Molecular Vision 2005; 11:1052-1060 <http://www.molvis.org/molvis/v11/a124/> Received 23 May 2005 | Accepted 30 November 2005 | Published 7 December 2005

Download Reprint

Spatial and temporal expression patterns of the choroideremia gene in the mouse retina

Nicholas W. Keiser, Waixing Tang, Zhangyong Wei, Jean Bennett
 
 

Department of Ophthalmology, F. M. Kirby Center for Molecular Ophthalmology, Scheie Eye Institute, University of Pennsylvania, Philadelphia, PA

Correspondence to: Jean Bennett, Department of Ophthalmology, F. M. Kirby Center for Molecular Ophthalmology, Scheie Eye Institute, University of Pennsylvania, Philadelphia, PA, 19104-6069; Phone: (215) 898-0915; FAX: (215) 573-7155; email: jebennet@mail.med.upenn.edu

Abstract

Purpose: Choroideremia (CHM), an X-linked retinal disease, is caused by mutations affecting the CHM gene. This gene encodes REP-1, which functions in the covalent modifications of proteins involved in vesicle trafficking. The disease affects several cell types in the retina, but it is not known which cell types contribute directly or indirectly to disease progression. A study of the expression patterns of Chm and the related gene Chml in the mouse retina was undertaken in order to address this issue.

Methods: The expression patterns of Chm and Chml were determined by in situ hybridization. The localization of the Chm protein product, Rep-1, was determined spatially and temporally in the mouse retina by immunohistochemistry.

Results: Chm and Chml mRNA were found in every major layer of the retina in adult mice. During development, Rep-1 protein localization changes from a fairly diffuse pattern during embryogenesis to a more specific pattern at the time of retinal differentiation. In adulthood, Rep-1 localizes to distinct cellular compartments in multiple retinal cell types.

Conclusions: Chm and Chml have the same broad expression profile in the mouse retina. In particular, the Chm transcript and corresponding protein are found in cell types other than those thought to be primarily affected in the human disease. These results have important implications for approaches with which to develop a relevant mouse model of choroideremia and for therapeutic strategies for this disease.

Introduction

Choroideremia (CHM) is a slowly progressing X-linked retinal disease characterized by degeneration of the choroid, the retinal pigment epithelium (RPE), and the neural retina. The disease is rare, representing 4% of all cases of inherited blindness, but can be readily diagnosed based on the appearance of the fundus of affected males. Individuals with CHM develop night blindness and loss of peripheral vision during the second decade of life, with progressive degeneration leading to complete blindness by about age 40. Due to random X-inactivation, female carriers of CHM can have patchy areas of degeneration, but tend to maintain good vision, although some vision loss later in life has been reported [1].

Choroideremia has been linked to the CHM gene on Xq21.2, which contains 15 exons spanning 150 kb [2-9]. The open reading frame encodes Rab Escort Protein-1 (REP-1), a 653-amino acid protein originally isolated as component A of Rab geranylgeranyl transferase (RGGT) [10-12]. Numerous types of mutations in CHM have been linked to the disease, including point mutations, insertions, and deletions ranging from one base pair to the entire locus [1,13-27]. Other disease-causing mutations include an L1 retrotransposon insertion, the addition of a cryptic exon, and exon skipping [28]. All mutations characterized to date lead to a truncation or a complete absence of REP-1 protein, rendering it undetectable by western blot analysis [29].

REP-1 shares 71% identity with another Rab escort protein, REP-2, which is encoded by the CHML gene [30,31]. These proteins perform redundant functions in mammals; both REPs bind to newly synthesized Rab GTPase proteins and escort them to RGGT, which covalently modifies the C-terminus of the Rab proteins with two geranylgeranyl isoprenoids [32]. This allows for their association with donor membranes. Over 60 Rab proteins have been identified in mammalian cells [33], and although REP-1 and REP-2 seem to bind Rabs with equal affinity, their efficiency in facilitating the conversion of certain unprenylated Rabs to their prenylated state varies [34,35]. Individuals with CHM mutations only develop ocular defects, so REP-2 probably compensates for REP-1 function in other organs. The two proteins appear to be present ubiquitously in systemic organs. However, differences between their relative levels have been found in rat cells [36], and differences in the amount of CHM mRNA between various eye compartments in non-human primates have been observed [37]. A previous report involving immunohistochemistry on sections of retina from a choroideremia carrier shows that REP-1 is restricted to amacrine cells and rod photoreceptors [38]. It is presently unknown whether CHML possesses a restrictive expression pattern in the retina. Cells not expressing CHML could potentially function abnormally if REP-1 was absent, and may therefore represent points of initiation of the disease. As a first step in expanding our knowledge of the progression of CHM at the molecular level, we performed detailed studies of Chm and Chml expression and Rep-1 protein localization in the mouse retina.

In order to determine potential initiation points of disease in choroideremia, and to establish potential targets of therapeutic intervention for a mouse model, we examined the specific cellular localization patterns of Chm and Chml mRNA in the adult mouse eye, and the localization patterns of Rep-1 protein at different stages of ocular development. We found no major differences in mRNA localization between Chm and Chml, but the pattern of Rep-1 localization in the mouse retina was found to be vastly different from what has been reported in the human [38], particularly in the inner retina and photoreceptors. The similarities between Chm and Chml expression patterns in the mouse and the difference in REP-1 localization between mouse and humans may explain the difficulties others have experienced in creating a mouse model for CHM [39,40]. Development of a viable murine model of choroideremia may require additional steps, such as the use of cell-specific knockouts of Chm or further analysis of specific Rep-1 targets.

Methods

Tissue preparation

All tissue used in these studies was from the CD-1 (ICR) mouse strain (Charles River Laboratories, Wilmington, MA). The CD-1 strain is derived from an albino non-inbred stock, and was chosen because it has no known retinal degenerative disease, displays high fertility, and is non-pigmented. The latter quality simplifies detection of cells that are positive by immunohistochemistry and in situ hybridization, particularly as one of the cell types of interest (retinal pigment epithelium (RPE) cells) are heavily pigmented in other strains. Animals were housed at the University of Pennsylvania's Animal Care Facility in compliance with instititutional and national regulations and with ARVO guidelines for animal care and use. Animals were housed in 12 h on/12 h off light with food and water provided ad libitum. Eyes from postnatal day (P)1, 7, 12, 14, and adult (4 months), and heads from embryonic day (E)14 were removed from cohorts of mice and fixed immediately in 4% paraformaldehyde in PBS overnight at 4 °C, cryoprotected with 30% sucrose, embedded in Tissue Freezing Medium (Triangle Biomedical Sciences, Durham, NC), and cyrosectioned.

Generation of RNA probes and in situ hybridization

RNA probes for in situ hybridization were constructed from PCR fragments of Chm and Chml amplified from total mouse brain RNA by RT-PCR. The Chm PCR product, corresponding to nucleotides 3307-3655, was amplified using the primers 5'-CATGGCTCCAAGGAAAAGCAC-3' and 5'-GGGCTTCAGGAAACAACAGC-3', and the Chml PCR product, corresponding to nucleotides 1158-1450, was amplified using the primers 5'-TACTCCGGCTGACTCTCTGG-3' and 5'-ATCTCCGGCTCTCCTTAACC-3'. The PCR products were cloned into the TOPO Zero Blunt II plasmid (Invitrogen, Carlsbad, CA). Flanking XhoI and EcoRI sites were created by a second round of PCR and used to clone the fragments into pBluescriptKS+/-(Stratagene, La Jolla, CA). Although the products were not sequenced, they were amplified using a high fidelity DNA polymerase and primers that had no matches to any other sequences. The amplified bands were of the predicted size. Sense and antisense DIG-labeled RNA probes were transcribed from pBluescriptKS^+/- using the DIG RNA Labeling Kit (Roche Molecular Diagnostics, Pleasanton, CA) with T7 or T3 RNA polymerase. In situ hybridization was carried out on 14 μm cryosections of eyes from adult CD-1 mice as previously described [41] with the following exceptions: wash buffers contained 0.2% Tween-20, all post-hybridization washes were extended by 5 min each, hybridization was performed at 55 °C, and post-hybridization washes were carried out at 50 °C and 45 °C. Following color development and post-fixation, slides were coverslipped with 70% glycerol and photographed under Nomarski differential interference contrast (DIC) microscopy.

Antibodies and lectins

Rabbit antiserum G2373 was generated by Genemed Synthesis, Inc. (South San Francisco, CA) to a synthetic peptide (CSESSVIPETNSETPK), corresponding to amino acids 638-652 in the mouse Rep-1 protein, and affinity purified on a column of Sulfolink beads (Pierce Technology Inc., Rockford, IL) coupled to the peptide. Affinity-purified whole rabbit IgG was purchased from Jackson Immunoresearch (West Grove, PA). The antibody to rhodopsin (4D2) was provided by Dr. Robert Molday (University of British Columbia, Vancouver, BC), and the mouse anti-rabbit CRALBP antibody was a gift from Dr. John Saari (University of Washington, Seattle, WA). Biotinylated peanut agglutinin lectin (PNA), mouse anti-cow calbindin D, mouse anti-calretinin, and mouse anti-PKCα (clone MC5) antibodies were purchased from Vector Laboratories, Inc. (Burlingame, CA), Sigma-Aldrich (St. Louis, MO), Chemicon International, Inc. (Temecula, CA), and BD Pharmingen (San Diego, CA), respectively.

Western blotting

Extract from an adult (4 months) CD-1 mouse eye was separated on a 4-12% Bis-Tris NuPAGE gel (Invitrogen, Carlsbad, CA) by electrophoresis and transferred to nitrocellulose membranes. Following blocking for 10 min in 5% milk/TBST (TBST is 50 mM Tris, 2 mM CaCl₂, 0.1% Tween-20), membranes were probed with serial dilutions of affinity-purified G2373 anti-mouse REP-1 antibody overnight at 4 °C. The membranes were washed in TBST and incubated with goat anti-rabbit peroxidase antibody diluted 1:5000 in 5% non fat dry milk/TBST, washed again in TBST, dried, exposed to ECL reagent (Amersham Biosciences, Piscataway, NJ), and exposed to film. The highest working dilution for the G2373 antibody was determined to be 1:5000. For peptide competition studies, the procedure was the same, except that excess Rep-1 peptide (5x, 10x, 50x, and 100x by mass) was pre-incubated with G2373 (1:5000) before probing overnight at 4 °C.

Immunohistochemistry

Freezing medium was removed from 8 μm cryosections of eyes from CD-1 mice of different developmental stages by incubation in PBS for 15 min, and immunostaining was performed using the Vectastain ABC Kit (Vector Laboratories, Inc.) with the purified G2373 antibody at a working dilution of 1:200 (1.4 μg/ml). As a control, affinity purified whole rabbit IgG (Jackson Immunoresearch, West Grove, PA) was incubated with the sections at the same concentration. Color development was achieved using the DAB Peroxidase Substrate Kit (Vector Laboratories, Inc.), and halted by incubation of the sections with water for 5 min. Slides were coverslipped using Fluoromount G (Southern Biotechnology Associates, Birmingham, AL) and photographed under Nomarski differential interference contrast (DIC) optics.

Co-immunofluorescence

CD-1 mouse eyes from P12, P14, and adult stages were fixed and cyrosectioned at 10 μm as described. Freezing medium was removes as described, and immunofluorescence with the following antibodies or lectins was performed using the Mouse on Mouse (M.O.M.) Kit (Vector Laboratories, Inc.): anti-rhodopsin (1:100), PNA (1:12.5), anti-CRALBP (1:500), anti-calbindin D (1:500), anti-calretinin (1:2000), and anti-PKCα (1:100) according to the manufacturer's instructions, except that avidin-Cy3 (1:1000, Jackson Immunoresearch, West Grove, PA) was used to detect the primary antibodies. Following two 30 min washes with PBS, co-staining for REP-1 was performed. Blocking was carried out for 1 h in 5% normal donkey serum, 0.1% Triton X-100, 0.5% BSA/PBS. Affinity-purified anti-mouse Rep-1 antibody (G2373) was diluted 1:200 in 0.1% Triton X-100, 0.5% BSA, PBS and was incubated with the sections overnight at 4 °C. After two 30 min washes with PBS, the sections were incubated with donkey anti-rabbit Alexa 488 (1:200, Molecular Probes, Eugene, OR) for 1 h and washed twice again with PBS for 30 min. Slides were coverslipped with Vectashield Mounting Medium (Vector Laboratories, Inc.) and sealed with nail polish. Images of Cy3 and Alexa 488 fluorescence were taken using a fluorescence microscope, and the images were merged to determine co-localization of the signals.

Results

Chm and Chml mRNA localize to the same regions of the adult mouse retina

In order to determine the distinct retinal cell types that express Chm and Chml, we performed in situ hybridization on cryosections of retinas from CD-1 mice. Sense and antisense probes were used that corresponded to nucleotides 3307-3655 in Chm and 1158-1450 in Chml. These regions were chosen because they had no significant homology to any other mouse mRNAs based on BLAST searches. We found that Chm localized strongly to the inner retina (Figure 1A,C), where both the ganglion cells and the cells of the inner nuclear layer (INL) were positive. Some cells in the inner nuclear layer contained noticeably higher levels of Chm RNA than surrounding cells (Figure 1A, INL layer, black arrow). In addition, the outer retina was also positive for Chm mRNA, with strong staining observed in the inner segments of the photoreceptors and the RPE (Figure 1A,C). Weaker staining was observed in the photoreceptor cell bodies and the choroid (Figure 1A,C, outer nuclear layer (ONL) and choroid (CH) layers, black arrows). No signal was detected when the sense probe was used as a control (Figure 1B,D). Similar staining patterns were observed in additional sections from C57Bl/6 mouse eyes (data not shown).

Chml transcripts were also detected in the ganglion cells and the inner nuclear layer, but the signal was weaker and more uniform than that of Chm (Figure 1E,G). Signal in the photoreceptor cell bodies was not visible, but moderate staining was observed in the photoreceptor inner segments and the RPE (Figure 1E,G). No staining was observed in sections where the sense probe was used (Figure 1F,H). Taken together, these results show that Chm and Chml transcripts were expressed in multiple cell types in a virtually identical pattern in the mouse retina, although the overall levels of the two transcripts appeared to be different.

Changes in the localization of Rep-1 during murine retinal development

In order to assess the localization of Rep-1 protein in the retina, we generated an antibody specific to the mouse Rep-1 protein. Using a peptide corresponding to amino acids 638-652 of mouse Rep-1, which shares no identity with the mouse Rep-2 protein, antisera was raised in rabbits (GeneMed Synthesis, Inc.), and polyclonal antibody (G2373) was purified by affinity chromatography. This peptide sequence is 86.7% similar to the analogous portion of human REP-1. The antibody was able to detect a single band on a western blot of extract from an adult CD-1 mouse eye at a dilution of 1:5000 (Figure 2A). The molecular weight of the band was consistent with the molecular weight of the REP-1 protein [10]. To prove the specificity of this antibody, we pre-incubated the diluted antibody with increasing mass amounts of competitor peptide before performing western blots on CD-1 mouse eye extract. The peptide successfully competed out the band, indicating that the purified antibody was specific to mouse Rep-1 (Figure 2B). Rep-2 is about 75 kDa and Rep-1 is about 80-90 kDa and the two molecules were separated by SDS-PAGE [36].

We next utilized this antibody to determine the localization pattern of the Rep-1 protein in the murine retina during development. Immunohistochemistry was performed on cryosections of CD-1 mouse eyes from different stages of development: embryonic day (E)14, postnatal day (P)1, P7, P12, P14, and adult (4 months). Control sections, preabsorbed with antibody, showed no staining (data not shown). At E14, Rep-1 was equally distributed across all neuroblasts cells in the retina and the RPE (Figure 2C). At the P1 stage, staining was observed in the neural progenitor layer, with a more intense signal present the inner retina and the newly formed ganglion cell layer (Figure 2E). By P7, when distinct layers of nuclei are visible in the neural retina, the intensity of staining had diminished in the cell bodies of outer retina, but remained present in the RPE, the inner plexiform layer (IPL), and the ganglion cell layer (Figure 2G). In addition, new layers of strong Rep-1 localization were present in the center and the outermost edge of the neuroblast layer. At the P12 stage, strong expression was observed in the ganglion cells, and weaker expression was observed in the IPL and around distinct cell bodies in the INL (Figure 2I). A strong signal was observed in the outer plexiform layer, but minimal expression was detected in the ONL. However, we did observe weak expression in the inner segments of the photoreceptors, which were visible in a distinct layer by this timepoint (Figure 2I). Expression in the RPE and choroid was also observed. The pattern seen at P14 was largely identical to that at P12 (Figure 2K), but the signal had intensified in the IPL, the outer plexiform layer, and the photoreceptor inner segments. By adulthood, strong localization was visible in the ganglion cell layer and the outer plexiform layer, with a high degree of staining also seen in the IPL (Figure 2M). Rep-1 was also present at a weak level in the cell bodies in the INL and ONL, with relatively strong staining in the inner segments of the photoreceptors. No staining was observed in the outer segments. We also observed strong staining in the RPE and choroid layers at the adult stage.

Cell type-specific localization of Rep-1 in the mouse retina

In order to elucidate the specific retinal cell types positive for Rep-1, we employed co-immunofluorescence for Rep-1 and various cell type-specific markers. To determine whether Rep-1 is restricted to the inner segments and excluded from the outer segments in the rod cells, we performed co-immunofluorescence for Rep-1 and rhodopsin. We found Rep-1 in the inner segments, with no detectable co-localization with rhodopsin in the adjacent outer segments (Figure 3A). This confirmed that these two proteins are confined to separate cellular compartments in the rod photoreceptors of adult mice. We next sought to determine if the same pattern could be observed in the cone photoreceptors, and so we employed PNA to label the cone sheaths. We found some co-localization between PNA labeling and Rep-1 labeling in the inner segments (Figure 3B and inset), indicating that Rep-1 is not excluded from the cone photoreceptors in the mouse. We also detected Rep-1 in the inner processes of Müller glia and in the RPE, as evidenced by the co-localization patterns between Rep-1 and CRALBP (Figure 3C).

We then attempted to determine exactly which cell types in the inner retina contained Rep-1 protein. By co-staining with an antibody to calbindin D, we were able to determine that Rep-1 was present at high levels in the processes of horizontal cells (Figure 3D). Rep-1 was also present in some amacrine and ganglion cells, as shown by co-localization with calretinin (Figure 3E). Lastly, we observed strong co-localization between Rep-1 and PKCα, a marker for rod bipolar cells, mostly within the bipolar cell bodies in the INL (Figure 3F).

Discussion

In this study, we sought to determine the localization patterns of Chm and Chml mRNA and Rep-1 protein in the retina, and whether Rep-1 localization patterns change over the course of differentiation of the various retinal cell layers. Using in situ hybridization, we found that both Chm and Chml mRNA are expressed throughout the adult mouse retina. Only subtle qualitative differences between expression patterns of the two genes were apparent: lower levels of expression were seen for Chml as compared to Chm, and Chml mRNA was not detected in the outer nuclear layer or the choroid, both of which expressed very low levels of Chm.

Widespread presence of the Chm protein product, Rep-1, was seen at early developmental timepoints. In the developing mouse retina (at E14), Rep-1 is found ubiquitously in all cell types and progenitors. From P1 to P12, Rep-1 can be found mostly in the RPE, the choroid, and the inner retina, although some strong localization is apparent in some neuroblasts at P7. At P12 it can be found in the developing inner segments of the photoreceptors. By adulthood, Rep-1 is distributed throughout a wide variety of cell types, including ganglion and amacrine cell bodies, the inner plexiform layer, the rod bipolar cell bodies and horizontal cell processes in the inner nuclear layer, the processes of Müller glia, the rod and cone inner segments, the RPE, and the choroid layer. Thus, there is some compartmentalization of Rep-1 during retinal development and maturity. Presumably, the localization of Rep-1 in the inner segments during photoreceptor differentiation reflects the role of this protein in growth or function of the outer segments. A dramatic increase in Rab prenylation is likely required to accommodate the large increase in protein transport that occurs as outer segment discs are formed. Rep-1 then persists throughout adulthood, and is presumably utilized in the repetitive cycles of disc membrane formation and shedding that occur throughout adulthood.

Rep-1 is present throughout both the undifferentiated and mature retina, and thus is not localized specifically to one of the cell types (photoreceptors, RPE and/or choroid) thought to be involved in disease progression in choroideremia. In a previous study, immunohistochemistry on retinal sections from an elderly human carrier for choroideremia showed strong abundance of REP-1 protein in rod photoreceptors and some amacrine cells [38], but no other retinal cell types. Our finding of high levels of Chm expression throughout the retina in mice may highlight a major difference in patterns of the REP-1 cellular specificity between mammalian species. Preliminary results with retinal tissue sections from rhesus monkeys have shown that CHM and CHML share the same widespread pattern of expression as their mouse homologues (data not shown), suggesting that humans may diverge from other species (even other primates) in their pattern of REP-1 localization.

We detected both Chm transcripts and Rep-1 in the mature photoreceptor inner segments of the mouse. Our findings in mice are consistent with the previously reported REP-1 localization pattern in the rod photoreceptors of the human retina [38], except that in the present study, in mice, both rods and cones contain Rep-1. It was postulated that the rod-specific expression pattern seen in humans accounts for the night blindness experienced by choroideremia patients in the early stages of the disease [38]. The finding that cone cells in mice also possess this protein may represent a major species difference. Alternatively, it may reflect differences in specificity of the different antibodies used in the two studies. In the human, REP-1 also localizes to the rod photoreceptor cell bodies, and is found at a high level in the synapses [38]. We observed mouse Chm transcript and Rep-1 protein in photoreceptor cell bodies, albeit at low levels compared to other cells.

All of these studies utilized eye tissue from the albino CD-1 mouse strain, which was chosen because its lack of pigment granules in the RPE and choroid would allow for easy detection of protein and mRNA in these tissues by our methods. It is unlikely that a pigmented mouse would show much deviation from the results reported here, especially in the tissue distribution of Chm and Chml, as our results show that even albino mice express both genes in all retinal layers, including the RPE and choroid. Where differences may occur is in the relative levels of REPs between pigmented and non-pigmented animals, as production and turnover of melanosomes in the RPE and choroid of the developing mouse eye may require a higher steady-state REP activity in these layers to aid in any increase in the prenylation of Rabs that may accompany an increase in melanosome trafficking. Therefore, differences between levels of REPs in the RPE and choroid of pigmented and non-pigmented animals would likely depend on whether the relative levels of Rab expression or prenylation was significantly different between the two mouse populations.

Previous studies show that CHM and CHML are co-expressed in many tissues, and this is likely to account for the lack of systemic disease in individuals affected with choroideremia [32]. The results of this study have shown that in the mouse retina, Chm and Chml are also co-expressed in retinal cells. The question arises as to why REP-2 does not suffice to replace the defective or absent REP-1 in the retinal cells of humans with choroideremia. One possibility is that one or more retinal cell types require a high steady-state level of general REP activity for proper function. Since our results and those of others [32] have shown that Rep-1 is expressed at a higher level than Rep-2 in the eye, the loss of REP-1 in particular may lead to a lower than average general REP activity in the cells in question, leading to a gradual accumulation of cellular defects. If this scenario holds true for humans, it could account for the slow progression of disease seen in individuals with CHM. This, of course, assumes that CHML expression patterns are as broad in the human as they are in the mouse. It is plausible that CHML is not co-expressed in the human retinal cells in which REP-1 function is necessary, and thus cannot substitute for REP-1 in individuals with CHM. However, preliminary data suggests that CHML has a broad pattern of expression in the retina (data not shown). Another possibility is that some Rab proteins that are specifically expressed in one or more retinal cell types require the presence of REP-1 for more efficient prenylation. Rab27a is expressed in the RPE and choroid, and while it binds REP-1 and REP-2 with equal efficiency, the REP-2/Rab27a complex has a significantly lower affinity for RGGT than the REP-1/Rab27a complex [34,35]. It is therefore possible that the buildup of unprenylated Rab27a or other Rabs with similar properties in retinal cells leads to global cellular defects and retinal degeneration. Analysis of the prenylation rates of other known Rab proteins in the presence of REP-1 or REP-2, in addition to the identification of new Rab proteins expressed in the retina, may aid in the determination of new potential downstream targets and give more insight into the progression of CHM.

One challenge in the field thus far has been the lack of an accurate mouse model for CHM. Recent work by Starr and colleagues [42] has identified a zebrafish mutant lacking REP-1 that exhibits both retinal degeneration and hair cell loss and dysfunction. This model may be very useful for examining the pathogenesis of CHM and determining specific Rabs that may be affected by REP-1 loss, but may not offer a way to test potential therapies that may be used in humans. Difficulties in generating a mouse model for CHM are complicated by apparent differences in systemic requirements for REP-1 and REP-2 between mice and humans [39,40]. Male mice lacking REP-1 died during embryonic development due to abnormal development of the extra-embryonic tissues, and defects were observed in trophoblast development and angiogenesis in the yolk sac and placenta [39]. Semi-quantitative RT-PCR analysis showed that Chml was expressed at a lower level than Chm in trophoblast cells, suggesting that Rab prenylation in some cell types may be entirely dependent on a specific REP [39]. Our data suggest that this may be the case in choroideremia, as Chm and Chml are expressed in the same cells in the mouse retina. It is clear from the results of Shi et al. [39] that a mouse model of choroideremia will require a conditional knockout of Chm so that the embryo can develop properly. Even so, a Chm knockout may not properly mimic the human disease, as our data have identified a clear species difference between mice and humans concerning REP-1 localization in the retina. The development of an animal model remains essential to a more complete understanding of the pathology of choroideremia.

Adenovirus-mediated gene transfer can be used in vitro to deliver wildtype CHM, and this rescues Rab geranylgeranyl transferase function in cells from affected individuals with choroideremia [43]. Given the broad expression patterns of Chm identified in this study, adenoviral vectors, which target RPE cells primarily [44,45], may not be the optimal vectors with which to treat choroideremia. Use of a modified adeno-associated virus (AAV) such as AAV2/5 would allow efficient targeting of both RPE and photoreceptor cells and may be necessary for optimal therapeutic effect with minimal toxicity [46].

Acknowledgements

This study was supported by NIH T32DK07748-06, the Foundation Fighting Blindness, The Lew R. Wasserman Merit Award from Research to Prevent Blindness, Inc., the Paul and Evanina Mackall Trust, and the F. M. Kirby Foundation.

References

1. Karna J. Choroideremia. A clinical and genetic study of 84 Finnish patients and 126 female carriers. Acta Ophthalmol Suppl 1986; 176:1-68.

2. Sankila EM, Lehner T, Eriksson AW, Forsius H, Karna J, Page D, Ott J, de la Chapelle A. Haplotype and multipoint linkage analysis in Finnish choroideremia families. Hum Genet 1989; 84:66-70.

3. Sankila EM, de la Chapelle A, Karna J, Forsius H, Frants R, Eriksson A. Choroideremia: close linkage to DXYS1 and DXYS12 demonstrated by segregation analysis and historical-genealogical evidence. Clin Genet 1987; 31:315-22.

4. Lesko JG, Lewis RA, Nussbaum RL. Multipoint linkage analysis of loci in the proximal long arm of the human X chromosome: application to mapping the choroideremia locus. Am J Hum Genet 1987; 40:303-11.

5. Schwartz M, Rosenberg T, Niebuhr E, Lundsteen C, Sardemann H, Andersen O, Yang HM, Lamm LU. Choroideremia: further evidence for assignment of the locus to Xq13-Xq21. Hum Genet 1986; 74:449-52.

6. Nussbaum RL, Lewis RA, Lesko JG, Ferrell R. Choroideremia is linked to the restriction fragment length polymorphism DXYS1 at XQ13-21. Am J Hum Genet 1985; 37:473-81.

7. Cremers FP, van de Pol DJ, van Kerkhoff LP, Wieringa B, Ropers HH. Cloning of a gene that is rearranged in patients with choroideraemia. Nature 1990; 347:674-7.

8. Merry DE, Janne PA, Landers JE, Lewis RA, Nussbaum RL. Isolation of a candidate gene for choroideremia. Proc Natl Acad Sci U S A 1992; 89:2135-9.

9. MacDonald IM, Sandre RM, Wong P, Hunter AG, Tenniswood MP. Linkage relationships of X-linked choroideremia to DXYS1 and DXS3. Hum Genet 1987; 77:233-5.

10. Andres DA, Seabra MC, Brown MS, Armstrong SA, Smeland TE, Cremers FP, Goldstein JL. cDNA cloning of component A of Rab geranylgeranyl transferase and demonstration of its role as a Rab escort protein. Cell 1993; 73:1091-9.

11. Seabra MC, Brown MS, Slaughter CA, Sudhof TC, Goldstein JL. Purification of component A of Rab geranylgeranyl transferase: possible identity with the choroideremia gene product. Cell 1992; 70:1049-57.

12. Seabra MC, Brown MS, Goldstein JL. Retinal degeneration in choroideremia: deficiency of rab geranylgeranyl transferase. Science 1993; 259:377-81.

13. MacDonald IM, Sandre RM, Hunter AG, Tenniswood MP. Gene mapping of X-linked choroideremia with restriction fragment-length polymorphisms. Can J Ophthalmol 1987; 22:310-5. Erratum in: Can J Ophthalmol 1988; 23:44.

14. Beaufrere L, Tuffery S, Hamel C, Arnaud B, Demaille J, Claustres M. A novel mutation (S558X) causing choroideremia. Hum Mutat 1996; 8:395.

15. Beaufrere L, Tuffery S, Hamel C, Bareil C, Arnaud B, Demaille J, Claustres M. An exonic polymorphism (381A/G) in the choroideremia gene. Genet Couns 1997; 8:223-5.

16. Cheng CK, Chen MS, Hou PK. Choroideremia: a study of two families. J Formos Med Assoc 1991; 90:1103-12.

17. Cremers FP, Brunsmann F, van de Pol TJ, Pawlowitzki IH, Paulsen K, Wieringa B, Ropers HH. Deletion of the DXS165 locus in patients with classical choroideremia. Clin Genet 1987; 32:421-3.

18. Cremers FP, Sankila EM, Brunsmann F, Jay M, Jay B, Wright A, Pinckers AJ, Schwartz M, van de Pol DJ, Wieringa B, de la Chapelle A, Pawlowitzki IH, Ropers HH. Deletions in patients with classical choroideremia vary in size from 45 kb to several megabases. Am J Hum Genet 1990; 47:622-8.

19. Forsythe P, Maguire A, Fujita R, Moen C, Swaroop A, Bennett J. A carboxy-terminal truncation of 99 amino acids resulting from a novel mutation (Arg555-->stop) in the CHM gene leads to choroideremia. Exp Eye Res 1997; 64:487-90.

20. Fujiki K, Hotta Y, Hayakawa M, Saito A, Mashima Y, Mori M, Yoshii M, Murakami A, Matsumoto M, Hayasaka S, Tagami N, Isashiki Y, Ohba N, Kanai A. REP-1 gene mutations in Japanese patients with choroideremia. Graefes Arch Clin Exp Ophthalmol 1999; 237:735-40.

21. McTaggart KE, Tran M, Mah DY, Lai SW, Nesslinger NJ, MacDonald IM. Mutational analysis of patients with the diagnosis of choroideremia. Hum Mutat 2002; 20:189-96.

22. Nesslinger N, Mitchell G, Strasberg P, MacDonald IM. Mutation analysis in Canadian families with choroideremia. Ophthalmic Genet 1996; 17:47-52.

23. Pascal O, Donnelly P, Fouanon C, Herbert O, Le Roux MG, Moisan JP. A new (old) deletion in the choroideremia gene. Hum Mol Genet 1993; 2:1489.

24. Schwartz M, Rosenberg T, van den Hurk JA, van de Pol DJ, Cremers FP. Identification of mutations in Danish choroideremia families. Hum Mutat 1993; 2:43-7.

25. Trujillo MJ, Sanz R, Rodriguez de Alba M, Lorda I, Ramos C, Ibanez A, Garcia-Sandoval B, Ayuso C. First mutation (S340X) in choroideremia gene in a Spanish family. Mutations in brief no. 173. Online. Hum Mutat 1998; 12:213.

26. van Bokhoven H, Schwartz M, Andreasson S, van den Hurk JA, Bogerd L, Jay M, Ruther K, Jay B, Pawlowitzki IH, Sankila EM, Wright A, Ropers HH, Rosenberg T, Cremers FPM. Mutation spectrum in the CHM gene of Danish and Swedish choroideremia patients. Hum Mol Genet 1994; 3:1047-51.

27. van den Hurk JA, Schwartz M, van Bokhoven H, van de Pol TJ, Bogerd L, Pinckers AJ, Bleeker-Wagemakers EM, Pawlowitzki IH, Ruther K, Ropers HH, Cremers FP. Molecular basis of choroideremia (CHM): mutations involving the Rab escort protein-1 (REP-1) gene. Hum Mutat 1997; 9:110-7.

28. van den Hurk JA, van de Pol DJ, Wissinger B, van Driel MA, Hoefsloot LH, de Wijs IJ, van den Born LI, Heckenlively JR, Brunner HG, Zrenner E, Ropers HH, Cremers FP. Novel types of mutation in the choroideremia (CHM) gene: a full-length L1 insertion and an intronic mutation activating a cryptic exon. Hum Genet 2003; 113:268-75.

29. MacDonald IM, Mah DY, Ho YK, Lewis RA, Seabra MC. A practical diagnostic test for choroideremia. Ophthalmology 1998; 105:1637-40.

30. Cremers FP, Armstrong SA, Seabra MC, Brown MS, Goldstein JL. REP-2, a Rab escort protein encoded by the choroideremia-like gene. J Biol Chem 1994; 269:2111-7.

31. Cremers FP, Molloy CM, van de Pol DJ, van den Hurk JA, Bach I, Geurts van Kessel AH, Ropers HH. An autosomal homologue of the choroideremia gene colocalizes with the Usher syndrome type II locus on the distal part of chromosome 1q. Hum Mol Genet 1992; 1:71-5.

32. Desnoyers L, Anant JS, Seabra MC. Geranylgeranylation of Rab proteins. Biochem Soc Trans 1996; 24:699-703.

33. Seabra MC, Mules EH, Hume AN. Rab GTPases, intracellular traffic and disease. Trends Mol Med 2002; 8:23-30.

34. Seabra MC, Ho YK, Anant JS. Deficient geranylgeranylation of Ram/Rab27 in choroideremia. J Biol Chem 1995; 270:24420-7.

35. Larijani B, Hume AN, Tarafder AK, Seabra MC. Multiple factors contribute to inefficient prenylation of Rab27a in Rab prenylation diseases. J Biol Chem 2003; 278:46798-804.

36. Chan D, Lin J, Raffaniello RD. Expression and localization of rab escort protein isoforms in parotid acinar cells from rat. J Cell Physiol 2000; 185:339-47.

37. Bernstein SL, Wong P. Regional expression of disease-related genes in human and monkey retina. Mol Vis 1998; 4:24 <http://www.molvis.org/molvis/v4/a24/>.

38. Syed N, Smith JE, John SK, Seabra MC, Aguirre GD, Milam AH. Evaluation of retinal photoreceptors and pigment epithelium in a female carrier of choroideremia. Ophthalmology 2001; 108:711-20.

39. Shi W, van den Hurk JA, Alamo-Bethencourt V, Mayer W, Winkens HJ, Ropers HH, Cremers FP, Fundele R. Choroideremia gene product affects trophoblast development and vascularization in mouse extra-embryonic tissues. Dev Biol 2004; 272:53-65.

40. van den Hurk JA, Hendriks W, van de Pol DJ, Oerlemans F, Jaissle G, Ruther K, Kohler K, Hartmann J, Zrenner E, van Bokhoven H, Wieringa B, Ropers HH, Cremers FP. Mouse choroideremia gene mutation causes photoreceptor cell degeneration and is not transmitted through the female germline. Hum Mol Genet 1997; 6:851-8.

41. Surace EM, Angeletti B, Ballabio A, Marigo V. Expression pattern of the ocular albinism type 1 (Oa1) gene in the murine retinal pigment epithelium. Invest Ophthalmol Vis Sci 2000; 41:4333-7.

42. Starr CJ, Kappler JA, Chan DK, Kollmar R, Hudspeth AJ. Mutation of the zebrafish choroideremia gene encoding Rab escort protein 1 devastates hair cells. Proc Natl Acad Sci U S A 2004; 101:2572-7.

43. Anand V, Barral DC, Zeng Y, Brunsmann F, Maguire AM, Seabra MC, Bennett J. Gene therapy for choroideremia: in vitro rescue mediated by recombinant adenovirus. Vision Res 2003; 43:919-26.

44. Bennett J, Wilson J, Sun D, Forbes B, Maguire A. Adenovirus vector-mediated in vivo gene transfer into adult murine retina. Invest Ophthalmol Vis Sci 1994; 35:2535-42.

45. Li T, Adamian M, Roof DJ, Berson EL, Dryja TP, Roessler BJ, Davidson BL. In vivo transfer of a reporter gene to the retina mediated by an adenoviral vector. Invest Ophthalmol Vis Sci 1994; 35:2543-9.

46. Auricchio A, Kobinger G, Anand V, Hildinger M, O'Connor E, Maguire AM, Wilson JM, Bennett J. Exchange of surface proteins impacts on viral vector cellular specificity and transduction characteristics: the retina as a model. Hum Mol Genet 2001; 10:3075-81.