Human Molecular Genetics, 2000, Vol. 9, No. 13 1919-1926
© 2000 Oxford University Press
Mutations in the N-terminus of the X-linked retinitis pigmentosa protein RP2 interfere with the normal targeting of the protein to the plasma membrane
Department of Pathology and 1Department of Molecular Genetics, Institute of Ophthalmology, University College London, 11–43 Bath Street, London EC1V 9EL, UK and 2Institute of Cancer Research, Chester Beatty Laboratories, London, UK
Received 2 May 2000; Revised and Accepted 23 June 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSACKOWLEDGEMENTSREFERENCES |
|---|
The X-linked retinitis pigmentosa (XLRP) gene, RP2, codes for a novel 350 amino acid protein of unknown function. We have identified putative sites for N-terminal acyl modification by myristoylation and palmitoylation in the RP2 protein. The RP2 protein is expressed ubiquitously in human tissues at relatively low levels (0.01% of total protein) and has a predominantly plasma membrane localization in cultured cells, as would be expected if the protein was subject to dual N-terminal acylation. Furthermore, mutagenesis of residues potentially required for N-terminal acylation prevents targeting of RP2 to the plasma membrane and the N-terminal 15 amino acids of the protein appear to be sufficient for this targeting. Our data suggest that the protein is dually acylated and that the palmitoyl moiety is responsible for targeting of the myristoylated protein from intracellular membranes to the plasma membrane. The effect of two mutations, which have been reported as causes of XLRP, R118H and
S6, were investigated. The R118H mutation does not affect the normal plasma membrane localization of RP2; in contrast, the S6 mutation interferes with the targeting of the protein to the plasma membrane. Therefore, the S6 mutation may cause XLRP because it prevents normal amounts of RP2 reaching the correct cellular locale, whereas the R118H mutation is in a region of the protein that is vital for another aspect of RP2 function in the retina. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSACKOWLEDGEMENTSREFERENCES |
|---|
X-linked retinitis pigmentosa (XLRP) is a severe form of retinal degeneration. Patients in the early stages of disease suffer from night blindness and constricted visual fields as a result of peripheral photoreceptor degeneration. As the disease progresses impairment of central vision occurs resulting in loss of visual acuity and blindness (1). The gene that causes one form of this heterogeneous X-linked disease, RP2, has recently been identified and shown to account for between 15 and 20% of XLRP (2). Screens of XLRP patients have identified >20 different pathogenic mutations in the RP2 gene including nonsense, missense, frameshift, insertion and deletion changes, with a prevalence towards protein truncation mutations (2–7).
The RP2 gene encodes a novel protein of 350 amino acids and its mRNA appears to be ubiquitously expressed (2). There are no functional data on this novel protein, and so far the only clues to its possible function are based on sequence analysis. The predicted amino acid sequence of RP2 was noted to have similarity to a previously identified protein, cofactor C, with 30.4% identity over 151 amino acids (2). Cofactor C is thought to act in the later stages of ß-tubulin folding (8,9) suggesting that RP2 may also have a role in tubulin biogenesis. The potential significance of this similarity is supported by the identification of pathogenic point mutations in RP2 at residues that are conserved with cofactor C (2,3). However, in addition to this reported similarity to cofactor C, we have identified potential sites for N-terminal acylation in the RP2 sequence and in this article we explore their functional significance (Fig. 1).
|
Myristoylation and palmitoylation are fatty acid modifications found at the N-termini of some proteins (reviewed in ref. 10). Myristate is a 14 carbon fatty acid that is covalently attached through an amide bond to the N-terminal glycine residue of proteins by N-myristoyl transferase (NMT) after cleavage of the initiating methionine. In addition to the essential glycine residue at amino acid position 2, RP2 has serine at position 6, lysine at position 7 and arginine at position 8, representing a potential recognition sequence for NMT (10). Myristoylation of the N-terminal glycine residue of a protein facilitates the attachment of the 16 carbon fatty acid palmitate through a thioester linkage to adjacent cysteine residues, normally at position 3. RP2 is therefore a candidate for both N-terminal myristoylation and palmitoylation.
Both myristoylation and palmitoylation are involved in the membrane association of proteins, including Src tyrosine kinases and G protein 
-subunits (10). Myristic and palmitic acid can insert hydrophobically into lipid bilayers attaching N-acylated proteins to the cytoplasmic face of the plasma membrane or other intracellular membranes. The dual acylation of Src tyrosine kinases and G protein -subunits has been shown to be essential in directing these proteins to the plasma membrane (11–13). It has been hypothesized that proteins with an N-terminal myristic acid transiently interact with multiple cellular membranes until they are palmitoylated by a plasma membrane-bound palmitoyl acyl transferase (PAT) and remain stably attached to the plasma membrane due to the dual fatty acid anchor (14). Experimental evidence has shown that mutations of the cysteine 3 palmitoylation site reduce the rate of membrane binding and redirect proteins to intracellular membranes (11,13).
Our major findings reported here are that RP2 is a ubiquitous protein that is predominantly localized to the plasma membrane of cultured cells, and mutagenesis of putative acylation sites prevents this targeting. In addition, we have examined the cellular localization of a deletion at serine 6 (S6) mutant, which has previously been reported in RP2 patients (2,5). The sequence requirement for the plasma membrane localization of RP2 is investigated and we consider the hypothesis that mutations in this region of the protein may cause retinitis pigmentosa due to aberrant protein localization and discuss the implications of our findings for the potential mechanisms of RP2 disease pathogenesis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSACKOWLEDGEMENTSREFERENCES |
|---|
Tissue and cellular expression of the RP2 protein
Human tissues were screened for RP2 protein expression by western analysis (Fig. 2A) using an affinity-purified sheep polyclonal serum (S974) raised against a recombinant human RP2 protein with a 9 x histidine tag, expressed and purified from Escherichia coli. The predicted molecular weight of the 350 amino acid RP2 is 39.6 kDa. A band of approximately this size was detected in all of the tissues screened. This was the only protein that S974 detected, except in heart where an additional smaller band (<20 kDa) was also present (data not shown). The pre-immune serum had no reactivity against human tissues by western blotting (data not shown). The levels of RP2 protein appeared to be similar in all of the tissues tested. From comparison with recombinant RP2 standards, we estimate that the RP2 protein represents ~0.01% of the total protein in human tissues. As RP2 appears to be ubiquitously expressed, we tested the human neuroblastoma cell line SH-SY5Y for the expression of the protein (Fig. 2B). Surprisingly, RP2 was present at significantly higher levels in SH-SY5Y cells than in any of the tissues investigated (~0.1% of total protein).
|
Serum S974 was tested for cross reactivity with RP2 orthologues from other species. The antibody detected a single protein band in Chinese hamster ovary (CHO) cells which was slightly smaller than human RP2 (see Fig. 6). This apparent RP2 orthologue was either present at very low levels or was recognized only weakly by the antisera compared with the human protein. The antibody also showed poor cross reactivity with mouse and rat tissues, although a protein band of similar size to human RP2 was again detected (data not shown).
|
Subcellular localization of RP2
We investigated the subcellular localization of RP2 in SH-SY5Y cells by immunofluorescence staining with S974 and confocal microscopy (Fig. 3A). RP2 was primarily localized to the plasma membrane of the cells. The intensity of plasma membrane staining was variable between cells and within the plasma membrane of individual cells. The most intense plasma membrane staining was visible where cells were in contact with each other, probably due to the close proximity of two plasma membranes, or possibly as a consequence of plasma membrane microdomains. In addition to the plasma membrane localization, a diffuse and slightly fibrous cytoplasmic stain was observed. Some cells also showed a punctate stain, which appeared to be in the nucleus. This staining pattern was observed with both methanol and formaldehyde fixations.
|
In order to confirm that RP2 was principally located in the plasma membrane of SH-SY5Y cells, subcellular fractionation was performed. Localization of RP2 was compared with the plasma membrane protein, neural cell adhesion molecule (N-CAM) (15,16). Following cell breakage and crude fractionation by centrifugation, levels of RP2 and N-CAM were analysed in pellet and supernatant fractions (Fig. 3B). RP2 and N-CAM were present in both fractions and both proteins were more abundant in the pellet than supernatant. The levels of each protein in these fractions appeared proportionally similar. This is consistent with this crude pellet fraction containing the majority of the plasma membrane, in addition to the nucleus and large cellular debris. To further investigate the co-partitioning of RP2 and N-CAM, the supernatant generated was fractionated using a well defined sucrose gradient centrifugation (SGC) method that we have previously characterized (17). The majority of RP2 partitioned with N-CAM in fractions 1–2 of the gradient (>38% sucrose). However, RP2 was also present at low levels in fractions throughout the gradient, in particular in fraction 16 (10% sucrose), demonstrating that not all of the protein was associated with the plasma membrane. To verify the membrane association of RP2 we treated an aliquot of the supernatant with the non-ionic detergent NP-40 prior to SGC. Detergent treatment significantly altered the distribution of RP2 in the gradient. The protein was exclusively in fractions 11–16 (<20% sucrose) of the gradient with a peak in fractions 14–15, demonstrating that RP2 in the high sucrose density fractions was detergent soluble. In summary, these data indicate that although RP2 is principally associated with the plasma membrane, it may also be associated with other membranes in the cell and/or the nucleus, but is also present in the cytoplasm.
Mutagenesis of putative acylation sites in RP2
To elucidate the mechanism of RP2 membrane association we investigated the effect of mutations in the putative N-acylation sites. Inspection of the N-terminal sequence of RP2 revealed a consensus myristoylation and palmitoylation motif (Fig. 1). The myristoyl moiety would be covalently attached to the glycine residue in the second position and the palmitoyl moiety to the cysteine residue in the third position. Full-length wild-type RP2 and three RP2 mutants glycine 2 to alanine (G2A), cysteine 3 to serine (C3S) and the double mutant G2A/C3S were cloned into the eucaryotic expression vector pBKCMV. In addition, RP2 with the pathogenic mutation arginine 118 to histidine (R118H) was created by site-directed mutagenesis and cloned into pBKCMV
The subcellular localization of the wild-type and mutant RP2 proteins was examined in CHO cells transfected with the different constructs and processed for immunofluorescence (Fig. 4). Under the conditions used, the staining intensity of endogenous CHO cell RP2 was not sufficient to be detected, although at higher antibody concentrations a similar pattern to that in SH-SY5Y cells was detected (data not shown). The wild-type protein (Fig. 4A and B) and R118H mutant protein (Fig. 4I and J) both localized predominantly to the plasma membrane, giving a similar staining pattern to that observed for endogenous human RP2 in SH-SY5Y cells. In contrast, the other mutant proteins did not localize to the plasma membrane. In cells transfected with the G2A (Fig. 4C and D) and G2A/C3S (Fig. 4G and H) mutations a diffuse cytoplasmic and nuclear staining was seen, showing that the RP2 protein was present throughout the cell. Cells expressing the C3S mutant RP2 (Fig. 4E and F) had a different staining pattern. C3S RP2 was not present in the nucleus, but was present exclusively in the cytoplasm with a staining pattern that was suggestive of RP2 being associated with intracellular membranes.
|
To further investigate the subcellular localization of the N-terminal RP2 mutants we generated stable cell lines using geneticin selection (Fig. 5A). Although CHO cells express a slightly smaller endogenous RP2 orthologue, it is detected weakly by antiserum S974 relative to the heterologous human RP2 and can only be detected if blots are exposed for longer periods of time. Basic subcellular fractionation of the stable cell lines revealed differences in the partitioning of RP2 wild-type and N-termini mutant proteins between pellets and supernatants (Fig. 5B). In the wild-type and C3S-expressing cells RP2 was more abundant in the pellet than supernatant fraction, whereas in the G2A- and G2A/C3S-expressing cells RP2 was more abundant in the supernatant fraction. Further washing of G2A and G2A/C3S pellets greatly reduced RP2 levels (data not shown), suggesting that in these cell lines RP2 was only loosely associated with the pellet. SGC of supernatant fractions from the cell lines also showed differences in the fractionation of wild-type and mutant proteins (Fig. 5C). Wild-type RP2 localized throughout the gradient in the same fractions as endogenous CHO cell RP2 (data not shown), although less endogenous RP2 than heterologous wild-type RP2 was present in fractions 14 and 15. This suggests that the levels of heterologous RP2 expression may be too high for efficient targeting of the protein to membranes and the excess may accumulate in the non-membrane fractions. The fractionation of endogenous RP2 in the gradient differed slightly between SH-SY5Y and CHO cells, probably as a consequence of altered membrane composition and shearing between the two cell lines not different RP2 processing. In our experience, these cell lines demonstrate subtle differences in their profile of membrane fractionation by SGC. G2A and G2A/C3S RP2 were present only in fractions 14–16 (<16% sucrose) of the gradient. The C3S mutant protein, like the wild-type protein, was present in fractions 14–15, and not fraction 16 (Fig. 5C). A longer exposure of the C3S blot suggested that a small amount of RP2 was present in other fractions (>16% sucrose), correlating with intracellular organelles.
|
These data demonstrate that mutations in putative N-acylation sites alter the subcellular membrane localization of RP2. They strongly suggest that the association of RP2 with the plasma membrane is a result of myristoylation and palmitoylation of the protein, whereas myristoylation alone targets the protein to intracellular membranes.
Subcellular localization of the pathogenic Ser6 RP2 mutant
A deletion of residue serine 6 (S6) has previously been reported in RP2 patients (2,5). Although the consensus sequence for N-terminal myristoylation is not absolute, it has been shown that a serine or threonine is preferred at amino acid position 6. Considering this observation and our results from the mutagenesis of putative acylation sites in RP2, we tested the effects of the S6 mutation on RP2 subcellular localization. RP2 S6 was created by site-directed mutagenesis and cloned into pBKCMV; however, because only low levels of mutant protein could be detected in transfected cells, immunofluorescence was not successful (data not shown). Therefore, a stable cell line expressing S6 RP2 was established. In these cells the levels of the S6 RP2 protein detected were equivalent to levels of the smaller endogenous CHO cell RP2 (Fig. 6) and thus appeared to be significantly lower than in wild-type and other mutant cell lines.
We investigated the localization of the S6 RP2 protein in the stable cell line using a modified simple subcellular fractionation (Fig. 6). RP2 levels were compared in the pellet and supernatant fractions of the S6 RP2 cells. Under these conditions, the endogenous hamster RP2 was detected only in the pellet fraction, whereas the S6 mutant protein was exclusively in the supernatant fraction (after washing of pellets). RP2 with a S6 mutation is, therefore, not associated with the same cellular compartment as wild-type RP2.
To confirm the cellular localization of S6 RP2, part of the N-terminal sequence of RP2 was fused to green fluorescent protein (GFP). The N-terminal 15 amino acids of wild-type and mutant RP2 were cloned in to the GFP expression vector pEGFP-N1, such that the peptide was tagged at its C-terminus with GFP. CHO cells were transfected with these constructs and confocal microscopy performed (Fig. 7). Wild-type RP2 (Fig. 7A) was mainly localized to the plasma membrane, whereas the S6 (Fig. 7E), G2A (Fig. 7B) and G2A/C3S (Fig. 7D) mutant proteins localized throughout the cell with strong staining in the nucleus. These staining patterns matched that observed when cells were transfected with GFP alone (Fig. 7F). In contrast, the C3S protein (Fig. 7C) localized in the cytoplasm and again appeared to be associated with intracellular membranes. These results indicate that S6 RP2 does not associate with the plasma membrane. They also demonstrate that the first 15 amino acids of RP2 are sufficient for membrane localization.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSACKOWLEDGEMENTSREFERENCES |
|---|
The RP2 protein is expressed at approximately the same level in all the human tissues that we examined, presenting no simple correlation between protein expression and disease pathogenesis. A major finding of this work is that a significant proportion of the RP2 protein is localized to the plasma membrane and that mutagenesis of the putative N-acylation sites prevents proper targeting. Results of N-terminal mutagenesis of RP2 are consistent with the protein being myristoylated and palmitoylated. A mutation at glycine 2 caused RP2 to be localized throughout the cell, whereas mutation at cysteine 3 resulted in a different staining pattern characteristic of attachment to intracellular membranes.
Many acylated proteins contain a combination of two membrane-anchoring features (10). This ‘two-signal’ mode of membrane binding of N-myristoylated proteins has been well investigated in the Src family of tyrosine kinases. Mutation analysis has shown that plasma membrane anchoring of p59fyn is dependent on the attachment of a myristoyl moiety at glycine 2 and a palmitoyl moiety at cysteine 3 (11). For example, the targeting of p59fyn to the plasma membrane has been reported to be conditional on palmitoylation of cysteine 3 (12) and that mutations of this residue have been shown to redirect the protein to intracellular membranes. Our results show that RP2 is also redirected from the plasma membrane to intracellular membranes on mutation of the cysteine 3 residue. Analysis of the contribution of the fatty acylated N-terminal domains from the Src family, GO and GAP-43 to subcellular localization has shown that GFP chimeras with a glycine 2/cysteine 3 motif localized to the plasma membrane and endosomal vesicles (13). Our results with GFP chimeras show that the N-terminal 15 amino acids of RP2 are sufficient for plasma membrane localization and confirm that the glycine 2/cysteine 3 motif is an absolute requirement. The minimal plasma membrane targeting region of RP2 remains to be fully delineated; however, we have demonstrated that it is contained within the 15 N-terminal amino acids, which is consistent with previous studies of other acylated proteins (13).
We also investigated the cellular localization of RP2 proteins containing mutations known to cause retinitis pigmentosa. The R118H mutation had no effect on the targeting of the protein to the plasma membrane. However, the S6 mutation was sufficient to inhibit membrane localization completely. Serine at position 6 is believed to be part of the preferred recognition sequence for NMT and it has been demonstrated that mutagenesis of serine 6 to alanine prevents both myristoylation and palmitoylation of the G protein Gi1 (18). In the first reported genotype–phenotype correlation for RP2 (5), RP2 patients with the R118H mutation appear to have a particularly aggressive form of disease. In contrast, the S6 mutation is associated with a milder phenotype, with delayed disease onset (5). Considering these clinical findings in the context of our data, the S6 mutation may affect the proper targeting of RP2 to the correct cellular location, whereas the R118H mutation affects the function of correctly localized protein. However, we also observed that the heterologously expressed full-length S6 mutant could only be detected at low levels when expressed in CHO cells. The mutant protein, therefore, may not accumulate to high levels in cells because it is misfolded and thus rapidly degraded. If this is also the case in the patients, then the correct targeting of RP2 to the membrane may not be essential for RP2 function in the retina, and the mild phenotype in these families could be a consequence of low levels of functional protein rather than improper targeting. However, the observation that S6 RP2 appeared to be present at low levels was based on immunofluorescence and western blotting using S974 and could also indicate that S974 cross-reacted poorly with this mutant. The GFP-tagged S6 mutant protein appeared to be present at normal levels, suggesting that the deletion of this residue does not in itself constitute a signal for proteolysis.
Several proteins including protein kinase A, recoverin and -transducin have been shown to be heterogeneously N-terminally modified with fatty acids in retinal photoreceptor cells (19,20). In the retina, proteins are modified at glycine 2 with the fatty acids 12:0, 14:0, 14:1n–9 and 14:2n–6 whereas similar proteins in other tissues are modified only with 14:0 (myristic acid). Therefore, if RP2 is N-terminally acylated in the retina, the fatty acids attached may be different to those in other tissues. This hypothesis may provide an explanation of why mutations in an apparently ubiquitously expressed protein cause a retinal-specific and not systemic disease, as it is possible that differential acylation of RP2 in retina results in the protein having a distinct localization or function. The results presented here clearly show that RP2 is associated with the plasma membrane through N-terminal acylation, it is important to clarify the nature of this N-terminal acylation, in particular in the retina.
In addition to mediating membrane associations, N-terminal acylation can have significant effects on protein function. For example, myristoylation can play a structural role in stabilizing three-dimensional protein conformation, and the orientation of the myristoyl moiety is not always static. Some N-myristoylated proteins exist in two conformations and the transition between these two states is regulated by a mechanism known as a ‘myristoyl switch’. This ‘switch’ can be triggered by ligand binding (e.g. recoverin), electrostatics (e.g. MARCKS) or proteolysis (e.g. HIV-1 Gag) (10). Whether RP2 undergoes such a ‘switch’ that mediates reversible membrane binding remains to be resolved, but a retinal-specific trigger may represent another possible mechanism for why mutations in this ubiquitous protein cause retinal degeneration.
Although the function of RP2 remains unknown it has a similarity over 150 amino acids to a previously described protein, cofactor C. Cofactor C was identified as a component of the ß-tubulin folding pathway (8) and RP2 is its first potential homologue. The production of native /ß-tubulin heterodimer depends on the action of the cytosolic chaperonin (CCT) and several cofactors, although the precise role of these cofactors is still unknown. The following model for the action of these cofactors in the post-chaperonin steps of tubulin folding has been proposed by Tian et al. (9): (i) after being released by CCT in a near-native state, the tubulin folding intermediates are captured by cofactors: A and D in the case of ß-tubulin; B and E in the case of -tubulin (or FAß FDß and FB and FE, respectively); (ii) the FB and FAß complexes act as reservoirs, capable of accepting or delivering their target protein to cofactors E and D, respectively; (iii) FE and FDß interact to form the species FE/FDß; and (iv) addition of cofactor C generates an /ß supercomplex, which hydrolyses GTP and produces native tubulin. This model, however, is controversial and other data on the function of these cofactors suggest that they may play a role in sequestering tubulin or modifying microtubule function. For example, the yeast homologue of cofactor A (Rbl2p) can sequester overexpressed ß-tubulin that would otherwise be lethal (21), whereas cofactor D in fission yeast (Alp1), in addition to assisting microtubule assembly, co-localizes with microtubules in cells and co-sediments with porcine brain microtubules in vitro (22), suggesting that it may be a microtubule-associated protein (MAP). Indeed, cofactor C may have other roles than that proposed in the above model. Unpublished data cited by Tian et al. (8) suggest that cofactor C is a chaperone, can associate with microtubules, and may itself be a MAP.
It is unlikely that RP2 functions exclusively in tubulin folding, as we have shown that it is primarily plasma membrane associated. However, it is possible that RP2 does still interact with tubulin and/or microtubules. Several proteins with different functions have been shown to interact with both the plasma membrane and microtubules. G proteins serve many functions involving the transfer of signals from the cell surface receptors to intracellular effector molecules and some G protein subunits are N-terminally acylated and interact with microtubules. Gi1, for example, is myristoylated and palmitoylated and is also able to interact with tubulin, inhibiting the polymerization of tubulin–GTP into microtubules (23). It has also been reported that proteins modified by saturated acyl groups are targeted to detergent-resistant membrane rafts in mammalian cells (24) and tubulin can be anchored to these rafts (25). Therefore, RP2 may still act as a MAP, possibly as a linker between membranes and the cytoskeleton facilitating membrane traffic, through the motif that is conserved with cofactor C.
The region of RP2 that is similar to cofactor C includes the R118H mutation and the arginine residue appears to be conserved in cofactor C across several species. Although it seems unlikely that RP2 is involved in tubulin biogenesis, this conserved motif may act as a common tubulin or microtubule binding site. In the retina this may include an association with the connecting cilia of photoreceptors, an organelle that has been shown to be of critical importance for photoreceptor viability. The fine structural localization of RP2 in the retina should help to clarify whether this is the case. Alternatively, the similarity between RP2 and cofactor C may reflect some other, previously unstudied, aspect of cofactor C function or a diversification and specialization of function from cofactor C.
In this report, we describe the first steps towards understanding the function of the RP2 protein and the consequences of mutations on the localization of RP2. Specifically, we have demonstrated that mutations in the N-terminus interfere with the normal targeting of the protein to the plasma membrane. The major challenge now is to ascertain more of the function of the protein, assess the significance of the similarity to cofactor C and evaluate its specific role in the retina.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSACKOWLEDGEMENTSREFERENCES |
|---|
Construction of plasmids
Full-length RP2 cDNA was amplified by PCR from a human brain cDNA library (Clontech, Palo Alto, CA) and cloned into the pGEMT vector, using oligonucleotide primers based on the untranslated regions of the mRNA. Restriction endonuclease recognition sites appropriate for subcloning and a 9 x histidine tag were introduced by using modified primers for PCR from pGEMT RP2 and subsequently cloned into the NcoI and BamHI site of pTrcHis, or the BamHI site of the expression vector pBKCMV. For mutagenesis of the coding region, RP2 was PCR amplified from the pGEMT RP2 construct using 5' primers where sequence alterations had been introduced to produce the required amino acid sequence changes G2A, C3S, G2A/C3S and
S6. The R118H mutation was introduced by site-directed PCR-mediated mutagenesis. The 15 amino acid N-terminal sequence of RP2 was cloned into the BamHI–AgeI site of pEGFP-N1 using annealed phosphorylated oligonucleotides. A BamHI site was introduced in the 5' end of the oligonucleotide and an AgeI site at the 3' end. This stratagem was also used to introduce the G2A, C3S, G2A/C3S and S6 mutations. All constructs were confirmed by sequencing.
Production of antisera and western blotting
Recombinant C-terminal 9 x histidine tagged RP2 expressed from pTrcHis was purified by metal affinity chromatography using Talon (Clontech) and used to produce a sheep polyclonal anti-RP2 serum, S974 (Scottish Antibody Production Unit, Carluke, UK). This serum was affinity purified using recombinant RP2 conjugated to activated Sepharose (Amersham Pharmacia, Little Chalfont, UK). The affinity-purified serum was used in western blotting at a titre of 1:500. Monoclonal anti-N-CAM (clone no. NCAM-OB11) was used at the recommended titre (Sigma, Poole, UK). Immune complexes were visualized by ECL detection (Amersham Pharmacia). Standards containing known amounts of the recombinant histidine-tagged RP2 were included on blots.
Preparation of tissue homogenates and SH-SY5Y lysate
Human tissues and SH-SY5Y cells were Dounce homogenized on ice in 20 mM Tris–HCl pH 7.5, 500 mM NaCl, 12.5 mM KCl, 1 mM EDTA, 1 mM dithiothreitol, containing a protease inhibitor cocktail (Sigma). The concentration of protein in homogenates was determined using the Bio-Rad (Hemel Hempstead, UK) DC assay.
Cell line maintenance and construction
The neuroblastoma cell line SH-SY5Y and CHO cells were grown in Dulbecco’s modified Eagle’s medium/F12 with 10% fetal calf serum. Cell lines expressing wild-type and mutant full-length RP2 were made based on the parent CHO cell line. Briefly, cells were transfected using calcium phosphate with pBKCMV containing wild-type or mutant RP2 downstream of the cytomegalovirus promoter. Stable cell lines were selected using geneticin, G418 (500 µg/ml). CHO cells were also transfected with RP2–GFP chimeras using calcium phosphate.
Confocal microscopy of cells expressing RP2 and RP2–GFP chimeras
Cells were seeded at 1 x 105 cells/ml into chamber slides, 24 h after seeding medium was replaced, and after 48 h cells were fixed in either ice-cold methanol or 3.7% (v/v) formaldehyde followed by detergent permeablization with 0.05% (v/v) Triton X-100. SHSY-5Y and CHO cells were then processed for confocal microscopy. Transfections were performed as described above, 24 h after seeding. For immunofluorescence affinity-purified S974 was used at a titre of 1:250 and fluorescein isothiocyanate-conjugated anti-sheep secondary antibody (Dako, Cambridge, UK) at a titre of 1:50. Cells were counterstained with the nuclear stain DAPI. The fluorescence was detected using a Zeiss laser scanning confocal microscope.
Basic subcellular fractionation
Cells were washed twice and scraped into phosphate-buffered saline (PBS). They were pelleted by centrifugation (1000 g) and resuspended in breaking buffer (250 mM sucrose, 10 mM triethanolamine, 10 mM acetic acid, 1 mM EDTA, pH to7.45 with NaOH, Sigma protease inhibitor cocktail). The cells were then allowed to swell for 10 min prior to breaking by passing 30 times through a steal ball bearing cell homogenizer (HY Enterprises, Redwood City, CA). After breakage, cells were centrifuged at 1850 g, and the supernatant fraction removed. The pellet was resuspended in an equal volume of buffer prior to comparison of pellet and supernatants by western analysis. The S6 experiment was as above with the following modifications: cell breakage was by Dounce homogenization in PBS (with Sigma protease inhibitor cocktail), with centrifugation at 12 000 g for 20 min prior to washing the pellet once with PBS.
Sucrose gradient centrifugation
Supernatant fractions from the basic subcellular fractionation were further separated on a linear sucrose gradient using a technique that produces a previously well characterized gradient (17). Briefly, a 14 ml 10–40% (w/w) sucrose gradient in 50 mM HEPES pH 7.2, 90 mM KCl was prepared with a 65% (w/v) sucrose cushion. Supernatants (1 ml) were loaded on to sucrose gradients and centrifuged using a swing-out rotor at an average of 78 800 g for 16 h at 4°C. For detergent treatment NP-40 was added to the supernatant fraction to a final concentration of 0.5%, 10 min prior to loading onto the gradient. After centrifugation 1 ml fractions were collected starting from the bottom of the tube (65% sucrose). Fractions were stored at –70°C until western analysis.
|
ACKOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSACKOWLEDGEMENTSREFERENCES |
|---|
We are grateful to the Brain Bank, Department of Neuropathology, Institute of Psychiatry, for providing human tissue samples and to Dr Peter Munro for his assistance with confocal microscopy. This work was supported by the Wellcome Trust and Fight for Sight; C.G. is a Fight for Sight Prize student.
|
FOOTNOTES |
|---|
+ To whom correspondence should be addressed. Tel: +44 20 7608 6944; Fax: +44 20 7608 6862; Email: michael.cheetham@ucl.ac.uk
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSACKOWLEDGEMENTSREFERENCES |
|---|
1 Bird, A.C. (1975) X-linked retinitis pigmentosa. Br. J. Ophthalmol., 59, 177–199.
2 Schwahn, U., Lenzner, S., Dong, J., Feil, S., Hinzmann, B., van Duijnhoven, G., Kirschner, R., Hemberger, M., Bergen, A.A., Rosenberg, T. et al. (1998) Positional cloning of the gene for X-linked retinitis pigmentosa 2. Nature Genet., 19, 327–332.[Web of Science][Medline]
3 Hardcastle, A.J., Thiselton, D.L., Van Maldergem, L., Saha, B.K., Jay, M., Plant, C., Taylor, R., Bird, A.C. and Bhattacharya, S.S. (1999) Mutations in the RP2 gene cause disease in 10% of families with familial X-linked retinitis pigmentosa assessed in this study. Am. J. Hum. Genet., 64, 1210–1215. [Web of Science][Medline]
4 Mears, A.J., Gieser, L., Yan, D., Chen, C., Fahrner, S., Hiriyanna, S., Fujita, R., Jacobson, S.G., Sieving, P.A. and Swaroop, A. (1999) Protein-truncation mutations in the RP2 gene in a North American cohort of families with X-linked retinitis pigmentosa. Am. J. Hum. Genet., 64, 897–900.[Web of Science][Medline]
5 Rosenberg, T., Schwahn, U., Feil, S. and Berger, W. (1999) Genotype-phenotype correlation in X-linked retinitis pigmentosa 2 (RP2). Ophthalmic Genet., 20, 161–172. [Medline]
6 Thiselton, D.L., Zito, I., Plant, C., Jay, M., Hodgson, S.V., Bird, A.C., Bhattacharya, S.S. and Hardcastle, A.J. (2000) Novel frameshift mutations in the RP2 gene and polymorphic variants. Hum. Mutat., 15, 580.
7 Wada, Y., Nakazawa, M., Abe, T. and Tamai, M. (2000) A new Leu253Arg mutation in the RP2 gene in a Japanese family with X-linked retinitis pigmentosa. Invest. Ophthalmol. Vis. Sci., 41, 290–293.
8 Tian, G., Huang, Y., Rommelaere, H., Vandekerckhove, J., Ampe, C. and Cowan, N.J. (1996) Pathway leading to correctly folded beta-tubulin. Cell, 86, 287–296.[Web of Science][Medline]
9 Tian, G., Lewis, S.A., Feierbach, B., Stearns, T., Rommelaere, H., Ampe, C. and Cowan, N.J. (1997) Tubulin subunits exist in an activated conformational state generated and maintained by protein cofactors. J. Cell Biol., 138, 821–832.
10 Resh, M.D. (1999) Fatty acylation of proteins: new insights into membrane targeting of myristoylated and palmitoylated proteins. Biochim. Biophys. Acta, 1451, 1–16.[Medline]
11 van’t Hof, W. and Resh, M.D. (1997) Rapid plasma membrane anchoring of newly synthesized p59fyn: selective requirement for NH2-terminal myristoylation and palmitoylation at cysteine-3. J. Cell Biol., 136, 1023–1035.
12 Wolven, A., Okamura, H., Rosenblatt, Y. and Resh, M.D. (1997) Palmitoylation of p59fyn is reversible and sufficient for plasma membrane association. Mol. Biol. Cell, 8, 1159–1173.[Abstract]
13 McCabe, J.B. and Berthiaume, L.G. (1999) Functional roles for fatty acylated amino terminal domains in subcellular localization. Mol. Biol. Cell, 10, 3771–3786.
14 Shahinian, S. and Silvius, J.R. (1995) Doubly-lipid-modified protein sequence motifs exhibit long-lived anchorage to lipid bilayer membranes. Biochemistry, 34, 3813–3822.[Medline]
15 Barbas, J.A., Chaix, J.C., Steinmetz, M. and Goridis, C. (1988) Differential splicing and alternative polyadenylation generates distinct NCAM transcripts and proteins in the mouse. EMBO J., 7, 625–632.[Web of Science][Medline]
16 Walsh, F.S. and Dickson, G. (1989) Generation of multiple N-CAM polypeptides from a single gene. Bioessays, 11, 83–88.[Web of Science][Medline]
17 Lewis, V.A., Hynes, G.M., Zheng, D., Saibil, H. and Willison, K. (1992) T-complex polypeptide-1 is a subunit of a heteromeric particle in the eukaryotic. Nature, 358, 249–252.[Medline]
18 Galbiati, F., Guzzi, F., Magee, A.I., Milligan, G. and Parenti, M. (1996) Chemical inhibition of myristoylation of the G-protein Gi1 alpha by 2-hydroxymyristate does not interfere with its palmitoylation or membrane association. Evidence that palmitoylation, but not myristoylation, regulates membrane attachment. Biochem. J., 313, 717–720.
19 DeMar Jr, J.C., Wensel, T.G. and Anderson, R.E. (1996) Biosynthesis of the unsaturated 14-carbon fatty acids found on the N termini of photoreceptor-specific proteins. J. Biol. Chem., 271, 5007–5016.
20 DeMar Jr, J.C. and Anderson, R.E. (1997) Identification and quantitation of the fatty acids composing the CoA ester pool of bovine retina, heart, and liver. J. Biol. Chem., 272, 31362–31368.
21 Archer, J.E., Vega, L.R. and Solomon, F. (1995) Rbl2p, a yeast protein that binds to beta-tubulin and participates in microtubule function in vivo. Cell, 82, 425–434.[Web of Science][Medline]
22 Hirata, D., Masuda, H., Eddison, M. and Toda, T. (1998) Essential role of tubulin-folding cofactor D in microtubule assembly and its association with microtubules in fission yeast. EMBO J., 17, 658–666.[Web of Science][Medline]
23 Roychowdhury, S., Panda, D., Wilson, L. and Rasenick, M.M. (1999) G protein alpha subunits activate tubulin GTPase and modulate microtubule polymerization dynamics. J. Biol. Chem., 274, 13485–13490.
24 Melkonian, K.A., Ostermeyer, A.G., Chen, J.Z., Roth, M.G. and Brown, D.A. (1999) Role of lipid modifications in targeting proteins to detergent-resistant membrane rafts. Many raft proteins are acylated, while few are prenylated. J. Biol. Chem., 274, 3910–3917.
25 Palestini, P., Pitto, M., Tedeschi, G., Ferraretto, A., Parenti, M., Brunner, J. and Masserini, M. (2000) Tubulin anchoring to glycolipid-enriched, detergent-resistant domains of the neuronal plasma membrane. J. Biol. Chem., 275, 9978–9985.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Prokisch, M. Hartig, R. Hellinger, T. Meitinger, and T. Rosenberg A Population-Based Epidemiological and Genetic Study of X-Linked Retinitis Pigmentosa Invest. Ophthalmol. Vis. Sci., September 1, 2007; 48(9): 4012 - 4018. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Garcia-Hoyos, B. Garcia-Sandoval, D. Cantalapiedra, R. Riveiro, I. Lorda-Sanchez, M. J. Trujillo-Tiebas, M. Rodriguez de Alba, J. M. Millan, M. Baiget, C. Ramos, et al. Mutational Screening of the RP2 and RPGR Genes in Spanish Families with X-Linked Retinitis Pigmentosa. Invest. Ophthalmol. Vis. Sci., September 1, 2006; 47(9): 3777 - 3782. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Barnes, J. Richards, M. Mancl, S. Hanash, L. Beretta, and P. M. Pitha Global and Distinct Targets of IRF-5 and IRF-7 during Innate Response to Viral Infection J. Biol. Chem., October 22, 2004; 279(43): 45194 - 45207. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S S Dandekar, N D Ebenezer, C Grayson, J P Chapple, C A Egan, G E Holder, S A Jenkins, F W Fitzke, M E Cheetham, A R Webster, et al. An atypical phenotype of macular and peripapillary retinal atrophy caused by a mutation in the RP2 gene Br J Ophthalmol, April 1, 2004; 88(4): 528 - 532. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I Zito, S M Downes, R J Patel, M E Cheetham, N D Ebenezer, S A Jenkins, S S Bhattacharya, A R Webster, G E Holder, A C Bird, et al. RPGR mutation associated with retinitis pigmentosa, impaired hearing, and sinorespiratory infections J. Med. Genet., August 1, 2003; 40(8): 609 - 615. [Full Text]
|
|
|
|
|
|
J. P. Chapple and M. E. Cheetham The Chaperone Environment at the Cytoplasmic Face of the Endoplasmic Reticulum Can Modulate Rhodopsin Processing and Inclusion Formation J. Biol. Chem., May 23, 2003; 278(21): 19087 - 19094. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Bader, O. Brandau, H. Achatz, E. Apfelstedt-Sylla, M. Hergersberg, B. Lorenz, B. Wissinger, B. Wittwer, G. Rudolph, A. Meindl, et al. X-linked Retinitis Pigmentosa: RPGR Mutations in Most Families with Definite X Linkage and Clustering of Mutations in a Short Sequence Stretch of Exon ORF15 Invest. Ophthalmol. Vis. Sci., April 1, 2003; 44(4): 1458 - 1463. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Grayson, F. Bartolini, J. P. Chapple, K. R. Willison, A. Bhamidipati, S. A. Lewis, P. J. Luthert, A. J. Hardcastle, N. J. Cowan, and M. E. Cheetham Localization in the human retina of the X-linked retinitis pigmentosa protein RP2, its homologue cofactor C and the RP2 interacting protein Arl3 Hum. Mol. Genet., November 15, 2002; 11(24): 3065 - 3074. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Saliba, P. M. G. Munro, P. J. Luthert, and M. E. Cheetham The cellular fate of mutant rhodopsin: quality control, degradation and aggresome formation J. Cell Sci., July 15, 2002; 115(14): 2907 - 2918. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Chapple, A. J. Hardcastle, C. Grayson, K. R. Willison, and M. E. Cheetham Delineation of the Plasma Membrane Targeting Domain of the X-Linked Retinitis Pigmentosa Protein RP2 Invest. Ophthalmol. Vis. Sci., June 1, 2002; 43(6): 2015 - 2020. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Bartolini, A. Bhamidipati, S. Thomas, U. Schwahn, S. A. Lewis, and N. J. Cowan Functional Overlap between Retinitis Pigmentosa 2 Protein and the Tubulin-specific Chaperone Cofactor C J. Biol. Chem., April 19, 2002; 277(17): 14629 - 14634. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C Grayson, J P Chapple, K R Willison, A R Webster, A J Hardcastle, and M E Cheetham In vitro analysis of aminoglycoside therapy for the Arg120stop nonsense mutation in RP2 patients J. Med. Genet., January 1, 2002; 39(1): 62 - 67. [Full Text] [PDF]
|
|
|
|
|
|
U. Schwahn, N. Paland, S. Techritz, S. Lenzner, and W. Berger Mutations in the X-linked RP2 gene cause intracellular misrouting and loss of the protein Hum. Mol. Genet., May 1, 2001; 10(11): 1177 - 1183. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||