- Email: [email protected]
DmFoxF, a novel Drosophila fork head factor expressed in visceral mesoderm
Mechanisms of Development 111 (2002) 163–166 www.elsevier.com/locate/modo
Gene expression pattern
DmFoxF, a novel Drosophila fork head factor expressed in visceral mesoderm Cristina Pe´rez Sa´nchez 1, Sergio Casas-Tinto´ 1, Lucas Sa´nchez, Javier Rey-Campos, Begon˜a Granadino* Centro Investigaciones Biolo´gicas CSIC, Vela´zquez 144, 28006 Madrid, Spain Received 26 July 2001; received in revised form 15 October 2001; accepted 24 October 2001
Abstract DmFoxF is a novel Drosophila fork head domain factor, which is expressed in the visceral mesoderm of the embryo. Our data suggest that DmFoxF is the fly orthologue of the vertebrates FOXF1 and FOXF2 transcription factors. DmFoxF shares homology with FOXF1 and FOXF2 in its fork head domain, and it is able to specifically bind DNA sequences recognized by these vertebrate fork head factors. In stage 10–11 embryos, the DmFoxF protein is detected into the nuclei of cells of the presumptive visceral mesoderm. It localizes at the segmental cell clusters of the mesoderm, which will eventually develop to surround the midgut endoderm. DmFoxF is also expressed in the proctodeal mesoderm, which will develop into the visceral mesoderm of the hindgut. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: DmFoxF; Fork head factors; Visceral mesoderm; FOXF1; FOXF2; Foxf1; Foxf2
1. Results and discussion The fork head family groups transcription factors that share a structurally related DNA binding domain, the fork head domain (Granadino et al., 2000). Several fork head domain (FD) genes, have been cloned and characterized in Drosophila. Most of them are involved in embryonic development (Hacker et al., 1992). In a search of the Drosophila DNA sequences database (Berkeley Drosophila Genome Project) with the amino acid sequence of the fork head domain, we identified a novel putative Drosophila fork head protein. This showed 63% homology, in its DNA binding domain, with the sequences of vertebrate FOXF1 and FOXF2 fork head factors (Pierrou et al., 1994). Based on this homology and other functional assays described below, we postulated that this novel Drosophila fork head factor was the fly orthologue of the vertebrate FOXF1 and FOXF2, and thus named it DmFoxF. We found that the DmFoxF sequence was included in a genomic clone (AC004351) from the third chromosome of Drosophila, which mapped the gene at the band 65D465E1. The putative exon-intron structure of DmFoxF was predicted by searching for canonical donor and acceptor * Corresponding author. Tel.: 134-91-5644562 ext.4416; fax: 134-915627518. E-mail address: [email protected] (B. Granadino). 1 These two authors have contributed equally to this work.
splice sites in this genomic sequence, and confirmed by RT-PCR experiments. The gene spans at least 4 Kbp and contains four exons (Fig. 1A). All splice junctions conformed to the GT/AG rule (Padgett et al., 1986). This indicates that the DmFoxF transcript spans at least 2582 nucleotides, with an open reading frame (ORF) of 1992 nucleotides. This would code for a 664 amino acid polypeptide, which includes the entire fork head DNA binding domain (Fig. 1B). Recently, we have found in the database a computerpredicted cDNA sequence (CG18647) that included part of the DmFoxF sequence. This cDNA included an extra exon of 90 nt, between exon 1 and exon 2. Although this extra exon would maintain the reading frame, it would add 30 extra residues to the predicted amino acid sequence, interrupting the fork head domain. This extra sequence has no correspondence within the human and mouse FOXF1 and FOXF2 proteins. Several attempts to find mRNAs with this extra exon by RT-PCR experiments, failed to provide evidence for the actual utilization of this alternative exon. However, we cannot discard that some rare Drosophila mRNA species may contain it. The homology between DmFoxF and vertebrate FOXF1 and FOXF2 is restricted only to the fork head domains (Fig. 2A), which show a rather low sequence similarity. However, the computer predicted three-dimensional structure of the DmFoxF fork head domain showed a general topology akin
0925-4773/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0925-477 3(01)00603-7
164
C. Pe´ rez Sa´ nchez et al. / Mechanisms of Development 111 (2002) 163–166
Fig. 1. DmFoxF gene and protein. (A) Exon/intron structure of the DmFoxFgene. The region encoding the fork head domain is shown in red. (B) Amino acid sequence of the DmFoxF protein. The fork head domain is shown in red.
stages of Drosophila: embryos, larvae, pupae and adults (Fig. 4A). A polyclonal antiserum, raised against the GST:DmFoxF fusion protein, revealed a single 73 kDa band in Western blot experiments with protein extracts of all these developmental stages (Fig. 4B). Northern blot experiments with total RNA from larvae and adults showed a single DmFoxF transcript of approximately 5 Kb (Fig. 4C). DmFoxF mRNA is first detected at the syncitial blastoderm stage, suggesting that this gene has maternal expression (data not shown). RT-PCR experiments with RNA isolated from unfertilized embryos confirmed this (Fig. 4A). However, Western blot experiments with the DmFoxF antiserum did not show expression of the DmFoxF protein in unfertilized embryos. The DmFoxF protein was first detected in stage 10–11 embryos, at the segmental cell clus-
to that of other fork head domains (Fig. 2B). In order to know whether the Drosophila factor was able to bind DNA similarly to the vertebrate homologues, we carried out electrophoretic mobility shifts assays. We used a recombinant Glutathione-S-transferase (GST):DmFoxF fusion protein, and a labeled oligonucleotide probe (F in Fig. 3) containing a consensus FOXF2 site (Pierrou et al., 1994). A retarded band was observed when this probe was incubated with the GST:DmFoxF fusion protein (Fig. 3), but not when purified GST alone was used. The same unlabelled oligonucleotide, but not an unrelated oligonucleotide, efficiently competed the formation of the retarded band. A related oligonucleotide (PE2) poorly competed the formation of the complex. This oligonucleotide contains a high affinity site for another fork head factor, FOXJ2, (Perez-Sanchez et al., 2000) and has a similar, but not identical, sequence. The GST:DmFoxF fusion protein bound, although not as efficiently, this oligonucleotide. Other oligonucleotide (NC in Fig. 3), containing a different sequence, also recognized by FOXJ2, but of a different type (Perez-Sanchez et al., 2000), was not bound by the GST:DmFoxF protein. DNA sequences recognized by FOXF2 contain a core element, GTAAACA, common to other fork head factors. Sequence specificity for FOXF2 is provided by specific nucleotides flaking this core element, at its 3 0 and 5 0 ends (Pierrou et al., 1994). The sequence of the PE2 oligonucleotide, but not those of GAS or NC oligonucleotides, contained the same core element. This could explain why DmFoxF is still able to bind the PE2 oligonucleotide, but not GAS or NC. A guanine residue 3 0 to the core element of PE2, (shown in inverted type in Fig. 3), could account for the lower binding affinity of DmFoxF to this sequence. The same position is rarely a guanine residue in human FOXF2 (7%) and FOXF1 (2%) PRC-selected sites (Peterson et al., 1997; Pierrou et al., 1994). The expression of DmFoxF during Drosophila development was analyzed by RT-PCR. The amplimers were sequenced for confirmation. These experiments showed that DmFoxF was expressed in all main developmental
Fig. 2. Structure of the fork head domain of DmFoxF. (A) Alignment of human and mouse FOXF1 and FOXF2, and Drosophila DmFoxF fork head domains. The a-helices, b-strands and W1 and W2 regions are indicated based on the data from the structure of the HNF3g fork head domain. The bipartite nuclear localization signals (NLS) are shown in red. Although the W2 region of DmFoxF is quite disparate from the human or murine FOXF1 and FOXF2, a basic motif (RRRPRGYRSK) within this region is still present, which resembles the NLS-C of the mammalian factors. (B) Computer-predicted model of the three-dimensional structure of the fork head domain of DmFoxF (red), built from its amino acid sequence, superimposed onto the three-dimensional structure of the HNF3g fork head domain (blue). The NMR spectroscopy-resolved structures of the fork head domains of rat Genesis (Marsden et al., 1998), human Freac-11 (van Dongen et al., 2000), human AFX (Weigelt et al., 2000>), and the X-ray structure of the fork head domain of rat HNF-3g (Clark et al., 1993), and knowledge-based computer assisted protein modeling methods were used for the three-dimensional prediction. The Cartesian coordinates of each structure were from the Brookhaven Protein Data Bank, with PDB codes 2HFH, 1D5V, 1E17, for Genesis, Freac-11 and AFX respectively, whereas the HNF-3 coordinates were from the Nucleic Acid Database Project (code PDT013). Computations were performed on a Power Challenge R10000 by using the BIOSYM software package, release 95.0 (Molecular Simulations, Inc., San Diego, CA).
C. Pe´ rez Sa´ nchez et al. / Mechanisms of Development 111 (2002) 163–166
165
DmFoxF protein is also expressed in mesodermal cells of the proctodeum, which will develop into visceral mesoderm of the hindgut. 2. Experimental procedures Accession number: the sequence data have been submitted to the GeneBank under the accession number: AY044243. 2.1. RNA analysis RNAs were extracted by the guanidinium acid-phenol chloroform method (Chomczynski and Sacchi, 1987) and separated on formaldehyde gels, transferred onto nylon membranes and hybridized with [ 32P]-labeled probes, following standard methods (Ausubel et al., 1998). RT-PCR analysis were performed with total RNA and primers DmFoxF 685 5 0 -AATGCGCACGGCATGCCCGTG -3 0 , DmFoxF 889 5 0 - CCAGAGAAACCAGCACTCAG-3 0 -DmFoxF 998 5 0 -ACGAATTCTTTCGAGGACCC-3 0 -DmFoxF 1206c 5 0 -GATCTTGGAGCGATAGCCAFig. 3. Binding of DmFoxF to DNA. EMSA with purified recombinant proteins and labeled double-stranded oligonucleotide probes shown below. Different unlabeled oligonucleotides were used as competitors at a 100-fold molar excess. Protein used were: F, GST::DmFoxF fusion protein. G, GST alone. Oligonucleotides used were: F, oligonucleotide containing the optimum sequence recognized by human FOXF2 (Pierrou et al., 1994); PE2 and NC, oligonucleotides containing binding sites for FOXJ2 (Perez-Sanchez et al., 2000); GAS, oligonucleotide containing an interferon g activating sequence of the FcgRI receptor gene (Shuai et al., 1994)-no competitor.
ters that include the presumptive midgut visceral mesoderm (Fig. 5). As development progresses, DmFoxF protein expression appears to be restricted to columnar cells, at the ventral side of the patch, which connect to each other to form a continuous band of visceral mesoderm (Fig. 5B).
Fig. 4. Expression of DmFoxF in different developmental stages. (A) Analysis of the expression of DmFoxF by RT-PCR experiments with total RNAs from different Drosophila developmental stages. (B) Western blot analysis with the DmFoxF polyclonal antiserum. (C) Northern blot analysis of 10 mg of total RNA isolated from larvae and adult flies. The blot was hybridized with a 32P-labelled probe derived from the DmFoxF cDNA. uE, unfertilized embryos; E, fertilized embryos; P, pupae; L, Larvae; A, adults.
Fig. 5. DmFoxF protein expression in whole embryos. Whole mount preparations of wild-type embryos were stained with the anti-DmFoxF antibody. (A,B) Stage 11 embryos photographed with Nomarski optics. Segmental cell clusters corresponding to presumptive visceral mesoderm are indicated with arrowheads. (C) Stage 13 embryo showing the visceral mesoderm continuous band. (D) Detail of a visceral mesoderm cluster showing the nuclear staining of the DmFoxF antibodies. (E,F) confocal microphotographs of immunofluorescence experiments with stages 10 and 14 embryos. VM, visceral mesoderm; PM, proctodeal mesoderm. Microscopy images were obtained with an inverted Nikon ES300 microscope with Nomarski interference optics and epifluorescence.
166
C. Pe´ rez Sa´ nchez et al. / Mechanisms of Development 111 (2002) 163–166
C-3 0 -DmFoxF 1577c 5 0 -GCCGCTCCCAGTGGCGAACTTG-3 0 -DmFoxF 1940c 5 0 -ATCATGGGGGGCGGGCTGATG-3 0 . 2.2. DNA binding assays. Antibody preparation and embryo immunohistochemistry A 519-base pair fragment of the DmFoxF cDNA, coding residues N-229 to I-402 including the fork head domain, was cloned in the pGEX-3X vector (Amersham Pharmacia Biotech) in frame with the GST open reading frame. The recombinant protein was purified by affinity chromatography onto glutathione-sepharose columns, and it was used for electrophoretic mobility shift assays (EMSA) as described (Perez-Sanchez et al., 2000). The purified recombinant protein was also used for antibody preparation. Immunizations of mice were done by subcutaneous injections of 100 mg of the purified fusion protein, following standard procedures (Harlow and Lane, 1988). Antibody staining of fly embryos was basically done following Azpiazu et al. (1996). Acknowledgements We thank J. Ferna´ ndez-Cabrera, L. Mateos and R. De Andre´ s for their technical assistance, I. Guerrero, I. Rodriguez, M. Ruiz-Go´ mez for helpful discussions, and M. Garcia de Lacoba for computer modeling. This work was supported by grants PM96-0005 and PM99-0103 from the Ministerio de Ciencia y TecnologI´a and PM08.9/0006/1998 from the Comunidad Auto´ noma de Madrid. References Ausubel, F.A., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., Struhl, K., 1998. Current Protocols in Molecular Biology, John Wiley and Sons, New York. Azpiazu, N., Lawrence, P.A., Vincent, J.P., Frasch, M., 1996. Segmentation
and specification of the Drosophila mesoderm. Genes Dev. 10, 3183– 3194. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156–159. Clark, K.L., Halay, E.D., Lai, E., Burley, S.K., 1993. Co-crystal structure of the HNF-3/fork head DNA-recognition motif resembles histone H5. Nature 364, 412–420. Granadino, B., Pe´ rez-Sa´ nchez, C., Rey-Campos, J., 2000. Fork head transcription factors. Curr. Genomics 1, 353–382. Hacker, U., Grossniklaus, U., Gehring, W.J., Jackle, H., 1992. Developmentally regulated Drosophila gene family encoding the fork head domain. Proc. Natl. Acad. Sci. U.S.A. 89, 8754–8758. Harlow, E., Lane, D. (Eds.), 1988. Antibodies: a Laboratory Manual C.S.H.L, Cold Spring Harbor, NY. Marsden, I., Jin, C., Liao, X., 1998. Structural changes in the region directly adjacent to the DNA-binding helix highlight a possible mechanism to explain the observed changes in the sequence-specific binding of winged helix proteins. J. Mol. Biol. 278, 293–299. Padgett, R.A., Grabowski, P.J., Konarska, M.M., Seiler, S., Sharp, P.A., 1986. Splicing of messenger RNA precursors. Annu. Rev. Biochem. 55, 1119–1150. Perez-Sanchez, C., Gomez-Ferreria, M.A., de La Fuente, C.A., Granadino, B., Velasco, G., Esteban-Gamboa, A., Rey-Campos, J., 2000. FHX, a novel fork head factor with a dual DNA binding specificity. J. Biol. Chem. 275, 12909–12916. Peterson, R.S., Clevidence, D.E., Ye, H., Costa, R.H., 1997. Hepatocyte nuclear factor-3 alpha promoter regulation involves recognition by cellspecific factors, thyroid transcription factor-1, and autoactivation. Cell Growth Differ. 8, 69–82. Pierrou, S., Hellqvist, M., Samuelsson, L., Enerback, S., Carlsson, P., 1994. Cloning and characterization of seven human forkhead proteins: binding site specificity and DNA bending. EMBO J. 13, 5002–5012. Shuai, K., Horvath, C.M., Huang, L.H., Qureshi, S.A., Cowburn, D., Darnell Jr., J.E., 1994. Interferon activation of the transcription factor Stat91 involves dimerization through SH2-phosphotyrosyl peptide interactions. Cell 76, 821–828. van Dongen, M.J., Cederberg, A., Carlsson, P., Gnerback, S., Wikstrom, M., 2000. Solution structure and dynamics of the DNA-binding domain of the adipocyte-transcription factor FREAC-II [in process citation]. J. Mol. Biol. 296, 351–359. Weigelt, J., Climent, I., Dahman-Wright, K., Wikstrom, M., 2000. 1H, 13C and 15N resonance assignments of the DNA binding domain of the human forkhead transcription factor AFX. J. Biomol. NMR 17, 181– 182.
Gene expression pattern
DmFoxF, a novel Drosophila fork head factor expressed in visceral mesoderm Cristina Pe´rez Sa´nchez 1, Sergio Casas-Tinto´ 1, Lucas Sa´nchez, Javier Rey-Campos, Begon˜a Granadino* Centro Investigaciones Biolo´gicas CSIC, Vela´zquez 144, 28006 Madrid, Spain Received 26 July 2001; received in revised form 15 October 2001; accepted 24 October 2001
Abstract DmFoxF is a novel Drosophila fork head domain factor, which is expressed in the visceral mesoderm of the embryo. Our data suggest that DmFoxF is the fly orthologue of the vertebrates FOXF1 and FOXF2 transcription factors. DmFoxF shares homology with FOXF1 and FOXF2 in its fork head domain, and it is able to specifically bind DNA sequences recognized by these vertebrate fork head factors. In stage 10–11 embryos, the DmFoxF protein is detected into the nuclei of cells of the presumptive visceral mesoderm. It localizes at the segmental cell clusters of the mesoderm, which will eventually develop to surround the midgut endoderm. DmFoxF is also expressed in the proctodeal mesoderm, which will develop into the visceral mesoderm of the hindgut. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: DmFoxF; Fork head factors; Visceral mesoderm; FOXF1; FOXF2; Foxf1; Foxf2
1. Results and discussion The fork head family groups transcription factors that share a structurally related DNA binding domain, the fork head domain (Granadino et al., 2000). Several fork head domain (FD) genes, have been cloned and characterized in Drosophila. Most of them are involved in embryonic development (Hacker et al., 1992). In a search of the Drosophila DNA sequences database (Berkeley Drosophila Genome Project) with the amino acid sequence of the fork head domain, we identified a novel putative Drosophila fork head protein. This showed 63% homology, in its DNA binding domain, with the sequences of vertebrate FOXF1 and FOXF2 fork head factors (Pierrou et al., 1994). Based on this homology and other functional assays described below, we postulated that this novel Drosophila fork head factor was the fly orthologue of the vertebrate FOXF1 and FOXF2, and thus named it DmFoxF. We found that the DmFoxF sequence was included in a genomic clone (AC004351) from the third chromosome of Drosophila, which mapped the gene at the band 65D465E1. The putative exon-intron structure of DmFoxF was predicted by searching for canonical donor and acceptor * Corresponding author. Tel.: 134-91-5644562 ext.4416; fax: 134-915627518. E-mail address: [email protected] (B. Granadino). 1 These two authors have contributed equally to this work.
splice sites in this genomic sequence, and confirmed by RT-PCR experiments. The gene spans at least 4 Kbp and contains four exons (Fig. 1A). All splice junctions conformed to the GT/AG rule (Padgett et al., 1986). This indicates that the DmFoxF transcript spans at least 2582 nucleotides, with an open reading frame (ORF) of 1992 nucleotides. This would code for a 664 amino acid polypeptide, which includes the entire fork head DNA binding domain (Fig. 1B). Recently, we have found in the database a computerpredicted cDNA sequence (CG18647) that included part of the DmFoxF sequence. This cDNA included an extra exon of 90 nt, between exon 1 and exon 2. Although this extra exon would maintain the reading frame, it would add 30 extra residues to the predicted amino acid sequence, interrupting the fork head domain. This extra sequence has no correspondence within the human and mouse FOXF1 and FOXF2 proteins. Several attempts to find mRNAs with this extra exon by RT-PCR experiments, failed to provide evidence for the actual utilization of this alternative exon. However, we cannot discard that some rare Drosophila mRNA species may contain it. The homology between DmFoxF and vertebrate FOXF1 and FOXF2 is restricted only to the fork head domains (Fig. 2A), which show a rather low sequence similarity. However, the computer predicted three-dimensional structure of the DmFoxF fork head domain showed a general topology akin
0925-4773/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0925-477 3(01)00603-7
164
C. Pe´ rez Sa´ nchez et al. / Mechanisms of Development 111 (2002) 163–166
Fig. 1. DmFoxF gene and protein. (A) Exon/intron structure of the DmFoxFgene. The region encoding the fork head domain is shown in red. (B) Amino acid sequence of the DmFoxF protein. The fork head domain is shown in red.
stages of Drosophila: embryos, larvae, pupae and adults (Fig. 4A). A polyclonal antiserum, raised against the GST:DmFoxF fusion protein, revealed a single 73 kDa band in Western blot experiments with protein extracts of all these developmental stages (Fig. 4B). Northern blot experiments with total RNA from larvae and adults showed a single DmFoxF transcript of approximately 5 Kb (Fig. 4C). DmFoxF mRNA is first detected at the syncitial blastoderm stage, suggesting that this gene has maternal expression (data not shown). RT-PCR experiments with RNA isolated from unfertilized embryos confirmed this (Fig. 4A). However, Western blot experiments with the DmFoxF antiserum did not show expression of the DmFoxF protein in unfertilized embryos. The DmFoxF protein was first detected in stage 10–11 embryos, at the segmental cell clus-
to that of other fork head domains (Fig. 2B). In order to know whether the Drosophila factor was able to bind DNA similarly to the vertebrate homologues, we carried out electrophoretic mobility shifts assays. We used a recombinant Glutathione-S-transferase (GST):DmFoxF fusion protein, and a labeled oligonucleotide probe (F in Fig. 3) containing a consensus FOXF2 site (Pierrou et al., 1994). A retarded band was observed when this probe was incubated with the GST:DmFoxF fusion protein (Fig. 3), but not when purified GST alone was used. The same unlabelled oligonucleotide, but not an unrelated oligonucleotide, efficiently competed the formation of the retarded band. A related oligonucleotide (PE2) poorly competed the formation of the complex. This oligonucleotide contains a high affinity site for another fork head factor, FOXJ2, (Perez-Sanchez et al., 2000) and has a similar, but not identical, sequence. The GST:DmFoxF fusion protein bound, although not as efficiently, this oligonucleotide. Other oligonucleotide (NC in Fig. 3), containing a different sequence, also recognized by FOXJ2, but of a different type (Perez-Sanchez et al., 2000), was not bound by the GST:DmFoxF protein. DNA sequences recognized by FOXF2 contain a core element, GTAAACA, common to other fork head factors. Sequence specificity for FOXF2 is provided by specific nucleotides flaking this core element, at its 3 0 and 5 0 ends (Pierrou et al., 1994). The sequence of the PE2 oligonucleotide, but not those of GAS or NC oligonucleotides, contained the same core element. This could explain why DmFoxF is still able to bind the PE2 oligonucleotide, but not GAS or NC. A guanine residue 3 0 to the core element of PE2, (shown in inverted type in Fig. 3), could account for the lower binding affinity of DmFoxF to this sequence. The same position is rarely a guanine residue in human FOXF2 (7%) and FOXF1 (2%) PRC-selected sites (Peterson et al., 1997; Pierrou et al., 1994). The expression of DmFoxF during Drosophila development was analyzed by RT-PCR. The amplimers were sequenced for confirmation. These experiments showed that DmFoxF was expressed in all main developmental
Fig. 2. Structure of the fork head domain of DmFoxF. (A) Alignment of human and mouse FOXF1 and FOXF2, and Drosophila DmFoxF fork head domains. The a-helices, b-strands and W1 and W2 regions are indicated based on the data from the structure of the HNF3g fork head domain. The bipartite nuclear localization signals (NLS) are shown in red. Although the W2 region of DmFoxF is quite disparate from the human or murine FOXF1 and FOXF2, a basic motif (RRRPRGYRSK) within this region is still present, which resembles the NLS-C of the mammalian factors. (B) Computer-predicted model of the three-dimensional structure of the fork head domain of DmFoxF (red), built from its amino acid sequence, superimposed onto the three-dimensional structure of the HNF3g fork head domain (blue). The NMR spectroscopy-resolved structures of the fork head domains of rat Genesis (Marsden et al., 1998), human Freac-11 (van Dongen et al., 2000), human AFX (Weigelt et al., 2000>), and the X-ray structure of the fork head domain of rat HNF-3g (Clark et al., 1993), and knowledge-based computer assisted protein modeling methods were used for the three-dimensional prediction. The Cartesian coordinates of each structure were from the Brookhaven Protein Data Bank, with PDB codes 2HFH, 1D5V, 1E17, for Genesis, Freac-11 and AFX respectively, whereas the HNF-3 coordinates were from the Nucleic Acid Database Project (code PDT013). Computations were performed on a Power Challenge R10000 by using the BIOSYM software package, release 95.0 (Molecular Simulations, Inc., San Diego, CA).
C. Pe´ rez Sa´ nchez et al. / Mechanisms of Development 111 (2002) 163–166
165
DmFoxF protein is also expressed in mesodermal cells of the proctodeum, which will develop into visceral mesoderm of the hindgut. 2. Experimental procedures Accession number: the sequence data have been submitted to the GeneBank under the accession number: AY044243. 2.1. RNA analysis RNAs were extracted by the guanidinium acid-phenol chloroform method (Chomczynski and Sacchi, 1987) and separated on formaldehyde gels, transferred onto nylon membranes and hybridized with [ 32P]-labeled probes, following standard methods (Ausubel et al., 1998). RT-PCR analysis were performed with total RNA and primers DmFoxF 685 5 0 -AATGCGCACGGCATGCCCGTG -3 0 , DmFoxF 889 5 0 - CCAGAGAAACCAGCACTCAG-3 0 -DmFoxF 998 5 0 -ACGAATTCTTTCGAGGACCC-3 0 -DmFoxF 1206c 5 0 -GATCTTGGAGCGATAGCCAFig. 3. Binding of DmFoxF to DNA. EMSA with purified recombinant proteins and labeled double-stranded oligonucleotide probes shown below. Different unlabeled oligonucleotides were used as competitors at a 100-fold molar excess. Protein used were: F, GST::DmFoxF fusion protein. G, GST alone. Oligonucleotides used were: F, oligonucleotide containing the optimum sequence recognized by human FOXF2 (Pierrou et al., 1994); PE2 and NC, oligonucleotides containing binding sites for FOXJ2 (Perez-Sanchez et al., 2000); GAS, oligonucleotide containing an interferon g activating sequence of the FcgRI receptor gene (Shuai et al., 1994)-no competitor.
ters that include the presumptive midgut visceral mesoderm (Fig. 5). As development progresses, DmFoxF protein expression appears to be restricted to columnar cells, at the ventral side of the patch, which connect to each other to form a continuous band of visceral mesoderm (Fig. 5B).
Fig. 4. Expression of DmFoxF in different developmental stages. (A) Analysis of the expression of DmFoxF by RT-PCR experiments with total RNAs from different Drosophila developmental stages. (B) Western blot analysis with the DmFoxF polyclonal antiserum. (C) Northern blot analysis of 10 mg of total RNA isolated from larvae and adult flies. The blot was hybridized with a 32P-labelled probe derived from the DmFoxF cDNA. uE, unfertilized embryos; E, fertilized embryos; P, pupae; L, Larvae; A, adults.
Fig. 5. DmFoxF protein expression in whole embryos. Whole mount preparations of wild-type embryos were stained with the anti-DmFoxF antibody. (A,B) Stage 11 embryos photographed with Nomarski optics. Segmental cell clusters corresponding to presumptive visceral mesoderm are indicated with arrowheads. (C) Stage 13 embryo showing the visceral mesoderm continuous band. (D) Detail of a visceral mesoderm cluster showing the nuclear staining of the DmFoxF antibodies. (E,F) confocal microphotographs of immunofluorescence experiments with stages 10 and 14 embryos. VM, visceral mesoderm; PM, proctodeal mesoderm. Microscopy images were obtained with an inverted Nikon ES300 microscope with Nomarski interference optics and epifluorescence.
166
C. Pe´ rez Sa´ nchez et al. / Mechanisms of Development 111 (2002) 163–166
C-3 0 -DmFoxF 1577c 5 0 -GCCGCTCCCAGTGGCGAACTTG-3 0 -DmFoxF 1940c 5 0 -ATCATGGGGGGCGGGCTGATG-3 0 . 2.2. DNA binding assays. Antibody preparation and embryo immunohistochemistry A 519-base pair fragment of the DmFoxF cDNA, coding residues N-229 to I-402 including the fork head domain, was cloned in the pGEX-3X vector (Amersham Pharmacia Biotech) in frame with the GST open reading frame. The recombinant protein was purified by affinity chromatography onto glutathione-sepharose columns, and it was used for electrophoretic mobility shift assays (EMSA) as described (Perez-Sanchez et al., 2000). The purified recombinant protein was also used for antibody preparation. Immunizations of mice were done by subcutaneous injections of 100 mg of the purified fusion protein, following standard procedures (Harlow and Lane, 1988). Antibody staining of fly embryos was basically done following Azpiazu et al. (1996). Acknowledgements We thank J. Ferna´ ndez-Cabrera, L. Mateos and R. De Andre´ s for their technical assistance, I. Guerrero, I. Rodriguez, M. Ruiz-Go´ mez for helpful discussions, and M. Garcia de Lacoba for computer modeling. This work was supported by grants PM96-0005 and PM99-0103 from the Ministerio de Ciencia y TecnologI´a and PM08.9/0006/1998 from the Comunidad Auto´ noma de Madrid. References Ausubel, F.A., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., Struhl, K., 1998. Current Protocols in Molecular Biology, John Wiley and Sons, New York. Azpiazu, N., Lawrence, P.A., Vincent, J.P., Frasch, M., 1996. Segmentation
and specification of the Drosophila mesoderm. Genes Dev. 10, 3183– 3194. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156–159. Clark, K.L., Halay, E.D., Lai, E., Burley, S.K., 1993. Co-crystal structure of the HNF-3/fork head DNA-recognition motif resembles histone H5. Nature 364, 412–420. Granadino, B., Pe´ rez-Sa´ nchez, C., Rey-Campos, J., 2000. Fork head transcription factors. Curr. Genomics 1, 353–382. Hacker, U., Grossniklaus, U., Gehring, W.J., Jackle, H., 1992. Developmentally regulated Drosophila gene family encoding the fork head domain. Proc. Natl. Acad. Sci. U.S.A. 89, 8754–8758. Harlow, E., Lane, D. (Eds.), 1988. Antibodies: a Laboratory Manual C.S.H.L, Cold Spring Harbor, NY. Marsden, I., Jin, C., Liao, X., 1998. Structural changes in the region directly adjacent to the DNA-binding helix highlight a possible mechanism to explain the observed changes in the sequence-specific binding of winged helix proteins. J. Mol. Biol. 278, 293–299. Padgett, R.A., Grabowski, P.J., Konarska, M.M., Seiler, S., Sharp, P.A., 1986. Splicing of messenger RNA precursors. Annu. Rev. Biochem. 55, 1119–1150. Perez-Sanchez, C., Gomez-Ferreria, M.A., de La Fuente, C.A., Granadino, B., Velasco, G., Esteban-Gamboa, A., Rey-Campos, J., 2000. FHX, a novel fork head factor with a dual DNA binding specificity. J. Biol. Chem. 275, 12909–12916. Peterson, R.S., Clevidence, D.E., Ye, H., Costa, R.H., 1997. Hepatocyte nuclear factor-3 alpha promoter regulation involves recognition by cellspecific factors, thyroid transcription factor-1, and autoactivation. Cell Growth Differ. 8, 69–82. Pierrou, S., Hellqvist, M., Samuelsson, L., Enerback, S., Carlsson, P., 1994. Cloning and characterization of seven human forkhead proteins: binding site specificity and DNA bending. EMBO J. 13, 5002–5012. Shuai, K., Horvath, C.M., Huang, L.H., Qureshi, S.A., Cowburn, D., Darnell Jr., J.E., 1994. Interferon activation of the transcription factor Stat91 involves dimerization through SH2-phosphotyrosyl peptide interactions. Cell 76, 821–828. van Dongen, M.J., Cederberg, A., Carlsson, P., Gnerback, S., Wikstrom, M., 2000. Solution structure and dynamics of the DNA-binding domain of the adipocyte-transcription factor FREAC-II [in process citation]. J. Mol. Biol. 296, 351–359. Weigelt, J., Climent, I., Dahman-Wright, K., Wikstrom, M., 2000. 1H, 13C and 15N resonance assignments of the DNA binding domain of the human forkhead transcription factor AFX. J. Biomol. NMR 17, 181– 182.