• Xenopus laevis;
  • Nrl;
  • Nr2e3;
  • retina;
  • lens;
  • cell fate;
  • transfection;
  • development


  1. Top of page
  2. Abstract
  6. Acknowledgements

Transformation of undifferentiated progenitors into specific cell types is largely dependent on temporal and spatial expression of a complex network of transcription factors. Here, we examined whether neural retina leucine zipper (Nrl) and photoreceptor-specific nuclear receptor Nr2e3 transcription factors contribute to cell fate determination. We cloned the Xenopus Nr2e3 gene and showed that its temporal and spatial expression is similar to its mammalian ortholog. We tested its in vivo function by misexpressing these transcription factors in Xenopus eye primordia, demonstrating that either human Nr2e3 or Nrl directed photoreceptor precursors to become rods at the expense of cones. Furthermore, overexpression of Xenopus Nrl dramatically increased the number of lens fibers, whereas human Nrl did not, suggesting evolutionary divergence of function of the Nrl gene family. Misexpression of Nrl and Nr2e3 together were more effective than either transcription factor alone in directing precursors to the rod fate. Developmental Dynamics 236:1970–1979, 2007. © 2007 Wiley-Liss, Inc.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Rods and cones are photoreceptors that carry out phototransduction in vertebrate retina and express distinct sets of genes for this purpose (Hisatomi and Tokunaga, 2002). These proteins are responsible for the functional differences in light responses: cones mediate color vision at high light levels, whereas rods are specialized for “monochrome” vision at low light levels. Photoreceptors are one of several cell types in retina (Dowling, 1987), which are generated from multipotent progenitors in a predictable sequence, starting with retinal ganglion cells followed by cone, amacrine, horizontal, bipolar cells, and ending with rod and Müller glia (Holt et al., 1988; Cepko et al., 1996). The competence model of cell fate determination proposes that early retinal progenitors become more restricted in their ability to produce “early born” cell types as they pass through the various developmental stages (Livesey and Cepko, 2001). The restriction of competence is thought to be regulated through continuous and precise control of sets of retina-specific and more widely expressed transcription factors (reviewed in Cepko et al., 1996; Cook, 2003). Several transcription factor families involved in retinal differentiation have been identified (Cepko, 1999; Dyer and Cepko, 2001; Moore et al., 2002; Wang and Harris, 2005). However, the precise combinations of transcription factors that specify individual cell types remain incompletely characterized. The major focus of the work reported here was to understand the role of Nr2e3, Nrl, and Otx5b genes in cell fate determination.

The Otx5b transcription factor, expressed in dividing retinal progenitors and adult photoreceptors, has been shown to regulate specification of both rods and cones (Viczian et al., 2003). Phylogenetic analysis shows that Otx5b and cone–rod homeobox (Crx) are orthologous (Germot et al., 2001; Sauka-Spengler et al., 2001). Crx knockout mice express rod and cone-specific genes (Furukawa et al., 1999), suggesting that Crx by itself is not a key determinant of photoreceptor cell fate and is likely to work cooperatively with other transcription factors such as Nrl and Nr2e3.

The rod photoreceptor-specific neural retinal leucine zipper (Nrl) gene, a transcription factor in the large Maf superfamily, is an essential regulator of photoreceptor differentiation (Mears et al., 2001; Swain et al., 2001) and an activator of rod gene expression (Rehemtulla et al., 1996; Lerner et al., 2001; Pittler et al., 2004; Yoshida et al., 2004). Nrl is evolutionarily conserved among vertebrates (Whitaker and Knox, 2004) and can interact with other transcription factors, such as Crx (Chen et al., 1997). In mouse, Nrl expression could be detected as early as embryonic day (E) 12, when rods begin to differentiate, and is maintained into adulthood (Swain et al., 2001; Akimoto et al., 2006). Targeted deletion of Nrl in mice (Nrl (−/−)) abolished formation of rods without changing the number of photoreceptors (Mears et al., 2001). These mice also have elevated expression of cone-specific genes (Mears et al., 2001), and recent electrophysiological and morphological studies suggest that the Nrl (−/−) photoreceptors are very similar to native cones (Nikonov et al., 2005). This “all cone” phenotype could be explained if the role of Nrl is to direct photoreceptor progenitors to become rods; in its absence, progenitor cells default to a cone fate. In this study, we examined whether progenitor cells are responsive to the hypothesized instructive role of Nrl.

The retina-specific orphan nuclear receptor (Nr2e3) (Kobayashi et al., 1999) also plays a crucial role in photoreceptor development and maintenance (Akhmedov et al., 2000; Haider et al., 2001; Yanagi et al., 2002). Recent evidence indicates that Nr2e3 is a transcription factor that activates rod-specific and represses cone-specific genes (Cheng et al., 2004; Chen et al., 2005; Corbo and Cepko, 2005; Peng et al., 2005). Based upon several observations, it appears that Nrl and Nr2e3 have similar or overlapping functions during photoreceptor differentiation. rd7 mice, which harbor mutations of Nr2e3, exhibit morphological similarities to Nrl (−/−) retinas, characterized by abnormal rods while forming excessive numbers of S-cones (Haider et al., 2001; Mears et al., 2001). However, Nrl (−/−) retinas exhibit an absence of rod gene transcripts, while Nr2e3rd7 photoreceptors abnormally express rod and cone genes within the same cell (Chen et al., 2005; Corbo and Cepko, 2005). In wild-type mice, initial Nrl expression precedes that of Nr2e3 but then overlaps within photoreceptor precursors. In humans, Nrl mutations are identified with autosomal-dominant retinitis pigmentosa (Mears et al., 2001; Bessant et al., 2003), whereas mutations in Nr2e3 lead to autosomal recessive retinopathy, “Enhanced S-cone Syndrome” (ESCS), characterized by impairment of both rods and L/M cones with concomitant increase in sensitivity of S-cones (Akhmedov et al., 2000; Haider et al., 2000; Haider et al., 2001; Milam et al., 2002; Jacobson et al., 2004).

Taken together, these observations suggest that Nrl and Nr2e3 may act as coregulators of rod development. In this study, we examined the developmental expression of a xNr2e3 and determined in vivo functions of Xenopus and mammalian Nrl and Nr2e3. We found that overexpression of each individual transcription factor in the developing Xenopus retina caused a significant increase in the number of rods at the expense of cones. Furthermore, overexpression of both transcription factors within the same cell resulted in an additional increase in rod number. Establishing the functional conservation between Xenopus and human Nrl-Nr2e3 provides numerous advantages for further mechanistic studies of development and disease through broadened experimental approaches. For example, Xenopus is ideal for studies of transcriptional regulation in developing retina due to rapid eye formation, abundance of cones (similar to humans; Grant et al., 1980), and relative ease in introduction of foreign genes into embryos by lipofection and transgenesis. Furthermore, transgenic approaches in Xenopus have proven to be powerful in the analysis of promoters (Knox et al., 1998; Mani et al., 1999, 2001; Zhu et al., 2002).


  1. Top of page
  2. Abstract
  6. Acknowledgements

Nr2e3 Gene Expression and Function Is Conserved in Tetrapods

To identify the Nr2e3 gene in Xenopus, we searched Xenopus tropicalis genome (,v.3.0) using the coding region of human Nr2e3 (accession no. AF121129). We identified a scaffold with long stretches of sequence homology. Gene-specific primers were designed to the most homologous regions and used in rapid amplification of cDNA ends (RACE) to identify exons 2–8 and the 3′-untranslated region (UTR) of the Xenopus laevis Nr2e3 gene. Unfortunately, strong RNA secondary structure prevented cloning of exon 1 (which ranges from 40 amino acids in humans and 27 amino acids in chicken) and the 5′-UTR despite numerous attempts.

Comparison of Nr2e3 sequences in various vertebrate species revealed a high degree of interspecies homology (data not shown). Total nucleotide and amino acid sequence identities of Xenopus and human Nr2e3 are 66.2% and 70.5%, respectively. xNr2e3 contains putative DNA binding (DBD) and ligand binding (LBD) domains, which are highly homologous to human Nr2e3 DBD and LBD (Fig. 1A).

thumbnail image

Figure 1. Xenopus Nr2e3 gene is highly conserved with other vertebrates. A: Comparison of human Nr2e3 with Xenopus Nr2e3. The sequence identities of DNA binding domains (DBD) and ligand binding domains (LBD) were revealed by using DNAStar (Stratagene). Numbers in the bars show the percentage of amino acid (AA) sequence identity to hNr2e3. The numbers above the bar indicate the boundaries of different regions based on AA residue position. Xenopus AA residue positions are relative due to the fact that exon 1 is missing. The dashed line represents the unidentified sequence. B: Phylogenetic relationship of Nr2e3 and Nr2e1 family members. Maximum parsimony algorithm with bootstraping was used to determine relationships. Species abbreviations are as follows: hs, Homo sapiens; mm, Mus musculus; xl, Xenopus laevis; gg, Gallus gallus. C: Reverse transcriptase-polymerase chain reaction (RT-PCR) of adult Xenopus heads (stages 20, 24, 36), eyes (stage 42). and retina (adult) RNA using xOtx2, xOtx5b, xL-Nrl, xNr2e3, xRho, and H4 (loading control) primers. D: Timing of Xenopus retinogenesis. E: In situ hybridizations of Xenopus laevis retina at different stages with xNr2e3 probe. Staining can be seen in the outer nuclear layer of the photoreceptors only when antisense probe is used. The negative control (using sense probe) does not produce any staining. Scale bar is 50 μm. F,G: Embryos at stage 17–18 were lipofected in eye anlagen with either CMV-empty vector (F) or CMV-hNr2e3 together with CMV-GFP (G) and immunostained with cone antibody (anti-calbindin) and DAPI. Cones appear red (untransfected) or yellow (transfected), while rods always appear green (F,G lower panels). The fates of lipofected cells were determined and quantified as the average percentage of each cell type. H: Student's t-test was used to determine significance. Asterisks denote P < 0.05. GC, ganglion; Am&BP, amacrine and bipolar, H, horizontal; R, rod; C, cone; Mu, Müller glia; RPE, retina pigment epithelia; OS, outer segment; ONL, outer nuclear layer; INL inner nuclear layer; IPL inner plexiform layer; GCL, ganglion cell layer.

Download figure to PowerPoint

To verify that this newly identified sequence belongs to the Nr2e3 family and not to other related families, we performed a phylogenetic analysis of the putative Xenopus Nr2e3 gene, various vertebrate Nr2e3 orthologs and their closest family members, Nr2e1 (Fig. 1B). The tree was obtained using the maximum parsimony algorithm (PAUP 4.0). There were 437 total characters, 139 characters were constant, 55 were variable and parsimony uninformative, 243 characters were parsimony informative. The most parsimonious tree had parameters: length = 511, consistency = 0.9687, and retention index = 9308. Two hundred bootstrapping replicates were completed and frequencies of occurrence (bootstrap values) are shown on each bipartition of the tree (Fig. 1B). The data revealed that the newly isolated Xenopus gene always grouped with other Nr2e3 family members (supported by 100% of bootstrap values) and never with Nr2e1 family members. Furthermore, BLAST records of the Xenopus Nr2e3 gene against the entire human genome returned Nr2e3 as the best hit for a homologous sequence. Ensembl also grouped xNr2e3 with its orthologs in different species. These results indicate that xNr2e3 is the amphibian counterpart of the mammalian Nr2e3.

Next, RT-PCR was performed to determine the temporal expression of Nr2e3 in relationship to other transcription factors involved in photoreceptor development (Fig. 1C). Total RNA isolated from Xenopus laevis heads or retinas (see the Experimental Procedures section) at various stages was reverse transcribed with random hexamers and the resulting cDNA was amplified in a polymerase chain reaction (PCR) reaction using primers for xOtx2, xOtx5b, xL-Nrl, xNr2e3, xRho. Primers for histone (H4) were used as a loading control (Table 1). RT-PCR revealed that Nr2e3 is expressed in postmitotic cells during and after retinal differentiation. In Xenopus, retinal cells are born within a very short period (Fig. 1D). First postmitotic cells appear at stage 24, more than 90% of photoreceptors are born by stage 34, and retinogenesis is largely complete by stage 38, when more than 95% of cells are postmitotic (Holt et al., 1988). One interesting difference between retinogenesis of Xenopus and higher vertebrates is that Xenopus rods develop slightly before inner nuclear layer neurons (Holt et al., 1988), while mammalian rods develop last. To precisely determine spatiotemporal Nr2e3 expression in Xenopus retina, we performed in situ hybridization (Fig. 1E). Nr2e3 expression was first observed at stage 34 exclusively in outer nuclear layer, similar to other vertebrates.

Table 1. Genes and Primers Used for Reverse Transcriptase-Polymerase Chain Reaction
GeneForward primerReverse primerAnnealing temp

Spatiotemporal expression pattern of Nr2e3 suggests that it mainly acts on postmitotic photoreceptor precursors. However, the role of Nr2e3 in Xenopus retinogenesis has not been examined. Furthermore, the possibility that Nr2e3 is sufficient to direct other retinal progenitors to become rods has not been explored. In this study, we tested these hypotheses by expressing Nr2e3 in different types of retinal progenitors in a cell autonomous manner. Because we were not able to isolate exon1 of xNr2e3, we used human Nr2e3. Stage 17–18 (neural fold stage) retinoblasts were lipofected in vivo with hNr2e3 DNA expression constructs under the control of a CMV promoter. The Nr2e3 gene was introduced into the retinal primordia, rather than postmitotic progenitors to allow sufficient time for transcription, translation, and posttranslational modifications to occur. Xenopus retinogenesis is extremely rapid, approximately 25 hr (Fig. 1D), so that the proteins are likely to be produced at the time when cells begin to differentiate. GFP DNA was co-injected with experimental DNA to identify the lipofected cells. GFP DNA co-injected with empty vector was used as a control. At stage 42, live tadpoles were examined for GFP expression under the fluorescent microscope. GFP-positive tadpoles were fixed, cryosectioned, immunostained and analyzed for retinal cell fates. Rods and cones were identified using immunohistochemistry with a cone marker anti-calbindin (Fig. 1F,G). Our conclusions about retinal cell fates were based on statistical analysis of at least 90 sections per condition. This large number of sections is necessary because each lipofected eye exhibits significant fluctuations of cell types from section to section. Nr2e3 significantly increased the proportion of GFP-positive rods (18%, n = 1,836 cells, 13 embryos; P < 0.003) compared with empty vector controls (10%, n = 1,602 cells, 9 embryos) at the expense of cones (12% empty vector, to 3% Nr2e3; Fig. 1H). It appears that in Xenopus, Nr2e3 is involved in directing photoreceptor precursors away from cone and toward rod cell fates.

It is possible that calbindin/GFP-positive cells could be “hybrid” photoreceptors, expressing both rod and cone genes within the same cell. To reliably identify rod–cone hybrid cells, several anti-rod and anti-cone antibodies should be used. Unfortunately, “hybrid” theory is difficult to test at this time due to the lack of anti-rod and anti-cone antibodies for Xenopus.

Together, our data suggest that Nr2e3 acts on photoreceptor precursors directing them to become rods. Furthermore, Nr2e3 overexpression in nonphotoreceptor progenitors appears insufficient to override their ultimate fate. A possible explanation could be that cells ectopically expressing Nr2e3 are determined before production of Nr2e3 protein and, therefore, restricted in their ability to respond to it. Alternatively, Nr2e3 may need the presence of other transcription factors such as Nrl and Otx5b to override nonphotoreceptor fated cells (see below).

Overexpression of xL-Nrl but Not hNrl in Developing Xenopus Eye Primordia Promotes Lens Fiber Cell Differentiation

The role of xL-Nrl in retinogenesis has not been examined. Previous studies have focused on understanding whether xL-Nrl is involved in lens formation. It has been shown that xL-Nrl is expressed in lens fiber cells during lens vesicle formation and that it can induce lens-specific markers such as crystallins in animal cap assays (Ishibashi and Yasuda, 2001). Interestingly, the mammalian Nrl gene has been shown to be involved in photoreceptor development (Mears et al., 2001). This involvement has been a cause for some initial confusion regarding the identity of the Xenopus ortholog of Nrl. xL-Nrl was originally named xL-Maf, because of its high homology to the chicken L-Maf amino acid sequence (91%) and its expression in the lens placode and later in lens fiber cells (Ogino and Yasuda, 1998). More stringent phylogenetic analysis as well as experimental data from several later studies showed that xL-Maf is in fact the Xenopus homolog of Nrl (Whitaker and Knox, 2004; Coolen et al., 2005) with functions in both the lens and retina.

To examine the role of xL-Nrl in retinogenesis and compare its function with the hNrl ortholog, we overexpressed each of these genes by DNA transfection similar to the above assay (Fig. 2). Overexpression of either hNrl or xL-Nrl promoted formation of rod cells at the expense of cones, similar to that seen with Nr2e3 (Fig. 2A–C). Lipofection of empty vector (n = 1,158 cells from 13 eyes) produced 10% rods, 15% cones, and 50% amacrine and bipolar cells. xL-Nrl (n = 1,218 cells from 20 eyes) increased the number of GFP-positive rods to 31% and decreased the number of GFP-positive cones to 4%. Furthermore, xL-Nrl seemed to significantly decrease the population of lipofected amacrine and bipolar cells to 24%. A similar switch was observed with hNrl (n = 749 cells from 9 eyes), where GFP-positive rods increased to a significant but more modest 16%, GFP-positive cone number decreased to 3%.

thumbnail image

Figure 2. A,B: Cells overexpressing hNrl or xL-Nrl differentiate into rods. Representative retinas lipofected with xL-Nrl (A) and hNrl (B) immunostained with anti-calbindin (cones) and 4′,6-diamidine-2-phenylidole-dihydrochloride (DAPI). C: The fates of lipofected cells were determined and quantified as the average percentage of each cell type. Analysis of variance (analysis of variance [ANOVA]), single-factor, was used to determine significance. Asterisks denote P < 0.05. GC, ganglion; Am&BP, amacrine and bipolar, H, horizontal; R, rod; C, cone; Mu, Müller glia. D,E: Many of the xL-Nrl lipofected tadpoles exhibited increased numbers of GFP-positive lens fiber cells compared with lipofections with empty vector or hNrl. Quantitative analysis (ANOVA, single-factor) revealed that the increase of GFP-positive lens fiber cells in xL-Nrl lipofected retinas is statistically significant (P < 0.01). G,H: While anti-rhodopsin antibody stained exclusively rods in empty vector lipofected animals (G), many of the xL-Nrl lipofected retinas exhibited rhodopsin expression in nonphotoreceptor cells (H, arrows).

Download figure to PowerPoint

The affect on amacrine/bipolar cells by xL-Nrl is somewhat surprising, however, there are several possible explanations for this result. xL-Nrl is expressed in dividing lens precursors as early as stage 24 (Ishibashi and Yasuda, 2001), while hNrl is found exclusively in postmitotic cells. This earlier expression of xL-Nrl, together with its additional function in lens formation (see below) suggests that xL-Nrl has additional functions in dividing cells that hNrl does not. In addition, several studies suggest that bipolar and photoreceptor cells may arise from the same progenitor (Altshuler and Cepko, 1992; Ezzeddine et al., 1997; Belliveau et al., 2000; Viczian et al., 2003). Thus, it is possible that the early presence of xL-Nrl in bipolar/photoreceptor progenitors may bias them to rod cell fate.

In addition to the cone-to-rod cell fate switch, we also noticed an increased number of GFP-labeled lens cells in the xL-Nrl lipofected animals (Fig. 2D,E). The lens cells in all experimental conditions were counted and plotted as a percentage of total lipofected cells. We clearly found that xL-Nrl significantly increased the number of GFP-positive lens fiber cells compared with empty vector controls or hNrl (Fig. 2F). This result supports the previously reported role of xL-Nrl in lens development (reviewed in Reza and Yasuda, 2004). There are at least two possible explanations for the observed increase in lens cells in embryos transfected with xL-Nrl. First, xL-Nrl could promote the proliferation of relatively few transfected lens fiber precursors, thus increasing the number of GFP-positive lens cells. Second, xL-Nrl could reprogram retinal to lens progenitor cells. The former seems more likely, because the latter explanation would require xL-Nrl to reprogram neuroectoderm into head ectoderm, a transformation that appears to be unprecedented. The results also show that the transfected progenitor cells that contribute to the lens are responsive to the Xenopus but not human ortholog. We conclude that the Nrl gene has undergone evolutionary divergence. In lower vertebrates, such as Xenopus, xL-Nrl regulates several functions including lens fiber differentiation and photoreceptor development. In mammals, Nrl is involved exclusively in photoreceptor differentiation while other large Maf proteins, such as c-Maf regulate lens formation (Coolen et al., 2005). In future experiments, it will be interesting to examine which part of the xL-Nrl gene is involved in the regulation of lens development.

In additional, we observed that some (but not all) cells overexpressing hNrl and xL-Nrl had stimulated transcription of rhodopsin. These included ganglion, amacrine, and Müller glia. No cone-specific calbindin (tested with anti-calbindin antibody, data not shown) was observed. As expected, nonrod cells overexpressing empty vector never produced rhodopsin (Fig. 2G,H). These Nrl-transfected rhodopsin-positive cells did not develop outer segments characteristic to rods, but rather rhodopsin appeared to reside in cellular membranes. This finding suggests that Nrl is capable of transcriptionally activating the rhodopsin promoter in nonphotoreceptor retinal cells. Previously, we found that Nrl could activate rhodopsin expression in non-neural tissues (Whitaker and Knox, 2004).

Overexpression of Multiple Transcription Factors Within the Same Cell Is More Efficient in Biasing Progenitors to Rod Cell Fate

The overlapping expression patterns of Nr2e3 and Nrl suggest that these genes may work together to direct retinal progenitors to become rods. An additional transcription factor Otx5b, an ortholog to the mammalian Crx (Germot et al., 2001), is likely to be a part of this regulatory network. Crx is involved in photoreceptor differentiation (Furukawa et al., 1997; Viczian et al., 2003) and has an overlapping expression pattern with Nr2e3 and Nrl during retinogenesis (Fig. 1C; Chen et al., 1997; Haider et al., 2001; Yoshida et al., 2004). Furthermore, a recent study has shown that Crx, Nrl, and Nr2e3 co-occupy promoters of photoreceptor-specific genes and that Nr2e3 binding to DNA is Crx-dependent (Peng and Chen, 2005). Therefore, we hypothesized that rod formation may be fine-tuned through the use of multiple transcription factors that direct the formation of rod photoreceptors. To study this further, we tested various combinations of these three transcription factors in lipofections (Fig. 3). Xenopus genes are potentially more effective than human orthologs in directing rod formation in Xenopus. This idea is partially supported by our data where xL-Nrl caused a larger increase of GFP-positive rods compared with hNrl. Thus, we used Xenopus genes whenever possible in the lipofection combinations to achieve the maximum switch to rods. We also tested the hNrl and hNr2e3 pair to compare their combined and individual effects on Xenopus retinogenesis. Overexpression of Otx5b alone (n = 1,136 cells, 8 eyes) significantly increased the population of GFP-positive rods (19%) and cones (24%; Fig. 3), confirming an earlier study (Viczian et al., 2003). To examine transcription factors in pairs, we lipofected hNr2e3+hNrl or Otx5b+xL-Nrl together into the developing retina. Retinas cotransfected with hNr2e3+hNrl (n = 783 cells, 8 eyes) had a significantly larger population of GFP-positive rods (32%) compared with cells lipofected with empty vector (10%), hNr2e3 (18%), hNrl (19%), or Otx5b (19%). The increase in rods came primarily at the expense of cones (47% decrease) and from a small but significant decrease in ganglion cells (18%). Our data suggest that human orthologs of Nrl and Nr2e3 work cooperatively, rather than alone in directing photoreceptor progenitors to become rods in Xenopus. Previous findings are consistent with this idea. For example, in Nr2e3rd7 mouse, the levels of Crx and Nrl are unaffected (Corbo and Cepko, 2005); however, photoreceptors in these mice express both rod and cone genes. Furthermore, ectopic expression of Nr2e3 in Nrl knockout mice also fails to produce functional rods (Cheng et al., 2006). In the case of Xenopus Nrl and Nr2e3 genes, the question remains as to whether they work cooperatively in directing rod development. This question can be addressed only after exon 1 of Nr2e3 is successfully isolated.

thumbnail image

Figure 3. Overexpressing multiple transcription factors within the same cell. AD: Representative retinas lipofected with Otx5b (A), Otx5b+xL-Nrl (B), hNrl+hNr2e3 (C), Otx5b+xL-Nrl+hNr2e3 (D) immunostained with anti-calbindin (cones) and 4′,6-diamidine-2-phenylidole-dihydrochloride (DAPI). E: The fates of lipofected cells were determined and quantified as the average percentage of each cell type. F: Summary of Tukey post hoc multiple comparison test. Statistically significant differences (P < 0.05) in cell numbers are shown in green (increase) or red (decrease). White blocks appear when there is no statistical difference.

Download figure to PowerPoint

Having all three transcription factors together did not promote further reprogramming. Cotransfection of Otx5b, xL-Nrl, and hNr2e3 (n = 795 cells, 6 eyes) increased the population of GFP-positive rods (34%); however, this change was not significantly different from that caused by cotransfection of hNr2e3+hNrl, Otx5b+xL-Nrl (Fig. 3), or xL-Nrl alone (Fig. 2). One potential explanation for this outcome is that ectopic expression of one transcription factor activates expression of its partner(s). Indirect support for this can be found in Nrl knockout mice, which do not express Nr2e3 transcripts (Mears et al., 2001). Thus, it is possible that transfection of xL-Nrl increases xNr2e3 levels, leading to similar levels of rod production regardless of whether other transcription factors are cotransfected. In the case of human orthologs, it is possible that hNrl is not as effective as xL-Nrl in activating Nr2e3 expression in Xenopus. Thus, cotransfecting hNr2e3 and hNrl would result in additive photoreceptor production. Finally, we found that not all retinal progenitors were responsive to these transcription factors, suggesting that Nrl and Nr2e3 alone, together, or in combination with Otx5b are not sufficient for rod formation. Thus, additional transcription factors, upstream of Nrl/Nr2e3, appear necessary for rod formation in Xenopus.

In summary, our data suggest that Nrl and Nr2e3 regulate photoreceptor development by activating rod and repressing cone genes in photoreceptor progenitors. These results are in agreement with previous in vitro studies (Peng et al., 2005) but differ somewhat from a recent publication that showed completely suppressed cone differentiation following ectopic expression of Nr2e3 in the Nrl−/− mouse retina (Cheng et al., 2006). In our experiments, 30–40% of cones were not affected by the presence of Nrl, Nr2e3, or both. There are several possible explanations for this difference. First, it could be a result of the limitations of the cotransfection technique: incomplete cotransfections, insufficient levels of expression, or different timing of the ectopic expression. Second, Nrl and Nr2e3 may only influence a subpopulation of photoreceptor progenitors. Third, there might be divergence in mechanisms of photoreceptor development between mammals and Xenopus. Finally, an additional coregulator or transcription factor upstream of Nrl and Nr2e3 could be necessary for a complete shut off of cone genes or turning on of rod genes. Therefore, the identification of such transcriptional regulators could be valuable for improving our understanding of retinogenesis as well as differences in developmental mechanisms between various species.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Cloning of Xenopus Nr2e3

Total RNA was isolated from adult Xenopus laevis retina with Trizol (Invitrogen). The 3′-RACE and 5′-RACE were performed using BD Smart RACE cDNA Amplification Kit (BD Biosciences) following the manufacturer's protocol. The scaffold sequence containing the Nr2e3 gene, which was found in Scaffold_637 in the Xenopus tropicalis genome site, was used to design gene-specific primers for RACE.

RNA Isolation and RT-PCR Assay

Total RNA isolated from heads of Xenopus embryos at stages 20, 24, 36, eyes at stage 42, and adult retinas was reverse transcribed with random hexamers and SuperScriptII polymerase (Invitrogen, Carlsbad, CA) and treated with DNAse (Promega, Madison, WI) following manufacturer's protocols. The resulting cDNA was amplified in a PCR reaction with primers designed to amplify different transcription factors (Table 1). PCR parameters were as follows: 5 min at 95°C to denature, then 35 cycles at 30 sec at 95°C, 30 sec 51–67°C, 1 min 72°C, and final extension of 5 min at 72°C.

Embryo Generation

Xenopus laevis eggs were obtained from hormone-induced adult frogs and then in vitro fertilized using standard methods. The treatment of all animals was in strict accordance with both Institutional Animal Care and Use Committee guidelines and those advocated by the National Institutes of Health.

In Vivo Lipofection of DNA

This protocol has been previously described (Viczian et al., 2003). Briefly, DNA was isolated by Qiagen maxi preps and diluted in nuclease-free water to a concentration of 1.5 mg/ml. These stocks were spun down at 14,000 × g for 10 min at 4°C before use. A total of 1 μl of each construct was mixed with 1 μl of pCS2+ green fluorescent protein (GFP) DNA to label transfected cells. GFP with pCS2+empty vector was used as a control. A total of 9 μl of DOTAP (Roche) was added to 3 μg of DNA and injected into eye primordia of stage 17–18 embryos. To normalize the amount of DNA introduced in double and triple lipofections, pCS2+empty vector was colipofected with experimental construct(s). Cotransfection efficiencies for double lipofection with similar experimental design have been shown to be >90% by several groups (Holt et al., 1990; Ohnuma et al., 2002a). If more than 90% of cells are cotransfected with any two plasmids, then the probability of the cell containing three plasmids in triple lipofection is ({9)3 = 0.85 (or 85%) and of the cell containing four plasmids during quadruple lipofection is ({9)4 = 0.81 (or 81%), assuming that the events are independent. Furthermore, numerous publications of successful double, triple, and quadruple lipofection studies in Xenopus (performed and analyzed at a similar stages) were reported previously (Moore et al., 2002; Ohnuma et al., 2002b; Viczian et al., 2003; Wang and Harris, 2005).

Fixation and Retinal Cell Fate Analysis

At stage 42, embryos were fixed for 1 hr at 4°C in 4% paraformaldehyde in phosphate buffer, cryoprotected in 20% sucrose overnight at 4°C, embedded in OCT compound (Tissue Tek), and cryostat sectioned (10 μm). Nuclei were stained using 4′,6-diamidine-2-phenylidole-dihydrochloride (DAPI). All GFP-positive cells from lens-containing sections were identified and counted. Peripheral retina was excluded from analysis because the individual cell layers in these sections were not discernible, making morphological analysis impossible. Identification of ganglion, amacrine, bipolar, horizontal cells, and Müller glia was based on the relative laminar position and morphology. Potentially, we may have inadvertently counted displaced amacrine cells as ganglion cells. However, these cells are not very abundant. Rods and cones were identified by antibody staining (either 4D2 [rod antibody] or anti-calbindin [cone antibody]). Several measures were taken to ensure that a GFP-positive cell was counted only once. First, we selected only weakly transfected retinas (less than 20% of total cells in the section were GFP-positive), so that the transfected cells were far enough apart from each other to ensure accurate assessment of morphology. Second, at stage 42, Xenopus retinal cells are approximately 10 μm in diameter, thus cutting at this thickness reduced the probability of the same cell being in more than one section. During photoreceptor type identification, a colocalization was determined by a high-magnification differential interference contrast microscopy image to confirm that there is only one outer segment within the z-plane of the section and overlay of red and green channels in high magnification. The contours of truly colocalized staining always overlap exactly. Finally, if axons and dendrite processes of the GFP-positive cells could not be seen in the section, the adjacent sections were checked to confirm the cell type and to prevent redundancy in counting.

The same randomly chosen eyes were used for assessment of lens phenotypes and for retinal cell fate, regardless of whether the lenses were GFP-positive or GFP-negative. For retinal cell fate analysis, GFP-positive lens cells were discarded. For lens phenotype analysis, GFP-positive cells in retina and lens were combined and differences in the in the lens were determined by plotting the GFP-positive lens cells as a percentage of total lipofected cells. The results were analyzed using either two-tailed Student's t-test, assuming equal variance, analysis of variance, or Tukey post hoc test and presented as mean ± SEM. There were three data sets, each shown in Figures 1–3. We were also able to compare all three data sets for the following reasons. First, each condition was repeated multiple days using different batches of embryos. Second, each experimental construct was always lipofected alongside empty vector control using the same batch of embryos. The statistical comparison of empty vector in all three data sets yielded insignificant differences in cell fates (small error bars with no statistical significance between trials). Finally, we only included in our analysis those batches of tadpoles that exhibited similar transfection efficiencies.


Cryostat sections (10 mm) were washed three times for 5 min in PBS + 0.1% Triton X-100 (PBST), blocked for 1 hr with PBST containing 5% heat-inactivated goat serum, then incubated overnight at 4° with primary antibody: monoclonal anti-rhodopsin (4D2) at 1:10,000 or anti-calbindin (Oncogene) 1:500. Sections were washed in blocking solution 3× for 5 min and incubated with a 1:500 dilution of the appropriate Cy3-conjugated secondary antibody for 2 hr. The sections were washed again in PBST, mounted in FluorSave (CalBioChem) containing 2% DABCO (Sigma), and dried overnight at room temperature.

DNA Construct Generation

Constructs pCS2.xOtx5b, pCS2.xL-Nrl, pCS2.Nrl, and pCS2.GFP were generated previously (Whitaker and Knox, 2004). pCS2.hNr2e3 was obtained by cloning the coding region of hNr2e3 contained in the EcoRI/XhoI fragment of PcDNA4c.hNr2e3 (a kind gift of A. Swaroop) into the same restriction sites in the pCS2 vector.

In Situ Hybridization

In situ hybridization of frozen sections was performed as described previously (Viczian et al., 2003). The 465-bp antisense and sense digoxigenin-labeled RNA probes corresponding to the 3′-UTR and part of ligand binding region of Nr2e3 were used. BM purple was used to detect an alkaline peroxidase-conjugated digoxigenin antibody.


  1. Top of page
  2. Abstract
  6. Acknowledgements

We thank Dr. Andrea Viczian for teaching us the in vivo targeted transfection technique as well as for stimulating discussions and suggestions, Dr. Zuber for valuable comments on the manuscript, and Dr. Swaroop for kind gift of PcDNA4c.hNr2e3. We also thank Amanda Leskovar for excellent technical support. This research was supported by National Eye Institute at NIH, Research to Prevent Blindness unrestricted grant to SUNY UMU Department of Ophthalmology and Lions of Central New York.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  • Akhmedov NB, Piriev NI, Chang B, Rapoport AL, Hawes NL, Nishina PM, Nusinowitz S, Heckenlively JR, Roderick TH, Kozak CA, Danciger M, Davisson MT, Farber DB. 2000. A deletion in a photoreceptor-specific nuclear receptor mRNA causes retinal degeneration in the rd7 mouse. Proc Natl Acad Sci U S A 97: 55515556.
  • Akimoto M, Cheng H, Zhu D, Brzezinski JA, Khanna R, Filippova E, Oh EC, Jing Y, Linares JL, Brooks M, Zareparsi S, Mears AJ, Hero A, Glaser T, Swaroop A. 2006. Targeting of GFP to newborn rods by Nrl promoter and temporal expression profiling of flow-sorted photoreceptors. Proc Natl Acad Sci U S A 103: 38903895.
  • Altshuler D, Cepko C. 1992. A temporally regulated, diffusible activity is required for rod photoreceptor development in vitro. Development 114: 947957.
  • Belliveau MJ, Young TL, Cepko CL. 2000. Late retinal progenitor cells show intrinsic limitations in the production of cell types and the kinetics of opsin synthesis. J Neurosci 20: 22472254.
  • Bessant DA, Holder GE, Fitzke FW, Payne AM, Bhattacharya SS, Bird AC. 2003. Phenotype of retinitis pigmentosa associated with the Ser50Thr mutation in the NRL gene. Arch Ophthalmol 121: 793802.
  • Cepko CL. 1999. The roles of intrinsic and extrinsic cues and bHLH genes in the determination of retinal cell fates. Curr Opin Neurobiol 9: 3746.
  • Cepko CL, Austin CP, Yang X, Alexiades M, Ezzeddine D. 1996. Cell fate determination in the vertebrate retina. Proc Natl Acad Sci U S A 93: 589595.
  • Chen S, Wang QL, Nie Z, Sun H, Lennon G, Copeland NG, Gilbert DJ, Jenkins NA, Zack DJ. 1997. Crx, a novel Otx-like paired-homeodomain protein, binds to and transactivates photoreceptor cell-specific genes. Neuron 19: 10171030.
  • Cheng H, Khanna H, Oh EC, Hicks D, Mitton KP, Swaroop A. 2004. Photoreceptor specific nuclear receptor NR2E3 functions as a transcriptional activator in rod photoreceptors. Hum Mol Genet 13: 15631575.
  • Chen J, Rattner A, Nathans J. 2005. The rod photoreceptor-specific nuclear receptor Nr2e3 represses transcription of multiple cone-specific genes. J Neurosci 25: 118129.
  • Cheng H, Aleman TS, Cideciyan AV, Khanna R, Jacobson SG, Swaroop A. 2006. In vivo function of the orphan nuclear receptor NR2E3 in establishing photoreceptor identity during mammalian retinal development. Hum Mol Genet 15: 25882602.
  • Cook T. 2003. Cell diversity in the retina: more than meets the eye. Bioessays 25: 921925.
  • Coolen M, Sii-Felice K, Bronchain O, Mazabraud A, Bourrat F, Retaux S, Felder-Schmittbuhl MP, Mazan S, Plouhinec JL. 2005. Phylogenomic analysis and expression patterns of large Maf genes in Xenopus tropicalis provide new insights into the functional evolution of the gene family in osteichthyans. Dev Genes Evol 215: 327339.
  • Corbo JC, Cepko CL. 2005. A hybrid photoreceptor expressing both rod and cone genes in a mouse model of enhanced S-cone syndrome. PLoS Genet 1: e11.
  • Dowling J. 1987. The retina: an approachable part of the brain. Cambridge, MA: Belknap Press. 282 p.
  • Dyer MA, Cepko CL. 2001. p27Kip1 and p57Kip2 regulate proliferation in distinct retinal progenitor cell populations. J Neurosci 21: 42594271.
  • Ezzeddine ZD, Yang X, DeChiara T, Yancopoulos G, Cepko CL. 1997. Postmitotic cells fated to become rod photoreceptors can be respecified by CNTF treatment of the retina. Development 124: 10551067.
  • Furukawa T, Morrow EM, Cepko CL. 1997. Crx, a novel otx-like homeobox gene, shows photoreceptor-specific expression and regulates photoreceptor differentiation. Cell 91: 531541.
  • Furukawa T, Morrow EM, Li T, Davis FC, Cepko CL. 1999. Retinopathy and attenuated circadian entrainment in Crx-deficient mice. Nat Genet 23: 466470.
  • Germot A, Lecointre G, Plouhinec JL, Le Mentec C, Girardot F, Mazan S. 2001. Structural evolution of Otx genes in craniates. Mol Biol Evol 18: 16681678.
  • Grant P, Rubin E, Cima C. 1980. Ontogeny of the retina and optic nerve in Xenopus laevis. I. Stages in the early development of the retina. J Comp Neurol 189: 593613.
  • Haider NB, Jacobson SG, Cideciyan AV, Swiderski R, Streb LM, Searby C, Beck G, Hockey R, Hanna DB, Gorman S, Duhl D, Carmi R, Bennett J, Weleber RG, Fishman GA, Wright AF, Stone EM, Sheffield VC. 2000. Mutation of a nuclear receptor gene, NR2E3, causes enhanced S cone syndrome, a disorder of retinal cell fate. Nat Genet 24: 127131.
  • Haider NB, Naggert JK, Nishina PM. 2001. Excess cone cell proliferation due to lack of a functional NR2E3 causes retinal dysplasia and degeneration in rd7/rd7 mice. Hum Mol Genet 10: 16191626.
  • Hisatomi O, Tokunaga F. 2002. Molecular evolution of proteins involved in vertebrate phototransduction. Comp Biochem Physiol B Biochem Mol Biol 133: 509522.
  • Holt CE, Bertsch TW, Ellis HM, Harris WA. 1988. Cellular determination in the Xenopus retina is independent of lineage and birth date. Neuron 1: 1526.
  • Holt CE, Garlick N, Cornel E. 1990. Lipofection of cDNAs in the embryonic vertebrate central nervous system. Neuron 4: 203214.
  • Ishibashi S, Yasuda K. 2001. Distinct roles of maf genes during Xenopus lens development. Mech Dev 101: 155166.
  • Jacobson SG, Sumaroka A, Aleman TS, Cideciyan AV, Schwartz SB, Roman AJ, McInnes RR, Sheffield VC, Stone EM, Swaroop A, Wright AF. 2004. Nuclear receptor NR2E3 gene mutations distort human retinal laminar architecture and cause an unusual degeneration. Hum Mol Genet 13: 18931902.
  • Knox BE, Schlueter C, Sanger BM, Green CB, Besharse JC. 1998. Transgene expression in Xenopus rods. FEBS Lett 423: 117121.
  • Kobayashi M, Takezawa S, Hara K, Yu RT, Umesono Y, Agata K, Taniwaki M, Yasuda K, Umesono K. 1999. Identification of a photoreceptor cell-specific nuclear receptor. Proc Natl Acad Sci U S A 96: 48144819.
  • Lerner LE, Gribanova YE, Ji M, Knox BE, Farber DB. 2001. Nrl and Sp nuclear proteins mediate transcription of rod-specific cGMP-phosphodiesterase beta-subunit gene: involvement of multiple response elements. J Biol Chem 276: 3499935007.
  • Livesey FJ, Cepko CL. 2001. Vertebrate neural cell-fate determination: lessons from the retina. Nat Rev Neurosci 2: 109118.
  • Mani SS, Besharse JC, Knox BE. 1999. Immediate upstream sequence of arrestin directs rod-specific expression in Xenopus. J Biol Chem 274: 1559015597.
  • Mani SS, Batni S, Whitaker L, Chen S, Engbretson G, Knox BE. 2001. Xenopus rhodopsin promoter. Identification of immediate upstream sequences necessary for high level, rod-specific transcription. J Biol Chem 276: 3655736565.
  • Mears AJ, Kondo M, Swain PK, Takada Y, Bush RA, Saunders TL, Sieving PA, Swaroop A. 2001. Nrl is required for rod photoreceptor development. Nat Genet 29: 447452.
  • Milam AH, Rose L, Cideciyan AV, Barakat MR, Tang WX, Gupta N, Aleman TS, Wright AF, Stone EM, Sheffield VC, Jacobson SG. 2002. The nuclear receptor NR2E3 plays a role in human retinal photoreceptor differentiation and degeneration. Proc Natl Acad Sci U S A 99: 473478.
  • Moore KB, Schneider ML, Vetter ML. 2002. Posttranslational mechanisms control the timing of bHLH function and regulate retinal cell fate. Neuron 34: 183195.
  • Nikonov SS, Daniele LL, Zhu X, Craft CM, Swaroop A, Pugh EN Jr. 2005. Photoreceptors of Nrl−/− mice coexpress functional S- and M-cone opsins having distinct inactivation mechanisms. J Gen Physiol 125: 287304.
  • Ogino H, Yasuda K. 1998. Induction of lens differentiation by activation of a bZIP transcription factor, L-Maf. Science 280: 115118.
  • Ohnuma S, Hopper S, Wang KC, Philpott A, Harris WA. 2002a. Co-ordinating retinal histogenesis: early cell cycle exit enhances early cell fate determination in the Xenopus retina. Development 129: 24352446.
  • Ohnuma S, Mann F, Boy S, Perron M, Harris WA. 2002b. Lipofection strategy for the study of Xenopus retinal development. Methods 28: 411419.
  • Peng GH, Chen S. 2005. Chromatin immunoprecipitation identifies photoreceptor transcription factor targets in mouse models of retinal degeneration: new findings and challenges. Vis Neurosci 22: 575586.
  • Peng GH, Ahmad O, Ahmad F, Liu J, Chen S. 2005. The photoreceptor-specific nuclear receptor Nr2e3 interacts with Crx and exerts opposing effects on the transcription of rod versus cone genes. Hum Mol Genet 14: 747764.
  • Pittler SJ, Zhang Y, Chen S, Mears AJ, Zack DJ, Ren Z, Swain PK, Yao S, Swaroop A, White JB. 2004. Functional analysis of the rod photoreceptor cGMP phosphodiesterase alpha-subunit gene promoter: Nrl and Crx are required for full transcriptional activity. J Biol Chem 279: 1980019807.
  • Rehemtulla A, Warwar R, Kumar R, Ji X, Zack DJ, Swaroop A. 1996. The basic motif-leucine zipper transcription factor Nrl can positively regulate rhodopsin gene expression. Proc Natl Acad Sci U S A 93: 191195.
  • Reza HM, Yasuda K. 2004. Roles of Maf family proteins in lens development. Dev Dyn 229: 440448.
  • Sauka-Spengler T, Baratte B, Shi L, Mazan S. 2001. Structure and expression of an Otx5-related gene in the dogfish Scyliorhinus canicula: evidence for a conserved role of Otx5 and Crx genes in the specification of photoreceptors. Dev Genes Evol 211: 533544.
  • Swain PK, Hicks D, Mears AJ, Apel IJ, Smith JE, John SK, Hendrickson A, Milam AH, Swaroop A. 2001. Multiple phosphorylated isoforms of NRL are expressed in rod photoreceptors. J Biol Chem 276: 3682436830.
  • Viczian AS, Vignali R, Zuber ME, Barsacchi G, Harris WA. 2003. XOtx5b and XOtx2 regulate photoreceptor and bipolar fates in the Xenopus retina. Development 130: 12811294.
  • Wang JC, Harris WA. 2005. The role of combinational coding by homeodomain and bHLH transcription factors in retinal cell fate specification. Dev Biol 285: 101115.
  • Whitaker SL, Knox BE. 2004. Conserved transcriptional activators of the Xenopus rhodopsin gene. J Biol Chem 279: 4901049018.
  • Yanagi Y, Takezawa S, Kato S. 2002. Distinct functions of photoreceptor cell-specific nuclear receptor, thyroid hormone receptor beta2 and CRX in one photoreceptor development. Invest Ophthalmol Vis Sci 43: 34893494.
  • Yoshida S, Mears AJ, Friedman JS, Carter T, He S, Oh E, Jing Y, Farjo R, Fleury G, Barlow C, Hero AO, Swaroop A. 2004. Expression profiling of the developing and mature Nrl−/− mouse retina: identification of retinal disease candidates and transcriptional regulatory targets of Nrl. Hum Mol Genet 13: 14871503.
  • Zhu X, Ma B, Babu S, Murage J, Knox BE, Craft CM. 2002. Mouse cone arrestin gene characterization: promoter targets expression to cone photoreceptors. FEBS Lett 524: 116122.