A monogenic dominant mutation in Rom1 generated by N-ethyl-N-nitrosourea mutagenesis causes retinal degeneration in mice

Generation and phenotype-based screening of mice for N-ethyl-N-nitrosourea-induced mutants

We performed the animal studies under the guidelines issued by the RIKEN BioResource Center (Tsukuba, Japan) in the “Outline for conducting animal experiments” (issued August 1999, revised October 2001). We obtained stock mice from CLEA Japan, Inc. (Tokyo, Japan). The method used for mouse ENU mutagenesis is available at RIKEN BioResource Center ENU Mutants website and has been described previously [20-24]. At 8–10 weeks of age, C57BL/6J male mice were intraperitoneally injected with 85 mg/kg or 100 mg/kg bodyweight ENU (Sigma, Tokyo, Japan) twice at weekly intervals. The injected male mice were mated with DBA/2J female mice after a sterile period (approximately 10–11 weeks). Phenotypic screenings were routinely performed on the first generation (G1) progeny. Funduscopy was performed at 45 weeks of age. Phenodeviant individuals (mutant candidates) were assigned M numbers.

Funduscopic examination

Pupils were dilated with 0.5% tropicamide and 0.5% phenylephrine (Santen Pharmaceutical Co., Ltd., Osaka, Japan) diluted 5–7.5 fold with physiologic saline. The fundi were examined by indirect ophthalmoscopy, and photographs were taken by using a GENESIS hand funduscope camera (Kowa Co., Ltd., Nagoya, Japan) with a Volk 90D condensing lens (Volk Optical, Mentor, OH), as described [25-27]. In short, the camera was fitted to a dissection microscope base for stability and a foot pedal was used to operate the shutter. Photographs were taken using conscious mice. One hand was used to hold the mouse while two fingers of the other hand were used to retract the eyelids. We used the condensing lens mounted between the camera and eye. The mouse was held on its side on the microscope platform beneath the lens. Focusing was achieved by moving the mouse.

Inheritance testing and gene mapping

To test for phenotype transmission and to genetically map the causative gene, we backcrossed M-1156 with DBA/2J animals and recovered 125 backcross (N2) progeny. Because it was difficult to determine whether 16 of the 125 mice had abnormal fundi at 25 weeks of age, they were not used in the chromosome mapping analysis. Genomic DNA was prepared from the tail tips of the 109 remaining backcross progeny by using the NA-2000 automatic nucleic acid isolation system (KURABO, Osaka, Japan). We used the db-SNP website for single nucleotide polymorphisms (SNPs) and microsatellite markers listed on the Mouse Genome Informatics (MGI website) for simple sequence length polymorphisms. After examining 118 polymorphic markers spaced evenly throughout the genome by the TaqMan MGB assay (Applied Biosystems, Tokyo, Japan) and ABI 7700 and ABI 7900 sequence detection systems, five markers (D19Mit109, D19Mit42, D19Mit69, D19Mit110, and rs3023477) on chromosome 19 were chosen for fine mapping of the causative gene.

Search of the M-1156 genome for the causative mutation

We used Positional MEDLINE (PosMed), a semantic Web system, to provide a list of ranked candidate genes within the region [28,29] to search for the causative mutation in the mapped chromosomal region. We designed primer pairs based on the genomic sequence of Rom1 (NC_000085) and sequenced the region from –110 bases upstream to 1,889 bases downstream of the Rom1 gene to search for mutations of the M-1156 genome.

Genotyping of the M-1156 Rom1 allele

The Rom1^Rgsc1156 allele was genotyped by allele-specific primer-PCR methods. The wild-type-specific primer (5′-AAG TAA CGG TTG CTG ACC GA-3′) or the Rgsc1156-specific primer (5′-AAG TAA CGG TTG CTG ACC GG-3′) was paired with a common opposite-strand primer (5′-GCT TTG CCA GTA AGT CTC AA-3′).

Electroretinogram

Single-flash electroretinograms (ERGs) of both eyes of Rom1^Rgsc1156 mutant mice were obtained at 7, 35, and 50 weeks of age. Mice were dark adapted overnight and then prepared for recording under dim red light. The mice were anesthetized with an intraperitoneal injection of physiologic saline containing ketamine (80 mg/kg; Daiichi Sankyo Propharma Co., Ltd., Tokyo, Japan) and xylazine (16 mg/kg; Bayer Medical Ltd., Tokyo, Japan). Their pupils were dilated with 0.5% tropicamide and 0.5% phenylephrine, and their corneas were anesthetized with 0.4% oxybuprocaine hydrochloride (Santen Pharmaceutical Co., Ltd.) to prevent blinking. Mice were placed prone on a temperature-regulated heating pad throughout the recording session. After coating the cornea with drops of 1.5% hydroxyethylcellulose gel (Senju Pharmaceutical Co., Ltd., Osaka, Japan), a 2.8-mm diameter customized white-light-emitting diode (LED) electrode (Mayo Corporation, Ltd., Inazawa, Japan) was placed on the surface of the cornea with a micromanipulator and used as a recording electrode. A stainless-steel needle was attached subcutaneously to the nasal bone as a reference electrode, and another stainless-steel needle was inserted subcutaneously near the tail as a ground electrode. The flash stimulus was controlled by a white LED luminescent device (LS-W: Mayo Corporation, Ltd.). Electrical signals were amplified with a Dual Bio Amp (Model ML135; ADInstruments Pty. Ltd., New South Wales, Australia) at a gain of 4,000 and bandpass filtered between 0.3 and 500 Hz (–3 dB points). Signals were digitized at a rate of 10 kHz with a data acquisition device (PowerLab2/25, Model ML825; ADInstruments Pty. Ltd.). Stimulus presentation and data acquisition were programmed and performed with data recording and analysis software (Scope v3.7; ADInstruments Pty. Ltd.). Flash stimuli were 5 ms in duration and had an intensity of 1,000 cd/m² measured at the eye. A total of five responses obtained at 1–min intervals were averaged for each eye. The amplitude of the a-wave was measured from the prestimulus baseline to the a-wave trough. The amplitude of the b-wave was measured from the a-wave trough to the peak of the b-wave or, if the a-wave was not measurable, from the prestimulus baseline.

Photopic ERGs were obtained from single eyes of Rom1^Rgsc1156 mutant mice at 7 weeks of age. Light-adapted responses were recorded with white flashes (5 ms duration and 1,000 cd/m² intensity) on the rod-saturating background (25.1 cd/ m2) after 10 min of exposure to the background light to allow complete light adaptation. A total of 16 responses obtained at 1-s intervals were computer averaged for each eye.

Light microscopy

Anesthetized mice with an intraperitoneal injection of physiologic saline containing pentobarbital sodium (90 mg/kg; Kyoritsu Seiyaku Co., Ltd., Tokyo, Japan) were intracardially perfused with 4% paraformaldehyde (Nacalai Tesque, Inc., Kyoto, Japan) /2.5% glutaraldehyde (Wako Pure Chemical Industries, Ltd., Osaka, Japan) in 0.1 M phosphate buffer (PB). Eyes were enucleated and fixed in fresh fixing buffer for 1 day at room temperature and then dehydrated by immersion in a series of increasing ethanol concentrations. After three changes of xylene, the eyes were embedded in paraffin wax. A mid-horizontal series of sections (3-µm thick) through the optic nerve head of each eye was prepared. The sections were stained with hematoxylin and eosin (Muto Pure Chemicals Co., Ltd., Tokyo, Japan) by the standard protocol and viewed with an AX80 microscope (Olympus Inc. Tokyo, Japan). The thickness of the outer nuclear layer and inner nuclear layer and the length of the outer segment of the photoreceptors were measured by light microscopy at the point where the ratio of the distance from the optic disc to the point to the distance from the optic disc to peripheral end of the retina was 0.2, as described elsewhere [30].

Transmission electron microscopy

Anesthetized mice were intracardially perfused with 2% paraformaldehyde/2.5% glutaraldehyde in 0.1 M PB. Eyes were enucleated and fixed in fresh fixing buffer for 1 day. The anterior segments of the eyes were removed, and the eyecups were postfixed for another day. After rinsing the eyecups, eyecups were immersed for 1 day in 0.1 M PB at 4 °C. The eyecups were osmicated for 2 h in 1% aqueous solution of osmium tetroxide (TAAB Laboratories Equipment Ltd., Aldermaston, Berks, England) at 4 °C and then dehydrated through ascending ethanol concentrations (50%, 70%, 80%, 90%, and 100% for 30 min each). After passing the specimens through propylene oxide (Wako Pure Chemical Industries, Ltd., Osaka, Japan) twice for 1 h each time, the specimens were embedded in pure Epon (TAAB Laboratories Equipment Ltd.) for 2 days at 60 °C. Ultrathin sections were cut with a microtome (U-5; LBK Instruments Ltd., Bromma, Sweden), and the sections (80–100 nm) were stained with saturated uranyl acetate and Reynolds lead citrate (Wako Pure Chemical Industries, Ltd.). The sections were examined with an electron microscope at 80 kV (JEM-100S; JEOL, Tokyo, Japan).

Preparation of anti-Rom1 and anti-Prph2 monoclonal antibodies

A synthetic 13-mer for Rom1 (LAHYKDTEVPGRC) and a 20-mer for Prph2 (CKSNQVEAEGADAGPAPEAG), corresponding to amino acids 141–153 in mouse Rom1 and 328–346 in mouse Prph2, respectively, were purchased from Invitrogen (Carlsbad, CA). The peptides were conjugated with keyhole limpet hemocyanin (KLH; Pierce, Rockford, IL) and injected into BALB/ cAJcl mice four times at 2-week intervals. Three days after the final booster shot, lymph node cells were fused with myeloma line P3U1 cells. The cells were cultured in GIT medium (Nihon Pharmaceutical Co. Ltd., Tokyo, Japan) for 14 days, and the culture supernatants were screened first by Enzyme-Linked ImmunoSorbent Assay (ELISA) against immunized peptides. Then, the positive samples were subjected to western blot analysis for extracts from Cos-7 cells overexpressing either Rom1 or Prph2. Hybridoma cells from the positive wells were cloned by the standard limiting dilution technique [31]. Anti-Rom1 (clones 1–38) and anti-Prph2 (clones 7–27) were used in this study.

Immunohistochemistry

Anesthetized mice were intracardially perfused with 4% paraformaldehyde in ice-cold 0.1 M PB, and the eyes were enucleated and fixed for 1 day at 4 °C. The anterior segments were removed, and the eyecups were immersed in a graded series of sucrose solutions (10% W/V, 20%, and 30%) in PBS (8.10 mM NaHPO₄, 1.47 mM KHPO₄, 137 mM sodium chloride, 2.68 mM KCl, pH 7.4) at 4 °C for 1 day per step; the eyecups were then embedded in Tissue-Tek OCT Compound (Sakura Finetek, Tokyo, Japan) and frozen in acetone at dry-ice temperature. A mid-horizontal series of 10-µm-thick sections through the optic nerve head of each eye was prepared, and the sections were allowed to dry in air. After incubation in 0.3% hydrogen peroxide (H₂O₂) for 10 min and rinsing in PBS for 5 min, the sections were blocked for 30 min with 1% bovine serum albumin (BSA) in PBS and washed three times with PBS for 5 min each time. The sections were then incubated for 16 h at 4 °C with Rom1 or Prph2 monoclonal antibody diluted 1:400 in 1% BSA/PBS. After washing three times in PBS for 5 min each time, the sections were treated with the Envision kit (DAKO, Tokyo, Japan) as follows: The sections were incubated for 16 h at 4 °C with a secondary goat-antimouse Immunoglobulin G (IgG; DAKO) conjugated to horseradish peroxidase (HRP, DAKO). After washing three times in PBS for 5 min each time, the sections were developed with 0.03% 3,3-diaminobenzidine tetrahydrochloride (DAKO) solution containing 0.005% H₂O₂. After visualization, sections were washed in water for 5 min and counterstained with hemotoxylin for 10 s. After washing in water for 10 min, the sections were dehydrated with ethanol and penetrated with xylene. The sections were examined with an AX80 microscope (Olympus Inc.). Peptide absorption tests of both monoclonal antibodies were performed.

Real-time quantitative PCR

Retinal expression of Rom1 and Prph2 mRNA was measured by real-time quantitative PCR (qPCR). The eyes of 8-week-old mice were enucleated, and the retinas were dissected from the eyecups in ice-cold RNAlater (Product number No.7020; Applied Biosystems). Total RNA was extracted from the retinas with TRIzol (Invitrogen). Real-time qPCR was performed by using the SYBR® Reverse-Transcriptase (RT)–PCR kit (product code RR045A; TaKaRa, Shiga, Japan), according to the manufacturer’s protocol. Briefly, single-stranded cDNA was synthesized from each RNA sample with Rnase H free MMLV RTase (Takara) and random 6-mers on a PCR Thermal Cycler GP (TaKaRa). The cDNA obtained was amplified on a Smart Cycler II rapid DNA amplification system (Cepheid, TaKaRa) with a SYBR® RT–PCR kit, which contains SYBR Premix Ex Taq, SYBR Green I, and deoxyribonucleotide triphosphates (dNTPs). The PCR reaction was run under the following conditions: denaturation at 95 °C for 10 s, followed by 45 cycles of 95 °C for 5 s, and 60 °C for 20 s. The levels of glyceraldehyde-3-phosphate dehydrogenase (Gapdh) mRNA and rhodopsin (Rho) mRNA were used as endogenous controls for each sample. The mRNA levels in each sample were determined by using a calibration curve. Separate calibration (standard) curves for each gene were constructed by using serial dilutions of mRNA from a mouse retina of the same strain that had not been exposed to ENU. Primer sets for amplifying each gene were chosen from exons separated by an intron. The primer sets used in this analysis were: for Rom1, forward primer 5′-GCC ATG AAG TGC TGC TGG AA-3′, reverse primer 5′-GTC TGC AAA TAC CGC AAA CCA AG-3′; for Prph2, forward primer 5′-ACC ATC GAC ATG CTC CAG ATT G-3′, reverse primer 5′-CCC ATC CAC GTT GCT CTT GA-3′; for Gapdh, forward primer 5′-AAA TGG TGA AGG TCG GTG TG-3′, reverse primer 5′-TGA AGG GGT CGT TGA TGG-3′; and for Rho, forward primer 5′-TTG GGC CCA CAG GCT GTA A-3′, reverse primer 5′-CCG AAG CGG AAG TTG CTC A-3′.

Western blotting

Retinas were dissected from eyes collected from 3-week-old mice and homogenized by sonication in ice-cold buffer (20 mM Tris [pH. 7.6], 137 mM NaCl, 10% glycerol, 1% NP-40) containing protease inhibitor (1 mM phenylmethylsulfonyl fluoride; Sigma-Aldrich, Tokyo, Japan, 10 μg/ml aprotinin ; Sigma-Aldrich, Tokyo, Japan, 1 μg/ml leupeptin ; Peptide Institute, Inc., Osaka, Japan). After centrifugation at 10,000× g for 15 min at 4 °C, the supernatant was collected and protein concentrations were determined with a bicinchoninic acid (BCA) protein assay kit (Product number 23225; Thermo Fisher Scientific, Rockford, IL), with BSA used as a standard. For western blotting of Rom1 and Prph2, 10 µg of the retinal protein was mixed with the same volume of 2× loading buffer (140 mM Tris [pH. 6.8], 140 mM NaCl, 60% glycerol, 4% sodium dodecyl sulfate, 0.02% bromophenol blue, 5.0% β-mercaptoethanol) and boiled for 10 min. The prepared protein samples were electrophoresed by 12.5% sodium dodecyl sulfate–PAGE and transferred to Immobilon transfer membranes (polyvinylidene difluoride membrane, Millipore, Tokyo, Japan). The membranes were blocked with 3% BSA in Tris-Buffered Saline Tween-20 (TBST; 20 mM Tris [pH. 7.6], 137 mM NaCl, 0.1% Tween-20) for 60 min and incubated for 60 min with mouse Rom1 antibody diluted 1:50 or mouse Prph2 antibody diluted 1:30 in TBST. After washing with TBST three times for 10 min each, the membranes were incubated with goat antimouse IgG conjugated to HRP (Product No. 62–6320; ZYMED Laboratories, San Francisco, CA). After washing with TBST three times for 10 min each, the membranes were immersed in Immobilon Western Chemiluminescent HRP Substrate (Millipore), and then immediately exposed to an LAS-3000 imaging system (FUJIFILM, Tokyo, Japan) to visualize the immunoreactive signals. Signal intensities were densitometrically quantitated with the ImageJ 1.41 software program (National Institutes of Health, Bethesda, MD). For western blotting of Rho, 0.1µg of the retinal protein was mixed with the same volume of 2× loading buffer. Samples were electrophoresed and transferred to a membrane without boiling. Mouse antibovine rhodopsin monoclonal antibody (Product No. MAB5356, clone name Rho 1D4; Chemicon, Temecula, CA) was diluted 1: 4,000 in high-salt TBST (20 mM Tris [pH.7.6], 1.14 M NaCl, 0.1% Tween-20), and immunoreaction with Rho antibody was allowed to proceed under the same conditions as with the Rom1 and Prph2 antibodies in high-salt TBST. The band between 25 kDa and 37 kDa was measured as the rhodopsin signal. The protein level in each sample was standardized to the level of β-actin, which was detected with a human β-actin monoclonal antibody conjugated to HRP (Product No. ab20272; Cosmo Bio Co., Ltd., Tokyo, Japan).

Statistical analysis

The results were statistically analyzed by a one-way analysis of variance (ANOVA) and expressed as means±standard error when comparing groups. All multiple pairwise comparisons were analyzed by the Tukey–Kramer multiple comparisons test. A p value <0.05 was considered statistically significant.

Discovery of a dominant mutant in Rom1 generated by N-ethyl-N-nitrosourea mutagenesis

During large-scale screening of ENU-mutagenized mice, we used a funduscopic camera to screen 2,515 first generation (G1) male mice for dominant abnormal eye phenotypes. Identified phenodeviants, i.e., animals that exhibited abnormal phenotypes, were subsequently backcrossed to DBA/2J or C3H/HeJ mice for inheritance testing. The inheritance tests confirmed that some independent phenotypes were clearly transmitted to the next generation. A catalog of the independent phenotypes will be described by us in near future. Comparison of the appearance of the fundi of heterozygous M-1156 mice with the wild-type mice showed that they were abnormal, and a mottled retina and narrowed arteries were observed in the abnormal fundi (Figure 1). The mutation exhibited a dominant inheritance pattern. No other abnormalities were observed in the M-1156 mice.

We mapped the mutant locus to the proximal region of chromosome 19 by linkage analysis of backcrossed progeny (Figure 2A). The mutated locus was mapped between rs3023477 and D19Mit69, close to D19Mit42 (Figure 2B). Although according to the public genome database (ensembl) there are more than 400 genes within the critical region, according to the PosMed web system Rom1, Stx3, and Best1, which encode rod outer segment membrane protein 1, syntaxin 3, and bestrophin 1, respectively, were strong candidate genes for retinal degeneration. We used genomic DNA from the mutant line to sequence their complete coding and flanking intron regions to determine whether the mutant phenotype was caused by a genetic alteration in Rom1, Stx3, or Best1. To exclude the possibility that base changes were pre-existing polymorphisms, we compared the sequences of the mutant mice with those of the wild-type (DBA/2J and C57BL/6J) mice. In addition, to exclude the possibility that base changes in Prph2 contributed to the phenotype, we compared the sequences of the mutant mice with those of the wild-type (DBA/2J and C57BL/6J) mice. A single-base substitution, T→C, at position 1,195 of Rom1 (M96760, National Center for Biotechnology Information [NCBI]) was identified in the M-1156 mutant line (Figure 3), and the base change caused Trp to Arg amino acid substitution at position 182 of the translated protein. The base substitution was found in exon 1 of Rom1, and the amino acid substitution was located in the large intradiscal loop of Rom1. We designated the new Rom1 mutant allele Rom1^Rgsc1156 (MGI accession number 3814074). No other alterations or polymorphisms in the Rom1 coding region were found when mutant mice and wild-type mice were compared, nor were any sequence changes in Stx3, Best1, or Prph2 detected in the mutant mice in a comparison with wild-type mice.

Characterization of the phenotype of the Rom1^Rgsc1156 mutant

Careful examination of photomicrographs of the retinas of Rom1^Rgsc1156 heterozygotes and Rom1^Rgsc1156 homozygotes (n=3 or 4 each; Figure 4) revealed that the outer nuclear layer was thinner and the outer segments of the photoreceptors were shorter in the heterozygotes than in wild-type mice. The alterations were more marked in homozygotes than in heterozygotes, and they progressed from 7 weeks of age to 35 weeks of age. There were only one or two rows of nuclei in the outer nuclear layer of 35-week-old homozygotes, and the inner and outer segments were almost entirely absent. There were no significant differences among the genotypes in the thickness of the inner nuclear layer at any age examined, but there were significant differences in the thickness of the outer nuclear layer and the length of the outer segments of the photoreceptors (Figure 5).

Transmission electron microscopy was used to examine the ultrastructure of photoreceptor outer segments (n=3 each; Figure 6). The outer segments of the photoreceptors were shorter in the heterozygotes than in the wild-type mice. Only remnants of the outer segments of the photoreceptors were observed in the homozygotes at 35 weeks of age. High magnification revealed that the outer segments of the photoreceptors of the heterozygotes were clearly disorganized (the disc diameters varied and the discs were more compactly arranged than in wild-type mice).

Finally, we performed electroretinography to evaluate retinal function. Single-flash ERGs showed that the amplitudes of the a-wave in the mutant mice were significantly lower than in wild-type mice (Figure 7). The degree of reduction was more severe in the homozygotes than in the heterozygotes and was greater in 35-week-old mice than in 7-week-old mice. In contrast, there were no significant differences among the genotypes in the amplitude of the b-wave of the photopic electroretinograms (Figure 8).

Expression of Rom1 and Prph2 in the Rom1^Rgsc1156 mutant

To examine whether the T→C single-base substitution in exon1 induces splice site changes, we sequenced retinal cDNA from both wild-type mice and Rom1^Rgsc1156 homozygotes. The mutant allele was expressed, and there were no splice site changes in Rom1^Rgsc1156 mutants (data not shown).

We also quantified the expression levels of both Rom1 and Prph2 in wild-type mice and Rom1^Rgsc1156 mutants by performing real-time PCR on samples from 8-week-old mice. Normalization of expression of Rom1 to expression of Gapdh revealed that an expression level in Rom1^Rgsc1156 heterozygotes was 68.6±8.7% (mean±standard error) of the level in the wild type, whereas the expression level in homozygotes was 40.4±3.9% of the wild-type level (Figure 9A). Expression of Prph2 relative to Gapdh in heterozygotes and homozygotes was 66.9±6.9% and 41.0±3.3%, respectively, of the level in the wild type. Rom1 and Prph2 mRNA levels relative to Gapdh mRNA differed significantly among the genotypes. When expression of Rom1 was normalized to Rho, the expression level in Rom1^Rgsc1156 heterozygotes was 80.5±7.0% of the wild-type level, whereas in homozygotes the expression level was 93.5±6.2% of the wild-type level (Figure 9B). The relative expression of Prph2 to Rho in heterozygotes and homozygotes was 78.1±5.0% and 94.6±4.6%, respectively, of expression in the wild type. No significant differences in Rom1 or Prph2 mRNA levels relative to Rho mRNA were detected among the genotypes.

Production of Rom1 and Prph2 in the Rom1^Rgsc1156 mutant

To investigate the production of Rom1 and Prph2 in Rom1^Rgsc1156 mutants, we performed immunohistochemistry and western blotting on samples obtained from mice at 3 weeks of age when there was still little change in the thickness of the outer nuclear layer and the length of the inner and outer segments of the photoreceptors at the light microscopic level. Rom1 immunoreactivity in the retina was decreased in heterozygotes compared to wild-type mice (Figure 10B), and Rom1 immunoreactivity was markedly reduced in the homozygotes (Figure 10C). Prph2 immunoreactivity was similarly reduced in both the heterozygotes and homozygotes (Figure 10F,G).

Western blot analysis showed a decreased density of Rom1 and Prph2 from the Rom1^Rgsc1156 mutants (Figure 11A), and the density of Rho from the Rom1^Rgsc1156 mutants was also decreased (Figure 11B). Semiquantitative analysis by densitometry showed that Rom1 immunoreactivity in Rom1^Rgsc1156 heterozygotes and homozygotes was 51.6±8.2% and 2.8±0.3%, respectively, of that in wild-type mice (Figure 11C). The immunoreactivity of Prph2 from Rom1^Rgsc1156 heterozygotes and homozygotes was 59.3±12.4% and 17.4±3.6%, respectively, of the immunoreactivity of Prph2 from wild-type mice. The immunoreactivity of Rho from the Rom1^Rgsc1156 heterozygotes and homozygotes was 70.4±6.3% and 25.9±4.7%, respectively, of the immunoreactivity of the Rho from wild-type mice. The level of each protein differed significantly among the genotypes.

Retinal degeneration in Rom1^Rgsc1156 mutants

Rom1^Rgsc1156 mutants have a T→C single base substitution at position 1,195 of Rom1 (M96760, NCBI) and no mutations in Prph2. Sequence analysis of retinal cDNA around the mutation showed that the mutant allele was expressed and that the mutation did not alter the splice site. The results of the real-time PCR suggested that the decrease in Rom1 and Prph2 expression relative to Gapdh expression was attributable to the decrease in thickness of the outer nuclear layer because there were no significant differences in Rom1 or Prph2 mRNA levels among the genotypes after normalization to Rho mRNA. In short, since the total expression levels of Rom1, Prph2, and Rho in the mutant appear to be associated with the thickness of the outer nuclear layer, a protein with an Arg substituted for Trp at position 182 of the normal Rom1 protein should be produced.

Immunohistochemistry at 3 weeks of age showed that the immunoreactivity of Prph2 as well as Rom1 was reduced in Rom1^Rgsc1156 mutants, and western blot analysis demonstrated that the Rom1 and Prph2 levels in Rom1^Rgsc1156 mutants were significantly decreased. Interestingly, the Rho level was also decreased in Rom1^Rgsc1156 mutants at 3 weeks of age. A study of knockout and transgenic mice showed that retinal degeneration occurs when the combined level of both Rom1 and Prph2 is only about 60% of their combined level in wild-type mice [32]. Our semiquantitative analysis of western blots demonstrated that the combined level of both Rom1 and Prph2 was already below that percentage in Rom1^Rgsc1156 mutants at 3 weeks of age, suggesting that photoreceptor degeneration in Rom1^Rgsc1156 mutants occurred at least by 3 weeks of age. Why was Prph2 decreased in Rom1^Rgsc1156 mutants? Prph2 and Rom1 interact noncovalently via their large intradiscal loops to produce homotetrameric and heterotetrameric core complexes [33-35]. Since the Trp¹⁸² of Rom1 is localized in the intradiscal loop, it may induce unstable heterotetramers and homotetramers in Rom1^Rgsc1156 mutants. There is little Rom1 or Prph2 immunoreactivity in the inner segment of Rom1^Rgsc1156 mutants, and immunoreactivity for both proteins is decreased in the outer segment in a genotype-dependent manner, suggesting that mutant Rom1–Prph2 heterotetramers and mutant Rom1 homotetramers are degraded immediately after subunit assembly. Unfortunately, because of the small amount of mutant Rom1 in the homozygotes, our immunoprecipitation analysis failed to show that mutant Rom1 binds Prph2. In vitro biochemical analyses of subunit assembly may elucidate which core complexes are components in Rom1^Rgsc1156 mutants.

The ERGs showed significant differences among the genotypes in the amplitude of the a-wave in single-flash ERGs but not in the amplitude of the b-wave in photopic ERGs. These findings suggested that there was little effect on the function of cone photoreceptors in Rom1^Rgsc1156 mutants at 7 weeks of age but that there was significant deterioration of the function of rod photoreceptors in Rom1^Rgsc1156 mutants. Rod photoreceptor degeneration should mainly have occurred in Rom1^Rgsc1156 mutants because Rom1 is produced exclusively in rod photoreceptors.

All of the above findings taken together suggest that Trp182Arg mutant Rom1 decreases the levels of wild-type Prph2 and Rom1, which in turn results in a reduced level of Prph2-containing tetramers in the disc rim region and ultimately results in unstable disorganized outer segments and photoreceptor degeneration [34,36-38].

Phenotypic severity in Rom1^Rgsc1156 mutants

Light microscopy showed that the thickness of the outer nuclear layer in 7-week-old Rom1^Rgsc1156/+ mice was almost the same as in 2-month-old rds/+ mice [39]; however, the electron microscopic appearance of the outer segments was extremely different. The disc diameters of the Rom1^Rgsc1156/+ mice varied and their discs were compactly stacked, while the outer segments of the rds/+ mice were whorls [7,40]. The length of the outer segments in the rds/+ mice was much shorter than in the in Rom1^Rgsc1156/+ mice and their length in Rom1^+/− mice was almost the same as their length in wild-type mice. According to the previous results [32], the levels of Prph2 and Rom1 in Rom1^+/− mice were 106±4% and 58±2%, respectively, of their level in the wild type, whereas their levels in rds/+ mice were 26±2% and 67±2%, respectively. In Rom1^Rgsc1156/+ mice, the levels of Prph2 and Rom1 were 59.3±12.4% and 51.6±8.2%, respectively. Taken together, these results suggest that the phenotypic severity in outer segments is due to the amount of Prph2-containing tetramers, which are required for higher order disulfide-linked oligomer formation because the level of oligomers is important for stable photoreceptor disc formation [36,37].

Rom1^Rgsc1156 mutants can serve as a new animal model of human retinitis pigmentosa

Sequence alterations in human ROM1 associated with RP have been reported. The three ROM1 mutations (p.Gly80fs, p.Leu114fs, p.Gly113Glu) are manifested by digenic RP in combination with a Leu185Pro mutation in PRPH2, and only one of these mutations has an essentially normal phenotype [15,16]. The p.Gly80fs and p.Leu114fs mutations result in frameshifts early in the coding region and a downstream premature stop at codon 131, suggesting that these mutations do not encode a functional protein and therefore are null alleles. Production of the p.Gly113Glu Rom1 protein is 20-fold lower than production of wild-type Rom1 when expressed in COS-1 cells, indicating that a mutation resulting in p.Gly113Glu behaves as a null allele [37]. Biochemical studies have demonstrated that the p.Leu185Pro Prph2 mutant assembles with wild-type Prph2 and Rom1 to form core tetramers, but unlike wild-type Prph2 is unable to self-assemble into the core homotetramers that are required for higher order disulfide-linked oligomer formation [34,37]. Based on these findings, it seems likely that the pathogenesis of digenic RP is attributable to a large decrease in the amount of core complexes, such as Prph2 homotetramers and Prph2-Rom1 heterotetramers.

Two amino acid substitutions (p.Pro60Thr and p.Thr108Met) in a single allele of ROM1 in an autosomal dominant RP patient and three different single amino acid substitutions (p.Gly75Asp, p.Leu114fs, p.Arg242Glu) in patients with sporadic RP have been reported [41-43]. None of these substitutions has been concluded to be pathogenic because the pedigrees reported have been so small that statistically significant co-segregation analyses are limited. Loss of one allele of Rom1 in mice does not induce photoreceptor degeneration, implying that p.Leu114fs is unlikely to be pathogenic in humans [5].

In conclusion, the evidence is convincing that the Trp182Arg substitution in Rom1^Rgsc1156 mutants causes retinal degeneration, suggesting that mutations in the large intradiscal loop of Rom1 may affect the formation of homotetrameric and heterotetrameric core complexes and their stability. The Rom1^Rgsc1156 mutant will not only provide additional information about subunit assembly but will also serve as a new animal model of human autosomal dominant RP. The prevalence of nondigenic dominant ROM1 mutations is less than 1% [43,44], but Rom1 may play a more important role in stabilizing the core complex than previously expected.