FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1520556951> ?p ?o ?g. }

• W1520556951 endingPage "17835" @default.
• W1520556951 startingPage "17830" @default.
• W1520556951 abstract "The 9G8 factor is a 30-kDa member of the SR splicing factor family. We report here the isolation and characterization of the human 9G8 gene. This gene spans 7745 nucleotides and consists of 8 exons and 7 introns within the coding sequence, thus contrasting with the organization of the SC35/PR264 or RBP1 SR genes. We have located the human 9G8 gene in the p22-21 region of chromosome 2. The 5′-flanking region is GC-rich and contains basal promoter sequences and potential regulatory elements. Transfection experiments show that the 400-base pair flanking sequence has a promoter activity. Northern blot analysis of poly(A)+ RNA isolated from human fetal tissues has allowed us to identify five different species, generated by alternative splicing of intron 3, which may be retained or excised as a shorter version, as well as the use of two polyadenylation sites. We also show that the different isoforms are differentially expressed in the fetal tissues. The persistence of sequences between exon 3 and 4 results in the synthesis of a 9G8 protein lacking the SR domain which is expected to be inactive in constitutive splicing. Thus, our results raise the possibility that alternative splicing of intron 3 provides a mechanism for modulation of the 9G8 function. The 9G8 factor is a 30-kDa member of the SR splicing factor family. We report here the isolation and characterization of the human 9G8 gene. This gene spans 7745 nucleotides and consists of 8 exons and 7 introns within the coding sequence, thus contrasting with the organization of the SC35/PR264 or RBP1 SR genes. We have located the human 9G8 gene in the p22-21 region of chromosome 2. The 5′-flanking region is GC-rich and contains basal promoter sequences and potential regulatory elements. Transfection experiments show that the 400-base pair flanking sequence has a promoter activity. Northern blot analysis of poly(A)+ RNA isolated from human fetal tissues has allowed us to identify five different species, generated by alternative splicing of intron 3, which may be retained or excised as a shorter version, as well as the use of two polyadenylation sites. We also show that the different isoforms are differentially expressed in the fetal tissues. The persistence of sequences between exon 3 and 4 results in the synthesis of a 9G8 protein lacking the SR domain which is expected to be inactive in constitutive splicing. Thus, our results raise the possibility that alternative splicing of intron 3 provides a mechanism for modulation of the 9G8 function. The splicing of nuclear pre-mRNA occurs in a multicomponent complex containing small nuclear ribonucleoproteins (U snRNP),1 1The abbreviations used are: snRNPsmall nuclear ribonucleoproteinRBDRNA binding domainntnucleotide(s)PCRpolymerase chain reactionkbkilobase(s)bpbase pair(s)CATchloramphenicol acetyltransferasePipes1,4-piperazinediethanesulfonic acid. 1The abbreviations used are: snRNPsmall nuclear ribonucleoproteinRBDRNA binding domainntnucleotide(s)PCRpolymerase chain reactionkbkilobase(s)bpbase pair(s)CATchloramphenicol acetyltransferasePipes1,4-piperazinediethanesulfonic acid. splicing factors and hnRNP proteins (for reviews, see 18Green M.R. Annu. Rev. Cell Biol. 1991; 7: 559-599Google Scholar and 33Moore J.M. Query C.C. Sharp P.A. Gesteland R.F. Atkins J.F. The RNA World. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1993: 303-358Google Scholar). Numerous splicing factors have been characterized in both lower (mainly the PRP factors) and higher eukaryotes. In higher eukaryotes, a unique set of factors, which belong to the family of the splicing factors (called the SR factors) rich in serine and arginine residues, is involved in the first steps of splice sites recognition. At present, seven SR factors have been identified and characterized: SF2/ASF (22Krainer A.R. Conway G.C. Kozak D. Genes & Dev. 1990; 4: 1158-1171Google Scholar, 23Krainer A.R. Conway G.C. Kozak D. Cell. 1990; 62: 35-42Google Scholar; 16Ge H. Manley J.L. Cell. 1990; 62: 25-34Google Scholar), SC35/PR264 (13Fu X.D. Maniatis T. Nature. 1990; 343: 437-441Google Scholar, 14Fu X.D. Maniatis T. Science. 1992; 256: 535-538Google Scholar; 40Sureau A. Soret J. Vellard M. Crochet J. Perbal B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 11683-11687Google Scholar), SRp20 (also named X16) (44Zahler A.M. Lane W.S. Stolk J.A. Roth M.B. Genes & Dev. 1992; 6: 837-847Google Scholar; 1Ayane M. Preuss U. Khler G. Nielsen P.J. Nucleic Acids Res. 1991; 19: 1273-1279Google Scholar), SRp55 (34Roth M.B. Zahler A.M. Stolk J.A. J. Cell Biol. 1991; 115: 587-596Google Scholar; 32Mayeda A. Zahler A.M. Krainer A.R. Roth M.B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 1301-1304Google Scholar), SRp75 (46Zahler A.M. Neugebauer K.M. Stolk J.A. Roth M.B. Mol. Cell. Biol. 1993; 13: 4023-4028Google Scholar), RBP1 (20Kim Y.-J. Zuo P. Manley J.L. Baker B.S. Genes & Dev. 1992; 6: 2569-2579Google Scholar), and the 9G8 factor (8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar). All these factors share two common characteristics: one or two RNA binding domains (RBD) near the amino terminus and a domain rich in serine/arginine (the SR domain) at the carboxyl terminus. The SR factors range in mass from 20 to 75 kDa and the best characterized are the 30-kDa SF2/ASF and SC35 (22Krainer A.R. Conway G.C. Kozak D. Genes & Dev. 1990; 4: 1158-1171Google Scholar, 23Krainer A.R. Conway G.C. Kozak D. Cell. 1990; 62: 35-42Google Scholar, 24Krainer A.R. Mayeda A. Kozak D. Binns G. Cell. 1991; 66: 383-394Google Scholar; 16Ge H. Manley J.L. Cell. 1990; 62: 25-34Google Scholar; 17Ge H. Zuo P. Manley J.L. Cell. 1991; 66: 373-382Google Scholar; 13Fu X.D. Maniatis T. Nature. 1990; 343: 437-441Google Scholar, 14Fu X.D. Maniatis T. Science. 1992; 256: 535-538Google Scholar; 32Mayeda A. Zahler A.M. Krainer A.R. Roth M.B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 1301-1304Google Scholar). small nuclear ribonucleoprotein RNA binding domain nucleotide(s) polymerase chain reaction kilobase(s) base pair(s) chloramphenicol acetyltransferase 1,4-piperazinediethanesulfonic acid. small nuclear ribonucleoprotein RNA binding domain nucleotide(s) polymerase chain reaction kilobase(s) base pair(s) chloramphenicol acetyltransferase 1,4-piperazinediethanesulfonic acid. It has been argued that the various SR factors are interchangeable in constitutive splicing because each is able to complement SR-deficient extracts (for instance a cytoplasmic S100 extract) (15Fu X.D. Mayeda A. Maniatis T. Krainer A.R. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 11224-11228Google Scholar; 32Mayeda A. Zahler A.M. Krainer A.R. Roth M.B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 1301-1304Google Scholar). Recently, it has been shown that SF2/ASF and SC35 are able to form commitment complexes with a pre-mRNA substrate (12Fu X.-D. Nature. 1993; 365: 82-85Google Scholar) and that they are required for the stable interaction of U1 snRNP with the 5′ splice site (21Kohtz J.D. Jamison S.F. Will C.L. Zuo P. Lhrmann R. Garcia-Blanco M.A. Manley J.L. Nature. 1994; 368: 119-124Google Scholar). In agreement with these results, the region of SF2/ASF containing the two RBDs is able to recognize a typical 5′ splice site in a short transcript (21Kohtz J.D. Jamison S.F. Will C.L. Zuo P. Lhrmann R. Garcia-Blanco M.A. Manley J.L. Nature. 1994; 368: 119-124Google Scholar). An interesting aspect of SR factors is that they may modulate alternative splicing in a concentration-dependent manner when several 5′ splice sites are in competition. Increasing amounts of SF2/ASF or SC35 generally result in the preferred selection of the more proximal 5′ splice site (23Krainer A.R. Conway G.C. Kozak D. Cell. 1990; 62: 35-42Google Scholar; 16Ge H. Manley J.L. Cell. 1990; 62: 25-34Google Scholar; 15Fu X.D. Mayeda A. Maniatis T. Krainer A.R. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 11224-11228Google Scholar). However, a more extended comparison, including SRp40, SRp55, and SRp75, indicates that each SR factor has a differential ability to modulate alternative splicing in vitro (45Zahler A.M. Neugebauer K.M. Lane W.S. Roth M.B. Science. 1993; 260: 219-222Google Scholar). Moreover, as these factors are differentially expressed in different tissues, 45Zahler A.M. Neugebauer K.M. Lane W.S. Roth M.B. Science. 1993; 260: 219-222Google Scholar proposed that the SR factors may be involved in tissue-specific regulation of alternative splicing in vivo. In support of this idea, overexpression of SF2/ASF by transfection experiments led to a modulation of alternative splicing in vivo (6Caceres J.F. Stamm S. Helfman D.M. Krainer A.R. Nature. 1994; 265: 1706-1709Google Scholar). We have isolated recently the 9G8 factor with a molecular mass (~30 kDa) similar to those of SF2/ASF and SC35 (8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar). However, its primary sequence in the RBD is only ~40% conserved relative to SF2/ASF and SC35. In addition, 9G8 presents some specific features since it contains an RRSRSXSX consensus sequence repeated six times in the SR domain and a CCHC zinc knuckle motif in its median region (8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar). The occurrence of a large family of SR splicing factors which are differentially expressed in organisms raises questions related to the structure of their genes, the existence of a common ancestral gene, and the molecular basis of their modulated expression. In contrast with the abundant data on the SC35/PR264 gene (42Vellard M. Sureau A. Soret J. Martinerie C. Perbal B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2511-2515Google Scholar; 42Vellard M. Sureau A. Soret J. Martinerie C. Perbal B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2511-2515Google Scholar), very little is known about the genomic structure and the expression of these factors. We report here the isolation and characterization of the 9G8 gene, the determination of the exon/intron organization, a succint analysis of the promoter, and the determination of the structure of the mRNA isoforms produced by alternative splicing and polyadenylation. A 5′ 32P-labeled 38-nt probe spanning positions +153/+190 of the cDNA (QE203) or a 32P-labeled random priming cDNA PCR product using the QO60 and QN140 oligonucleotides (+682/+902) have been used to screen a human placental genomic library in λGEM 12. Duplicate plaque lifts were prepared and probed as described (37Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). Three positive genomic clones containing inserts of 17 (λ9G8-I), 15 (λ9G8-II), and 17 kb (λ9G8-III) were isolated from 8.105 recombinant phages. cDNA 1 probe, a 330-bp BstBI-BglII fragment (265/595) and cDNA 2 probe, a 709-bp EcoRI-EcoRI fragment (262/971) were obtained from the 9G8 cDNA clone 3 (8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar); 3′-untranslated region, a 383-bp NdeI-AvaII fragment (1003/1386), was obtained from the 3.2-kb SacI subclone of λ9G8-III. The probes used for the recognition of intron 3 were obtained by an amplification by PCR of a fragment of 1064 bp containing the total intron 3 and short sequences of the surrounding exons using the primers QM95 (5′-TTTGATAGACCACCTGCC-3′) and QP101 (5′-TTCGTCCCCTGCTCCTGCTGC-3′). The resulting fragment was then cleaved with RsaI and XbaI, and two DNA bands of 450 bp (IVS 3 up) and 327 bp (IVS 3 down) were gel-purified. Each probe was labeled with [32P]dCTP by random priming. For the Southern blot analysis of the recombinant phages, 2 μg of each DNA were digested with SacI or EcoR I restriction enzymes. One μg was run on 5% polyacrylamide gel and then transferred to a nylon membrane filter Hybond N+ with 0.4 M NaOH as described by the supplier. The phage DNA digested by SacI was used to subclone, by shotgun technique, the different fragments produced. For the genomic Southern analysis, 15 μg of human placental DNA digested by EcoRI and SacI were electrophoresed on 0.8% agarose gel and transferred to a nylon membrane (Hybond N+). The blot containing the recombinant phages DNA was probed with the 38-mer QE203 oligonucleotide or the QO60/QN140 PCR product (+682/+902), labeled with 32P. The blot containing human placental DNA was probed with the 32P-labeled random priming cDNA 1 and cDNA 2 probes. The hybridization was performed at 42°C for 16 h in the hybridization solution (2 × SSC, 0.1% SDS, 1 × Denhardt's solution, 30% formamide, and 10 μg/ml salmon sperm DNA), except for the QE203 probe where formamide was omitted. The filters were then washed with 0.2 × SSC and 0.1% SDS at 45°C (QE203 probe) or at 60°C (other probes) and subjected to an overnight autoradiography. The 5′32P labeled QM94 (5′-GCAGCGCCCAGGGCTCGAGTGAC-3′) or QO14 (5′-GTAACGCGACATGATGACAGACC-3′) were annealed overnight with aliquots of 1 μg of 293 cells poly(A)+ RNA in a solution 1 × NPES (250 mM NaCl, 40 mM Pipes, pH 6.4, 5 mM EDTA, 0.2% SDS). After precipitation and washing, the extension reaction was performed with 10 units of avian myeloblastosis virus reverse transcriptase for 30 min at 42°C. The RNA was then degraded by 0.2 M NaOH at 42°C and the extension DNA product was electrophoresed on a 6% denaturing polyacrylamide gel containing 8 M urea. Artificial restriction sites HindIII and BamHI were inserted by PCR technique at positions −420 and +25, respectively. After endonuclease digestion, the PCR product was inserted in the corresponding sites of the pBLCAT3 (30Luckow B. Schtz G. Nucleic Acids Res. 1987; 15: 5490Google Scholar), giving the 9G8 “full-length” construct (9G8FL) (Fig. 6B). This region contains one StuI site at position −205 and a NotI site at position −38. Deletions mutants 9G8 ΔH/S and 9G8ΔS/N were obtained by releasing respectively HindIII-StuI and StuI-NotI fragments from the 9G8FL vector, blunting, and religating the vector backbone. Deletion mutant 9G8ΔS/-72 was obtained by releasing the StuI-NotI fragment from the 9G8FL vector and inserting the phosphorylated double-stranded 36-mer oligonucleotide corresponding to the +72/NotI fragment of the wild type promoter. The 9G8Δ-88/-34 mutant was generated by cutting the 9G8FL construct with SphI and digesting with Bal-31 for various times. DNA was then blunted with the Klenow fragment of DNA polymerase and religated. All constructs were confirmed by restriction analysis and sequencing. The pE1A, pSVCREMτ, and pSVBmyb clones that express the 293-amino acid protein of adenoviral E1A unit, CREMτ, and c-Myb, respectively, were described previously (27Leff T. Elkaim R. Goding C.R. Jalinot P. Sassone-Corsi P. Perricaudet M. Kedinger C. Chambon P. Proc. Natl. Acad. Sci. U. S. A. 1984; 81: 4381-4385Google Scholar; 11Foulkes N.S. Mellstrm B. Benusiglio E. Sassone-Corsi P. Nature. 1992; 355: 80-84Google Scholar; 40Sureau A. Soret J. Vellard M. Crochet J. Perbal B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 11683-11687Google Scholar). JEG-3 human choriocarcinoma cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and were transfected by calcium phosphate coprecipitation technique. Cells were plated at a density of 106 cells/10-cm plate and transfected with 10 μg of total plasmid DNA. 3 μg of reporter plasmid was included in each transfection sample together with 1 μg of pSVCREMτ, pE1A, or pSVBmyb expression plasmids. CAT activity was assayed as described previously (38Sassone-Corsi P. Lamph W.W. Kamps M. Verma I.M. Cell. 1988; 54: 553-560Google Scholar) and was quantified by PhosphorImager counting. The activity values correspond to the percentage of chloramphenicol modified by the chloramphenicol acetyltransferase. The human fetal multiple tissue Northern blot, containing 2 μg of poly(A)+ RNA from each tissue was obtained from Clontech (catalog number 7761-1). The membrane was prehybridized and hybridized in pre/hybridization solution (5 × SSC, 10 × Denhardt's solution, 45% formamide, 1.5% SDS, and 100 μg/ml salmon sperm DNA) at 42°C. After a 20-h hybridization with radiolabeled probe (2 × 106 cpm/ml), the blot was washed 30 min at 55°C in 0.1 × SSC and 0.1% SDS and exposed 16 h with Kodak X-Omat film or was quantified by PhosphorImager counting. We previously cloned a cDNA encoding the 9G8 SR factor (8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar). To isolate the corresponding genomic clone, a 38-nt oligonucleotide probe encompassing an amino acid sequence of the RBD not present in the other SR factors was used to screen a placental human genomic library in λGEM12. We obtained two genomic clones λ9G8-I and −II containing inserts of 17 and 15 kb, respectively (see also 8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar). However, preliminary analysis indicated that they do not cover the entire open reading frame of the 9G8 mRNA. Therefore, a 211-bp PCR product, from positions 692 to 902 of the 9G8 mRNA, covering the C-terminal region of the 9G8 factor, allowed the isolation of another 17-kb clone (λ9G8-III). Analysis of the three clones by restriction endonuclease mapping and Southern blotting using the specific probes mentioned above revealed that the inserts overlap and together cover the entire open reading frame of the 9G8 gene (Fig. 1). The 5.5-kb SacI fragment at the 3′ terminus of the clone I insert, as well as the 2.6-kb internal fragment and 3.2-kb fragment at the 3′ terminus of the clone III insert were subcloned into pBluescript SK+ and used for further characterization and sequencing analysis. Complete exon-intron organization of the 9G8 gene was determined by sequencing the totality of the gene. Using the previously cloned 9G8 cDNA sequence as the reference mRNA sequence, we have determined that the 9G8 gene is 7745 nucleotides long and contains 8 exons and 7 introns (Table 1). The exons range from 36 (exon 7) to 1572 bp (last exon), and the intron sizes vary from 308 to 1298 bp (Table 1). From the sizes of exons and introns as well as the splice site sequences which fulfill the GT-AG rule (4Breathnach R. Chambon P. Annu. Rev. Biochem. 1981; 50: 349-383Google Scholar), we deduced that the 9G8 gene exhibits features typical of many eukaryotic genes. The translated sequence of 9G8 is highly cut up as the open reading frame is distributed over the 8 exons. Interestingly, the RBD is contained in exon 2, and the exon 3 encodes for the zinc knuckle motif (8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar). In contrast, the SR domain which is 110 amino acids in size covers exons 4-8. A study of the DNA sequence downstream the stop codon reveals two polyadenylation signals at 610 and 1511 bp downstream of the stop codon.TABLE I Open table in a new tab Southern blot analysis of human genomic DNA digested by EcoR I or SacI was performed using the two cDNA probes cDNA 1 (positions 265/595) covering exons 2 and 3 and cDNA 2 (positions 262/971) covering exon 2 to exon 8 (Fig. 2). The short cDNA probe revealed a single band after digestion with either EcoRI (2.7 kb) or SacI (7.5 kb), in agreement with what was expected from structure of genomic clones (Fig. 1). In contrast, the extended probe (cDNA 2) detected an additional fragment in the EcoRI (~8 kb) and SacI (2.5 kb) restriction digests (Fig. 2), consistent with the structure of the λ9G8-III clone shown in Fig. 1. Thus, these results confirm the structural organization of the 9G8 gene. They indicate also that 9G8 is encoded by a single copy gene and that no pseudogenes are present in the genome. The chromosomal location of the 9G8 gene was determined by in situ hybridization using the cDNA 2 probe. Analysis of 100 metaphase cells revealed a total of 237 silver grains on chromosomes, and 51 of these (21.5%) were located on chromosome 2. Analysis of the grain distribution indicated that 36 out of 51 (70.6%) of these mapped to the p22-21 region on the short arm of chromosome 2, with an intense localization in the p21 band (Fig. 3). This result allows to unequivocally assign the 9G8 gene to the p22-21 region of human chromosome 2.Figure 3:Idiogram of distribution of signals on chromosome 2. In situ hybridization of human metaphase chromosomal spreads were performed as described (31Mattei M.G. Bruce B. Karsenty G. Genomics. 1993; 16: 786-788Google Scholar). 70.6% of grains located on chromosome 2 mapped to the p22-21 region of short arm, with a maximum in the p21 band.View Large Image Figure ViewerDownload (PPT) To determine the transcription initiation site, we performed a primer extension analysis (Fig. 4). Primer extension on poly(A)+ RNA isolated from 293 cells, using two 23-mer oligonucleotides QM94 (upstream the AUG codon) and QO14 (encompassing the AUG codon) resulted in the synthesis of cDNAs of 72-70 residues (Fig. 4) or 117-115 residues (not shown), respectively. Comparing the extension products with a sequence ladder generated by extending the same primers from a plasmid containing incomplete 9G8 cDNA localized the initiation site at a G residue, downstream from the CT-rich sequence CTCTTCCTC/G + 1. It is not known if the cDNA beginning 2 residues downstream of the longer cDNA (Fig. 4) corresponds to a true initiation site at an A residue or to a premature stop of the primer extension. Analysis of the sequence of the 5′-flanking region of the 9G8 gene contained in the 5.5-kb SacI subfragment reveals a high GC content (57%) and several promoter elements (Fig. 5). A TATA motif (TATATAA) is present at position −29, and three potential SP1 binding sites (GGCGGG) are found at positions −87, −148, and −224. Computer search reveals also putative regulatory elements (Locker and Buzard, 1990; Faisst and Meyer, 1992). Two sequence motifs for liver-specific factors A1 (LFA1), TGAACC and TGACCC, are present at positions −149 and −345, and one possible AP-2 motif (GCCTGGg), which deverges from the AP-2 consensus by one nucleotide, is located at position −298. In addition, an ATGACGcA sequence, which exhibits a good match with the consensus ATF site is present at position −59 and overlaps a TGACGcat sequence, with a significant homology to the CRE motif. Sequences identical to the core consensus for Ets (GGAAPu) are also present at positions −266, −261, −208, and −115. Finally, minimal consensus sites for Myb (AACNG) are located at positions −96, −121, −239, and −375. To characterize the upstream region of the 9G8 gene, a DNA fragment spanning from −414 to +26 (clone 9G8FL), as well as fragments containing various deletions, were fused to the bacterial CAT reporter gene in the pBLCAT3 plasmid (Fig. 6A). These constructs were transiently expressed into JEG-3 cells, which express 9G8 mRNA at levels similar to those of HeLa cells and cellular extracts were assayed for CAT activity 36 h after transfection. In Fig. 6B, we show that the upstream region was able to induce significant CAT activity. However, a deletion of the −202 to −36 region (clone 9G8ΔS/N), which leaves the TATA motif intact, results in an 8-fold reduction of the CAT activity (compare the first and the fifth bars). A similar reduction was observed with the almost complete deletion of the 9G8 upstream region from −414 to −36 (not shown). Moreover, two smaller deletions −202/-72 (clone 9G8ΔS/-72) and −88/-34 (clone 9G8Δ-88/-34), which cover the −202 to −36 deletion (Fig. 6A), resulted in both cases in a ~2.5-fold reduction compared with the wild type construct. In contrast, deletion of the region −414 to −203 (clone 9G8ΔH/S) induced a 2-fold stimulation of the promoter activity (compare the first and the second bars), indicating that the most upstream sequences do not contain strong positive acting elements and that some negative elements may be present. Thus, this deletional analysis shows that the upstream region of the 9G8 gene is effective in activating a CAT reporter gene and that the positive acting elements are mainly concentrated all along the −202 to −34 region, immediately upstream of the TATA box. The presence of many putative regulatory elements served as a guide for testing, in a preliminary manner, several trans-acting factors. The effect of CREMτ+PKA, Myb, and E1A factors, which are known to activate many cellular genes, were analyzed by cotransfection experiments. We found that each factor stimulates the transcription around three fold (not shown), suggesting that the 9G8 promoter is able to respond to different trans-acting factors. Northern blot analysis of human fetal poly(A)+ RNA isolated from the brain, liver, kidney, and lung (Clontech) was carried out using a 330-nt cDNA probe, covering exons 2 and 3 of the 9G8 RNA (cDNA 1). We detect five mRNAs of approximately 1.3, 2.0, 2.4, 2.6, and 3.8 kb in size, respectively, the 2.4- and 2.6-kb being very close (Fig. 7, lanes 1-4). We observed that the same five transcripts are present in adult poly(A)+ RNAs (Multiple Tissue Northern blot 1, Clontech), suggesting that none of the isoforms is specific for a developmental stage (not shown). To further analyze the different isoforms, the blot was hybridized with DNA probes specific for (i) the 3′-noncoding region between the two potential poly(A) signals (3′ UTR; Fig. 7, lanes 5-8). We have used this probe because the distance observed between the two potential polyadenylation signals (~1 kb) could account for different sizes of mRNA observed. (ii) The intron 3, since further characterization of the cDNA clones for 9G8 isolated previously (8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar) indicated that two out of nine cDNA clones contained this intron (not shown). We observed that the 3.8- and 2.6-kb mRNA species use the distal poly(A) signal (Fig. 7, second panel), and that only the 3.8-, 2.4-, and 2.0-kb species contain intron 3 sequences (not shown). Nevertheless, the 2.0-kb species is too short to contain the entire intron 3. Looking for consensus signals within this intron, we have found one potential 3′ splice site in its middle. By using DNA probes specific for the sequences downstream (Fig. 7, lanes 9-12) and upstream (Fig. 7, lanes 13-16) of this splice site, we show that the 2.0-kb mRNA is generated by the use of this alternative 3′ splice site and that the 3.8- and 2.4-kb mRNA contain the totality of intron 3 sequences. The putative structure of all the mRNA isoforms is given in Fig. 7. A quantitative analysis of the relative abundance of the different species in various fetal tissues indicates that the 1.3-kb species, which encodes the whole 9G8 factor, is highly predominant in the liver, but it appears as a minor isoform in the kidney. In contrast, the 2.4-kb isoform, which contains the entire intron 3, is the predominant species in kidney. A quantitative estimation of the 9G8 mRNA isoforms indicates that the relative ratio of the intron 3 minus transcripts (1.3- and 2.6-kb species) to the intron 3 plus transcripts (2.0-, 2.4-, and 3.8-kb species) varies from about 1 to 5 between the kidney and the liver, respectively. The 9G8 gene is divided into eight exons and seven introns, and the coding sequence is highly divided, since it starts in exon 1 and stops in exon 8. Thus, its exon/intron organization is very different from those of the two other SR factors SC35/PR264 (39Sureau A. Perbal B. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 932-936Google Scholar) or RBP1 (20Kim Y.-J. Zuo P. Manley J.L. Baker B.S. Genes & Dev. 1992; 6: 2569-2579Google Scholar), since in these genes, the coding sequence is contained only in two exons. Previous comparison of amino acid sequences of RBD has shown that 9G8 presents a good homology with SRp20 and RBP1 (8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar), suggesting that several SR factors may originate from a common ancestral gene, as already proposed (2Birney A. Sanjay K. Krainer A.R. Nucleic Acids Res. 1993; 21: 5803-5816Google Scholar). However, the very different organization of 9G8 and RBP1 genes indicates that profound modifications have occurred after ancient gene duplication. It has been proposed that the intron sequences frequently demarcate important functional or structural domains of proteins. We observe indeed that the RNA binding domain of the 9G8 factor covers precisely exon 2 sequences, whereas the middle region containing the specific CCHC zinc knuckle is located in the third exon. Finally, the SR domain of 9G8 is distributed in the exons 4-8, but a specific region of this domain, not found in the other SR factors, is encoded precisely in exon 5. This region contains four conserved repetitions of the consensus RRSRSXSX (8Cavaloc Y. Popielarz M. Fuchs J.-P. Gattoni R. Stvenin J. EMBO J. 1994; 13: 2639-2649Google Scholar) and might originate from intragenic recombination occurred during evolution. We have shown that the 9G8 gene is located in human on the chromosome 2p22-21. This region is the site of several known genes, including T cell leukemia virus enhancer factor (HTLF) (28Li C. Lusis A.J. Sparkes R. Tran S.-M. Gaynor R. Genomics. 1992; 13: 658-664Google Scholar), and translocations of chromosome 2p22-16 with chromosome 11p23 have been reported in human leukemia (3Bloomfield C.D. de la Chappelle A. Semin. Oncol. 1987; 14: 372-383Google Scholar). Examination of the 400-bp region upstream of the 9G8 gene shows several interesting features. The G + C content of 57% and the CpG content of 7% are indicators that the promoter region of the 9G8 gene is in a CpG island (25Larsen F. Gundersen G. Lopez R. Prydz H. Genomics. 1992; 13: 1095-1107Google Scholar). However, the presence of a TATA box and many potential regulatory elements does not allow us to classify this gene as a typical housekeeping gene (Locker and Buzard, 1990; Faisst and Meyer, 1992). The promoter of the 9G8 gene has been shown to be functional in JEG-3 cells and to respond to several trans-acting factors. In this respect, the 9G8 promoter resembles the SC35/PR264 gene which contains several Myb-responsive elements (42Vellard M. Sureau A. Soret J. Martinerie C. Perbal B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2511-2515Google Scholar). In fact, 9G8 expression, similar to that of the other SR factors, is likely ubiquitous, but it has been shown that the SR factors recognized by mAb104 (SRp20, SRp30, SRp40, SRp55, and SRp75) are expressed at different levels in various calf tissues (45Zahler A.M. Neugebauer K.M. Lane W.S. Roth M.B. Science. 1993; 260: 219-222Google Scholar). We show in this paper that the expression of the 9G8 mRNA is the target of different regulations such as alternative splicing and alternative polyadenylation, leading to five well detectable species from 1.3 to 3.8 kb. One interesting feature is the retention of the entire or a part of the intron 3, because it leads to the translation of a truncated form of the 9G8 protein devoid of the SR domain by introducing a stop codon downstream of the exon 3/intron 3 junction (Fig. 8). In fact, although there are no important variations in the total amounts of 9G8 transcripts within the tested fetal tissues (Fig. 7), some important changes within the distribution of each species occur. The existence of the alternative splicing of intron 3, most likely due to the existence of a suboptimal 3′ splice site at the end of this intron (see the sequence in Fig. 8) seems puzzling if it is thought that maximal levels of SR factors are required for an efficient splicing machinery. However, we have to take into account that levels of SR factors are variable (45Zahler A.M. Neugebauer K.M. Lane W.S. Roth M.B. Science. 1993; 260: 219-222Google Scholar) and may participate to the modulation of alternative splicing, as proposed by these authors. Moreover, our data raise the interesting possibility that the splicing of intron 3 might be submitted to regulation, because its 3′ splice site is similar to other weak 3′ splice sites, such as the female-specific 3′ splice site of double-sex pre-mRNA (35Ryner L.C. Baker B.S. Genes & Dev. 1991; 5: 2071-2085Google Scholar; 19Hedley M.L. Maniatis T. Cell. 1991; 65: 579-586Google Scholar) or the fibronectin ED1 exon (7Caputi M. Casari M. Guenzi S. Tagliabue R. Sidoli A. Melo C.A. Baralle F.E. Nucleic Acids Res. 1994; 22: 1018-1022Google Scholar). In this respect, purine-rich motifs of the form CAGGAGGAA, CAGCAGGAG, and CAGGGACGAAG, located downstream within the exon 4 of this gene, resemble cis-acting motifs found in the M2 exon of IgM gene (41Tanaka K. Watakabe A. Shimura Y. Mol. Cell. Biol. 1994; 14: 1347-1354Google Scholar), troponin (43Xu R. Teng J. Cooper T.A. Mol. Cell. Biol. 1993; 13: 3660-3674Google Scholar), fibronectin (26Lavigueur A. La Branche H. Kornblihtt A.R. Chabot B. Genes & Dev. 1993; 7: 2405-2417Google Scholar; 7Caputi M. Casari M. Guenzi S. Tagliabue R. Sidoli A. Melo C.A. Baralle F.E. Nucleic Acids Res. 1994; 22: 1018-1022Google Scholar), or bovine growth hormon gene (9Dirksen W.P. Hampson R.K. Sun Q. Rottman F.M. J. Biol. Chem. 1994; 269: 6431-6436Google Scholar). They may be important for the excision of the whole intron 3. In fact, a mRNA variant containing a retained intron had also been previously described for the ASF/SF2 protein (17Ge H. Zuo P. Manley J.L. Cell. 1991; 66: 373-382Google Scholar), but the amount of this species was small (<5%). Interestingly, the ΔSR splice variant of ASF/SF2 conserves all its ability to modulate the alternative splicing, but loses its characteristics of constitutive splicing factor (47Zuo P. Manley J.L. EMBO J. 1993; 12: 4727-4737Google Scholar; 5Caceres J.F. Krainer A.R. EMBO J. 1993; 12: 4715-4726Google Scholar). Nevertheless, it is still unknown whether this type of truncated SR factors is involved in the regulation of alternative splicing pathways in vivo. We also show that the 9G8 pre-mRNA is alternatively processed in its 3′-untranslated region. Looking at polyadenylation signals within the genomic primary sequence, we find one ATTAAA motif at position 1433 and one AATAAA motif at position 2334 that are used in vivo. It had been observed that SC35/PR264 pre-mRNA was submitted to an alternative polyadenylation coupled to alternative splicing within its 3′-untranslated region (39Sureau A. Perbal B. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 932-936Google Scholar). Differences in the stability of the SC35/PR264 mRNA species have been demonstrated in agreement with studies showing that sequences contained in the 3′-untranslated region are involved in the stability of the mRNA or in the control of translation (36Sachs A.B. Cell. 1993; 74: 413-421Google Scholar). In conclusion, we have identified splice variants of the 9G8 transcript that may allow the synthesis of significant and variable amounts of 9G8 SR factor deleted of the SR domain. It will be interesting to determine whether the differential expression of the 9G8 factor may be involved in the modulation of alternative splicing in vivo. We are grateful to Dr. I. Davidson for critical reading of the manuscript. We thank G. Hildwein for excellent technical assistance, the cell culture group for growing cells, the photographic staff for preparation of the manuscript, B. Chatton, N. Foulkes, and J. Soret for the gift of clones." @default.
• W1520556951 created "2016-06-24" @default.
• W1520556951 creator A5017378450 @default.
• W1520556951 creator A5049791919 @default.
• W1520556951 creator A5071625833 @default.
• W1520556951 creator A5073626645 @default.
• W1520556951 creator A5090334058 @default.
• W1520556951 date "1995-07-01" @default.
• W1520556951 modified "2023-10-16" @default.
• W1520556951 title "The Gene Encoding Human Splicing Factor 9G8" @default.
• W1520556951 cites W101837116 @default.
• W1520556951 cites W1512969508 @default.
• W1520556951 cites W1969600436 @default.
• W1520556951 cites W1971421846 @default.
• W1520556951 cites W1971956957 @default.
• W1520556951 cites W1982559739 @default.
• W1520556951 cites W1982598173 @default.
• W1520556951 cites W1983800032 @default.
• W1520556951 cites W1984680322 @default.
• W1520556951 cites W1990989603 @default.
• W1520556951 cites W1995012997 @default.
• W1520556951 cites W2002986711 @default.
• W1520556951 cites W2013002081 @default.
• W1520556951 cites W2013955971 @default.
• W1520556951 cites W2014791263 @default.
• W1520556951 cites W2019802229 @default.
• W1520556951 cites W2021334674 @default.
• W1520556951 cites W2023199595 @default.
• W1520556951 cites W2031156037 @default.
• W1520556951 cites W2036045539 @default.
• W1520556951 cites W2040478918 @default.
• W1520556951 cites W2041847911 @default.
• W1520556951 cites W2057051846 @default.
• W1520556951 cites W2057479447 @default.
• W1520556951 cites W2069584591 @default.
• W1520556951 cites W2071336621 @default.
• W1520556951 cites W2076530408 @default.
• W1520556951 cites W2076849957 @default.
• W1520556951 cites W2085324214 @default.
• W1520556951 cites W2085368553 @default.
• W1520556951 cites W2088431556 @default.
• W1520556951 cites W2089398626 @default.
• W1520556951 cites W2091955128 @default.
• W1520556951 cites W2092451011 @default.
• W1520556951 cites W2094419922 @default.
• W1520556951 cites W2116884417 @default.
• W1520556951 cites W2117976997 @default.
• W1520556951 cites W2132978049 @default.
• W1520556951 cites W2164032670 @default.
• W1520556951 cites W235098460 @default.
• W1520556951 cites W268783301 @default.
• W1520556951 doi "https://doi.org/10.1074/jbc.270.30.17830" @default.
• W1520556951 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/7629084" @default.
• W1520556951 hasPublicationYear "1995" @default.
• W1520556951 type Work @default.
• W1520556951 sameAs 1520556951 @default.
• W1520556951 citedByCount "41" @default.
• W1520556951 countsByYear W15205569512013 @default.
• W1520556951 countsByYear W15205569512014 @default.
• W1520556951 countsByYear W15205569512018 @default.
• W1520556951 countsByYear W15205569512020 @default.
• W1520556951 countsByYear W15205569512021 @default.
• W1520556951 countsByYear W15205569512022 @default.
• W1520556951 countsByYear W15205569512023 @default.
• W1520556951 crossrefType "journal-article" @default.
• W1520556951 hasAuthorship W1520556951A5017378450 @default.
• W1520556951 hasAuthorship W1520556951A5049791919 @default.
• W1520556951 hasAuthorship W1520556951A5071625833 @default.
• W1520556951 hasAuthorship W1520556951A5073626645 @default.
• W1520556951 hasAuthorship W1520556951A5090334058 @default.
• W1520556951 hasBestOaLocation W15205569511 @default.
• W1520556951 hasConcept C104317684 @default.
• W1520556951 hasConcept C125411270 @default.
• W1520556951 hasConcept C141231307 @default.
• W1520556951 hasConcept C169760540 @default.
• W1520556951 hasConcept C194583182 @default.
• W1520556951 hasConcept C197077220 @default.
• W1520556951 hasConcept C199360897 @default.
• W1520556951 hasConcept C2781039887 @default.
• W1520556951 hasConcept C36823959 @default.
• W1520556951 hasConcept C41008148 @default.
• W1520556951 hasConcept C54355233 @default.
• W1520556951 hasConcept C54458228 @default.
• W1520556951 hasConcept C67705224 @default.
• W1520556951 hasConcept C70721500 @default.
• W1520556951 hasConcept C73758832 @default.
• W1520556951 hasConcept C86803240 @default.
• W1520556951 hasConceptScore W1520556951C104317684 @default.
• W1520556951 hasConceptScore W1520556951C125411270 @default.
• W1520556951 hasConceptScore W1520556951C141231307 @default.
• W1520556951 hasConceptScore W1520556951C169760540 @default.
• W1520556951 hasConceptScore W1520556951C194583182 @default.
• W1520556951 hasConceptScore W1520556951C197077220 @default.
• W1520556951 hasConceptScore W1520556951C199360897 @default.
• W1520556951 hasConceptScore W1520556951C2781039887 @default.
• W1520556951 hasConceptScore W1520556951C36823959 @default.
• W1520556951 hasConceptScore W1520556951C41008148 @default.
• W1520556951 hasConceptScore W1520556951C54355233 @default.

• next