Retinoid X Receptor Is Necessary to Establish the S-Opsin Gradient In

Home , Retinoid receptor, Retinoid X receptor, Thyroid hormone receptor

Retinoid X Receptor ␥ Is Necessary to Establish the S-opsin Gradient in Cone Photoreceptors of the Developing Mouse Retina

Melanie R. Roberts,1,2 Anita Hendrickson,2 Christopher R. McGuire,2 and Thomas A. Reh2

PURPOSE. The retinoid X receptors (RXRs) are members of the tion, they also have demonstrated some isoform-specific family of ligand-dependent nuclear hormone receptors. One of functions. For example, RXR␣ heterodimerizes with retinoic these genes, RXR␥, is expressed in highly restricted regions of acid receptors (RARs) to regulate retinal growth and cardiac the developing central nervous system (CNS), including the development.4 retina. Although previous studies have localized RXR␥ to de- Although all three isoforms are expressed in the developing veloping cone photoreceptors in several species, its function nervous system, RXR␥ has the most restricted and develop- in these cells is unknown. A prior study showed that thyroid mentally regulated expression. It is expressed in the develop- hormone receptor ␤2 (TR␤2) is necessary to establish proper ing striatum, part of the tegmentum, the pituitary, the ventral cone patterning in mice by activating medium-wavelength (M) horns of the spinal cord,3,5 and the retina.6 RXR␥-null mice cone opsin and suppressing short-wavelength (S) cone opsin. show isoform-specific defects in both striatal and hippocampal Thyroid hormone receptors often regulate gene transcription function, although RXR␤ can partially compensate for the loss as heterodimeric complexes with RXRs. of RXR␥ in the striatum.2,7,8 RXR␥ has also been identified in 6,9,10 METHODS. To determine whether RXR␥ cooperates with TR␤2 the developing retina of Xenopus, chicks, and mice. It has to regulate cone opsin patterning, the developmental expres- a highly restricted pattern of expression; and, in both chicks ␥ sion of RXR␥ was examined, and cone opsin expression in and mice, RXR is expressed in developing cones. Despite the RXR␥-null mice was analyzed. conserved expression pattern in developing cone photorecep- ␥ ␥ tors, the function of RXR in cone development is not known. RESULTS. RXR was expressed in postmitotic cones and was We have shown that thyroid hormone receptor ␤2 (TR␤2), transiently downregulated at the time of S-opsin onset in both ␥ another member of the same family of nuclear hormone recep- mouse and human cones. RXR -null mice expressed S-opsin in tors, has a role in the developmental “choice” of which opsin all cones, similar to the TR␤2-null mice. Unlike TR␤2-null mice, 11 ␥ gene to express. Like most mammals, mice have two opsin which did not express M-opsin, RXR -null mice had a normal proteins that respond maximally to different wavelengths: S- pattern of M-opsin expression. opsin to short wavelengths and M-opsin to longer wavelengths. CONCLUSIONS. RXR␥ is essential (along with TR␤2) for suppress- However, the ratio and distribution of cone opsins varies dra- ing S-opsin in all immature cones and in dorsal cones in the matically among species (for review, see Ref. 12). For example, mature retina, but it is not necessary for M-opsin regulation. S-opsin is excluded from the central fovea of human reti- These results demonstrate a critical role for RXRs in regulating nas.13,14 In mice, however, there is an opposing gradient of M- cell differentiation in the CNS and highlight a remarkable and S-opsin, so that M-opsin expression is highest in dorsal conservation of opsin regulation from Drosophila to mammals. cones, whereas S-opsin predominates in the ventral retina.15 (Invest Ophthalmol Vis Sci. 2005;46:2897–2904) DOI: TR␤2-null mice show a profound disruption in the cone opsin 10.1167/iovs.05-0093 gradient. In these mice, M-opsin is not expressed in any cones, indicating that TR␤2 is necessary to upregulate M-opsin. In he retinoid X receptors (RXRs) are members of the family addition, all cones in TR␤2-null mice express S-opsin, which Tof ligand-dependent nuclear hormone receptors that regu- indicates that TR␤2 is also necessary to suppress S-opsin. Be- late gene expression either as homodimers or more frequently cause TRs often regulate transcription as heterodimers with as heterodimeric complexes with other nuclear hormone re- RXRs, we hypothesized that RXR␥ may also play a part in the ceptors (for review, see Ref. 1). Studies of mice with targeted determination of cone opsin expression. gene mutations in each of the three RXR isoforms—␣, ␤, and To determine whether RXR␥ is necessary for proper cone ␥—have demonstrated that RXRs regulate several developmen- photoreceptor development, we examined the expression of tal processes, including cardiogenesis and neurogenesis.2,3 Al- RXR␥ during retinal development and analyzed the cone phe- though these studies have revealed some redundancy of func- notype of RXR␥-deficient mice. We found that RXR␥ is expressed in developing cones in both the mouse and human retina and that RXR␥ is downregulated at the time of S-opsin From the 1Graduate Program in Neurobiology and Behavior and onset in both species. Like TR␤2-deficient mice, RXR␥-defi- the 2Department of Biological Structure, University of Washington, cient mice express S-opsin in every cone. Unlike TR␤2-defi- Seattle, Washington. cient mice, however, RXR␥-deficient mice have a normal M- Supported by National Institutes of Health Grants NS28308 (TAR) opsin gradient. Thus, RXR␥ is involved in the regulation of the and T32 EY07031 (MRR). expression of S-opsin expression, but not that of M-opsin, Submitted for publication January 25, 2005; revised March 28, which suggests that M- and S-opsin are independently regu- 2005; accepted March 31, 2005. lated by different nuclear hormone receptor complexes. Disclosure: M.R. Roberts, None; A. Hendrickson, None; C.R. McGuire, None; T.A. Reh, None The publication costs of this article were defrayed in part by page METHODS charge payment. This article must therefore be marked “advertise- ment” in accordance with 18 U.S.C. §1734 solely to indicate this fact. Animals Corresponding author: Thomas A. Reh, University of Washington, Department of Biological Structure, Box 357420, Seattle, WA 98195; RXR␥-null mice (C57BL/6; from Ronald Evans, The Salk Institute, La [email protected]. Jolla, CA)2 were genotyped with separate DNA primer pairs for the

Downloaded from iovs.arvojournals.org on 09/26/2021 2898 Roberts et al. IOVS, August 2005, Vol. 46, No. 8

wild-type (5Ј-ACTAGTGGATCCACTCCACTTCC-3Ј,5Ј-ACTGCGCTGG- or lowest S opsin expression was 0° dorsal. The average percentage of TGCGAATGATTGT-3Ј) and the RXR␥-null allele (5Ј GACACCAGAC- opsinϩ cones (ϮSEM) was reported for fields in the dorsal, central, and CAACTGGTAATGG-3Ј,5Ј-GCATCGAGCTGGGTAATAAGCG-3Ј). Reac- ventral thirds of each retina. Statistical significance was determined tions containing all four primers were amplified for 28 cycles (30 with a one-way ANOVA. seconds at 94°C, 1 minute at 60°C, and 1 minute at 72°C). The day of birth was designated postnatal day (P)0. Human fetal eyes were ob- RESULTS tained from the Human Embryology Laboratory (University of Wash- ington School of Medicine) or from ABR, Inc. (Alameda, CA). All RXR␥ in Mouse and Human Cones experiments were conducted in accordance with the institutional ␥ guidelines of the University of Washington School of Medicine and the Previous reports indicate that one of the RXR isoforms, RXR , ARVO Statement for the Use of Animals in Ophthalmic and Vision is expressed in both mature mouse cones and in putative cones 6,17 Research. of the developing mouse retina. We used both immunofluorescence and in situ hybridization to confirm these reports. Tissue Preparation We also used an immunofluorescence to assay for RXR␥ expression in developing human cones. For immunohistochemistry, retinas were dissected in cold phosphate- In the embryonic mouse retinas, RXR␥ antibodies labeled buffered saline (PBS), fixed for 2 hours in 4% paraformaldehyde, rinsed nuclei in the developing ganglion cell layer and at the scleral in PBS, and equilibrated in 30% sucrose/PBS overnight at 4°C. Retinas surface (developing photoreceptor layer) as early as embryonic were frozen in optimal cutting temperature compound (OCT, Tissue- day (E)14.5. RXR␥ expression extended along the length of the ␮ Tek; Sakura Finetek, Torrance, CA) and cryosectioned at 12 m retina in these regions by E15.5 (Fig. 1A). In situ hybridization ␮ (mouse) or 20 m (human). for RXR␥ mRNA showed a pattern of expression similar to that observed with the antiserum (Fig. 1B). The scleral surface of Immunohistochemistry and Flatmounts the developing retina contains both dividing progenitor cells Slides were incubated overnight in primary antibodies—1:200 rabbit and immature photoreceptors. To test directly whether the RXR␥ (sc-555, lot A1111), 1:150 goat S-sensitive opsin (sc-14363; Santa cells that express RXR␥ are progenitors or immature photore- Cruz Biotechnology, Santa Cruz, CA), 1:500 rabbit S-opsin or M-opsin ceptors, we double-labeled retinal sections from E16 mice with (AB5407, AB5745; Chemicon International, Temecula, CA), 1:10,000 RXR␥ and an antibody to phosphohistone H3, which is ex- rabbit TR␤2 (a gift from Lily Ng and Douglas Forrest), or 1:200 rat pressed specifically in mitotic progenitors (Fig. 1C). We did not monoclonal phosphohistone H3 (Ab10543; Novus Biologicals, Little- find any phosphohistone-expressing cells at the scleral surface ton, CO)—and for 2 hours in secondary antibodies (1:500 AlexaFluor of E16 retinas that also expressed RXR␥, indicating that RXR␥ goat anti-rabbit 568, goat anti-rat 488, donkey anti-rabbit 568, or is not expressed in progenitors. Thus, the RXR␥-positive cells chicken anti-goat 488; Molecular Probes, Inc., Eugene, OR). To visual- were most likely immature photoreceptors. ize cone outer segments, we incubated sections with FITC-conjugated There are no known markers to discriminate between im- peanut lectin (Sigma-Aldrich) for 2 hours. Sections were rinsed thor- mature rods and cones in mouse retinas. However, we suggest oughly in PBS before they were coverslipped (Fluoromount-G; South- that these immature photoreceptors are cones, based on two ern Biotechnology, Birmingham, AL). Slides were viewed on a confocal lines of evidence. First, by E16, most of the cones but only a 18 microscope (Axioplan 2 or LSM5 Pascal; Carl Zeiss Meditec, Dublin, small percentage of the rods have been generated. Although CA), and photographed with a digital camera (Spot 2.3.1; Diagnostic we cannot rule out the possibility that a few of the early-born Instruments, Sterling Heights, MI). rods express RXR␥, many of the RXR␥-expressing photorecep- For flatmount studies, retinas were marked on the dorsal side with tors must be cones. Second, as the photoreceptors matured, tissue-marking dye (Cancer Diagnostics, Inc., Birmingham, MI), flat- we were able to confirm that RXR␥ is expressed specifically in tened on a glass slide, fixed with 4% paraformaldehyde for 15 minutes, cone nuclei by double-labeling with peanut lectin, which binds and fixed for an additional hour in a 24-well plate. Retinas were specifically to cone outer segments, and with cone-specific incubated in primary antibodies at the concentrations noted earlier for opsins (described later). Every lectin-positive cone had an 48 hours at 4°C and in secondary antibodies overnight at 4°C. To label RXR␥-positive nucleus (Fig. 1D) and RXR␥ was not present in all cones, we incubated retinas in FITC-conjugated peanut lectin for 2 rod nuclei (not shown). hours (PNA; Sigma-Aldrich). Retinas were mounted (Fluoromount-G; We also analyzed the expression of RXR␥ in human retinas, Southern Biotechnology) between coverslips separated by an imaging where it is much easier to distinguish immature cones from chamber (Secure-Seal; Grace Bio-Laboratories, Bend, OR). rods because of differences in their morphology and protein expression. An early marker of cone photoreceptors in human In Situ Hybridization retina, tubby-like protein 1 (TULP-1), is expressed in cones 19 ␥ before the onset of cone opsin expression. By double-label Riboprobes were made to the full 1.6-Mb RXR transcript (Ronald immunofluorescence, we found that all TULP-expressing cones Evans, The Salk Institute). Tissue preparation and riboprobe synthesis ␥ 16 had RXR -positive nuclei by fetal week (FW)16 (Fig. 2A). was described previously. ␣-Transducin is another cone-specific marker that labels both S- and M-opsin-expressing cones in fetal human retina.20 At Quantitative PCR FW16, ␣-transducin was visible in the cytoplasm of cones We prepared cDNA from wild-type and RXR␥-null retinas. Levels of across much of the retina. These ␣-transducin expressing RXR␣ and -␤ in P5 retinas and of M-opsin in adult retinas were cones also have RXR␥ nuclei (Fig. 2B). Thus, our data suggest measured by quantitative PCR, as described elsewhere.16 We analyzed that RXR␥ is an early cone-specific marker. three to five individuals of each genotype and age. Primer sequences are available on request. Dynamic Expression of RXR␥ during Cone Development Data Analysis To analyze further the developmental time course of RXR␥ To determine the percentage of opsinϩ cones in flatmounts, we expression, we labeled mouse retinal sections from animals at counted the total number of PNAϩ and opsinϩ cones for at least ages E12.5 through adult. We found RXR␥ expression in a few sixteen 1.16-mm2 fields in four to five retinas for each genotype. The newly generated cones at the scleral surface and also in the field within the dye-labeled dorsal quadrant with the highest M opsin ganglion cell layer beginning at E14.5 (Fig. 3A). By E15.5, RXR␥

Downloaded from iovs.arvojournals.org on 09/26/2021 IOVS, August 2005, Vol. 46, No. 8 RXR-Mediated S-opsin Gradient in Cones 2899

sient. By P5 all cones re-expressed RXR␥ (Fig. 3E). RXR␥ expression persisted in cones through adult stages, though the intensity of expression was somewhat reduced after the first postnatal week (Fig. 3F). The transient downregulation of RXR␥ in cones correlated with the onset of S-opsin expression, which begins at E18. We hypothesized that downregulation of RXR␥ is necessary to allow induction of S-opsin. To determine whether the expression of these genes is consistent with the hypothesis, we labeled retinal sections with both S-opsin and RXR␥ antisera at E19 (Figs. 3G–I). Although we occasionally observed cells that expressed both S-opsin and RXR␥ at this age, nearly all of the cones expressed either RXR␥ or S-opsin, but not both. Most S-opsin-expressing cones were found in regions where RXR␥ had been downregulated. Moreover, cones expressing the highest levels of S-opsin were usually RXR␥Ϫ or expressed very low levels of RXR␥. This downregulation was transient; by P5, RXR␥ was re-expressed in all S-opsin-expressing cones (Figs. 3J–L). The double-labeling results in mouse retina are consistent with the hypothesis that a transient downregulation of RXR␥ in perinatal cones promotes the onset of S-opsin. In human retinas, we also observed a transient downregulation of RXR␥ in S-opsin-expressing cones at the time of S-opsin onset. At the beginning of S-opsin expression at FW14 and FW16 (Fig. 4), the S-cones had undetectable or low levels of RXR␥. However, by FW18, when S-opsin expression has reached the edge of the retina,13 most S-opsin-expressing cones also re-expressed RXR␥ (not shown). Patterns of RXR␥ and TR␤2 during Cone Maturation We reported earlier that TR␤2 is expressed in developing cones.11 To determine whether TR␤2 has the same developmental pattern as RXR␥, we labeled retinal sections and flatmounts with TR␤2 antiserum, which was raised against an N-terminal polypeptide of mouse TR␤2 (Lily Ng, Iwan Jones, and Douglas Forrest). We observed TR␤2 in postmitotic cones at E14.5, corresponding with the time of RXR␥ onset (data not shown). By E15.5, TR␤2 and RXR␥ were uniformly expressed in developing cones (Fig. 5A, compare with RXR␥ in Fig. 1A). However, between E19 and P2, we observed a downregulation of TR␤2 in cones, similar to that described for RXR␥ (Fig. 5B). At P5, TR␤2 was re-expressed in S-opsin-positive cones (Figs. 5C, 5D). TR␤2 expression declined after the first postnatal week and was present at very low levels in adult cones (not shown).

FIGURE 1. RXR␥ was expressed in postmitotic photoreceptors. (A) Immunoﬂuorescence for RXR␥ in a transverse section of an E15.5 mouse retina revealed RXR␥ϩ nuclei in the developing ganglion cell layer (GCL) and at the scleral surface (arrows). (B) In situ hybridization for RXR␥ mRNA in an E15 mouse retina conﬁrmed that RXR␥ mRNA was expressed in the same regions as RXR␥ protein. (C) Double-label of an E16 mouse retina with phosphohistone H3 (PH3; green), which labeled dividing cells, and RXR␥ (red). No nuclei were RXR␥ϩ/PH3ϩ (D). All cones in the adult retina (outer segments labeled with peanut lectin, green) had RXR␥ϩ nuclei (red) Scale bar, 100 ␮m.

was expressed uniformly across the retina (Fig. 1A) and re- mained robust until E17.5 (Fig. 3B). However, there was a dramatic decline of RXR␥ expression in cones just before birth. From E19 (Fig. 3C) to P2 (Fig. 3D), we observed only small ␥ patches of isolated RXR -positive cones. We could not detect FIGURE 2. RXR␥ was expressed in human cones. Double-label immu- a consistent spatial pattern in the downregulation of RXR␥, and noﬂuorescence in FW16 human retinal sections revealed that (A) RXR␥ the patches of RXR␥ varied in number and location, even labeled the nuclei of all TULP-1-expressing and (B) ␣-transducin-ex- between littermates. The downregulation in RXR␥ was tran- pressing cones. Scale bar, 200 ␮m.

Downloaded from iovs.arvojournals.org on 09/26/2021 2900 Roberts et al. IOVS, August 2005, Vol. 46, No. 8

FIGURE 3. RXR␥ was downregulated at the onset of S-opsin expression in mouse retinas. (A–F) Retinal sections at various developmental time points were labeled with RXR␥ antiserum. RXR␥ protein was detected in cone nuclei at E14.5 (A). The number or RXR␥-expressing cones increased until E17.5 (B). RXR␥ was transiently downregulated between E19 (C) and P2 (D). RXR␥ was re-expressed in all cones by P5 (E) and persisted in adult cones (F). At E19 (G–I), S-opsin (green) turned on in cones with downregulated RXR␥ (red, arrowheads). Cells that retained high levels of RXR␥ (arrows) had not yet turned on S-opsin. By P5 (J–L), RXR␥ (red) was re-expressed in all S-opsin-positive (green) cones.

S- and M-opsin Expression in in the normal pattern of S-opsin expression: S-opsin was ex- RXR␥-Deficient Mice pressed at nearly equal levels in both dorsal and ventral cones (Figs. 6B, 6D). The ectopic expression of S-opsin was main- To determine whether RXR␥ has a role in the regulation of tained in RXR␥-null adult retinas. In adult RXR␥-null retinas, cone opsin expression, we analyzed RXR␥-null mice for nearly all cones expressed S-opsin (Figs. 6F, 6H), whereas in changes in opsin expression. TR␤2-null mice do not express M-opsin, but all cones express S-opsin.11 Because RXR␥ het- wild-type retinas, few dorsal cones expressed S-opsin (Figs. 6E, ␤ 6G). This phenotype was similar to that of TR␤2-null mice, and erodimerizes with TR 2 and both receptors are expressed in ␤ ␥ developing cone nuclei, we hypothesized that RXR␥ is neces- supports our hypothesis that a TR 2-RXR heterodimer is sary, along with TR␤2, to regulate cone opsin expression and necessary to establish the S-opsin gradient by suppressing its predicted that RXR␥-null mice would have a phenotype similar expression in a subset of dorsal cones. to that of TR␤2-null mice. Because mice deficient in TR␤2 lack M-opsin, we also ex- In wild-type mice, the S-opsin-expressing cones were con- amined RXR␥-null mice for changes in M-opsin expression. centrated in the ventral retina, and there were few S-opsin- Remarkably, we did not find the same M-opsin phenotype in positive cones in the most dorsal retina. This gradient is estab- RXR␥-null retinas that we had in TR␤2-null retinas. M-opsin lished during development. Even in P0 retinas, many more expression in RXR␥-null retinal sections (Figs. 6J, 6L) was ventral cones than dorsal cones expressed S-opsin (Figs. 6A, indistinguishable from the gradient in wild-type sections (Fig. 6C). By contrast, P0 RXR␥-null mice had a profound disruption 6I, 6K). We also used quantitative PCR to confirm that the level

Downloaded from iovs.arvojournals.org on 09/26/2021 IOVS, August 2005, Vol. 46, No. 8 RXR-Mediated S-opsin Gradient in Cones 2901

DISCUSSION Other reports have shown expression of RXR␥ in cone photoreceptors of several species.6,9,10 To determine whether RXR␥ has a role in cone development, we examined its spatial and temporal expression and the effect of RXR␥ loss of function on opsin expression in the mouse retina. We found that RXR␥ is necessary for the developmental determination of cone opsin

FIGURE 4. RXR␥ was downregulated in S-opsin-expressing cones in developing human retinas. Sections from FW16 human retinas were colabeled with RXR␥ and S-opsin antisera. A three-dimensional confocal image of the photoreceptor surface revealed numerous RXR␥ϩ nuclei (red). However, S-opsin expressing cones (green) did not express RXR␥ (✽). Scale bar, 10 ␮m.

of M-opsin expression is similar between wild-type and RXR␥- null retinas (data not shown). To quantify the opsin gradients in wild-type, RXR␥ϩ/Ϫ, and RXR␥-null retinas, we coimmunolabeled retinal flatmounts with an opsin antibody and FITC-peanut lectin (PNA), which labeled all cones (Figs. 7A–D). We determined the percentage of cones that expressed S-opsin in the dorsal, central, and ventral regions of the retina (Fig. 7E). In wild-type retinas, 19% of dorsal cones, 63% of central cones, and 86% of ventral cones expressed S-opsin. By contrast, almost all cones in RXR␥-null retinas expressed S-opsin; 93%, 98%, and 96% of cones expressed S-opsin in the dorsal, central, and ventral retina, respectively. We found a significant gene-dosage effect of RXR␥ inactivation on the S-opsin gradient. In RXR␥ϩ/Ϫ animals, S- opsin was expressed in 40% of dorsal, 86% of central, and 92% of ventral cones. Counts of M-opsin-expressing cones showed that 94%, 79%, and 34% of wild-type cones expressed M-opsin, compared with 88%, 63%, and 34% of RXR␥-null cones in the dorsal, central, and ventral retinas, respectively (Fig. 7E). RXR␣ and RXR␤ are also expressed in the retina6 and have been shown to compensate for RXR␥ function in other systems.7 To determine whether other RXR isoforms are upregulated in RXR␥-null retinas, we performed quantitative PCR to compare the levels of RXR␣ and -␤ in wild-type and RXR␥-null retinas at P5. RXR␣ expression was not affected in RXR␥-null retinas. We found a small increase (3.6-fold) of RXR␤ in RXR␥- null retinas. However, because of high variation between animals, this increase was not significantly significant. To examine further whether RXR␤ compensates for RXR␥ in M-opsin regulation, we counted the number of M-opsin-expressing cones in RXR␤/␥ double-null mice. M-opsin expression in RXR␤/␥ double-retinas was still graded, but M-opsin expression was significantly reduced compared to RXR␥-null retinas (data not shown). FIGURE 5. TR␤2 had an expression profile similar to that of RXR␥ in Taken together, our results indicate that both RXR␥ and the developing mouse retina. TR␤2 expression was examined by im- TR␤2 are necessary to establish the ventral–dorsal gradient of munohistochemistry (compare with RXR␥ expression in Figs. 1, 2). (A) By E15.5, TR␤2 was expressed in immature cone nuclei (arrows). Note S-opsin by suppressing its expression in a subset of dorsal ␥ ␤ ␤ ␥ that, unlike RXR ,TR 2 was not expressed in the ganglion cell layer cones. However, there is a need for TR 2, but not for RXR ,in (GCL). Downregulation of TR␤2 was coincident with S-opsin onset. (B) the regulation of M-opsin expression. Thus, the expression of In an E19 retinal section, only a few patches of TR␤2 were visible at the M- and S-opsin in developing cone photoreceptors is regulated scleral surface (arrows). (C) By P5, TR␤ 2(red) was re-expressed in all by different nuclear receptor complexes. S-opsin-expressing cones (green). Scale bar: (A) 100 ␮m; (D)25␮m.

Downloaded from iovs.arvojournals.org on 09/26/2021 2902 Roberts et al. IOVS, August 2005, Vol. 46, No. 8

FIGURE 7. Quantification of opsin gradients. Four-week-old retinal flatmounts were labeled with peanut lectin (PNA) to label cones and S- or M-opsin to determine the percentage of cones expressing each opsin in the dorsal, central, or ventral region. Dorsal and ventral fields double labeled with PNA (green) and S-ospin (red) are shown for wild-type (A, C) and RXR␥Ϫ/Ϫ (B, D) retinas. (E) S-opsin was upregulated in both RXR␥ϩ/Ϫ and -Ϫ/Ϫ dorsal retinas and also in the central retina of RXR␥Ϫ/Ϫ mice. The gradient of M-opsin was similar between wild-type and RXR␥Ϫ/Ϫ retinas. There was no significant difference in M-opsin expression in either the dorsal or ventral retina. However, the 17% decrease of M-opsin in the mutant central retina is statistically significant. *P Ͻ 0.05, **P Ͻ 0.01, ***P Ͻ 0.0001; one-way ANOVA. Scale bar, 20 ␮m.

expression. Mice deﬁcient in RXR␥ express S-opsin in nearly every cone, resulting in a loss of normal cone opsin patterning. Although we conﬁrmed the previous reports that RXR␥ is expressed in developing and adult cone photoreceptors, we

retinas (E–L), opsin was mostly localized to cone outer segments Ϫ/Ϫ FIGURE 6. RXR␥ mice had upregulated S-opsin and normal M- (arrows), though S-opsin is also visible in a few cell bodies (arrow- opsin expression. The expression of S-opsin and M-opsin was exam- head). Adult wild-type retinas have few S-opsin-expressing cones in the Ϫ Ϫ ined by immunohistochemistry. At P0 (A–D), S-opsin was visible in the dorsal retina (E) and many in the ventral retina (G), whereas RXR␥ / cell body and processes of the cones. P0 wild-type retinas had few retinas had many S-opsin-expressing cones in both dorsal (F) and S-opsin-expressing cones in the dorsal retina (A) and many in the ventral (H) retina. M-opsin expression (I–L) appeared to be similar in Ϫ Ϫ ventral retina (C). By contrast, P0 RXR␥Ϫ/Ϫ retinas had many S-opsin- both wild-type (I, K) and RXR␥ / (J, L) retinas. In both, M-opsin was expressing cones in both the dorsal (B) and ventral (D) retina. In adult highest in the dorsal retina (I, J) and lowest in the ventral retina (K, L).

Downloaded from iovs.arvojournals.org on 09/26/2021 IOVS, August 2005, Vol. 46, No. 8 RXR-Mediated S-opsin Gradient in Cones 2903

S-syndrome, and mouse mutants in this gene have an increased number of S-cones.28–30 Targeted deletion of Nrl, a leucine zipper transcription factor upstream of Nr2e3, has an even more extreme phenotype. Rods fail to develop and nearly all photoreceptors become S-opsin-expressing cones.31 We have shown that TR␤2 is critical for the developmental choice between S- and M-opsin in cones,11 and now we show that RXR␥ is also a critical regulator of this step. In our model, RXR␥/TR␤2 heterodimers suppress S-opsin expression in a subset of cones, whereas M-opsin is regulated by either a TR␤2 homodimer or TR␤2 with an unknown heterodimeric partner. As noted in the introduction, mice have an opposing gradi- FIGURE 8. Proposed model of photoreceptor development. Multipo- ent of M- and S-opsin, so that M-opsin expression is highest in tent retinal progenitor cells give rise to rods, S-cones, and M-cones. The the dorsal retina, whereas S-opsin-expressing cones predomi- particular opsin gene expressed in the cell is controlled in part by 12,15 nuclear receptor transcription factors, Nr2e3, TR␤2, and RXR␥. Nr2e3 nate in the ventral retina. By contrast, the human central 13,14 suppresses S-opsin expression in rods, whereas a heterodimer between fovea lacks S-opsin-expressing cones. Because we did not RXR␥ and TR␤2 suppresses S-opsin in M-cones. Because TR␤2 is also observe a gradient RXR␥ or TR␤2 expression in developing necessary for M-opsin expression in cones, either a TR␤2 homodimer, mouse retinas, it is unlikely that regulation of receptor expres- or a heterodimer with an unknown partner, is postulated to activate sion alone controls the dorsal–ventral cone opsin gradient. M-opsin expression. Rather, we propose that a gradient in one or both of the ligands for these receptors controls cone opsin patterning. There is found a striking, but transient, downregulation of RXR␥ and evidence of at least two ligands for RXRs: 9-cis retinoic acid and TR␤2 in the perinatal retina that coincided with S-opsin onset. the polyunsaturated fatty acid docosahexanoic acid.32,33 Al- We also found that RXR␥ was expressed in developing human though several studies have described gradients in all-trans- cone photoreceptors, suggesting that the function of this gene retinoic acid,34,35 the ligand for RARs, the distribution of RXR in cone development is conserved. ligands in the retina has not been reported. Nevertheless, there We propose that RXR␥ and TR␤2 cooperate to regulate the is evidence that ligand-dependent RXR activation occurs nor- pattern of S-opsin expression in developing cones. Studies in mally in the developing retina,9 consistent with the possibility many systems have shown that TR/RXR heterodimers can reg- that a gradient in an RXR ligand controls the graded pattern of ulate transcription.21–24 We have reported that mice deficient opsin expression. in TR␤2 show a profound disruption in the normal pattern of The transcriptional mechanisms that control photoreceptor opsin expression in cone photoreceptors.11 We now show that specification are remarkably well conserved among species. both TR␤2 and RXR␥ are present in developing cones and that Notably, the Drosophila homologue of RXR, ultraspiracle loss of either receptor results in a similar disruption in the (usp), regulates opsin transcription in a heterodimeric com- pattern of S-opsin expression. Thus, it is likely that het- plex with the ecdysone receptor, the fly homologue of thyroid erodimers of RXR␥ and TR␤2 normally establish the pattern of hormone receptor. Flies have eight photoreceptors (R1–8) S-opsin expression by inhibiting it in many of the dorsal retinal that express one of five opsins. The R7 photoreceptor ex- cones. Of interest, both RXR␥ and TR␤2 were transiently presses either Rh-3 or Rh-4 opsin (for review, see Ref. 36). downregulated at S-opsin onset in both mice and humans, Remarkably, R7 photoreceptors in usp-null flies express only suggesting that this downregulation is necessary for the timing Rh-3.37,38 Thus, usp may suppress Rh-3 in fly photoreceptors, of S-opsin expression. just as RXR␥ suppresses S-opsin in mice. Because the transcrip- In contrast with our previous analysis of the TR␤2-null tional machinery controlling opsin expression is well con- mice, we found that RXR␥ was not necessary for activation of served, RXR, TR␤2, and their ligands may be used to generate M-opsin expression. Although TR␤2-null cones do not express highly varying photoreceptor patterns in a multitude of spe- M opsin, RXR␥-null retinas have normal M-opsin expression. cies, from flies to humans. Thus, S- and M-opsin must be regulated by different nuclear receptor complexes. Because RXR␤ and -␣ are also expressed Acknowledgments in the retina,6 we hypothesized that one of these isoforms may The authors thank Pierre Chambon for the RXR␤-null mice and Melissa dimerize with TR␤2 and compensate for RXR␥’s loss, much as Phillips for technical assistance and helpful comments on the manu- RXR␤ does in the striatum.7 However, we did not find an script. upregulation of either RXR␣ or -␤ in RXR␥-null retinas and found that M-opsin expression was graded but reduced in References RXR␤/␥-double-null retinas. We could not examine M-opsin in 5 RXR␣-null mice because they die in utero. Thus, we cannot 1. McKenna NJ, O’Malley BW. Combinatorial control of gene expres- rule out that RXR␣ alone is sufficient to compensate for the sion by nuclear receptors and coregulators. Cell. 2002;108:465– loss of both RXR␥ and -␤ without being significantly upregu- 474. lated. Incomplete compensation by RXR␣ could explain the 2. Chiang MY, Misner D, Kempermann G, et al. An essential role for small but significant decrease of M-opsin in the central retina of retinoid receptors RARbeta and RXRgamma in long-term potenti- RXR␥-null mice. ation and depression. Neuron. 1998;21:1353–1361. Our model of how RXR␥ fits into photoreceptor develop- 3. Saga Y, Kobayashi M, Ohta H, et al. Impaired extrapyramidal ment is shown in Figure 8. Cones are generated in mice from function caused by the targeted disruption of retinoid X receptor RXRgamma1 isoform. Genes Cells. 1999;4:219–228. multipotent progenitors from E10.5 to E16.18 Shortly after 4. Kastner P, Grondona JM, Mark M, et al. Genetic analysis of RXR withdrawal from the cell cycle, both rods and cones express alpha developmental function: convergence of RXR and RAR sig- photoreceptor-specific transcription factors such as crx and 25 naling pathways in heart and eye morphogenesis. Cell. 1994;78: NeuroD1. Several nuclear hormone receptors are involved in 987–1003. regulating the next stages of photoreceptor differentiation. 5. Dolle P, Fraulob V, Kastner P, Chambon P. Developmental expres- Nr2e3 is an orphan nuclear receptor that is expressed in sion of murine retinoid X receptor (RXR) genes. Mech Dev. 1994; rods.26,27 Mutations in Nr2e3 underlie the human enhanced 45:91–104.

Downloaded from iovs.arvojournals.org on 09/26/2021 2904 Roberts et al. IOVS, August 2005, Vol. 46, No. 8

6. Mori M, Ghyselinck NB, Chambon P, Mark M. Systematic immu- by the structure of the hormone response element. Mol Cell Biol. nolocalization of retinoid receptors in developing and adult mouse 1995;15:4718–4726. eyes. Invest Ophthalmol Vis Sci. 2001;42:1312–1318. 23. Puzianowska-Kuznicka M, Damjanovski S, Shi YB. Both thyroid 7. Krezel W, Ghyselinck N, Samad TA, et al. Impaired locomotion and hormone and 9-cis retinoic acid receptors are required to effi- dopamine signaling in retinoid receptor mutant mice. Science. ciently mediate the effects of thyroid hormone on embryonic 1998;279:863–867. development and specific gene regulation in Xenopus laevis. Mol 8. Samad TA, Krezel W, Chambon P, Borrelli E. Regulation of dopa- Cell Biol. 1997;17:4738–4749. minergic pathways by retinoids: activation of the D2 receptor 24. Campos-Barros A, Erway LC, Krezel W, et al. Absence of thyroid promoter by members of the retinoic acid receptor-retinoid X hormone receptor beta-retinoid X receptor interactions in auditory receptor family. Proc Natl Acad Sci USA. 1997;94:14349–14354. function and in the pituitary-thyroid axis. Neuroreport. 1998;9: 9. Luria A, Furlow JD. Spatiotemporal retinoid-X receptor activation 2933–2937. detected in live vertebrate embryos. Proc Natl Acad Sci USA. 25. Hatakeyama J, Kageyama R. Retinal cell fate determination and 2004;101:8987–8992. bHLH factors. Semin Cell Dev Biol. 2004;15:83–89. 10. Hoover F, Seleiro EA, Kielland A, Brickell PM, Glover JC. Retinoid 26. Bumsted O’Brien KM, Cheng H, Jiang Y, Schulte D, Swaroop A, X receptor gamma gene transcripts are expressed by a subset of Hendrickson AE. Expression of photoreceptor-specific nuclear re- early generated retinal cells and eventually restricted to photoreceptor NR2E3 in rod photoreceptors of fetal human retina. Invest ceptors. J Comp Neurol. 1998;391:204–213. Ophthalmol Vis Sci. 2004;45:2807–2812. 11. Ng L, Hurley JB, Dierks B, et al. A thyroid hormone receptor that 27. Cheng H, Khanna H, Oh EC, Hicks D, Mitton KP, Swaroop A. is required for the development of green cone photoreceptors. Photoreceptor-specific nuclear receptor NR2E3 functions as a tran- Nat Genet. 2001;27:94–98. scriptional activator in rod photoreceptors. Hum Mol Genet. 2004; 13:1563–1575. 12. Szel A, Rohlich P, Caffe AR, van Veen T. Distribution of cone photoreceptors in the mammalian retina. Microsc Res Tech. 1996; 28. Milam AH, Rose L, Cideciyan AV, et al. The nuclear receptor 35:445–462. NR2E3 plays a role in human retinal photoreceptor differentiation and degeneration. Proc Natl Acad Sci USA. 2002;99:473–478. 13. Xiao M, Hendrickson A. Spatial and temporal expression of short, 29. Haider NB, Jacobson SG, Cideciyan AV, et al. Mutation of a nuclear long/medium, or both opsins in human fetal cones. J Comp Neu- receptor gene, NR2E3, causes enhanced S cone syndrome, a dis- rol. 2000;425:545–559. order of retinal cell fate. Nat Genet. 2000;24:127–131. 14. Bumsted K, Hendrickson A. Distribution and development of 30. Haider NB, Naggert JK, Nishina PM. Excess cone cell proliferation short-wavelength cones differ between Macaca monkey and hu- due to lack of a functional NR2E3 causes retinal dysplasia and man fovea. J Comp Neurol. 1999;403:502–516. degeneration in rd7/rd7 mice. Hum Mol Genet. 2001;10:1619– 15. Applebury ML, Antoch MP, Baxter LC, et al. The murine cone 1626. photoreceptor: a single cone type expresses both S and M opsins 31. Mears AJ, Kondo M, Swain PK, et al. Nrl is required for rod with retinal spatial patterning. Neuron. 2000;27:513–523. photoreceptor development. Nat Genet. 2001;29:447–452. 16. Kubota R, McGuire C, Dierks B, Reh TA. Identification of ciliary 32. Heyman RA, Mangelsdorf DJ, Dyck JA, et al. 9-cis retinoic acid is a epithelial-specific genes using subtractive libraries and cDNA ar- high affinity ligand for the retinoid X receptor. Cell. 1992;68:397– rays in the avian eye. Dev Dyn. 2004;229:529–540. 406. 17. Janssen JJ, Kuhlmann ED, van Vugt AH, et al. Retinoic acid recep- 33. de Urquiza AM, Liu S, Sjoberg M, et al. Docosahexaenoic acid, a tors and retinoid X receptors in the mature retina: subtype deter- ligand for the retinoid X receptor in mouse brain. Science. 2000; mination and cellular distribution. Curr Eye Res. 1999;19:338– 290:2140–2144. 347. 34. McCaffery P, Lee MO, Wagner MA, Sladek NE, Drager UC. Asym- 18. Carter-Dawson LD, LaVail MM. Rods and cones in the mouse metrical retinoic acid synthesis in the dorsoventral axis of the retina. II. Autoradiographic analysis of cell generation using triti- retina. Development. 1992;115:371–382. ated thymidine. J Comp Neurol. 1979;188:263–272. 35. McCaffery P, Posch KC, Napoli JL, Gudas L, Drager UC. Changing 19. Milam AH, Hendrickson AE, Xiao M, et al. Localization of tubby-like patterns of the retinoic acid system in the developing retina. Dev protein 1 in developing and adult human retinas. Invest Ophthal- Biol. 1993;158:390–399. mol Vis Sci. 2000;41:2352–2356. 36. Cook T, Desplan C. Photoreceptor subtype specification: from flies 20. Cornish EE, Xiao M, Yang Z, Provis JM, Hendrickson AE. The role to humans. Semin Cell Dev Biol. 2001;12:509–518. of opsin expression and apoptosis in determination of cone types 37. Oro AE, McKeown M, Evans RM. The Drosophila retinoid X recep- in human retina. Exp Eye Res. 2004;78:1143–1154. tor homolog ultraspiracle functions in both female reproduction 21. Zhang XK, Hoffmann B, Tran PB, Graupner G, Pfahl M. Retinoid X and eye morphogenesis. Development. 1992;115:449–462. receptor is an auxiliary protein for thyroid hormone and retinoic 38. Zelhof AC, Ghbeish N, Tsai C, Evans RM, McKeown M. A role for acid receptors. Nature. 1992;355:441–446. ultraspiracle, the Drosophila RXR, in morphogenetic furrow move- 22. Sjoberg M, Vennstrom B. Ligand-dependent and -independent ment and photoreceptor cluster formation. Development. 1997; transactivation by thyroid hormone receptor beta 2 is determined 124:2499–2506.

Downloaded from iovs.arvojournals.org on 09/26/2021