crocodile

EVOLUTIONARY HOMOLOGS

Drosophila sloppy-paired 1 and 2 both have Forkhead domains (Hacker, 1992). Sloppy paired belongs to a different class of forkhead domain proteins than does either Forkhead itself or HNF-3beta. Sloppy paired's closest homolog is BF-1. The crocodile gene is required for the establishment of head structures. Crocodile's closest mammalian homolog is FD1. Forkhead belongs to the same class as HNF-3alpha, HNF-3beta, HNF-3gamma, XFKH1/XFD1, and XFD1/pintallavis (Sasaki, 1993 and Hacker, 1995).

Six forkhead gene family members have been cloned from a mouse genomic library in addition to the mouse equivalents of the genes for HNF3 alpha, -beta, and -gamma. The six genes, termed fkh-1 to fkh-6, share a high degree of similarity with the Drosophila forkhead gene, having 57-67% amino acid identity within the forkhead domain. fkh-1 seems to be the mammalian homolog of the Drosophila FD1 gene (crocodile), since the sequences are 86% identical. fkh-1 to fkh-6 show distinct spatial patterns of expression in adult tissues and are expressed during embryogenesis (Kaestner, 1993).

A mouse gene, MFH-1 (mesenchyme fork head 1) is related to the Drosophila fork head and rat HNF3 genes. MFH-1 encodes a distinct fork head domain that is classified into a distinct subfamily. A recombinant MFH-1 protein can bind to the HNF3 binding site. MFH-1 is expressed temporally in developing embryos, first in the non-notochordal mesoderm and later in areas of mesenchymal condensation in the trunk, head, and limbs. These results suggest that MFH-1 might be involved in the formation of special mesenchymal tissues (Miura, 1993).

Both mouse fkh-6 and MFH-1 genes are expressed in embryonic mesoderm from the headfold stage onward. Transcripts for both genes are localized mainly to mesenchymal tissues; fkh-6 mRNA is enriched in the mesenchyme of the gut, lung, tongue and head, whereas MFH-1 is expressed in somitic mesoderm, in the endocardium and blood vessels as well as the condensing mesenchyme of the bones and kidney and in head mesenchyme. Both genes are located within a 10 kb region (on mouse chromosome 8 at 5.26 +/- 2.56 cM telomeric to Actsk1. The close physical linkage of these two winged helix genes is conserved in humans, where the two genes map to chromosome 16q22-24. This tandem arrangement suggests the common use of regulatory mechanisms. The fkh-6/MFH-1 locus maps close to the mouse mutation amputated, which is characterised by abnormal development of somitic and facial mesoderm. Based on the expression patterns it is suggested that a mutation in MFH-1, not fkh-6 is the possible cause for the amputated phenotype (Kaestner, 1996).

A member of the fork head family, the mesenchyme fork head-1 (MFH-1) gene, is expressed in developing mesenchyme. The mouse (MFH-1) and human (FKHL14) chromosomal MFH-1 genes have been isolated and the gene and protein structures of MFH-1 determined. The MFH-1 gene has no introns and the identity of the amino acid sequences of mouse and human MFH-1 proteins is 94%. Both mouse and human MFH-1 proteins act as positive transactivators (Miura, 1997).

The gene mfh1, encoding a winged helix/forkhead domain transcription factor, is expressed in a dynamic pattern in paraxial and presomitic mesoderm and developing somites during mouse embryogenesis. Expression later becomes restricted to condensing mesenchyme of the vertebrae, head, limbs, and kidney. A targeted disruption of the gene was generated by homologous recombination in embryonic stem cells. Most homozygous mfh1 null embryos die prenatally but some survive to birth, with multiple craniofacial and vertebral column defects. Using molecular markers, it has been shown that the initial formation and patterning of somites occurs normally in mutants. Differentiation of sclerotome-derived cells also appears unaffected, although a reduction of the level of some markers [e.g., mtwist, mf1, scleraxis, and alpha1(II) collagen] is seen in the anterior of homozygous mutants. The most significant difference, however, is a marked reduction in the proliferation of sclerotome-derived cells, as judged by BrdU incorporation. This proliferation defect is also seen in micromass cultures of somite-derived cells treated with transforming growth factor beta1 and fibroblast growth factors. These findings establish a requirement for a winged helix/forkhead domain transcription factor in the development of the paraxial mesoderm. A model is proposed for the role of mfh1 in regulating the proliferation and differentiation of cell lineages giving rise to the axial skeleton and skull (Winnier, 1997).

The 'winged helix' or 'forkhead' transcription factor gene family is defined by a common 100-amino-acid DNA-binding motif. The chromosomal position, start site of transcription, sequence and adult expression pattern of the mouse Fkh1/Mf1 (Forkhead homolog 1/mesoderm/mesenchyme forkhead1) gene is described. This gene contains one exon and encodes a protein of 553 amino acids that is highly related to the mouse MFH1 protein. Fkh1/Mf1 mRNA is expressed widely in adult tissues. Linkage analysis shows that the Fkh1/Mf1 gene is localized to chromosome 13 at 17.02cM from the centromer, in close proximity to Bmp6 and Hfh1, another distinct member of the winged helix gene family (Hiemisch, 1998a).

The expression patterns of the mouse Forkhead homolog 1/mesoderm/mesenchyme forkhead 1 (Fkh1/Mf1) gene were compared with those of the highly related winged helix gene Mfh1 in late gestation mouse embryos. Transcripts for both genes are restricted to derivatives of the mesoderm. Co-expression is found in cartilage primordia of the head, ribs, vertebra and bones. However, in several structures analyzed, Fkh1/Mf1 signals are lower in the inner layers of the developing cartilage than those of Mfh1 (Hiemisch, 1998b).

A cDNA encoding a novel Xenopus winged helix transcription factor termed XFD-11 (Xenopus fork head domain) has been identified. The DNA binding domain is most closely related to those of human or murine FREAC-3 (FKHL7/MF-1/FKH-1) proteins. The XFD-11 gene is activated at late blastula/early gastrula and transcription proceeds throughout embryogenesis. Early expression is found in ventral and lateral but not in dorsal mesoderm. At neurula stages, transcripts are found in posterior mesoderm except for the dorsal midline, and the gene is also transcribed at the lateral border of the neural plate and within anterior neuroectoderm. At later stages of development, transcripts are detected within the pronephros, the heart, within neural crest cells surrounding the eye, in the mandibular, hyoid and branchial arches, and within the tail (Koster, 1998).

Genetic linkage, genome mismatch scanning, and analysis of patients with alterations of chromosome 6 have indicated that a major locus for development of the anterior segment of the eye, IRID1, is located at 6p25. Abnormalities of this locus lead to glaucoma. FKHL7 (also called 'FREAC3'), a member of the forkhead/winged-helix transcription-factor family, has also been mapped to 6p25. DNA sequencing of FKHL7 in five IRID1 families and 16 sporadic patients with anterior-segment defects reveal three mutations: a 10-bp deletion predicted to cause a frameshift and premature protein truncation prior to the FKHL7 forkhead DNA-binding domain, as well as two missense mutations of conserved amino acids within the FKHL7 forkhead domain. Mf1, the murine homolog of FKHL7, is expressed in the developing brain, skeletal system, and eye, consistent with FKHL7 having a role in ocular development. However, mutational screening and genetic-linkage analyses exclude FKHL7 from underlying the anterior-segment disorders in two IRID1 families with linkage to 6p25. These findings demonstrate that, although mutations of FKHL7 result in anterior-segment defects and glaucoma in some patients, it is probable that at least one more locus involved in the regulation of eye development is also located at 6p25 (Mears, 1998).

Mf1 encodes a forkhead/winged helix transcription factor expressed in many embryonic tissues, including prechondrogenic mesenchyme, periocular mesenchyme, meninges, endothelial cells, and kidney. Homozygous null Mf1lacZ mice die at birth with hydrocephalus, eye defects, and multiple skeletal abnormalities identical to those of the classical mutant, congenital hydrocephalus. Congenital hydrocephalus involves a point mutation in Mf1, generating a truncated protein lacking the DNA-binding domain. Mesenchyme cells from Mf1lacZ embryos differentiate poorly into cartilage in micromass culture and do not respond to added BMP2 and TGFbeta1. The differentiation of arachnoid cells in the mutant meninges is also abnormal. The human Mf1 homolog FREAC3 is a candidate gene for ocular dysgenesis and glaucoma mapping to chromosome 6p25-pter, and deletions of this region are associated with multiple developmental disorders, including hydrocephaly and eye defects (Kume, 1998).

A number of different eye disorders with the presence of early-onset glaucoma as a component of the phenotype have been mapped to human chromosome 6p25. These disorders have been postulated to be either allelic to one another or associated with a cluster of tightly linked genes. Two primary congenital glaucoma (PCG) patients with chromosomal anomalies involving 6p25 have been identified. In order to identify a gene involved in PCG, the chromosomal breakpoints in a patient with a balanced translocation between 6p25 and 13q22 were cloned. Cloning of the 6p25 breakpoint led to the identification of two candidate genes based on proximity to the breakpoint. One of these, FKHL7, encoding a forkhead transcription factor, is in close proximity to the breakpoint in the balanced translocation patient and is deleted in a second PCG patient with partial 6p monosomy. Furthermore, FKHL7 harbors mutations in patients diagnosed with Rieger anomaly (RA), Axenfeld anomaly (AA) and iris hypoplasia (IH). This study demonstrates that mutations in FKHL7 cause a spectrum of glaucoma phenotypes (Nishimura, 1998).

Congenital hydrocephalus is an etiologically diverse, poorly understood, but relatively common birth defect. Most human cases are sporadic with familial forms showing considerable phenotypic and etiologic heterogeneity. The autosomal recessive mouse mutation congenital hydrocephalus has been studied to identify candidate human hydrocephalus genes and their modifiers. ch mice have a congenital, lethal hydrocephalus in association with multiple developmental defects, notably skeletal defects, in tissues derived from the cephalic neural crest. Positional cloning methods were used to map ch in the vicinity of D13Mit294 and confirm that the ch phenotype is caused by homozygosity for a nonsense mutation in a gene encoding a winged helix/forkhead transcription factor(Mf1). Based on linked genetic markers, detailed phenotypic characterization of mutant homozygotes and heterozygotes was performed to demonstrate the pleiotropic effects of the mutant gene. Surprisingly, ch heterozygotes have the glaucoma-related distinct phenotype of multiple anterior segment defects resembling Axenfeld-Rieger anomaly. A second member of this gene family (Hfh1), a candidate for other developmental defects, is found approximately 470 kb proximal to Mf1 (Hong, 1999).

Mf1, which encodes a winged-helix/forkhead transcription factor related to Drosophila Crocodile, is the murine homolog of human FKHL7, mutated in individuals with autosomal dominant inherited dysgenesis of the anterior segment of the eye (Axenfeld-Reiger anomaly). Mouse embryos homozygous for null mutations in Mf1 (Mf1lacZ and Mf1ch) show severely abnormal development of the anterior segment. The cornea fails to separate from the lens, resulting in the complete absence of an anterior chamber. There is no differentiation of the inner corneal endothelial layer, as judged by electron microscopy and by absence of labeling with monoclonal antibody to zonula occludens protein 1, a normal component of occluding junctions in wild-type endothelial cells. In addition, the mutant corneal stroma is disorganized and the epithelium thicker than normal. The Mf1 gene is normally expressed in the periocular mesenchyme at E11.5 but is downregulated as the corneal endothelium differentiates. In contrast, Mf1lacZ expression persists longer in mutant corneal mesenchyme, and abnormal expression is also seen in the mutant corneal epithelium. Based on classical studies with the chick embryonic eye, a model is proposed for the differentiation of the mammalian corneal endothelium from mesenchyme in response to putative signals from the lens. The mesenchyme is thought to be derived from the head neural crest. During early eye development, factors secreted by the lens induce endothelial differentiation in the adjacent presumptive corneal mesenchyme. It is likely that Mf1 is part of the downstream signaling cascade mediating this inductive process. Alternatively, Mf1 may play a role in regulating the fate of the neural crest cells that move in front of the developing eye so that these cells are specified as presumptive corneal mesenchyme rather than as presumptive scleral mesenchyme, for example (Kidson, 1999).

The murine genes, Foxc1 and Foxc2 (previously, Mf1 and Mfh1), encode forkhead/winged helix transcription factors with virtually identical DNA-binding domains and overlapping expression patterns in various embryonic tissues. The forkhead domain of these proteins is 76% identical to that of Drosophila Crocodile. Foxc1/Mf1 is disrupted in the mutant, congenital hydrocephalus (Foxc1/Mf1^ch), which has multiple developmental defects. Depending on the genetic background, most Foxc1 homozygous mutants are born with abnormalities of the metanephric kidney, including duplex kidneys and double ureters, one of which is a hydroureter. Analysis of embryos reveals that Foxc1 homozygotes have ectopic mesonephric tubules and ectopic anterior ureteric buds. Moreover, expression in the intermediate mesoderm of Glial cell-derived neurotrophic factor (Gdnf), a primary inducer of the ureteric bud, is expanded more anteriorly in Foxc1 homozygous mutants compared with wild type. These findings support the hypothesis of Mackie and Stephens concerning the etiology of duplex kidney and hydroureter in human infants with congenital kidney abnormalities (Mackie, G. G. and Stephens, F. G. (1975) J. Urol. 114: 274-280). It is hypothesized that, in some congenital abnormalities, an ectopic ureter bud is induced more anteriorly than normal. Reciprocal interactions between the ectopic bud and the adjacent nephrogenic mesenchyme give rise to an ectopic kidney that fuses with the normal kidney, giving a duplex kidney. Ultimately, the ectopic ureter opens into the urethra and not the trigone, leading to abnormal outflow of urine and development of hydroureter. Previous studies established that most Foxc1^lacZFoxc2^tm1 compound heterozygotes have the same spectrum of cardiovascular defects as single homozygous null mutants, demonstrating interaction between the two genes in the cardiovascular system. Most compound heterozygotes have hypoplastic kidneys and a single hydroureter, while all heterozygotes are normal. This provides evidence that the two genes interact in kidney as well as heart development (Kume, 2000).

Mutations of the human EYA1 gene, a homolog of the Drosophila eyes absent (eya) gene, are associated with the dominant inherited disorder, branchio-oto-renal (BOR) syndrome in which very variable defects in kidney and urinary tract development are seen. Eya1 expression overlaps with that of Gdnf during kidney development and mice homozygous for a null mutation in Eya1 lack the outgrowth of the ureteric bud, the same phenotype seen in Gdnf mutants. In addition, in Eya1 homozygous mutant embryos, Gdnf is not detected in the metanephric mesenchyme, suggesting that Eya1 controls the genetic regulatory cascade upstream of Gdnf. At 10.5 dpc the Eya1 domain also extends more anteriorly in Foxc1/Mf1^ch mutants compared to the wild type . This suggests that Foxc1/Mf1 regulates either Eya1 or more upstream genes in the regulatory cascade in the intermediate mesoderm rather than Gdnf itself (Kume, 2000).

Several models can be considered for the role of Foxc1/Mf1 in the formation of ectopic anterior ureter buds and mesonephric tubules. For example, it is hypothesized that both phenotypes are the result of the persistence of Gdnf transcription in nephrogenic mesenchyme cells that normally only transiently express the gene. During normal development, Gdnf is first expressed in the nephrogenic cord at 8.5 dpc and then in the mesonephric and metanephric mesenchyme while they differentiate alongside the Wolffian duct. According to the model presented, Gdnf expression and/or Gdnf-expressing cells are normally lost from the region between somites 16 and 25 in the mouse; by 10.5 dpc, expression is only seen in the metanephric mesenchyme around the region of the future ureter bud. The more anterior expression of Gdnf seen in Foxc1/Mf1 homozygous mutants suggests that these mutants have defects in the mechanism(s) that normally downregulates Gdnf expression anterior to the region around somite 25. Possible mechanisms for the programmed suppression of Gdnf include the withdrawal of a positive factor normally inducing or maintaining gene expression, or the activation of a negative factor actively repressing Gdnf. If such mechanisms exist, then Foxc1/Mf1 might function upstream or downstream of the factor(s) normally regulating Gdnf expression in posterior mesonephric mesenchyme. The model does not distinguished between direct or indirect regulation of Gdnf. The finding that expression of Eya1 is also seen more anteriorly in Foxc1/Mf1 homozygous mutants than in the wild type, raises the possibility that Foxc1 negatively regulates Eya1 rather than Gdnf. Evidence that Eya1 is upstream of Gdnf comes from the recent finding that Gdnf is not detected in Eya1 mutant embryos (Kume, 2000).

Four insertional alleles were identified of foxi one (foo), an embryonic lethal mutation in zebrafish that displays defects in both otic placode and the jaw. In foo/foo embryos the otic placode is split into two smaller placodes and mutant embryos show a dorsoventral (DV) cartilage defect manifested as a reduced hyomandibular and reduced third and fourth branchial arches. foxi one (foo), the zebrafish ortholog of Foxi1 (FREAC6, FKHL10, HFH-3, Fkh10) and a member of the forkhead domain transcriptional regulator family, has been identified as the gene mutated in foo/foo embryos. foo is expressed in otic placode precursor cells, and foo/foo embryos lack placodal pax8 expression and have disorganized otic expression of pax2.1 and dlx3. Third stream neural crest cell migration, detected by dlx2 and krox20 expression, is aberrant in that it invades the otic placode territory. foo is expressed in pharyngeal pouch endoderm and is required for pouch expression of pax8 and proper patterning of other markers in the pouch, such as nkx2.3. In foo/foo embryos, a failure to maintain fgf3 expression in the pouches is observed, followed by apoptosis of neural crest cells in adjacent arches. It is concluded that foo expression is essential for pax8 expression probably downstream of Fgf signaling in a conserved pathway jointly required for integrity of patterning in the otic placode and pharyngeal pouches. It is proposed that correct placement of survival/proliferation cues is essential for shaping the pharyngeal cartilages and that evolutionary links between jaw and ear formation can be traced to Fgf-Foxi1-Pax8 pathways (Nissen, 2003).

Facial abnormalities in human SHH mutants have implicated the Hedgehog (Hh) pathway in craniofacial development, but early defects in mouse Shh mutants have precluded the experimental analysis of this phenotype. Hh-responsiveness has been removed specifically in neural crest cells (NCCs), the multipotent cell type that gives rise to much of the skeleton and connective tissue of the head. Hh-responsiveness was removed from the entire neural crest lineage by crossing mice harboring Wnt1-Cre with those that are conditionally null for Smoothened (Smo), an obligatory and cell-autonomous component of Hh signal transduction in responding tissue. In these mutants, many of the NCC-derived skeletal and nonskeletal components are missing, but the NCC-derived neuronal cell types are unaffected. Although the initial formation of branchial arches (BAs) is normal, expression of several Fox genes, specific targets of Hh signaling in cranial NCCs, is lost in the mutant. The spatially restricted expression of Fox genes suggests that they may play an important role in BA patterning. Removing Hh signaling in NCCs also leads to increased apoptosis and decreased cell proliferation in the BAs, which results in facial truncation that is evident by embryonic day 11.5 (E11.5). Together, these results demonstrate that Hh signaling in NCCs is essential for normal patterning and growth of the face. Further, this analysis of Shh-Fox gene regulatory interactions leads to the proposal that Fox genes mediate the action of Shh in facial development (Jeong, 2004).

These data indicate that Hh signaling regulates ectomesenchymal expression of five Fox genes, Foxc2, Foxd1, Foxd2, Foxf1, and Foxf2. Although several of these have been reported to be induced by Shh in somites, foregut, or tissue culture, little attention has been given to their expression in facial primordia. Consequently, prior to this work, Foxc2 was the only one of these members that had been shown to be transcribed downstream of Hh signaling in the ectomesenchyme. Based on these findings, it is proposed that the Fox genes are the major mediators of the function of Hh signaling in craniofacial morphogenesis. Further support for this model comes from the mutant phenotype of Foxc2. The head skeleton of Foxc2 mutants exhibits defects that overlap those of Wnt1-Cre;Smo^n/c mutants, suggesting that the loss of Foxc2 expression can account for at least part of the phenotype of Wnt1-Cre;Smo^n/c embryos. In particular, the absence of the palate components (palatal process of the maxilla and palatine) and the middle ear ossicles (incus and stapes) correlates with the expression of Foxc2 in the MXA and second BAs. Foxf2 mutants also have a cleft palate, although this is likely to be secondary to the influence of Foxf2 on tongue morphogenesis. In contrast, no craniofacial abnormalities were reported in the mutants of either Foxd1 or Foxd2. This lack of an overt phenotype could be due to a functional redundancy between these or other Fox family members that obscures their importance. Unfortunately, the early lethality caused by mutation of Foxf1 precludes an assessment of its role in facial development (Jeong, 2004).

Although the transcription of Foxc2, Foxd1, Foxd2, Foxf1, and Foxf2 are clearly all under the positive regulation by Hh signaling in facial primordia, the Fox genes are dissimilar from one another in their normal expression patterns. Furthermore, in the ectomesenchyme of Wnt1-Cre;R26SmoM2 embryos, the distribution and level of each Fox gene transcripts are spatially regulated despite the uniform activation of the Hh pathway; in the MNA, Foxc2 and Foxd1 are expressed ubiquitously except at the midline, whereas Foxf1 is excluded from the lateral ends. Foxd2 and Foxf2 are both expressed along the entire mediolateral axis, but Foxd2 has an increasing, and Foxf2 a decreasing, gradient of intensity from medial to lateral. These observations suggest that Fox genes may be at the regulatory intersection between a Hh pathway input and that of another signaling activity present in a mediolateral gradient in the MNA. For example, if a hypothetical signaling molecule forms an increasing concentration gradient from medial to lateral, then induction of Foxc2, Foxd1, and Foxd2, and repression of Foxf1 and Foxf2 at different thresholds could result in the Fox gene expression patterns described above (Jeong, 2004).

How do Fox genes participate in craniofacial development? First, they may be functionally redundant permissive factors that serve common needs of cells such as survival or proliferation. In this case, a certain amount of Fox protein may be required in order for a cranial neural crest cell to participate in facial morphogenesis, but the exact combination of Fox proteins may not be important. Further, when more than one Fox gene is expressed in the same cell, inactivating one of these may or may not produce abnormalities, depending on its expression level and the sensitivity of the particular cell to the overall Fox gene dosage. Alternatively, certain combinations of Fox genes may have instructive information specifying distinct cell fates. When combined together, the unique expression patterns of each Fox gene make an intriguing map of 'Fox codes' in the developing face. How these domains defined by different Fox codes correlate with facial structures is not clear, because a fate map of facial development is not yet available. However, if one assumes that the relative positions of the facial element precursor domains at E10.5 are the same as those of the facial elements in the newborns, this leads to some interesting predictions. For example, the mesenchyme around the first pharyngeal cleft is expected to make the skeleton associated with the otic capsule, such as stapes, malleus, gonial bone, and tympanic ring. This mesenchyme expresses Foxc2 + Foxd1 + Foxd2. On the other hand, the tongue arises at the midline of the MNAs, where Foxf1, but none of these three Fox genes, is expressed. The domain anterior to the tongue, where the lower incisors form, has still another Fox code, Foxd1 + Foxd2 + Foxf1 + Foxf2. All these facial structures are lost in Wnt1-Cre;Smo^n/c embryos, consistent with all the Fox codes being lost. The absence of craniofacial defects in Foxd1 or Foxd2 mutants could be explained by some degree of tolerance in the Fox codes, which would allow more than one combination to encode the same element. The molars and the body of the dentary apparently develop outside of the Fox gene expression domains. Accordingly, they are present in Wnt1-Cre;Smo^n/c heads, suggesting that they are specified by other mechanisms. Similar correlations can be found for the MXA-derived elements, but not for the FNP-derived ones. The FNP derivatives (nasal bone, nasal cartilage, premaxilla, and upper incisor) suffer relatively mild defects in Wnt1-Cre;Smo^n/c embryos, where none of them are completely lost. Furthermore, no defects in these structures are observed in Foxc2^–/– embryos. Therefore, it is speculated that unlike the first and second BAs, specification of individual skeletal elements in the FNP is independent of Hh signaling or Fox gene expression in the ectomesenchyme, though FNP growth is dependent on Hh signaling. Clearly, distinguishing between the two models for Fox gene function will require additional loss-of-function, gain-of-function, and gene swapping experiments (Jeong, 2004).

Notch signaling, wt1 and foxc2 are key regulators of the podocyte gene regulatory network in Xenopus

Podocytes are highly specialized cells in the vertebrate kidney. They participate in the formation of the size-exclusion barrier of the glomerulus/glomus and recruit mesangial and endothelial cells to form a mature glomerulus. At least six transcription factors (wt1, foxc2, hey1, tcf21, lmx1b and mafb) are known to be involved in podocyte specification, but how they interact to drive the differentiation program is unknown. The Xenopus pronephros was used as a paradigm to address this question. All six podocyte transcription factors were systematically eliminated by antisense morpholino oligomers. Changes in the expression of the podocyte transcription factors and of four selected markers of terminal differentiation (nphs1, kirrel, ptpru and nphs2) were analyzed by in situ hybridization. The data were assembled into a transcriptional regulatory network for podocyte development. Although eliminating the six transcription factors individually interfered with aspects of podocyte development, no single gene regulated the entire differentiation program. Only the combined knockdown of wt1 and foxc2 resulted in a loss of all podocyte marker gene expression. Gain-of-function studies showed that wt1 and foxc2 were sufficient to increase podocyte gene expression within the glomus proper. However, the combination of wt1, foxc2 and Notch signaling was required for ectopic expression in ventral marginal zone explants. Together, this approach demonstrates how complex interactions are required for the correct spatiotemporal execution of the podocyte gene expression program (White, 2010).

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.