August 2006
Volume 47, Issue 8
Free
Biochemistry and Molecular Biology  |   August 2006
Characterization of Rhodopsin P23H-Induced Retinal Degeneration in a Xenopus laevis Model of Retinitis Pigmentosa
Author Affiliations
  • Beatrice M. Tam
    From the Department of Ophthalmology and Visual Sciences, University of British Columbia, Vancouver, BC, Canada.
  • Orson L. Moritz
    From the Department of Ophthalmology and Visual Sciences, University of British Columbia, Vancouver, BC, Canada.
Investigative Ophthalmology & Visual Science August 2006, Vol.47, 3234-3241. doi:10.1167/iovs.06-0213
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Beatrice M. Tam, Orson L. Moritz; Characterization of Rhodopsin P23H-Induced Retinal Degeneration in a Xenopus laevis Model of Retinitis Pigmentosa. Invest. Ophthalmol. Vis. Sci. 2006;47(8):3234-3241. doi: 10.1167/iovs.06-0213.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. To investigate the pathogenic mechanisms that underlie retinal degeneration induced by the rhodopsin mutation P23H in a Xenopus laevis model of RP.

methods. Transgenic X. laevis were generated that expressed the rhodopsin mutants rhoP23H and rhoP23H/K29R (a variant incapable of transducin activation). Using quantitative dot blot assay, transgenic rhodopsin levels and the extent of retinal degeneration were determined. The contribution of rhodopsin signal transduction to cell death was assessed by comparison of rhoP23H and rhoP23H/K296R effects and by dark rearing of rhoP23H tadpoles. Intracellular localization and the oligomeric state of rhoP23H were determined by confocal immunofluorescence microscopy and Western blot analysis.

results. RhoP23H induced retinal degeneration in a dose-dependent manner whereas expression of a control rhodopsin did not, indicating that rod photoreceptor death was specific to the P23H mutation and was not caused by the overexpression of rhodopsin. Neither abolishment of rhoP23H photosensitivity and ability to activate transducin nor dark rearing rescued rod viability. RhoP23H was localized primarily to the endoplasmic reticulum (ER) of inner segments. Western blot analysis of transgenic retinas showed that rhoP23H was prone to form dimers and higher molecular weight oligomers. However, aggresomes were not observed in rhoP23H transgenic retinal sections, despite their being reported in cultured cells expressing rhoP23H.

conclusions. These results support a role for rhoP23H misfolding and inner segment accumulation in rod death, possibly by ER overload or other cellular stress pathways rather than by altered rhodopsin signal transduction or aggresome formation.

Retinitis pigmentosa (RP) is a heterogeneous group of inherited degenerative disorders characterized by the progressive loss of rod and cone photoreceptors, leading to blindness. More than 40 genes are linked to RP, including rhodopsin, a heptahelical G protein–coupled receptor that resides in a specialized photoreceptor organelle called the outer segment (OS). On photon capture, rhodopsin adopts an active conformation through isomerization of its chromophore and initiates the phototransduction cascade by the activation of transducin. 
More than 100 rhodopsin mutations are associated with RP and account for approximately 30% of autosomal-dominant cases. Rhodopsin mutations are classified based on their properties in vitro when expressed in cultured epithelial cells. 1 2 Class 1 mutants resemble wild-type rhodopsin in expression levels, fold correctly, and form functional photopigment. Class 2 mutants exhibit low expression and stability, variable ability to regenerate chromophore, and inefficient transport to the plasma membrane. Class 2 mutants also tend to misfold and are retained in the ER. The class 2 mutant P23H (rhoP23H) is the most prevalent cause of RP in North America. 
The pathogenic mechanisms underlying rhoP23H toxicity in vivo remain unclear. Several studies suggest that for rhoP23H, a toxic gain of function may be associated with visual transduction because rhoP23H trafficks normally to the OS 3 4 5 6 and aberrantly to the synaptic layer in transgenic rodents expressing mutant murine or human rhodopsin, 4 5 humans and transgenic mice expressing rhoP23H display prolonged dark adaptation, 7 8 and retinas of P23H transgenic mice are more susceptible to light damage than those of controls. 8 9 However, other studies suggest that rhoP23H misfolding is the major cause of cell death because expression of rhoP23H in cultured cells impairs/overloads normal protein degradation pathways and results in the formation of aggresomes, 10 11 rhoP23H is unable to rescue OS formation when expressed in rhodopsin knockout mice, 12 and partially folded rhoP23H may disrupt disk morphogenesis. 12 13  
In this study, we used transgenic Xenopus laevis to address mechanisms of rhoP23H pathology in vivo. Transgenic X. laevis make an excellent experimental animal for modeling RP because of their ease of generation, large photoreceptors, and rod/cone ratio similar to that of humans. We compared retinal degeneration (RD) induced by rhoP23H alone and in combination with a mutation that abolishes the ability of rhodopsin to activate transducin, and we determined the effect of dark rearing on RD. Subcellular localization of rhoP23H was also examined. Our results support a role for rhoP23H misfolding and ER retention in rod death by a mechanism independent of rhodopsin visual transduction. This mechanism may proceed through ER overload or other cellular stress pathways without the involvement of aggresomes. 
Materials and Methods
Molecular Biology
X. laevis expression constructs were based on pXOP0.8-eGFP-N1. 14 Isolation and mutagenesis of the X. laevis rhodopsin cDNA were previously described. 14 The 1D4 epitope tag VSKTETSQVAPA was introduced by PCR. Double and triple mutants were constructed using a combination of standard subcloning procedures and additional rounds of site-directed mutagenesis. Completed mutant rhodopsin cDNAs were verified by DNA sequencing. In the final cloning step, the enhanced green fluorescent protein (eGFP) cDNA was removed and the expressed protein was not a GFP fusion. For integration into sperm genomic DNA, expression vectors were digested with FseI (New England Biolabs, Beverly, MA) and were purified (QIAquick Gel Extraction Kit; Qiagen, Valencia, CA). For expression in cell culture, the mutant rhodopsin cDNAs were cloned into the EcoRI/NotI sites of pMT3. 15  
Generation and Rearing of Transgenic X. laevis
Transgenic X. laevis were generated using the nuclear transplantation method of Kroll and Amaya, 16 modified as previously described. 14 17 Embryos were housed in 4-L tanks in an 18°C incubator on a 12-hour dark/12-hour light cycle. After 24 hours, embryos were treated with 18 μg/mL G418 (Invitrogen, Carlsbad, CA) for 120 hours, as previously described. 18 Dark-reared animals were housed in the same incubator in tanks wrapped in several layers of aluminum foil. At 14 days postfertilization (dpf) corresponding to developmental stage approximately 48, 19 normally developed X. laevis were killed, and one eye was fixed in 4% paraformaldehyde while the contralateral eye was solubilized as previously described. 14 These studies were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 
Cell Culture, Transfection, and Western Blot
HEK293S cells were cultured and transfected as previously described. 14 Samples from either transfected HEK293S cells or transgenic retinas were separated on a 12% acrylamide gel and transferred to Immobilon-P membrane (Millipore, Billerica, MA). Membranes were probed with a 1:10 dilution of mAb B630N (cell culture supernatant; gift of Paul A. Hargrave, University of Florida) or 1:5000 dilution of mAb 1D4 (Chemicon, Temecula, CA), followed by a 1:10,000 antibody dilution (IRDye800-labeled secondary antibody; Rockland, Gilbertsville, PA) and then imaged (Odyssey imaging system; Li-Cor Biosciences, Lincoln, NE). 
Dot Blots
Dot blots were prepared as previously described 14 and were imaged as described for Western blots. The ratio of mAb 1D4 (recognizes transgenic rhodopsin only) to mAb B630N (recognizes total rhodopsin) labeling was used to estimate the level of mutant rhodopsin expression in transgenic eyes, based on linear interpolation using ratios obtained from control samples. Plots of expression level versus log total rhodopsin content were fit to dose–response curves using graphics software (Sigma Plot; Systat, Richmond, CA). 
Immunohistochemistry and Confocal Microscopy
Fixed eyes were embedded and cryosectioned as previously described. 14 Frozen sections were labeled with 2B2 antibody (cell culture supernatant, gift of Robert S. Molday, University of British Columbia) at a 1:10 dilution, 514-18 (cell culture supernatant; gift of Paul A. Hargrave, University of Florida) at 1:10 dilution, or anti-calnexin at 1:50 dilution (Stressgen Biotechnologies, Victoria, BC, Canada) followed by 1:750 dilution of Cy3-conjugated secondary antibody (Jackson Immunoresearch, West Grove, PA). Sections were also counterstained with Alexa 488– conjugated wheat germ agglutinin (Molecular Probes, Eugene, OR) and Hoechst 33342 (Sigma-Aldrich, St. Louis, MO) as previously described. 17 Sections were imaged using a laser scanning confocal microscope (Zeiss 510; Carl Zeiss, Oberkochen, Germany). More than five animals were analyzed per construct. 
Results
Expression of rhoP23H Causes Retinal Degeneration in Transgenic X. laevis
To examine the ambiguities regarding rhoP23H localization and mechanisms of toxicity, we sought to generate an X. laevis model of rhoP23H-induced RP. We designed a modified control rhodopsin (c-rho) in which a conservative M13F mutation introduces an N-terminal binding site for mAb 2B2, and substitution of the X. laevis 12 C-terminal amino acids for the corresponding bovine sequence introduces a binding site for mAb 1D4 (Fig. 1C) . Neither mAb 2B2 nor mAb 1D4 recognizes wild-type endogenous rhodopsin. 14 Transgenic X. laevis were generated in which c-rho was expressed in rod photoreceptors under the control of the X. laevis opsin promoter. At 14 dpf, the retinas of these animals were examined by confocal microscopy of frozen sections. Retinas expressing c-rho (Figs. 2C 2D)were essentially identical to wild-type retinas (Figs. 2A 2B) ; rod OS were fully elaborated and closely packed together. Thus, neither the 1D4 tag nor the M13F substitution was inherently deleterious. In contrast, retinas expressing the P23H variant (c-rhoP23H) exhibited a range of retinal degenerations as seen by a deficiency of rods in the central retina (Figs. 2E 2F 2G 2H) . Generally, the OSs of c-rhoP23H retinas were either entirely missing or were shorter than those of the control retinas. Thus, the human RP mutation rhoP23H also induces RD in transgenic X. laevis
RhoP23H Activation of Transducin Is Not Required to Induce Rod Cell Death
In transgenic mice, rhoP23H has been localized to the photoreceptor synaptic layer, 4 suggesting a role for abnormal synaptic transmission in rhoP23H-induced rod death. Moreover, in cultured photoreceptors, mislocalized rhodopsin causes cell death by inappropriate activation of nonvisual G protein–coupled signaling cascades in the IS. 20 To test whether aberrant rhoP23H signaling was involved in RD, we introduced a mutation (K296R) that generates a rhodopsin that folds properly but is unable to bind 11-cis retinal and is constitutively inactive. 14 21 We expressed rhodopsins containing the P23H and K296R mutations singly and together as a double mutant. RD caused specifically by rhoP23H activation of G-protein signaling should be blocked by K296R when both mutations are present in the same molecule. However, at 14 dpf, transgenic retinas expressing c-rhoP23H/K296R (Figs. 3C 3D)exhibited rod death similar to that in retinas expressing c-rhoP23H (Figs. 2E 2F 2G 2H) . Transgenic retinas expressing c-rhoK296R (Figs. 3A 3B)did not exhibit significant RD when compared to c-rho or wild-type retinas (Figs. 2A 2B 2C 2D) . Thus, abolishing the ability of rhodopsin to activate transducin neither induced RD by itself nor rescued rods from P23H-induced cell death. 
Quantitative Analysis of rhoP23H-Induced Degeneration
To more rigorously establish the relationship between the transgenic rhodopsins and cell death, we applied quantitative analysis to the observed c-rhoP23H and c-rhoP23H/K296R-induced RD. The transgenic X. laevis model is highly amenable to a quantitative approach because many primary transgenics can be generated in a single experiment. Each primary transgenic is the product of a different integration event; therefore, our results represent a wide range of expression levels and minimize effects caused by genetic background or integration site. Transgenic tadpoles from all four groups (c-rho, c-rhoP23H, c-rhoK296R, c-rhoP23H/K296R) were killed at 14 dpf. Solubilized extracts from one eye were analyzed by dot blot (Figs. 4A 4B) . Control samples of WT X. laevis rhodopsin (from solubilized nontransgenic retinas) and c-rho (from transfected HEK293S cells) were included, enabling us to determine the relative expression levels of transgenic to endogenous rhodopsin. Strikingly, average expression levels of c-rhoP23H and c-rhoP23H/K296R (<1% of total rhodopsin, accounting for decreased rhodopsin levels from degeneration) were significantly lower than those of c-rho and c-rhoK296R (7.7% of total rhodopsin; P = 0.0000002, Mann-Whitney U test). In plots of expression level versus total rhodopsin (Fig. 4C) , increasing expression levels of c-rhoP23H and c-rhoP23H/K296R correlated with decreasing total rhodopsin content, indicative of RD (loss of rods resulting in less total rhodopsin). In contrast, c-rho and c-rhoK296R expression did not correlate with a decrease in total rhodopsin content. 
We performed Kruskal-Wallis nonparametric analysis on the rank-transformed values obtained for the B630N signals shown in Figure 4A(Tables 1 and 2) . Nonparametric tests were used because of the likelihood that the data would not be normally distributed. The difference between groups was highly significant (P = 0.0003). A subsequent multiple comparisons test 22 indicated that the total rhodopsin contents of c-rhoP23H and c-rhoP23H/K296R eyes were significantly lower than for c-rho and c-rhoK296R eyes. However, c-rhoP23H and c-rhoP23H/K296R eyes were not significantly different from each other, and there was no significant difference between c-rho and c-rhoK296R eyes. 
Overexpression of control rhodopsins (c-rho and c-rhoK296R) did not cause significant retinal degeneration. Furthermore, total rhodopsin did not increase significantly with increased transgene expression, suggesting that in X. laevis rods, rhodopsin mRNA levels are not limiting and increased transgene expression primarily alters the ratio of transgenic to endogenous rhodopsin. It is also possible that degeneration caused by overexpression may occur at later time points. 
Altogether, these results demonstrate that the rod death observed with c-rhoP23H exhibited a dose-dependent relationship specific to the P23H mutation and was not the result of rhodopsin overexpression. Furthermore, these data confirm that the toxic effects of P23H were not ameliorated by blocking the ability of c-rhoP23H to activate transducin. Quantitative results from the dot blot correlated very well with the qualitative histologic results obtained from frozen sections. 
Dark Rearing Does Not Rescue P23H-Induced Retinal Degeneration
Dark rearing transgenic mice expressing rhoP23H retards the rate of RD, suggesting that rhodopsin signal transduction may be implicated in P23H toxicity. 8 However, in transgenic P23H rats, there was no protective effect of dark rearing. 23 Moreover, in the present study, we were unable to rescue degeneration with K296R, indicating that direct signaling through the mutant rhodopsin was not involved in P23H-induced RD. To assess whether dark rearing confers indirect protective effects on rhoP23H retinas, we reared tadpoles expressing c-rhoP23H either in total darkness or in cyclic light (12 hours light/12 hours dark). At 3 dpf (before development of the retina), we transferred embryos to a lightproof container until 14 dpf, at which time we killed the tadpoles. Duplicate dot blots of solubilized retina samples were probed with mAb B630N (total rhodopsin) or mAb 1D4 (transgenic rhodopsin). Quantification of the dot blots is shown in Figure 5 . The total rhodopsin content of c-rhoP23H–expressing retinas was lower (approximately 30%) in dark-reared tadpoles than in their light-reared littermates. This is consistent with previous studies showing that dark-reared wild-type X. laevis exhibits a decrease in rhodopsin synthesis of approximately 30% compared with that reared in cyclic light. 14 24 Nonetheless, c-rhoP23H still induced a dose-dependent RD irrespective of the light-rearing conditions. Thus, in transgenic X. laevis, rhoP23H-induced RD is not rescued by direct prevention of rhoP23H rhodopsin signaling or by indirect physiologic effects associated with dark rearing. 
RhoP23H Is Retained in the Rod Inner Segment
We analyzed the distribution of c-rhoP23H in transgenic X. laevis rods. The large size of the frog rod IS and OS and the lack of multiple layers of nuclei allowed us to identify entire individual rod cells and to visualize fine subcellular detail at the light microscopic level. Frozen sections of transgenic retinas were labeled with mAb 2B2 (specific for the N terminus of the transgenic rhodopsins). Both c-rho (Figs. 6C 6D)and c-rhoK296R (Figs. 6E 6F)localized primarily to the rod OS as expected and to Golgi membranes of the IS. In contrast, both c-rhoP23H (Figs. 6G 6H)and c-rhoP23H/K296R (Figs. 6I 6J)localized predominantly to the rod IS; they exhibited a diffuse pattern of labeling throughout the IS that excluded the nucleus and Golgi membranes. We obtained similar results with mAb 1D4, which recognizes the C terminus of c-rhoP23H (data not shown). This distribution is entirely consistent with the distribution of ER in the X. laevis photoreceptor. Double labeling with 2B2 and anti-calnexin (an ER resident protein) show complete colocalization (Figs. 6K 6L 6M) . C-rhoP23H and c-rhoP23H/K296R were largely absent from the OS, though we occasionally observed narrow bands of c-rhoP23H in some OS (Fig. 6M) , demonstrating that although the rods were actively expressing the mutant rhodopsin (as indicated by mAb 2B2 labeling of the IS), little or no protein was able to escape the quality control mechanisms of the cell. Aggresomes (predicted to form adjacent to the microtubule organizing center) or other large aggregate-like structures were not observed. Thus, most P23H rhodopsin was retained and degraded in the ER, though a very small fraction escaped quality control and did traffic to the OS. 
Anti-rhodopsin labeling of X. laevis retinas was typified by intense labeling of the periphery/outside margin of the OS (Figs. 6A 6B) . Because of the extremely high rhodopsin concentration in the OS plasma membrane and disk membranes and the tight packing of disks within the OS, the antibody–antigen complex at the outside edge might have created a physical barrier that prevented further antibody access to the interior of the OS. 
RhoP23H Is More Prone to Form Dimers/Aggregates than WT Rhodopsin
Because class 2 (misfolding) mutants such as rhoP23H tend to form dimers and higher order oligomers in cultured cells, 2 we examined the oligomeric state of rhoP23H in transgenic retinas. We performed Western blot analysis on solubilized eye extracts from transgenic and wild-type dpf14 tadpoles. The blots were probed with either mAb 1D4 (to detect c-rhoP23H) or mAb B630N (to detect wild-type rhodopsin) (Figs. 7A 7B) . Wild-type rhodopsin was detected mainly as monomers, but some dimers, tetramers, and higher order complexes were also detected as is typically seen for this protein. c-Rho from a transgenic retina was indistinguishable from wild-type. In contrast, c-rhoP23H was detected only as dimers, trimers, or higher order multimers (Fig. 7B) . Similarly, c-rhoP23H expressed in transfected HEK293 cells was much more likely to form dimers and high molecular weight species than the control, c-rho. These results provide further evidence that c-rhoP23H exists in an improperly folded state in transgenic X. laevis retinas. 
Discussion
There are multiple and potentially conflicting reports concerning the fate of rhoP23H and the mechanisms by which it induces RD. 3 5 12 It is, therefore, crucial to determine the exact subcellular distribution of rhoP23H because localization largely defines the possible pathogenic mechanisms. The present study provides strong in vivo evidence demonstrating that rhoP23H is misfolded and retained in the IS. In transgenic mice, VPPrho (a modified murine rhodopsin that contains three N-terminal mutations, including P23H) expressed on a rhodopsin knockout background was unable to support OS formation and was localized exclusively to the IS. 12 However, this scenario does not mimic most human RP cases in which one normal rhodopsin gene is present. Localization of the mutant VPPrho was not determined in the presence of wild-type rhodopsin and OS. Moreover, in a separate report of transgenic mice carrying the same rhoVPP transgene, rhoVPP was localized to the OS when expressed on a rhodopsin+/+ background. 5 In this study of transgenic X. laevis, we have clearly demonstrated that in rods capable of supporting OS formation, rhoP23H localized predominantly to the IS in a pattern consistent with retention in the ER. Unlike c-rho, which trafficked principally to the OS, there was a distinct lack of c-rhoP23H or c-rhoP23H/K296R in the OS except in thin, sporadic bands. Our quantitative results also support the immunolocalization results (i.e., the low levels of c-rhoP23H reflect its absence from the OS). The observation that c-rhoP23H formed dimers, trimers, and higher molecular weight aggregates provides a third line of evidence supporting the accumulation of c-rhoP23H in a misfolded state in the IS. It is interesting that although wild-type rhodopsin and c-rho formed dimers and tetramers at low levels, trimers were not observed, which suggests a different mechanism of aggregation of P23H rhodopsin. 
Abolishing rhodopsin photoactivation did not rescue rods from P23H-induced degeneration, indicating that visual transduction functions of c-rhoP23H were not involved in pathogenesis. Although we detected only small amounts of c-rhoP23H in the OS, it introduced the possibility that this subpopulation could have been responsible for rod degeneration. Altered visual transduction properties have been observed in a transgenic mouse line that expresses VPPrho. 7 In addition, studies of primary retinal cultures suggest that rhodopsin that mislocalizes to the IS may be toxic because of inappropriate activation of adenylate cyclase. 20 Thus, c-rhoP23H could potentially activate IS signaling pathways. Our experiments involving the K296R mutation negated both these possibilities. Because the K296R mutation blocks retinal binding and G-protein activation, 14 21 it precludes the involvement of transducin and likely other G-proteins in P23H-induced degeneration. The inability to rescue photoreceptors expressing c-rhoP23H by dark rearing further supports this argument. Dark rearing partially rescues retinal degeneration in VPP mice 8 but not transgenic rhoP23H rats. 23 The protective effects seen in mice may represent a secondary effect associated with alterations in the physiology of dark-adapted murine retinas as opposed to direct effects of rhodopsin signaling. This effect may not be evolutionarily conserved. Because the P23H/K296R double mutant misfolds in vitro and is retained in the ER at very low levels, we did not assay its ability to activate transducin. However, there is no reason to suspect that P23H would counteract constitutive inactivation by K296R because P23 does not participate in the interaction between K296 and E113, which stabilizes the inactive conformation of rhadopson, or in the binding of chromophore. Furthermore, because of the lack of chromophore binding, P23H/K296R rhodopsin can adopt only a constitutively active or inactive conformation. If constitutively active, it would be inactivated in vivo through arrestin binding, as has been demonstrated for K296E. 25  
IS accumulation of c-rhoP23H strongly indicates that the observed RD involved disruption of the ER in some capacity, possibly by ER stress and the unfolded protein response (UPR). The UPR controls transcriptional regulation of target genes resulting in decreased total protein synthesis to slow the accumulation of misfolded proteins, upregulation of chaperones, and upregulation of proteins required for ER-associated degradation (ERAD) in which nonnative polypeptides are degraded by cytosolic proteasomes. ER stress responses to unfolded proteins can include upregulation of the proapoptotic transcription factor CHOP and activation of ER-resident caspase-12. Induction of the proinflammatory transcription factor NF-κB as part of the ER overload response may also be involved, as has been observed for ΔF508-CFTR. 26 27  
RhoP23H expressed in epithelial cells is a substrate for the ubiquitin-proteasome system, and eventually leads to its impairment. 10 11 Overloading the proteasome results in the transport of small, undegraded protein aggregates to the microtubule organizing center (MTOC), where they coalesce to form large complexes known as aggresomes. 10 28 It is controversial whether aggresome formation is a toxic 29 or a protective 30 31 32 mechanism. Regardless, we did not observe aggresomes irrespective of rhoP23H expression levels or the state of RD. The MTOC of photoreceptors deals with extremely high volumes of vesicular traffic (destined for the OS), and photoreceptors might have developed mechanisms to restrict aggresome formation. Furthermore, aggresomes are typically associated with intermediate filaments, which have not been observed in photoreceptors. Our results indicate that aggresomes are not required for rhoP23H toxicity. 
Alternatively, the accumulation of rhoP23H in the ER may interfere with wild-type rhodopsin synthesis. RhoP23H may saturate components of protein biosynthesis/quality control (e.g., calnexin, BiP, or the sec61 translocon), sequestering them from normal rhodopsin or other essential proteins. In transgenic mice, knockout of one rhodopsin allele supports OS formation but induces slow RD, whereas loss of both alleles abolishes OS formation and causes rapid and severe degeneration. 33 Thus, heterozygsity for rhoP23H (i.e., loss of half the normal complement of rhodopsin) combined with diminution of wild-type rhodopsin biosynthesis 34 may represent an intermediate state between heterozygous and homozygous null. Disruption of wild-type rhodopsin biosynthesis is not mutually exclusive from (and may occur in conjunction with) ER stress. 
It is important to consider the discrepancies between this report and previous studies of transgenic mice in which rhoP23H is reported predominantly in the OS. 3 5 One possible reason for the discrepancies is that rhoP23H may fold more efficiently in mice than in frogs. The stringency of quality control mechanisms or the availability of chaperones may vary between species. Furthermore, because rhodopsins from different species (murine, human, and frog) have been studied, the results may reflect differences in primary sequences and thus in stability. Yet another possibility is the effect of lower temperature, which could influence the kinetics of rhodopsin folding, though low temperature could also promote folding and prevent denaturation associated with thermal instability. Nonetheless, we propose that misfolded rhoP23H is retained in the IS of transgenic mice at very low levels. Antibodies used in the transgenic mouse studies might not have been sufficiently sensitive to detect this population. 3 5 Furthermore, some localization studies used antibodies that cannot distinguish mutant from wild-type rhodopsin. 12 13 However, because rhoP23H is associated with RD in all studies, it is likely that rhoP23H toxicity involves the same mechanisms regardless of animal model. 
The occasional narrow bands of c-rhoP23H detected in the OS are intriguing. If c-rhoP23H molecules randomly fold sufficiently to exit the ER, a low but uniform level would be present in the OS. Instead, the presence of distinct c-rhoP23H bands represents brief intervals of successful c-rhoP23H trafficking, suggesting transient alteration of the intacellular milieu favoring increased folding efficiency or decreased quality control. From a therapeutic standpoint, it may be important to determine what cellular conditions affect the ability of c-rhoP23H’s to “escape” ERAD. Our X. laevis model provides an excellent system for investigating this. 
Cultured photoreceptors lose their structural integrity and epithelial cells lack the unique features of photoreceptors; hence, the ability to investigate pathogenic mechanisms in vivo is crucial to the understanding of RP. Because the results derived from individual animal models are not always applicable to human disease conditions, it is important to accumulate data from as many sources as possible. Thus, the establishment of transgenic X. laevis as a model of RP provides another valuable tool for disease analysis and complements existing rodent models. The advantages of this system include excellent resolution of intracellular detail in light microscopy applications and the potential to generate and analyze large numbers of primary transgenic animals, allowing comparison of transgene effects. Given that photoreceptor structure, retinal organization, phototransduction pathways, and rhodopsin sequence are all highly conserved among vertebrates, this model is likely to be relevant to human disease, and comparison of differences between this and existing animal models is likely to shed light on disease mechanisms. Innumerable potential confounding effects exist in all animal models (e.g., temperature, cone density, size of eye, nocturnal behavior); therefore, comparisons that establish common features are very important. The present study provides strong in vivo evidence supporting protein misfolding, not aberrant rhoP23H visual transduction, as the major mechanism involved in rhoP23H-induced photoreceptor degeneration. Even low-level misfolding of rhoP23H in the human retina (or other animal models) could impact rod viability and should be considered a potential mechanism for pathogenicity. 
 
Figure 1.
 
Rod photoreceptor, rhodopsin, and transgenic rhodopsin sequence. (A) Depiction of the structure of a rod photoreceptor. The rod is a highly polarized neuronal cell consisting of a synaptic terminal (S), a cell body or an inner segment (IS) containing the cell’s biosynthetic machinery, and a specialized organelle, the outer segment (OS), which consists of a stack of flattened membranous sealed disks discontinuous from the plasma membrane. Rhodopsin is synthesized in the IS and is subsequently incorporated into the OS disks, where it participates in visual transduction. Loss of polarized rhodopsin localization is detrimental to the cell. (B) Representation of a wild-type rhodopsin molecule. Rhodopsin is a heptahelical transmembrane protein. Its N terminus contains two glycan moieties, and the C terminus contains two palmitoylated cysteines. Rhodopsin’s chromophore, 11-cis-retinal, is bound to the lysine at residue 296. The P23H mutation resides in the intradiskal/extracellular N terminus. (C) Engineered sequence alterations at the N and C terminus of the transgenic rhodopsins. Residue 13 and the extreme C terminus (boxed regions) were substituted with bovine sequences to introduce antibody epitopes 2B2 and 1D4, respectively. Numbers refer to the residue in the bovine sequence. Arrows indicate the amino acids that were mutated. Given that both these changes were conservative, neither was expected to affect the structure/function of the molecule.
Figure 1.
 
Rod photoreceptor, rhodopsin, and transgenic rhodopsin sequence. (A) Depiction of the structure of a rod photoreceptor. The rod is a highly polarized neuronal cell consisting of a synaptic terminal (S), a cell body or an inner segment (IS) containing the cell’s biosynthetic machinery, and a specialized organelle, the outer segment (OS), which consists of a stack of flattened membranous sealed disks discontinuous from the plasma membrane. Rhodopsin is synthesized in the IS and is subsequently incorporated into the OS disks, where it participates in visual transduction. Loss of polarized rhodopsin localization is detrimental to the cell. (B) Representation of a wild-type rhodopsin molecule. Rhodopsin is a heptahelical transmembrane protein. Its N terminus contains two glycan moieties, and the C terminus contains two palmitoylated cysteines. Rhodopsin’s chromophore, 11-cis-retinal, is bound to the lysine at residue 296. The P23H mutation resides in the intradiskal/extracellular N terminus. (C) Engineered sequence alterations at the N and C terminus of the transgenic rhodopsins. Residue 13 and the extreme C terminus (boxed regions) were substituted with bovine sequences to introduce antibody epitopes 2B2 and 1D4, respectively. Numbers refer to the residue in the bovine sequence. Arrows indicate the amino acids that were mutated. Given that both these changes were conservative, neither was expected to affect the structure/function of the molecule.
Figure 2.
 
Expression of c-rhoP23H causes retinal degeneration in transgenic X. laevis. Confocal micrographs merged with DIC images of dpf 14 (approximately stage 47/48) wild-type retinas (A, B) or transgenic retinas expressing c-rho (C, D) or c-rhoP23H (EH). Retinal sections were stained with Alexa fluor 488 wheat germ agglutinin to outline cell layers (red) and Hoechst 33342 nuclear stain (blue). Right: Enlargements of the central retina shown in the corresponding left panel. Rod OS from retinas expressing c-rho (D) are similar in length and packing density to wild-type retinas (B). However, retinas expressing c-rhoP23H exhibited varying degrees of retinal degeneration (F, H). Rod OS are short and intermittent (F) or almost entirely missing (H). Cones (asterisks) predominate in areas devoid of rods. Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C, E, G); 2 μM (B, D, F, H).
Figure 2.
 
Expression of c-rhoP23H causes retinal degeneration in transgenic X. laevis. Confocal micrographs merged with DIC images of dpf 14 (approximately stage 47/48) wild-type retinas (A, B) or transgenic retinas expressing c-rho (C, D) or c-rhoP23H (EH). Retinal sections were stained with Alexa fluor 488 wheat germ agglutinin to outline cell layers (red) and Hoechst 33342 nuclear stain (blue). Right: Enlargements of the central retina shown in the corresponding left panel. Rod OS from retinas expressing c-rho (D) are similar in length and packing density to wild-type retinas (B). However, retinas expressing c-rhoP23H exhibited varying degrees of retinal degeneration (F, H). Rod OS are short and intermittent (F) or almost entirely missing (H). Cones (asterisks) predominate in areas devoid of rods. Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C, E, G); 2 μM (B, D, F, H).
Figure 3.
 
The rhodopsin-inactivating mutation K296R does not rescue P23H-induced retinal degeneration. Confocal micrographs merged with DIC images of transgenic retinas expressing c-rhoK296R (A, B) or c-rhoP23H/K296R (C, D). Retinal sections were stained with Alexa fluor 488 wheat germ agglutinin to outline cell layers (red) and Hoechst 33342 nuclear stain (blue). (A, C) Enlargements of the central retina (B, D). Retinas expressing c-rhoK296R (A, B) are comparable to wild-type retinas (Figs. 2A 2B) . Rod OS are full length and closely apposed. However, in retinas expressing c-rhoP23H/K296R (C, D), there are significantly fewer and shorter rod OS. Cones (asterisks) predominate in areas devoid of rods. This pattern of retinal degeneration is similar to c-rhoP3H (Figs. 2E 2F 2G 2H) . Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C); 2 μM (B, D).
Figure 3.
 
The rhodopsin-inactivating mutation K296R does not rescue P23H-induced retinal degeneration. Confocal micrographs merged with DIC images of transgenic retinas expressing c-rhoK296R (A, B) or c-rhoP23H/K296R (C, D). Retinal sections were stained with Alexa fluor 488 wheat germ agglutinin to outline cell layers (red) and Hoechst 33342 nuclear stain (blue). (A, C) Enlargements of the central retina (B, D). Retinas expressing c-rhoK296R (A, B) are comparable to wild-type retinas (Figs. 2A 2B) . Rod OS are full length and closely apposed. However, in retinas expressing c-rhoP23H/K296R (C, D), there are significantly fewer and shorter rod OS. Cones (asterisks) predominate in areas devoid of rods. This pattern of retinal degeneration is similar to c-rhoP3H (Figs. 2E 2F 2G 2H) . Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C); 2 μM (B, D).
Figure 4.
 
Quantitative analysis of retinal degeneration induced by c-rhoP23H and c-rhoP23H/K296R. Solubilized transgenic eyes were blotted onto membranes and probed with antibodies that recognized total rhodopsin (mAb B630N; A) or only transgenic rhodopsin (mAb 1D4; B). Control samples contained 100% wild-type rhodopsin from wild-type retinas (WT) or 100% transgenic c-rho from transfected HEK293S cells (293). Low mAb B630N signals indicate low levels of rhodopsin and thus retinal degeneration. (C) Dose–response curve generated from quantification of the fluorescent signals from (A) and (B). Expression of control rhodopsins c-rho and c-rhoK296R did not result in significant loss of total rhodopsin at the expression levels observed. c-RhoP23H and c-rhoP23H/K296R caused similar rhodopsin loss. In c-rho and c-rhoK296R controls, variability in total rhodopsin primarily resulted from variation in the size of the eyes. This variability was greater than in wild-type animals and likely was associated with stresses from the transgenesis and selection procedures.
Figure 4.
 
Quantitative analysis of retinal degeneration induced by c-rhoP23H and c-rhoP23H/K296R. Solubilized transgenic eyes were blotted onto membranes and probed with antibodies that recognized total rhodopsin (mAb B630N; A) or only transgenic rhodopsin (mAb 1D4; B). Control samples contained 100% wild-type rhodopsin from wild-type retinas (WT) or 100% transgenic c-rho from transfected HEK293S cells (293). Low mAb B630N signals indicate low levels of rhodopsin and thus retinal degeneration. (C) Dose–response curve generated from quantification of the fluorescent signals from (A) and (B). Expression of control rhodopsins c-rho and c-rhoK296R did not result in significant loss of total rhodopsin at the expression levels observed. c-RhoP23H and c-rhoP23H/K296R caused similar rhodopsin loss. In c-rho and c-rhoK296R controls, variability in total rhodopsin primarily resulted from variation in the size of the eyes. This variability was greater than in wild-type animals and likely was associated with stresses from the transgenesis and selection procedures.
Table 1.
 
Kruskal-Wallis Analysis of Total Rhodopsin Levels
Table 1.
 
Kruskal-Wallis Analysis of Total Rhodopsin Levels
Group Transgene n Sum of Ranks Average Rank Average Total Rhodopsin
1 c-rho 22 1298 58.0 100
2 c-rhoK296R 22 1117 54.3 88.2
3 c-rhoP23H 22 797 38.6 67.6
4 c-rhoP23H/K296R 22 704 27.1 62.1
Table 2.
 
Multiple Comparisons Analysis of Total Rhodopsin Levels
Table 2.
 
Multiple Comparisons Analysis of Total Rhodopsin Levels
Comparison P Conclusion
Group 1 vs. 4 0.00024 1 > 4
Group 1 vs. 3 0.0017 1 > 3
Group 1 vs. 2 0.259 Not significantly different
Group 2 vs. 4 0.0091 2 > 4
Group 2 vs. 3 0.042 2 > 3
Group 3 vs. 4 0.550 Not significantly different
Figure 5.
 
Dark rearing of rhoP23H-expressing tadpoles did not rescue rod degeneration. Tadpoles expressing c-rhoP23H were raised in cyclic light (12 hours light/12 hours dark) or in total darkness. Duplicate dot blots of solubilized transgenic eyes were probed with mAb B630N or mAb 1D4, and the results were quantified and plotted. Dark rearing did not protect from rhoP23H-induced retinal degeneration compared with cyclic light rearing, suggesting that activation of the visual transduction cascade does not play a role in rhoP23H-induced retinal degeneration.
Figure 5.
 
Dark rearing of rhoP23H-expressing tadpoles did not rescue rod degeneration. Tadpoles expressing c-rhoP23H were raised in cyclic light (12 hours light/12 hours dark) or in total darkness. Duplicate dot blots of solubilized transgenic eyes were probed with mAb B630N or mAb 1D4, and the results were quantified and plotted. Dark rearing did not protect from rhoP23H-induced retinal degeneration compared with cyclic light rearing, suggesting that activation of the visual transduction cascade does not play a role in rhoP23H-induced retinal degeneration.
Figure 6.
 
RhoP23H is retained in the ER of rod IS. Confocal micrographs of transgenic tadpole retinas at 14 dpf. (AJ) Cryosections from fixed eyes were labeled with 514-18 (A, B) or 2B2 (CJ) (green) to visualize wild-type or transgenic rhodopsin, respectively. Post-Golgi membranes and nuclei were stained with wheat germ agglutinin (red) and Hoechst 33342 (blue), respectively. Wild-type retina (A, B) shows the normal distribution of rhodopsin in the OS and the Golgi membranes (arrows). The distribution of c-rho (C, D) and c-rhoK296R (E, F) resembles that of wild-type. However, c-rhoP23H (G, H) and c-rhoP23H/K296R (I, J) were both localized diffusely throughout the IS. (KM) Retina expressing c-rhoP23H double labeled with anticalnexin (red, K) and mAb 2B2 (green, L). Colocalization (yellow, M) indicates that the mutant protein was retained in the ER. Occasional narrow bands of rhoP23H were observed in the OS (arrowhead). Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C, E, G, I, K, L); 2 μM (B, D, F, H, J, M).
Figure 6.
 
RhoP23H is retained in the ER of rod IS. Confocal micrographs of transgenic tadpole retinas at 14 dpf. (AJ) Cryosections from fixed eyes were labeled with 514-18 (A, B) or 2B2 (CJ) (green) to visualize wild-type or transgenic rhodopsin, respectively. Post-Golgi membranes and nuclei were stained with wheat germ agglutinin (red) and Hoechst 33342 (blue), respectively. Wild-type retina (A, B) shows the normal distribution of rhodopsin in the OS and the Golgi membranes (arrows). The distribution of c-rho (C, D) and c-rhoK296R (E, F) resembles that of wild-type. However, c-rhoP23H (G, H) and c-rhoP23H/K296R (I, J) were both localized diffusely throughout the IS. (KM) Retina expressing c-rhoP23H double labeled with anticalnexin (red, K) and mAb 2B2 (green, L). Colocalization (yellow, M) indicates that the mutant protein was retained in the ER. Occasional narrow bands of rhoP23H were observed in the OS (arrowhead). Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C, E, G, I, K, L); 2 μM (B, D, F, H, J, M).
Figure 7.
 
RhoP23H forms high molecular weight aggregates when expressed in transgenic X. laevis retinas. Samples from solubilized wild-type retina (lane 1, WT), c-rho expressing transgenic retina (lane 2), c-rhoP23H-expressing transgenic retinas (lanes 514) and solubilized HEK293S cells transfected with c-rho (lane 3) or c-rhoP23H (lane 4) analyzed by Western blot. Duplicate blots were probed with mAb B630N (A) and mAb 1D4 (B). Wild-type rhodopsin appears primarily as a monomer (approximately 35 kDa) and, to a lesser extent, as dimers (approximately 70 kDa) and tetramers (approximately 120 kDa). mAb 1D4 exclusively recognizes epitope-tagged rhodopsins (B, lanes 214) but not wild-type X. laevis rhodopsin (B, lane 1). c-Rho expressed in transgenic retinas appeared identical with wild-type, whereas c-rhoP23H was present mostly as dimers (approximately 70 kDa) and trimers (approximately 90 kDa) regardless of expression level. c-Rho and c-rhoP23H expressed in HEK293S cells exhibited similar behavior. Percentages given represent the proportion of total rhodopsin that constituted transgenic rhodopsin.
Figure 7.
 
RhoP23H forms high molecular weight aggregates when expressed in transgenic X. laevis retinas. Samples from solubilized wild-type retina (lane 1, WT), c-rho expressing transgenic retina (lane 2), c-rhoP23H-expressing transgenic retinas (lanes 514) and solubilized HEK293S cells transfected with c-rho (lane 3) or c-rhoP23H (lane 4) analyzed by Western blot. Duplicate blots were probed with mAb B630N (A) and mAb 1D4 (B). Wild-type rhodopsin appears primarily as a monomer (approximately 35 kDa) and, to a lesser extent, as dimers (approximately 70 kDa) and tetramers (approximately 120 kDa). mAb 1D4 exclusively recognizes epitope-tagged rhodopsins (B, lanes 214) but not wild-type X. laevis rhodopsin (B, lane 1). c-Rho expressed in transgenic retinas appeared identical with wild-type, whereas c-rhoP23H was present mostly as dimers (approximately 70 kDa) and trimers (approximately 90 kDa) regardless of expression level. c-Rho and c-rhoP23H expressed in HEK293S cells exhibited similar behavior. Percentages given represent the proportion of total rhodopsin that constituted transgenic rhodopsin.
The authors thank Daniel D. Oprian for helpful discussions. 
SungCH, DavenportCM, NathansJ. Rhodopsin mutations responsible for autosomal dominant retinitis pigmentosa: clustering of functional classes along the polypeptide chain. J Biol Chem. 1993;268:26645–26649. [PubMed]
SungCH, SchneiderBG, AgarwalN, PapermasterDS, NathansJ. Functional heterogeneity of mutant rhodopsins responsible for autosomal dominant retinitis pigmentosa. Proc Natl Acad Sci USA. 1991;88:8840–8844. [CrossRef] [PubMed]
OlssonJE, GordonJW, PawlykBS, et al. Transgenic mice with a rhodopsin mutation (Pro23His): a mouse model of autosomal dominant retinitis pigmentosa. Neuron. 1992;9:815–830. [CrossRef] [PubMed]
RoofDJ, AdamianM, HayesA. Rhodopsin accumulation at abnormal sites in retinas of mice with a human P23H rhodopsin transgene. Invest Ophthalmol Vis Sci. 1994;35:4049–4062. [PubMed]
WuTH, TingTD, OkajimaTI, et al. Opsin localization and rhodopsin photochemistry in a transgenic mouse model of retinitis pigmentosa. Neuroscience. 1998;87:709–717. [CrossRef] [PubMed]
AblonczyZ, KnappDR, DarrowR, OrganisciakDT, CrouchRK. Mass spectrometric analysis of rhodopsin from light damaged rats. Mol Vis. 2000;6:109–115. [PubMed]
GotoY, PeacheyNS, RippsH, NaashMI. Functional abnormalities in transgenic mice expressing a mutant rhodopsin gene. Invest Ophthalmol Vis Sci. 1995;36:62–71. [PubMed]
NaashML, PeacheyNS, LiZY, et al. Light-induced acceleration of photoreceptor degeneration in transgenic mice expressing mutant rhodopsin. Invest Ophthalmol Vis Sci. 1996;37:775–782. [PubMed]
OrganisciakDT, DarrowRM, BarsalouL, KuttyRK, WiggertB. Susceptibility to retinal light damage in transgenic rats with rhodopsin mutations. Invest Ophthalmol Vis Sci. 2003;44:486–492. [CrossRef] [PubMed]
IllingME, RajanRS, BenceNF, KopitoRR. A rhodopsin mutant linked to autosomal dominant retinitis pigmentosa is prone to aggregate and interacts with the ubiquitin proteasome system. J Biol Chem. 2002;277:34150–34160. [CrossRef] [PubMed]
SalibaRS, MunroPM, LuthertPJ, CheethamME. The cellular fate of mutant rhodopsin: quality control, degradation and aggresome formation. J Cell Sci. 2002;115:2907–2918. [PubMed]
FrederickJM, KrasnoperovaNV, HoffmannK, et al. Mutant rhodopsin transgene expression on a null background. Invest Ophthalmol Vis Sci. 2001;42:826–833. [PubMed]
LiuX, WuTH, StoweS, et al. Defective phototransductive disk membrane morphogenesis in transgenic mice expressing opsin with a mutated N-terminal domain. J Cell Sci. 1997;110(part 20):2589–2597.
TamBM, XieG, OprianDD, MoritzOL. Mislocalized rhodopsin does not require activation to cause retinal degeneration and neurite outgrowth in Xenopus laevis. J Neurosci. 2006;26:203–209. [CrossRef] [PubMed]
FrankeRR, SakmarTP, OprianDD, KhoranaHG. A single amino acid substitution in rhodopsin (lysine 248-leucine) prevents activation of transducin. J Biol Chem. 1988;263:2119–2122. [PubMed]
KrollKL, AmayaE. Transgenic Xenopus embryos from sperm nuclear transplantations reveal FGF signaling requirements during gastrulation. Development. 1996;122:3173–3183. [PubMed]
MoritzOL, TamBM, KnoxBE, PapermasterDS. Fluorescent photoreceptors of transgenic Xenopus laevis imaged in vivo by two microscopy techniques. Invest Ophthalmol Vis Sci. 1999;40:3276–3280. [PubMed]
MoritzOL, BiddleKE, TamBM. Selection of transgenic Xenopus laevis using antibiotic resistance. Transgenic Res. 2002;11:315–319. [CrossRef] [PubMed]
NieuwkoopP, FaberJ. Normal Table of Xenopus laevis (Daudin): A Systematical and Chronological Survey of the Development from the Fertilized Egg till the End of Metamorphosis. 1994;Garland Publishing New York.
AlfinitoPD, Townes-AndersonE. Activation of mislocalized opsin kills rod cells: a novel mechanism for rod cell death in retinal disease. Proc Natl Acad Sci USA. 2002;99:5655–5660. [CrossRef] [PubMed]
CohenGB, YangT, RobinsonPR, OprianDD. Constitutive activation of opsin: influence of charge at position 134 and size at position 296. Biochemistry. 1993;32:6111–6115. [CrossRef] [PubMed]
ConoverW. Practical Nonparametric Statistics. 1999;John Wiley & Sons New York.
GreenES, MenzMD, LaVailMM, FlanneryJG. Characterization of rhodopsin mis-sorting and constitutive activation in a transgenic rat model of retinitis pigmentosa. Invest Ophthalmol Vis Sci. 2000;41:1546–1553. [PubMed]
BesharseJC, HollyfieldJG, RaybornME. Turnover of rod photoreceptor outer segments, II: membrane addition and loss in relationship to light. J Cell Biol. 1977;75:507–527. [CrossRef] [PubMed]
LiT, FransonWK, GordonJW, BersonEL, DryjaTP. Constitutive activation of phototransduction by K296E opsin is not a cause of photoreceptor degeneration. Proc Natl Acad Sci USA. 1995;92:3551–3555. [CrossRef] [PubMed]
KnorreA, WagnerM, SchaeferHE, ColledgeWH, PahlHL. DeltaF508-CFTR causes constitutive NF-kappaB activation through an ER-overload response in cystic fibrosis lungs. Biol Chem. 2002;383:271–282. [PubMed]
PahlHL, BaeuerlePA. A novel signal transduction pathway from the endoplasmic reticulum to the nucleus is mediated by transcription factor NF-kappa B. EMBO J. 1995;14:2580–2588. [PubMed]
JohnstonJA, WardCL, KopitoRR. Aggresomes: a cellular response to misfolded proteins. J Cell Biol. 1998;143:1883–1898. [CrossRef] [PubMed]
WaelterS, BoeddrichA, LurzR, et al. Accumulation of mutant huntingtin fragments in aggresome-like inclusion bodies as a result of insufficient protein degradation. Mol Biol Cell. 2001;12:1393–1407. [CrossRef] [PubMed]
Garcia-MataR, BebokZ, SorscherEJ, SztulES. Characterization and dynamics of aggresome formation by a cytosolic GFP-chimera. J Cell Biol. 1999;146:1239–1254. [CrossRef] [PubMed]
KawaguchiY, KovacsJJ, McLaurinA, VanceJM, ItoA, YaoTP. The deacetylase HDAC6 regulates aggresome formation and cell viability in response to misfolded protein stress. Cell. 2003;115:727–738. [CrossRef] [PubMed]
TaylorJP, TanakaF, RobitschekJ, et al. Aggresomes protect cells by enhancing the degradation of toxic polyglutamine-containing protein. Hum Mol Genet. 2003;12:749–757. [CrossRef] [PubMed]
LemJ, KrasnoperovaNV, CalvertPD, et al. Morphological, physiological, and biochemical changes in rhodopsin knockout mice. Proc Natl Acad Sci USA. 1999;96:736–741. [CrossRef] [PubMed]
RajanRS, KopitoRR. Suppression of wild-type rhodopsin maturation by mutants linked to autosomal dominant retinitis pigmentosa. J Biol Chem. 2005;280:1284–1291. [CrossRef] [PubMed]
Figure 1.
 
Rod photoreceptor, rhodopsin, and transgenic rhodopsin sequence. (A) Depiction of the structure of a rod photoreceptor. The rod is a highly polarized neuronal cell consisting of a synaptic terminal (S), a cell body or an inner segment (IS) containing the cell’s biosynthetic machinery, and a specialized organelle, the outer segment (OS), which consists of a stack of flattened membranous sealed disks discontinuous from the plasma membrane. Rhodopsin is synthesized in the IS and is subsequently incorporated into the OS disks, where it participates in visual transduction. Loss of polarized rhodopsin localization is detrimental to the cell. (B) Representation of a wild-type rhodopsin molecule. Rhodopsin is a heptahelical transmembrane protein. Its N terminus contains two glycan moieties, and the C terminus contains two palmitoylated cysteines. Rhodopsin’s chromophore, 11-cis-retinal, is bound to the lysine at residue 296. The P23H mutation resides in the intradiskal/extracellular N terminus. (C) Engineered sequence alterations at the N and C terminus of the transgenic rhodopsins. Residue 13 and the extreme C terminus (boxed regions) were substituted with bovine sequences to introduce antibody epitopes 2B2 and 1D4, respectively. Numbers refer to the residue in the bovine sequence. Arrows indicate the amino acids that were mutated. Given that both these changes were conservative, neither was expected to affect the structure/function of the molecule.
Figure 1.
 
Rod photoreceptor, rhodopsin, and transgenic rhodopsin sequence. (A) Depiction of the structure of a rod photoreceptor. The rod is a highly polarized neuronal cell consisting of a synaptic terminal (S), a cell body or an inner segment (IS) containing the cell’s biosynthetic machinery, and a specialized organelle, the outer segment (OS), which consists of a stack of flattened membranous sealed disks discontinuous from the plasma membrane. Rhodopsin is synthesized in the IS and is subsequently incorporated into the OS disks, where it participates in visual transduction. Loss of polarized rhodopsin localization is detrimental to the cell. (B) Representation of a wild-type rhodopsin molecule. Rhodopsin is a heptahelical transmembrane protein. Its N terminus contains two glycan moieties, and the C terminus contains two palmitoylated cysteines. Rhodopsin’s chromophore, 11-cis-retinal, is bound to the lysine at residue 296. The P23H mutation resides in the intradiskal/extracellular N terminus. (C) Engineered sequence alterations at the N and C terminus of the transgenic rhodopsins. Residue 13 and the extreme C terminus (boxed regions) were substituted with bovine sequences to introduce antibody epitopes 2B2 and 1D4, respectively. Numbers refer to the residue in the bovine sequence. Arrows indicate the amino acids that were mutated. Given that both these changes were conservative, neither was expected to affect the structure/function of the molecule.
Figure 2.
 
Expression of c-rhoP23H causes retinal degeneration in transgenic X. laevis. Confocal micrographs merged with DIC images of dpf 14 (approximately stage 47/48) wild-type retinas (A, B) or transgenic retinas expressing c-rho (C, D) or c-rhoP23H (EH). Retinal sections were stained with Alexa fluor 488 wheat germ agglutinin to outline cell layers (red) and Hoechst 33342 nuclear stain (blue). Right: Enlargements of the central retina shown in the corresponding left panel. Rod OS from retinas expressing c-rho (D) are similar in length and packing density to wild-type retinas (B). However, retinas expressing c-rhoP23H exhibited varying degrees of retinal degeneration (F, H). Rod OS are short and intermittent (F) or almost entirely missing (H). Cones (asterisks) predominate in areas devoid of rods. Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C, E, G); 2 μM (B, D, F, H).
Figure 2.
 
Expression of c-rhoP23H causes retinal degeneration in transgenic X. laevis. Confocal micrographs merged with DIC images of dpf 14 (approximately stage 47/48) wild-type retinas (A, B) or transgenic retinas expressing c-rho (C, D) or c-rhoP23H (EH). Retinal sections were stained with Alexa fluor 488 wheat germ agglutinin to outline cell layers (red) and Hoechst 33342 nuclear stain (blue). Right: Enlargements of the central retina shown in the corresponding left panel. Rod OS from retinas expressing c-rho (D) are similar in length and packing density to wild-type retinas (B). However, retinas expressing c-rhoP23H exhibited varying degrees of retinal degeneration (F, H). Rod OS are short and intermittent (F) or almost entirely missing (H). Cones (asterisks) predominate in areas devoid of rods. Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C, E, G); 2 μM (B, D, F, H).
Figure 3.
 
The rhodopsin-inactivating mutation K296R does not rescue P23H-induced retinal degeneration. Confocal micrographs merged with DIC images of transgenic retinas expressing c-rhoK296R (A, B) or c-rhoP23H/K296R (C, D). Retinal sections were stained with Alexa fluor 488 wheat germ agglutinin to outline cell layers (red) and Hoechst 33342 nuclear stain (blue). (A, C) Enlargements of the central retina (B, D). Retinas expressing c-rhoK296R (A, B) are comparable to wild-type retinas (Figs. 2A 2B) . Rod OS are full length and closely apposed. However, in retinas expressing c-rhoP23H/K296R (C, D), there are significantly fewer and shorter rod OS. Cones (asterisks) predominate in areas devoid of rods. This pattern of retinal degeneration is similar to c-rhoP3H (Figs. 2E 2F 2G 2H) . Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C); 2 μM (B, D).
Figure 3.
 
The rhodopsin-inactivating mutation K296R does not rescue P23H-induced retinal degeneration. Confocal micrographs merged with DIC images of transgenic retinas expressing c-rhoK296R (A, B) or c-rhoP23H/K296R (C, D). Retinal sections were stained with Alexa fluor 488 wheat germ agglutinin to outline cell layers (red) and Hoechst 33342 nuclear stain (blue). (A, C) Enlargements of the central retina (B, D). Retinas expressing c-rhoK296R (A, B) are comparable to wild-type retinas (Figs. 2A 2B) . Rod OS are full length and closely apposed. However, in retinas expressing c-rhoP23H/K296R (C, D), there are significantly fewer and shorter rod OS. Cones (asterisks) predominate in areas devoid of rods. This pattern of retinal degeneration is similar to c-rhoP3H (Figs. 2E 2F 2G 2H) . Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C); 2 μM (B, D).
Figure 4.
 
Quantitative analysis of retinal degeneration induced by c-rhoP23H and c-rhoP23H/K296R. Solubilized transgenic eyes were blotted onto membranes and probed with antibodies that recognized total rhodopsin (mAb B630N; A) or only transgenic rhodopsin (mAb 1D4; B). Control samples contained 100% wild-type rhodopsin from wild-type retinas (WT) or 100% transgenic c-rho from transfected HEK293S cells (293). Low mAb B630N signals indicate low levels of rhodopsin and thus retinal degeneration. (C) Dose–response curve generated from quantification of the fluorescent signals from (A) and (B). Expression of control rhodopsins c-rho and c-rhoK296R did not result in significant loss of total rhodopsin at the expression levels observed. c-RhoP23H and c-rhoP23H/K296R caused similar rhodopsin loss. In c-rho and c-rhoK296R controls, variability in total rhodopsin primarily resulted from variation in the size of the eyes. This variability was greater than in wild-type animals and likely was associated with stresses from the transgenesis and selection procedures.
Figure 4.
 
Quantitative analysis of retinal degeneration induced by c-rhoP23H and c-rhoP23H/K296R. Solubilized transgenic eyes were blotted onto membranes and probed with antibodies that recognized total rhodopsin (mAb B630N; A) or only transgenic rhodopsin (mAb 1D4; B). Control samples contained 100% wild-type rhodopsin from wild-type retinas (WT) or 100% transgenic c-rho from transfected HEK293S cells (293). Low mAb B630N signals indicate low levels of rhodopsin and thus retinal degeneration. (C) Dose–response curve generated from quantification of the fluorescent signals from (A) and (B). Expression of control rhodopsins c-rho and c-rhoK296R did not result in significant loss of total rhodopsin at the expression levels observed. c-RhoP23H and c-rhoP23H/K296R caused similar rhodopsin loss. In c-rho and c-rhoK296R controls, variability in total rhodopsin primarily resulted from variation in the size of the eyes. This variability was greater than in wild-type animals and likely was associated with stresses from the transgenesis and selection procedures.
Figure 5.
 
Dark rearing of rhoP23H-expressing tadpoles did not rescue rod degeneration. Tadpoles expressing c-rhoP23H were raised in cyclic light (12 hours light/12 hours dark) or in total darkness. Duplicate dot blots of solubilized transgenic eyes were probed with mAb B630N or mAb 1D4, and the results were quantified and plotted. Dark rearing did not protect from rhoP23H-induced retinal degeneration compared with cyclic light rearing, suggesting that activation of the visual transduction cascade does not play a role in rhoP23H-induced retinal degeneration.
Figure 5.
 
Dark rearing of rhoP23H-expressing tadpoles did not rescue rod degeneration. Tadpoles expressing c-rhoP23H were raised in cyclic light (12 hours light/12 hours dark) or in total darkness. Duplicate dot blots of solubilized transgenic eyes were probed with mAb B630N or mAb 1D4, and the results were quantified and plotted. Dark rearing did not protect from rhoP23H-induced retinal degeneration compared with cyclic light rearing, suggesting that activation of the visual transduction cascade does not play a role in rhoP23H-induced retinal degeneration.
Figure 6.
 
RhoP23H is retained in the ER of rod IS. Confocal micrographs of transgenic tadpole retinas at 14 dpf. (AJ) Cryosections from fixed eyes were labeled with 514-18 (A, B) or 2B2 (CJ) (green) to visualize wild-type or transgenic rhodopsin, respectively. Post-Golgi membranes and nuclei were stained with wheat germ agglutinin (red) and Hoechst 33342 (blue), respectively. Wild-type retina (A, B) shows the normal distribution of rhodopsin in the OS and the Golgi membranes (arrows). The distribution of c-rho (C, D) and c-rhoK296R (E, F) resembles that of wild-type. However, c-rhoP23H (G, H) and c-rhoP23H/K296R (I, J) were both localized diffusely throughout the IS. (KM) Retina expressing c-rhoP23H double labeled with anticalnexin (red, K) and mAb 2B2 (green, L). Colocalization (yellow, M) indicates that the mutant protein was retained in the ER. Occasional narrow bands of rhoP23H were observed in the OS (arrowhead). Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C, E, G, I, K, L); 2 μM (B, D, F, H, J, M).
Figure 6.
 
RhoP23H is retained in the ER of rod IS. Confocal micrographs of transgenic tadpole retinas at 14 dpf. (AJ) Cryosections from fixed eyes were labeled with 514-18 (A, B) or 2B2 (CJ) (green) to visualize wild-type or transgenic rhodopsin, respectively. Post-Golgi membranes and nuclei were stained with wheat germ agglutinin (red) and Hoechst 33342 (blue), respectively. Wild-type retina (A, B) shows the normal distribution of rhodopsin in the OS and the Golgi membranes (arrows). The distribution of c-rho (C, D) and c-rhoK296R (E, F) resembles that of wild-type. However, c-rhoP23H (G, H) and c-rhoP23H/K296R (I, J) were both localized diffusely throughout the IS. (KM) Retina expressing c-rhoP23H double labeled with anticalnexin (red, K) and mAb 2B2 (green, L). Colocalization (yellow, M) indicates that the mutant protein was retained in the ER. Occasional narrow bands of rhoP23H were observed in the OS (arrowhead). Rpe, retinal pigment epithelium; os, outer segments; is, inner segments; n, photoreceptor nuclei. Magnification bars: 10 μM (A, C, E, G, I, K, L); 2 μM (B, D, F, H, J, M).
Figure 7.
 
RhoP23H forms high molecular weight aggregates when expressed in transgenic X. laevis retinas. Samples from solubilized wild-type retina (lane 1, WT), c-rho expressing transgenic retina (lane 2), c-rhoP23H-expressing transgenic retinas (lanes 514) and solubilized HEK293S cells transfected with c-rho (lane 3) or c-rhoP23H (lane 4) analyzed by Western blot. Duplicate blots were probed with mAb B630N (A) and mAb 1D4 (B). Wild-type rhodopsin appears primarily as a monomer (approximately 35 kDa) and, to a lesser extent, as dimers (approximately 70 kDa) and tetramers (approximately 120 kDa). mAb 1D4 exclusively recognizes epitope-tagged rhodopsins (B, lanes 214) but not wild-type X. laevis rhodopsin (B, lane 1). c-Rho expressed in transgenic retinas appeared identical with wild-type, whereas c-rhoP23H was present mostly as dimers (approximately 70 kDa) and trimers (approximately 90 kDa) regardless of expression level. c-Rho and c-rhoP23H expressed in HEK293S cells exhibited similar behavior. Percentages given represent the proportion of total rhodopsin that constituted transgenic rhodopsin.
Figure 7.
 
RhoP23H forms high molecular weight aggregates when expressed in transgenic X. laevis retinas. Samples from solubilized wild-type retina (lane 1, WT), c-rho expressing transgenic retina (lane 2), c-rhoP23H-expressing transgenic retinas (lanes 514) and solubilized HEK293S cells transfected with c-rho (lane 3) or c-rhoP23H (lane 4) analyzed by Western blot. Duplicate blots were probed with mAb B630N (A) and mAb 1D4 (B). Wild-type rhodopsin appears primarily as a monomer (approximately 35 kDa) and, to a lesser extent, as dimers (approximately 70 kDa) and tetramers (approximately 120 kDa). mAb 1D4 exclusively recognizes epitope-tagged rhodopsins (B, lanes 214) but not wild-type X. laevis rhodopsin (B, lane 1). c-Rho expressed in transgenic retinas appeared identical with wild-type, whereas c-rhoP23H was present mostly as dimers (approximately 70 kDa) and trimers (approximately 90 kDa) regardless of expression level. c-Rho and c-rhoP23H expressed in HEK293S cells exhibited similar behavior. Percentages given represent the proportion of total rhodopsin that constituted transgenic rhodopsin.
Table 1.
 
Kruskal-Wallis Analysis of Total Rhodopsin Levels
Table 1.
 
Kruskal-Wallis Analysis of Total Rhodopsin Levels
Group Transgene n Sum of Ranks Average Rank Average Total Rhodopsin
1 c-rho 22 1298 58.0 100
2 c-rhoK296R 22 1117 54.3 88.2
3 c-rhoP23H 22 797 38.6 67.6
4 c-rhoP23H/K296R 22 704 27.1 62.1
Table 2.
 
Multiple Comparisons Analysis of Total Rhodopsin Levels
Table 2.
 
Multiple Comparisons Analysis of Total Rhodopsin Levels
Comparison P Conclusion
Group 1 vs. 4 0.00024 1 > 4
Group 1 vs. 3 0.0017 1 > 3
Group 1 vs. 2 0.259 Not significantly different
Group 2 vs. 4 0.0091 2 > 4
Group 2 vs. 3 0.042 2 > 3
Group 3 vs. 4 0.550 Not significantly different
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×