the drosophila g protein y subunit gene (d-gyi) produces three

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 267, No. 9, Issue of March 25, pp. 6086-6092,1992 Printed in U.S.A.

The Drosophila G Protein y Subunit Gene (D-GyI) Produces Three Developmentally Regulated Transcripts and Is Predominantly Expressed in the Central Nervous System”

(Received for publication, October 16, 1991)

Kausik Ray$ and Ranjan GangulyQ From the Department of Zoology and the Cell, Molecular, and Developmental Biology Program, University of Tennessee, Knoxville, Tennessee 37996

A genomic clone, 536, located at the 44CD region of polytene chromosomes of Drosophila rnelanogaster, has been characterized for its neurobiological importance. We found that this clone contains a gene which produces 2.6-, 1.3- and 1.1-kilobase (kb) RNAs. While the 2.6-kb RNA is expressed only in the head, the 1.3- kb RNA is present exclusively in the body. The 1.1-kb RNA, however, is found in both the head and body, but in much higher concentration in the head. DNA sequence analysis of a 2.6-kb RNA-specific cDNA showed that this gene encodes a 70-amino acid polypeptide which is the putative Drosophila homologue to the y subunit of the bovine G-protein. The Drosophila protein, named D-Grl, shares 46, 43, and 28% identity, and 59, 62, and 60% similarity, with the 72,73, and yt proteins of bovine G proteins, respectively. Sequencing of the 1.1-kb RNA-specific cDNA clone revealed that the 1.1-kb RNA is produced from the 2.6-kb transcription unit by usage of an alternative polyadenylation site, and has a coding region identical to that of the 2.6-kb RNA. Genomic Southern blot hybridization indicated that the Drosophila genome has only one D-Wl gene. Throughout development the 1.1-kb RNA is found to be the most prevalent species; its level peaks between 9 and 12 h of embryogenesis. As is the case for the other G protein genes of Dro- sophila, the D-G-yI gene is predominantly expressed in the central nervous system of the fly.

Transduction of extracellular signals to intracellular effec- tor molecules is mediated by a family of GTP-binding proteins, the G proteins, which are heterotrimeric proteins com- posed of a, p, and y subunits (see Gilman (1987) and Simon et al. (1991) for review). When extracellular stimulants such as hormones, neurotransmitters, light, etc., interact with their specific receptors, the a subunit of the G proteins associates with the activated receptor and exchanges bound GDP for

* This work was supported by a University of Tennessee Faculty Research Award (to R. G.) and the Department of Zoology, University of Tennessee. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in thispaper has been submitted to the GenBankTM/EMBL Data Bank with accession number(s) M85042. + This work will be submitted for partial fulfillment of the degree of Doctor of Philosophy.

5 To whom all correspondence should be addressed Dept. of Zo- ology, University of Tennessee, Knoxville, T N 37996. Tel.: 615-974- 2371.

GTP. The activated a subunit, in turn, dissociates from the 8-y heterodimer and interacts with various intracellular effectors such as phosphodiesterase, phospholipase C , adenyl cyclase, phospholipase A2, and ion channels, generating second messengers and eliciting dramatic intracellular changes. Fol- lowing this interaction, the intrinsic GTPase activity of the a subunit hydrolyses GTP to GDP, inactivating the a subunit. The inactivated a subunit then reassociates with the Py complex.

Although the role of P-y complex in the signal transduction process is not clearly understood, it has been suggested that intracellular effectors such as phospholipase A2 and potassium channels may be directly stimulated by the B-y complex (Jelsema and Axelford, 1987; Logothetis et al., 1987; Kim et al., 1989). Also, the y protein may be involved in membrane anchoring of the G protein (Sanford et al., 1991). Gene cloning coupled with polymerase chain reaction has demonstrated that the p and y subunits exhibit amino acid sequence diversity in mammalian cells, and such diversity is also found for the a subunit of the G proteins (see Simon et al. (1991) for review). It is speculated that different combinatorial associa- tions of the three subunits may generate diverse types of G proteins, which may then interact with the variety of receptors involved in G protein-coupled signal transduction pathways.

By nature of their functions, G proteins are expected to be present in most cell types. Since not every cell type is exposed to similar kind of stimulants, variation in G proteins is expected. Moreover, the abundance and varieties of G proteins will also vary in different tissues depending on their physiol- ogy. In neural cells G proteins are abundant and have varieties of isotypes of a subunit (see Simon (1991) for review) with distinct functions. For example, in dorsal root ganglia, the Ga, regulates various potassium as well as calcium channels, which are known to be involved in neurotransmission (Yatani et al., 1988, 1990; Hescheler et al., 1987). The calcium channel is also regulated by Ga, (Yatani et al., 1988). The Ga, and Gaol* (Jones and Reed, 1989), which are abundant in the neural cells of olfactory epithelia, activate adenyl cyclase and in- crease the cAMP level. Interestingly, in Drosophila, a mutation of the dunce gene, coding for CAMP-phosphodiesterase, leads to elevated levels of cAMP and defects in memory and learning (Davis and Dauwalder, 1991). These findings imply that G proteins are involved in important biochemical pathways which control various neural functions.

Since the head of Drosophila is rich in neural tissues, studies on genes exclusively or predominantly expressed in the head will improve our understanding about the structure, function, and regulation of neurobiologically important genes, including the G protein genes. Often, a “reverse genetic” approach has been employed to isolate and study Drosophila homologues of

6086

y Subunit Gene of Drosophila G Protein 6087

various mammalian genes. Several groups have used the mammalian G protein gene probes to isolate the a (de Sousa et al., 1989; Quan et al., 1989; Thambi et al., 1989; Yoon et al., 1989) and @ (Yarfitz et al., 1988) subunit genes of Drosophila and demonstrated that, as in mammals, the a subunit in this species also shows amino acid sequence diversity. The y subunit gene of Drosophila, however, has not been isolated.

Instead of using a heterologous probe, we took a different approach to isolate Drosophila genes with neurobiological importance. Levy et al. (1982) isolated 20 cloned genomic DNA sequences containing genes of Drosophila melarwguster which are predominantly expressed in the head, rather than in the body of the fly. Since the major portion of the head is occupied by the eye and brain, it is expected that genes present in these clones may code for eye- and/or brain-specific proteins, including G proteins. Indeed, genes in one class of these 20 clones, showing predominant expression in the head of the fly but virtually no expression during embryogenesis (Levy and Manning, 1982), have been found to encode eye-specific proteins (Montell et al., 1985; Zipursky et al., 1985; Fryxell and Meyerowitz, 1987; Montell and Rubin, 1988; Krishnan and Ganguly, 1990; Yamada et al., 1990; Krishnan et al., 1991). We presumed that the genes in the class of clones which show peak expression during embryogenesis as well as in the adult head might have neural cell-specific functions, since genes with similar functions are known to be highly active during embryogenesis (Campos-Ortega, 1990). Thus, we initiated our study with clone 536, belonging to the latter class of clones and discovered that it contains the putative y subunit gene of Drosophila G protein. We demonstrate that the Drosophila y subunit gene (D-Gyl ) codes for three RNA species, of which two are produced by developmentally regulated alternative polyadenylation process. Like the other G protein genes of Drosophila, the D-Gyl gene is expressed predominantly in the nervous system of the fly.

MATERIALS AND METHODS

Fly Stocks, Molecular Biology Reagents, Radioactive Compounds, and Recombinant DNA Libraries-D. melanogaster, strain Oregon R, was used for DNA and RNA isolation. Flies were raised on standard corn meal-agar-molasses medium at 23 "C. The eym (eye missing, Lindsley and Zimm, 1985) mutant strain was obtained from Y. Inoue, Mishima, Japan. Restriction enzymes and all molecular biology reagents were purchased from Promega (Madison, WI) except the Sequenase version 2.0, which was obtained from U. S. Biochemical Corp. The reagents for DNA sequencing gels were from Bio-Rad, 32P- labeled compounds were from ICN (Irvine, CA) and 35S- and 3H- labeled deoxyribonucleotides were obtained from Du Pont-New Eng- land Nuclear. The genomic library of the Canton S strain of D. melanogaster was made at the BamHI site of XEMBW vector as described previously (Krishnan et al., 1991). The D. melanogaster head cDNA library, made in a AS WAJ2 vector (Palazzolo and Mey- erowitz, 1987), was a gift from M. Palazzolo and E. Meyerowitz, In this library, the 5'- and 3'-ends of the cDNA inserts are directed toward the EcoRI and XbaI cloning sites, respectively. The oligo(dT)- primed 3-12-h embryonic cDNA library of D. melanogaster was obtained from T. Kornberg.

Nucleic Acids Isolation and Blot Hybridization-Genomic DNA and RNA were isolated according to Krishnan et al. (1991). Plasmid DNA was isolated according to standard procedure (Maniatis et al., 1982), and bacteriophage X DNA was isolated as described by Helms et al. (1982). DNA fragments, to be used as probes, were labeled with radioactive deoxynucleotides by nick translation (Maniatis et al., 1982) or the random hexamer-primed labeling procedure (Cobianchi and Wilson, 1987). For Southern blot hybridization, DNA digested with restriction enzymes was blotted onto HyBond nylon membrane (Amersham Corp.), baked, and hybridized in 6 X SSC, 5 X Denhardt's solution, 0.1 mg/ml salmon sperm DNA, and 3ZP-labeled DNA probe at 68 'C for 20 h. Blots were washed at 60 "C in 0.1 X SSC containing

0.1% SDS' (high stringency washing) and exposed for autoradiography at -80 "C. In the case of low stringency conditions, blots were hybridized at 50 "C in the hybridization solution and washed at 45 "C in 2 X SSC and 0.1% SDS.

For Northern blot hybridization, RNA samples were fractionated in 1.2% formaldehyde-agarose gel as described (Maniatis et al., 1982) and blotted onto HyBond nylon membranes. The blots were hybridized in a solution containing 6 X SSC, 5 X Denhardt's solution, 0.1 M PIPES, pH 6.4, 50% formamide (Fluka Chemicals), 10% dextran sulfate, 0.1 mg/ml salmon sperm DNA, and labeled DNA probe. Following hybridization at 42 'C for 18 h the blots were washed four times in 2 X SSC at room temperature, four times in 0.2 X SSC and 0.1% SDS at 60 "C, and exposed for autoradiography at -80 "C.

Transcription Mapping-To map the transcribed regions within a genomic clone, cloned DNA was digested with various restriction enzymes, electrophoresed on agarose gel, and Southern blotted. The resulting blot was then hybridized with 32P-labeled, random hexamer- primed, single-stranded heterogeneous cDNA probes made against Drosophila head poly(A)+ RNA population (Krishnan et al., 1991). The transcribed strand of a given DNA fragment was determined by generating strand-specific riboprobes and hybridizing them individually with the Northern blots of head poly(A)+ RNA as described previously (Krishnan et al. 1991).

DNA Sequencing-The restriction fragments of the cDNA and genomic DNA were subcloned in pBSK+ plasmid. Overlapping dele- tions of the cDNA and genomic DNA fragments were generated by using EroIII/Sl nuclease deletion system (Promega) and sequencing was done by dideoxy chain termination method (Sanger et al., 1977) using Sequenase version 2.0 (U. S. Biochemical Corp.). Wherever needed, specific oligodeoxyribonucleotide primers were used for sequencing. The sequencing strategy is shown in Fig. 1B. Nucleotide and amino acid sequence analyses were performed by using the TFASTA, GAP, and PILEUP programs of the University of Wiscon- sin Genetics Computer Group (UWGCG).

In Situ Hybridization-For tissue localization of RNA, fly heads were embedded in OCT compound (Tissue-Tek), frozen in liquid nitrogen, and sectioned (8 pm) by using a Reichert cryostat. Sections were placed on subbed slides and fixed in 4% paraformaldehyde. Subsequent treatments of the sections, hybridization with 35S- or 3H- labeled probes, and washings were done as described (Ashburner, 1989). The 36S- and 3H-Iabeled probes, with specific activities of 1-2 X lo9 cpm/pg and 2-3 X lo7 cpm/pg, respectively, were generated by the random hexamer priming method (Cobianchi and Wilson, 1987). Hybridization signals were detected by autoradiography.

RESULTS

Identification of the Transcription Units in Clone 536 S h w - ing Head-specific Expression-Clone 536 has been mapped to the region 44CD of polytene chromosomes (Levy et al., 1982). By hybridizing immobilized DNA of clone 536 with 32P- labeled head and body poly(A)+ RNA of adult Drosophila Levy et al. (1982) showed that the RNA encoded by genes present in this clone are at least 10-fold more abundant in the head than in the body. This, however, did not reveal the number and location of different genes present in this clone, or their levels of expression in the head and body of the fly. To determine this, single-stranded cDNA probe made against poly(A)+ RNA of adult head was hybridized with EcoRI- digested DNA of clone 536. Only two EcoRI fragments, 1.5 and 2.0 kb, located next to each other (Fig. lA, fragments A and B ) , hybridized with the cDNA probe. Thus, the transcription units active in the head of the fly are located in these two EcoRI fragments.

To determine the extent of the transcription unit and its expression in head and body, sub-fragments of fragment A (Fig. lA, fragments C and D ) and fragment B were used individually as probes on Northern blots of head and body poly(A)+ RNA. In order to detect the hybridization signals with the body RNA, five times more body RNA than head

The abbreviations used are: SDS, sodium dodecyl sulfate; PIPES, 1,4-piperazineethanesulfonic acid kb, kilobase pair(s); nt, nucleotide(s); ORF, open reading frame; UTR, untranslated region.

6088 y Subunit Gene of Drosophila G Protein

A t 5'

H R H H A R K K SH I I I1

* R P S H H R P P R K K S H S H H R *

536.2

I I1 I I l l

I kb A- c- - 0

- G - - T E

-0

536-02 An-

P S H H R P

536.Cl R P

An & 536-A5 An er

P S H H R

An 536.62 cDNA - " "e -

A n d 5 3 6 C l cDNA

" _j

"

FIG. 1. Panel A , restriction maps of 536 genomic and cDNA clones. Genomic clone 536-1 was isolated by Levy et al. (1982). The other genomic clone, 536-2, was isolated from a genomic library constructed in XEMBL3 vector. The hatched bars in the genomic clones indicate the transcribed regions, the boundaries of which have not been determined. The PstI sites within the two EcoRI fragments, A and B, of clone 536-1 were determined in the subclones of these two DNA fragments. These sites in the entire clone 536-1 and 536-2 have not been determined. The open bars represent different cDNA clones of the D-G-yl gene. The direction of transcription was determined by hybridizing strand-specific riboprobes of fragment D to the head and body poly(A)+ RNA. Fragments A-D are from 536-1, and fragments E-F are from 536-B2 clones. These fragments correspond to the respective restriction fragments shown in the maps of 536-1 and 536- B2 clones. Restriction sites with asterisks are the cloning sites in XCharon 4 vector. R, EcoRI; H, HindIII; S, SstI; P, PstI; K, KpnI. Panel B, diagram showing the strategy of sequencing the 536 genomic and cDNA clones. The solid bones indicate the coding regions. The open triangle on the genomic DNA represents an intron in the 5'- UTR. Arrows with boxes indicate sequencing with synthetic oligodeoxyribonucleotide primers.

RNA was used. While fragment C hybridized only with a 2.6- kb RNA (Fig. 2 A ) , both fragments D and B hybridized with 2.6-, 1.3-, and 1.1-kb RNAs (Fig. 2, B and C). It is also evident from the results that the 2.6-kb RNA is head-specific, the 1.3- kb RNA is body-specific, and the 1.1-kb RNA is expressed both in head and body, but its level is much greater in the head.

Clone 536 Contains the Drosophila G Protein y Subunit Gene-Since the expression of the 2.6-kb RNA is head-specific, we decided to characterize the cDNA clones specific for this RNA. We used fragment C (Fig. LA), which hybridizes only with the 2.6-kb RNA, as a probe to screen a Drosophila head cDNA library and isolated several cDNA clones. The restriction pattern of the longest (2535 base pairs) cDNA clone, 536-B2, is identical to the restriction pattern of the transcribed regions within clone 536 (Fig. lA). Also, the patterns of hybridization (data not shown), with respect to the sizes and levels, of the head and body RNA with the sub-

'"k B l E HMBS H BJ

26-

?

1.3 - 1.1 -

B C

FIG. 2. Hybridization of head and body poly(A)+ RNA with different regions of the D-G-yl genomic DNA. All DNA fragments were labeled with "*P by the nick-translation procedure. Blot shown in panel B was reused for panel A after removal of the probe. DNA fragments (see Fig. L4) used as probes were: panel A, fragment C; panel B, fragment D; panel C, fragment B. Lunes H, 1 pg of head RNA; lunes B l , 1 pg of body RNA; lunes B5,5 pg of body RNA.

fragments of 536-B2 cDNA, i.e. fragments G, F, and E (Fig. LA), are identical to those observed with fragments C, D, and B of 536-1 clone (Fig. 2), respectively. Although the precise locations of the 5'- and 3'-ends of 536-B2 cDNA within the genomic clone are not known, this cDNA hybridizes only with the 6.0- and 2.0-kb EcoRI fragments of the 536-2 genomic clone (Fig. lA). It is possible that the 2.6-kb transcription unit is fully contained within these two EcoRI fragments.

Nucleic acid sequencing results showed that 536-B2 cDNA is almost a full-length copy of the 2.6-kb RNA (Fig. 3). Comparison of the genomic and cDNA sequences (Fig. 3) revealed that no intron is present between nt 77 and the first HindIII site located at n t 997, the region which fully contains the coding sequence. However, an intron, the length of which has not been determined, is present between nt 76 and 77 of the DNA. Although we have not sequenced the genomic DNA downstream to the first HindIII site, the spacings between restriction enzyme sites on the genomic DNA and on the cDNA sequence (Fig. lA) are very similar, indicating that if there is any intron present downstream to the first HindIII site, it must be very small. It is likely that an AATAAA motif located at nt 2477 is used for polyadenylation of the 2.6-kb RNA.

Conceptual translation of the nucleotide sequence of 536- B2 cDNA showed that the longest open reading frame (ORF) is located between nt 96 and 308 of the cDNA with coding capacity of 70 amino acid residues (Fig. 3). Thus, the 2.6-kb RNA has 95- and 2227-nt-long 5'- and 3"untranslated regions (UTR), respectively (Fig. 3). For three reasons we believe that the ORF between nt 96 and 308 is meaningful. First, the ATG codon of this ORF is preceded by a tetranu- cleotide sequence, 5'-CATC-3', which shows close match with the consensus sequences for Drosophila ((C/A)AA(A/C)) and vertebrate (CANC) translation initiation (Cavener, 1987). Second, the sequences upstream of the first ATG have stop codons in all three reading frames. Third, the codon usage within this reading frame matches the Drosophila codon usage preference.

That the longest ORF is biologically meaningful became clear when a data base search showed similarities (Fig. 4) between the deduced amino acid sequence of the longest ORF and the y subunits of bovine G proteins reported by other groups (Hurley et al., 1984; Robishaw et al., 1989; Gautam et al., 1990). Alignment of the deduced amino acid sequence of the longest ORF and different y subunits of bovine G proteins is shown in Fig. 4. The Drosophila protein is found to be 46, 43, and 28% identical (and 59,52, and 60% similar) to the 72, 73 and yt proteins of bovine G proteins, respectively. The deduced molecular mass (8 kDa) of the Drosophila protein is also very similar to that of the bovine y proteins. Like the


FIG. 3. Nucleotide sequence of the 636-B2 cDNA. The open reading frame starts at nt 96. Amino acids are denoted by single-letter codes. The AA- TAAA polyadenylation signals are underscored with heauy lines. A potential polyadenylation signal (AATATA) is underscored with broken lines. The ar- rowheads indicate the ends of the 1.1-kb RNA-specific 5 3 6 4 1 cDNA. Nucleotide sequence derived from the genomic DNA is underlined. The downward arrow in the 5'-UTR indicates the presence of an intron, the length of which is not known. Nucleotide substitutions in the 15th codon and nt 311, found in the 536-C1 clone, are shown in boldface type. A few of the A(T)nA motifs are underscored with dots.

1

89

96

162

228

294

3 0 9

3 9 1

485

573

6 6 1

'1 49

8 3 1

925

1013

1 1 0 1

1 1 8 9

1 2 1 1

1365

1453

1 5 4 1

1629

1117

1805

1893

1 9 8 1

2069

2 1 5 1

2245

2333

2 4 2 1

2509

CATTCGTCAACTTTGCTCGATTTTAGCCCGTAGCCCGTAGAGTTCGAGATTTAAAAAAAAAAAAACATATAAGCCGGAGCTGG~

AGsZAK A

A E GAC TCC CTG CAG CAG CAG CGC GTC GTG GTG GAG CAG CTG CGT CGC GAG G U

+ A

M D V M S S S L Q Q Q R V V V E Q L R R E A

G.€LAK GAC CGC CAG A C G ATC K G TCA TGC G C I B A G ATG ATG AAG TAC ATC ACG GAG CAC GBG

A I D R Q T I S E S C A K M M K Y I T E H E

CAG GAG GAC T S C T G CTC ACC GC,G TTC ACC AGC CAG ABC: GTG BAT CCC TTC CGC GAG AAG TCG TCC

Q E D Y L L T G F T S Q K V N P F H E K S S - C T V L *

GGAGGCGGTGAGGATGGTCCGCCGTTGCCG&G%EA!&GATCGCCGGCG T

E € m L

I ,,

1 I ,

G A C A A T T A A T G C G C T G A C T A T A G C A C C T A C T A C ' T C ' r A C C T A A C C C R ' I c , ,

- A H i n d U I

CGAGGCGRGTGTRAGAAGGAAAAGTTG'TATT GAAAATATAAGTATTTTGAAGCTT ATTAACCACC ""

~ I I AGCGGGTGGAACGACACCACAAACAACAARTGTACGAAAATATCTTA'r'rGAATTACTTAAAAGCTTTAACACGAAGCAGAAATGC'~AC

CCACAACAACCAGCTAGTTGAACTTATCGCAATATTTAGTGCGTTTATATCAGTTTTTCTTAG'TCAAAG~ATATAACGAACGTTGCAA

TAAGCGACAAGAGCATATATATACCAAATACATTCAAAATGTAATCAATATTTGTCTCT'rAGCATAG'rTTC'rTCA'rTTTAATTCACT~

CGGCGTACGAAGCCGATATCCAATTGAATGATAGCAAAAAGTTTCGCAAGACCAATAACTTATTTCATTCCCGTAACAACACAACCAA

CACACAGATGTCCTCAGAAACTCGAAAGC'rGTGGCGTTTCTCTAGTTTGGCGAAAATGCCGAGAAGAATGTGTGATTCT~CCTCC'~AA

ACGCATCCATTCTAATCATATATGTAAACGTGTTGAGCGCAATTAATTAAAGTGTTATTGGCATGAGAGAAAAGTAACCACACAAGCA

CAGTTATTTTTGTACAAACTAACATGGCATAGCCGACCTTTGTTCGGTATTGGAATGATAGACGCAATGATTATCCTGTAGGGAGAGC

ACGCTATGATATCAGCATACACCCACCAAGTCGTACATACACACATATTGCATCCCGAACTTACGATTCTCGTCCCGCTGAGACGAAC

TGTCCAAGTCGCCAGCTCAGTGTGGATGGGAATGAGGAAGGTATAGTTTATTTGATCGGTTCTTCCAAAAT'~GCA'~T'rAACAAGAGTT

TTTGAATTGGCGGTGATTTTAGT'rCTTAACGTAGCAAAAATTCAGACAACGTTTACTTAGTACCAAGAATCACACGTTTTTGGAGACG

.........

...............

............ ..............

GACATTTTCTATTCGGAAACCTATATAAAAGGTTTTGA~CTCCAGTTTGCTGTCG~GAGCAAACAGCAAAATGTGGTCCAAAGCCGCC m

ATTGCGAAAGGCTTTCATGGGAGCCGCCCTATCAATCGAAACTGCAGTCGGTTACGGGAGTCCTCCAGTTTCGGCTTAAACATTCCGC

ATTCAACAGTTTTCGATCCAACGGACACTTTGCATCCAAACTT'TAGTTATCC'rGAA~G'rTATTAGCCCGAGGACAGATTGAACCACC~

AATGGAGCACGGATGGTTTGGTTTCGTCCTCAGATGTGTGTCATTTTCTATAACTTAT~CATAGAGGGTAAATCGCATTTGGAAAGGAAG

GGCATTAGATGCACACGTACTCAAAATCAATCATGGTCATCGAATACACTTTAACACCTTTCACTTCAAATGAAAAGCGACAAC(;ACA

TCGTATTTTATTCGAAAATTAA'rACAACTCTTTTGCGATGCAAACGGGGACGAGTCCTCACAAATATTATTGAAATGTA~C'~GCGAAT

AAGCATTTGAAATATTGTACTCAACAAAATGAATAATCGTTTGCTTGAATTTCAAA~AAGCGATAACTTACTTAGATTCGTTA

TCTAGCGGTGGCACCGGGTTCGGTGCGA20

&LL

..............

FIG. 4. Comparison of the deduced amino acid sequence of the D-Gyl gene with the published amino acid sequence of bovine 72 (Robishaw et al., 1989; Gautam et al., 1990), y3 (Gautam et al., 1990), and retinal yt (Hurley et al., 1984) proteins. The dark and open boxes represent identical and similar amino acids between D-Gyl and any one bovine protein. Regions with greater similarity are ouerlined. The PILEUP program of UWGCG software package was used for sequence alignment.

gamma subunit of transducin, Drosophila protein is also an acidic protein, with calculated PI of 4.6. Hydropathy analysis showed that the Drosophila and bovine y proteins share similarly placed hydrophobic and hydrophilic domains (data not shown). Three regions of the Drosophila protein (Fig. 4, regions I-III), one in the amino-terminal and the other two

in the carboxyl-terminal halves, show greater than 75% identity with the bovine brain 7 2 and y3 proteins. However, less similarity between the Drosophila protein and bovine retinal y t proteins are found in these regions (Fig. 4). Regions I1 and I11 are also highly conserved between two bovine brain y proteins (Gautam et al., 1990) and speculated to be functionally important (Robishaw et al., 1990). Finally, as in all four bovine y subunits (Gautam et al., 1990), Drosophila protein also has a CAAX motif (where A represents amino acids with aliphatic side chain and X is any amino acid) at the carboxyl- terminal end (Fig. 4) which has been shown to be required for prenylation of the y protein by farnesyl and geranylgeranyl groups (Lai et al., 1990; Finegold et al., 1991; Sanford et al., 1991). These similarities led us to conclude that the 536-B2 cDNA encodes the Drosophila homologue to the y subunit of bovine G proteins. Thus, we designated the locus corresponding to this cDNA clone D-Gyl.

The Drosophila Genome Has Only One D-Gyl Gene-To determine whether sequences similar to D-Gyl gene are present elsewhere in the genome, we probed Southern blots of the genomic DNA of Oregon R strain with coding region containing DNA fragment E (Fig. lA) , under high and low stringent


conditions (Fig. 5 ) . Under both conditions, major hyhridization was observed with 2.0-kh EcoRI, 8.0-kh HindIII, and 2.0- a n d 1.4-kh PstI fragments (Fig. 5 ) . I t is evident from the restriction maps of the genomic clones (Fig. 1) tha t the fragments showing major hybridization originated from the locus 536. In the lane containing RarnHI-digested DNA, hybridization of only a single fragment was observed (Fig. 5, lane 4 ) . An 8.0-kh EcoRI fragment, showing faint hyhridization with the probe (Fig. 3, lane I ) , is either a product of partial EcoRI digestion or probably a result of EcoRI poly- morphism in the locus 536 present in a small population of Oregon R flies. Therefore, it appears that sequences similar to those in fragment E containing the coding region of the D- G y l gene are not present elsewhere in the genome and all three transcripts are products of the same gene. This agrees with the finding that clone 536 maps to a single cytological location 44CD (Levy et al., 1982) of polytene chromosomes. Furthermore, when strand-specific rihoprohes made from fragment D were used on Northern hlots of head and body RNA, hybridization of all three RNAs was observed with the rihoprohe made from one strand only hut not from the other (data not shown). These results indicate that the three transcripts are not products of nested transcription units.

The D-Gyl Gene Uses Alternative Polyadenylation Site to Produce More Than One Transcript-The origin of the three species of RNA from the D-Cy1 gene may he the result of alternative polyadenylation. This notion is based on the oh- servation that in addition to the polyadenylation signal for t h e 2.6-kb RNA present at nt 2477, another polyadenylation signal (AATAAA) is located at nt 845 (Fig. 3). It appears that utilization of the latter polyadenylation signal will yield the 1.1-kh RNA. To investigate this possihility, we co-screened a 3-12-h emhryonic cDNA lihrary with a 2.6-kh RNA-specific probe (Fig. 1, fragment C) and a prohe which hybridizes with all three RNA (Fig. 1, fragment D ) . We chose this library because in 3-12-h embryos the 2.6-kh RNAs is not detectahle and t he 1.1-kh RNA is the most prevalent species (Fig. 6). Several cDNA clones which hyhridized with fragment D, hut not with C, were isolated. cDNA clones of the head library which hybridized only with fragment D but not C were also isolated hy using the same strategy. Sequencing of an 817-nt- long embryonic cDNA clone (536-C1, Fig. L4) revealed that its 3'-end is located a t n t 870 which is 20 nt downstream from the first polyadenylation signal (Fig. 3). Except for two hase

8.0-

4.0- 3.-

2*

1 .o-

1 FIG. 5 . Southern hlot hybridization of t h e genomic DNA

digested with various restrict ion enzymes. I m w I, I~YJRI; lanr 2, /~,'roRI + HindIII; lnnr 3 , Ifindlll; lnnr 4, IhrnHI; lane 5 , P s t I . All lanes contained approximately 4 pg of DNA except lanes 3 and 4, which had 2 pg of DNA. For hybridization, fragment E (see Fig. 1) was labeled with ."I' by nick translation and used as a probe. lnft pnnrl. the hlot was hyhridized and washed under low stringent condition as described under "Materials and Methods." Rightpond, after autoradiographic exposure, the blot shown in Irft pnnrl was washed under high stringent condition and exposed. Both blots were exposed Sor 48 h.

v 2 . . 0 4 I 6 r 8

- flmOLrr -c

Fit. 6. Developmental express ion of the D-Crl gene. Ap- proximately 10 pg of t o t a l R S A isolat(4 from ditferent stnges wrrc fractionated on formaldehyde-agarose gel, transferrpd 10 nylnn membrane, and hybridized with '.'P-l;lheled fragment I.: (see Fig. 1 ) o f the .536-R2 cDNA. The arrom indicate the I.:<- and I . l -kh RNAs. / A ~ P I, 0-3-h embryo; lanr 2, :1-6-h emtxyo: lnnr 3 . Ci-9-h embryo: lrrnr* ./. 9-12-h emhryo; lanr ,5, 12-1.5-h embryo; lnnr 6, 15-18-h emhryo; lnnv 7, third instar larva; lanr 8. unstagetl pupae.

substitutions, nucleotide sequence of the 536-Cl cDNA is identical to that of the 2.6-kh RNA-specific cDNA (536-R2) between nt 51 and 870, including the coding region (Fig. 3) . The two hase suhstitutions are prohahly a result of polymor- phism. Sequencing of a 352-nt-long cDNA clone isolated from the head cDNA library (536-A5, Fig. 1A ) showed that the 3'- end of this cDNA is also located at nt 870. Sequence comparison revealed that this cDNA is identical to the 352 n t of the 3'-end of 536-C1 cDNA. These findings, therefore, indicate that the polyadenylation signal located at nt 845 is used to produce the 1.1-kh RNA, which has the same coding region a s t he 2.6-kh RNA. It is to he noted that A hexanucleotide sequence, AATATA, located a t n t 982, can be used as a polyadenylation signal to produce the 1.3-kh RNA. AATATA motif is believed to he used for polyadenylation in the case of maternal mRNA D7 of Xmopus (Smith et ai.. 1988).

Temporal and Spatial Expression of the I)-Cy1 Gme-Fig. 6 shows the expression of the D-Gy1 gene during emhryogenesis. A similar profile was obtained when dot hlots of the .536 genomic clone were prohed with "T-labeled RNA of different developmental stages of D. rnelanogustcv (Levy and Mmning, 1982). Present results, however, show the levels of different D-Gyl RNAs during development which were not determined in the previous investigation (Levy and Manning, 1982). The results show that during emhryogenesis the predominant species is the 1.1-kb RNA, the level of which peaks between 9 and 12 h of development. Since this RNA is first detectahle after 3 h of emhryogenesis, it is zygotically expressed. T h e 2.6-kh RNA is not detectahle during emhryonic development, and the 1.3-kh RNA is found as early as in 0-3 h embryos (Fig. 6). Since very few genes are transcribed in the first 3 h of emhryogenesis (Edgar and Schubiger, 1986), it is possible t h a t t h e 1.3-kh RNA is maternally transcribed and deposited into the egg.

Northern blot analysis (Fig. 2) indicated that the 2.6- and 1.1-kh RNAs of the D-Gy1 gene are predominantly expressed in the head. To determine whether expression of any of these two transcripts is eye-specific, RNA isolated from the heads of eye missing (eym) mutant flies was prohed with fragment D which hybridizes with all three RNAs. Fig. 7 shows that both 2.6- and 1.1-kh RNA are present in the Pyrn flies. Since t h e eyrn flies lack the compound eyes (Inoue, 1980; Lindslev and Zimm, 1985), the results imply that the 2.6- and 1.1-kh RNA are not expressed solely in the eye; they are either expressed both in the hrain and eve, or in the hrain onlv.

T h e specific site of expression of the 2.6-kb RNA in the head was determined hy in situ hyhridization of the 2.6-kh RNA-specific prohe (Fig. 1, fragment C ) to the RNA in the head sections. Heavy hybridization was ohserved over the cortical regions of the hrain and the optic lohe (Fig. 8. A and R ) , which contain the cell hodies of neurons. No significant


terminal amino acid is required for isoprenylation by the geranylgeranyl group (Finegold et al. 1991) which is thought to help the y protein for membrane anchoring (Sanford et al., 1991). Both cysteine and the terminal leucine are conserved in the CAAX motif of the D-Gyl protein, suggesting that this protein is also post-translationally modified for membrane anchoring. These similarities indicate that the D-Gyl is the putative Drosophila homologue of the y subunit of the mam-

FIG. 8. In situ hybridization of different regions of the 536- B2 cDNA clone with cryostat sections of head of wild type Drosophila. Panels A and C, bright field; B and D, dark field. ""S- Labeled fragment G (Fig. l.4) was used as a probe for the section shown in panel A . For the section in panel C , 3H-labeled fragment F (see Fig. 1A) was used as probe. r, retina; 1, lamina; 0, lobula; n, neuropile of the brain. The arrows indicate cortical regions of the brain and lobula.

hybridization signals were observed over the neuropile region of the brain. When similar head sections were hybridized with fragment D (Fig. l) , an identical hybridization pattern was observed (Fig. 8, C and D ) . Since fragment D hybridizes with both the 2.6- and 1.1-kb RNAs (see Fig. 2), the results of in situ hybridization imply that the 1.1-kb RNA is also expressed in the cortical regions of the brain and optic lobe. Both probes also hybridized lightly over the lamina and retina. These results indicate that the 2.6- and 1.1-kb RNAs of the D-Gyl gene are expressed predominantly in the central nervous system of adult D. melanogaster.

DISCUSSION

The results presented in this report demonstrate that the D-G-yl protein is similar to bovine y proteins in several respects. First, the D-Gyl protein shows 52-60% amino acid sequence similarity with the y subunits of bovine brain G proteins and transducin. Second, the deduced molecular weight of the polypeptide is also similar to that of the y subunit of mammalian G proteins. Third, the D-G-yl also has a carboxyl terminal CAAX motif, which is found in all three bovine y proteins, ras proteins, and yeast mating factor (Brake et al., 1985; Powers et al., 1986; Robishaw et al., 1989; Gautam et al., 1990). The cysteine residue in the CAAX motif has been shown to be the site for polyisoprenylation by farnesyl or geranylgeranyl groups (Lai et al., 1990; Sanford et al., 1991; Finegold et al., 1991). Furthermore, leucine as the

identical) than-they are to the y t protein (Gautam et al., 1990). Three regions of the D-Gyl protein of Drosophila have greater than 75% identity with the corresponding regions of the bovine brain y2 and 73 proteins, suggesting that these regions are functionally important and perhaps conserved in evolution. Further, like the D-Gyl gene, the bovine 72 and 73 genes are also predominantly expressed in the brain (Gau- tam et al., 1989, 1990). It is possible that the tissue-specific expression of the Drosophila and bovine brain y subunit genes may be regulated by similar mechanisms.

The D-Gyl gene has an intronless coding region and produces three RNAs. Our results show that, of these three RNAs, at least two (1.1 and 2.6 kb) are produced by the usage of alternative polyadenylation signals located at n t 845 and 2477, respectively. Since the relative levels of these two RNAs vary during embryogenesis and in the head, the usage of the two alternative polyadenylation sites is temporally and tissue- specifically regulated. Besides the alternative polyadenylation, selective mRNA degradation might be another mechanism which controls the level of the 2.6-kb RNA. This asser- tion is derived from the fact that the 3' UTR of this RNA is AT-rich (65%) and contains many A(T)nA motifs which are known to cause mRNA degradation (Shaw and Kamen, 1986). If such mechanism is operating, it is also temporally and tissue-specifically regulated because the 2.6-kb RNA is found only in the head. Furthermore, we believe that the 1.3-kb RNA is also produced by alternative polyadenylation by uti- lizing the hexanucleotide sequence AATATA located at nucleotide 986 because the AATATA motif is believed to be used for polyadenylation of maternal mRNA D7 of Xenopus (Smith et al., 1988). However, sequencing of 1.3-kb RNA-specific cDNA will resolve this issue. Further investigation is needed to clarify the significance of the alternative polyadenylation process and the length of the 3'-UTR as to the function of the D-Gyl gene.

The expression pattern of the D-Gyl gene is very similar to those of the Goa (de Sousa et al., 1989), Gia (Provost et al., 1988), and 6 subunit (Yarfitz et al., 1988) genes of Drosophila. They show a high level of expression in the head of the fly, and, interestingly, they all code for more than one RNA. Production of multiple mRNA may be a pattern common to the genes involved in neural functions. The expression patterns of the D-Gyl and Goa (de Sousa et al., 1989) genes are identical; both are expressed in the cortical region of the brain and optic lobe. Another interesting feature is that one of the many RNAs encoded by the a and 6 genes shows expression in 0-3-h embryos, suggesting that these RNAs are maternally inherited and that G proteins may have some role in early development and oogenesis. This is supported by the fact that mutation of concertina (cta) gene, encoding an a subunit of Drosophila G protein (Parks and Wieschaus, 1991), causes female sterility and embryonic lethality due to abnormal gastrulation (Schupbach and Wieschaus, 1989). High levels


of cta RNA are found in early embryo and ovary (Parks and Wieschaus, 1991). The D-Cy1 gene, reported here, also produces an RNA (1.3 kb), the level of which is very high in 0- 3-h embryos (Fig. 7). It is possible that the D-Cy1 gene is also transcribed in the ovary and maternally inherited. From the pattern of expression of the a, p, and y subunit genes it appears that in Drosophila G proteins are involved in at least two diverse functions, neuronal signal transduction and differentiation. It is believed that various forms of a, p, and y subunits in mammals may generate diverse types of G proteins in combinatorial fashion (Simon et al., 1991). In Drosophila, however, diversity in G protein may not be as great as in the mammals because only one y (this study) and one /3 (Yarfitz et al., 1988) subunit genes are believed to be present in this organism. Thus, different a subunits involved in different functions, such as neural signal transduction and gastrulation, will combine with only one type of y and p protein to form the heterotrimer. Future investigation will resolve the role of the same y and p subunits in diverse cellular functions in Drosophila.

Biochemical aspects of signal transduction and the role of G proteins in this process have been studied extensively in the mammalian system (Gilman, 1987; Simon et al., 1991) . However, the role of G proteins in various cellular functions is not well understood, although studies in other systems including Drosophila (Parks and Wiehaus, 1991; Wang et al., 1990; Walker and Bourguignon, 1990; Bengtsson et al., 1990; Hadwiger et al., 1991) imply that the G proteins may play important roles in diverse biological phenomena such as cy- toskeletal structure, cell-cell communication, differentiation, and fertility. The molecular basis of temporal and tissue- specific regulation of G protein gene expression in metazoan organisms is also not understood. Well established genetics and availability of various sophisticated genetic tools in Dro- sophila, including targeted mutagenesis (Ballinger and Ben- zer, 1989) and targeted gene replacement (Gloor et al., 1991), will be helpful in elucidating the various roles G proteins play in cell function.

Acknowledgments-We thank Mary Ann Handel and Bruce McKee for critical comments on the manuscript and A. C. Echternacht for his encouragements. We thank Sankar Mitra for his generous gift of oligonucleotide primers.

REFERENCES Ashburner, M. (1989) Drosophila: A Laboratory Manual, 1st Ed., pp.

173-180, Cold Spring Harbor Laboratory Press, Cold Spring Har- bor, NY

Ballinger, D. G., and Benzer, S. (1989) Proc. Natl. Acad. Sci. U. S. A.

Bengtsson, T., Sarndahl, E., Stendahl, O., and Anderson, T. (1990) Proc. Natl. Acad. Sci. U. S. A. 8 7 , 2921-2925

Brake, A., Brenner, C., Natrajan, R., Laybourn, P., and Merry- weather, J. (1985) Current Communications in Molecular Biology, pp. 103-108, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

86,9402-9406

Campos-Ortega, J. A. (1990) Adu. Genet. 27,403-453 Cavener, D. R. (1987) Nucleic Acids Res. 15, 1353-1361 Cobianchi, F., and Wilson, S. H. (1987) Methods Enzymol. 152,94-

Davis, R. L., and Dauwalder, B. (1991) Trends Genet. 7 , 224-229 de Sousa, S. M., Hoveland, L. L., Yarfitz, S., and Hurley, J. B. (1989)

Edgar, B. A., and Schubiger, G. (1986) Cell 44,871-877 Finegold, A. A., Johnson, D. I., Farnsworth, C. C., Gelb, M. H., Judd,

S. R., Glomset, J. A., and Tamanoi, F. (1991) Proc. Natl. Acad. Sci.

110

J. Biol. Chem. 264, 18544-18551

U. S. A. 88, 4448-4452 Fryxell, K. J., and Meyerowitz, E. M. (1987) EMBO J. 6,443-451

Gautam, N., Manfred, B., Aebersold, R., and Simno, M. I. (1989)

Gautam, N., Northup, J., Tamir, H., and Simon, M. (1990) Proc.

Gilman, A. G. (1987) Annu. Reu. Biochem. 56, 615-649 Gloor, G. B., Nassif, N. A., Johnson-Schlitz, D. M., Preston, C. R.,

and Engels, W. R. (1991) Science 2 5 3 , 1110-1117 Hadwiger, J. A., Wilkie, T. M., Strathmann, M., and Firtel, R. A.

(1991) Proc. Natl. Acad. Sci. U. S. A. 88,8213-8217 Helms, C., Graham, M. Y., Dutchik, J. E., and Olson, M. (1982) DNA

4,39-49 Hescheler, J., Rosenthal, W., Trantwein, W., Schultz, G. (1987)

Nature 325,445-446 Hurley, J. B., Fong, H. K., Teplow, D. B., Dreyer, W. J., and Simon,

M. I. (1984) Proc. Natl. Acad. Sci. U. S. A. 81,6948-6952 Inoue, Y. (1980) Drosophila Information Service 55,206 Jelsema. C. L.. and Axelrod. J. 11987) Proc. Natl. Acad. Sci. U. S. A.

Science 244,971-974

Natl. Acad. Sci. U. S. A. 8 7 , 7973-7977

, . I

84,3623-3627 Jones. D. T.. and Reed. R. R. (1989) Science 2 4 4 . 790-795 Kim, D., Lewis, D. L.,'Graziadei, L., Neer, E. J.; Bar-Sagi, D., and

Krishnan, R., and Ganguly, R. (1990) Nucleic Acids Res. 18,5894 Krishnan, R. K., Swanson, K. D., and Ganguly, R. (1991) Chromo-

Lai, R. K., Perez-Sala, D., Canada, F. J., and Rando, R. R. (1990)

Levy, L. S., and Manning, J. E. (1982) Dev. Biol. 94,465-476 Levy, L. S., Ganguly, R., Ganguly, N., and Manning, J . E. (1982) Deu.

Biol. 9 4 , 451-464 Lindsley, D. L., and Zimm, G. (1985) Drosophila Information Service.

62,113 Logothetis, D. E., Kurachi, Y., Galper, J., Neer, E. J., and Clapham,

D. E. (1987) Nature 325,321-326 Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular

Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

Clapham, D. E. (1989) Nature 337 , 557-560

SO^ 100 , 125-133

Proc. Natl. Acad. Sei. U. S. A. 8 7 , 7673-7677

Montell, C., and Rubin, G. M. (1988) Cell 5 2 , 757-772 Montell, C., Jones, K., Hafen, E., and Rubin, G. (1985) Science 2 3 0 ,

Palazzolo, M., and Meyerowitz, E. M. (1987) Gene (Amst.) 5 2 , 197-

Parks, S., and Wieschaus, E. (1991) Cell 64,447-458 Powers, S., Michaelis, S., Broek, D., Santa Anna, S., Field, J., Her-

skowitz, I., and Wigler, M. (1986) Cell 47,413-422 Provost, N. M., Somers, D. E., and Hurley, J. B. (1988) J. Biol. Chem.

263,12070-12076 Quan. F., Wolfgang, W. J., and Forte. M. A. (1989) Proc. Natl. Acad.

1040-1043

206

- Sci.' U.'S. A. 86,4321-4325 Robishaw, J. D., Kalman, V. K., Moomaw, C. R., and Slaughter, C.

Sanford. J.. Codina. J.. and Birnbaumer. L. (1991) J. Biol. Chem. A. (1989) J. Biol. Chem. 2 6 4 , 15758-15761

266,9570-9579 , , , . ,

Saneer. F.. Nicklen. S.. and Coulson. A. R. (1977) Proc. Natl. Acad. Sii. U. S. AI ~74,5463-5467

. .

Shaw, G., and Kamen, R. (1986) Cell 46,659-667 Schupbach, G. M., and Wieschaus, E. (1989) Genetics 121,101-117 Simon. M. I.. Strathmann. M. P.. and Gautam. N. (1991) Science , . .

252; 802-808

Deu. 2,1296-1306

(1989) J. Biol. Chem. 264,18552-18560

Smith, R. C., Dworkin, M. B., and Dworkin-Rastl, E. (1988) Genes

Thambi, N. C., Quan, F., Wolfgang, W. J., Spiegel, A., and Forte, M.

Walker, G., and Bourguignon, L. Y. (1990) FASEB J. 4,2924-2933 Wang, N., Yan, K., and Rasenick, M. M. (1990) J. Biol. Chem. 265,

Yamada, T., Takeuchi, Y., Komori, N., Kobayashi, H., Sakai, Y., Hotta, Y., and Matsumoto, H. (1990) Science 248,483-486

Yarfitz, S. Provost, N. M., and Hurley, J. B. (1988) Proc. Natl. Acad. Sci. U. S. A. 85,7134-7138

Yatani, A., Imoto, Y., Codina, J., Hamilton, S. L., Brown, A. M., and Birnbaumer, L. (1988) J. Biol. Chem. 263,9887-9895

Yatani, A., Okabe, K., Codina, J., Birnbaumer, L., and Brown, A. M. (1990) Science 249, 1163-1168

Yoon, J. S., Shortridge, R. D., Bloomquist, B. T., Schneuwly, S., Perdew, M. H., and Pak, W. L. (1989) J. Biol. Chem. 264,18536- 18543

Zipursky, S. L., Venkatesh, T. R., and Benzer, S. (1985) Proc. Natl. Acad. Sei. U. S. A. 82, 1855-1859

1239-1242

the drosophila g protein y subunit gene (d-gyi) produces three

Documents