|Home | About | Journals | Submit | Contact Us | Français|
The retinal transcription factor Nr2e3 plays a key role in photoreceptor development and function. In this study we examine gene expression in the retina of Nr2e3rd7/rd7 mutants with respect to wild-type control mice, to identify genes that are misregulated and hence potentially function in the Nr2e3 transcriptional network. Quantitative candidate gene real time PCR and subtractive hybridization approaches were used to identify transcripts that were misregulated in Nr2e3rd7/rd7 mice. Chromatin immunoprecipitation assays were then used to determine which of the misregulated transcripts were direct targets of NR2E3. We identified 24 potential targets of NR2E3. In the developing retina, NR2E3 targets transcription factors such as Ror1, Rorg, and the nuclear hormone receptors Nr1d1 and Nr2c1. In the mature retina NR2E3 targets several genes including the rod specific gene Gnb1 and cone specific genes blue opsin, and two of the cone transducin subunits, Gnat2 and Gnb3. In addition, we identified 5 novel transcripts that are targeted by NR2E3. While mislocalization of proteins between rods and cones was not observed, we did observe diminished concentration of GNB1 protein in adult Nr2e3rd7/rd7 retinas. These studies identified novel transcriptional pathways that are potentially targeted by Nr2e3 in the retina and specifically demonstrate a novel role for NR2E3 in regulating genes involved in phototransduction.
Enhanced-S-Cone Syndrome (ESCS) is an autosomal recessive retinopathy caused by mutations in the orphan receptor NR2E3 (Haider et al., 2000; Nakamura et al., 2004). While most hereditary human retinal diseases affect the topography of mature photoreceptor cells through apoptotic loss of cells, ESCS manifests as a gain of function in S (short, blue) cone cells; the least populous of the photoreceptor cells (Haider et al., 2000). Patients also exhibit early night blindness, retinal tearing, neovascularization, and varying degrees of L (long, red) cone and M (middle, green) cone vision and of retinal degeneration (Jacobson et al., 1990; Marmor et al., 1990; Jacobson et al., 1991; Haider et al., 2000; Milam et al., 2002; Jacobson et al., 2004). Mutations within NR2E3 are also associated with a related retinal disease, Goldman Favre syndrome, clumped pigmentary retinopathy, and more recently in autosomal dominant retinitis pigmentosa as well as a milder form of ESCS (Favre, 1958; Fishman et al., 1976; Gerber et al., 2000; Sharon et al., 2003; Chavala et al., 2005; Hayashi et al., 2005; Coppieters et al., 2007; Lam et al., 2007; Gire et al., 2007).
The retinal degeneration 7 (rd7, Nr2e3rd7/rd7) mutant mouse which lacks NR2E3, has been shown to have an increase in blue opsin expression without a concomitant increase in green opsin or significant reduction in rhodopsin expression (Hawes et al., 1999; Akhmedov et al., 2000; Haider et al., 2001). Our recent studies show that the increase in blue opsin expressing cones of the Nr2e3rd7/rd7 mouse results from abnormal proliferation due to a defect in mitotic progenitors to suppress their cone generation program (Haider et al., 2006). Persistent ectopic progenitors are thus likely to lead to the gain of S-cone function observed in ESCS patients.
While earlier whole genome studies have identified genes that are differentially expressed in Nr2e3rd7/rd7 retinas compared to controls (Chen et al., 2005; Corbo and Cepko, 2005; Peng et al., 2005; Cheng et al., 2006) the transcriptional network through which NR2E3 functions still remains poorly understood. While Nr2e3 action alone does not have a profound effect on rod cell differentiation, in combination with Nrl, it promotes rod development (Hennig et al., 2008; McIlvain and Knox, 2007). Recent studies suggest that Nr2e3 functions as a dual transcriptional regulator and lack of Nr2e3 leads to hybrid photoreceptor cells that de-repress cone genes and reduce rod gene expression (Corbo and Cepko 2005; Cheng et al., 2006, Webber et al. 2008). In this study we identified differentially expressed genes by subtractive hybridization and performed quantitative real time PCR (qRT-PCR) to evaluate those as well as over 100 genes that have a putative or known function in the development or maintenance of the retina, or for which mutations are associated with human retinal disease. We further used chromatin immunoprecipitation to identify which of the differentially expressed genes Nr2e3 directly regulated. This study was performed at two postnatal time points: P2 and P21. P2 was chosen as a developmental time point during which Nr2e3 is expressed. We chose P21 to examine genes differentially expressed and targeted by Nr2e3 in the adult retina. Differential gene expression in the adult may occur as a secondary effect of retinal dysplasia apparent in the rd7 mouse at P21, however differentially expressed genes that are direct targets of NR2E3 likely contribute the adult rd7 phenotype. Through our comprehensive study, we identified genes potentially regulated by Nr2e3 such as the nuclear receptor Nr1d1, which was previously reported to be a co-factor of Nr2e3 (Cheng et al., 2004), the cone specific phototransduction genes blue opsin and the α and β cone transducin subunits Gnat2 and Gnb3 respectively, and the rod specific gene, guanine nucleotide binding protein 1 (Gnb1). While the expression level of these genes is altered in Nr2e3rd7/rd7 mice, at the message and protein level, they are not mislocalized to other cells. This suggests that NR2E3 functions by regulating genes in a temporal and cell-type specific manner in independent transcriptional networks in rod and cone cells.
The mice used in this study were bred and maintained under standard conditions in the research vivariums at The Jackson Laboratory and at the University of Nebraska Medical Center. Tissues were harvested from B6.Cg-Nr2e3rd7/rd7 and C57BL/6J (B6) at the following postnatal (P) 2, and P21. A minimum of three Nr2e3rd7/rd7 and three B6 mice were analyzed for each time point. Mice were genotyped for the Nr2e3rd7/rd7 mutation as previously described (Haider et al., 2001). Adult mice were phenotyped by indirect ophthalmoscopy for the characteristic Nr2e3rd7/rd7 panretinal spots prior to additional analysis.
Subtractive hybridization was performed using Clontech PCR-Select cDNA Subtraction kit (BD biosciences Clontech) per manufacturer’s instructions. Briefly, total RNA was isolated from P2 and P21 eyes of control and Nr2e3rd7/rd7 mice. Three pooled samples (n=10 eyes in each pool) were generated for each genotype and time point. cDNAs were synthesized from each sample and tester and driver cDNAs were digested with Rsa1 to generate blunt ended molecules. The tester cDNAs were then subdivided into two portions and each was ligated with a different cDNA adaptor. Two hybridizations were performed for each tester cDNA to normalize differentially expressed genes and then to generate PCR templates from differentially expressed sequences. PCR amplified products were cloned into Topo-XL and sequenced. All clones (approximately 200) were sequenced and analyzed using Sequencher (V. 4.6, GeneCodes, Ann Arbor, MI) and NCBI nucleotide BLAST.
Total RNA was isolated from seven P2 and P21 pairs of eyes from Nr2e3rd7/rd7 and control (C57BL/6J, B6) mice using Trizol. Two μg of total RNA was reverse transcribed in independent triplicates using Retroscript (Ambion). cDNA samples were diluted 1:100 and real time PCR was performed in triplicate for each primer using SYBR Green PCR Master Mix (Applied Biosystems). PCR products of approximately 200 nucleotides were amplified with primers selected using Primer Express 2.0 software program. Reactions were quantified using an ABI 7500 Real Time PCR instrument and analyzed with accompanying software. Relative expression levels were determined by normalizing cycle threshold values for each primer to the amount of β-Actin expressed (1000/2^(Ct-gene-Ct-:β-Actin). Relative fold change was calculated from normalized values. Significant expression differences were assessed by the student’s T-Test with a minimum p-value of 0.05. Supplemental Table 1 (S1) lists all primers used for real time PCR.
Chromatin immunoprecipitation (ChIP) was performed using reagents from the ChIP kit (Upstate, Millipore) and a modified protocol. Briefly, retinas were dissected from P2 or P21 C57BL/6J eyes (8–10 eyes per ChIP; three independent assays per time point) and placed into 400 μl of PI buffer (1 tablet protease inhibitor cocktail, Roche) in 10ml phosphate buffered saline (PBS) on ice. Chilled retinas were dissociated and crosslinked in 37% formaldehyde (1% final concentration) for 20 minutes at room temperature on a rotating platform. Retinas were incubated for an additional 5 minutes in 1M glycine (final concentration 0.125mM) and rinsed twice with PI buffer before lysis. Retinas were lysed on ice and sonicated on ice using a Misonix Sonicator 3000 (power 5, pulse 10×, 30 second pause between pulses; 150 total pulses). Retina lysates were pre-cleared using a salmon sperm DNA/protein A Agarose-50% slurry and immunoprecipitation was performed overnight with approximately one μg of NR2E3 antibody (Haider et al. 2006) at 4°C with rotation. Immunoprecipitated samples were eluted with 1%SDS, 0.1M NaHCO3 buffer and reverse cross-linked with NaCl (200 mM final concentration) and incubated with 10 mg RNAse A at 65°C for five hours. Input (control) samples were isolated similarly and incubated with pre-clear solution instead of antibody or with goat IgG antibody (negative control). PCR reactions were performed using 1 μl of 1:10 diluted Input or Experimental (NR2E3 precipitated) sample under standard PCR conditions as previously described (Haider et al., 2001) using an annealing temperature of 58°C and 35 cycles. PCR reactions were electrophoresed in a 2% agarose gel and visualized by ethidium bromide staining. Primers approximately 100 nucleotides 5′ or 3′ from a putative response element sequence were selected to amplify approximately 200 base pair products (bp) (Supplemental Table 2, S2).
Immunohistochemical analysis was performed on 6 μm serial sections of B6 and Nr2e3rd7/rd7 eyes as previously described (Haider et al. 2006). Briefly, paraffin embedded eyes were fixed with either 4% PFA or Methanol/Acetic acid (3:1) and maintained at a dorsal/ventral orientation. After blocking with 2% normal goat or horse serum (Vector, CA) in PBS, sections were incubated with the following primary antibodies: Gnb1 (rabbit, 1:100, Protein Tech Group), rhodopsin (mouse, 1:500, Leinco Technologies), peanut agglutanin (1:200, Sigma) at 4°C overnight. The next day, samples were rinsed and incubated in the appropriate secondary antibody (1:400, AlexaFluor by Molecular Probes, Invitrogen) for 1–2 hours. Images from sections were collected on a Zeiss Axioplan II fluorescent microscope equipped with the appropriate bandpass filters for each flurochrome.
To characterize the transcriptional networks associated with NR2E3 function in developing and mature photoreceptors, we determined the expression profile of genes in mice lacking Nr2e3. Subtractive hybridization was performed using P2 and P21 retina samples to identify genes involved in development and maintenance of photoreceptors. At each time point, B6 cDNAs were subtracted from Nr2e3rd7/rd7 cDNAs to identify transcripts up-regulated in Nr2e3rd7/rd7 mice, and conversely, Nr2e3rd7/rd7 cDNAs were subtracted from B6 cDNAs to identify the transcripts which are down-regulated in Nr2e3rd7/rd7 mice. We sequenced approximately 200 clones and identified 14 up-regulated and 12 down-regulated transcripts at P2, and 10 up-regulated and 6 down-regulated transcripts at P21 (Supplemental Table S3). The observed expression changes identified through subtractive hybridization were then confirmed by real time PCR.
Additionally, we took a more targeted approach and performed real time PCR testing over 100 genes known to function in photoreceptor development, structure or maintenance of the retina, in transport of proteins in the retina, and approximately 20 orphan receptors, to determine if any known retinal genes were misregulated in Nr2e3rd7/rd7 mice (Supplemental Table S4). Overall, from this combined effort, 30 genes at P2 and P21 were found to be misregulated in the Nr2e3rd7/rd7 compared to wildtype retinas (Figure 1, Table 1). While 10 of these genes were also reported in previous studies (Corbo and Cepko 2005, Chen et al, 2005 Oh et al. 2008) only three were previously identified as targets of NR2E3 (Peng et al. 2005). At P2, thirteen genes were significantly up regulated and three genes that are downregulated in Nr2e3rd7/rd7 retinas relative to controls. The majority of these genes were transcriptional regulators, four were signaling molecules, and several novel transcripts with unknown function were also identified. Six of the sixteen genes were identified by both subtractive hybridization and targeted expression profiling approaches. At P21, we identified twenty misregulated genes. In contrast to the genes identified at P2, the majority of these genes were not transcription factors. Expression of several signaling molecules were altered in mutant retinas. Also, while most of the genes identified at P2 were up-regulated, an equal number of genes were either up or down regulated at P21 in the Nr2e3rd7/rd7 retinas relative to controls. Six genes were differentially expressed at P2 and P21.
In order to determine which of the misregulated genes were potential targets of Nr2e3, we performed chromatin immunoprecipitation (ChIP) using P2 and P21 B6 retinas. First, we evaluated approximately 30 kb of 5′ untranslated sequence from each of the 30 misregulated genes (from Figure 1, Table 2) to identify potential Nr2e3 binding/response element sequences. These sites were chosen based on earlier reports (Kobayashi et al. 1999, Chen et al. 2005) that defined a putative response element for NR2E3 (AAGTCAnAAGTCA), which included two identical motifs separated by a spacer. At P2, ten putative targets NR2E3 binding sites were evaluated, and eight genes were identified as directly targeted by Nr2e3 (Figure 2, Table 1). Among these genes were novel transcripts with unknown function (Riken cDNA gene 2610101N10 (Rik1n10), and accession id BC07393 (BC7393)), two were known transcription factors (Ror1, Rorg), and one was a signaling protein (Irbp). At P21, ten of the eleven binding sites we tested were positive (Figure 2). When examining the expression at P2 and P21, six genes were misregulated at both P2 and P21 (Nr2e3, Nr1d1, Nr2c1, Rgr, Gnb3, and Rora). Interestingly, NR2E3 appears to regulate its own expression in both the developing and mature retina (also in Peng et al. 2005). While many of the response elements we identified were not identical dimeric motifs, they were greater 75% similar to the putative target sequence. The spacer sequences varied from one to four nucleotides, with one RE sequence, for the gene Ankmy2, containing a perfect dimeric repeat with a 13-nucleotide spacer sequence.
To determine if misregulation of gene expression resulted in an alteration of temporal or spatial protein expression of proteins directly targeted by NR2E3, we performed immunohistochemical analyses for the rod specific gene GNB1. As expected, in control B6 retinas, GNB1 co-localized with rhodopsin (Figure 3A–C) and not with the cone cell marker peanut agglutinin (PNA) (Figure 3G–I). Further, in the NR2e3rd7/rd7 retinas, GNB1 expression (Figure 3D–F, J–L) was restricted to rod outer segments, whereas in the B6 retina, expression was observed in both the outer and inner segments (Figure 3A–C, G–I). We did not observe mislocalization of GNB1 to cone cells in NR2e3rd7/rd7 retinas. While altered protein localization and expression can be a secondary effect of the morphological changes apparent in the mature rd7 retina, that Gnb1 expression is potentially regulated by NR2E3 suggests that suggests in this case, altered expression may be due to Nr2e3 function in the mature rod cell.
Enhanced S-Cone Syndrome has the unusual phenotype of a gain of function of the least populous photoreceptor subtype. Although this is a rare human disorder, the nuclear receptor NR2E3, which is mutated in this disorder, is critical to photoreceptor development, and specifically photoreceptor generation. Our previous studies using Nr2e3rd7/rd7 mice as a model for ESCS show that the mechanism through which this phenotype occurs is over-proliferation of ectopic retinal progenitors (Haider et al., 2006). A similar enhanced S-cone phenotype is also observed in the tlx mice, lacking Nr2e1, a nuclear receptor very similar to Nr2e3. Nr2e1 modulates retinal cell proliferation and cell-cycle re-entry by inhibiting Pten expression, a negative regulator of neural stem cell proliferation (Zhang et al., 2006; Sun et al., 2007).
In this study, we identified several genes whose expression is regulated by Nr2e3 in the developing and mature retina. While the genes directly regulated by NR2E3 overlap to some extent in the developing and adult retina, the majority were unique subsets, suggesting a unique role for NR2E3 both in the development and maintenance of photoreceptor cells. Twenty-four direct targets of NR2E3 were identified, only three of which were previously reported as NR2E3 targets (Peng et al. 2005). Several classes of genes emerged suggesting NR2E3 provides input into multiple transcription networks simultaneously. These include transcription factors (e.g. Nr1d1, Nr2c1, and Rorg) and phototransduction genes (e.g. blue opsin, Gnat2, Gnb1, and Gnb3). In addition, we identified several novel genes contributing to these networks through regulation by NR2E3. While the majority of the genes we identified were up-regulated in Nr2e3rd7/rd7 retinas, we did identify several downregulated targets, suggesting that Nr2e3 can function both as a suppressor and/or enhancer of transcription. We identified several transcription factors targeted by Nr2e3 in the developing retina. Further work will need to be done to determine whether these genes are specific to mitotic retinal progenitors or postmitotic, differentiating cells that are committed to a cone or rod lineage. In addition, at P2 Nr2e3 regulates photoreceptor specific genes such as Irbp, recoverin, and rhodopsin kinase. While the eye is not yet open at P2, and the outer segments of photoreceptors, where these genes ultimately function, are still being developed, Nr2e3 regulated gene expression apparently prepares the retina for visual transduction.
Recent studies have shown Nr2e3 action alone does not have a profound effect on rod cell differentiation, however in combination with Nrl it promotes rod development (Hennig et al., 2008; McIlvain and Knox, 2007). Earlier gene expression studies have shown that lack of Nr2e3 results in misexpression of rod and cone genes; and specifically downregulation of rod genes and upregulation of cone genes (Corbo and Cepko 2005; Cheng et al., 2006). Similar to these reports, our study confirms that in Nr2e3rd7/rd7 mice, cone specific genes such as blue opsin, and the alpha and beta transducins (Gnat2 and Gnb3) are up-regulated while rod specific genes such as Gnb1 are down regulated compared to normal. The altered mRNA expression level translates into altered protein expression for GNB1 as seen in this study as well as GNAT2 (Haider et al. 2006) and PRPH2/RDS (Nystuen et al. 2008). Additionally, we also observed several genes expressed in both rod and cone cells were differentially expressed in the adult Nr2e3rd7/rd7 retina. Interestingly, at P2, most of the differentially expressed genes were up-regulated in Nr2e3rd7/rd7 retinas compared to normal, while in the adult retina, an approximately equal number of genes were increased or decreased. The six potential NR2E3 targets at P2 and P21 are all up-regulated in the P2 retina however only 4 are up-regulated in the P21 retina (Nr2e3, Nr2c1, Nr1d1, Gnb3) while two are down-regulated (Rgr, Rora). The down-regulation of Rgr and Rora in the Nr2e3rd7/rd7 retina could be due to loss of photoreceptor cells and not a direct consequence of lack of Nr2e3. These results suggest that perhaps in the developing retina Nr2e3 primarily functions as a repressor of transcription while in the adult retina it may function equally as a repressor or activator. Differential expression in the adult retina however may also be due to the primary developmental defects and thus do not accurately reflect how Nr2e3 functions. Additional experiments to confirm which of the targets identified in this study are directly targeted by Nr2e3 versus those that indirect targets or those that are differentially expressed due to secondary factors will be essential. Further, we are examining Nr2e3 function in a conditional mutant mouse where normal Nr2e3 expression persists in the developing retina and loss of Nr2e3 occurs only in mature rod or cone cells. These studies will provide valuable insight into delineating the role of Nr2e3 in the mature retina. In vitro and gain of function studies using the Nrl−/− mice suggest that Nr2e3 functions with Nrl in rod development (Cheng et al., 2004; Cheng et al., 2006). Our current studies demonstrate Nr2e3 potentially regulates the expression of rod and cone specific genes suggesting its role is not limited to suppression of cone genes in rod photoreceptor cells.
The molecular mechanisms through which Nr2e3 functions to direct cone cell generation, rod differentiation, and visual transduction remain to be determined. As a member of the steroid hormone family of nuclear receptors, Nr2e3 contains several conserved domains including DNA binding, dimerization, and ligand binding. The nuclear hormone receptor genes may function as heterodimers or homodimers and while many require ligand activation for function, they can also be recruited to the nucleus by cofactors. Nr2e3 is in the class of orphan receptors for which no ligand has been identified to date. Nr2e3 has been shown to interact with the nuclear receptor Nr1d1 (Cheng et al., 2004) and more recently, a novel cell-cycle co-repressor was identified for Nr2e3, RetCoR that expresses in the brain and retina (Takezawa et al., 2007). Nr2e3 appears to recruit RetCoR to repress Cyclin D1 expression, which is required for proliferation of retinal progenitor cells. While there is currently no known endogeneous ligand for NR2E3, high throughput assays have been performed and identified agonists of NR2E3 (Kapitskaya et al. 2006, Wolkenberg et al. 2006). It is well established that two ligands, retinoic acid and thyroid hormone, play a significant role in the developing retina and in particular in the differentiation of photoreceptor cells, however they have not yet been associated with NR2E3. In fact, thyroid hormone levels appear normal in the Nr2e3rd7/rd7 retina (Yanagi et al. 2002) suggesting it is not the ligand for Nr2e3. Our previous studies show that NR2E3 expression in the developing retina appears to be entirely nuclear while there is nuclear and predominantly perinuclear staining in the mature retina (Haider et al. 2006). This would suggest that perhaps in the developing retina, this nuclear receptor may be constitutively active and not require ligand or co-factor for activation and localization to the nucleus while additional aid from ligand activator or co-factor recruitment may be required in the mature retina. ChIP data presented in this study showing NR2E3 recruitment to its own promoter site may suggest an autoregulatory mechanism is present and may contribute to its highest expression in the adult retina (our unpublished data) and its higher message levels in the Nr2e3rd7/rd7 retina. Additional studies to examine NR2E3 complexes in the developing versus mature retina will provide further understanding of how it is recruited to the nucleus. These findings support a multifunctional role for Nr2e3 as dual repressor/activator to regulate retinal progenitors as well as postmitotic rod and cone cells.
This study provides new information about the transcriptional networks that regulate photoreceptor generation and function. The identification of transcription factors that are direct targets for Nr2e3 provides entry points into the biological pathways regulating photoreceptor generation and possibly differentiation. These studies also suggest a novel function for Nr2e3 in the mature retina in regulation of phototransduction genes in a cell dependent manner. Further studies will be performed to determine the cell type specific regulation of the Nr2e3-regulated genes.
This study was supported by the following grants from the National Eye Institute (EY11996 (PMN)), TJL Cancer Core Grant (CA-34196), NRSA Postdoctoral training grant (F32 EY07080-01A (NBH)), the Center for Biomedical Excellence Award through the National Center for Research Resources, NIH (NIH 5 P20 RRO18788-02 (NBH)), and the Nebraska Tobacco Settlement Biomedical Research Development, Hope for Vision (NBH).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.