- Email: [email protected]
Cloning and analysis of the murine Foxi2 transcription factor
Biochimica et Biophysica Acta 1731 (2005) 133 – 138 http://www.elsevier.com/locate/bba
Short sequence paper
Cloning and analysis of the murine Foxi2 transcription factor Patrick J.E.C. Wijchers, Marco F.M. Hoekman, J. Peter H. Burbach, Marten P. Smidt * Rudolf Magnus Institute of Neuroscience, Department of Pharmacology and Anatomy, University Medical Center Utrecht, Universiteitsweg 100, 3584 CG Utrecht, The Netherlands Received 14 June 2005; received in revised form 31 August 2005; accepted 13 September 2005 Available online 5 October 2005
Abstract Forkhead transcription factors comprise a large family of key regulators of embryonic development. Here, we describe the cloning and analysis of the murine Foxi2 gene, coding for a putative 311 amino acid protein resembling Foxi subfamily members in mice and other species. Expression analysis during the final stages of embryonic development revealed that Foxi2 expression is mainly confined to subsets of cells in epithelial structures and particular ducts, in addition to the developing forebrain and neural retina. Since FoxI factors are thought to be implicated in the regulation of cell fate, the highly restricted expression pattern of Foxi2 suggestive of a possible role in controlling cellular identity. D 2005 Elsevier B.V. All rights reserved. Keywords: Forkhead; Transcription factor; Foxi2; Development; Cell identity; Differentiation
Forkhead transcription factors direct various developmental processes through regulation of tissue-specific expression of target genes. They share a highly conserved DNA binding domain called forkhead domain, based on the founder member in Drosophila [1 –3]. To date, over 100 forkhead genes have been identified in a large variety of species, implicated in diverse biological functions [4,5]. In embryonic development in particular, forkhead factors have been shown to be important in many tissues, including brain [6 –10], pancreas [11], lung [12,13] and kidney [14 – 17]. In addition, roles in cell cycle regulation and survival have been described [18 –20]. The importance of these transcription factors is highlighted by their involvement in a range of human diseases [5,21]. The FoxI1 gene is the only isolated member of the mammalian FoxI subclass of forkhead transcription factors. In mice, Foxi1 (formerly known as fkh10) shows restricted expression in the developing otic vesicle, and mice with a null mutation for Foxi1 exhibit defects in inner ear formation, resulting in hearing and balance impairment [22]. Expression is also present in the developing and adult kidney [23], where Foxi1 targeting leads to disruption of acid/base homeostasis [16]. Parallels between the defects in cells expressing Foxi1 in these tissues suggest a function for Foxi1 in determining cellular identity. * Corresponding author. Tel.: +31 30 2538809; fax: +31 30 2539032. E-mail address: [email protected] (M.P. Smidt). 0167-4781/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.bbaexp.2005.09.003
In zebrafish, the foxi class consist of at least three members [24], of which foxi1 has a conserved role in inner ear development [25 –28]. The non-overlapping expression patterns of zebrafish foxi1-i3b argue for the existence of additional mammalian FoxI family members [24]. Recently, a homologybased search of the mouse genome has resulted in the in silico identification of putative Foxi2 and Foxi3 genes [29]. Expression pattern analysis of Foxi2 and Foxi3 from embryonic day (E) 6.5 to 10.5 suggests an important role for Foxi factors in craniofacial development. In this study, we report the cloning and detailed characterization of the murine Foxi2 gene and describe the expression pattern during the later stages of development which is mainly restricted to subsets of cells in epithelial structures and the developing forebrain. We performed a RT-PCR based screen for forkhead transcription factors expressed in ventral midbrain tissue of adult mice (C57/Bl6), from which total RNA was isolated using TRIzol Reagent (Invitrogen) and reverse transcribed using Superscript II (Invitrogen) in combination with both oligo(dT) and random hexamer primers. A 151-bp fragment was isolated in a PCR (annealing temperature of 45 -C) using degenerate oligonucleotides 5V-MRARGCCVMYCTACTCTTA-3V (forward) and 5V-GCTGGMAGAAYWSYVTKCG-3V (reverse), deduced from a subset of forkhead transcription factors. This fragment showed high sequence homology to the mouse Foxi1 gene, and database homology analysis predicted the existence
134
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138
of a s Foxi-class gene. Based on the sequence information obtained from two mouse-derived ESTs (accession numbers BQ900150 and BB643511), the complete coding sequence ( 72 until STOP codon) was predicted, and isolated from the same cDNA library using a Qiagen One-Step RT-PCR kit (forward primer: 5V-AGGGTGAATGGAGCTCCCATGAGC3V; reverse primer: 5V-TCAAACTTCAGTCCCATCCCGGC3V. The cDNA (Genbank accession number DQ078268) corresponded to the forkhead transcription factor recently identified in silico, and designated as Foxi2 [29]. However, we deviate from their initial design and propose an additional level of complexity in the composition of the Foxi2 gene and the regulation of the resulting protein levels. Fig. 1 shows the Foxi2 cDNA sequence isolated, the coding sequence comprising 933 bp, coding for a 311 amino acid protein, containing a conserved forkhead DNA binding
domain (DBD). Two upstream ATG sequences are located in the 72-bp preceding the putative start codon (dotted underline in Fig. 1), and two additional upstream ATGs can be identified in silico within 130 bp of the putative start codon, none of which are embedded in a favorable context for initiation of translation [30], nor are they conserved in other species (not shown). In contrast, conservation with Foxi2 genes from human (designed in silico based on an EST with accession number AW271332) and zebrafish starts at the designated methionine (see also Fig. 2), coded for by an AUG codon surrounded by preferred nucleotides at key positions for initiation (G at pos 3 and G at pos +4 [30]). Additionally, in the human sequence, no potential methionines can be observed N-terminal to the designated starting methionine (not shown). Thus, whereas previous reports designated the in frame AUG at position 44 as the starting codon, our analysis
Fig. 1. Nucleotide and deduced amino acid sequence of murine Foxi2. Amino acids residues are depicted in single-letter codes. The underlined sequence indicates the forkhead DNA binding domain. The >< refers to the interruption of the two coding exons by a single 658 bp intron. Note the additional ATG sequences upstream of the ATGSTART (dotted underline). Other 5Vand 3Vuntranslated region features referred to in the text have been identified in silico. Nucleotide and deduced amino acid sequence of murine Foxi2. Amino acids residues are depicted in single-letter codes.
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138
135
Fig. 2. Alignment of the FoxI2 proteins from Homo sapiens (hs), Mus musculus (mm) and Danio rerio (dr). Amino acid residues 151 to 244 comprise the forkhead DNA binding domain (boxed in blue). Despite the low overall conservation across all three species, several short conserved regions can be identified (orange boxes), including the N- and C-termini. Note the extended N-terminus of the zebrafish Foxi2 protein, outside the conserved region. Color coding refers to the relative similarity between amino acid residues based on their physiochemical properties (e.g., aromatic residues F, Y and W are depicted in red). Dots represent gaps in the sequence alignment.
suggests that the predicted methionine at position +1 is the likely translational starting site. Although these upstream ATGs are located in a context not likely to adequately function as translational starting sites [30], we cannot exclude the possibility of N-terminal extended isoforms. Nonetheless, upstream AUGs are regarded as an elegant tool to regulate functional protein levels synthesized from mRNA, particularly in genes encoding potent regulators, including transcription factors [31,32]. Tight regulation by such upstream AUGs has been reported for a number of murine transcription factors such as Rx/Rax [33], GATA-1 [34] and C/EBP [35]. Finally, a polyadenylation signal (AATAAA) was identified in silico at 1320 bp of the STOP codon. The mouse Foxi2 gene is located on chromosome 7qF3, where the two coding exons are separated by a 658-bp intron (Fig. 1). The possibility of additional non-coding introns and exons, however, cannot yet be excluded. This chromosomal region is syntenic to human chromosome 10q26.2, where the putative FOXI2 is located. Though several disorders have been linked to this region, there is no evidence yet to support involvement of FOXI2 in any of these. The DNA binding domain of Foxi2 is highly conserved from human to zebrafish, exhibiting 98% and 92% identity relative to murine Foxi2 (Fig. 2). These high conservation levels (>92% identical) also apply to the forkhead domains of the murine homologues, Foxi1 and the putative Foxi3 (predicted in silico based on an EST with accession number
Fig. 3. In situ hybridisation for Foxi2 shows expression in craniofacial structures in E16.5 sagittal cryosections. (A-AV) Expression is present in a subset of cells in the olfactory epithelium. (B-BV) Weak expression was observed in the whiskers. (C-CV) In the tooth, Foxi2 is expressed exclusively in the dental epithelium. (D-DV) Foxi2 mRNA was detected in epithelial cells of the endolymphatic duct of the inner ear. de, dental epithelium; eld, endolymphatic duct; oe, olfactory epithelium; wh, whiskers. Boxes delineate regions magnified in V-figures (e.g., 3CV).
CF736134). Along the entire length, Foxi2 orthologues display a lower level of identity, sharing 79% (putative Homo sapiens) and 60% (Danio rerio) with the murine Foxi2. Despite lower
136
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138
Fig. 4. Foxi2 localization in epithelial structures at E18.5. Expression is visible in the kidney (A-AV) and hair follicles (A, AVV). We also observed staining in the thymus (B-BV). (C-CV) Expression in the mandibular gland, marking cells of the collecting ducts, ending up in the sublingual and submaxillary ducts of the salivary gland (D-DV). hf, hair follicles; kid, kidney; mg, mandibular gland; sg, salivary gland; th, thymus. Boxes delineate regions magnified in V-figures (e.g., 4CV).
levels of conservation outside the forkhead domain, several other short regions exist which show relatively high levels of conservation across all species. These include regions adjacent to the forkhead domain, and N-terminal and C-terminal residues. Interestingly, translation of the putative zebrafish mRNA is thought to initiate at an earlier less beneficial AUG codon (G at pos 3) than the murine and putative human orthologues [24], despite a lack of conservation in the region preceding the conserved methionine (not shown). Compared to the Foxi1 and putative Foxi3, Foxi2 amino acid conservation along the entire length does not exceed 65%. The complete Foxi2 coding cDNA was subcloned into the pGEMT-easy vector (Promega) for digoxigenin labeling in in situ hybridisation experiments. Full-length sense and antisense cRNA probes were synthesized by in vitro transcription using T7- or Sp6-RNA polymerase in combination with a DIG-RNA Labeling kit (Roche). In situ hybridisation was performed on 16 Am cryostat sections as described previously [36]. We have never observed any positive signal using the sense probe. Though forkhead domains are highly conserved between Foxi1 and Foxi2 at the amino acid level, an overall nucleotide conservation level of 53% (82% in the forkhead domain) ensures high specificity for Foxi2 target mRNA. We analyzed the expression patterns of Foxi2 during later stages of mouse embryonic development. From embryonic day 12.5 until day 18.5, there is little variation in the expression pattern. Representative is the expression pattern at E16.5, where expression is present in the olfactory epithelium, albeit only in a subset of cells lining the nasal cavity (Fig. 3A-AV).
Fig. 5. Foxi2 distribution in the central nervous system. Foxi2 specifically marks the developing forebrain at E16.5 (A) including cortical cell layers (AV). Forebrain expression is maintained from E14.5, when faint expression can also be detected in the roof of the midbrain (B-BV). Expression (here at E18.5) is also apparent in the neural layer of the retina (C-CV). cx, cortex; nlr, neural layer of retina; rm, rood of midbrain; tel, telencephalon. Boxes delineate regions magnified in V-figures (e.g., 5CV).
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138
Foxi2 is also faintly expressed in the developing whiskers (Fig. 3B-BV). In the tooth, Foxi2 transcript was detected in the dental epithelium, but was excluded from the dental mesenchyme (Fig. 3C-CV). Elsewhere in the developing cranium, we found expression in cells lining the endolymphatic duct in the inner ear (Fig. 3D-DV). The latter structure was already positive at E12.5 (Not shown), suggesting an age at onset between E10.5, when the otic placode is reportedly negative for Foxi2 [29], and E12.5. No apparent changes in the expression in these regions can be observed at E18.5 (not shown). At E18.5, when certain structures show more detail, expression is observed in the kidney (Fig. 4A-AV), in a regular pattern of the hair follicles (Fig. 4A, AVV) and in the thymus (Fig. 4B-BV). Foxi2 is also detected in the mandibular gland, where it seems to be confined to collecting ducts that transport saliva to the oral cavity (Fig. 4C-CV). This becomes more apparent in the sublingual and submaxillary ducts of the salivary gland, where the speckled pattern suggests expression in a subset of epithelial cells lining the lumen (Fig. 4D-DV). In the developing central nervous system (CNS), restricted Foxi2 expression in the cortical layers and ganglionic eminence of the telencephalon (Fig. 5A-AV) and the preoptic area (Fig. 5B) seems to delineate a boundary between prosomeres p5 and p6 in the diencephalon. This was already observed at E14.5, however, at this stage, there is additional faint expression in the roof of the midbrain (Fig. 5B-BV). Though still present at E16.5, the intensity of the signal is decreased to such extent that it is hardly visible (not shown). Despite the cloning of Foxi2 from adult brain tissue, we could not detect any signal at that stage using in situ hybridisation (not shown), indicating that Foxi2 transcript numbers have decreased to levels undetectable by classic ISH, requiring highly sensitive RTPCR methods for detection. Expression in the CNS is not limited to the brain, since Foxi2 is extensively expressed in the neural layer of the retina (Fig. 5C-CV). We have isolated the Foxi2 gene, a novel member of the murine FoxI class of forkhead transcription factors, and analyzed its expression during late development. We identified Foxi2 as a 936-bp coding sequence, coding for a 311-amino acid protein, with a DNA binding domain that is highly conserved in Foxi1 and the putative Foxi3. Interestingly, Foxi2 is expressed in several tissues positive for Foxi1 during later stages of development such as the endolymphatic duct and kidney. In addition, Foxi2 is expressed in submandibular and salivary ducts. Within these ducts, the composition of the secretion is altered by sodium, potassium and bicarbonate ion exchange, similar to processes taking place in the collecting ducts of the kidney and endolymphatic duct. Strikingly, in the latter structures, Foxi1 is expressed and important for development of cells responsible for the maintenance of acid –base homeostasis [16,37]. It is therefore tempting to speculate that Foxi2 might serve a similar function as Foxi1 in regulating cellular specification of cells that control acid/base homeostasis in these collection ducts. Despite this overlap, however, Foxi1 null mutants show defects in the development of both tissues [16,38], suggesting partially distinct functions of these closely related factors.
137
In the brain, Foxi2 expression seems to delineate a neuromeric boundary between prosomeres p5 and p6 in the diencephalon, supporting a role in patterning of the developing forebrain. Strikingly, Foxi2 is often expressed in epithelial structures, sites of continued birth of cells, all requiring specification of their developmental fate. Altogether, Foxi2 displays a strikingly restricted expression pattern, fuelling speculation that FoxI family members are implicated in developmental regulation of cellular identity. Acknowledgements This work was supported by an NWO grant (903-42-190) and a Korczak Foundation research fellowship. References [1] D. Weigel, H. Jackle, The fork head domain—A novel DNA-binding motif of eukaryotic transcription factors, Cell (1990) 455 – 456. [2] D. Weigel, H.J. Bellen, G. Jurgens, H. Jackle, Primordium specific requirement of the homeotic gene fork head in the developing gut of the drosophila embryo, Roux’s Arch. Dev. Biol. (1989) 201 – 210. [3] E. Lai, V.R. Prezioso, W.F. Tao, W.S. Chen, J.E. Darnell, Hepatocyte nuclear factor-3-alpha belongs to a gene family in mammals that is homologous to the drosophila homeotic gene fork-head, Gene Dev. (1991) 416 – 427. [4] E. Kaufmann, W. Knochel, Five years on the wings of fork head, Mech. Dev. (1996) 3 – 20. [5] P. Carlsson, M. Mahlapuu, Forkhead transcription factors: key players in development and metabolism, Dev. Biol. (2002) 1 – 23. [6] S.H. Xuan, C.A. Baptista, G. Balas, W.F. Tao, V.C. Soares, E. Lai, Winged helix transcription factor Bf-1 is essential for the development of the cerebral hemispheres, Neuron (1995) 1141 – 1152. [7] S. Filosa, J.A. RiveraPerez, A.P. Gomez, A. Gansmuller, H. Sasaki, R.R. Behringer, S.L. Ang, goosecoid and HNF-3 beta genetically interact to regulate neural tube patterning during mouse embryogenesis, Development (1997) 2843 – 2854. [8] P.A. Labosky, G.E. Winnier, T.L. Jetton, L. Hargett, A.K. Ryan, M.G. Rosenfeld, A.F. Parlow, B.L.M. Hogan, The winged helix gene, Mf3, is required for normal development of the diencephalon and midbrain, postnatal growth and the milk-ejection reflex, Development (1997) 1263 – 1274. [9] R. Wehr, A. Mansouri, T. de Maeyer, P. Gruss, Fkh5-deficient mice show dysgenesis in the caudal midbrain and hypothalamic mammillary body, Development (1997) 4447 – 4456. [10] E. Herrera, R. Marcus, S. Li, S.E. Williams, L. Erskine, E. Lai, C. Mason, Foxd1 is required for proper formation of the optic chiasm, Development (2004) 5727 – 5739. [11] K.A. Lantz, K.H. Kaestner, Winged-helix transcription factors and pancreatic, Dev. Clin. Sci. (2005) 195 – 204. [12] D. Warburton, M. Schwarz, D. Tefft, G. Flores-Delgado, K.D. Anderson, W.V. Cardoso, The molecular basis of lung morphogenesis, Mech. Dev. (2000) 55 – 81. [13] R.H. Costa, V.V. Kalinichenko, L. Lim, Transcription factors in mouse lung development and function, Am. J. Physiol., Lung C. (2001) L823 – L838. [14] T. Kume, K.Y. Deng, B.L.M. Hogan, Murine forkhead/winged helix genes Foxc1 (Mf1) and Foxc2 (Mfh1) are required for the early organogenesis of the kidney and urinary tract, Development (2000) 1387 – 1395. [15] T. Kume, K.Y. Deng, B.L.M. Hogan, Minimal phenotype of mice homozygous for a null mutation in the forkhead/winged helix gene, Mf2, Mol. Cell. Biol. (2000) 1419 – 1425. [16] S.R. Blomqvist, H. Vidarsson, S. Fitzgerald, B.R. Johansson, A. Ollerstam, R. Brown, A.E.G. Persson, G. Bergstrom, S. Enerback, Distal renal tubular acidosis in mice that lack the forkhead transcription factor Foxi1, J. Clin. Invest. (2004) 1560 – 1570. [17] R.S. Levinson, E. Batourina, C. Chol, M. Vorontchikhina, J. Kitajewski, C.L. Mendelsohn, Foxd1-dependent signals control cellularity in the renal
138
[18] [19]
[20]
[21] [22]
[23]
[24] [25]
[26]
[27]
[28]
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138 capsule, a structure required for normal renal development, Development (2005) 529 – 539. B.M.T. Burgering, G.J.P.L. Kops, Cell cycle and death control: long live Forkheads, Trends Biochem. Sci. (2002) 352 – 360. J. Laoukili, M.R.H. Kooistra, A. Bras, J. Kauw, R.M. Kerkhoven, A. Morrison, H. Clevers, R.H. Medema, FoxM1 is required for execution of the mitotic programme and chromosome stability, Nat. Cell Biol. (2005) U126 – U134. A. Brunet, S.R. Datta, M.E. Greenberg, Transcription-dependent and -independent control of neuronal survival by the PI3K-Akt signaling pathway, Curr. Opin. Neurobiol. (2001) 297 – 305. O.J. Lehmann, J.C. Sowden, P. Carlsson, T. Jordan, S.S. Bhattacharya, Fox’s in development and disease, Trends Genet. (2003) 339 – 344. M. Hulander, W. Wurst, P. Carlsson, S. Enerback, The winged helix transcription factor Fkh10 is required for normal development of the inner ear, Nat. Genet. (1998) 374 – 376. D.G. Overdier, H.G. Ye, R.S. Peterson, D.E. Clevidence, R.H. Costa, The winged helix transcriptional activator HFH-3 is expressed in the distal tubules of embryonic and adult mouse kidney, J. Biol. Chem. (1997) 13725 – 13730. K.S. Solomon, J.M. Logsdon, A. Fritz, Expression and phylogenetic analyses of three zebrafish foxl class genes, Dev. Dyn. (2003) 301 – 307. S.A. Lee, E.L. Shen, A. Fiser, A. Sali, S. Guo, The zebrafish forkhead transcription factor Foxi1 specifies epibranchial placode-derived sensory neurons, Development (2003) 2669 – 2679. R.M. Nissen, J.Z. Yan, A. Amsterdam, N. Hopkins, S.M. Burgess, Zebrafish foxi one modulates cellular responses to Fgf signaling required for the integrity of ear and jaw patterning, Development (2003) 2543 – 2554. K.S. Solomon, T. Kudoh, I.B. Dawid, A. Fritz, Zebrafish foxi1 mediates otic placode formation and jaw development, Development (2003) 929 – 940. S. Hans, D. Liu, M. Westerfield, Pax8 and Pax2a function synergistically in otic specification, downstream of the Foxi1 and Dlx3b transcription factors, Development (2004) 5091 – 5102.
[29] T. Ohyama, A.K. Groves, Expression of mouse Foxi class genes in early craniofacial development, Dev. Dyn. (2004) 640 – 646. [30] M. Kozak, Interpreting cDNA sequences: some insights from studies on translation, Mamm. Genome (1996) 563 – 574. [31] M. Kozak, Pushing the limits of the scanning mechanism for initiation of translation, Gene (2002) 1 – 34. [32] H.A. Meijer, A.A.M. Thomas, Control of eukaryotic protein synthesis by upstream open reading frames in the 5V-untranslated region of an mRNA, Biochem. J. (2002) 1 – 11. [33] P. Tucker, L. Laemle, A. Munson, S. Kanekar, E.R. Oliver, N. Brown, H. Schlecht, M. Vetter, T. Glaser, The eyeless mouse mutation (ey1) removes an alternative start codon from the Rx/rax homeobox gene, Genesis (2001) 43 – 53. [34] R. Calligaris, S. Bottardi, S. Cogoi, I. Apezteguia, C. Santoro, Alternative translation initiation site usage results in 2 functionally distinct forms of the gata-1 transcription factor, Proc. Natl. Acad. Sci. U. S. A. (1995) 11598 – 11602. [35] V. Ossipow, P. Descombes, U. Schibler, CCAAT enhancer-binding protein messenger-RNA is translated into multiple proteins with different transcription activation potentials, Proc. Natl. Acad. Sci. U. S. A. (1993) 8219 – 8223. [36] F.M.J. Jacobs, L.P. van der Heide, P.J.E.C. Wijchers, J.P.H. Burbach, M.F.M. Hoekman, M.P. Smidt, FoxO6, a novel member of the FoxO class of transcription factors with distinct shuttling dynamics, J. Biol. Chem. (2003) 35959 – 35967. [37] M. Hulander, A.E. Kiernan, S.R. Blomqvist, P. Carlsson, E.J. Samuelsson, B.R. Johansson, K.P. Steel, S. Enerback, Lack of pendrin expression leads to deafness and expansion of the endolymphatic compartment in inner ears of Foxi1 null mutant mice, Development (2003) 2013 – 2025. [38] M. Hulander, A.E. Kiernan, S.R. Blomqvist, P. Carlsson, E.J. Samuelsson, B.R. Johansson, K.P. Steel, S. Enerback, Lack of pendrin expression leads to deafness and expansion of the endolymphatic compartment in inner ears of Foxi1 null mutant mice, Development (2003) 2013 – 2025.
Short sequence paper
Cloning and analysis of the murine Foxi2 transcription factor Patrick J.E.C. Wijchers, Marco F.M. Hoekman, J. Peter H. Burbach, Marten P. Smidt * Rudolf Magnus Institute of Neuroscience, Department of Pharmacology and Anatomy, University Medical Center Utrecht, Universiteitsweg 100, 3584 CG Utrecht, The Netherlands Received 14 June 2005; received in revised form 31 August 2005; accepted 13 September 2005 Available online 5 October 2005
Abstract Forkhead transcription factors comprise a large family of key regulators of embryonic development. Here, we describe the cloning and analysis of the murine Foxi2 gene, coding for a putative 311 amino acid protein resembling Foxi subfamily members in mice and other species. Expression analysis during the final stages of embryonic development revealed that Foxi2 expression is mainly confined to subsets of cells in epithelial structures and particular ducts, in addition to the developing forebrain and neural retina. Since FoxI factors are thought to be implicated in the regulation of cell fate, the highly restricted expression pattern of Foxi2 suggestive of a possible role in controlling cellular identity. D 2005 Elsevier B.V. All rights reserved. Keywords: Forkhead; Transcription factor; Foxi2; Development; Cell identity; Differentiation
Forkhead transcription factors direct various developmental processes through regulation of tissue-specific expression of target genes. They share a highly conserved DNA binding domain called forkhead domain, based on the founder member in Drosophila [1 –3]. To date, over 100 forkhead genes have been identified in a large variety of species, implicated in diverse biological functions [4,5]. In embryonic development in particular, forkhead factors have been shown to be important in many tissues, including brain [6 –10], pancreas [11], lung [12,13] and kidney [14 – 17]. In addition, roles in cell cycle regulation and survival have been described [18 –20]. The importance of these transcription factors is highlighted by their involvement in a range of human diseases [5,21]. The FoxI1 gene is the only isolated member of the mammalian FoxI subclass of forkhead transcription factors. In mice, Foxi1 (formerly known as fkh10) shows restricted expression in the developing otic vesicle, and mice with a null mutation for Foxi1 exhibit defects in inner ear formation, resulting in hearing and balance impairment [22]. Expression is also present in the developing and adult kidney [23], where Foxi1 targeting leads to disruption of acid/base homeostasis [16]. Parallels between the defects in cells expressing Foxi1 in these tissues suggest a function for Foxi1 in determining cellular identity. * Corresponding author. Tel.: +31 30 2538809; fax: +31 30 2539032. E-mail address: [email protected] (M.P. Smidt). 0167-4781/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.bbaexp.2005.09.003
In zebrafish, the foxi class consist of at least three members [24], of which foxi1 has a conserved role in inner ear development [25 –28]. The non-overlapping expression patterns of zebrafish foxi1-i3b argue for the existence of additional mammalian FoxI family members [24]. Recently, a homologybased search of the mouse genome has resulted in the in silico identification of putative Foxi2 and Foxi3 genes [29]. Expression pattern analysis of Foxi2 and Foxi3 from embryonic day (E) 6.5 to 10.5 suggests an important role for Foxi factors in craniofacial development. In this study, we report the cloning and detailed characterization of the murine Foxi2 gene and describe the expression pattern during the later stages of development which is mainly restricted to subsets of cells in epithelial structures and the developing forebrain. We performed a RT-PCR based screen for forkhead transcription factors expressed in ventral midbrain tissue of adult mice (C57/Bl6), from which total RNA was isolated using TRIzol Reagent (Invitrogen) and reverse transcribed using Superscript II (Invitrogen) in combination with both oligo(dT) and random hexamer primers. A 151-bp fragment was isolated in a PCR (annealing temperature of 45 -C) using degenerate oligonucleotides 5V-MRARGCCVMYCTACTCTTA-3V (forward) and 5V-GCTGGMAGAAYWSYVTKCG-3V (reverse), deduced from a subset of forkhead transcription factors. This fragment showed high sequence homology to the mouse Foxi1 gene, and database homology analysis predicted the existence
134
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138
of a s Foxi-class gene. Based on the sequence information obtained from two mouse-derived ESTs (accession numbers BQ900150 and BB643511), the complete coding sequence ( 72 until STOP codon) was predicted, and isolated from the same cDNA library using a Qiagen One-Step RT-PCR kit (forward primer: 5V-AGGGTGAATGGAGCTCCCATGAGC3V; reverse primer: 5V-TCAAACTTCAGTCCCATCCCGGC3V. The cDNA (Genbank accession number DQ078268) corresponded to the forkhead transcription factor recently identified in silico, and designated as Foxi2 [29]. However, we deviate from their initial design and propose an additional level of complexity in the composition of the Foxi2 gene and the regulation of the resulting protein levels. Fig. 1 shows the Foxi2 cDNA sequence isolated, the coding sequence comprising 933 bp, coding for a 311 amino acid protein, containing a conserved forkhead DNA binding
domain (DBD). Two upstream ATG sequences are located in the 72-bp preceding the putative start codon (dotted underline in Fig. 1), and two additional upstream ATGs can be identified in silico within 130 bp of the putative start codon, none of which are embedded in a favorable context for initiation of translation [30], nor are they conserved in other species (not shown). In contrast, conservation with Foxi2 genes from human (designed in silico based on an EST with accession number AW271332) and zebrafish starts at the designated methionine (see also Fig. 2), coded for by an AUG codon surrounded by preferred nucleotides at key positions for initiation (G at pos 3 and G at pos +4 [30]). Additionally, in the human sequence, no potential methionines can be observed N-terminal to the designated starting methionine (not shown). Thus, whereas previous reports designated the in frame AUG at position 44 as the starting codon, our analysis
Fig. 1. Nucleotide and deduced amino acid sequence of murine Foxi2. Amino acids residues are depicted in single-letter codes. The underlined sequence indicates the forkhead DNA binding domain. The >< refers to the interruption of the two coding exons by a single 658 bp intron. Note the additional ATG sequences upstream of the ATGSTART (dotted underline). Other 5Vand 3Vuntranslated region features referred to in the text have been identified in silico. Nucleotide and deduced amino acid sequence of murine Foxi2. Amino acids residues are depicted in single-letter codes.
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138
135
Fig. 2. Alignment of the FoxI2 proteins from Homo sapiens (hs), Mus musculus (mm) and Danio rerio (dr). Amino acid residues 151 to 244 comprise the forkhead DNA binding domain (boxed in blue). Despite the low overall conservation across all three species, several short conserved regions can be identified (orange boxes), including the N- and C-termini. Note the extended N-terminus of the zebrafish Foxi2 protein, outside the conserved region. Color coding refers to the relative similarity between amino acid residues based on their physiochemical properties (e.g., aromatic residues F, Y and W are depicted in red). Dots represent gaps in the sequence alignment.
suggests that the predicted methionine at position +1 is the likely translational starting site. Although these upstream ATGs are located in a context not likely to adequately function as translational starting sites [30], we cannot exclude the possibility of N-terminal extended isoforms. Nonetheless, upstream AUGs are regarded as an elegant tool to regulate functional protein levels synthesized from mRNA, particularly in genes encoding potent regulators, including transcription factors [31,32]. Tight regulation by such upstream AUGs has been reported for a number of murine transcription factors such as Rx/Rax [33], GATA-1 [34] and C/EBP [35]. Finally, a polyadenylation signal (AATAAA) was identified in silico at 1320 bp of the STOP codon. The mouse Foxi2 gene is located on chromosome 7qF3, where the two coding exons are separated by a 658-bp intron (Fig. 1). The possibility of additional non-coding introns and exons, however, cannot yet be excluded. This chromosomal region is syntenic to human chromosome 10q26.2, where the putative FOXI2 is located. Though several disorders have been linked to this region, there is no evidence yet to support involvement of FOXI2 in any of these. The DNA binding domain of Foxi2 is highly conserved from human to zebrafish, exhibiting 98% and 92% identity relative to murine Foxi2 (Fig. 2). These high conservation levels (>92% identical) also apply to the forkhead domains of the murine homologues, Foxi1 and the putative Foxi3 (predicted in silico based on an EST with accession number
Fig. 3. In situ hybridisation for Foxi2 shows expression in craniofacial structures in E16.5 sagittal cryosections. (A-AV) Expression is present in a subset of cells in the olfactory epithelium. (B-BV) Weak expression was observed in the whiskers. (C-CV) In the tooth, Foxi2 is expressed exclusively in the dental epithelium. (D-DV) Foxi2 mRNA was detected in epithelial cells of the endolymphatic duct of the inner ear. de, dental epithelium; eld, endolymphatic duct; oe, olfactory epithelium; wh, whiskers. Boxes delineate regions magnified in V-figures (e.g., 3CV).
CF736134). Along the entire length, Foxi2 orthologues display a lower level of identity, sharing 79% (putative Homo sapiens) and 60% (Danio rerio) with the murine Foxi2. Despite lower
136
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138
Fig. 4. Foxi2 localization in epithelial structures at E18.5. Expression is visible in the kidney (A-AV) and hair follicles (A, AVV). We also observed staining in the thymus (B-BV). (C-CV) Expression in the mandibular gland, marking cells of the collecting ducts, ending up in the sublingual and submaxillary ducts of the salivary gland (D-DV). hf, hair follicles; kid, kidney; mg, mandibular gland; sg, salivary gland; th, thymus. Boxes delineate regions magnified in V-figures (e.g., 4CV).
levels of conservation outside the forkhead domain, several other short regions exist which show relatively high levels of conservation across all species. These include regions adjacent to the forkhead domain, and N-terminal and C-terminal residues. Interestingly, translation of the putative zebrafish mRNA is thought to initiate at an earlier less beneficial AUG codon (G at pos 3) than the murine and putative human orthologues [24], despite a lack of conservation in the region preceding the conserved methionine (not shown). Compared to the Foxi1 and putative Foxi3, Foxi2 amino acid conservation along the entire length does not exceed 65%. The complete Foxi2 coding cDNA was subcloned into the pGEMT-easy vector (Promega) for digoxigenin labeling in in situ hybridisation experiments. Full-length sense and antisense cRNA probes were synthesized by in vitro transcription using T7- or Sp6-RNA polymerase in combination with a DIG-RNA Labeling kit (Roche). In situ hybridisation was performed on 16 Am cryostat sections as described previously [36]. We have never observed any positive signal using the sense probe. Though forkhead domains are highly conserved between Foxi1 and Foxi2 at the amino acid level, an overall nucleotide conservation level of 53% (82% in the forkhead domain) ensures high specificity for Foxi2 target mRNA. We analyzed the expression patterns of Foxi2 during later stages of mouse embryonic development. From embryonic day 12.5 until day 18.5, there is little variation in the expression pattern. Representative is the expression pattern at E16.5, where expression is present in the olfactory epithelium, albeit only in a subset of cells lining the nasal cavity (Fig. 3A-AV).
Fig. 5. Foxi2 distribution in the central nervous system. Foxi2 specifically marks the developing forebrain at E16.5 (A) including cortical cell layers (AV). Forebrain expression is maintained from E14.5, when faint expression can also be detected in the roof of the midbrain (B-BV). Expression (here at E18.5) is also apparent in the neural layer of the retina (C-CV). cx, cortex; nlr, neural layer of retina; rm, rood of midbrain; tel, telencephalon. Boxes delineate regions magnified in V-figures (e.g., 5CV).
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138
Foxi2 is also faintly expressed in the developing whiskers (Fig. 3B-BV). In the tooth, Foxi2 transcript was detected in the dental epithelium, but was excluded from the dental mesenchyme (Fig. 3C-CV). Elsewhere in the developing cranium, we found expression in cells lining the endolymphatic duct in the inner ear (Fig. 3D-DV). The latter structure was already positive at E12.5 (Not shown), suggesting an age at onset between E10.5, when the otic placode is reportedly negative for Foxi2 [29], and E12.5. No apparent changes in the expression in these regions can be observed at E18.5 (not shown). At E18.5, when certain structures show more detail, expression is observed in the kidney (Fig. 4A-AV), in a regular pattern of the hair follicles (Fig. 4A, AVV) and in the thymus (Fig. 4B-BV). Foxi2 is also detected in the mandibular gland, where it seems to be confined to collecting ducts that transport saliva to the oral cavity (Fig. 4C-CV). This becomes more apparent in the sublingual and submaxillary ducts of the salivary gland, where the speckled pattern suggests expression in a subset of epithelial cells lining the lumen (Fig. 4D-DV). In the developing central nervous system (CNS), restricted Foxi2 expression in the cortical layers and ganglionic eminence of the telencephalon (Fig. 5A-AV) and the preoptic area (Fig. 5B) seems to delineate a boundary between prosomeres p5 and p6 in the diencephalon. This was already observed at E14.5, however, at this stage, there is additional faint expression in the roof of the midbrain (Fig. 5B-BV). Though still present at E16.5, the intensity of the signal is decreased to such extent that it is hardly visible (not shown). Despite the cloning of Foxi2 from adult brain tissue, we could not detect any signal at that stage using in situ hybridisation (not shown), indicating that Foxi2 transcript numbers have decreased to levels undetectable by classic ISH, requiring highly sensitive RTPCR methods for detection. Expression in the CNS is not limited to the brain, since Foxi2 is extensively expressed in the neural layer of the retina (Fig. 5C-CV). We have isolated the Foxi2 gene, a novel member of the murine FoxI class of forkhead transcription factors, and analyzed its expression during late development. We identified Foxi2 as a 936-bp coding sequence, coding for a 311-amino acid protein, with a DNA binding domain that is highly conserved in Foxi1 and the putative Foxi3. Interestingly, Foxi2 is expressed in several tissues positive for Foxi1 during later stages of development such as the endolymphatic duct and kidney. In addition, Foxi2 is expressed in submandibular and salivary ducts. Within these ducts, the composition of the secretion is altered by sodium, potassium and bicarbonate ion exchange, similar to processes taking place in the collecting ducts of the kidney and endolymphatic duct. Strikingly, in the latter structures, Foxi1 is expressed and important for development of cells responsible for the maintenance of acid –base homeostasis [16,37]. It is therefore tempting to speculate that Foxi2 might serve a similar function as Foxi1 in regulating cellular specification of cells that control acid/base homeostasis in these collection ducts. Despite this overlap, however, Foxi1 null mutants show defects in the development of both tissues [16,38], suggesting partially distinct functions of these closely related factors.
137
In the brain, Foxi2 expression seems to delineate a neuromeric boundary between prosomeres p5 and p6 in the diencephalon, supporting a role in patterning of the developing forebrain. Strikingly, Foxi2 is often expressed in epithelial structures, sites of continued birth of cells, all requiring specification of their developmental fate. Altogether, Foxi2 displays a strikingly restricted expression pattern, fuelling speculation that FoxI family members are implicated in developmental regulation of cellular identity. Acknowledgements This work was supported by an NWO grant (903-42-190) and a Korczak Foundation research fellowship. References [1] D. Weigel, H. Jackle, The fork head domain—A novel DNA-binding motif of eukaryotic transcription factors, Cell (1990) 455 – 456. [2] D. Weigel, H.J. Bellen, G. Jurgens, H. Jackle, Primordium specific requirement of the homeotic gene fork head in the developing gut of the drosophila embryo, Roux’s Arch. Dev. Biol. (1989) 201 – 210. [3] E. Lai, V.R. Prezioso, W.F. Tao, W.S. Chen, J.E. Darnell, Hepatocyte nuclear factor-3-alpha belongs to a gene family in mammals that is homologous to the drosophila homeotic gene fork-head, Gene Dev. (1991) 416 – 427. [4] E. Kaufmann, W. Knochel, Five years on the wings of fork head, Mech. Dev. (1996) 3 – 20. [5] P. Carlsson, M. Mahlapuu, Forkhead transcription factors: key players in development and metabolism, Dev. Biol. (2002) 1 – 23. [6] S.H. Xuan, C.A. Baptista, G. Balas, W.F. Tao, V.C. Soares, E. Lai, Winged helix transcription factor Bf-1 is essential for the development of the cerebral hemispheres, Neuron (1995) 1141 – 1152. [7] S. Filosa, J.A. RiveraPerez, A.P. Gomez, A. Gansmuller, H. Sasaki, R.R. Behringer, S.L. Ang, goosecoid and HNF-3 beta genetically interact to regulate neural tube patterning during mouse embryogenesis, Development (1997) 2843 – 2854. [8] P.A. Labosky, G.E. Winnier, T.L. Jetton, L. Hargett, A.K. Ryan, M.G. Rosenfeld, A.F. Parlow, B.L.M. Hogan, The winged helix gene, Mf3, is required for normal development of the diencephalon and midbrain, postnatal growth and the milk-ejection reflex, Development (1997) 1263 – 1274. [9] R. Wehr, A. Mansouri, T. de Maeyer, P. Gruss, Fkh5-deficient mice show dysgenesis in the caudal midbrain and hypothalamic mammillary body, Development (1997) 4447 – 4456. [10] E. Herrera, R. Marcus, S. Li, S.E. Williams, L. Erskine, E. Lai, C. Mason, Foxd1 is required for proper formation of the optic chiasm, Development (2004) 5727 – 5739. [11] K.A. Lantz, K.H. Kaestner, Winged-helix transcription factors and pancreatic, Dev. Clin. Sci. (2005) 195 – 204. [12] D. Warburton, M. Schwarz, D. Tefft, G. Flores-Delgado, K.D. Anderson, W.V. Cardoso, The molecular basis of lung morphogenesis, Mech. Dev. (2000) 55 – 81. [13] R.H. Costa, V.V. Kalinichenko, L. Lim, Transcription factors in mouse lung development and function, Am. J. Physiol., Lung C. (2001) L823 – L838. [14] T. Kume, K.Y. Deng, B.L.M. Hogan, Murine forkhead/winged helix genes Foxc1 (Mf1) and Foxc2 (Mfh1) are required for the early organogenesis of the kidney and urinary tract, Development (2000) 1387 – 1395. [15] T. Kume, K.Y. Deng, B.L.M. Hogan, Minimal phenotype of mice homozygous for a null mutation in the forkhead/winged helix gene, Mf2, Mol. Cell. Biol. (2000) 1419 – 1425. [16] S.R. Blomqvist, H. Vidarsson, S. Fitzgerald, B.R. Johansson, A. Ollerstam, R. Brown, A.E.G. Persson, G. Bergstrom, S. Enerback, Distal renal tubular acidosis in mice that lack the forkhead transcription factor Foxi1, J. Clin. Invest. (2004) 1560 – 1570. [17] R.S. Levinson, E. Batourina, C. Chol, M. Vorontchikhina, J. Kitajewski, C.L. Mendelsohn, Foxd1-dependent signals control cellularity in the renal
138
[18] [19]
[20]
[21] [22]
[23]
[24] [25]
[26]
[27]
[28]
P.J.E.C. Wijchers et al. / Biochimica et Biophysica Acta 1731 (2005) 133 – 138 capsule, a structure required for normal renal development, Development (2005) 529 – 539. B.M.T. Burgering, G.J.P.L. Kops, Cell cycle and death control: long live Forkheads, Trends Biochem. Sci. (2002) 352 – 360. J. Laoukili, M.R.H. Kooistra, A. Bras, J. Kauw, R.M. Kerkhoven, A. Morrison, H. Clevers, R.H. Medema, FoxM1 is required for execution of the mitotic programme and chromosome stability, Nat. Cell Biol. (2005) U126 – U134. A. Brunet, S.R. Datta, M.E. Greenberg, Transcription-dependent and -independent control of neuronal survival by the PI3K-Akt signaling pathway, Curr. Opin. Neurobiol. (2001) 297 – 305. O.J. Lehmann, J.C. Sowden, P. Carlsson, T. Jordan, S.S. Bhattacharya, Fox’s in development and disease, Trends Genet. (2003) 339 – 344. M. Hulander, W. Wurst, P. Carlsson, S. Enerback, The winged helix transcription factor Fkh10 is required for normal development of the inner ear, Nat. Genet. (1998) 374 – 376. D.G. Overdier, H.G. Ye, R.S. Peterson, D.E. Clevidence, R.H. Costa, The winged helix transcriptional activator HFH-3 is expressed in the distal tubules of embryonic and adult mouse kidney, J. Biol. Chem. (1997) 13725 – 13730. K.S. Solomon, J.M. Logsdon, A. Fritz, Expression and phylogenetic analyses of three zebrafish foxl class genes, Dev. Dyn. (2003) 301 – 307. S.A. Lee, E.L. Shen, A. Fiser, A. Sali, S. Guo, The zebrafish forkhead transcription factor Foxi1 specifies epibranchial placode-derived sensory neurons, Development (2003) 2669 – 2679. R.M. Nissen, J.Z. Yan, A. Amsterdam, N. Hopkins, S.M. Burgess, Zebrafish foxi one modulates cellular responses to Fgf signaling required for the integrity of ear and jaw patterning, Development (2003) 2543 – 2554. K.S. Solomon, T. Kudoh, I.B. Dawid, A. Fritz, Zebrafish foxi1 mediates otic placode formation and jaw development, Development (2003) 929 – 940. S. Hans, D. Liu, M. Westerfield, Pax8 and Pax2a function synergistically in otic specification, downstream of the Foxi1 and Dlx3b transcription factors, Development (2004) 5091 – 5102.
[29] T. Ohyama, A.K. Groves, Expression of mouse Foxi class genes in early craniofacial development, Dev. Dyn. (2004) 640 – 646. [30] M. Kozak, Interpreting cDNA sequences: some insights from studies on translation, Mamm. Genome (1996) 563 – 574. [31] M. Kozak, Pushing the limits of the scanning mechanism for initiation of translation, Gene (2002) 1 – 34. [32] H.A. Meijer, A.A.M. Thomas, Control of eukaryotic protein synthesis by upstream open reading frames in the 5V-untranslated region of an mRNA, Biochem. J. (2002) 1 – 11. [33] P. Tucker, L. Laemle, A. Munson, S. Kanekar, E.R. Oliver, N. Brown, H. Schlecht, M. Vetter, T. Glaser, The eyeless mouse mutation (ey1) removes an alternative start codon from the Rx/rax homeobox gene, Genesis (2001) 43 – 53. [34] R. Calligaris, S. Bottardi, S. Cogoi, I. Apezteguia, C. Santoro, Alternative translation initiation site usage results in 2 functionally distinct forms of the gata-1 transcription factor, Proc. Natl. Acad. Sci. U. S. A. (1995) 11598 – 11602. [35] V. Ossipow, P. Descombes, U. Schibler, CCAAT enhancer-binding protein messenger-RNA is translated into multiple proteins with different transcription activation potentials, Proc. Natl. Acad. Sci. U. S. A. (1993) 8219 – 8223. [36] F.M.J. Jacobs, L.P. van der Heide, P.J.E.C. Wijchers, J.P.H. Burbach, M.F.M. Hoekman, M.P. Smidt, FoxO6, a novel member of the FoxO class of transcription factors with distinct shuttling dynamics, J. Biol. Chem. (2003) 35959 – 35967. [37] M. Hulander, A.E. Kiernan, S.R. Blomqvist, P. Carlsson, E.J. Samuelsson, B.R. Johansson, K.P. Steel, S. Enerback, Lack of pendrin expression leads to deafness and expansion of the endolymphatic compartment in inner ears of Foxi1 null mutant mice, Development (2003) 2013 – 2025. [38] M. Hulander, A.E. Kiernan, S.R. Blomqvist, P. Carlsson, E.J. Samuelsson, B.R. Johansson, K.P. Steel, S. Enerback, Lack of pendrin expression leads to deafness and expansion of the endolymphatic compartment in inner ears of Foxi1 null mutant mice, Development (2003) 2013 – 2025.