Identification of RDH10, an All-trans Retinol Dehydrogenase, in Retinal Müller Cells

Abstract

purpose. To investigate the expression of RDH10, an all-trans retinol dehydrogenase identified in the retinal pigment epithelium (RPE), in retinal Müller cells.

methods. The RDH10 protein levels in mouse eyecups and bovine tissues were examined by Western blot analysis using a polyclonal antibody against RDH10. The cellular localization in the retina was determined by immunohistochemistry. Expression of RDH10 in rMC-1, a cell line derived from rat Müller cells, was determined by RT-PCR and Western blot analysis. All-trans retinol dehydrogenase activity assays were performed using lysates from rMC-1 cells. The generation of all-trans retinal from tritiated all-trans retinol was analyzed by HPLC.

results. RDH10, retinal G protein-coupled receptor (RGR), and RPE65 all had higher expression levels in the eyecups of BALB/c than in C57Bl/6 mice. In addition to the RPE, RDH10 was also detected at lower levels in the retina and liver. Immunohistochemistry showed that RDH10 was localized in Müller cells in retinal sections. RDH10 was detected in rMC-1 cells, at both the RNA and protein levels. The rat RDH10 cDNA containing the full-length coding region was cloned from rMC-1 cells. The rat RDH10 cDNA encodes a protein of 341 amino acids and shares 99% sequence identity with human, bovine, and mouse RDH10 at the amino acid level. In rMC-1 cells, all-trans retinol dehydrogenase activity was detected in the microsomal fraction. NADP was shown to be the preferred cofactor, which is identical with the cofactor preference of the recombinant RDH10.

conclusions. RDH10 was expressed in retinal Müller cells, in addition to the RPE. RDH10 generates all-trans retinal, which is the substrate for the photoisomerase RGR in Müller cells.

In vertebrates, vision begins with the absorption of light by rhodopsin, causing the isomerization of the chromophore 11-cis retinal into all-trans retinal. The free all-trans retinal from visual pigments is then reduced to all-trans retinol in photoreceptors. To maintain visual function, 11-cis retinal needs to be regenerated to form rhodopsin. Previous studies ^{1 2 3} showed that both retinal pigment epithelium (RPE) and retinal Müller cells can convert all-trans retinoids into the 11-cis isomers. 

In the RPE, all-trans retinol is isomerized and oxidized into 11-cis retinal. Two pathways have been identified for this process in the RPE: the isomerohydrolase cycle in the dark and the photic cycle in the light. ^{1 4 5 6} In the dark, all-trans retinol is first esterified by lecithin-retinol acyltransferase (LRAT). ⁷ The retinyl ester is isomerohydrolized to 11-cis retinol by a yet-to-be-identified isomerohydrolase. The oxidation of 11-cis retinol is catalyzed by the 11-cis retinol dehydrogenase, RDH5 and RDH11, to generate 11-cis retinal for the regeneration of visual pigments. ^{8 9} A light-dependent visual cycle generating 11-cis retinal catalyzed by the RPE retinal G protein-coupled receptor (RGR) has also been identified. ⁴ RGR is a 32-kDa opsin primarily localized in the RPE and at lower levels in retinal Müller cells. ^{10 11} When exposed to 470 nm or near-UV light, RGR can isomerize all-trans retinal to 11-cis retinal. ^{12 13} The significance of RGR in the photic visual cycle for normal vision under continuous light has been confirmed recently using the RGR knockout (Rgr ^−/−) mice. ⁴  

RDH10 (GenBank accession no. AY178865; http://www.ncbi.nlm.nih.gov/Genbank; provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD) is a newly identified all-trans retinol dehydrogenase predominantly expressed in the RPE. ¹⁴ Like other retinol dehydrogenases or retinal reductases previously identified in the visual cycle, RDH10 also belongs to the short-chain dehydrogenase/reductase (SDR) superfamily. RDH10 cDNAs were cloned from the human, cow, and mouse; each encoded a protein of 341 amino acids. Recombinant human RDH10 expressed in COS cells specifically oxidized all-trans retinol to all-trans retinal in a NADP-dependent manner. ¹⁴ Therefore, RDH10 has the capacity to generate all-trans retinal, a substrate for RGR, in the photic visual cycle. 

Although the retinoid visual cycle in the RPE has been intensively studied, the retinoid metabolism in the Müller cell is still poorly understood. Previous studies ^{2 3} showed that 11-cis retinoids can be generated in Müller cells. An important early study ² showed that cultured chicken Müller cells can synthesize 11-cis retinoids from all-trans retinol. Recently, in an elegant series of experiments, Mata and colleagues ³ have shown that a novel pathway for 11-cis retinoid synthesis may exist in Müller cells. Müller cells also contain essential proteins identified in the visual cycles, such as cellular retinol-binding protein (CRBP) and cellular retinal-binding protein, a protein binding 11-cis retinal. ^{15 16} Similarly, RGR, the key protein in the photic visual cycle, has also been detected in Müller cells. ^{10 11} The present study demonstrates that RDH10 is also expressed in retinal Müller cells. 

Methods

Sample Preparation and Western Blot Analysis

All procedures concerning animals in this study were in strict agreement with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and approval was received from the institutional animal care and use committee. 

To examine mouse strain differences, BALB/c, C57Bl/6, RPE65 knockout (Rpe65 ^−/−) and RGR knockout (Rgr ^−/−) mice reared in 12-hour light/dark cycles were used. The background strain for the Rpe65 ^−/− and Rgr ^−/− mice is predominantly C57BL/6. Eyecups including the retina were dissected, pooled, and homogenized in 1% SDS buffer. Fifty μg total proteins from each lysate were loaded onto 12% polyacrylamide gels and subjected to Western blot analysis using a 1:2000 dilution of primary antibodies. Densitometry was analyzed by Quantity One program (Bio-Rad Laboratories, Inc., Hercules, CA). 

For the tissue distribution of RDH10 in bovine, fresh bovine tissues (from the local abattoir) were homogenized with a Teflon-glass homogenizer and then sonicated in 1% SDS buffer. Proteins (50 μg) from various homogenates were analyzed by Western blot analysis using the anti-RDH10 antibody as described above. 

The anti-RDH10 polyclonal antibody was raised in a rabbit using the RDH10 C-terminal peptide (position 327–341) as described previously, ¹⁴ and affinity-purified with a column of the antigen peptide. The RPE65 antibody was raised and characterized in previous studies. ¹⁷ The antibody for RGR is a generous gift from Henry Fong at University of Southern California. 

For detection of RDH10 in rMC-1 cells, cultured rMC-1 cells and C57Bl/6 eyecups (without the retina) were sonicated in 0.1 M sodium phosphate buffer, pH 7.4, containing 1 mM DTT and 0.32 M sucrose. The lysate was centrifuged for 20 minutes at 20,000g to remove tissue debris, unbroken cells, nuclei, and mitochondria. The supernatant was centrifuged at 100,000g for 1 hour. The microsomal pellet was washed five times with the same buffer for eyecup sonication. The final microsomal pellet was resuspended in the RDH activity assay buffer (10 mM BTP, 0.1 M NaCl, 1 mM DTT). Protein concentration was measured by the Bio-Rad Protein Assay (Bio-Rad Laboratories) according to the protocol of the manufacturer. 

Immunohistochemistry

The retinal sections from 2-month-old C57Bl/6 mice were fixed and embedded in the optimal cutting temperature (OCT) compound, as described previously. ¹⁸ The sections were washed twice with washing buffer (30 mM Tris, 150 mM NaCl, pH 7.5) at room temperature for 10 minutes. For additional permeabilization, sections were incubated in 10% methanol for 15 minutes. After three 1-minute washes in PBS, the sections were preincubated with the blocking buffer (4% human serum, 1% BSA, and 0.4% Triton X-100) for 1 hour. The sections were incubated with the anti-RDH10 antibody (1:1000 dilution) at 4°C overnight, followed by incubation with biotinylated anti-rabbit secondary antibody (1:300 dilution) at room temperature for 2 hours. The immunostaining was subsequently detected using the Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA) with DAB (0.05% diaminobenzidine in 0.1 M Tris, pH 7.5, and 0.003% hydrogen peroxide) as a substrate. 

Reverse Transcription–Polymerase Chain Reaction (RT-PCR)

The rMC-1 cells, an immortalized rat retinal Müller cell line, were a generous gift from Vijay Sarthy at Northwestern University and cultured as described previously. ¹⁹ Total RNA was isolated from rMC-1 cells and mouse eyecups with Trizol reagent (Invitrogen, Carlsbad, CA). cDNAs were reverse transcribed from 5 μg total RNA from rMC-1 cells and mouse eyecup. PCR was preformed using the Smart Cycler (Cepheid, Sunnyvale, CA) and a SYBR green PCR kit (PE, Warrington, UK). The RDH10 PCR was carried out with a sense primer mF8 (5′-CACGCAGAGCAATGAGGAGAC-3′) and an antisense primer T45 (5′-CGTCACAAGTGTAGGTAAAAACC-3′) for 33 cycles (94°C, 30 seconds; 56°C, 30 seconds; and 72°C, 1 minute). The PCR product was visualized by agarose gel electrophoresis. 

Cloning of Rat RDH10

Using the cDNA from rMC-1 cells as the template, a 1.1-kb cDNA containing the full-length coding region was amplified by PCR with a sense primer (5′-GTCGCGATGAACATC GTGGTG-3′) and an antisense primer (5′-TCGGCATGGGGTTCGAATCCAGAC-3′). PCR was carried out first at 94°C for 3 minutes, followed by 45 cycles of 94°C for 45 seconds, 64°C for 1 minute, and 72°C for 3 minutes. The PCR products were cloned into a vector (pCRII; Invitrogen); the positive colonies were selected and sequenced. All products were confirmed by sequencing the complementary strand and were verified in another clone from an independent PCR. 

All-trans Retinol Dehydrogenase Activity Assay

RDH activity assays were performed under a dim red light. For each reaction, rMC-1 cell microsomal (in 0.1 M sodium phosphate buffer, pH 7.4) and soluble fractions (500 μg of total protein) were added into 200 μL of the activity assay buffer containing 1% BSA and 400 μM NADP or NAD. The reaction was started by the addition of 2 μCi of all-trans [11,12-³H]-retinol (specific activity 47 Ci/mM; NEN Life Science Products, Boston, MA) in ethanol solution. After 1 hour at 37°C, retinoids were extracted by the addition of 300 μL of cold methanol and 300 μL hexane, vortexing and centrifuging 5 minutes at 10,000g. The upper organic layer was collected for HPLC analysis. Retinoids were analyzed using a normal phase HPLC 5 μm column (Lichrosphere SI-60; Alltech, Deerfield, IL) and isocratic solvent of 11.2% ethyl acetate, 2.0% dioxane, and 1.4% octanol in hexane. The products were analyzed with an in-line flow scintillation analyzer (Packard Radiomatic 500TR; Perkin Elmer, Wellesley, MA). Elution peaks were identified by spiking with the corresponding retinoid standards. Radioactive retinoid peaks were calculated as a percentage of total radioactivity, using the Millenium32 software (Waters Corporation, Milford, MA). 

Results

Strain Difference in RDH10 Expression Level

Equal amounts (50 μg) of eyecup proteins from BALB/c, C57Bl/6, Rpe65 ^−/− and Rgr ^−/− mice were analyzed by Western blot analysis using antibodies specific to RDH10, RGR, and RPE65 proteins. As shown in Figure 1 , BALB/c eyecups contain a higher RDH10 level than those of the C57Bl/6 mice. Correlating with the RDH10 levels, the RGR and RPE65 levels are also higher in BALB/c mice than in C57Bl/6 mice. Rpe65 ^−/− mice had RDH10 and RGR expression levels similar to C57Bl/6 mice. Rgr ^−/− mice showed similar expression levels of RDH10, but approximately 50% lower levels of RPE65, compared with wild-type C57Bl/6 mice, after normalization to β-actin (Fig. 1) . RDH10 was also detected in the livers of BALB/c and C57Bl/6 mice. Unlike in the eyecups, these two strains of mice showed similar RDH10 levels in the liver (Fig. 1E) , suggesting that the strain difference in RDH10 levels is ocular tissue-specific. 

Tissue Distribution Pattern of RDH10 Protein

The tissue distribution pattern of RDH10 protein in bovine tissues was determined by Western blot analysis using the anti-RDH10 antibody. While the highest expression level was detected in bovine RPE, RDH10 was also found in the retina and liver (Fig. 2) . All other tissues analyzed, including brain, lung, kidney, pancreas, and skeletal muscle, appeared devoid of detectable RDH10 protein. Protein integrity in all the tissues was demonstrated by Western blot analysis using the anti-β-actin antibody (data not shown). 

Cellular Localization of RDH10 in Retina

The anti-RDH10 antibody was used to determine the cellular localization of RDH10 in the retina by immunohistochemistry. In addition to the RPE, immunoreactivity was also detected in filamentous processes in the retina (Fig. 3) . The interspersed staining among the somata of retinal neurons corresponded to the known features of Müller cells with their radial trunks extending from the nerve fiber layer to the outer plexiform layer. 

Identification of RDH10 in Müller Cell Line

To further confirm the expression of RDH10 in Müller cells, we used a cell line derived from rat retinal Müller cells, rMC-1. Using primers specific to the consensus sequences of mouse and human RDH10 sequences, RT-PCR amplified a single band of RDH10 fragment from rMC-1 cells and mouse eyecup (Fig. 4A) . The PCR products of 150 bp matched the expected length based on the human and mouse RDH10 cDNA sequences. The size of 150 bp excluded that the PCR products were amplified from genomic DNA because the two PCR primers span a 1.5-kb intron, according to human and mouse genomic sequences of RDH10 (unpublished results). This result showed that the RDH10 mRNA exists in this Müller cell line. 

The expression of RDH10 in rMC-1 cells was also confirmed at the protein level by Western blot analysis, using the anti-RDH10 antibody. In the microsomal fraction of rMC-1 cells, the antibody recognized a single band of 39 kDa, which was identical with the band in the microsomal fraction of the eyecup without the retina (Fig. 4B) . 

Sequence Analysis of Rat RDH10 from rMC-1 Cells

To confirm that the RT-PCR product amplified from rMC-1 cells is from the RDH10 mRNA, the RDH10 cDNA containing the full-length coding region was synthesized by RT-PCR from rMC-1 cells. As the rMC-1 is a rat cell line and rat RDH10 had not been cloned previously, we cloned the RT-PCR product and determined its sequence. The cloned rat RDH10 sequence did not match any known genes in GenBank, but showed the highest sequence homology (95.8% identity at the nucleic acid level) with the mouse RDH10 cDNA. At the amino acid level, the cloned rat sequence shared 98.8%, 99.4%, and 99.4% identity with the mouse, human, and bovine RDH10, respectively. The rat RDH10 also retained all the motifs of the SDR superfamily members, such as the “TGXXXGXG” and “YXXXK” signature motifs of the SDR superfamily (Fig. 5A) . Moreover, rat RDH10 also had significant homology with other SDR enzymes, which have a role in retinoid metabolism and/or have been identified in the retina (Fig. 5B) . 

All-trans Retinol Dehydrogenase Activity in rMC-1 Cells

The presence of all-trans retinol dehydrogenase activity in rMC-1 cells was confirmed by an in vitro assay. The microsomal fraction of rMC-1 cells was washed three times with PBS to eliminate soluble intracellular nucleotide cofactors. This fraction generated only minimal amounts of all-trans retinal in the dehydrogenase activity assay in the absence of exogenous cofactors (Fig. 6A) . However, the addition of NADP resulted in the formation of a major peak of all-trans retinal (Fig. 6B , peak 1). NAD was found to be a much less efficient cofactor (Fig. 6C) . As a negative control, incubation of all-trans retinal with NADP alone in the absence of rMC-1 proteins did not generate any detectable all-trans retinal, demonstrating that the conversion of all-trans retinol to all-trans retinal is an enzymatic reaction. Little atRDH activity was detected in the cytosolic fraction of the rMC-1 cell lysate, even in the presence of NADP (Fig. 6E) . The characteristics of retinol dehydrogenase activity from rMC-1 cells (i.e., presence in the microsomal fraction and cofactor preference) are identical with the activity attributed to that of recombinant RDH10 expressed in COS cells. ¹⁴  

Discussion

Retinal Müller cells are suggested to play a role in retinoid metabolism. ^{2 3} In the present study, we demonstrated that RDH10, an all-trans retinol dehydrogenase, is expressed in retinal Müller cells in addition to the RPE. This expression pattern is identical with that of the photoisomerase RGR. ²⁰  

More than ten enzymes in the SDR family have been identified in the retina and RPE. ^{8 9 14 21 22 23} Among them, retSDR1 and prRDH are located in the photoreceptor outer segments and may play roles in reducing all-trans retinal to all-trans retinol. ^{21 23} RDH5, RDH10, and RDH11 are expressed in the RPE. RDH5 is a specific cis-RDH and is found predominantly in the RPE. ⁸ RDH11 has recently been found to be expressed in the RPE and Müller cells. ⁹ In in vitro assays, RDH11 displayed both cis-RDH and atRDH activities. However, the cis-RDH activity of RDH11 may be more important because RDH11 interacts with RDH5. RPE RDH11 has been suggested to be responsible for the remaining cis-RDH activity in the RPE of RDH5 knockout mice. ⁹ Moreover, the atRDH activity of RDH11 was inhibited by CRBP, an all-trans retinol binding protein, which is abundantly expressed in the RPE. Therefore, it is unlikely that RDH11 is a major contributor to the atRDH activity in the RPE. 

RDH10 is a specific all-trans retinol dehydrogenase identified in the RPE and now in the Müller cells. ¹⁴ We propose that one of the possible physiological functions of RDH10 is to synthesize the all-trans retinal chromophore for RGR and thus to act as a functional partner of RGR in the photic visual cycle. As RGR is known to be expressed in both the RPE and Müller cells, it was necessary to determine whether RDH10 co-expresses with RGR in both the RPE and Müller cells. Western blot analysis demonstrated that RDH10 is expressed at a high level in the RPE and a lower level in the retina (Fig. 2) . The anti-RDH10 antibody stained a cell type in the inner retina. The morphology of the stained cells and their location in the retina are consistent with those of Müller cells. ¹⁶ To further confirm the observation, we used rMC-1 cells, a well-established and characterized retinal Müller cell line. ¹⁹ RDH10 expression in this cell line was detected at both the mRNA and protein levels (Fig. 4) . Moreover, we identified the RDH10 activity in rMC-1 cells (Fig. 6) . The subcellular localization and co-factor preference of the atRDH activity in the rMC-1 cells was consistent with those of the recombinant RDH10 expressed in COS cells. ¹⁴ The RDH10 cDNA was cloned from rMC-1 cells, and the rat RDH10 sequence isolated from rMC-1 cells shared high sequence homology with that from human, bovine, and mouse RDH10. The rat RDH10 amino acid sequence differed from the human and bovine RDH10 only by two amino acids. Similar to RDH10 cloned from other species, the rat RDH10 contained a hydrophobic domain and is likely a trans-membrane protein. Sequence conservation across species underscores its functional significance. Taken together, these studies indicate that a functional RDH10 is expressed in retinal Müller cells. 

RDH10 expression levels in Müller cells are lower than those in the RPE, which is consistent with the pattern of RGR expression. ²⁰ As RDH10 is capable of converting all-trans retinol into all-trans retinal, and thus provide the substrate for RGR in the photoisomerization in Müller cells, Müller cells should be able to generate 11-cis retinal from all-trans retinol in the light. 

RDH10 displayed a strain difference in mice in its expression level in the RPE (i.e., higher in BALB/c and lower in C57Bl/6). This strain difference paralleled the expression levels of RGR and RPE65 in these strains. Previous studies ^{24 25} show that BALB/c mice have higher RPE65 protein levels, which correlates with their higher isomerase activity compared to C57Bl/6 mice. These differences have been suggested as a reason to explain why BALB/c mice are more susceptible to light damage. ²⁴ The present study demonstrated that the photic cycle proteins (i.e., RGR and RDH10) also had higher levels in BALB/c than in C57Bl/6 mice. The higher photoisomerase activity may be yet another factor contributing to the higher susceptibility to light damage in BALB/c mice. Noteworthy, we found that RPE65 levels in the Rgr ^−/− mouse eyecup are approximately 50% of that in the wild-type C57Bl/6 mice (Fig. 1) . This finding is different from the result reported in a recent publication, ²⁶ which showed similar RPE65 levels in Rgr ^−/− and C57Bl/6 mice. A possible explanation for the disparity between our result and that of Maeda and co-workers ²⁶ is the condition for raising animals. In the report by Maeda and co-workers, ²⁶ “all animals were maintained in complete darkness and all manipulations were carried out under dim red light.” In contrast, we reared the mice in normal 12-hour light/dark cycles. Whether the lower expression level of RPE65 is directly related to the RGR gene disruption or is a nonspecific change in Rgr ^−/− mice remains unclear at the present time. 

Accumulating evidence ³ shows that the retinal Müller cell may be an important source of 11-cis chromophore for cone photoreceptors. As the maximal regeneration rate of cone pigments is approximately 2000× higher than that of rhodopsin, ^{27 28} cones may need a more efficient supply system of 11-cis chromophore. Furthermore, as cones have shorter outer segments than rods, cone outer segments are more distal from the RPE, which may impede the transfer of the 11-cis chromophore. Thus, to be functional under light conditions, cones may need an alternative system to supply 11-cis chromophore in addition to the RPE pathway. As Müller cells may generate 11-cis retinal under intense light, this photic visual cycle in Müller cells may serve as an alternative 11-cis chromophore supply for cones to meet the high demands for the continuous supply of 11-cis retinal. 

 Supported by Grants EY12231, EY015650, EY13520, EY14793 and EY04939 from the National Institutes of Health; a grant from Foundation Fighting Blindness, Owings Mills, MD; and an unrestricted grant to MUSC-SEI from Research to Prevent Blindness, Inc., New York, New York. RKC is a RBP Senior Scientific Investigator.

 Submitted for publication December 1, 2003; revised February 24, 2004; accepted March 2, 2004.

 Disclosure: B.X. Wu, None; G. Moiseyev, None; Y. Chen, None; B. Rohrer, None; R.K. Crouch, None; J.-X. Ma, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

RDH10 expression levels in the eyecups from different strains of mice. Eyecups containing the retinas were dissected and homogenized separately from three mice of each strain. Equal amounts (50 μg) of total proteins from each mouse were analyzed by Western blot analysis using the antibodies specific for RDH10, RGR, RPE65, and β-actin. (A) A representative Western blot result from one mouse of each strain. (B-D) Densitometry analysis using the Quantity One software from Bio-Rad Laboratories. Values are average densities (mean ± SEM, n = 3) and expressed as percentages of the average values in BALB/c. (E) Western blot analysis of RDH10 expression in the livers of C57Bl/6 and BALB/c mice.

Figure 2.

 

Distribution of the RDH10 protein in bovine tissues. The same amount of total proteins (50 μg) from various bovine tissues was loaded for Western blot analysis using the anti-RDH10 antibody (1:2000 dilution). Lane 1, RPE; lane 2, retina; lane 3, brain; lane 4, lung; lane 5, liver; lane 6, kidney; lane 7, pancreas; lane 8, skeletal muscle.

Figure 3.

 

Immunohistochemical localization of RDH10 in the retina. Immunolabeling of retinal sections using the anti-RDH10 antibody showed intense signals in the RPE and weaker signal in Müller cells. GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium.

Figure 4.

 

Detection of RDH10 mRNA and protein in rMC-1 cells. (A) RT-PCR analysis: RNAs from the rMC-1 cells and mouse eyecups (without the retina) were used as templates for RT-PCR using RDH10-specific primers. PCR amplified a specific band (150 bp) from rMC-1 cells and from mouse eyecups. (B) Western blot analysis: The anti-RDH10 antibody (1:2000 dilution) recognized a 39-kDa band in both the rMC-1 cells and mouse eyecup.

Figure 5.

 

Sequence analysis of rat RDH10. (A) Amino acid sequence alignment of rat (r), human (h), bovine (b), and mouse (m) RDH10. The deduced amino acid sequence of rat RDH10 was aligned with the human, bovine, and mouse RDH10 sequences using the Wisconsin Genetic Computer Group (GCG) program. Rat RDH10 was used as the template and only the amino acids different from the rat RDH10 are indicated in human, bovine, and mouse sequences. Note that bovine and human RDH10 have identical amino acid sequences. (B) Evolutionary relationships between RDH10 and other SDR family members. A phylogenetic tree of RDH10 in the mammalian SDR family was plotted by the GCG program using the Kimura protein distance correction. The length of each horizontal line in the tree is proportional to the difference of the amino acid sequences.

Figure 6.

 

All-trans retinol dehydrogenase activity in Müller cells. The all-trans retinol dehydrogenase (atRDH) activity was measured in microsomal fractions of rMC-1 cells (500 μg in 0.1 M phosphate buffer, pH 7.4) using 2 μCi all-trans [11,12-³H]-retinol (peak 2) as a substrate, with and without cofactors. (A) Only a trace amount of all-trans retinal was formed without the addition of an exogenous cofactor. (B) The addition of 400 μM NADP to the reaction resulted in the generation of a major peak of all-trans retinal (peak 1). (C) The addition of 400 μM NAD generated only a minor peak of all-trans retinal. (D) The same amount of NADP alone in the absence of cellular proteins did not generate any detectable all-trans retinal. (E) Only low atRDH activity was detected in the cytosolic fraction of rMC-1 cells when the same amount of NADP was added as a cofactor.

The authors thank Dr. Vijay Sarthy at the Northwestern University for kindly providing the rMC-1 cell line; Dr. Henry Fong at the University of Southern California for Rgr ^−/− mice and the anti-RGR antibody; and Dr. Michael Redmond at the National Eye Institute for the Rpe65 ^−/− mice.