• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jun 25, 2002; 99(13): 8719–8724.
Published online Jun 18, 2002. doi:  10.1073/pnas.132195699
PMCID: PMC124365
Developmental Biology

Mutually regulated expression of Pax6 and Six3 and its implications for the Pax6 haploinsufficient lens phenotype

Abstract

Pax6 is a key regulator of eye development in vertebrates and invertebrates, and heterozygous loss-of-function mutations of the mouse Pax6 gene result in the Small eye phenotype, in which a small lens is a constant feature. To provide an understanding of the mechanisms underlying this haploinsufficient phenotype, we evaluated in Pax6 heterozygous mice the effects of reduced Pax6 gene dosage on the activity of other transcription factors regulating eye formation. We found that Six3 expression was specifically reduced in lenses of Pax6 heterozygous mouse embryos. Interactions between orthologous genes from the Pax and Six families have been identified in Drosophila and vertebrate species, and we examined the control of Pax6 and Six3 gene expression in the developing mouse lens. Using in vitro and transgenic approaches, we found that either transcription factor binds regulatory sequences from the counterpart gene and that both genes mutually activate their expression. These studies define a functional relationship in the lens in which Six3 expression is dosage-dependent on Pax6 and where, conversely, Six3 activates Pax6. Accordingly, we show a rescue of the Pax6 haploinsufficient lens phenotype after lens-specific expression of Six3 in transgenic mice. This phenotypic rescue was accompanied by cell proliferation and activation of the platelet-derived growth factor α-R/cyclin D1 signaling pathway. Our findings thus provide a mechanism implicating gene regulatory interactions between Pax6 and Six3 in the tissue-specific defects found in Pax6 heterozygous mice.

Keywords: homeodomain proteins‖transcription factors‖eye development

The Pax gene family encodes a group of transcription factors containing a paired DNA-binding domain and, in some cases, a paired-type homeodomain (1). In vertebrates, Pax genes include nine members that hold essential developmental roles, as exemplified by the phenotypes found in homozygous null mutants (2). An interesting feature of Pax genes, and one that is shared with a restricted group of gene families (3), is that dominant phenotypes frequently accompany heterozygous loss-of-function mutations. Typically, the phenotype only partly reflects the expression pattern, and the affected organ is reduced in size in the heterozygote and is absent in the homozygote. A well-studied example of this dosage-dependent effect is found with the Pax6 gene (4, 5). Mice that are homozygous for Pax6 null mutations are anophthalmic, whereas a 50% reduction of Pax6 gene dosage causes a microphthalmic phenotype (4). Overexpression of Pax6 in eye structures of transgenic mice also causes a microphthalmic phenotype (6), indicating that stringent controls must be set on Pax6 expression levels for normal eye development to occur.

An understanding of the mechanisms by which single allele mutations of certain genes, but not of others, lead to an haploinsufficient syndrome remains an unresolved issue (3). It has been proposed that Pax6 exerts a dosage effect (7) by which certain effector genes may not respond predictably to Pax6 protein levels that are below a threshold value. Eye development in vertebrates and invertebrates is regulated by a conserved network of transcription factors and nuclear proteins (8), and we tested the hypothesis that reduced Pax6 gene dosage may lead to abnormal gene expression within this network. We were particularly interested in studying the effects of reduced Pax6 gene dosage on the activity of the Six3 gene, because both genes are expressed in an overlapping fashion during mouse eye development (9), and regulatory networks involving genes orthologous to Pax6 and Six3 have been demonstrated in eyes of vertebrates and Drosophila (10, 11). Moreover, disruption of Pax6 function in the prospective lens ectoderm of the mouse abolishes Six3 expression, suggesting a dependence of Six3 on Pax6 function in this structure (12).

Here, we describe the regulatory interactions between Pax6 and Six3 in the developing mouse lens. Our results reveal a gene relationship in which Six3 expression is dosage-dependent on Pax6 function and where, conversely, Six3 activates Pax6 expression. We found this gene relationship to be tissue-specific. Furthermore, we show that the Pax6 haploinsufficient lens phenotype is rescued in transgenic mice expressing Six3 under the regulation of a lens-specific gene promoter. Our results thus provide a mechanism explaining how heterozygous loss-of-function mutations of the mouse Pax6 gene may lead to the tissue-specific defects observed in the Small eye mutant. Moreover, these findings suggest that similar regulatory interactions may underlie haploinsufficient syndromes resulting from single-allele mutations of other transcription factor-encoding genes.

Materials and Methods

Transgene Constructs and Generation of Transgenic Mice.

The αASix3 transgene was generated by inserting a 1.4-kb Six3 cDNA fragment into polylinker restriction sites located between the αA-crystallin promoter and simian virus 40 (SV40) intron and polyadenylation sequences of plasmid pACP3 (13) (see Fig. Fig.22A). The transgene was linearized with SacII and readied for microinjection. A similar approach was used to construct a modified αA-crystallin promoter/Pax6 cDNA transgene (αAmPax6), except that two additional Pax6 binding sites were inserted in the promoter region (L.W.R., unpublished work). Transgenic mice were generated (14) and bred in the FVB/N background. Two αASix3 transgenic lines (nos. 40 and 53) were studied. Both displayed the same phenotype based on morphology and molecular analysis. Eight αAmPax6 lines were generated; analysis was done on line OVE1078. NMRI mice harboring a targeted mutation of the Pax6 gene (Pax6lacZ) (15) were backcrossed for at least four generations in the FVB/N background before mating with αASix3 transgenic mice. For staging of embryos, noon of the day of vaginal plug formation was considered as embryonic day 0.5 (E0.5). Genotyping was done by PCR (13) or Southern blotting (15) with genomic DNA extracted from mouse tails.

Figure 2
Structure and expression of the αASix3 transgene. (A) Diagram of the DNA construct used for the generation of αASix3 transgenic mice. The αASix3 transgene was generated by inserting the full-length coding sequences of the murine ...

Histology and in Situ Hybridization.

Embryos were harvested, fixed in 4% paraformaldehyde, and processed in Paraplast. Transverse sections of the eyes were done at 8-μ thickness as described (16). Eyes from 3-week-old mice were dissected, fixed in Carnoy's solution, processed in JB-4 plastic (Polysciences), and sectioned, according to a previously published protocol (17). In situ hybridization experiments using 35S-labeled riboprobes were performed as described (16). A transgene-specific probe (probe B, see Fig. Fig.22A) consisted of a 220-bp BglII–BamHI fragment, containing SV40 polyadenylation sequences. The Six3 (9), FoxE3 (18), Prox1 (19), and Eya1 and 2 (provided by R. Maas, Harvard University, Cambridge, MA; ref. 20) riboprobes have been described.

Image Analysis.

Bright- and dark-field images were captured with an Olympus BX60 microscope fitted with a charge-coupled device, and analyzed with the imaging software analysis (Soft-Image System, Münster, Germany). For gray-scale measurements, dark-field images were captured at identical light intensities and exposure times and saved as 8-bit black and white images. Selected ocular regions were outlined, the remaining image was masked, and gray-scale values were measured on individual pixels. Results were expressed as percentages, with values of 0 and 100 being given to black and white, respectively.

Immunohistochemistry.

The following antibodies were used: anti-β- and anti-γ-crystallins, anti-αA-crystallin, anti-MIP26, anti-filensin, and anti-CP49 [kindly provided, respectively, by J. S. Zigler (National Eye Institute, Bethesda), K. Kato (Institute for Developmental Research, Aichi, Japan), J. Horwitz (Jules Stein Eye Institute, Los Angeles), and R. A. Quinlan (University of Dundee, Dundee, U.K.)]. Anti-Cdc2, Cdk4, Cdkn1b, Cdkn1c, platelet-derived growth factor (PDGF)α-R, fibroblast growth factor-R types 1 and 2 antibodies (Santa Cruz Biotechnology), anti-insulin-like growth factor I-R (GroPep, Adelaide, Australia), anti-cyclin D1 (Zymed), anti-Pax6 (American Type Culture Collection), anti-Six3 (kindly provided by G. Oliver, St. Jude Children's Research Hospital, Memphis, TN; ref. 21), anti-Sox1 (obtained from R. Lovell-Badge, National Institute for Medical Research, London; ref. 22), and anti-phospho p44/42 mitogen-activated protein kinase (Cell Signaling Technology, Beverly, MA). In all cases, appropriate Alexa-488-conjugated secondary antibodies (anti-mouse, -rabbit, or -goat) were used and cell nuclei were counterstained with propidium iodide (all from Molecular Probes). Fluorescence was recorded with a SLM410 Zeiss confocal microscope.

BrdUrd Labeling and Terminal Deoxynucleotidyltransferase-Mediated dUTP Nick-End Labeling (TUNEL).

For BrdUrd labeling, gravid females were injected i.p. with a BrdUrd/fluorodeoxyuridine mix (10 μm/1 μm, Amersham Pharmacia) at 0.01 ml/g of body weight, and embryos were harvested 3 h afterward. After processing in Paraplast, sections were reacted with a mouse monoclonal anti-BrdUrd antibody (Roche Molecular Biochemicals) using conditions suggested by the manufacturer. TUNEL staining was done by using ApopTag Plus Fluorescein Apoptosis Detection reagents (Intergen, Purchase, NY) as specified.

Electrophoretic Mobility-Shift Assays (EMSAs) and Cell Transfections.

EMSAs were done according to a previously published protocol (23). Expression vectors used for cell transfections were cloned in the cytomegalovirus promoter-driven pCS2+ vector. The Pax6-luciferase reporter contained six tandem copies of a 20-bp sequence (5′-CTCATTGCCCATTCAAATAC-3′) derived from a Pax6 lens enhancer (24). The Six3-luciferase reporter contains four tandem copies of a 45-bp sequence containing an inverted Pax6-binding site (5′-CCGCCCAGAGCCTTGCAGTTGAGCGGAAAAGCACTCCTGGTGCAG-3′), located approximately 2 kbp upstream of exon 1 of mouse Six3. Both reporter constructs were cloned in pGL3-promoter (Promega). Cultures and transfections of COS7 cells were prepared as described (23). Transfection efficiency was estimated by including a constant amount (0.3 μg) of cytomegalovirus promoter-LacZ plasmid while the total amount of transfected DNA was kept constant (2 μg) in all assays. Forty hours after transfection, cells were harvested, and luciferase assay was carried out by using a Berthold (Nashua, NH) LB9501 luminometer. β-Galactosidase activity was measured with the Galacto-Light kit (Tropix, Bedford, MA). For all experiments three samples were transfected and processed independently. Each experiment was performed at least twice.

Results

Six3 Expression in the Lens Requires Pax6 Function in a Dosage-Dependent Manner.

We examined the regulatory hierarchy of genes involved in lens development by comparing the expression of eye-specific transcription or nuclear factors in the eyes of E12.5 wild-type and Pax6 heterozygous embryos. Reduced Six3 expression was found in lenses of Pax6lacZ embryos (Fig. (Fig.11 A′, B′, and G), whereas expression levels of Eya1 and 2, FoxE3, Prox1, and Sox1 (Fig. (Fig.11 CF′ and data not shown) were the same in wild-type and Pax6lacZ embryos. In contrast, Six3 expression in the retina of Pax6lacZ embryos was unchanged (Fig. (Fig.11 A′, B′, and G). The lens-specific reduction of Six3 expression in Pax6lacZ embryos indicates a dosage dependence of Six3 expression on Pax6 levels in the lens, whereas Pax6 function is not essential for Six3 expression in the retina (23).

Figure 1
Dosage-dependent reduction of Six3 expression in lenses of Pax6lacZ embryos. Reduced Six3 expression is found in lenses of E12.5 Pax6lacZ embryos (B′), compared with controls (A′). Six3 expression in the retina (ret) is unchanged. Expression ...

Lens-Specific Expression of Six3 Rescues the Pax6 Haploinsufficient Lens Phenotype.

Because of its specificity, we speculated that this dosage-dependent reduction of Six3 expression might contribute to the Pax6 haploinsufficient lens phenotype. We therefore attempted to rescue this lens defect by increasing Six3 gene dosage, using a transgene (αASix3, Fig. Fig.22A) linking a 0.4-kbp murine αA-crystallin gene promoter to a 1.4-kbp mouse Six3 cDNA (9). With a transgene-specific riboprobe, embryonic expression was initially detected at E12.0 (Fig. (Fig.22B). Expression of the transgene in lens fiber cells overlapped that of the endogenous Six3 gene from E12.0 to E13.5 and was maintained in the differentiated fiber cells (Fig. (Fig.22 BE). We found that transgene expression levels were lower in αASix3/Pax6lacZ embryos than in αASix3/Pax6+/+ embryos (not shown), this difference being likely caused by the presence of a Pax6-responsive site in the αA-crystallin promoter sequences used in the transgene construct (25).

Lenses from E14.5 αASix3/Pax6lacZ embryos were larger than those of age-matched Pax6lacZ embryos, but remained smaller than wild-type controls (Fig. (Fig.33 AC). This phenotypic rescue was found in Pax6lacZ mice hemizygous for the αASix3 transgene and was fully penetrant in the FVB/N background. The increased lens size was maintained in 3 week-old αASix3/Pax6lacZ mice (Fig. (Fig.33 IK), although at this later stage lens morphology was abnormal and included the presence of vacuoles and ballooned fiber cells (Fig. (Fig.33 K and O). To explain this phenotypic rescue, we first examined the effects of Six3 expression on lens fiber cell proliferation, differentiation, and death. We studied cell proliferation in embryos by using the BrdUrd incorporation assay, which detects cells entering the S phase. BrdUrd-incorporating lens fiber cells were present in E12.5 (Fig. (Fig.33S) and E13.5 αASix3/Pax6lacZ embryos, but not in the lens fiber cell compartment of controls (Fig. (Fig.33 Q and R). We next examined the effects of Six3 expression on lens differentiation and cell death. At stage E14.5, the expression of the fiber cell differentiation markers αA-, β- and γ-crystallins, MIP-26, filensin, and CP-49 (Fig. (Fig.33 UW and data not shown), as well as the number of TUNEL assay-positive fiber cells (Fig. (Fig.33 YZ′), were the same in fiber cells of αASix3/Pax6lacZ and control embryos. Altogether, these results indicate that the improved lens size in αASix3/Pax6lacZ embryos resulted, in part, from fiber cell proliferation. Importantly, differentiation and survival of fiber cells in αASix3/Pax6lacZ embryos were normal, leading to overall lens growth.

Figure 3
Rescue of the Pax6 haploinsufficient lens phenotype in αASix3/Pax6lacZ transgenic mice. (AP) Effects of Six3 expression on lens morphology. Transverse sections from eyes of E14.5 (A--H) and 3-week-old (IP) mice. Note the increased ...

Six3 Overexpression Causes Defective Lens Maturation and Fiber Cell Death.

In contrast to the improved lens size in αASix3/Pax6lacZ transgenic mice, αASix3/Pax6+/+ transgenic mice were microphakic (Fig. (Fig.33 D and L). The severity of the microphakia was greater in embryos homozygous for the αASix3 transgene (not shown). The latter effect was found in either the Pax6+/+ or Pax6lacZ background, indicating that Six3 gene dosage was the primary factor affecting the lens phenotype. αASix3/Pax6lacZ transgenic embryos were microphakic from stage E14.5 onward (Fig. (Fig.33D). In sections done on adult eyes, the lenses had a disorganized fiber cell arrangement accompanied by posterior rupture of their content into the vitreous chamber (Fig. (Fig.33 L and P).

BrdUrd incorporation assay revealed lens fiber cell proliferation in αASix3/Pax6+/+ transgenic embryos, an effect similar to that observed in αASix3/Pax6lacZ embryos (Fig. (Fig.33 S and T). We examined the effect of increased Six3 expression on lens fiber cell maturation and survival. From E14.5 onward, a heterogeneous distribution of lens differentiation markers was observed in lenses of αASix3/Pax6+/+ embryos, when compared with controls (Fig. (Fig.33 U and X). Lastly, from stage E14.5 onward, vacuoles were present in fiber cells of αASix3/Pax6+/+ embryos, but not in age-matched controls (Fig. (Fig.33 E and H), indicating a process of cell death. However, fiber cells were rarely stained with TUNEL (Fig. (Fig.33Z′′). Cell death thus resulted either from necrosis or from a process differing from the common forms of apoptosis. These results thus suggest that, in addition to defective maturation, Six3 overexpression causes fiber cell death.

Mutually Regulated Expression of Pax6 and Six3 in the Lens.

Members of the vertebrate Pax and Six gene families participate in gene feedback loops regulating eye, kidney, ear, and muscle formation (reviewed in ref. 26). To further address the effects of Six3 expression in the developing lens, we examined the regulatory relationship between Pax6 and Six3. In lenses of transgenic embryos, expression of Six3 up-regulated that of Pax6 (Fig. (Fig.44 AF). Reciprocally, up-regulated Six3 expression was found in fiber cells of αA-crystallin (modified)/Pax6AmPax6) transgenic embryos (Fig. (Fig.44 GL). Expression of other eye-specific factors (Eya1, Eya2, FoxE3, Prox1, and Sox1) was unchanged in lenses of transgenic embryos and controls (not shown).

Figure 4
Mutually regulated expression of Pax6 and Six3 in the developing lens. (AL) Mutual activation of Six3 and Pax6 expression in transgenic embryos. Fiber cell nuclei from E14.5 αASix3 transgenic embryos are immunoreactive for the Pax6 protein ...

A 107-bp BglII–AccI DNA fragment, located 5 kbp upstream of the Pax6 (P0) transcription start site, has previously been shown to direct reporter gene activity in lenses of transgenic embryos (24). This fragment was examined by EMSA for the presence of Six3-binding sites. We observed the shift of electromobility of a 20-bp sequence (5′-CTCATTGCCCATTCAAATAC-3′) derived from the 5′ part of the enhancer (not shown). We examined whether this sequence has Six3-dependent transcriptional activity. In cell transfection assays, luciferase activity was increased after cotransfecting a reporter construct containing a multimer of this sequence with a Six3 expression vector (Fig. (Fig.44M). Additionally, overlapping phage clones containing 27 kbp of the Six3 gene were sequenced and evaluated for the presence of sequences related to the Pax6-paired domain DNA-binding sites (27). An oligonucleotide derived from one potential inverted site (5′-TGCAGTTGAGCGGAAAAG-3′), located 2 kbp upstream of the first coding exon, shifted the Pax6 protein in EMSA (not shown). Moreover, a luciferase reporter construct containing a multimerized oligonucleotide was activated after cotransfection with a Pax6 expression vector (Fig. (Fig.44N). The results of EMSAs and cell transfection assays, combined with those in transgenic embryos, clearly indicate that each gene constitutes a direct target for the counterpart transcription factor, and that Pax6 and Six3 expression is mutually activated during mouse lens development.

Up-Regulation of PDGFα-R and Cyclin D1 After Lens-Specific Expression of Six3 or Pax6 in Transgenic Mice.

We further evaluated the cell proliferation in αASix3/Pax6lacZ and αASix3/Pax6+/+ transgenic embryos by studying the expression of cell cycle regulatory proteins in fiber cells. We found increased levels of cyclin D1 (Fig. (Fig.55 AD), whereas expression levels of Cdkn1c, Cdkn1b, Cdc2, and Cdk4 were unchanged (not shown). The up-regulated cyclin D1 activity was accompanied by an increase in the activated form of p44/42 mitogen-activated protein kinase (ERK), indicating a fiber cell response to an extracellular mitogenic signal (not shown). We further evaluated this signaling pathway by studying the lens expression levels of receptor tyrosine kinases that act upstream of ERK. Expression levels of PDGFα-R were up-regulated in fiber cells of αASix3 transgenic embryos, whereas expression levels of insulin-like growth factor I-R, and fibroblast growth factor types 1 and 2 were unchanged (Fig. (Fig.55 IL and data not shown). Moreover, fiber cell proliferation, ERK activation, and up-regulated expression of PDGFα-R and cyclin D1 were found in lenses of αAmPax6 transgenic embryos (Fig. (Fig.55 EH and MN, and data not shown). Defective fiber cell maturation was also demonstrated in αAmPax6 transgenic embryos (Fig. (Fig.55 O and P), an effect similar to that observed after Six3 overexpression (see Fig. Fig.33 U and X). Therefore, in gain-of-function studies, the effects of either transcription factor in the lens were similar, as a likely consequence of the cross-regulation between both genes. These results also argue that both transcription factors participate, along with FoxE3, Prox1, and Cdkn1c (18, 19, 28), in a gene network controlling cell proliferation in the lens. In this context, the mutual activation of Pax6 and Six3 may serve to limit the fluctuations of expression levels of either transcription factor (3). Stable expression of Pax6 and Six3 may be required for the coordinate expression of various components of the cyclin D1/PDGFα-R signaling pathway, thus facilitating the progression between different phases of the cell cycle.

Figure 5
PDGFα-R and cyclin D1 up-regulation in transgenic embryos expressing Six3 or Pax6 in the lens. (AH) Cyclin D1 immunochemistry. Increased numbers of cyclin D1-reactive nuclei are detected in the fiber cell compartment of E12.5 α ...

Discussion

In this study, we described the mechanisms regulating the expression of Pax6 and Six3 in the developing mouse lens. Using both in vitro and transgenic approaches, each transcription factor was shown to directly interact with the counterpart gene, and mutual activation of gene expression was demonstrated. Importantly, Six3 expression was specifically reduced in lenses of Pax6 heterozygous embryos. Hence, our findings invoke a model of gene regulation in the lens in which Six3 expression is dosage-dependent on Pax6, and where Six3 activates Pax6 transcription. This model indicates a pivotal role for the Six3 transcription factor in this regulatory pathway. Moreover, it provides an explanation for the tissue-specific defects observed in haploinsufficient syndromes. Accordingly, we have tested this regulatory model and shown, by means of a transgenic experiment, that Six3 expression in lenses of Pax6 heterozygous embryos restores this structure to a near-normal size.

Previous studies have highlighted some of the mechanisms underlying haploinsufficient syndromes (3, 7, 29). The Pax6 haploinsufficient lens phenotype has been related to a delay in lens formation, possibly because of an initial failure of the Pax6 protein to reach a threshold level (7). Although the mechanisms controlling the expression of Pax6 and Six3 were not directly evaluated at early stages of lens formation, prior findings show that both genes interact at these developmental stages, and therefore likely contribute to the early lens defect observed in Pax6 heterozygous embryos. Indeed, experiments involving the conditional disruption of the Pax6 gene in the prospective lens ectoderm (12) have shown that Pax6 function is required for the initiation of Six3 expression in this structure. However, it remains unclear whether Six3 regulates Pax6 function at early stages of lens formation, because the consequences of Six3 disruption in the mouse have not been reported. Indirect evidence for such an interaction however comes from the targeted deletion of a 340-bp lens-specific Pax6 enhancer (30), which includes the 107-bp fragment analyzed in our study. This study confirms that this 340-bp enhancer controls early (E9.0–12.0) lens-specific Pax6 expression, implying a role for the Six3-binding site it contains in the regulation of early Pax6 expression. This study also indicates the existence of lens-specific enhancer(s) controlling Pax6 expression at later (>E12.0) developmental stages. The increased Pax6 expression resulting from the activation of the additional lens-specific enhancer(s) leads to a rescue of the early microphakic phenotype observed in homozygous and heterozygous mutant embryos. These results are therefore analogous to the phenotypic rescue observed in our study and provide independent proof that genetic manipulations at later stages of lens formation can correct early developmental defects.

Prior studies have indicated distinct roles for Pax6 during eye formation (31, 32). The function of Six3, however, remains unclear. Previous studies have indicated roles for the closely related XSix6/Optx2 gene (33) and the Drosophila ortholog sine oculis (34) in proliferative control. We have shown that, in lenses of transgenic mouse embryos, Six3 expression induces cell proliferation. Similarly, lens fiber cell proliferation was observed in αAmPax6 transgenic embryos, this common effect likely reflecting the cross-regulation between both genes. Our results show that a signaling pathway involving PDGFα-R, ERK-1, and cyclin D1 mediates the effects of Six3 or Pax6 on cell proliferation. Expression levels of these factors were increased in transgenic mice, whereas levels of other receptor tyrosine kinases (insulin-like growth factor I-R, fibroblast growth factor-R types 1 and 2) or factors known to regulate lens fiber cell proliferation (Cdkn1c, Prox1, and FoxE3) (18, 19, 28) were unchanged in transgenic embryos and controls. Although our results would predict critical functions for both PDGFα-R and D-type cyclins in lens development, disruption of PDGFα-R (35) or cyclin D1 (36) in the mouse have not resulted in any lens defects. However, inactivation of individual PDGF receptors or cyclins in mice appears to be functionally compensated (36, 37), and further analysis would be warranted to determine their exact function during lens development.

Our results indicate that the mutual activation of Pax6 and Six3 expression results from direct interactions between each transcription factor and the counterpart gene. These findings agree with studies suggesting a general role for Pax6 as an activator of transcription (38), but appear in conflict with recent studies indicating that Six3 acts as a transcriptional repressor (39, 40). Six3-mediated repression seems dependent on its interaction with vertebrate members of the Groucho-related family of corepressors (39). Transcription factors that interact with Groucho-related proteins are subdivided into two main subclasses (41). A first class consists of constitutive repressors that bind Groucho proteins in a stable fashion. In contrast, conditional repressors are converted from activators to repressors of transcription depending on the presence or absence of a coactivator protein, or whether the context of the DNA-binding site facilitates the recruitment of Groucho-related proteins. Our results suggest that Six3 functions as a conditional repressor. Similar to the Drosophila Dorsal (42) and the mammalian Pax5 (43) proteins, the activity of Six3 appears context-dependent, because it can function as a transcriptional activator after binding to native DNA sequences. These results indicate an added complexity in the regulation of target genes by Six3. Moreover, members from all three subgroups of the Six family have been shown to interact with Groucho-related proteins (39). Our findings thus suggest that some, if not all, of these factors may exert dual effects on transcription.

Altogether, our results indicate a conservation of the regulatory mechanisms regulating the expression of Pax6 and Six3 orthologs in the eyes of mouse and Drosophila (10, 34). During early mouse lens development, Six3 acts as a downstream target of Pax6 function (12). Subsequently, we show that both genes directly interact to positively regulate their gene expression levels. A similar regulatory strategy has been described in the developing Drosophila eye between Pax6/eyeless and sine oculis, and the Eya coregulatory protein (10, 34). On the basis of nucleic acid comparisons, Six3 and the related Six6/Optx2 share close homologies to the Drosophila Optix gene (44), whose activity is independent of that of Pax6/eyeless (45), and are more distantly related to sine oculis. A possible interpretation is that Sine oculis and Six3 have retained from a common ancestral gene their regulatory relationship to Pax6/eyeless, whereas this same strategy has been lost for Optix and Six6/Optx2. This common gene remains to be identified, although the search for such an ancestral gene in a basal organism (46) should provide novel perspectives on the evolution of both Pax and Six genes.

Acknowledgments

We thank A. Cvekl for the pACP3 plasmid, L. St-Onge for Pax6lacZ mice, M. K. Duncan and E. Saló for discussing results before publication, and P. Collombat and D. Treichel for sequence analysis and graphics assistance, respectively. The transgenic line OVE1078 was generated while L.W.R. was a postdoctoral fellow in the laboratory of P. Overbeek (Baylor College of Medicine, Houston). This work was supported by Research to Prevent Blindness, National Institutes of Health Grant EY13146 (to L.W.R.), and the Max Planck Society.

Abbreviations

EMSA
electrophoretic mobility-shift assay
TUNEL
terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling
PDGF
platelet-derived growth factor
En
embryonic day n
SV40
simian virus 40

Footnotes

This paper was submitted directly (Track II) to the PNAS office.

References

1. Noll M. Curr Opin Genet Dev. 1993;3:595–605. [PubMed]
2. Chi N, Epstein J A. Trends Genet. 2002;18:41–47. [PubMed]
3. Cook D L, Gerber L N, Tapscott S J. Proc Natl Acad Sci USA. 1998;95:15641–15646. [PMC free article] [PubMed]
4. Hill R E, Favor J, Hogan B L, Ton C C, Saunders G F, Hanson I M, Prosser J, Jordan T, Hastie N D, van Heyningen V. Nature (London) 1991;354:522–525. [PubMed]
5. Walther C, Gruss P. Development (Cambridge, UK) 1991;113:1435–1449. [PubMed]
6. Schedl A, Ross A, Lee M, Engelkamp D, Rashbass P, van Heyningen V, Hastie N D. Cell. 1996;86:71–82. [PubMed]
7. van Raamsdonk C D, Tilghman S M. Development (Cambridge, UK) 2000;127:5439–5448. [PubMed]
8. Treisman J E. BioEssays. 1999;21:843–850. [PubMed]
9. Oliver G, Mailhos A, Wehr R, Copeland N G, Jenkins N A, Gruss P. Development (Cambridge, UK) 1995;121:4045–4055. [PubMed]
10. Niimi T, Seimiya M, Kloter U, Flister S, Gehring W J. Development (Cambridge, UK) 1999;126:2253–2260. [PubMed]
11. Loosli F, Winkler S, Wittbrodt J. Genes Dev. 1999;13:649–654. [PMC free article] [PubMed]
12. Ashery-Padan R, Marquardt T, Zhou X, Gruss P. Genes Dev. 2000;14:2701–2711. [PMC free article] [PubMed]
13. Duncan M K, Cvekl A, Li X, Piatigorsky J. Invest Ophthalmol Visual Sci. 2000;41:464–473. [PubMed]
14. Hogan B, Beddington R, Costantini F. Manipulating the Mouse Embryo: A Laboratory Manual. Plainview, NY: Cold Spring Harbor Lab. Press; 1994.
15. St-Onge L, Sosa-Pineda B, Chowdhury K, Mansouri A, Gruss P. Nature (London) 1997;387:406–409. [PubMed]
16. Stoykova A, Gruss P. J Neurosci. 1994;14:1395–1412. [PubMed]
17. Klopp N, Löster J, Graw J. Invest Ophthalmol Visual Sci. 2001;42:183–187. [PubMed]
18. Blixt A, Mahlapuu M, Aitola M, Pelto-Huikko M, Enerbäck S, Carlsson P. Genes Dev. 2000;14:245–254. [PMC free article] [PubMed]
19. Wigle J T, Chowdhury K, Gruss P, Oliver G. Nat Genet. 1999;21:318–322. [PubMed]
20. Xu P X, Woo I, Her H, Beier D R, Maas R L. Development (Cambridge, UK) 1997;124:219–231. [PubMed]
21. Lagutin O, Zhu C Q C, Furuta Y, Rowitch D H, McMahon A P, Oliver G. Dev Dyn. 2001;221:342–349. [PubMed]
22. Pevny L H, Sockanathan S, Placzek M, Lovell-Badge R. Development (Cambridge, UK) 1998;125:1967–1978. [PubMed]
23. Marquardt T, Ashery-Padan R, Andrejewski N, Scardigli R, Guillemot F, Gruss P. Cell. 2001;105:43–55. [PubMed]
24. Kammandel B, Chowdhury K, Stoykova A, Aparicio S, Brenner S, Gruss P. Dev Biol. 1999;205:79–97. [PubMed]
25. Cvekl A, Kashanchi F, Sax C M, Brady J N, Piatigorsky J. Mol Cell Biol. 1995;15:653–660. [PMC free article] [PubMed]
26. Relaix F, Buckingham M. Genes Dev. 1999;13:3171–3178. [PubMed]
27. Epstein J, Cai J, Glaser T, Jepeal L, Maas R. J Biol Chem. 1994;269:8355–8361. [PubMed]
28. Zhang P, Wong C, DePinho R A, Harper J W, Elledge S J. Genes Dev. 1998;12:3162–3167. [PMC free article] [PubMed]
29. Collinson J M, Quinn J C, Buchanan M A, Kaufman M H, Wedden S E, West J D, Hill R E. Proc Natl Acad Sci USA. 2001;98:9688–9693. [PMC free article] [PubMed]
30. Dimanlig P V, Faber S C, Auerbach W, Makarenkova H P, Lang R A. Development (Cambridge, UK) 2001;128:4415–4424. [PubMed]
31. Pichaud F, Treisman J, Desplan C. Cell. 2001;105:9–12. [PubMed]
32. Gehring W J, Ikeo K. Trends Genet. 1999;15:371–377. [PubMed]
33. Zuber M E, Perron M, Philpott A, Bang A, Harris W A. Cell. 1999;98:341–352. [PubMed]
34. Pignoni F, Hu B R, Zavitz K H, Xiao J A, Garrity P A, Zipursky S L. Cell. 1997;91:881–891. [PubMed]
35. Soriano P. Development (Cambridge, UK) 1997;124:2691–2700. [PubMed]
36. Sicinski P, Donaher J L, Parker S B, Li T S, Gardner H, Haslam S Z, Bronson R T, Elledge S J, Weinberg R A. Cell. 1995;82:621–630. [PubMed]
37. Soriano P. Genes Dev. 1994;8:1888–1896. [PubMed]
38. Matsunaga E, Araki I, Nakamura H. Development (Cambridge, UK) 2000;127:2357–2365. [PubMed]
39. Kobayashi M, Nishikawa K, Suzuki T, Yamamoto M. Dev Biol. 2001;232:315–326. [PubMed]
40. Lengler J, Krausz E, Tomarev S, Prescott A, Quinlan R A, Graw J. Nucleic Acids Res. 2001;29:515–526. [PMC free article] [PubMed]
41. Chen G Q, Courey A J. Gene. 2000;249:1–16. [PubMed]
42. Pan D, Courey A J. EMBO J. 1992;11:1837–1842. [PMC free article] [PubMed]
43. Eberhard D, Jimenez G, Heavey B, Busslinger M. EMBO J. 2000;19:2292–2303. [PMC free article] [PubMed]
44. Toy J, Yang J M, Leppert G S, Sundin O H. Proc Natl Acad Sci USA. 1998;95:10643–10648. [PMC free article] [PubMed]
45. Seimiya M, Gehring W J. Development (Cambridge, UK) 2000;127:1879–1886. [PubMed]
46. Pineda D, Gonzalez J, Callaerts P, Ikeo K, Gehring W J, Saló E. Proc Natl Acad Sci USA. 2000;97:4525–4529. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • Gene
    Gene
    Gene links
  • Gene (nucleotide)
    Gene (nucleotide)
    Records in Gene identified from shared sequence links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • HomoloGene
    HomoloGene
    HomoloGene links
  • MedGen
    MedGen
    Related information in MedGen
  • Nucleotide
    Nucleotide
    Published Nucleotide sequences
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Protein
    Published protein sequences
  • PubMed
    PubMed
    PubMed citations for these articles
  • Taxonomy
    Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...