- Email: [email protected]
Alternative splicing of the Drosophila melanogaster rotundRacGAP gene
Gene, 168 (1996) 135-141 ©1996 Elsevier Science B.V. All fights reserved. 0378-1119/96/$15.00
135
G E N E 09450
Alternative splicing of the Drosophila melanogaster rotundRacGAP gene (GTPase-activating protein; genomic organization; RT-PCR; RACE; tissue specificity; nucleic acid sequence)
C a r o l i n e D. H o e m a n n * ' * * , E v e l y n e Bergeret*, A n n a b e l G u i c h a r d a n d R u t h G r i f f i n - S h e a CEA--Laboratoire de Biologic MolOculaire du Cycle Cellulaire, INSERM Unit~ 309, Dkpartement de Biologic Molkeulaire et Structurale, Centre d'Etudes NuclOaires, Grenoble, 38054 Grenoble Cedex 9, France
Received by J.K.C. Knowles: 27 March 1995; Revised/Accepted: 3 August/22 August 1995; Received at publishers: 16 October 1995
SUMMARY
The rotund (rn) gene in Drosophila melanogaster codes for a RacGTPase-activating protein, RnRacGAP. Cellular studies have shown that RacGAP proteins function as negative regulators of substrate Rac proteins which, in turn, control the localization and polymerization state of actin within the cell. Previous sequence analysis of rn genomic DNA and incomplete eDNA clones suggested that at least two differentially spliced forms of the transcript exist, rnRacGAP(1) and rnRacGAP(2). Using nested reverse transcription-polymerase chain reaction (RT-PCR) methods, we have cloned missing exon and intron sequences, and detected differences between rnRacGAP(1) and rnRacGAP(2) involving 24 nucleotides (nt) of coding sequence and 119 nt of 3' UTR. This translates to a difference of seven amino acids at the C-termini of the polypeptide products. Utilization, in RT-PCR analysis, of form-specific primers provided a simple assay for the tissue specificity of expression of the two forms, rnRacGAP(1) is the predominant species in the testes and is expressed at a low level in the ovary and somatic tissues, rnRacGAP(2) is only very weakly expressed and is detectable solely in the testes.
INTRODUCTION
The small G proteins of the Rac family cycle between an active GTP-bound, and an inactive GDP-bound, state to control actin cytoskeletal organization and membrane Correspondence to: Dr. R. Griffin-Shea, INSERM Unit6 309, BMCC/DBMS, CEN-G, Grenoble 38054 Cedex 9, France. Tel. (33-76) 883-090; Fax (33-76) 885-100; e-mail: [email protected] * These authors have contributed equally to this work. ** Present address: Laboratory of Molecular Biology, Institut de Recherches Cliniques de Montreal, 110 Pine Avenue West, Montreal, Quebec H2W IR7, Canada. Tel. (1-514) 987-5570.
Abbreviations: aa, amino acid(s); bp, base pair(s); CAT, chloramphenicol acetyltransferase; eDNA, DNA complementary to RNA; DEPC, diethyl pyrocarbonate; GAP, GTPase-activating protein;kb, kilobases(s) or 1000 bp; nt, nucleotide(s), oligo, oligodeoxyribonucleotide; PCR, polymerase chain reaction; r-, ribosomal; Rac, Ras-related C3 botulinum toxin substrate; RACE, rapid amplification of eDNA ends; rn, rotund gene; Rn, Rotund protein; RT, reverse transcription; SSC, 0.15M NaC1/0.015 M Na3-citrate pH7.6; tsp, transcription start point(s); UTR, untranslated region(s). SSD! 0378-1119(95)00747-4
movement in cells (Ridley et al., 1992; Hall, 1994). GTPase-activating proteins specific for the Rac subfamily (RacGAPs) act as downregulators of Rac activity by specifically catalyzing the intrinsic GTPase activity of Rac proteins thereby returning them to the inactive GDP-bound state (Lamarche and Hall, 1994). In Drosophila, a single RacGAP protein, RnRacGAP, has been characterized (Guichard et al., 1995). RnRacGAP is encoded by one of two transcripts of the rotund (rn) locus in region 84D3,4 (Agnel et al., 1989). Genetic analyses of the locus have defined the rn null mutant phenotype as male sterility and an absence of structures in the sub-distal parts of the appendages (Kerridge and Thomas-Cavallin, 1988); mutant analyses suggest that the second transcript, m5.3, may be responsible for the appendage morphogenetic function (Agnel et al., 1992). The specific function of the rnRacGAP transcript remains to be determined. In situ hybridization experiments showed that rnRacGAP is expressed in both pupal and adult testes, specifically in the primary sperma-
136 (a) ~1071
]i
aaacccacaccctcaaatgctcacgcacacatggcacaccggcgLcaatqcatcLggtac -1011 gcctaaaatacgactacaaaaaagcataaaaatccaattcccgcctLgCtcgLtaatttg -951 cgccgctgctcttcaattcacagccaaaactgaaaagtgccatgataattttgt[tttac -891 tctatccaaacttaaaatcgcataaaaatcaatqccaatgatgccattaaatcaacagac -831 agatccatgtttcgttttccacaaaagttccaagctcgaagtcctttaaattctaatttt -771 tgtqtttttttagcaqcacqtttactcaacattttcgctttctgacgttcccgcctcaca -711 tgaactgcaqcattttttcgLgctacatttttgctggttgcattctaLg%ctggtggccg ~Sl
caaaac~ttttccatgaatttttcatccatttagcggcgccattgagccattttctggg
lcgcttaq~pattc~tagatactttageaaattgtttttttagttcqattttctacattct ttttgttttta atttttagatttigtatcgtgacattgaatctcggtgtgattgatt gtctttaac~tcgaaaatataaaaagatcaacactgacttgtttatatg aagaagacagggcaaatatctgttattgtt~ggcacaaaaaatgtttdtaatadtttgat tcactttttttttqtattgaaaacagcatttttgatatttcttegtaacaccttagtcgc ccattaaatttctgcggcqccctaqatqtqtqtatqaattatccaqaaatatttacaaat ,qtagqacccacaqtaqq~TCTTTATGGGTGATCGAATTTTTGT_~CTGC~AGAACTATT TAAGCTATCAGCTTTCACCCGATrCAAATAAACCGTAAGAGTAAATAAAC~TTTCGGTCG AGTATTCGGAGAAGCGAGCCGATAACTTCTGCGTACGTGACTCA~C6GA'Ad'Td'G~TCT
-591 -531 -471 -411 -351 -291 -231 -171 111 -51 10 70 130 190 250 310 370 430 490 550 610 670 730 79O 850 910 970
tocytes, in addition to a weak accumulation in the imagihal discs, larval structures which develop into the adult appendages (Agnel et al., 1989; 1992). In this paper, we establish that there is alternative splicing of the rnRacGAP primary transcript; we define the complete sequences of the rnRacGAP(1) and rnRacGAP(2) transcript forms and characterize the pattern of their specific expression in germline and somatic tissues.
CTTCGGAA~TT~CC~ACZ~ACCAGAZ6_m?.~T~A~CAAr~CCCA~6~¥AGTG
GCTCTGGGAGTCGGACGCCCTCAAATCG~CTTTATTTGTCGCCCGTTCGCCCCACGATGC AGAACAAACGGAGGCTTCTGCGGGAGTACAGGTCCTACGATGACCTCAGCGAGCACTACC GCATGTTTGGATCGCAGTCCCTGGACTC~ATCGCGTGGACATGAATCCCAGTG GCTGCGATGGCTTATCCACG-GACGGACTCGACTTCTGCTCGCAGTCCCACTCGGGGCTGC TGAGGGAGCACAATTTCAAGATTAAGTCCTACTACTACAACGTGGGGAACTGCGTACACT
GCCGCAAAA(~tgagttgecatcaccgaatggtggaacttctgagaggatttc~aattc~ gggaaaacatgaaaagttaaataacatcccaaccctctct~accaacatttacctcecct ,aaCCaCqtOtccctqqtcttct%cca~GATCCGCTTCGCCATGGCCAGCCTGAGGTGTCG GGCGTGTCCGCTCCGCTGCCACATCGGTTGCTGCCGCCAGCTGACGGTGAACTGCATCCC CCAGCCTCAAATCGGTACGAAGAGAGGATGTCTCAGCGATTATGCGCCTCGGGTGGCGCC CAT~GTGCCGGCC~TAATTGTCCATT(]TGTAACdf;AGATC(~AGGC(;CCGGGATTGCAGCA GGAGGGACTCTACCGGGTGTCCTCCACCCGAGACAAATGCAAACCACTGCGTCGGAAGC7 GCTGCGTGGCAAGTCCACGCCGCATTTGGGCAATAAAGACACCCACACACTGTGCTGTTG TGTTAAGGACTTTCTTCGGCACTTGGTTCACCCTCTGATTCCCATTTACCACCGAA~GGA TTTCGAGGAGGCCACGCGGCACGAAGATCGTCTGGCTGTGGAGATGGCGGTATATTTAGC GGTTCTGGACCTCCATCAGGCTCACAGGGATACGTTGGCCTACTTAATGCTTCACTGGCA GAAAATTGCCCAAAGTCCTGCTGTCCGCATGACGGTCAACAATCTTGCCGTGCT~ TC--'6~ACTCT.C~TTCGGAGATCTGGATCTAACCCTCGAAAATGTCG~'CACTTGGCAGCGGGT GCTGAAGGTGCTACTTCTGATGCCTGCCGGATTTTGGTCTCAGTTCTTGGAGGTCCATCC CCTTCCCACTTCGTTGGGCAGCACCTATGACTTTGAGGACCGGTACAATCACCGTCACTG GGACAGCTCGTCCAATCTGGGATGGTCTTCTGTTAAGACCTACT T T A G A T C A A T ~ a g
1030 1090 1150 1210 1270 acctatggtttccctt~--a~{~g-g~'a'g-a'~'t't'~atattcgaataagccaaac "ca 1330 at @,caaqcaatcaaacat~ cat aca aa Let tt cact qaat act t cctqt tt ca 1390 CCTTAGCAGTACTCACCT~A~GA~AGTCTTATSACTGAGACArCGAGATGGATGCAGT 1450 "CA~'C~'T'~TCA'~T~'ATTTTG TAT TT T2'TTTC AT T AT TC AC TAC AG T TGAAG A A ~ ~ ; I 1510 AATAATTGGATCTCTAA tgt tgtqt a ~ t ta~.gcqgct aacaaq~g 1570 ta 1572
(b) (1266) g t a
aga o c t
atg gtt
tcc
ctt[~]
(1289)
(377) V a l A r g P r o Met V a i
S e r Leu * * *
(1386) g t c
act cac cta ~
aac e t t
agc a g t
(377) V a l A s n L e u Ser Ser Thr His L e u * * * (C)
(-1130)
370
575
(-583)
Form2
(383) (1415) Form 1 ***
(384)
(364)
Genomic DNA
137
Form 2 Exo l
v
E
2
m2
(1386)
Fig. 1. The rnRacGAP gene. (a) Genomic nt sequence. Positive numbering starts from the A pf the ATG start codon, which is encircled. The first EcoRI fragment and Exons 1, 2 and 3 are outlined. Exons are written in upper case, flanking sequences and introns are in lower case. Restriction sites are surrounded by boxes with outlines that are solid, EcoRI; dotted, BglI; dashed, PvuII; or notched, PstI. The tsp starts at the boxed A at nt -212. The start of Form-1 cDNA is indicated by an open circle at nt -120. PCR primer regions are underlined with an arrowhead indicating the polarity, except for as2. i, overlined for clarity. For the nested primers, the overlapping nt are enclosed on three sides. The Antp binding site at nt - 4 6 0 is within an oval outlin£ The ATG start codon is encircled. Stop codons are bracketed from above: for Form 1 at nt 1287 and for Form 2 at nt 1410. The polyadenylation addition site at nt 1348 in intron 2 is indicated by two upper case A's. (b) Deduced aa sequences of the nt sequences differing between Forms
RESULTS AND DISCUSSION
(a) Identification of alternatively spliced products of rnRacGAP by 3' RACE Sequence analyses of two incomplete, but overlapping, cDNA clones coding for RnRacGAP (pcl.7 and pcl.7d; Agnel et al., 1992), indicated the presence of differing 3' ends, suggesting that the primary transcript was being
1 and 2. The numbers in parentheses indicate the position of the nt as in panel a or for the aa as in Agnel et al. (1992). The stop codons are overbracketed. (c) Genomic organization of the rnRacGAP gene. The positions of the EcoRI sites, and asl and as2 primers are indicated by numbers in parentheses, and all other primers, within their symbols. The numbering is from the first 5' nt in the sequence as given in panel a. Top line: previously unpublished sequence data are represented by heavy lines and include: 5' flanking sequence (548 nt); intron I (137 nt); intron 2 (120 nt); Exon 3 (143 nt with 121 nt of new sequence); and 3' flanking sequence (44 nt). EcoRl (circles), BglI (diamond), PvulI (triangle) and PstI (square) restriction sites are indicated. Middle line: Exon-intron organization of rnRacGAP(2). The predicted exons are indicated by boxes, with coding regions shaded. Sense primers are encircled; antisense primers are indicated as clear (common coding sequence primers) or hatched (exon-specific primers) boxes bearing shortened versions of primer names. Introns are drawn as inverted triangles. Bottom line: Exon-intron organization of rnRacGAP(1). Symbols as for the Form 2. The large arrow indicates the 92-nt distance to the start of the original pcl.7 cDNA clone. Methods: Mini-preparation of plasmid DNA by the alkaline lysis method, subcloning procedures and sequencing by the Sanger dideoxy method are described in Sambrook et al. (1989). For RACE, cDNA was made from wild-type red e H female adult poly(A)+RNA using the Amersham cDNA synthesis kit and primer T17-Xho to prime cDNA synthesis. First-round RACE was performed using a primer, GAP1, located in the GAP-homology domain; second-round RACE with with the nested sense primer GAP2. Reaction conditions: 100 gl reaction volume containing Vent buffer supplied by the manufacturer (NE BioLabs, Beverly, MA, USA) with l gg BSA 500 ng each primer 1 unit Vent polymerase. 1 Ixl of first round RACE products were used for second round. Exon 3 product was identified using Southern and restriction analysis and subcloned by blunt-end ligation into Bluescript (Stratagene, La Jolla, CA, USA) for sequence analysis. The nt sequence data reported in this paper will appear in the EMBL, GenBank and DDBJ Nucleotide Sequence Databases under accession No. U22539. Primers: asl, 5'-GGTGAGTACTGCTAAGGTTGAC; as2, 5'-GGGAAACCATAGGTCTTACC; GAP1, 5'- CTTGCCGTGATCTTCGCTCC; GAP2, 5'-TCTTCGCTCCAACTCTGTT; Tl7-Xho, 5'-GACTCGAGTCGACATCGA(T)17.
137 alternatively spliced. The C-terminal end of the pcl.7 clone encoded 25 nt of an incomplete third exon. We isolated this exon in its entirety by the 3' RACE method (O'Hara et al., 1989) from female adult poly(A)+RNA. RACE reactions were performed with an oligo(dT) polymer containing 3' anchor primer and nested primers located in the GAP-homology domain, and GAP 2 (Fig. 1). Sequence analysis revealed that exon 3 contained 143 nt: 24 nt corresponding to coding sequence followed by two consecutive in-frame stop codons. The complete coding sequence defined by exons 1, 2 and 3 is now referred to as rnRacGAP(1), or Form 1. We also established the exact sizes of intron 2 (120bp), intron 1 (137 bp) exon 2 (808 bp) and exon 3 (143 bp), as well as portions of the 5' (548 bp) and 3' (44 bp) flanking regions. The genomic sequence containing the rnRacGAP gene is given in Fig. la and schematicized in Fig. lc At the polypeptide level, the 3' divergence in the nt sequences of rnRacGAP(1) and the previously characterized eDNA clone (Agnel et al., 1992) coding for rnRacGAP(2) or Form 2, translated to differences of only seven, or six, amino acids, respectively, at the C-termini of the protein products (Fig. lb). No specific subcellular localization signals were found within the differing C-termini. (h) Characterization of rnRacGAP 5' terminus In order to ensure that the differences in rnRacGAP(1) and rnRacGAP(2) were limited to the 3' ends, we undertook a series of in vivo and in vitro experiments to characterize the 5' limit of transcription.
(1) In vivo analysis Initially, we mapped the size of the rnRacGAP 5' terminus of Form 1 by RNAse H cleavage of a hybrid formed between adult male and female poly(A)+RNA and an antisense primer (prPst) 268 bp downstream from the tsp of clone pcl.7. Northern blot analysis followed by hybridization with a 5'-specific probe shows a single band in adult male mRNA of about 340 bp (Fig. 2a; lane 2). No signal was detectable in female flies (lane 1); non-specific degradation was ruled out by a control RT-PCR (bottom) with r-gene rp49 (O'Connell and Rosbash, 1984). These results define approx. 80 bp of additional upstream sequence. To more accurately determine the position of the tsp, we utilized the primer-extension method with a primer, prPvuII, located 150 nt upstream from the prPst primer and adult fly poly(A)+RNA. As shown in Fig. 2b, a fragment was detected corresponding to the size predicted by the RNase H experiment, defining a 212-nt 5' UTR. Finally, we verified that rnRacGAP(I) and rnRacGAP(2) differed only at their 3' ends by RT-PCR
analysis of adult male poly(A)+RNA with a sense primer (prexls5') 10 bp downstream from the tsp, in combination with either the Form-l-specific asl primer, or nested Form-2-specific primers (as2 and as2.1). We obtained products (Fig. 2c) of the expected sizes (thicker arrows) for both Forms 1 (lane 3) and Form 2 (lane 1); the highermolecular-weight bands correspond to genomic DNA (thinner arrows; lanes 4 and 2, respectively). Comparative restriction analyses (data not shown) confirmed that Forms 1 and 2 were colinear. (2) In vitro transcription To verify the functionality of the putative rnRacGAP promoter, a heterologous promoter construct was made containing a genomic fragment that fused 1700 bp of upstream sequence and the first 154 bp (site of BglI cleavage) of the 212 bp 5' UTR, with the bacterial cat gene. This genomic rnRacGAP DNA was cloned in both sense and antisense orientations with respect to the cat reporter. Supercoiled plasmid DNA was incubated with Drosophila embryo nuclear extract for 30 rain at 37°C, and the resulting in vitro transcripts were analyzed by primer extension using a cat-specific primer according to Kadonga (1990). As a positive control, a promoter construct encoding the Xenopus histone H1 ° promoter fused to cat (Khochbin and Wolffe, 1993) was processed in parallel with the rnRacGAP-cat constructs. A primerextension product corresponding to the predicted tsp was observed (Fig. 3, arrow, lane 2), confirming that the rnRacGAP region contains a promoter. The absence of primer-extension products from the rnRacGAP-cat antisense construct (lane A) demonstrates the specificity of the in vitro sense transcription products. These combined results, schematically represented in Fig. 2d, point to an authentic rnRacGAP tsp site as occurring 212bp upstream from the ATG start codon. Structural features surrounding this tsp are consistent with previously identified tsp. The tsp itself (GCA(+ 1)CTCT) conforms to a mammalian TFIID binding site (Y, Y, A(+ 1), N, W, Y, Y; Javahery et al., 1994) and is nearly identical with a canonical cap site identified for Drosophila transcripts (D, Y, A(+ 1), K, T, G; Purnell et al., 1994) (where D--A or G or T, K = G or T, W = A or T and Y = C or T). A Hogness box motif similar to that described for the dsx gene (Fig. 1, TACAAATGTA; Burtis et al., 1989) is located 24 bp upstream from the mapped tsp (Fig. 1, at -237). A TATA-box element was found - 2 3 0 bp upstream (Fig. 1, nt - 4 4 2 ) from the mapped tsp but no G+C-rich, nor CAAT box, sequence was found. At -251 bp upstream from the tsp (Fig. 1, nt -460), a consensus sequence was identified for the Antennapedia homeotic protein-binding site (ATTTAATTGA; Hanes and Brent, 1991).
138 (a)
(c)
(b) 1
2
(d)
bp 404
Mr
1
2
3
4
--
kb
in vivo
7.6 ~> 5.3 ~> 2.8 D,
308 --
[] bp
242-238 --
2000
I --
RNase H fragment -~
217 --
1.9 ~> 1.6 ~>
1600--
201--
5' [fiR 212
EXON1 | ~147~ i AUG1 1 "~" 3 4 0
190 -180 --
1.0 D,
160 --
1000--
/
147 --
0.6
D"
0.4
D-
Primer extension fragment in vitro
122
--
Ii0
--
0.3 {> rp49
?
500
rn [ ~ I
cat
J
m
s48 //
370
~
,
[:2>
[i~i ~ i72~
>
in vitro transcription fragment 4
154
!99
[
90--
76-67-Fig. 2. Analysis of the rnRacGAP transcript 5' terminus. (a) Northern blot analysis of female (lane 1) and male (lane 2) poly(A) + RNA after incubation with a 5' primer located at +268 nt from the 5' start site of the published cDNA clone pcl.7 (Agnel et al., 1992) and RNAse H treatment. The filter was stripped and re-hybridized with a random-prime labeled PCR fragment of the rp49 gene (lower panel). Primers: prPst, 5'-ATCCTGCAGCGAGTCCAGGG; RP49-antisense, 5'-GTGTATTCCGACCACGTTACA; RP49-sense, 5'-TCCTACCAGCTTCAAGATGAC. (b) Primer extension product from adult fly poly(A)+RNA. Primer: prPvulI, 5'-GTTGGCGAATTGATTCAGCTG. (c) RT-PCR amplification of rnRacGAP(1) and rnRacGAP(2). The RT-PCR reaction was carried out with a sense primer + 10 bp from the mapped tsp site (exls5') and form-specific primers. Lanes 1 and 2: Form-2-specific PCR products; lanes 3 and 4: Form-l-specific PCR products. The templates were oligo(dT)-primed cDNA generated from poly(A)+RNA from adult males in lanes 1 and 3, and cloned genomic DNA in lanes 2 and 4. The thicker arrows mark the position of the form-specific spliced products; the thinner arrows, the positions of amplified genomic DNA. Equal amounts of PCR product were separated on 1.0% agarose gels in the presence of ethidium bromide. Molecular weight marker is the Boehringer 1-kb ladder. The image was produced with an Is-700 Camera, using Is-1000 Digital Imaging System (Alpha lnnotech, San Leandro, CA, USA). primers: exls5', 5'-GGGTGATCGAATTTTTGTGC; as2.1, 5'-GAAATTCTATCCAAATCCG. (d) Schematic representation of combined results in vivo and in vitro analysis of the 5' ends of Forms 1 and 2. Symbols as in Fig. 1; dark square is HindlII site. Methods: For the RNase H experiment, RNA was isolated according to Chomczynski and Sacchi (1987) or using RNAzol B (Bioprobe Systems, Montreuil-Sous-Bios, France). Poly(A) + RNA was isolated using oligo(dT) cellulose columns (Sigma). RNA pellets containing approx. 10 ixg poly(A)+RNA were resuspended in 50 gl DEPC-treated water containing RNAsin (Promega, Madison, WI, USA) and 1 gg of a 21-nt primer prPst. RNA and oligo were heated at 65°C for 5 min, then placed on ice. RNase H treatment was performed according to Ausubel (1991) with an incubation at 38°C for 20 rain at 1 unit RNase H. RNA was separated on a 2.2 M formaldehyde-l.8% agarose gel, transferred to a nylon filter and hybridized with a random-prime labeled probe (Feinberg and Vogelstein, 1983) from the first 300 bp of the rnRacGAP gene (upper panel). For the primer extension reaction: 5' end labeling of the PvulI primer (100 ng) and the primer extension reaction from adult fly poly(A)+RNA were performed as in Khochbin and Wolffe (1993). For the PCR reactions: The cDNA template was made from 1 gg poly(A)+RNA and oligo(dT) (Amersham cDNA synthesis kit). PCR reaction conditions are as follows: 50 gl total volume with undiluted Cetus (Perkin-Elmer) Taq buffer/200 mM dNTP/20 pmol of each primer was heated with the cDNA template for 5 rain at 94°C under paraffin oil. 1 unit of Taq polymerase was added and the reaction was cycled 30 times at 94°C for 1 min, 55°C for 2 rain and 72°C for 3 min.
(c) Tissue-specific expression of rnRacGAP(1) and
produced
rnRacGAP(2)
signal from adult females in the RNase H experiment, we
i n t h e testes, a n d i n l i g h t o f t h e a b s e n c e o f
had
t e s t e d R N A f r o m m a l e a n d f e m a l e flies, as well as dis-
shown that the majority of rnRacGAP transcripts are
sected ovaries, testes and the remaining carcasses, for the
As p r e v i o u s
in situ hybridization
experiments
139
1A2
3456
7
A
tlt m
i¸i;!ii !!ii
i !iii
!ii~~i~i?i!~~i~!~i!~illii~~
presence of both rnRacGAP(1) and rnRacGAP(2). Form-specific amplification was achieved with the antisense primers, asl and as2, which are equidistant (232 bp) from the sense primer, GAP1 (Fig. 1). As shown in Fig. 4A, rnRacGAP(1) produces signals from DNA and RNA that can be directly distinguished by their migration position, while rnRacGAP(2), colinear with the genomic DNA in this region, produces PCR products of the same size from both RNA and potentially-contaminating DNA. To remove DNA-associated signals we treated aliquots of all RNA preparations with DNase followed by phenol extraction prior to RT-PCR reactions (Fig. 4B). As a control for the extent of DNase action, excess rn genomic DNA was added to an aliquot of the male carcass RNA sample, and the mixture then treated in parallel with experimental samples; the DNA-related products from Form-l-specific (lane 13) and Form-2-specific (lane 15) amplification are completely removed by DNase treatment (lanes 14 and 16, respectively). Thus, the residual signals present in lanes 1-12 are RNA-specific and show that the DNase treatment removes all rnRacGAP(2)-generated signals (lanes 2, 6, 8, 10 and 12) except for a relatively weak signal still present in the testes (lane 4). The highly predominant form in the testes is rnRacGAP(1) (lane 3). This expression appears to account for the intense signal in whole male flies (lane 11), as the male carcass sample shows considerably smaller amounts of rnRacGAP(1) (lane 7). Low levels of Form-1 expression are also detected in female carcasses (lane 5) and ovaries (lanes 1). While transcript is not detectable here in whole female flies (lanes 9-10), a 3-times longer exposure renders rnRacGAP(1) detectable (data not shown). These results suggest that rnRacGAP(1) expression
cat primer, 5 ' - G G T A T A T C C A G T G A T T T T T T T C T C C A T . Methods: A HindlII fragment from the rn locus was amplified with PCR primers
Fig. 3. In vitro experiments to determine the 5' tsp. Autoradiography of primer extension products of in vitro transcripts generated from heterologous rnRacGAP-cat fusion genes and a control histone HI"cat gene: histone H1°-cat (lane 1), antisense rnRacGAP-cat (lane A) and sense rnRacGAP-cat (lane 2; arrow). Lanes 3-6, rnRacGAP-cat sequencing reaction. Lane 7, end-labeled pBR322-MspI molecular weight marker, prextBgll primer, 5'-AGGCCACTTCGGCTTGAGTC;
prextBglI and the T7 primer (Boehringer) to generate a blunt-end fragment containing 1700-bp 5' flanking sequence and 154-bp 5 ' U T R of the rnRacGAP gene. This fragment was ligated into the BasicCAT (Prornega) vector Klenow-filled-in HindlII cloning site in both sense and antisense orientation. Sequencing of the promoter region showed the absence of mutations from vent amplification. A control plasmid carried the histone H1 ~" promoter (Khochbin and Wolffe, 1993). Plasmids were purified twice over CsCI gradients to recover supercoiled DNA which was used to generate in vitro transcripts with the Prornega Drosophila Embryo Nuclear Extract Transcription System and accompanying protocol. Primer extension of the resulting transcripts was performed as in Khochbin and Wolffe (1993). The ethanol-precipitated DNA was resuspended in 3 gl loading dye mix (2 lal loading dye from US Biochemical (Cleveland, OH, USA) sequencing kit and 1 p.l 0.1 M NaOH), heated for 5 min at 100°C and loaded onto a 7 M urea-6% polyacrylamide sequencing gel along with a sequencing reaction performed with the US Biochemical sequencing kit, using the rnRacGAPcat sense construct as template, unlabeled dNTPs and the same endlabeled cat primer used in the primer-extension reaction.
140 1
2
3
4
5
15
7
8
9
I0
II
12
13
14
15 15
I
J Fig. 4. Tissue-specific expression of rnRacGAP(1) and rnRacGAP(2). Total RNA was isolated from the indicated tissues, subjected to RT-PCR and subsequent amplification and tested by Southern blot analyses: 80 ovaries, lanes 1 and 2; 60 testes, lanes 3 and 4; 10 female carcasses, lanes 5 and 6; 10 male carcasses, lanes 7 and 8; 10 adult females, lanes 9 and 10; 10 adult males, lanes 11 and 12. Products in the odd-numbered lanes from 1 12 were generated from the Form-lspecific primer pair, GAP 1-as l; products in even-numbered lanes from 1-12, from the Form-2-specific primer pair, GAPI-as2. (A) Autoradiogram of samples before DNase treatment probed with DNA from the 3' common coding region of the rnRacGAP transcript. Controls for the size of the genomic DNA and the Form-specific RNAs are presented: amplification of genomic DNA, lane 13; amplification of cloned rnRacGAP(1) cDNA, lane 14; and amplification of cloned rnRacGAP(2) cDNA, lane 15. (B) Autoradiogram of samples after DNase treatment to remove DNA-associated signals; probe as in A. Controls for the extent of DNase action are shown in lanes 13 16. Genomic DNA was added to an aliquot of the male carcass RNA preparation to be treated in parallel: lane 13, Form-l-specific amplification, before DNase treatment; lane 14, after DNase treatment; lane 15, Form-2-specific amplification before DNase treatment; lane 16, after DNase treatment. Note complete disappearance of the DNA-generated signal in these two samples indicating completeness of the DNase reaction. Lower panel: Autoradiograms of PCR products produced With primers specific for the ribosomal protein rp49 gene and probed with rp49 DNA as control for sample addition. These samples were not treated with DNase. Methods: For DNase (RNase-free; Stratagene) reaction, 2 gl RNA (0.2 gg/gl) was resuspended in a buffered 20 gl final reaction volume containing 1 Ixl RNasin and 2 gl of DNase (10 units/l~l) and incubated for 30 min at 37°C. The samples were then phenolextracted and ethanol-precipitated prior to PCR reactions. PCR reaction conditions are as described in the legend to Fig. 2. PCR products were analysed on a 2.0% agarose gel.
may be necessary for testes function. In order to test this hypothesis, we are currently introducing rnRacGAP transgenes into flies deficient for the rn locus. Preliminary results suggest that testes-associated function may, indeed, be contained in the Form 1 differing 3' end: Form 2 transcript alone cannot rescue male sterility, while a genomic fragment containing complete RnRacGAP sequence encompassing both 3' ends can restore male fertility (A.G., unpublished data).
(d) Conclusions (I) Cloning and sequencing of the alternative exon present in rnRacGAP(1) confirm that an alternative splice pathway operates at the 3' end of the rnRacGAP primary transcript. In vivo and in vitro experiments to characterize the 5' ends of rnRacGAP(1) and rnRacGAP(2) defined a 212-bp 5' UTR. The remaining sequence between the 5' and 3' ends of rnRacGAP(1) and rnRacGAP(2) was determined to be colinear. Thus, the differences between the two Forms are restricted to the 3' ends and involve 143 nt for Form 1, and 83 nt for Form 2. At the polypeptide level, these differences translate, respectively, to seven, or six, alternative C-terminal aa. (2) Previous in situ hybridization experiments (Agnel et al., 1992) had shown that in adult males, the rnRacGAP sequence was specifically produced in the primary spermatocytes; at this level of sensitivity, no signal could be detected in females. In this study, we show by PCR analysis of male and female flies, and isolated germ tissue and carcasses, that rnRacGAP(1) is the predominant species in the testes and is expressed at a low level in the ovary, and the somatic tissues of both sexes, rnRacGAP(2) is only very weakly expressed and is detectable solely in the testes.
ACKNOWLEDGEMENTS
These experiments were carried out in the laboratory of Dr. J.J. Lawrence, whom we thank for support. Our appreciation to Saadi Khochbin for helpful comments and critical reading of the manuscript. C.D.H. was supported by postdoctoral fellowships from the CEA, Centre d'Etudes Nucl6aires (Grenoble, France) and the International Fogarty-INSERM exchange program (IF06TW01881-01).
REFERENCES Agnel, M., Kerridge, S., Vola, C. and Griffin-Shea, R.: Two transcripts from the rotund region of Drosophila show similar positional specificities in imaginal disc tissues. Genes Dev. 3 (1989) 85-95. Agnel, M., Roder, L., Vola, C. and Griffin-Shea, R.: A Drosophila rotund transcript expressed during spermatogenesis and imaginal disc morphogenesis encodes a protein which is similar to human Rac GTPase-activating (racGAP) proteins, Mol. Cell Biol. 12 (1992) 511l 5122. Ausubel, F.M., Brent, R., Kingston, R.G., Moore, D.D., Smith, J.A., Leidman, J.G. and Struhl, K.: Current Protocols in Molecular Biology. Greene Publishing Associates and Wiley Interscience, New York, NY, 1991. Burtis, K.C. and Baker, B.S.: Drosophila doublesex gene controls somatic sexual differentiation by producing alternatively spliced mRNAs encoding related sex-specific polypeptides. Cell 56 (1989) 997 1010.
141 Chomczynski, P. and Sacchi, N.: Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chlorofrm extraction. Anal. Biochem. (1987) 156-159. Feinberg, A.P. and Vogelstein, B.: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132 (1983) 6-13. Guichard, A., Bergeret, E. and Griffin-Shea, R.: RotundRacGAP. In: Zerial, M. and Huber, L.A. (Eds.), Guidebook to the Small GTPases. Oxford University Press, Oxford, 1995, in press. Hall, A.: Small GTP-binding proteins and the regulation of the actin cytoskeleton. Annu. Rev. Cell Biol. 10 (1994) 31 54. Hanes, S.D. and Brent, R.: A genetic model for interaction of the homeodomain recognition helix with DNA. Science 251 (1991) 426-430. Javahery, R., Khachi, A., Lo, K., Zenzie-Gregory, B. and Smale, S.T.: DNA sequence requirements for transcriptional initiator activity i~ mammalian cells. Mol. Cell. Biol. 14 (1994) 116-127. Kadonaga, J.T.: Assembly and disassembly of the Drosophila RNA polymerase II complex during transcription. J. Biol. Chem. 265 (1990) 2624 2631. Kerridge, S. and Thomas-Cavallin, M.: Appendage morphogenesis in Drosophila: a developmental study of the rotund (rn) gene. Wilhelm Roux's Arch. Dev. Biol. 197 (1988) 19-26.
Khochbin, S. and Wolffe, A.P.: Developmental regulation and butyrateinducible transcription of the Xenopus histone HI ° promoter. Gene 128 (1993) 173-180. Lamarche, N. and Hall, A.: GAPs for rho-related GTPases. Trends Genet. 10 (1994) 436-440. O'Connell, P.O. and Rosbash, M.: Sequence, structure, and codon preference of the Drosophila ribosomal protein 49 gene. Nucleic Acids Res. 12 (1984) 5495-5513. O'Hara, O., Dorit, R.L. and Gilbert, W.: One-sided polymerase chain reaction: the amplification of cDNA. Proc. Natl. Acad. Sci. USA 86 (1989) 5673 5677. Purnell, B.A., Emanuel, P.A. and Gilmour, D.S.: TFIID sequence recognition of the initiator and sequences farther downstream in Drosophila class II genes. Genes Dev. 8 (1994) 830-842. Ridley, A.J., Paterson, H.F., Johnston, C.L., Diekmann, D. and Hall, A.: The small GTP-binding protein rac regulates growth factorinduced membrane ruffling. Cell 70 (1992) 401 410. Sambrook, J., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989.
135
G E N E 09450
Alternative splicing of the Drosophila melanogaster rotundRacGAP gene (GTPase-activating protein; genomic organization; RT-PCR; RACE; tissue specificity; nucleic acid sequence)
C a r o l i n e D. H o e m a n n * ' * * , E v e l y n e Bergeret*, A n n a b e l G u i c h a r d a n d R u t h G r i f f i n - S h e a CEA--Laboratoire de Biologic MolOculaire du Cycle Cellulaire, INSERM Unit~ 309, Dkpartement de Biologic Molkeulaire et Structurale, Centre d'Etudes NuclOaires, Grenoble, 38054 Grenoble Cedex 9, France
Received by J.K.C. Knowles: 27 March 1995; Revised/Accepted: 3 August/22 August 1995; Received at publishers: 16 October 1995
SUMMARY
The rotund (rn) gene in Drosophila melanogaster codes for a RacGTPase-activating protein, RnRacGAP. Cellular studies have shown that RacGAP proteins function as negative regulators of substrate Rac proteins which, in turn, control the localization and polymerization state of actin within the cell. Previous sequence analysis of rn genomic DNA and incomplete eDNA clones suggested that at least two differentially spliced forms of the transcript exist, rnRacGAP(1) and rnRacGAP(2). Using nested reverse transcription-polymerase chain reaction (RT-PCR) methods, we have cloned missing exon and intron sequences, and detected differences between rnRacGAP(1) and rnRacGAP(2) involving 24 nucleotides (nt) of coding sequence and 119 nt of 3' UTR. This translates to a difference of seven amino acids at the C-termini of the polypeptide products. Utilization, in RT-PCR analysis, of form-specific primers provided a simple assay for the tissue specificity of expression of the two forms, rnRacGAP(1) is the predominant species in the testes and is expressed at a low level in the ovary and somatic tissues, rnRacGAP(2) is only very weakly expressed and is detectable solely in the testes.
INTRODUCTION
The small G proteins of the Rac family cycle between an active GTP-bound, and an inactive GDP-bound, state to control actin cytoskeletal organization and membrane Correspondence to: Dr. R. Griffin-Shea, INSERM Unit6 309, BMCC/DBMS, CEN-G, Grenoble 38054 Cedex 9, France. Tel. (33-76) 883-090; Fax (33-76) 885-100; e-mail: [email protected] * These authors have contributed equally to this work. ** Present address: Laboratory of Molecular Biology, Institut de Recherches Cliniques de Montreal, 110 Pine Avenue West, Montreal, Quebec H2W IR7, Canada. Tel. (1-514) 987-5570.
Abbreviations: aa, amino acid(s); bp, base pair(s); CAT, chloramphenicol acetyltransferase; eDNA, DNA complementary to RNA; DEPC, diethyl pyrocarbonate; GAP, GTPase-activating protein;kb, kilobases(s) or 1000 bp; nt, nucleotide(s), oligo, oligodeoxyribonucleotide; PCR, polymerase chain reaction; r-, ribosomal; Rac, Ras-related C3 botulinum toxin substrate; RACE, rapid amplification of eDNA ends; rn, rotund gene; Rn, Rotund protein; RT, reverse transcription; SSC, 0.15M NaC1/0.015 M Na3-citrate pH7.6; tsp, transcription start point(s); UTR, untranslated region(s). SSD! 0378-1119(95)00747-4
movement in cells (Ridley et al., 1992; Hall, 1994). GTPase-activating proteins specific for the Rac subfamily (RacGAPs) act as downregulators of Rac activity by specifically catalyzing the intrinsic GTPase activity of Rac proteins thereby returning them to the inactive GDP-bound state (Lamarche and Hall, 1994). In Drosophila, a single RacGAP protein, RnRacGAP, has been characterized (Guichard et al., 1995). RnRacGAP is encoded by one of two transcripts of the rotund (rn) locus in region 84D3,4 (Agnel et al., 1989). Genetic analyses of the locus have defined the rn null mutant phenotype as male sterility and an absence of structures in the sub-distal parts of the appendages (Kerridge and Thomas-Cavallin, 1988); mutant analyses suggest that the second transcript, m5.3, may be responsible for the appendage morphogenetic function (Agnel et al., 1992). The specific function of the rnRacGAP transcript remains to be determined. In situ hybridization experiments showed that rnRacGAP is expressed in both pupal and adult testes, specifically in the primary sperma-
136 (a) ~1071
]i
aaacccacaccctcaaatgctcacgcacacatggcacaccggcgLcaatqcatcLggtac -1011 gcctaaaatacgactacaaaaaagcataaaaatccaattcccgcctLgCtcgLtaatttg -951 cgccgctgctcttcaattcacagccaaaactgaaaagtgccatgataattttgt[tttac -891 tctatccaaacttaaaatcgcataaaaatcaatqccaatgatgccattaaatcaacagac -831 agatccatgtttcgttttccacaaaagttccaagctcgaagtcctttaaattctaatttt -771 tgtqtttttttagcaqcacqtttactcaacattttcgctttctgacgttcccgcctcaca -711 tgaactgcaqcattttttcgLgctacatttttgctggttgcattctaLg%ctggtggccg ~Sl
caaaac~ttttccatgaatttttcatccatttagcggcgccattgagccattttctggg
lcgcttaq~pattc~tagatactttageaaattgtttttttagttcqattttctacattct ttttgttttta atttttagatttigtatcgtgacattgaatctcggtgtgattgatt gtctttaac~tcgaaaatataaaaagatcaacactgacttgtttatatg aagaagacagggcaaatatctgttattgtt~ggcacaaaaaatgtttdtaatadtttgat tcactttttttttqtattgaaaacagcatttttgatatttcttegtaacaccttagtcgc ccattaaatttctgcggcqccctaqatqtqtqtatqaattatccaqaaatatttacaaat ,qtagqacccacaqtaqq~TCTTTATGGGTGATCGAATTTTTGT_~CTGC~AGAACTATT TAAGCTATCAGCTTTCACCCGATrCAAATAAACCGTAAGAGTAAATAAAC~TTTCGGTCG AGTATTCGGAGAAGCGAGCCGATAACTTCTGCGTACGTGACTCA~C6GA'Ad'Td'G~TCT
-591 -531 -471 -411 -351 -291 -231 -171 111 -51 10 70 130 190 250 310 370 430 490 550 610 670 730 79O 850 910 970
tocytes, in addition to a weak accumulation in the imagihal discs, larval structures which develop into the adult appendages (Agnel et al., 1989; 1992). In this paper, we establish that there is alternative splicing of the rnRacGAP primary transcript; we define the complete sequences of the rnRacGAP(1) and rnRacGAP(2) transcript forms and characterize the pattern of their specific expression in germline and somatic tissues.
CTTCGGAA~TT~CC~ACZ~ACCAGAZ6_m?.~T~A~CAAr~CCCA~6~¥AGTG
GCTCTGGGAGTCGGACGCCCTCAAATCG~CTTTATTTGTCGCCCGTTCGCCCCACGATGC AGAACAAACGGAGGCTTCTGCGGGAGTACAGGTCCTACGATGACCTCAGCGAGCACTACC GCATGTTTGGATCGCAGTCCCTGGACTC~ATCGCGTGGACATGAATCCCAGTG GCTGCGATGGCTTATCCACG-GACGGACTCGACTTCTGCTCGCAGTCCCACTCGGGGCTGC TGAGGGAGCACAATTTCAAGATTAAGTCCTACTACTACAACGTGGGGAACTGCGTACACT
GCCGCAAAA(~tgagttgecatcaccgaatggtggaacttctgagaggatttc~aattc~ gggaaaacatgaaaagttaaataacatcccaaccctctct~accaacatttacctcecct ,aaCCaCqtOtccctqqtcttct%cca~GATCCGCTTCGCCATGGCCAGCCTGAGGTGTCG GGCGTGTCCGCTCCGCTGCCACATCGGTTGCTGCCGCCAGCTGACGGTGAACTGCATCCC CCAGCCTCAAATCGGTACGAAGAGAGGATGTCTCAGCGATTATGCGCCTCGGGTGGCGCC CAT~GTGCCGGCC~TAATTGTCCATT(]TGTAACdf;AGATC(~AGGC(;CCGGGATTGCAGCA GGAGGGACTCTACCGGGTGTCCTCCACCCGAGACAAATGCAAACCACTGCGTCGGAAGC7 GCTGCGTGGCAAGTCCACGCCGCATTTGGGCAATAAAGACACCCACACACTGTGCTGTTG TGTTAAGGACTTTCTTCGGCACTTGGTTCACCCTCTGATTCCCATTTACCACCGAA~GGA TTTCGAGGAGGCCACGCGGCACGAAGATCGTCTGGCTGTGGAGATGGCGGTATATTTAGC GGTTCTGGACCTCCATCAGGCTCACAGGGATACGTTGGCCTACTTAATGCTTCACTGGCA GAAAATTGCCCAAAGTCCTGCTGTCCGCATGACGGTCAACAATCTTGCCGTGCT~ TC--'6~ACTCT.C~TTCGGAGATCTGGATCTAACCCTCGAAAATGTCG~'CACTTGGCAGCGGGT GCTGAAGGTGCTACTTCTGATGCCTGCCGGATTTTGGTCTCAGTTCTTGGAGGTCCATCC CCTTCCCACTTCGTTGGGCAGCACCTATGACTTTGAGGACCGGTACAATCACCGTCACTG GGACAGCTCGTCCAATCTGGGATGGTCTTCTGTTAAGACCTACT T T A G A T C A A T ~ a g
1030 1090 1150 1210 1270 acctatggtttccctt~--a~{~g-g~'a'g-a'~'t't'~atattcgaataagccaaac "ca 1330 at @,caaqcaatcaaacat~ cat aca aa Let tt cact qaat act t cctqt tt ca 1390 CCTTAGCAGTACTCACCT~A~GA~AGTCTTATSACTGAGACArCGAGATGGATGCAGT 1450 "CA~'C~'T'~TCA'~T~'ATTTTG TAT TT T2'TTTC AT T AT TC AC TAC AG T TGAAG A A ~ ~ ; I 1510 AATAATTGGATCTCTAA tgt tgtqt a ~ t ta~.gcqgct aacaaq~g 1570 ta 1572
(b) (1266) g t a
aga o c t
atg gtt
tcc
ctt[~]
(1289)
(377) V a l A r g P r o Met V a i
S e r Leu * * *
(1386) g t c
act cac cta ~
aac e t t
agc a g t
(377) V a l A s n L e u Ser Ser Thr His L e u * * * (C)
(-1130)
370
575
(-583)
Form2
(383) (1415) Form 1 ***
(384)
(364)
Genomic DNA
137
Form 2 Exo l
v
E
2
m2
(1386)
Fig. 1. The rnRacGAP gene. (a) Genomic nt sequence. Positive numbering starts from the A pf the ATG start codon, which is encircled. The first EcoRI fragment and Exons 1, 2 and 3 are outlined. Exons are written in upper case, flanking sequences and introns are in lower case. Restriction sites are surrounded by boxes with outlines that are solid, EcoRI; dotted, BglI; dashed, PvuII; or notched, PstI. The tsp starts at the boxed A at nt -212. The start of Form-1 cDNA is indicated by an open circle at nt -120. PCR primer regions are underlined with an arrowhead indicating the polarity, except for as2. i, overlined for clarity. For the nested primers, the overlapping nt are enclosed on three sides. The Antp binding site at nt - 4 6 0 is within an oval outlin£ The ATG start codon is encircled. Stop codons are bracketed from above: for Form 1 at nt 1287 and for Form 2 at nt 1410. The polyadenylation addition site at nt 1348 in intron 2 is indicated by two upper case A's. (b) Deduced aa sequences of the nt sequences differing between Forms
RESULTS AND DISCUSSION
(a) Identification of alternatively spliced products of rnRacGAP by 3' RACE Sequence analyses of two incomplete, but overlapping, cDNA clones coding for RnRacGAP (pcl.7 and pcl.7d; Agnel et al., 1992), indicated the presence of differing 3' ends, suggesting that the primary transcript was being
1 and 2. The numbers in parentheses indicate the position of the nt as in panel a or for the aa as in Agnel et al. (1992). The stop codons are overbracketed. (c) Genomic organization of the rnRacGAP gene. The positions of the EcoRI sites, and asl and as2 primers are indicated by numbers in parentheses, and all other primers, within their symbols. The numbering is from the first 5' nt in the sequence as given in panel a. Top line: previously unpublished sequence data are represented by heavy lines and include: 5' flanking sequence (548 nt); intron I (137 nt); intron 2 (120 nt); Exon 3 (143 nt with 121 nt of new sequence); and 3' flanking sequence (44 nt). EcoRl (circles), BglI (diamond), PvulI (triangle) and PstI (square) restriction sites are indicated. Middle line: Exon-intron organization of rnRacGAP(2). The predicted exons are indicated by boxes, with coding regions shaded. Sense primers are encircled; antisense primers are indicated as clear (common coding sequence primers) or hatched (exon-specific primers) boxes bearing shortened versions of primer names. Introns are drawn as inverted triangles. Bottom line: Exon-intron organization of rnRacGAP(1). Symbols as for the Form 2. The large arrow indicates the 92-nt distance to the start of the original pcl.7 cDNA clone. Methods: Mini-preparation of plasmid DNA by the alkaline lysis method, subcloning procedures and sequencing by the Sanger dideoxy method are described in Sambrook et al. (1989). For RACE, cDNA was made from wild-type red e H female adult poly(A)+RNA using the Amersham cDNA synthesis kit and primer T17-Xho to prime cDNA synthesis. First-round RACE was performed using a primer, GAP1, located in the GAP-homology domain; second-round RACE with with the nested sense primer GAP2. Reaction conditions: 100 gl reaction volume containing Vent buffer supplied by the manufacturer (NE BioLabs, Beverly, MA, USA) with l gg BSA 500 ng each primer 1 unit Vent polymerase. 1 Ixl of first round RACE products were used for second round. Exon 3 product was identified using Southern and restriction analysis and subcloned by blunt-end ligation into Bluescript (Stratagene, La Jolla, CA, USA) for sequence analysis. The nt sequence data reported in this paper will appear in the EMBL, GenBank and DDBJ Nucleotide Sequence Databases under accession No. U22539. Primers: asl, 5'-GGTGAGTACTGCTAAGGTTGAC; as2, 5'-GGGAAACCATAGGTCTTACC; GAP1, 5'- CTTGCCGTGATCTTCGCTCC; GAP2, 5'-TCTTCGCTCCAACTCTGTT; Tl7-Xho, 5'-GACTCGAGTCGACATCGA(T)17.
137 alternatively spliced. The C-terminal end of the pcl.7 clone encoded 25 nt of an incomplete third exon. We isolated this exon in its entirety by the 3' RACE method (O'Hara et al., 1989) from female adult poly(A)+RNA. RACE reactions were performed with an oligo(dT) polymer containing 3' anchor primer and nested primers located in the GAP-homology domain, and GAP 2 (Fig. 1). Sequence analysis revealed that exon 3 contained 143 nt: 24 nt corresponding to coding sequence followed by two consecutive in-frame stop codons. The complete coding sequence defined by exons 1, 2 and 3 is now referred to as rnRacGAP(1), or Form 1. We also established the exact sizes of intron 2 (120bp), intron 1 (137 bp) exon 2 (808 bp) and exon 3 (143 bp), as well as portions of the 5' (548 bp) and 3' (44 bp) flanking regions. The genomic sequence containing the rnRacGAP gene is given in Fig. la and schematicized in Fig. lc At the polypeptide level, the 3' divergence in the nt sequences of rnRacGAP(1) and the previously characterized eDNA clone (Agnel et al., 1992) coding for rnRacGAP(2) or Form 2, translated to differences of only seven, or six, amino acids, respectively, at the C-termini of the protein products (Fig. lb). No specific subcellular localization signals were found within the differing C-termini. (h) Characterization of rnRacGAP 5' terminus In order to ensure that the differences in rnRacGAP(1) and rnRacGAP(2) were limited to the 3' ends, we undertook a series of in vivo and in vitro experiments to characterize the 5' limit of transcription.
(1) In vivo analysis Initially, we mapped the size of the rnRacGAP 5' terminus of Form 1 by RNAse H cleavage of a hybrid formed between adult male and female poly(A)+RNA and an antisense primer (prPst) 268 bp downstream from the tsp of clone pcl.7. Northern blot analysis followed by hybridization with a 5'-specific probe shows a single band in adult male mRNA of about 340 bp (Fig. 2a; lane 2). No signal was detectable in female flies (lane 1); non-specific degradation was ruled out by a control RT-PCR (bottom) with r-gene rp49 (O'Connell and Rosbash, 1984). These results define approx. 80 bp of additional upstream sequence. To more accurately determine the position of the tsp, we utilized the primer-extension method with a primer, prPvuII, located 150 nt upstream from the prPst primer and adult fly poly(A)+RNA. As shown in Fig. 2b, a fragment was detected corresponding to the size predicted by the RNase H experiment, defining a 212-nt 5' UTR. Finally, we verified that rnRacGAP(I) and rnRacGAP(2) differed only at their 3' ends by RT-PCR
analysis of adult male poly(A)+RNA with a sense primer (prexls5') 10 bp downstream from the tsp, in combination with either the Form-l-specific asl primer, or nested Form-2-specific primers (as2 and as2.1). We obtained products (Fig. 2c) of the expected sizes (thicker arrows) for both Forms 1 (lane 3) and Form 2 (lane 1); the highermolecular-weight bands correspond to genomic DNA (thinner arrows; lanes 4 and 2, respectively). Comparative restriction analyses (data not shown) confirmed that Forms 1 and 2 were colinear. (2) In vitro transcription To verify the functionality of the putative rnRacGAP promoter, a heterologous promoter construct was made containing a genomic fragment that fused 1700 bp of upstream sequence and the first 154 bp (site of BglI cleavage) of the 212 bp 5' UTR, with the bacterial cat gene. This genomic rnRacGAP DNA was cloned in both sense and antisense orientations with respect to the cat reporter. Supercoiled plasmid DNA was incubated with Drosophila embryo nuclear extract for 30 rain at 37°C, and the resulting in vitro transcripts were analyzed by primer extension using a cat-specific primer according to Kadonga (1990). As a positive control, a promoter construct encoding the Xenopus histone H1 ° promoter fused to cat (Khochbin and Wolffe, 1993) was processed in parallel with the rnRacGAP-cat constructs. A primerextension product corresponding to the predicted tsp was observed (Fig. 3, arrow, lane 2), confirming that the rnRacGAP region contains a promoter. The absence of primer-extension products from the rnRacGAP-cat antisense construct (lane A) demonstrates the specificity of the in vitro sense transcription products. These combined results, schematically represented in Fig. 2d, point to an authentic rnRacGAP tsp site as occurring 212bp upstream from the ATG start codon. Structural features surrounding this tsp are consistent with previously identified tsp. The tsp itself (GCA(+ 1)CTCT) conforms to a mammalian TFIID binding site (Y, Y, A(+ 1), N, W, Y, Y; Javahery et al., 1994) and is nearly identical with a canonical cap site identified for Drosophila transcripts (D, Y, A(+ 1), K, T, G; Purnell et al., 1994) (where D--A or G or T, K = G or T, W = A or T and Y = C or T). A Hogness box motif similar to that described for the dsx gene (Fig. 1, TACAAATGTA; Burtis et al., 1989) is located 24 bp upstream from the mapped tsp (Fig. 1, at -237). A TATA-box element was found - 2 3 0 bp upstream (Fig. 1, nt - 4 4 2 ) from the mapped tsp but no G+C-rich, nor CAAT box, sequence was found. At -251 bp upstream from the tsp (Fig. 1, nt -460), a consensus sequence was identified for the Antennapedia homeotic protein-binding site (ATTTAATTGA; Hanes and Brent, 1991).
138 (a)
(c)
(b) 1
2
(d)
bp 404
Mr
1
2
3
4
--
kb
in vivo
7.6 ~> 5.3 ~> 2.8 D,
308 --
[] bp
242-238 --
2000
I --
RNase H fragment -~
217 --
1.9 ~> 1.6 ~>
1600--
201--
5' [fiR 212
EXON1 | ~147~ i AUG1 1 "~" 3 4 0
190 -180 --
1.0 D,
160 --
1000--
/
147 --
0.6
D"
0.4
D-
Primer extension fragment in vitro
122
--
Ii0
--
0.3 {> rp49
?
500
rn [ ~ I
cat
J
m
s48 //
370
~
,
[:2>
[i~i ~ i72~
>
in vitro transcription fragment 4
154
!99
[
90--
76-67-Fig. 2. Analysis of the rnRacGAP transcript 5' terminus. (a) Northern blot analysis of female (lane 1) and male (lane 2) poly(A) + RNA after incubation with a 5' primer located at +268 nt from the 5' start site of the published cDNA clone pcl.7 (Agnel et al., 1992) and RNAse H treatment. The filter was stripped and re-hybridized with a random-prime labeled PCR fragment of the rp49 gene (lower panel). Primers: prPst, 5'-ATCCTGCAGCGAGTCCAGGG; RP49-antisense, 5'-GTGTATTCCGACCACGTTACA; RP49-sense, 5'-TCCTACCAGCTTCAAGATGAC. (b) Primer extension product from adult fly poly(A)+RNA. Primer: prPvulI, 5'-GTTGGCGAATTGATTCAGCTG. (c) RT-PCR amplification of rnRacGAP(1) and rnRacGAP(2). The RT-PCR reaction was carried out with a sense primer + 10 bp from the mapped tsp site (exls5') and form-specific primers. Lanes 1 and 2: Form-2-specific PCR products; lanes 3 and 4: Form-l-specific PCR products. The templates were oligo(dT)-primed cDNA generated from poly(A)+RNA from adult males in lanes 1 and 3, and cloned genomic DNA in lanes 2 and 4. The thicker arrows mark the position of the form-specific spliced products; the thinner arrows, the positions of amplified genomic DNA. Equal amounts of PCR product were separated on 1.0% agarose gels in the presence of ethidium bromide. Molecular weight marker is the Boehringer 1-kb ladder. The image was produced with an Is-700 Camera, using Is-1000 Digital Imaging System (Alpha lnnotech, San Leandro, CA, USA). primers: exls5', 5'-GGGTGATCGAATTTTTGTGC; as2.1, 5'-GAAATTCTATCCAAATCCG. (d) Schematic representation of combined results in vivo and in vitro analysis of the 5' ends of Forms 1 and 2. Symbols as in Fig. 1; dark square is HindlII site. Methods: For the RNase H experiment, RNA was isolated according to Chomczynski and Sacchi (1987) or using RNAzol B (Bioprobe Systems, Montreuil-Sous-Bios, France). Poly(A) + RNA was isolated using oligo(dT) cellulose columns (Sigma). RNA pellets containing approx. 10 ixg poly(A)+RNA were resuspended in 50 gl DEPC-treated water containing RNAsin (Promega, Madison, WI, USA) and 1 gg of a 21-nt primer prPst. RNA and oligo were heated at 65°C for 5 min, then placed on ice. RNase H treatment was performed according to Ausubel (1991) with an incubation at 38°C for 20 rain at 1 unit RNase H. RNA was separated on a 2.2 M formaldehyde-l.8% agarose gel, transferred to a nylon filter and hybridized with a random-prime labeled probe (Feinberg and Vogelstein, 1983) from the first 300 bp of the rnRacGAP gene (upper panel). For the primer extension reaction: 5' end labeling of the PvulI primer (100 ng) and the primer extension reaction from adult fly poly(A)+RNA were performed as in Khochbin and Wolffe (1993). For the PCR reactions: The cDNA template was made from 1 gg poly(A)+RNA and oligo(dT) (Amersham cDNA synthesis kit). PCR reaction conditions are as follows: 50 gl total volume with undiluted Cetus (Perkin-Elmer) Taq buffer/200 mM dNTP/20 pmol of each primer was heated with the cDNA template for 5 rain at 94°C under paraffin oil. 1 unit of Taq polymerase was added and the reaction was cycled 30 times at 94°C for 1 min, 55°C for 2 rain and 72°C for 3 min.
(c) Tissue-specific expression of rnRacGAP(1) and
produced
rnRacGAP(2)
signal from adult females in the RNase H experiment, we
i n t h e testes, a n d i n l i g h t o f t h e a b s e n c e o f
had
t e s t e d R N A f r o m m a l e a n d f e m a l e flies, as well as dis-
shown that the majority of rnRacGAP transcripts are
sected ovaries, testes and the remaining carcasses, for the
As p r e v i o u s
in situ hybridization
experiments
139
1A2
3456
7
A
tlt m
i¸i;!ii !!ii
i !iii
!ii~~i~i?i!~~i~!~i!~illii~~
presence of both rnRacGAP(1) and rnRacGAP(2). Form-specific amplification was achieved with the antisense primers, asl and as2, which are equidistant (232 bp) from the sense primer, GAP1 (Fig. 1). As shown in Fig. 4A, rnRacGAP(1) produces signals from DNA and RNA that can be directly distinguished by their migration position, while rnRacGAP(2), colinear with the genomic DNA in this region, produces PCR products of the same size from both RNA and potentially-contaminating DNA. To remove DNA-associated signals we treated aliquots of all RNA preparations with DNase followed by phenol extraction prior to RT-PCR reactions (Fig. 4B). As a control for the extent of DNase action, excess rn genomic DNA was added to an aliquot of the male carcass RNA sample, and the mixture then treated in parallel with experimental samples; the DNA-related products from Form-l-specific (lane 13) and Form-2-specific (lane 15) amplification are completely removed by DNase treatment (lanes 14 and 16, respectively). Thus, the residual signals present in lanes 1-12 are RNA-specific and show that the DNase treatment removes all rnRacGAP(2)-generated signals (lanes 2, 6, 8, 10 and 12) except for a relatively weak signal still present in the testes (lane 4). The highly predominant form in the testes is rnRacGAP(1) (lane 3). This expression appears to account for the intense signal in whole male flies (lane 11), as the male carcass sample shows considerably smaller amounts of rnRacGAP(1) (lane 7). Low levels of Form-1 expression are also detected in female carcasses (lane 5) and ovaries (lanes 1). While transcript is not detectable here in whole female flies (lanes 9-10), a 3-times longer exposure renders rnRacGAP(1) detectable (data not shown). These results suggest that rnRacGAP(1) expression
cat primer, 5 ' - G G T A T A T C C A G T G A T T T T T T T C T C C A T . Methods: A HindlII fragment from the rn locus was amplified with PCR primers
Fig. 3. In vitro experiments to determine the 5' tsp. Autoradiography of primer extension products of in vitro transcripts generated from heterologous rnRacGAP-cat fusion genes and a control histone HI"cat gene: histone H1°-cat (lane 1), antisense rnRacGAP-cat (lane A) and sense rnRacGAP-cat (lane 2; arrow). Lanes 3-6, rnRacGAP-cat sequencing reaction. Lane 7, end-labeled pBR322-MspI molecular weight marker, prextBgll primer, 5'-AGGCCACTTCGGCTTGAGTC;
prextBglI and the T7 primer (Boehringer) to generate a blunt-end fragment containing 1700-bp 5' flanking sequence and 154-bp 5 ' U T R of the rnRacGAP gene. This fragment was ligated into the BasicCAT (Prornega) vector Klenow-filled-in HindlII cloning site in both sense and antisense orientation. Sequencing of the promoter region showed the absence of mutations from vent amplification. A control plasmid carried the histone H1 ~" promoter (Khochbin and Wolffe, 1993). Plasmids were purified twice over CsCI gradients to recover supercoiled DNA which was used to generate in vitro transcripts with the Prornega Drosophila Embryo Nuclear Extract Transcription System and accompanying protocol. Primer extension of the resulting transcripts was performed as in Khochbin and Wolffe (1993). The ethanol-precipitated DNA was resuspended in 3 gl loading dye mix (2 lal loading dye from US Biochemical (Cleveland, OH, USA) sequencing kit and 1 p.l 0.1 M NaOH), heated for 5 min at 100°C and loaded onto a 7 M urea-6% polyacrylamide sequencing gel along with a sequencing reaction performed with the US Biochemical sequencing kit, using the rnRacGAPcat sense construct as template, unlabeled dNTPs and the same endlabeled cat primer used in the primer-extension reaction.
140 1
2
3
4
5
15
7
8
9
I0
II
12
13
14
15 15
I
J Fig. 4. Tissue-specific expression of rnRacGAP(1) and rnRacGAP(2). Total RNA was isolated from the indicated tissues, subjected to RT-PCR and subsequent amplification and tested by Southern blot analyses: 80 ovaries, lanes 1 and 2; 60 testes, lanes 3 and 4; 10 female carcasses, lanes 5 and 6; 10 male carcasses, lanes 7 and 8; 10 adult females, lanes 9 and 10; 10 adult males, lanes 11 and 12. Products in the odd-numbered lanes from 1 12 were generated from the Form-lspecific primer pair, GAP 1-as l; products in even-numbered lanes from 1-12, from the Form-2-specific primer pair, GAPI-as2. (A) Autoradiogram of samples before DNase treatment probed with DNA from the 3' common coding region of the rnRacGAP transcript. Controls for the size of the genomic DNA and the Form-specific RNAs are presented: amplification of genomic DNA, lane 13; amplification of cloned rnRacGAP(1) cDNA, lane 14; and amplification of cloned rnRacGAP(2) cDNA, lane 15. (B) Autoradiogram of samples after DNase treatment to remove DNA-associated signals; probe as in A. Controls for the extent of DNase action are shown in lanes 13 16. Genomic DNA was added to an aliquot of the male carcass RNA preparation to be treated in parallel: lane 13, Form-l-specific amplification, before DNase treatment; lane 14, after DNase treatment; lane 15, Form-2-specific amplification before DNase treatment; lane 16, after DNase treatment. Note complete disappearance of the DNA-generated signal in these two samples indicating completeness of the DNase reaction. Lower panel: Autoradiograms of PCR products produced With primers specific for the ribosomal protein rp49 gene and probed with rp49 DNA as control for sample addition. These samples were not treated with DNase. Methods: For DNase (RNase-free; Stratagene) reaction, 2 gl RNA (0.2 gg/gl) was resuspended in a buffered 20 gl final reaction volume containing 1 Ixl RNasin and 2 gl of DNase (10 units/l~l) and incubated for 30 min at 37°C. The samples were then phenolextracted and ethanol-precipitated prior to PCR reactions. PCR reaction conditions are as described in the legend to Fig. 2. PCR products were analysed on a 2.0% agarose gel.
may be necessary for testes function. In order to test this hypothesis, we are currently introducing rnRacGAP transgenes into flies deficient for the rn locus. Preliminary results suggest that testes-associated function may, indeed, be contained in the Form 1 differing 3' end: Form 2 transcript alone cannot rescue male sterility, while a genomic fragment containing complete RnRacGAP sequence encompassing both 3' ends can restore male fertility (A.G., unpublished data).
(d) Conclusions (I) Cloning and sequencing of the alternative exon present in rnRacGAP(1) confirm that an alternative splice pathway operates at the 3' end of the rnRacGAP primary transcript. In vivo and in vitro experiments to characterize the 5' ends of rnRacGAP(1) and rnRacGAP(2) defined a 212-bp 5' UTR. The remaining sequence between the 5' and 3' ends of rnRacGAP(1) and rnRacGAP(2) was determined to be colinear. Thus, the differences between the two Forms are restricted to the 3' ends and involve 143 nt for Form 1, and 83 nt for Form 2. At the polypeptide level, these differences translate, respectively, to seven, or six, alternative C-terminal aa. (2) Previous in situ hybridization experiments (Agnel et al., 1992) had shown that in adult males, the rnRacGAP sequence was specifically produced in the primary spermatocytes; at this level of sensitivity, no signal could be detected in females. In this study, we show by PCR analysis of male and female flies, and isolated germ tissue and carcasses, that rnRacGAP(1) is the predominant species in the testes and is expressed at a low level in the ovary, and the somatic tissues of both sexes, rnRacGAP(2) is only very weakly expressed and is detectable solely in the testes.
ACKNOWLEDGEMENTS
These experiments were carried out in the laboratory of Dr. J.J. Lawrence, whom we thank for support. Our appreciation to Saadi Khochbin for helpful comments and critical reading of the manuscript. C.D.H. was supported by postdoctoral fellowships from the CEA, Centre d'Etudes Nucl6aires (Grenoble, France) and the International Fogarty-INSERM exchange program (IF06TW01881-01).
REFERENCES Agnel, M., Kerridge, S., Vola, C. and Griffin-Shea, R.: Two transcripts from the rotund region of Drosophila show similar positional specificities in imaginal disc tissues. Genes Dev. 3 (1989) 85-95. Agnel, M., Roder, L., Vola, C. and Griffin-Shea, R.: A Drosophila rotund transcript expressed during spermatogenesis and imaginal disc morphogenesis encodes a protein which is similar to human Rac GTPase-activating (racGAP) proteins, Mol. Cell Biol. 12 (1992) 511l 5122. Ausubel, F.M., Brent, R., Kingston, R.G., Moore, D.D., Smith, J.A., Leidman, J.G. and Struhl, K.: Current Protocols in Molecular Biology. Greene Publishing Associates and Wiley Interscience, New York, NY, 1991. Burtis, K.C. and Baker, B.S.: Drosophila doublesex gene controls somatic sexual differentiation by producing alternatively spliced mRNAs encoding related sex-specific polypeptides. Cell 56 (1989) 997 1010.
141 Chomczynski, P. and Sacchi, N.: Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chlorofrm extraction. Anal. Biochem. (1987) 156-159. Feinberg, A.P. and Vogelstein, B.: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132 (1983) 6-13. Guichard, A., Bergeret, E. and Griffin-Shea, R.: RotundRacGAP. In: Zerial, M. and Huber, L.A. (Eds.), Guidebook to the Small GTPases. Oxford University Press, Oxford, 1995, in press. Hall, A.: Small GTP-binding proteins and the regulation of the actin cytoskeleton. Annu. Rev. Cell Biol. 10 (1994) 31 54. Hanes, S.D. and Brent, R.: A genetic model for interaction of the homeodomain recognition helix with DNA. Science 251 (1991) 426-430. Javahery, R., Khachi, A., Lo, K., Zenzie-Gregory, B. and Smale, S.T.: DNA sequence requirements for transcriptional initiator activity i~ mammalian cells. Mol. Cell. Biol. 14 (1994) 116-127. Kadonaga, J.T.: Assembly and disassembly of the Drosophila RNA polymerase II complex during transcription. J. Biol. Chem. 265 (1990) 2624 2631. Kerridge, S. and Thomas-Cavallin, M.: Appendage morphogenesis in Drosophila: a developmental study of the rotund (rn) gene. Wilhelm Roux's Arch. Dev. Biol. 197 (1988) 19-26.
Khochbin, S. and Wolffe, A.P.: Developmental regulation and butyrateinducible transcription of the Xenopus histone HI ° promoter. Gene 128 (1993) 173-180. Lamarche, N. and Hall, A.: GAPs for rho-related GTPases. Trends Genet. 10 (1994) 436-440. O'Connell, P.O. and Rosbash, M.: Sequence, structure, and codon preference of the Drosophila ribosomal protein 49 gene. Nucleic Acids Res. 12 (1984) 5495-5513. O'Hara, O., Dorit, R.L. and Gilbert, W.: One-sided polymerase chain reaction: the amplification of cDNA. Proc. Natl. Acad. Sci. USA 86 (1989) 5673 5677. Purnell, B.A., Emanuel, P.A. and Gilmour, D.S.: TFIID sequence recognition of the initiator and sequences farther downstream in Drosophila class II genes. Genes Dev. 8 (1994) 830-842. Ridley, A.J., Paterson, H.F., Johnston, C.L., Diekmann, D. and Hall, A.: The small GTP-binding protein rac regulates growth factorinduced membrane ruffling. Cell 70 (1992) 401 410. Sambrook, J., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989.