nematodes were boiled for 5 min and the dis-solved proteins were fractionated by SDS electrophoresison a 10Yo acrylamide/0.27% bis-acrylamide gel (Laemmli,

1970). The proteins were electroblotted onto a nitrocellulosefilter by the method of Towbin et al. (1979) in 25 mM Tris-HCI, 192 mM glycine buffer, pH 8.3, 20% methanol at 200mA for 2 h. The filter was blocked in TBS (20 mM Tris-HCI,pH 7.5, 500 mM NaCI)/3% BSA for 30 min and incubatedin the primary antibody for 1 h at room temperature in TBS/1% BSA. The filter was washed with TBS for 30 min andthen incubated in the secondary antibody for 1 h at roomtemperature. Goat anti-rabbit IgG (Fc) conjugated to alkalinephosphatase (Promega, Madison, WI) was used as the sec-ondary antibody at a dilution of 1:7500 in TBS/1 % BSA.After washing in TBS for 10 min, we developed the blot withnitro blue tetrazolium (NBT) and 5-bromo4-chloro-3-indolylphosphate (BCIP) according to the manufacturer's instruc-tions.

Isolation and mapping of cosmid clonesA cosmid library was constructed from N2 (Bristol) genomicDNA that was partially digested with Xho II. The DNA wassize-fractionated by field inversion gel electrophoresis andsubcloned into the BamHl site of pWEKan (J. Mendel, un-published). The resulting cosmid library contains about threeor four genome equivalents, and the average insert size is35-50 kilobase pairs. The library was gridded in microtiterplates and replicated to nitrocellulose filters. PCR fragmentsfrom the gpa-1, gpa-2, and gpa-3 genes were labeled witha Multiprime kit (Amersham) using a-32P-dCTP and hybrid-ized to the nitrocellulose filter replicas in 1% BSA, 7% SDS,1 mM EDTA, 0.5% formamide, 0.2M NaPO4, pH 7.2 at 65°Cfor 12 h using 4x 105 cpm/ml of probe. The filters werewashed in 2x SSC at 500C for 30 min, 0.2x SSC at 500Cfor 30 min, 0.2x SSC at 650C for 30 min and then exposedto X-ray film. Positive clones were picked, DNA was preparedfrom them, and their identities confirmed by PCR analysis.The cosmid clones were mapped by Alan Coulson andJohn Sulston (MRC, Cambridge, England) using a finger-printing technique that compares the pattern of HindIll-Sau3A1 restriction fragments of a cosmid to that of anoverlapping set of cosmids (contigs) that now represents>95% of the C. elegans genome (Coulson et at., 1986; A.Coulson, J. Sulston, and R. Waterston, personal communi-cation).

DNA sequence analysisDNA sequencing reactions were performed using the Se-quenase, version 2.0 kit according to the instructions pro-vided by the manufacturer (United States Biochemical,Cleveland, OH). Double stranded miniprep DNA for use asa sequencing template was prepared by a modification ofthe boiling lysis method of Holmes and Quigley (1981). Inthis procedure 1.5-ml cultures were grown in Luria brothwith ampicillin (100 yg/ml) for 6-8 h or just until the cultureswere saturated. The cells were pelleted in 1.5-mi microfugetubes and resuspended in 300 ,l STET (8% sucrose, 5%Triton X-100, 50 mM EDTA, 50 mM Tris-HCI, pH 8.0).Twenty microliters of lysozyme solution (10 mg/ml in 50mM Tris-HCL, pH 8.0) were added. After incubation at roomtemperature for 1 min, the mixture was boiled for 2 minand immediately microfuged at room temperature for 5min. The pellet was removed with a sterile toothpick and300 Ml 75% isopropanol, 2.5 M ammonium acetate wasadded. After mixing, the tubes were microfuged at roomtemperature for 5 min. The supernatant was aspirated andthe pellet was washed in 1 ml 70% ethanol, 1 ml 100%ethanol, then dried in a Spin-Vac. The pellet was resus-pended in 50 Al H20. The plasmid DNA was denatured


M.A. Lochrie et al.

TGTACTTCTAGAACTTCAAGATTCACATTTATATTTTTTTAGTTTTCTATCAAGTCTTGA 2040sPheLeuSerSerLeuAs

TCGTATCAAGTTAC CG&TTATAATCCAACTGkCAGGI'KTTCTTCTGTCTAGAATCAA 2100pArgIleLysLeuProAspTyrAsnProThrGluGlnAspIleLeuLeuSerArgIleLy

GACAACTGGAATTGTTGAAGTTAAATTTCAAATGAAAAGTGTAGATTTTAGGTGTGAATA 2160sThrThrGlyIleValGluValLysPheGlnMetLysSerValAspPheAr

GTTAATGATTCATGAGTTTGCAAGACATGAGTTTCAGGGTATTGGACGTTGGAGGTCAAC 2220gValPheAspValGlyGlyGlnA

GAT AGAAATGGATTCATTGTTTTGAAGATGTTAATGCAATTATTTTCATCG 2280rgSerGiuArgLysLysTrpIleH.sCysPheGluAspValAsnAlaIleIlePheIleA

CTGCAATTTCAGAATATGATCAAGTTCTGTTTGAAG&TGAGACGACGGTTAGAACAAGGA 2340laAlaIleSerGluTyrAspGlnValLeuPheGluAspGluThrThr

AAACTTTATGGTTTATTTGAGAATTGTTTTTTTTTTTGCAGAATCGAATGATTGAATCAA 2400AsnArgMetIleGluSerM

TGAGGCTGTTTGAATCAATTTGTAATTCACGATGGTTCATCAATACTTCAATGATTCTTT 2460etArgLeuPheGluSerIleCysAsnSerArgTrpPheIleAsnThrSerMetIleLeuE

TCTTAAATAAGAAGGATTTGTTTGCCGAGAAGATTAAAAGAACTTCAATTAAATCCGCAT 2520heLeuAsnLysLysAspLeuPheAlaGluLysIleLysArgThrSerIleLysSerAlaP

TCCCAGACTATAAAGGTAAGATTTAATTGAACTTTTTCACACTCTGCCCATATATTTTCA 2580heProAspTyrLysG

GGTGCACAAACCTACGACGAATCGTGTCGATATATTGAGGAAAAGTTTGATGGATTGAAT 2640lyAlaGlnThrTyrAspGluSerCysArgTyrIleGluGluLysPheAspGlyLeuAsn

GCAAATC A AGAT TTACATGCATCAGA GTGTGCAAC GATACAGATCAAGTA 2700AlaAsnProGluLysThrIleTyrMetHisGlnThrCysAlamhr spThrAspGlnVal

CAAATGATTTTGGACTCAGTGATTGATATGATTATCCAGGCC ACTTGCAAGGATGCGGT 2760GlnMetIleLeuAspSerValIleAspMetIleIleGlnAlaAsnLeuGlnGlyCysGly

TTGTACTGAAATGTTTGTGCCTTTCTCTCTCTCTCTCTTTCTCTCGCTATTTCTATCACT 2820LeuTyrEnd

TGGAAATACACATAGGGAAAGCATTTTCAGAAAAAAAACTGGATTACTTTCTGTTAAGAA 2880GTAACCTGAAAATATACTTAATCGAAATTTAAGACAAACATGGTAAAGAAGGAAACATGG 2940GTAATGCCAGGTCAATTCTTTGTTATGA 2970ATGCATTCATTTTGAGTCTATTTTTGTATGCTTCTTTTTTATGTCTAAGCATTCGTCGAC 60ATTTGAAATATGCGTGAAATiiATTTTCCAGTGGAACTAGAGCTGGGGGGCTTCCTATTGA 120*TTTTTGATTGATTTTTGAAAAAGACT?TCGACCACGAATATCAATAAAATAATCTAAGTTT 180CCGGGAAGACTCGAAACATCAAAAAGAAACGACTGGAACTATTCCATCGAACAATAGTCT 240ATCTTTCGTTCATTTAGATGTGAAATTCATGATGATGTACAAATTGATTCTATTCGAAAG 300

Figure 9

CELL REGULATION148

Ga in C. elegans

ATTAATGTATTATTGAACATTTTCAGCTGAATCGAACATCTTTTATCACTCATTATTATG 360ATATTTGTTCATTTTCAATGCGTCACAGCTTGGTGTCGTTTTGTTGAATTGGCAGCACCA 420TTTCTTGTGAAAAAATGCTTGCTGTCGTCTTTCATTATTGTCGAGGTGAAACAG-AAAAAG 480AAAGAGACACACGAGATATCAACAACACAAAATTCTGAATATCTGACTTTGGCTCTGACC 540GCCGTCGCCGCCCGCCCGCTAACGAGCATTTTCTCATTTTCTTCTCATTTACCCCGTGTC 600TACTTCTTTTCTTATTTTTCGTTTCGTTAAACTTCGGTTGTCGCATAATCAGGCCTGCTT 660TTCATTTTTCACTTTTTGTTTGAACTTTTGGGCTTTTTAGAGTTGTGCTTCATTCAAATC 720

TATATTCAATTACAGAAAATTATTAGCCAAAGAACAATTATGGG&TTATGCCaATCTGCA 780MetGlyLeuCysGlnSerAla

GAGGACAAAGAACTGACGTTGAAATCAAAAGCG TCGATAAGGAAATGATGCAGAACCAT 840GluAspLysGluLeuThrLeuLysSerLysAlaIleAspLysGluMetMetGlnAsnHis

ATGTCACATCAAAAAGToTGAAGCTTCTGCTTTTGGGTAAAATTTTAAAACTTATTTTT 900MetSerGlnGlnLysValValLysLeuLeuLeuLeuG

TGTATTCAGCTTCGCGCAGTCAGAAATTGTTTTACGGTAATCATATGACTCATTGAGAAC 960TTGTTTGACTTTTTGATAATACGCCGGAATCAAATACAAGCAAAACAAAAGTGTCGTGTG 1020AGAAATTGTTACCCGCTTTTGTTGAAAAGTCCAAAAAATTTAATTTATGAGAACGTTGGG 1080ATAAAAAATATTATCGTTTCAGAAGAAATGTTCCCTTCTGTTCTCGAACACTTTTTCAAA 1140AAATTACTTCAAATTTGTAATAACTGTAATTAATAATTCAGAGAATGAAAGAGCTGTGAT 1200TTTTAAATACTATATTTACAGACAAAAAGGATAAGTCAGAGAAAATGAGAAATGATCAGT 1260AATAAGAATAAAGAACATCAACTCAATTTTCTCGAATCATATAAAATAATTATTAAAGTT 1320TTAAGTTTCTTCTCTCACCCGGCTTCTTGCAAACACTACTTTTTTAACTGCCCGCAAAAA 1380TTTTCATGTGTTCTAAATATTAATAATCAAATTATTTATCAAGTTATAATTCTCTTTTGA 1440AGCTTATCAAAGAACTACAAACTACGTAGAACCCGTTGATGTCAAATTTTCAGATATCAA 1500GTGTGCGTACGCTTAGGTTTTCAGCTGTTTCTTTCATTTGATTCTGAGAATTTAGAAGTT 1560GAAAAATAAATTAGCTGGGGTTTGTGACAAGTTGGGCAAAATTAAACTCTTACTAATTTC 1620

AGGTGCTGGAGAATGTGGAAAAAGTACGGTTCTG AACAAATG&GGATTCTTCACGATCA 1680lyAlaGlyGluCysGlyLysSerThrValLeuLysGlnMetArgIleLeuHisAspHi

CGGATTTACTGCAGAAGAGGCTGAACAACA GAAAAGTGTcTTTCAATAATACCCTTCA 1740sGlyPheThrAlaGluGluAlaGluGlnGlnLysSerValValPheAsnAsnThrLeuGl

AGCAATGACTGCAATTCTGAAAGGAATGGAAGCACTTCGAATGACCTTTGATAAGCCAAT 1800nAlaMetThrAlaIleLeuLysGlyMetGluAlaLeuArgMetThrPheAspLysProIl

TCGAGAAAATGATGCAAAATTTGTGATGGAGTCTCATAAAATGCTCCAAGAAGCGAAAQT 1860eArgGluAsnAspAlaLysPheValMetGluSerHisLysMetLeuGlnGluAlaLysva

TTTCCCAGAAGAATTAGCAAATGCCATTCAAGCATTATGGAATGATAAAGCCGTTCAGCA 1920lPheProGluGluLeuAlaAsnAlaIleGlnAlaLeuTrpAsnAspLysAlaValGlnGl

AGTTATTGCAAJ.AAGGAAATGAGTTCCAAATGCCTGAAAGTGCACCACAGTAAGATTTTAC 1980nValIleAlaLysGlyAsnGluPheGlnMetProGluSerAlaProHi

Figure 9. DNA sequence of C. elegans gpa-3 gene. The predicted amino acid sequence of the gpa-3 protein is shown belowthe DNA sequence. The standard three-letter amino acid abbreviations are given in the legend to Figure 2. Coding sequencesare shown in boldface lettering. The primers used in PCR reactions are underlined. oMP19 corresponds to nucleotide positions2235-2252, oMP20/oMP21 to positions 2460-2477, and Ta29 to positions 2674-2691.


M.A. Lochrie et al.

N 718 269

4443 127 541

N 1020 45 46 49 45 49 268 76 C

I 444 4 4 4 I

N 745 54 46 54 46 C

N C

I 44 444+4 4X* *

Figure 10. Intron sizes and locations in gpa-1, gpa-2, and gpa-3 coding sequences. The solid lines represent the codingregions of gpa-1, gpa-2, gpa-3, and Ga (mammalian Gia, G.a, or Ta). The arrows indicate the location of introns and thenumbers indicate their size in base pairs. The asterisks indicate intron positions that are conserved in all of the genes.

Abbreviations used are N, amino terminus; C, carboxyl terminus.

with alkali by adding 5 ,ul 2 M NaOH and 2 mM EDTA andincubating at room temperature for 5 min. The solutionwas neutralized by adding 25,gl 0.9 M sodium acetate, pH5.3. Plasmid DNA was precipitated by adding 200 IA

ethanol, mixing, incubating at -700C for 5 min, and mi-crofuging for 10 min at 4°C. The supernatant was aspiratedand the pellet was washed in 70% ethanol and dried in a

Spin-Vac. The denatured DNA was resuspended in 20 ,iH20. For DNA sequencing reactions, 7 ,l denatured tem-plate was annealed to 1 Ml (10 ng) primer with 2 ,ul 5xSequenase buffer at 370C for 20 min. For DNA sequenceanalysis of the Goa cDNA, restriction fragments of XCe6-2 were subcloned into pBluescript KS/-. Most of the se-

quence of these inserts was determined using sequencingprimers that are complementary to pBluescript KS/-. Theremaining sequence was determined using synthetic oli-gonucleotides as insert specific sequencing primers. A T7DNA polymerase stop obscuring the sequence of positions171-181 was resolved by the use of the TaqTrack DNAsequencing system (Promega) at a reaction temperatureof 80°C. For DNA sequence analysis of the gpa-1 and gpa-3 genes, HindlIl restriction fragments of the gpa-1 and gpa-3 cosmid clones that hybridize to the gpa- 1 and gpa-3 PCR-derived probes were identified by Southern blotting andsubcloned into pBluescript KS/- (see Figures 6A and 8A).To obtain the complete sequence of the gpa-3 gene, it wasalso necessary to subclone a Nsi I restriction fragmentthat overlaps the HindlIl fragment. The DNA sequence of

the gpa-1 and gpa-3 genes was determined by a combi-nation of three methods. First, specific restriction frag-ments of the original subclones were subcloned furtherand sequenced as described above. Second, yb transpo-sons were inserted at random sites in subclones con-structed in pMOB (Strathmann et aL., 1991). The locationof the yb transposons was determined by PCR, and DNAsequences were determined by the use of transposon-specific primers corresponding to each of the differentends of -y6. Because yb generates a five base pair dupli-cation at the site of insertion, it is possible to obtain se-quences that overlap at the site of -yb insertion. Third, anyremaining sequences that were not obtained by the firsttwo methods were determined by constructing syntheticprimers based on available sequence and using these assequencing primers. Sequence compressions were re-solved using dITP instead of dGTP in the sequencing re-actions; T7 polymerase stops were resolved by the use ofthe TaqTrack system at a reaction temperature of 80°C.DNA sequences were analyzed using Pustell programs(Pustell and Kafatos, 1984) run on an IBM PC or the Uni-versity of Wisconsin Genetics Computer Group sequenceanalysis software package (Devereux et aL., 1984) run ona VAXstation 2000.

AcknowledgmentsWe thank Mike Strathmann for advice on PCR reactionsand for the provision of materials for the -6 transposon-

CELL REGULATION

56

4172 C

gpa-J

gpa-2

gpa-3

Ga

* *

I

150

1: MSN ES ELV AQAKQNK I LDNILKAKKTD ENI LLLLG AG KSTV2: MS L e w S EREKVGTLKSIIIID E IKQLQTSE ERT LLLLG AG GKSTVI3: MG eoSEK ELTLKSK1 D KEMQNHMSQ QKV KLLLL GEGKSTE

1: IKQIKITHNSG FST1 ISNKR N VQA ALLD Q LQF STRVC2: (HLLTSKQ YF LTQAKN VIE vADHL GPA AGLQ F SDPTR3: ERILHDHG FT QQK S LQA AILK A LRM DKPIR

1: NAHEKLIRET2: EHDVHMLTLY3: ENDAKFVMES

LNDKAEYGfIKDMQH-K1HKMLQEAKJ

SDAMFNALTEQQDAADHVEKPEELANAIQA

CA YDKR FFYLRL YAER LNIRQV IAKGtQMP

1: H--DSAK DRI HTPN TE GK2: DIGDNTE ENL ISKED EA LffGIi3: -- -ESAP SSL IKLPD E L

1: DVGGQRS2: VFDVGGQRS3: VFDVGGQRS

IKKWIHCIKKWIHC I E NEtKKWIHC E Ig lSuL2l

1: I L V K V L F L K SLG wSFS S2: L Si R I L FLN L K- ENIH EEYR -

3: SI3 R SML FLNj -ITSIK SYF-

1 LNAAVAFV EIYRSMAEj K I CATD LD

2: VNYAETVAFI K EALSN ETF CATD iV3: STYDESCRYI EF DGLNAJ TI CATD ID

247249247

297297295

347346344

1: ST GCL 3572: QS K GL 3563: Q'@ 354

Figure 11. Comparison of the amino acid sequences of C. elegans gpa-1, gpa-2, and gpa-3. Sequences were aligned usingthe FASTP program of Lipman and Pearson (1985). Dashes indicate gaps introduced to obtain an optimal alignment. One-letter amino acid abbreviations are given in the legend to Figure 3. Amino acids that are found in all three sequences at thesame position are boxed.

based sequencing method, the Caltech Microchemical Fa-cility for oligonucleotide synthesis, Andrea Holboke for C.elegans genomic DNA, Alan Coulson and John Sulston forgene mapping data, Eva Neer and Graeme Milligan for G.aantibodies, Stuart Kim and Russell Hill for the C. eleganscDNA library, and Yasumi Ohshima and Ron Plasterk forgenerous communication of results before publication.P.W.S. is an investigator of the Howard Hughes MedicalInstitute. Supported by a Developmental Biology Grant fromthe Lucille P. Markey Foundation and by USPHS grantsGM34236 to M.I.S. and HD23690 to P.W.S.

Received: October 5, 1990.Revised and accepted: November 16, 1990.

ReferencesBallinger, D.G., and Benzer, S. (1989). Targeted gene mu-tations in Drosophila. Proc. NatI. Acad. Sci. USA 86, 9402-9406.

Breathnach, R., and Chambon, P. (1 981). Organization andexpression of eucaryotic split genes coding for proteins.Annu. Rev. Biochem. 50, 349-383.

Brenner, S. (1974). The genetics of Caenorhabditis elegans.Genetics 77, 71-94

Buss, J.E., Mumby, S., Casey, P.J., Gilman, A.G. and Sefton,B. (1987). Myristoylated a subunits of guanine nucleotide-binding regulatory proteins. Proc. Natl. Acad. Sci. USA 84,7493-7497.Caron, M.G. (1989). The guanine nucleotide regulatory pro-tein-coupled receptors for nucleosides, nucleotides, aminoacids, and amine neurotransmitters. Curr. Opinion Cell Biol.1, 159-166.

Casey, P., Fong, H.K.W., Simon, M.l. and Gilman, A.G.(1990). G,, a guanine nucleotide-binding protein with uniquebiochemical properties. J. Biol. Chem. 265, 2383-2390.

Cerione, R.A., Regan, J.W., Nakata, H., Codina, J., Benovic,J.L., Gierschik, P., Somers, R., Spiegel, A.M., Birnbaumer,

Vol. 2, February 1991

G., in C. elegans

505050

100100100

149149150

197199197

151

M.A. Lochrie et al.

L., Lefkowitz, R.J., and Caron, M.G. (1986). Functional re-constitution of the a2-adrenergic receptor with guanine nu-cleotide regulatory proteins in phospholipid vesicles. J. Biol.Chem. 261, 3901-3909.

Chalfie, M., and White, J. (1988). The nervous system. In:The Nematode Caenorhabditis elegans, ed. W.B. Wood, ColdSpring Harbor, New York: Cold Spring Harbor Laboratory,337-391.

Coulson, A., Sulston, J., Brenner, S., and Karn, J. (1986).Toward a physical map of the genome of the nematodeCaenorhabditis elegans. Proc. NatI. Acad. Sci. USA 83,7821-7825.

Coulson, A., Waterson, R., Kiff, J., Sulston, J., and Kohara,Y. (1988). Genome linking with yeast artificial chromosomes.Nature 335, 184-186.

deSousa, S.M., Hoveland, L.L., Yarfitz, S., and Hurley, J.B.(1989). The Drosophila G,a-like G protein gene producesmultiple transcripts and is expressed in the nervous systemand in ovaries. J. Biol. Chem. 264, 18544-18551.

Devereux, J., Haeberli, P., and Smithies, 0. (1984). A com-prehensive set of sequence analysis programs for the VAX.Nucleic Acids Res. 12, 387-395.

Emmons, S.W. (1988). The genome. In: The Nematode Cae-norhabditis elegans, ed. W.B. Wood, Cold Spring Harbor,New York: Cold Spring Harbor Laboratory, 47-80.

Emmons, S.W., Rosenzweig, B. and Hirsh, D. (1980). Ar-rangement of repeated sequences in the DNA of the nem-atode Caenorhabditis elegans. J. Mol. Biol. 144, 481-500.

Feinberg, A.P. and Vogelstein, B. (1983). A technique forradiolabeling DNA restriction fragments to high specific ac-tivity. Anal. Biochem. 132, 6-13.

Fino Silva, I., and Plasterk, R.H.A. (1990). Characterizationof a G-protein a subunit gene from the nematode Caeno-rhabditis elegans. J. Mol. Biol. 215, 483-487.

Fire, A. (1986). Integrative transformation of Caenorhabditiselegans. EMBO J. 5, 2673-2680.

Freissmuth, M., and Gilman, A.G. (1989). Mutations of G,adesigned to alter the reactivity of the protein with bacterialtoxins. Substitutions at arg187 result in loss of GTPase ac-tivity. J. Biol. Chem. 264, 21907-21914.

Gabrion, J., Brabet, P., Dao, B.N.T., Homburger, V., Dumuis,A., Sebben, M., Rouot, B., and Bockaert, J. (1989). Ultra-structural localization of the GTP-binding protein G. in neu-rons. Cell Signaling 1, 107-123.

Gillespie, P.G., and Beavo, J.A. (1988). Characterization ofa bovine cone photoreceptor phosphodiesterase purifiedby cyclic GMP-sepharose chromatography. J. Biol. Chem.263, 8133-8141.

Gilman, A.G. (1987) G proteins: transducers of receptor-generated signals. Annu. Rev. Biochem. 56, 615-650.

Graziano, M.P. and Gilman, A.G. (1989). Synthesis in Esch-erichia coli of GTPase-deficient mutants of G.a. J. Biol. Chem.264, 15475-15482.

Gunderson, R.E., and Devreotes, P.N. (1990). In vivo recep-tor-mediated phosphorylation of a G protein in Dictyostelium.Science 248, 591-593.

Harris-Warrick, R.M,, Hammond, C., Paupardin-Tritsch, D.,Homburger, V., Rouot, B., Bockaert, J., and Gerschenfeld,H.M. (1988). An a40 subunit of a GTP-binding protein im-munologically related to G. mediates a dopamine-induced

decrease of Ca2l current in snail neurons. Neuron 1, 27-32.

Hescheler, J., Rosenthal, W., Trautwein, W., and Schultz, G.(1987). The GTP-binding protein, G., regulates neuronalcalcium channels. Nature 325, 445-447.

Hodgkin, J., Edgley, M., Riddle, D.L., and Albertson, D.G.(1988). Genetic map. In: The Nematode Caenorhabditis ele-gans, ed. W.B. Wood, Cold Spring Harbor, New York: ColdSpring Harbor Laboratory, 491-583.

Hodgkin, J., Horvitz, H.R., and Brenner, S. (1979). Non-dis-junction mutants of the nematode Caenorhabditis elegans.Genetics 91, 67-94.

Holmes, D.S., and Quigley, M. (1981). A rapid boiling methodfor the preparation of bacterial plasmids. Anal. Biochem.114, 193-197.

Hsu, W.H., Rudolph, U., Sanford, J., Bertrand, P., Olate, J.,Nelson, C., Moss, L.G., Boyd, A.E., III, Codina, J., and Birn-baumer, L. (1990). Molecular cloning of a novel splice variantof the a subunit of the mammalian G, protein. J. Biol. Chem.265, 11220-11226.

Huang, X.-Y., and Hirsh, D. (1989). A second trans-splicedRNA leader sequence in the nematode Caenorhabditis e/e-gans. Proc. NatI. Acad. Sci. USA 86, 8640-8644.

Itoh, H., Toyama, R., Kozasa, T., Tsukamoto, T., Matsuoka,M., and Kaziro, Y. (f988). Presence of three distinct molec-ular species of G, protein a subunit. Structure of rat cDNAsand human genomic DNAs. J. Biol. Chem. 263, 6656-6664.

Jones, D.T., and Reed, R.R. (1987). Molecular cloning of fiveGTP-binding protein cDNA species from rat olfactory neu-roepithelium. J. Biol. Chem. 262, 14241-14249.

Jones, T.L.Z., Simonds, W.F., Merendino, J.J., Jr., Brann,M.R., and Spiegel, A.M. (1990). Myristoylation of an inhib-itory GTP-binding protein a subunit is essential for its mem-brane attachment. Proc. Natl. Acad. Sci. USA 87, 568-572.

Kahano, Y., Tsai, S.-C., Adamik, R., Hewlett, E.L., Moss, J.,and Vaughan, M. (1984). Rhodopsin-enhanced GTPase ac-tivity of the inhibitory GTP-binding protein of adenylate cy-clase. J. Biol. Chem. 259, 7378-7381.

Kaiser, K., and Goodwin, S.F. (1990). "Site-selected" trans-poson mutagenesis of Drosophila. Proc. NatI. Acad. Sci. USA87, 1686-1690.

Kaziro, Y., Itoh, H., Kozasa, T., Toyama, R., Tsukamoto, T.,Matsuoka, M., Nakafuku, M., Obara, T., Takagi, T., and Her-nandez, R. (1988). Structures of the genes coding for G-protein a subunits from mammalian and yeast cells. ColdSpring Harbor Symp. Quant. Biol. 53, 209-220.

Kenyon, C. (1988). The nematode Caenorhabditis elegans.Science 240, 1448-1453.

Krause, M., and Hirsh, D. (1987). A trans-spliced leader se-quence on actin mRNA in C. elegans. Cell 49, 753-761.

Laemmli, U. (1970). Cleavage of structural proteins duringthe assembly of the head of bacteriophage T4. Nature 227,680-685.

Landis, C.A., Masters, S.B., Spada, A.M., Bourne, H.R., andVallar, L. (1989). GTPase inhibiting mutations activate thea chain of G. and stimulate adenylate cyclase in human pi-tuitary tumours. Nature 340, 692-696.

Lavu, S., Clark, J., Swarup, R., Matsushima, K., Paturu, K.,Moss, J., and Kung, H.-F. (1988). Molecular cloning and DNAsequence analysis of the human guanine nucleotide binding

CELL REGULATION152

Gar in C. elegans

protein GOa. Biochem. Biophys. Res. Commun. 150, 811-815 and erratum 153, 487.

Lipman, D.J., and Pearson, W.R. (1985). Rapid and sensitiveprotein similarity searches. Science 227, 1435-1 441.

Lochrie, M.A., and Simon, M.I. (1988). G protein multiplicityin eukaryotic organisms. Biochemistry 27, 4957-4965.

Maniatis, T., Fritsch, E.F., and Sambrook, J. (1982). MolecularCloning: A Laboratory Manual, Cold Spring Harbor, NewYork: Cold Spring Harbor Laboratory.

Masters, S.B., Miller, R.T., Chi, M.-H., Chang, F.-H., Bei-derman, B., Lopez, N.G., and Bourne, H.R. (1989). Mutationsin the GTP-binding site of G8a alter stimulation of adenylylcyclase. J. Biol. Chem. 264, 15467-15474.

McFadzean, I., Mullaney, I., Brown, D.A., and Milligan, G.(1989). Antibodies to the GTP binding protein, Go, antago-nize noradrenaline-induced calcium current inhibition inNG108-15 hybrid cells. Neuron 3, 177-182.

Milburn, M.V., Tong, L., deVos, A.M., Brunger, A., Yamai-zumi, Z., Nishimura, S., and Kim, S.-H. (1990). Molecularswitch for signal transduction: structural differences be-tween active and inactive forms of the protooncogenic rasproteins. Science 247, 939-945.

Moriarty, T.M., Padrell, E., Carty, D.J., Omri, G., Landau,E.M., and lyengar, R. (1989). Go protein as signal transducerin the pertussis toxin-sensitive phosphatidylinositiol path-way. Nature 343, 79-82.

Moss, J., and Vaughan, M. (1988). ADP-ribosylation of guanylnucleotide-binding regulatory proteins by bacterial toxins.Adv. Enzymol. 61, 303-379.

Mumby, S.M., Heukeroth, R.O., Gordon, J.l., and Gilman,A.G. (1990). G-protein a-subunit expression, myristoylation,and membrane association in COS cells. Proc. NatI. Acad.Sci. USA 87, 728-732.

Neer, E.J., Lok, J.M., and Wolf, L.G. (1984). Purification andproperties of the inhibitory guanine nucleotide regulatoryunit of brain adenylate cyclase. J. Biol. Chem. 259, 14222-14229.

Olate, J., Jorquera, H., Purcell, P., Codina, J., Birnbaumer,L., and Allende, J.E. (1989). Molecular cloning and sequencedetermination of a cDNA coding for the a-subunit of a G.-type protein ofXenopus laevis oocytes. FEBS Lett. 244, 188-192.

Proudfoot, N.J., and Brownlee, G.G. (1976). 3' Non-codingregion sequences in eukaryotic messenger RNA. Nature263, 211-214.

Provost, N.M., Somers, D.E., and Hurley, J.B. (1988). A Dro-sophila melanogaster G protein a subunit gene is expressedprimarily in embryos and pupae. J. Biol. Chem. 263, 12070-12076.

Pustell, J., and Kafatos, F.C. (1984). A convenient andadaptable package of computer programs for DNA and pro-tein sequence management, analysis, and homology deter-mination. Nucleic Acids Res. 12, 643-655.

Raport, C.J., Dere, B., and Hurley, J.B. (1989). Characteriza-tion of the mouse rod transducin a subunit gene. J. Biol.Chem. 264, 7122-7128.

Sagi-Eisenberg, R. (1989). GTP-binding proteins as possibletargets for protein kinase C action. Trends Biochem. Sci.14, 355-357.

Schmidt, C.J., Garen-Fazio, S., Chow, Y.-K., and Neer, E.(1989). Neuronal expression of a newly identified Drosophilamelanogaster G protein aO subunit. Cell Regul. 1, 125-134.

Senogles, S.E., Spiegel, A.M., Padrell, E., lyengar, R., andCaron, M. (1990). Specificity of receptor G-protein inter-actions. Discrimination of Gi subtypes by the D2 dopaminereceptor in a reconstituted system. J. Biol. Chem. 265,4507-4514.

Serventi, I.M., Price, S.R., Murtagh, J.J., and Moss, J. (1990).Structure and function of G.a. In: ADP-Ribosylating Toxinsand G Proteins, ed. J. Moss and M. Vaughan, Washington,DC: American Society for Microbiology, ,325-348.Sternweis, P.C., and Robishaw, J.D. (1984). Isolation of twoproteins with high affinity for guanine nucleotides frommembranes of bovine brain. J. Biol. Chem. 259, 13806-13813.

Strathmann, M., Hamilton, B.A., Mayeda, C., Simon, M.l.,Meyerowitz, E.M., and Palazzolo, M.J. (1991). Transposon-facilitated DNA sequencing. Proc. NatI. Acad. Sci. USA (inpress).Strathmann, M., and Simon, M.l. (1990). G protein diversity:a distinct class of alpha subunits is present in vertebratesand invertebrates. Proc. NatI. Acad. Sci. USA 87, 9113-9117.

Strathmann, M.P., Wilke, T.M., and Simon, M.l. (1989). Di-versity of the G-protein family: sequences from five addi-tional a subunits in mouse. Proc. Natl. Acad. Sci. USA 86,7407-7409.

Strathmann, M., Wilke, T.M., and Simon, M.l. (1990). Alter-native splicing produces transcripts encoding two forms ofthe a subunit of GTP-binding protein G.. Proc. Nati. Acad.Sci. USA 87, 6477-6481.

Strittmatter, S.M., Valenzuela, D., Kennedy, T.E., Neer, E.,and Fishman, M. (1990). G. is a major protein of growthcone protein subject to regulation by GAP-43. Nature 344,836-841.

Sulston, J. (1988) Cell lineage. In: The Nematode Caeno-rhabditis elegans, ed. W.B. Wood, Cold Spring Harbor, NewYork: Cold Spring Harbor Laboratory, 123-155.

Sulston, J., and Hodgkin, J. (1988). Methods. In: The Nem-atode Caenorhabditis elegans, ed. W.B. Wood. Cold SpringHarbor, New York: Cold Spring Harbor Laboratory, 587-606.

Sulston, J., Horvitz, H.R., and Kimble, J. (1988). Cell lineage.In: The Nematode Caenorhabditis elegans, ed. W.B. Wood,Cold Spring Harbor, New York: Cold Spring Harbor Labo-ratory, 457-489.

Thambi, N.C., Quan, F., Wolfgang, W.J., Spiegel, A., andForte, M. (1989). Immunological and molecular character-ization of G.a-like proteins in the Drosophila central nervoussystem. J. Biol. Chem. 264, 18552-18560.Towbin, H., Staehlin, T., and Gordon, J. (1979). Electropho-retic transfer of proteins from polyacrylamide gels to nitro-cellulose sheets: procedure and some applications. Proc.Natl. Acad. Sci. USA 76, 4350-4354.Towler, D.A., Gordon, J.l., Adams, S.P., Glaser, L. (1988).The biology and enzymology of eukaryotic protein acylation.Annu. Rev. Biochem. 57, 69-99.Ueda, H., Uno, S., Harada, J., Kobayashi, I., Katada, T., Ui,M., and Satoh, M. (1990). Evidence for receptor-mediatedinhibition of intrinsic activity of GTP-binding protein, Gi1 and


M.A. Lochrie et al.

G12, but not G. in reconstitution experiments. FEBS Lett.266, 178-182.

Van derVoorn, L., Gebbink, M., Plasterk, R.H.A., and Ploegh,H.L. (1990). Characterization of a G-protein #-subunit genefrom the nematode Caenorhabditis elegans. J. Mol. Biol. 213,17-26.

Van Dongen, A.M.J., Codina, J., Olate, J., Mattera, R., Joho,R., Birnbaumer, L., and Brown, A.M. (1988). Newly identifiedbrain potassium channels gated by the guanine nucleotidebinding protein G,. Science 242, 1433-1437.

Van Meurs, K.P., Angus, C.W., Lavu, S., Kung, H.-F., Czar-necki, S.K., Moss, J., and Vaughan, M. (1987). Deducedamino acid sequence of bovine retinal G.a: homologies toother guanine nucleotide-binding proteins. Proc. Nati. Acad.Sci. USA 84, 3107-31 1 1.

Weinstein, L.S., Spiegel, A.M., and Carter, A.D. (1988).Cloning and characterization of the human gene for the a-subunit of Gj2, a GTP-binding signal transduction protein.FEBS Lett. 232, 333-340.

White, J., Southgate, E., and Durbin, R. (1988). Neuroanat-omy. In: The Nematode Caenorhabditis elegans, ed. W.B.Wood, Cold Spring Harbor, New York: Cold Spring HarborLaboratory, 433-455.

White, J., Southgate, E., Thomson, J.N., and Brenner, S.(1986). The structure of the nervous system of the nematodeCaenorhabditis elegans. Philos. Trans. R. Soc. Lond. Biol.Sci. 314, 1-340.Wickens, M., and Stephenson, P. (1984). Role of the con-served AAUAAA sequence: Four AAUAAA point mutationsprevent messenger RNA 3' end formation. Science 226,1045-1051.Wood, W.B. (1988). Introduction to Caenorhabditis elegansbiology. In: The Nematode Caenorhabditis elegans, ed. W.B.Wood, Cold Spring Harbor, New York: Cold Spring HarborLaboratory, 1-16.Worley, P.F., Baraban, J. M., Van Dop, C., Neer, E.J., andSnyder, S.H. (1986). G., a guanine nucleotide-binding pro-tein: immunohistochemical localization in rat brain resem-bles distribution of second messenger systems. Proc. Natl.Acad. Sci. USA 83, 4561-4565.Yatani, A., Mattera, R., Codina, J., Graf, R., Okabe, K., Padrell,E., lyengar, R., Brown, A., and Birnbaumer, L. (1988). TheG protein-gated atrial K' channel is stimulated by three dis-tinct Gia-subunits. Nature 336, 680-682.Yoon, J., Shortridge, R.D., Bloomquist, B.T., Schneuwly, S.,Perdew, M.H., and Pak, W.L. (1989). Molecular character-ization of Drosophila gene encoding G.a subunit homologue.J. Biol. Chem. 264, 18536-18543.

CELL REGULATION154

homologous uniqueg caenorhabditis elegans

Documents