A genetic analysis of VSX2 function in the mouse eye

Vsx2 is a homeodomain-containing transcription factor essential for maintenance of neuroretinal identity and neurogenesis. Vsx2 is believed to function via mechanisms that are strictly cell-intrinsic. However, recent research reveals evidence for involvement of Vsx2-mediated cell non-autonomous mechanisms. This leads to the hypothesis that Vsx2 regulates cell non-autonomous pathways that play a role in maintaining neuroretinal identity and neurogenesis. To test this hypothesis, genetic chimeras were generated by morula aggregations of transgenic ActB:EYFP transgenic mice homozygous for wild type alleles of Vsx2 [Tg(EYFP/EYFP);Vsx2] and mice that are homozygous for null alleles of Vsx2 (Vsx2orj/orj). Maintenance of neuroretinal identity is assessed by immunostaining for the expression of MITF, a retinal pigmented epithelium (RPE) gene that regulates many pigmentation pathway genes. Neurogenesis is assessed through analysis of cell differentiation using immunostaining for several cell-type specific markers. Analysis of the retinas of tg(EYFP/EYFP);Vsx2 ↔ Vsx2orj/orj chimeras reveals the presence of MITF expression in Vsx2orJ/orJ cells but not Vsx2+/+ cells indicating that neighboring Vsx2+/+ cells are unable to rescue Vsx2orJ/orJ cells from loss of neuroretinal identity. Analysis also reveals that Vsx2orJ/orJ cell differentiation remains delayed indicating that neighboring Vsx2+/+ cells are unable to rescue Vsx2orJ/orJ cells from delayed neurogenesis. These findings suggest that Vsx2 regulates neuroretinal identity and neurogenesis by cell-autonomous mechanisms.

May 2012 A GENETIC ANALYSIS OF VSX2 FUNCTION IN THE MOUSE EYE by Massiell German A Senior Honors Thesis Submitted to the Faculty of The University of Utah In Partial Fulfillment for the Requirements for the Honors Degree in Bachelor of Science In Biology Approved: ____________________ ____________________ Edward Levine Neil J. Vickers Supervisor Chair, Dept. of Biology ____________________ ____________________ Rosemary Gray Dr. Sylvia D. Torti Department Honors Advisor Dean, Honors College ii ABSTRACT Vsx2 is a homeodomain-containing transcription factor essential for maintenance of neuroretinal identity and neurogenesis. Vsx2 is believed to function via mechanisms that are strictly cell-intrinsic. However, recent research reveals evidence for involvement of Vsx2-mediated cell non-autonomous mechanisms. This leads to the hypothesis that Vsx2 regulates cell non-autonomous pathways that play a role in maintaining neuroretinal identity and neurogenesis. To test this hypothesis, genetic chimeras were generated by morula aggregations of transgenic ActB:EYFP transgenic mice homozygous for wild type alleles of Vsx2 [Tg(EYFP/EYFP);Vsx2] and mice that are homozygous for null alleles of Vsx2 (Vsx2orj/orj). Maintenance of neuroretinal identity is assessed by immunostaining for the expression of MITF, a retinal pigmented epithelium (RPE) gene that regulates many pigmentation pathway genes. Neurogenesis is assessed through analysis of cell differentiation using immunostaining for several cell-type specific markers. Analysis of the retinas of Tg(EYFP/EYFP);Vsx2 ↔ Vsx2orj/orj chimeras reveals the presence of MITF expression in Vsx2orJ/orJ cells but not Vsx2+/+ cells indicating that neighboring Vsx2+/+ cells are unable to rescue Vsx2orJ/orJ cells from loss of neuroretinal identity. Analysis also reveals that Vsx2orJ/orJ cell differentiation remains delayed indicating that neighboring Vsx2+/+ cells are unable to rescue Vsx2orJ/orJ cells from delayed neurogenesis. These findings suggest that Vsx2 regulates neuroretinal identity and neurogenesis by cell-autonomous mechanisms. iii TABLE OF CONTENTS ABSTRACT ii LIST OF FIGURES iv ACKNOWLEDGEMENTS v INTRODUCTION 1 MATERIALS AND METHODS 9 RESULTS 15 DISCUSSION 26 REFERENCES 30 iv LIST OF FIGURES 1. The retina 2 2. The Vsx2 gene 3 3. Cell autonomous and cell non-autonomous mechanisms 7 4. Chimera generation techniques 11 5. Chimeric tissue 16 6. Neuroretinal identity 17 7. Neurogenesis 18 8. MITF expression in the wild type and orJ mutant retina 18 9. MITF expression in the wild type and orJ mutant chimeric retina 19 10. Immunohistochemical results at E12.5 1 22 11. Immunohistochemical results at E12.5 2 23 12. Immunohistochemical results at E15.0 1 24 13. Immunohistochemical results at E15.0 2 25 v ACKNOWLEDGEMENTS My sincerest appreciation goes to Dr. Edward Levine for allowing me the privilege and opportunity to study in his laboratory during my undergraduate education, for always being available to answer questions and for always expanding beyond the answer to explain the grander implications of details. Endless thanks to Crystal Sigulinsky for guiding me through all of the learning curves and for being a tremendous mentor and friend. Thank you Dr. Rosemary Gray for introducing me to Dr. Levine and opening the door to an invaluable research opportunity. 1 INTRODUCTION: The retina is the sensory tissue of the eye. This tissue develops from cells in the neural plate, the part of the ectoderm that gives rise to the nervous system, that evaginates to form the optic vesicle which in turn invaginates to form the optic cup. The optic cup is composed of two cell layers: the outer epithelial layer becomes the retinal pigmented epithelium (RPE) and the inner neuroepithelial layer becomes the neuroretina. The cells that compose the inner neuroepithelial layer are the multipotent retinal progenitor cells (RPCs). The RPCs in the neuroretina differentiate to form the six major neurons: rod photoreceptor cells, cone photoreceptor cells, horizontal cells, bipolar cells, amacrine cells and ganglion cells (figure 1). One major glial cell type, Muller glia, is found in the retina (Reh & Levine 1998). RPCs must undergo extensive proliferation in order to supply enough cells to differentiate into each retinal cell type. RPCs differentiate sequentially with the retinal ganglion cells (RGCs) being the first cells to differentiate followed by the majority of cone, horizontal and amacrine cells which differentiate at approximately the same time. The majority of rod cells differentiate soon afterwards and the bipolar cells are among the last to differentiate (Cepko et al., 1996; Das et al., 2009; Marquardt, 2003). While there is a trend in the order at which retinal cells differentiate there is considerable overlap illustrating the complexity involved in the development of the retina. 2 Figure 1: Diagram of the cellular structure of the retina. Vsx2 is a homeobox gene essential for proper retinal development that is expressed in the RPCs and bipolar cells and is important in the regulation of RPC behaviors including proliferation and differentiation (Bone-Larson et al., 2000; Burmeister et al., 1996; Green et al., 2003; Horsford et al., 2005; Rowan et al., 2004; Sigulinsky et al., 2008). Vsx2 is highly conserved throughout biology with consensus sequences found in human, mouse, fish, Drosophila and nematode. The regions include the homeodomain, an octapeptide, the OAR domain (named after the genes OTP, Aristaless and Rax which are homeobox-containing proteins that also house the domain) and the CVC domain (also named after the homeobox-containing proteins in which it is found: Chx10 - the former name for Vsx2, Vsx1 and Ceh10) (figure 2). The 3 homeodomain and the CVC domain are identical in mouse and human (Liang and Sandell, 2008). Figure 2: The domains in Vsx2. The octapeptide is shown in green, the homeodomain in red, the CVC domain in blue and the OAR domain in yellow. The arrow shows the approximate location for the point mutation in the Vsx2orJ allele. Ocular retardation J (orJ) mice are homozygous for the Vsx2orJ allele that contains a nonsense mutation (Liang and Sandell, 2008). Since the VSX2 protein is not expressed, the Vsx2orJ allele is considered a null allele. OrJ mice exhibit blindness, microphthalmia (a small eye phenotype) a cataractous lens and a thin and poorly differentiated retina (Burmeister et al., 1996). Humans with mutations in the Vsx2 gene present clinically with microphthalmia which is reported in 11% of congenitally blind children (Bar-Yosef et al., 2004; Ferda Percin et al., 2000; Verma and Fitzpatrick, 2007). Vsx2 is believed to function via cell intrinsic mechanisms due to its role as a transcription factor (as the mechanism of transcription is inherently a cell intrinsic process) and the fact that loss of function phenotypes manifest in RPCs and bipolar cells (other retinal cell types cease to express Vsx2 upon differentiation) where Vsx2 is expressed. However, because the orJ phenotype is so severe, it is possible that Vsx2 may 4 regulate pathways that involve cell signaling. If this is the case, the orJ phenotype may be the manifestation of disrupted cell autonomous and cell non-autonomous mechanisms. One study attempted to assess the autonomy of Vsx2 phenotypes (Kindiakov and Koniukhov, 1986) however the results were difficult to interpret and largely inferred due to a lack of biotechnological advances available today. Thus, the subject of cell non-autonomous mechanisms of Vsx2 function requires re-evaluation. Cell non-autonomous mechanisms and response to cell extrinsic factors are important for ocular development (Dorsky et al., 1997; Dorsky et al., 1995; Furukawa et al., 2000; Hashimoto et al., 2006; Moshiri et al., 2008; Mu et al., 2004; Shkumatava et al., 2004) and recent work from our laboratory reveals evidence of Vsx2-mediated cell non-autonomous mechanisms involved in the regulation of RPC proliferation via Hedgehog signaling (Sigulinsky et al., 2008). For these reasons we hypothesize that Vsx2 regulates neuroretinal identity and neurogenesis by cell non-autonomous mechanisms as the neuroretina should not be considered exempt from a possible cell non-autonomous mechanism, specifically a Vsx2- mediated mechanism, which regulates RPC behavior. Neuroretinal identity In the developing optic cup, optic neuroepithelial RPE precursor cells express MITF but not VSX2 while neuroretina precursor cells express VSX2 but not MITF. In orJ mice however, optic neuroepithelial neuroretina precursor cells express both Mitf and Vsx2 mRNA (the mutant Vsx2 gene is transcribed but not translated due to the premature stop codon). Mitf transcriptionally regulates many pigmentation pathway genes in RPE precursor cells. In addition, ectopic Mitf expression likely plays a role in the 5 hyperpigmentation of the orJ neuroretina. The ectopic expression of Mitf in orJ neuroretinal cell precursors suggests that the identity of the cells is compromised (Horsford et al., 2005). Although Vsx2 is shown to transcriptionally repress Mitf (Bharti et al., 2008; Horsford et al., 2005) and repression of Mitf by Vsx2 is a cell autonomous process, this does not exclude the possibility of a Vsx2-mediated cell non-autonomous mechanism in regards to maintenance of neuroretinal identity. Neurogenesis Retinal neurogenesis, the development of the six major types of neurons in the retina from the RPCs, begins at about embryonic day 11.5 (E11.5) in the wild type mouse and occurs in a central to peripheral wave (Aung et al., 2008; Das et al., 2009). In the orJ mutant, retina neurogenesis is delayed by about two days (Sigulinsky et al., 2008). Retinal neurogenesis can be assessed by analysis of class III β-Tubulin (AcTUBB3). AcTUBB3 is one of the six tubulin types expressed in mammalian tissue (Das et al., 2009; Sullivan et al., 1985). The acetylated form of tubulin is less dynamic and is found in cells that experience less change, such as postmitotic neuron precursors. Hence, assessment of AcTUBB3 expression can be used as an early indicator of RPC differentiation. In addition, retinal cell class-specific markers such as Pou4f2, Islet 1, Otx2 and Ptf1a can be used to assess RPC differentiation. Pou4f2 expression is essential for the differentiation of RPCs to ganglion cells as Pou4f2-/- mice show a 70% decrease in ganglion cell number (Gan et al., 1996). Like Vsx2, Islet 1, Otx2 and Ptf1a are also homeobox genes. Islet1 expression is essential for the differentiation of RPCs to amacrine, horizontal and bipolar cells as there is a significant reduction of these cell types 6 in conditional Islet 1 mutants (Elshatory et al., 2007; Galli-Resta et al., 1997). Otx2 is expressed in photoreceptor cells. Otx2 conditional knockout mice exhibited a complete loss of photoreceptors; additionally, retroviral gene transfer of Otx2 into mouse RPCs stimulated the RPCs to become photoreceptors (Nishida et al., 2003). Ptf1a expression is essential for the differentiation of RPCs to horizontal and amacrine cells as there is a lack of development of horizontal and amacrine cells in mice in which the Ptf1a gene had been inactivated and as evidenced by the transdifferentiation of already existing mouse amacrine and horizontal cells to retinal ganglion cells upon Ptf1a inactivation (Das et al., 2009; Fujitani et al., 2006). In addition, amacrine and horizontal cells are completely absent in Ptf1a knockout mouse retinal explants (Nakhai et al., 2007). BHLHB5 is required for bipolar and amacrine cell development as studies show a significant reduction in bipolar and amacrine cells upon deletion of BHLHB5 (Feng et al., 2006). The specificity of these transcripts to differentiated cell types makes it possible to trace and compare RPC differentiation in orJ and wild type retinas and to assess the autonomy of the mechanism by which Vsx2 regulates neurogenesis. 7 Figure 3: (A) In a Vsx2-mediated cell autonomous mechanism for maintenance of neuroretinal identity or neurogenesis, Vsx2 directly represses the determining factor(s) for the development of the phenotype. (B) In a Vsx2-mediated cell non autonomous mechanism for maintenance of neuroretinal identity or neurogenesis, Vsx2 plays a role in the regulation of an exogenous compound which may serve directly as the ligand for a receptor that in turn represses the factor(s) that determine the development of the phenotype, or indirectly as a member of a signaling pathway that ultimately produces the ligand for the receptor that in turn represses the factor(s) that determine the development of the phenotype. Experimental Approach Chimeras are animals whose tissues exhibit cells of different genotypes due to the different embryonic origins of the cells. The first mouse chimeras were developed in the 1960s by aggregating two morulae at the eight cell stage (Tam and Rossant, 2003). This technique has since been routinely used in the analysis of genetic function including the analysis of the autonomy of specific genes (figure 4). The chimera is an efficient tool for determining the autonomy of genes because it allows for the coexistence of two different genotypes in the same tissue, in vivo. Chimeras create a system in which genes are separated by cell boundaries but the proteins they code for are not (with the exception of membrane proteins) and this makes it possible to determine whether certain proteins function via a cell autonomous or cell non-autonomous mechanism. Genetic chimeras 8 are produced to assess the autonomy of the neuroretinal identity and neurogenesis phenotypes. The cells of the chimeric animals are of two genotypes: orJ cells that are homozygous for the null Vsx2 allele (Vsx2orJ/orJ) and wild type cells (Vsx2+/+) that are homozygous for the wild type Vsx2 allele. In order to distinguish the orJ cells from the wild type cells, transgenic mice that express endogenous yellow fluorescent protein (Tg(EYFP/EYFP)) are used as the source of the wild type cells. This system of mutant and wild type cells coexisting within the same organism while remaining distinguishable based on their fluorescent properties allows for the assessment of orJ cell behavior in a wild type cell environment. If the neuroretinal identity and/or neurogenesis phenotypes are regulated by cell non-autonomous mechanisms the orJ cells in Tg(EYFP/EYFP);Vsx2+/+ ↔ Vsx2orJ/orJ (chimeric animals whose tissue is composed of cells from Vsx2orJ/orJ mice and transgenic mice that express EYFP and are homozygous for the wild type Vsx2 allele) chimeras will exhibit the wild type cell phenotypes of maintained neuroretinal identity and/or a normal rate of neurogenesis. 9 MATERIALS AND METHODS Chimera Generation The animals used were obtained from Jackson Labs (Bar Harbor, ME) and were on a 129/Sv background strain. In preparation for chimera generation, Vsx2+/+ and Vsx2orJ/orJ mice are superovulated by injection of pregnant mares serum (PMS), which stimulates follicle formation and ovulation in the mouse, between 8:00 and 10:00 AM. 48 h later mice receive an injection of human chorionic gonadotropin hormone (HCG; ) which serves to promote maintenance of the endometrial lining of the uterus and ovulation. These mice are immediately mated with males of the same genotype to avoid the generation of heterozygous embryos and hence facilitate distinguishing mutant and wild type chimeras. The next morning, females are checked for a copulatory plug. Plugged females are sacrificed, their uteruses removed, washed and transferred to the laboratory of the University of Utah Transgenic/Gene Targeting facility where the remainder of the chimera generation procedure is performed. At the Transgenic/Gene Targeting Facility the zona pelucida is removed from the morulae of our mice and those of Tg(EYFP/EYFP);Vsx2+/+ (transgenic mice that express endogenous yellow fluorescent protein and are homozygous for the wild type Vsx2 allele) mice from the University of Utah Transgenic/Gene Targeting facility (these mice are superovulated and mated following the same procedure described above). Once the zona pelucida is removed, one Vsx2orJ/orJ and one Tg(EYFP/EYFP);Vsx2+/+ or one Vsx2+/+ and one Tg(EYFP/EYFP);Vsx2+/+ (for the generation of control chimeras) morula are placed in close proximity overnight to aggregate. These embryos are then implanted in the uterus 10 of a pseudopregnant female (a female mated with a sterile male) and allowed to develop until E12.5. In other rounds of chimera generation, different procedures are used in an attempt to produce chimeras with a greater ratio of orJ to wild type cells. The technique is slightly altered when two Vsx2orJ/orJ morulae and one Tg(EYFP/EYFP);Vsx2+/+ morula are left overnight to aggregate and form an embryo; the rest of the chimera generation procedure is followed as described above. In another technique, EYFP embryonic stem cells are injected into an orJ blastocyst and the blastocyst is implanted into a pseudopregnant female as described above (figure 4). 11 Figure 4: Diagram illustrating the procedures for chimera generation. 12 Table 1: Primary Antibodies Tissue Harvesting and Preparation Whole eyes or isolated retinas are dissected from the eyes of the embryos in Hank's buffered saline solution (HBSS) at room temperature. Tissue used for immunohistochemistry with MITF antibodies is fixed in 4% paraformaldehyde (PFA) for 2 hs and retinas used for immunohistochemistry with all other antibodies is fixed in 4% PFA for 30 min. After fixation, tissue is washed twice with 1X phosphate buffered saline (PBS: 320 g NaCl, 8 g KCl, 57.6 g Na2HPO4, 9.6 g KH2PO4 dissolved in 3.2 L of H20 and 10 N NaOH to reach pH of 7.4 and a volume of 4 L) and dehydrated by exposing tissue to a series of sucrose solutions of 5%, 10% and 20% for two washes of 5 min, 10 min and 20 min respectively. Following dehydration, tissue is cryprotected in optimum cutting temperature compound (Sakura Finetek, Torrance CA) and stored at -80 °C. 12.0 μm cryosections are placed on glass microscope slides and then stored at -20 °C. Antigen Host Target Dilution Source Pou4f2 Goat Retinal Ganglion cell precursors 1:50 Santa Cruz Biotechnology Mitf Mouse RPE 1:400 Exalpha Biologicals AcTUBB3 Mouse Neuronal precursors 1:1000 Covance Islet Mouse Amacrine cells and retinal ganglion cells 1:50 DSHB Otx2 Rabbit Photoreceptor and amacrine precursors 1:50,000 Santa Cruz Ptf1a Rabbit Amacrine and horizontal cell precursors 1:800 Helena Edlund 13 Immunohistochemistry Sections are washed with PBS for 3 min, three times in order to remove the cryoprotectant from the tissue sections before pretreatment in blocking buffer (2% goat or donkey serum, 0.15% TritonX, 0.01% sodium azide in PBS) for 30 min. The primary antibodies used in this experiment are goat anti-Pou4f2 (Santa Cruz Biotechnology, CA), mouse anti-Mitf (Exalpha Biologicals, MA), mouse anti-Acetylated Class III Beta- Tubulin3 (Covance, CA), mouse anti-Islet1 (Developmental Studies Hybridoma Bank, IA), rabbit anti-Otx2 (Santa Cruz Biotechnology, CA), and rabbit anti-Ptf1a (gift from Helena Edlund) (table 1). Sections are incubated in the primary antibody solution at room temperature for 1 h and then overnight at -4 °C. The primary antibody solution is removed by rinsing the slides with PBS for 10 min, two times. Subsequently, the slides are incubated with species-specific secondary antibodies conjugated to AlexaFlour 568 (Invitrogen, Carlsbad, CA) diluted at 1:1000 in blocking solution for 1 h at room temperature. Afterwards, sections are rinsed twice with PBS for 10 min at room temperature. Sections are counterstained for nuclei by incubating for 30 min in 4,6-diamidino-2-phenylindole (DAPI; Fluka, Switzerland) diluted 1:4000 in H2O and then rinsed for 3 min, three times in distilled H2O at room temperature. Sections are allowed to dry for 1 h before mounting with Vectashield® mounting medium (Burlingame, CA) and covering with glass coverslips in preparation for microscopy. 14 Imaging, Image Analyses and Data Quantification All sections are imaged with an Olympus Fluoview Confocal 1000 microscope. The images are edited by uniform, artificial coloring of positive cells using Photoshop CS5 (Adobe). All other aspects of the images are untouched. Artificial coloring allows cells positively labeled by immunohistochemistry to be more easily distinguished from that of EYFP positive cells. Cell counts are performed using Photoshop CS5 (Adobe) and ImageJ (NIH). 15 RESULTS Cell distribution patterns of chimeric tissue are important for analysis of cell behavior in different cell environments. Several rounds of chimera generation were completed to obtain a variety of cell distribution patterns. Initial rounds of chimera generation yielded chimeras with a lower orJ:wild type cell ratio (Figs. 5B,C) than desired for analyses of clusters of mutant cells in a wild type cell environment (Fig. 5A). Therefore, two other methods of chimera generation were performed to obtain chimeras of higher orJ:wild type cell ratios. Blastocyst injection chimera generation and a triple morulae aggregation technique in which orJ:wild type morulae in a ratio of 2:1, a technique which has not been quoted before, were used to generate chimeras. Most embryos generated using the blastocyst injection technique were unusable due to either the absence of wild type cells (Fig. 5F) or too high a orJ:wild type cell ratio (Figs. 5E,G). Embryos generated by the triple morulae aggregation technique were usable and exhibit a higher orJ:wild type cell ratio (Figs. G-I). The chimeras from the three morulae aggregation technique exhibited a higher ratio of orJ to wild type cells therefore allowing for the analysis of groups of orJ cells in a wildtype cell environment. 16 Figure 5: A,B,C) Retinas from chimeric mice generated from the 1:1 orJ:wild type morula aggregation technique. D,E,F) Chimeric limbs from mice generated using the blastocyst injection technique. Only image D shows significant chimerism. Image E shows very little chimerism while image F is not chimeric. Images D and E are representative of the majority of the animals generated using the blastocyst injection technique. G,H,I) Retinas from chimeric mice generated from 2:1 orJ:wild type morula aggregation technique. Tissue from this technique yielded a greater variety of cell distribution patterns than the 1:1 orJ:wild type morula aggregation technique. 17 Cell non-autonomous Cell autonomous EYFP positive wild type cell EYFP negative orJ cell MITF positive MITF negative To determine whether the Vsx2 gene plays a role in the regulation of the neuroretinal identity (figure 6) and neurogenesis (figure 7) phenotypes via cell autonomous or cell non-autonomous mechanisms in mouse, orJ cells are observed in a wild type cell environment. If orJ cells respond to the wild type cell environment by exhibiting changes in protein expression that reflect a wild type cell phenotype, this would suggest that Vsx2 regulates retinal cell phenotypes via cell non-autonomous mechanisms. The opposite response in which orJ cells maintain their mutant phenotype despite their close proximity to wild type cells, would suggest that Vsx2 regulates retinal cell phenotypes via cell autonomous mechanisms. Figure 6: Schematic illustrating the two possible outcomes for the assessment of the autonomy of the neuroretinal identity phenotype. 18 Wild type (EYFP+) orJ (EYFP-) Progenitor layer (RPCs) Differentiated cell layer (DCL) Cell autonomous control Cell non-autonomous input Figure 7: Diagram illustrating the two possible outcomes for the assessment of the autonomy of the neurogenesis phenotype. Generously contributed by Crystal Sigulinsky (University of Utah, Salt Lake City, UT). Figure 8: A) Wild type retina showing the repression of MITF in the neuroretina and the expression of MITF in the RPE. B) orJ retina showing the ectopic expression of MITF in the neuroretina and the expression of MITF in the RPE. Scale bars 100 um. 19 Figure 9: A,B,C) Control chimera shows no ectopic expression of MITF in the neuroretina. D,E,F,G,H,I) orJ chimera shows ectopic expression of MITF in the orJ mutant cells whereas surrounding wild type cells do not. Scale bars 100 um. 20 The wild type retina does not express MITF whereas the orJ retina ectopically expresses MITF (figure 8). In the wild type chimeric retina there are no MITF positive cells regardless of whether they are EYFP+ or EYFP- (figure 9B). In the mutant chimera, orJ cells continue to ectopically express MITF (figures 9E,H). Clusters of orJ cells (figure 9E) as well as single orJ cells surrounded by wild type cells (figure 9H) express MITF suggesting that an orJ cell's proximity to other mutant cells is irrelevant to whether the cell maintains the mutant phenotype. MITF expression in orJ cells is present in the peripheral as well as the central retina (where transdifferentiation is rare) showing that even in the peripheral retina where transdifferentiation is frequent, orJ cells maintain their mutant phenotype. Retinal neurons differentiate normally in the wild type chimeric retina. At E12.5 wild type chimera central RPCs differentiate to form ganglion, amacrine, photoreceptor and bipolar cell precursors (figures 10A-E and K-O). At this age, neurogenesis has not yet reached the peripheral retina. By E15.0, neurogenesis in the wild type chimeric retina reaches the peripheral retina and lamination is evident in the arrangement of the retinal cell types highlighted by individual immunohistochemical stainings. Positive cells in the wild type chimeric retina are both EYFP+ and EYFP- (figure 12). Retinal neuron differentiation is delayed in the mutant chimeric retina. At E12.5 EYFP+ wild type chimera central RPCs differentiate to form ganglion, amacrine, photoreceptor and bipolar cell precursors whereas there are no positive, neural precursor cells in the EYFP- mutant cells (figures 10F-J and P-T, figure 11F-J and P-T). As with wild type cells in the wild type chimeric retina, neurogenesis in wild type cells in the mutant chimera has not yet reached the peripheral retina. By E15.0, EYFP- mutant cells 21 in the chimeric retina exhibit neurogenesis (figure 13). Lamination is evident in the arrangement of the retinal cell types (figure 13). 22 Figure 10: Neurogenesis in chimeras at age E12.5. Positive cells include Ptf1a, Pou4f2, BHLHB5 and Islet1 positive cells. (B,C,D,E) Positive cells are present in the wild type chimeric retina in both EYFP+ and EYFP- wild type cells. (G,H,I,J) Positive cells are present in EYFP+, wild type cells. Positive cells are absent in the EYFP-, orJ cells. (A,F) The size of the orJ retina is severely reduced as compared to the wild type retina. (K,P) The dashed line demarcates the retina. (L,M,N,O) AcTUBB3 positive cells are present in the central retina in both EYFP+ and EYFP- wild type cells. (Q,R,S,T) AcTUBB3 positive cells are present in central retina in EYFP+ cells. AcTUBB3 positive cells are absent in the EYFP-, orJ cells. (A,B,F,G,K,L,P,Q) Scale bars 100 um. (C,H,M,R) Scale bars 40 um. 23 Figure 11: Neurogenesis in chimeras at age E12.5. (A,B,C,D,E) Pou4F2 positive cells are present throughout the retina in both EYFP+ and EYFP- wild type cells. (F) Tissue to the right of the dashed line is retinal tissue. Note that the size of the mutant chimera is severely reduced as compared to the wild type retina. (G,H,I,J) Pou4F2 positive cells are present in EYFP+, wild type cells. Pou4F2 positive cells are absent in EYFP-, orJ cells. (K,L,M,N,O) Otx2 positive cells are present in both EYFP+ and EYFP-, wild type cells. (P,Q,R,S,T) Otx2 positive cells are present in several EYFP+ wild type cells. Otx2 positive cells are absent in EYFP-, orJ cells. (K,P) The size of the orJ retina is severely reduced as compared to the wild type retina. (A,B,F,G,K,L,P,Q) Scale bars 100 um. (C,H,M,R) Scale bars 40 um. 24 Figure 12: Neurogenesis in wild type chimeras at age E15.0. Scale bars 100 um. 25 Figure 13: Neurogenesis in mutant chimeras at age E15.0. Scale bars 100 um. 26 DISCUSSION The data presented in this study demonstrate that orJ cells in the chimeric retina continue to ectopically express MITF and hence maintain a compromised neuroretinal identity in the chimeric retina. OrJ cells also maintain delayed neurogenesis as evidenced by the absence of retinal neurons at E12.5 in the chimeric retina, an age when neurogenesis has begun progression in the wild type chimeric retina. These results suggest that Vsx2 functions via a cell autonomous mechanism in the regulation of RPC maintenance of neuroretinal identity and neurogenesis. Our results disprove the hypothesis that Vsx2 regulates cell non-autonomous pathways that play a role in maintaining neuroretinal identity and neurogenesis. This does not exclude the possibility that Vsx2 plays a role in the regulation of cell non-autonomous mechanisms that control other aspects of the orJ phenotype. One example of an aspect of the orJ phenotype that may be regulated by cell non-autonomous mechanisms is that of RPC proliferation in the orJ retina. Recent unpublished data from our lab shows that the proliferation phenotype of the orJ chimeric retina exhibits a cell non-autonomous component. OrJ cells in the mutant chimeric retina show increased proliferation in the peripheral retina compared to the germline orJ mutant suggesting that Vsx2 plays a cell non-autonomous role in the RPC proliferation phenotype. Interestingly, RPC proliferation worsens in central retina of the mutant chimera as compared to the germ line mutant. This is unexpected as previous research (Sigulinksy et. al, 2008) suggests that hedgehog signaling, which promotes RPC proliferation, progresses in a central to peripheral wave in the retina and hence RPC proliferation is also expected to progress in a central to peripheral wave and not to cause isolated proliferation. 27 A full understanding of the role of Vsx2 in the development of the retina would require understanding Vsx2 in the context of other retinal development signaling pathways. Studies have shown that the RAF/MEK/ERK pathway is a key pathway in the regulation of retinal development. The presumptive RPE of the optic cup of a developing retina when exposed to fibroblast growth factors (FGFs) 1 and 2 does not develop into RPE; instead, a structure resembling a second retina develops. Like a presumptive retina, this structure does not express Mitf, however Vsx2 is expressed and exhibits cell proliferation (Nguyen and Arnheiter, 2000). Removal of the surface ectoderm, the source of FGFs, results in the development of a presumptive retina that expresses Mitf but not Vsx2 and resembles a RPE (Nguyen and Arnheiter, 2000). FGF9 expression in the presumptive mouse RPE also results in the development of a second retina (Zhao et al., 2001). This suggests that FGFs are sufficient and necessary for the development of retina and that FGFs, like Vsx2, downregulate Mitf. FGFs are ligands for receptor tyrosine kinases of which one signaling pathway is the MAPK pathway (Szebenyi and Fallon, 1999). When Ras is activated in the presumptive RPE, it develops into a second retina (Zhao et al., 2001). This suggests that FGFs signal for the optic cup to become retina by activating the MAPK pathway. Therefore, FGF ligands activate the MAPK pathway which results in the upregulation of Vsx2 which in turn has been shown to transcriptionally repress Mitf (Bharti et al., 2008; Horsford et al., 2005). FGF signaling is also involved in the maintenance of neuroretinal identity; the question then arises, where does Vsx2 fit in the context of the FGF signaling 28 pathway? Is it possible that Vsx2 does not directly repress Mitf but that Vsx2 activates FGF signaling or is it more probable that FGF activates Vsx2? If Vsx2 activated FGF signaling it would be expected that orJ cells in a chimeric retina would not express MITF as neighboring wild type cells would presumably produce the FGFs necessary to repress MITF and hence rescue the orJ cells from a compromised neuroretinal identity. However, results show that orJ cells maintain a compromised neuroretinal identity as MITF continues to be ectopically expressed. Therefore, it is more probable that FGF activates Vsx2. Future studies in which orJ chimeric retinas are treated with an FGF agonist such as SU5402 would elucidate the relationship between FGF signaling and Vsx2. Future studies may also benefit from modifications to the animal model as the experimental method of chimera generation is not the only technique that could have been used to determine whether Vsx2 functions via a cell autonomous or non-autonomous mechanism in the orJ retina. Genetic mosaic and DNA recombination techniques could be used to determine whether Vsx2 functions via cell autonomous or cell non-autonomous mechanisms. Like chimeric animals, the tissues of genetic mosaic animals exhibit cells of different genotypes. However, the source of the differences in genotype between cells in genetic mosaic organisms is different than that of chimeric organisms. Genetic mosaic organisms are generated by altering the genotype of the cells early in development. In most cases genetic alteration is accomplished through DNA recombination techniques using floxed alleles and cre recombinase enzymes. Because only the cells from a single organism are used to produce the genetic mosaics, variables 29 that would be introduced by differing background strains of cells (as with chimera generation) are eliminated. DNA recombination techniques using floxed alleles and cre recombinase could be employed to assess how Vsx2 regulates neuroretinal phenotypes temporally. That is, using DNA recombination techniques it is possible to knock out Vsx2 at a specific stage in embryonic development and observe the effect on the development of the neuroretina at that point in time. The generation of genetic mosaics and the use of DNA recombination techniques in future experiments would allow for observation of the effect of loss of Vsx2 at different time points which may reveal novel Vsx2 functions. In conclusion, orJ mutant cells maintain a compromised neuroretinal identity and delayed neurogenesis in the chimeric retina suggesting Vsx2 regulates the neuroretinal identity and neurogenesis phenotypes via a cell autonomous mechanism. These results represent a step in developing an understanding Vsx2 in the context of other retinal development pathways and ultimately understanding how the retina as a whole develops. 30 References Aung, T., et al., 2008. Molecular analysis of CHX10 and MFRP in Chinese subjects with primary angle closure glaucoma and short axial length eyes. Mol Vis. 14, 1313-8. Bar-Yosef, U., et al., 2004. CHX10 mutations cause non-syndromic microphthalmia/ anophthalmia in Arab and Jewish kindreds. Hum Genet. 115, 302-9. Bharti, K., et al., 2008. Alternative promoter use in eye development: the complex role and regulation of the transcription factor MITF. Development. 135, 1169-78. Bone-Larson, C., et al., 2000. Partial rescue of the ocular retardation phenotype by genetic modifiers. J Neurobiol. 42, 232-47. Burmeister, M., et al., 1996. Ocular retardation mouse caused by Chx10 homeobox null allele: impaired retinal progenitor proliferation and bipolar cell differentiation. Nat Genet. 12, 376-84. Cepko, C. L., et al., 1996. Cell fate determination in the vertebrate retina. Proc Natl Acad Sci U S A. 93, 589-95. Das, G., et al., 2009. Cyclin D1 fine-tunes the neurogenic output of embryonic retinal progenitor cells. Neural Dev. 4, 15. Dorsky, R. I., et al., 1997. Regulation of neuronal diversity in the Xenopus retina by Delta signalling. Nature. 385, 67-70. Dorsky, R. I., et al., 1995. Xotch inhibits cell differentiation in the Xenopus retina. Neuron. 14, 487-96. Elshatory, Y., et al., 2007. Islet-1 controls the differentiation of retinal bipolar and cholinergic amacrine cells. J Neurosci. 27, 12707-20. Feng, L., et al., 2006. Requirement for Bhlhb5 in the specification of amacrine and cone bipolar subtypes in mouse retina. Development. 133, 4815-25. Ferda Percin, E., et al., 2000. Human microphthalmia associated with mutations in the retinal homeobox gene CHX10. Nat Genet. 25, 397-401. Fujitani, Y., et al., 2006. Ptf1a determines horizontal and amacrine cell fates during mouse retinal development. Development. 133, 4439-50. Furukawa, T., et al., 2000. rax, Hes1, and notch1 promote the formation of Muller glia by postnatal retinal progenitor cells. Neuron. 26, 383-94. Galli-Resta, L., et al., 1997. Mosaics of islet-1-expressing amacrine cells assembled by short-range cellular interactions. J Neurosci. 17, 7831-8. Gan, L., et al., 1996. POU domain factor Brn-3b is required for the development of a large set of retinal ganglion cells. Proc Natl Acad Sci U S A. 93, 3920-5. Green, E. S., et al., 2003. Genetic rescue of cell number in a mouse model of microphthalmia: interactions between Chx10 and G1-phase cell cycle regulators. Development. 130, 539-52. Hashimoto, T., et al., 2006. VEGF activates divergent intracellular signaling components to regulate retinal progenitor cell proliferation and neuronal differentiation. Development. 133, 2201-10. Horsford, D. J., et al., 2005. Chx10 repression of Mitf is required for the maintenance of mammalian neuroretinal identity. Development. 132, 177-87. Kindiakov, B. N., Koniukhov, B. V., 1986. [Mutant gene expression in murine aggregation chimeras. 5. The ocular retardation and fidget genes]. Ontogenez. 17, 47-55. 31 Liang, L., Sandell, J. H., 2008. Focus on molecules: homeobox protein Chx10. Exp Eye Res. 86, 541-2. Marquardt, T., 2003. Transcriptional control of neuronal diversification in the retina. Prog Retin Eye Res. 22, 567-77. Moshiri, A., et al., 2008. Near complete loss of retinal ganglion cells in the math5/brn3b double knockout elicits severe reductions of other cell types during retinal development. Dev Biol. 316, 214-27. Mu, X., et al., 2004. Discrete gene sets depend on POU domain transcription factor Brn3b/Brn-3.2/POU4f2 for their expression in the mouse embryonic retina. Development. 131, 1197-210. Nakhai, H., et al., 2007. Ptf1a is essential for the differentiation of GABAergic and glycinergic amacrine cells and horizontal cells in the mouse retina. Development. 134, 1151-60. Nguyen, M., Arnheiter, H., 2000. Signaling and transcriptional regulation in early mammalian eye development: a link between FGF and MITF. Development. 127, 3581-91. Nishida, A., et al., 2003. Otx2 homeobox gene controls retinal photoreceptor cell fate and pineal gland development. Nat Neurosci. 6, 1255-63. Rowan, S., et al., 2004. Transdifferentiation of the retina into pigmented cells in ocular retardation mice defines a new function of the homeodomain gene Chx10. Development. 131, 5139-52. Shkumatava, A., et al., 2004. Sonic hedgehog, secreted by amacrine cells, acts as a short-range signal to direct differentiation and lamination in the zebrafish retina. Development. 131, 3849-58. Sigulinsky, C. L., et al., 2008. Vsx2/Chx10 ensures the correct timing and magnitude of Hedgehog signaling in the mouse retina. Dev Biol. 317, 560-75. Sullivan, K. F., et al., 1985. Apparent gene conversion between beta-tubulin genes yields multiple regulatory pathways for a single beta-tubulin polypeptide isotype. Mol Cell Biol. 5, 2454-65. Szebenyi, G., Fallon, J. F., 1999. Fibroblast growth factors as multifunctional signaling factors. Int Rev Cytol. 185, 45-106. Tam, P. P., Rossant, J., 2003. Mouse embryonic chimeras: tools for studying mammalian development. Development. 130, 6155-63. Verma, A. S., Fitzpatrick, D. R., 2007. Anophthalmia and microphthalmia. Orphanet J Rare Dis. 2, 47. Zhao, S., et al., 2001. Patterning the optic neuroepithelium by FGF signaling and Ras activation. Development. 128, 5051-60.