- Email: [email protected]
Isolation, cloning, and expression of a new murine zinc finger encoding gene1
Biochimica et Biophysica Acta 1447 (1999) 278^283 www.elsevier.com/locate/bba
Short sequence-paper
Isolation, cloning, and expression of a new murine zinc ¢nger encoding gene1 Jean-Franc°ois Prost, Didier Ne©gre, Franc°oise Cornet-Javaux, Jean-Claude Cortay, Alain J. Cozzone, Daniel Herbage, Fre¨de¨ric Mallein-Gerin * Institut de Biologie et Chimie des Prote¨ines, CNRS UPR 412, 7 passage du Vercors 69367 Lyon Cedex 07, France Received 17 May 1999; received in revised form 13 August 1999; accepted 16 August 1999
Abstract With the aim of identifying genes involved in cartilage differentiation, we have used a subtractive hybridization strategy with cDNAs from a chondrocytic cell line (MC615) and mRNAs from a mesenchymal precursor cell line (10T1/2). We have isolated a cDNA clone representing a novel mouse gene. The predicted 368-amino acid protein, designated ZF-12, contains four C2 H2 -type zinc finger motifs and one region homologous to the LeR domain, a finger-associated structural domain. ZF-12 mRNAs are expressed during embryonic development and in different organs in adult, including rib cartilage. These data suggest that ZF-12 might play an important role not only in cartilage differentiation, but also in basic cellular processes. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: Mouse; Cartilage; cDNA sequence; Zinc ¢nger motif; LeR domain
Cloning of genes that control chondrocyte di¡erentiation has been hampered because chondrocytes are unstable in culture and dedi¡erentiate when cultured at low density for a long time or on a substratum favoring cell spreading [1]. We previously established a mouse chondrocyte cell line, termed MC615, that is able to express markers characteristic of cartilage in vivo, such as collagen types II, IX, and XI, the link protein and the cartilage aggrecan [2], and to form cartilage nodules when cultured in suspension over agarose [3]. With the aim of identifying genes involved in cartilage di¡erentiation, we have now used a subtractive hybridization strategy with the * Corresponding author. Fax: +33-4-7272-2602; E-mail: [email protected] 1 The nucleotide sequence of ZF-12 will appear in GenBank nucleotide sequence database (accession number AF149093).
MC615 cell line and the mouse embryonic mesenchymal cell line, 10T1/2. 10T1/2 is a multipotential cell line that can be induced, upon treatment with azacytidine, to di¡erentiate into myoblasts, adipocytes and, to a lesser extent, chondrocytes [4]. Therefore, this cell line represents a mesodermal precursor, and a subtractive hybridization screening has been successfully applied with parental 10T1/2 cells and myoblast cells to isolate the cDNA of the ¢rst basic helix-loop-helix myogenic regulatory gene, MyoD [5]. Cytoplasmic RNA from MC615 or 10T1/2 cells grown in monolayer was isolated and poly(A ) RNA was selected using oligo(dT)-cellulose columns. Poly(A ) MC615 RNA was reverse-transcribed with Superscript reverse transcriptase (Gibco-BRL), in presence of [K-32 P]dCTP (800 Ci/mmol). In order to prepare a probe enriched with cDNA sequences
0167-4781 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 9 9 ) 0 0 1 5 7 - 8
BBAEXP 91304 19-10-99
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
Fig. 1. Schematic representation of cDNA clones of ZF-12. ZF12L is the cDNA cloned from the MC615 library. ZF-12R was obtained after RT-PCR. ZF-12L and ZF-12R overlap. The open reading frame is 1104 bp long. The four zinc ¢nger motifs and the LeR domain are represented. The HindIII fragment in the 3P-untranslated region corresponds to the probe used for Northern blotting analysis. Also are indicated the positions of the ZF-12S and ZF-12AS oligonucleotides used for ZF-12 gene expression analysis by RT-PCR in cartilage.
characteristic of MC615 expression, a 10-fold excess of poly(A ) 10T1/2 RNA was hybridized for 28 h at 68³C in 120 mM sodium phosphate, pH 6.8, 820 mM NaCl, 10 mM EDTA, pH 8, and in presence of 10 Wg/ml polyU (15 mer) with the MC615 cDNA molecules. The single-stranded subtracted cDNA probe was eluted at 60³C from an hydroxyapatite column with 120 mM sodium phosphate, pH 2, 0.1% SDS. This subtracted cDNA probe was labeled with [K-32 P]dCTP (3000 Ci/mmol) by random priming and used to screen a MC615 cDNA library constructed in UNI-ZAPXR vector (Stratagene). Positive plaques were submitted to in vivo excision and recircularization into pBluescript phagemid (Stratagene) carrying the cloned DNA insert, according to the manufacturer's instructions. A positive clone termed ZF-12L, carrying a 2-kbp cDNA insert (Fig. 1), was further characterized and its description is reported here. In order to extend the cDNA sequence corresponding to ZF-12L, we carried out 5P rapid ampli¢cation of cDNA ends (RACE) using the 5P-RACE kit (version 2, Life Technologies). The ¢rst strand cDNA was synthesized from 1 Wg mouse liver poly(A ) RNA with the gene-speci¢c primer GSP1 (5P-GTGTTGTTTCTTTCTAGAAGGATTT-3P, positions 1107^1131 in Fig. 2). Following cDNA synthesis, the ¢rst strand product was puri¢ed from non-incorporated dNTPs and GSP1 by spin-column chromatography, and tailed at its 3P-end with termi-
279
nal deoxynucleotidyl transferase and dCTP. This tailed cDNA was then ampli¢ed by PCR using the GSP1 primer and the Abridged Anchor Primer AAP (5P-GGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG-3P). Products from the ¢rst PCR were then ampli¢ed by a second PCR reaction in the presence of the nested gene-speci¢c primer, GSP2 (5P-TTTCAAACCTACCCTTGGGAAA-3P, positions 1078^1099 in Fig. 2) and the Abridged Universal Ampli¢cation Primer AUAP (5P-GGCCACGCGTCGACTAGTAC-3P). By following this procedure, one major PCR product of 1100 bp was resolved on a 5% non-denaturing polyacrylamide gel, then eluted and inserted by TA cloning into plasmid pCR2.1-TOPO (Invitrogen). Sequencing was performed using the dideoxynucleotide chain termination procedure on double-stranded DNA, in the laboratory, or by automated DNA sequencing (Genome Express and ESGS from ACT Gene). The DNA sequences were analyzed by the DNAid computer program [6]. Alignments and identity studies were performed via the IBCP site server2 with the Antheprot [7] program. The two cDNAs isolated from the MC615 cDNA library (ZF-12L, from position 1033 to the 3P-end of the poly-A tail shown in Fig. 2) and from the RACE (ZF-12R, positions 1^1099 in Fig. 2) are diagrammed in Fig. 1. They represent a sequence encoding a putative open reading frame of 1104 bp which is followed by a polyadenylation signal and poly-A tail (Fig. 2). Conceptual translation of this open reading frame yielded a protein of 368 amino acids with four zinc ¢ngers of the C2 H2 type. The C2 H2 motif represents a DNA-binding structure ¢rst discovered in Xenopus RNA polymerase III transcription factor TFIIIA [8,9], and also found in other transcription factors, such as human SP1 [10] and Drosophila Kru«ppel [11]. The zinc ¢nger sequences indicated by a box in Fig. 1 are located in the carboxy terminal region of the predicted protein. Alignment of the zinc ¢nger motifs shows that the invariant cysteines and histidines are clearly conserved (Fig. 3A). Each ¢nger motif conforms closely to the consensus sequence CX2 CX3 FX5 LX2 HX3 H [12] and the four motifs are
2
The IBCP site server is accessible via the World Wide Web (http://www.ibcp.fr).
BBAEXP 91304 19-10-99
280
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
BBAEXP 91304 19-10-99
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
281
Fig. 2. Nucleotide and deduced amino acid sequences of ZF-12. The nucleotide sequence is numbered from the 5P-end of ZF-12R clone, and the amino acid sequence is numbered from the putative translation initiation site. ZF-12L and ZF-12R overlap between nucleotides 1033 and 1099. Below ZF-12 amino acid sequence are indicated the only amino acids derived from a human cDNA sequence (sp accession number 014754) that di¡er from ZF-12. The poly(A ) signal is underlined. Zinc ¢nger encoding sequences are indicated by boxes and numbered. The LeR domain is underlined with a thick line. The termination codon is indicated by . The 3P-untranslated end of ZF-12 cDNA clone which is about 1550 bp long, was not completely sequenced, and is partially represented by a dotted line.
6
connected by highly conserved H/C links, whose consensus sequence is TGEKPYX [13]. Analysis of the N-terminal portion of ZF-12 revealed that the protein contains the leucine-rich region LeR [14], which is highly conserved in the Nterminal portion of several C2 H2 zinc ¢nger proteins, such as SRE-ZBP [15], Zfp-38 [16], RU49 [17], p18 [14], p20 [14], ZNF174 [18], LD5-1 [19] and MZF-2 [20]. This element is 77 amino acids long (Fig. 3B). Although the function of the LeR domain has not yet been determined, the high content of leucine res-
idues with the possible folding of K-helices and the striking conservation of this domain among di¡erent members of zinc ¢nger proteins suggest that it may mediate protein^protein interaction with components of the transcriptional machinery. Moreover, analysis of the predicted amino acid sequence of ZF-12 showed 93% identity with a human sequence listed with accession numbers O14754 in the Swissprot database and P17028 (sequence ZN24) in the Genbank database [21]. Therefore, ZF-12 likely represents the murine counterpart of this human cDNA (Fig. 2).
Fig. 3. (A) Zinc ¢nger domains and inter domains of ZF-12. The four zinc ¢nger motifs of the conceptual translation of ZF-12 cDNA are shown aligned beneath the consensus sequence of mouse C2 H2 ¢ngers, as previously described [12]. Upper case letters indicate the most highly conserved residues, and the lower case letters indicate more variable positions. In positions where either one of two amino acids is likely to be present, these two amino acids are indicated. Residues identical to those of the consensus motif are indicated with a dash. (B) ZF-12 amino acid sequence of the LeR domain aligned with the consensus sequence for the LeR domain found in other zinc ¢nger proteins [20]. In positions where two amino acids are likely to be present in the consensus sequence, both are indicated on the top. The only amino acids of ZF-12 that di¡er from the LeR consensus are indicated at the bottom.
BBAEXP 91304 19-10-99
282
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
Fig. 4. ZF-12 gene expression analysis. (A) Northern blot hybridization of total RNAs isolated from MC615 chondrocytes or 10T1/2 cells with the random prime labeled HindIII cDNA fragment, corresponding to the 3P-untranslated region of ZF-12. A 20-Wg amount of each RNA was loaded per lane. Exposure: 1 week. (B) RT-PCR analysis on total RNA extracted from newborn mouse rib cartilage. After ultracentrifugation, total RNA was digested with RNAse-free DNAse, and reverse transcribed with oligodT primers. Synthesized cDNA was ampli¢ed with ZF-12S and ZF-12AS primers for 35 cycles, with denaturation at 94³C (1 min), annealing at 57³C (1 min) and extension at 72³C (1 min). The PCR reaction generates a 315-bp product, visualized by ethidium bromide staining of a 2% agarose gel. To eliminate the possibility that residual genomic DNA might be contributing to the ampli¢ed product, RNA preparation was also analyzed in the absence of RT and yielded no product (not shown). The positions of 100-bp DNA ladder (Promega) are indicated on the left. (C) Northern blot (Clontech), containing 2 Wg of poly(A ) RNA per lane from di¡erent mouse tissues. The blot was hybridized with the random prime labeled HindIII cDNA fragment, corresponding to the 3P-untranslated region of ZF-12. Exposure: 1 week. (D) Northern blot (Clontech) containing 2 Wg of poly(A ) RNA per lane from mouse embryos at di¡erent developmental stages. The blot was hybridized with the random prime labeled HindIII cDNA fragment, corresponding to the 3P-untranslated region of ZF-12. The positions of RNA size markers are shown on the left. Exposure: 1 week.
This observation is in agreement with the fact that the amino acid sequence of the zinc ¢nger transcription factors is usually well conserved between human and mouse [22,23]. Northern blot analysis using an HindIII cDNA fragment as a probe located in the 3P-non-translated region (see Fig. 1) revealed that ZF-12 gene expression was higher in MC615 cells than in 10 T1/2 cells (Fig. 4A), as expected by using our subtractive hybridization strategy. RT-PCR analysis on total RNA isolated from newborn-mouse rib cartilage using the sense ZF-12S primer (5P-CTTGATGACCCTGGTCAGCC-3P, positions 778^798 in Fig. 2) and the antisense ZF-12AS primer (5P-CCTACCCTTGGGAAAACAGG-3P, positions 1073^1092 in Fig. 2) located between the LeR domain and the zinc ¢nger
domains, indicated that ZF-12 is expressed in vivo in skeletal tissues (Fig. 4B). Because rib cartilage undergoes endochondral ossi¢cation after birth, it will be important to analyze ZF-12 gene expression in developmental skeletal tissues by in situ hybridization, to determine whether ZF-12 is expressed by chondrocytes or osteoblasts, or both. To characterize further the pattern of ZF-12 gene expression in mouse, the HindIII cDNA fragment was hybridized to mRNAs prepared from a panel of adult tissues (Clontech mouse multiple tissue blot). ZF-12 transcripts were detected in heart, brain, liver, skeletal muscle, kidney and testis, and to a very low level, in spleen and lung (Fig. 4C). This signal could be further resolved as transcripts of 3.7 and 6.4 kb, with a stronger signal for the shorter transcript. ZF-12 expression was also
BBAEXP 91304 19-10-99
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
examined by Northern blot analysis of mRNA from total embryo (Clontech mouse embryo multiple tissue blot). ZF-12 expression was hardly detected at day 7, peaked at day 11 and remained constant at the di¡erent stages examined subsequently (Fig. 4C), suggesting that ZF-12 is developmentally regulated. Each transcript appeared to be of the same relative abundance in the di¡erent samples. It is possible that the 3.7-kb and the faint 6.4-kb transcripts could arise as products of alternative processing, from alternative promoters of the same gene, or as products of a gene or genes related to ZF-12. Isolation and characterization of genomic clones should help to achieve this determination. In any case, the pattern of tissue expression of ZF-12 suggests that this new member of the zinc ¢nger proteins might have a biological role not restricted to skeletal di¡erentiation. In conclusion, the main result of this study is that the presumptive protein product of ZF-12 contains zinc ¢nger domains and a LeR motif, and is thus considered a member of a class of proteins capable of binding speci¢c DNA sequences and regulating cell growth, di¡erentiation or development [24]. It would be interesting to determine and characterize the genomic target, or targets, of ZF-12 to elucidate the function of this particular zinc ¢nger gene. In addition, its expression in murine cell lines, such as MC615 and 10T1/2, should provide a valuable tool to study its own regulation. We thank Dr. Catherine Tomasetto for her precious help during the subtractive hybridization procedure, and Dr. Marie-Christine Rio and Dr. Paul Basset for their kind hospitality in their laboratory (Institut de Ge¨ne¨tique et de Biologie Mole¨culaire et Cellulaire CNRS-INSERM U184, Illkirch, France). F.M.G. was supported by the `Association pour la Recherche sur le Cancer' (ARC).
283
[2] F. Mallein-Gerin, B.R. Olsen, Proc. Natl. Acad. Sci. USA 90 (1993) 3289^3293. [3] F. Mallein-Gerin, F. Ruggiero, T.M. Quinn, F. Bard, A.J. Grodzinsky, B.R. Olsen, M. van der Rest, Exp. Cell Res. 219 (1995) 257^265. [4] S.M. Taylor, P.A. Jones, Cell 17 (1979) 771^779. [5] R.L. Davis, H. Weintraub, A.B. Lassar, Cell 51 (1987) 987^ 1000. [6] F. Dardel, P. Bensoussan, Comput. Appl. Biosci. 4 (1988) 483^486. [7] G. Dele¨age, C. Geourjon, Comput. Appl. Biosci. 9 (1993) 197^199. [8] J. Miller, A.D. McLachlan, A. Klug, EMBO J. 4 (1985) 1609^1614. [9] R.S. Brown, C. Sander, P. Argos, FEBS Lett. 186 (1985) 271^274. [10] J.T. Kadonaga, K.R. Carner, F.R. Masiarz, R. Tjian, Cell 51 (1987) 1079^1090. [11] U.B. Rosenberg, C. Schroder, A. Preiss, A. Kienlin, S.I.R. Cote, H. Jackle, Nature 319 (1986) 336^339. [12] P.S. Burke, D.J. Wolgemuth, Nucleic Acids Res. 20 (1992) 2827^2834. [13] R. Schuh, W. Aicher, U. Gaul, S. Cote, A. Preiss, D. Maier, E. Seifert, U. Naubert, C. Schroder, R. Kemler, H. Jackle, Cell 47 (1986) 1025^1032. [14] G. Pengue, V. Calabro, P.C. Bartoli, A. Pagliuca, L. Lania, Nucleic Acids Res. 22 (1994) 2908^2914. [15] R.M. Attar, M.Z. Gilman, Mol. Cell. Biol. 12 (1992) 2432^ 2443. [16] K. Chowdhury, M. Goulding, C. Walther, K. Imai, H. Fickenscher, Mech. Dev. 39 (1992) 129^142. [17] X.W. Yang, R. Zhong, N. Heintz, Development 122 (1996) 555^566. [18] A.J. Williams, L.M. Khachigian, T. Shows, T. Collins, J. Biol. Chem. 270 (1995) 22143^22152. [19] E. Beutler, T. Gelbart, C. West, W. Kuhl, P. Lee, Blood Cells Mol. Dis. 21 (1995) 207^216. [20] K. Murai, H. Murakami, S. Nagata, Genes Cell 2 (1997) 581^591. [21] H.J. Thiesen, New Biol. 2 (1990) 363^374. [22] A.J. Buckler, J. Pelletier, D.A. Haber, T. Glaser, D.E. Housman, Mol. Cell. Biol. 11 (1991) 1707^1712. [23] A. Molnar, P. Wu, D.A. Largespada, A. Vortkamp, S. Scherer, N.G. Copeland, N.A. Jenkins, G. Bruns, K. Georgopoulos, J. Immunol. 156 (1996) 585^592. [24] T. El-Baradi, T. Pieler, Mech. Dev. 35 (1991) 155^169.
References [1] L. Labourdette, D. Herbage, F. Mallein-Gerin, Cell. Eng. 1 (1996) 200^208.
BBAEXP 91304 19-10-99
Short sequence-paper
Isolation, cloning, and expression of a new murine zinc ¢nger encoding gene1 Jean-Franc°ois Prost, Didier Ne©gre, Franc°oise Cornet-Javaux, Jean-Claude Cortay, Alain J. Cozzone, Daniel Herbage, Fre¨de¨ric Mallein-Gerin * Institut de Biologie et Chimie des Prote¨ines, CNRS UPR 412, 7 passage du Vercors 69367 Lyon Cedex 07, France Received 17 May 1999; received in revised form 13 August 1999; accepted 16 August 1999
Abstract With the aim of identifying genes involved in cartilage differentiation, we have used a subtractive hybridization strategy with cDNAs from a chondrocytic cell line (MC615) and mRNAs from a mesenchymal precursor cell line (10T1/2). We have isolated a cDNA clone representing a novel mouse gene. The predicted 368-amino acid protein, designated ZF-12, contains four C2 H2 -type zinc finger motifs and one region homologous to the LeR domain, a finger-associated structural domain. ZF-12 mRNAs are expressed during embryonic development and in different organs in adult, including rib cartilage. These data suggest that ZF-12 might play an important role not only in cartilage differentiation, but also in basic cellular processes. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: Mouse; Cartilage; cDNA sequence; Zinc ¢nger motif; LeR domain
Cloning of genes that control chondrocyte di¡erentiation has been hampered because chondrocytes are unstable in culture and dedi¡erentiate when cultured at low density for a long time or on a substratum favoring cell spreading [1]. We previously established a mouse chondrocyte cell line, termed MC615, that is able to express markers characteristic of cartilage in vivo, such as collagen types II, IX, and XI, the link protein and the cartilage aggrecan [2], and to form cartilage nodules when cultured in suspension over agarose [3]. With the aim of identifying genes involved in cartilage di¡erentiation, we have now used a subtractive hybridization strategy with the * Corresponding author. Fax: +33-4-7272-2602; E-mail: [email protected] 1 The nucleotide sequence of ZF-12 will appear in GenBank nucleotide sequence database (accession number AF149093).
MC615 cell line and the mouse embryonic mesenchymal cell line, 10T1/2. 10T1/2 is a multipotential cell line that can be induced, upon treatment with azacytidine, to di¡erentiate into myoblasts, adipocytes and, to a lesser extent, chondrocytes [4]. Therefore, this cell line represents a mesodermal precursor, and a subtractive hybridization screening has been successfully applied with parental 10T1/2 cells and myoblast cells to isolate the cDNA of the ¢rst basic helix-loop-helix myogenic regulatory gene, MyoD [5]. Cytoplasmic RNA from MC615 or 10T1/2 cells grown in monolayer was isolated and poly(A ) RNA was selected using oligo(dT)-cellulose columns. Poly(A ) MC615 RNA was reverse-transcribed with Superscript reverse transcriptase (Gibco-BRL), in presence of [K-32 P]dCTP (800 Ci/mmol). In order to prepare a probe enriched with cDNA sequences
0167-4781 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 9 9 ) 0 0 1 5 7 - 8
BBAEXP 91304 19-10-99
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
Fig. 1. Schematic representation of cDNA clones of ZF-12. ZF12L is the cDNA cloned from the MC615 library. ZF-12R was obtained after RT-PCR. ZF-12L and ZF-12R overlap. The open reading frame is 1104 bp long. The four zinc ¢nger motifs and the LeR domain are represented. The HindIII fragment in the 3P-untranslated region corresponds to the probe used for Northern blotting analysis. Also are indicated the positions of the ZF-12S and ZF-12AS oligonucleotides used for ZF-12 gene expression analysis by RT-PCR in cartilage.
characteristic of MC615 expression, a 10-fold excess of poly(A ) 10T1/2 RNA was hybridized for 28 h at 68³C in 120 mM sodium phosphate, pH 6.8, 820 mM NaCl, 10 mM EDTA, pH 8, and in presence of 10 Wg/ml polyU (15 mer) with the MC615 cDNA molecules. The single-stranded subtracted cDNA probe was eluted at 60³C from an hydroxyapatite column with 120 mM sodium phosphate, pH 2, 0.1% SDS. This subtracted cDNA probe was labeled with [K-32 P]dCTP (3000 Ci/mmol) by random priming and used to screen a MC615 cDNA library constructed in UNI-ZAPXR vector (Stratagene). Positive plaques were submitted to in vivo excision and recircularization into pBluescript phagemid (Stratagene) carrying the cloned DNA insert, according to the manufacturer's instructions. A positive clone termed ZF-12L, carrying a 2-kbp cDNA insert (Fig. 1), was further characterized and its description is reported here. In order to extend the cDNA sequence corresponding to ZF-12L, we carried out 5P rapid ampli¢cation of cDNA ends (RACE) using the 5P-RACE kit (version 2, Life Technologies). The ¢rst strand cDNA was synthesized from 1 Wg mouse liver poly(A ) RNA with the gene-speci¢c primer GSP1 (5P-GTGTTGTTTCTTTCTAGAAGGATTT-3P, positions 1107^1131 in Fig. 2). Following cDNA synthesis, the ¢rst strand product was puri¢ed from non-incorporated dNTPs and GSP1 by spin-column chromatography, and tailed at its 3P-end with termi-
279
nal deoxynucleotidyl transferase and dCTP. This tailed cDNA was then ampli¢ed by PCR using the GSP1 primer and the Abridged Anchor Primer AAP (5P-GGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG-3P). Products from the ¢rst PCR were then ampli¢ed by a second PCR reaction in the presence of the nested gene-speci¢c primer, GSP2 (5P-TTTCAAACCTACCCTTGGGAAA-3P, positions 1078^1099 in Fig. 2) and the Abridged Universal Ampli¢cation Primer AUAP (5P-GGCCACGCGTCGACTAGTAC-3P). By following this procedure, one major PCR product of 1100 bp was resolved on a 5% non-denaturing polyacrylamide gel, then eluted and inserted by TA cloning into plasmid pCR2.1-TOPO (Invitrogen). Sequencing was performed using the dideoxynucleotide chain termination procedure on double-stranded DNA, in the laboratory, or by automated DNA sequencing (Genome Express and ESGS from ACT Gene). The DNA sequences were analyzed by the DNAid computer program [6]. Alignments and identity studies were performed via the IBCP site server2 with the Antheprot [7] program. The two cDNAs isolated from the MC615 cDNA library (ZF-12L, from position 1033 to the 3P-end of the poly-A tail shown in Fig. 2) and from the RACE (ZF-12R, positions 1^1099 in Fig. 2) are diagrammed in Fig. 1. They represent a sequence encoding a putative open reading frame of 1104 bp which is followed by a polyadenylation signal and poly-A tail (Fig. 2). Conceptual translation of this open reading frame yielded a protein of 368 amino acids with four zinc ¢ngers of the C2 H2 type. The C2 H2 motif represents a DNA-binding structure ¢rst discovered in Xenopus RNA polymerase III transcription factor TFIIIA [8,9], and also found in other transcription factors, such as human SP1 [10] and Drosophila Kru«ppel [11]. The zinc ¢nger sequences indicated by a box in Fig. 1 are located in the carboxy terminal region of the predicted protein. Alignment of the zinc ¢nger motifs shows that the invariant cysteines and histidines are clearly conserved (Fig. 3A). Each ¢nger motif conforms closely to the consensus sequence CX2 CX3 FX5 LX2 HX3 H [12] and the four motifs are
2
The IBCP site server is accessible via the World Wide Web (http://www.ibcp.fr).
BBAEXP 91304 19-10-99
280
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
BBAEXP 91304 19-10-99
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
281
Fig. 2. Nucleotide and deduced amino acid sequences of ZF-12. The nucleotide sequence is numbered from the 5P-end of ZF-12R clone, and the amino acid sequence is numbered from the putative translation initiation site. ZF-12L and ZF-12R overlap between nucleotides 1033 and 1099. Below ZF-12 amino acid sequence are indicated the only amino acids derived from a human cDNA sequence (sp accession number 014754) that di¡er from ZF-12. The poly(A ) signal is underlined. Zinc ¢nger encoding sequences are indicated by boxes and numbered. The LeR domain is underlined with a thick line. The termination codon is indicated by . The 3P-untranslated end of ZF-12 cDNA clone which is about 1550 bp long, was not completely sequenced, and is partially represented by a dotted line.
6
connected by highly conserved H/C links, whose consensus sequence is TGEKPYX [13]. Analysis of the N-terminal portion of ZF-12 revealed that the protein contains the leucine-rich region LeR [14], which is highly conserved in the Nterminal portion of several C2 H2 zinc ¢nger proteins, such as SRE-ZBP [15], Zfp-38 [16], RU49 [17], p18 [14], p20 [14], ZNF174 [18], LD5-1 [19] and MZF-2 [20]. This element is 77 amino acids long (Fig. 3B). Although the function of the LeR domain has not yet been determined, the high content of leucine res-
idues with the possible folding of K-helices and the striking conservation of this domain among di¡erent members of zinc ¢nger proteins suggest that it may mediate protein^protein interaction with components of the transcriptional machinery. Moreover, analysis of the predicted amino acid sequence of ZF-12 showed 93% identity with a human sequence listed with accession numbers O14754 in the Swissprot database and P17028 (sequence ZN24) in the Genbank database [21]. Therefore, ZF-12 likely represents the murine counterpart of this human cDNA (Fig. 2).
Fig. 3. (A) Zinc ¢nger domains and inter domains of ZF-12. The four zinc ¢nger motifs of the conceptual translation of ZF-12 cDNA are shown aligned beneath the consensus sequence of mouse C2 H2 ¢ngers, as previously described [12]. Upper case letters indicate the most highly conserved residues, and the lower case letters indicate more variable positions. In positions where either one of two amino acids is likely to be present, these two amino acids are indicated. Residues identical to those of the consensus motif are indicated with a dash. (B) ZF-12 amino acid sequence of the LeR domain aligned with the consensus sequence for the LeR domain found in other zinc ¢nger proteins [20]. In positions where two amino acids are likely to be present in the consensus sequence, both are indicated on the top. The only amino acids of ZF-12 that di¡er from the LeR consensus are indicated at the bottom.
BBAEXP 91304 19-10-99
282
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
Fig. 4. ZF-12 gene expression analysis. (A) Northern blot hybridization of total RNAs isolated from MC615 chondrocytes or 10T1/2 cells with the random prime labeled HindIII cDNA fragment, corresponding to the 3P-untranslated region of ZF-12. A 20-Wg amount of each RNA was loaded per lane. Exposure: 1 week. (B) RT-PCR analysis on total RNA extracted from newborn mouse rib cartilage. After ultracentrifugation, total RNA was digested with RNAse-free DNAse, and reverse transcribed with oligodT primers. Synthesized cDNA was ampli¢ed with ZF-12S and ZF-12AS primers for 35 cycles, with denaturation at 94³C (1 min), annealing at 57³C (1 min) and extension at 72³C (1 min). The PCR reaction generates a 315-bp product, visualized by ethidium bromide staining of a 2% agarose gel. To eliminate the possibility that residual genomic DNA might be contributing to the ampli¢ed product, RNA preparation was also analyzed in the absence of RT and yielded no product (not shown). The positions of 100-bp DNA ladder (Promega) are indicated on the left. (C) Northern blot (Clontech), containing 2 Wg of poly(A ) RNA per lane from di¡erent mouse tissues. The blot was hybridized with the random prime labeled HindIII cDNA fragment, corresponding to the 3P-untranslated region of ZF-12. Exposure: 1 week. (D) Northern blot (Clontech) containing 2 Wg of poly(A ) RNA per lane from mouse embryos at di¡erent developmental stages. The blot was hybridized with the random prime labeled HindIII cDNA fragment, corresponding to the 3P-untranslated region of ZF-12. The positions of RNA size markers are shown on the left. Exposure: 1 week.
This observation is in agreement with the fact that the amino acid sequence of the zinc ¢nger transcription factors is usually well conserved between human and mouse [22,23]. Northern blot analysis using an HindIII cDNA fragment as a probe located in the 3P-non-translated region (see Fig. 1) revealed that ZF-12 gene expression was higher in MC615 cells than in 10 T1/2 cells (Fig. 4A), as expected by using our subtractive hybridization strategy. RT-PCR analysis on total RNA isolated from newborn-mouse rib cartilage using the sense ZF-12S primer (5P-CTTGATGACCCTGGTCAGCC-3P, positions 778^798 in Fig. 2) and the antisense ZF-12AS primer (5P-CCTACCCTTGGGAAAACAGG-3P, positions 1073^1092 in Fig. 2) located between the LeR domain and the zinc ¢nger
domains, indicated that ZF-12 is expressed in vivo in skeletal tissues (Fig. 4B). Because rib cartilage undergoes endochondral ossi¢cation after birth, it will be important to analyze ZF-12 gene expression in developmental skeletal tissues by in situ hybridization, to determine whether ZF-12 is expressed by chondrocytes or osteoblasts, or both. To characterize further the pattern of ZF-12 gene expression in mouse, the HindIII cDNA fragment was hybridized to mRNAs prepared from a panel of adult tissues (Clontech mouse multiple tissue blot). ZF-12 transcripts were detected in heart, brain, liver, skeletal muscle, kidney and testis, and to a very low level, in spleen and lung (Fig. 4C). This signal could be further resolved as transcripts of 3.7 and 6.4 kb, with a stronger signal for the shorter transcript. ZF-12 expression was also
BBAEXP 91304 19-10-99
J.-F. Prost et al. / Biochimica et Biophysica Acta 1447 (1999) 278^283
examined by Northern blot analysis of mRNA from total embryo (Clontech mouse embryo multiple tissue blot). ZF-12 expression was hardly detected at day 7, peaked at day 11 and remained constant at the di¡erent stages examined subsequently (Fig. 4C), suggesting that ZF-12 is developmentally regulated. Each transcript appeared to be of the same relative abundance in the di¡erent samples. It is possible that the 3.7-kb and the faint 6.4-kb transcripts could arise as products of alternative processing, from alternative promoters of the same gene, or as products of a gene or genes related to ZF-12. Isolation and characterization of genomic clones should help to achieve this determination. In any case, the pattern of tissue expression of ZF-12 suggests that this new member of the zinc ¢nger proteins might have a biological role not restricted to skeletal di¡erentiation. In conclusion, the main result of this study is that the presumptive protein product of ZF-12 contains zinc ¢nger domains and a LeR motif, and is thus considered a member of a class of proteins capable of binding speci¢c DNA sequences and regulating cell growth, di¡erentiation or development [24]. It would be interesting to determine and characterize the genomic target, or targets, of ZF-12 to elucidate the function of this particular zinc ¢nger gene. In addition, its expression in murine cell lines, such as MC615 and 10T1/2, should provide a valuable tool to study its own regulation. We thank Dr. Catherine Tomasetto for her precious help during the subtractive hybridization procedure, and Dr. Marie-Christine Rio and Dr. Paul Basset for their kind hospitality in their laboratory (Institut de Ge¨ne¨tique et de Biologie Mole¨culaire et Cellulaire CNRS-INSERM U184, Illkirch, France). F.M.G. was supported by the `Association pour la Recherche sur le Cancer' (ARC).
283
[2] F. Mallein-Gerin, B.R. Olsen, Proc. Natl. Acad. Sci. USA 90 (1993) 3289^3293. [3] F. Mallein-Gerin, F. Ruggiero, T.M. Quinn, F. Bard, A.J. Grodzinsky, B.R. Olsen, M. van der Rest, Exp. Cell Res. 219 (1995) 257^265. [4] S.M. Taylor, P.A. Jones, Cell 17 (1979) 771^779. [5] R.L. Davis, H. Weintraub, A.B. Lassar, Cell 51 (1987) 987^ 1000. [6] F. Dardel, P. Bensoussan, Comput. Appl. Biosci. 4 (1988) 483^486. [7] G. Dele¨age, C. Geourjon, Comput. Appl. Biosci. 9 (1993) 197^199. [8] J. Miller, A.D. McLachlan, A. Klug, EMBO J. 4 (1985) 1609^1614. [9] R.S. Brown, C. Sander, P. Argos, FEBS Lett. 186 (1985) 271^274. [10] J.T. Kadonaga, K.R. Carner, F.R. Masiarz, R. Tjian, Cell 51 (1987) 1079^1090. [11] U.B. Rosenberg, C. Schroder, A. Preiss, A. Kienlin, S.I.R. Cote, H. Jackle, Nature 319 (1986) 336^339. [12] P.S. Burke, D.J. Wolgemuth, Nucleic Acids Res. 20 (1992) 2827^2834. [13] R. Schuh, W. Aicher, U. Gaul, S. Cote, A. Preiss, D. Maier, E. Seifert, U. Naubert, C. Schroder, R. Kemler, H. Jackle, Cell 47 (1986) 1025^1032. [14] G. Pengue, V. Calabro, P.C. Bartoli, A. Pagliuca, L. Lania, Nucleic Acids Res. 22 (1994) 2908^2914. [15] R.M. Attar, M.Z. Gilman, Mol. Cell. Biol. 12 (1992) 2432^ 2443. [16] K. Chowdhury, M. Goulding, C. Walther, K. Imai, H. Fickenscher, Mech. Dev. 39 (1992) 129^142. [17] X.W. Yang, R. Zhong, N. Heintz, Development 122 (1996) 555^566. [18] A.J. Williams, L.M. Khachigian, T. Shows, T. Collins, J. Biol. Chem. 270 (1995) 22143^22152. [19] E. Beutler, T. Gelbart, C. West, W. Kuhl, P. Lee, Blood Cells Mol. Dis. 21 (1995) 207^216. [20] K. Murai, H. Murakami, S. Nagata, Genes Cell 2 (1997) 581^591. [21] H.J. Thiesen, New Biol. 2 (1990) 363^374. [22] A.J. Buckler, J. Pelletier, D.A. Haber, T. Glaser, D.E. Housman, Mol. Cell. Biol. 11 (1991) 1707^1712. [23] A. Molnar, P. Wu, D.A. Largespada, A. Vortkamp, S. Scherer, N.G. Copeland, N.A. Jenkins, G. Bruns, K. Georgopoulos, J. Immunol. 156 (1996) 585^592. [24] T. El-Baradi, T. Pieler, Mech. Dev. 35 (1991) 155^169.
References [1] L. Labourdette, D. Herbage, F. Mallein-Gerin, Cell. Eng. 1 (1996) 200^208.
BBAEXP 91304 19-10-99