Human Molecular Genetics, 2000, Vol. 9, No. 20 3065-3073
© 2000 Oxford University Press
Functional characterization of missense mutations at codon 838 in retinal guanylate cyclase correlates with disease severity in patients with autosomal dominant cone–rod dystrophy
Department of Molecular Genetics, Institute of Ophthalmology, University College London, Bath Street, London EC1V 9EL, UK, 1School of Biological Sciences, Queen Mary University of London, Mile End Road, London E1 4NS, UK and 2Howard Hughes Medical Institute and Department of Biochemistry, Box 357370, University of Washington, Seattle, WA 98195, USA
Received 22 August 2000; Revised and Accepted 4 October 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Three different mutations in codon 838 of GUCY2D, the gene for retinal guanylate cyclase 1, have been linked to autosomal dominant cone–rod dystrophy at the CORD6 locus. To examine the relationship between enzyme activity and disease severity, the three disease-causing substitutions (R838C, R838H and R838S) and four artificial mutations (R838A, R838E, R838L and R838K) were generated. Assay of GCAP1-stimulated cyclase activity in vitro shows that, compared with wild-type, R838E, R838L and R838K possess only low activity, whereas R838A, R838C, R838H and R838S have activity equal or superior to wild-type at low Ca2+ concentrations. These four latter mutants showed a higher apparent affinity for GCAP1 than did wild-type. The Ca2+ sensitivity of the GCAP1 activation was also altered with marked residual activity at high Ca2+, the effect increasing: wild-type < R838C < R838H << R838A < R838S. Within the photoreceptor, this would result in a failure to inactivate cyclase activity at high physiological Ca2+ concentrations. Amongst the three disease-associated mutations, the effect correlates directly with disease severity. The wild-type and R838H mutant displayed a difference in pH sensitivity, with the mutant showing a higher specific activity with pH > 6.0. Site 838 is in the dimerization domain that forms a coiled-coil in the active protein. A computer-aided structure prediction of this region indicates that R838 in the wild-type breaks the structure at four helical turns, and there is an increasing tendency for the structure to continue for further turns in the order R838C < R838H,S,K << R838E < R838A < R838L.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Autosomal dominant cone–rod dystrophy (adCORD) is a distinct type of chorioretinal disease causing initial degeneration of cone photoreceptors followed by loss of rods (1,2). The disease is characterized by an early loss of visual acuity and colour discrimination associated with loss of cones, followed by nyctalopia and progressive peripheral field loss as the rods subsequently degenerate (3). These clinical features are typically accompanied by electroretinographic abnormalities that reveal early loss of photopic (cone) responses and later, progressive loss of scotopic (rod) responses. Molecular genetic analysis has linked adCORD with mutations in GUCY2D, the gene for retinal guanylate cyclase 1 (RetGC1), on chromosome 17p12–p13 (CORD6) (1).
RetGC1 is one member of a pair of membrane-bound guanylate cyclases, RetGC1 and RetGC2, which synthesize cGMP from GTP in mammalian photoreceptor cells. The membrane potential and signalling states of rod and cone cells are controlled by cGMP-gated cation channels in the plasma membrane. Light stimulates degradation of cGMP, causing the cation channels to close, resulting in a lowering of the Na+ and Ca2+ concentrations, hyperpolarization of the cell and a slowing of neurotransmitter release. The lowered Ca2+ concentration then allows the Ca2+ binding guanylate cyclase activator proteins (GCAP1, GCAP2 and GCAP3) to stimulate the RetGCs and restore the cGMP level. As a result the cation channels reopen and photosensitivity is restored to the cell.
The membrane-bound guanylate cyclases all comprise an extracellular domain, a small transmembrane domain and an intracellular domain consisting of a kinase homology domain, a dimerization domain and a catalytic domain. Four adCORD mutations in RetGC1 have been described to date: two single, missense mutations, R838C (1) and R838H (4) (A.M. Payne, A.G. Morris, S.M. Downes, A.C. Bird, A.T. Moore, S.S. Bhattacharya and D.M. Hunt, in preparation), a double mutation E837D/R838S (4,5) and a triple mutation E837D/R838C/T839M (2). All these mutations occur in a three codon sequence within the putative dimerization domain of the protein and all include a mutation in codon 838. A study of a panel of 90 unrelated cone, cone–rod and macular dystrophy patients revealed that as many as 6.7% had mutations in codon 838 of RetGC1, which appears to represent a mutation hot spot (A.M. Payne et al., in preparation). A biochemical analysis of the R838C mutation indicated that it appears to increase the apparent affinity of RetGC1 for GCAP1 and alters the Ca2+ sensitivity of the GCAP1 response, allowing the mutant to be stimulated by GCAP1 at higher Ca2+ concentrations than the wild-type (6). Other mutations elsewhere in the RetGC1 gene have been linked to the more severe, autosomal recessive retinopathy, Leber congenital amaurosis (LCA1) (7).
Detailed clinical assessment of patients with the R838C, R838H and E837D/R838S mutations have revealed variations in phenotype associated with the different mutations. Patients with the E837D/R838S mutation exhibit the most severe phenotype with early onset of the disease (in the first decade) and marked photophobia, nyctalopia and varying degrees of peripheral field loss and associated electrophysiological evidence of moderate to severe rod involvement by the fourth decade (5). In contrast, patients with the R838C and R838H mutations display a milder phenotype, with many remaining asymptomatic until their 20s and, even by the fourth decade, most displaying a predominantly cone dystrophy with minimal rod loss (S.M. Downes, A.M. Payne, R. Kelsell, F.F.W. Fitzke, G.E. Holder, M. Warren, D. Hunt, A.T. Moore and A.C. Bird, in preparation).
Activation of RetGC1 by GCAP involves dimerization of two RetGC1 monomers (8). In this study, a series of experiments has been undertaken to investigate whether a correlation can be made between the amino acid substitution at position 838 in RetGC1 and its effect on the activity of the enzyme. Point mutations at position 838 corresponding to the three disease-causing mutations found in adCORD patients (R838C, R838H and R838S) as well as a number of other ‘artificial’ mutations (R838A, R838E, R838L and R838K) were therefore generated and expressed. In this report we focus on the variation in activation of the RetGC1 R838 mutants by GCAP1, since GCAP1 appears to be the main activator of RetGC1 in photoreceptor outer segments (9).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The adCORD mutations examined in this study all occur at position 838 in the dimerization domain of RetGC1, a region encompassing amino acids 817–857. A series of point mutations was generated in which both conservative and non-conservative substitutions were made for the basic R838 residue present in wild-type RetGC1. Thus, R838 was substituted by polar residues cysteine and serine, non-polar residues alanine and leucine, the acidic residue glutamate, another basic residue lysine and the imidazole-containing residue histidine. Wild-type RetGC1 and the R838 mutants were cloned into the expression plasmid pRC-CMV, expressed in human embryonic kidney (HEK) 293-T cells and the expressed proteins collected in membrane preparations.
In this study the objective was to compare the functional activity of each of the R838 mutants with that of the wild-type RetGC1 and it was thus important to express and characterize the recombinant proteins under identical conditions. In order to minimize preparation variations arising, for example, from differences in the transfection efficiency of the cells, wild-type RetGC1 and the complete series of mutants were expressed in aliquots of the same 293-T cells and purified simultaneously. The expression level of RetGC1 protein in each preparation in comparison with the wild-type preparation was then estimated semi-quantitatively by western analysis using an antibody to the kinase homology domain of RetGC1 (10) (Fig. 1). This indicated only slight differences in expression levels between mutants and these differences were not consistent between preparations made on different dates. By conducting three series of assays using preparations made on different dates and estimating mean activity values, the influence of minor variations in protein expression was minimized. As a further safeguard against possible loss of activity of membrane preparations on storage, activity assays on proteins from each series of membrane preparations were conducted simultaneously using equal amounts of preparation and within 4 days of the preparation being made.
|
Overall RetGC activity varies widely with the identity of the 838 substitution
The overall catalytic activity of wild-type RetGC1 and each mutant was first estimated as: (i) basal activity (in the absence of Ca2+ and GCAP1); (ii) activity in the presence of 8 µM GCAP1 at low Ca2+; and (iii) activity in the presence of Mn2+/Triton X-100 (which activates membrane guanylate cyclases constitutively and provides an estimate of general catalytic ability) (11). For each series of membrane preparations, the maximum absolute activity was obtained for wild-type RetGC1 in the presence of Mn2+/Triton X-100. Thus, for the purposes of comparison, basal, GCAP1-stimulated and Mn2+-stimulated activities for wild-type and each mutant were calculated for each series of preparations with reference to this value, and mean values of relative RetGC1 activity from three independent series of assays were then calculated in each case. The results are shown in Figure 2. Basal activity and Mn2+-stimulated activity of all the mutants was reduced to at best 50% of wild-type, with the R838L mutant showing very little activity and the R838C, R838E and R838K mutants having
20% or less activity with respect to wild-type. However, in the presence of GCAP1, membrane preparations containing R838A, R838H or R838S had a specific activity which was higher or at least equal to wild-type, whereas mutant R838C had only slightly lower activity. Mutants R838L, R838E and R838K had a greatly reduced specific activity, indicating that these mutations prevent maximal stimulation of the enzyme. Further studies were confined to the four mutants having similar or greater activity than the wild-type, namely R838A, R838C, R838H and R838S. It is important to note that the latter three are the disease-causing mutations.
|
Maximum specific activation of R838A, R838H and R838S mutants by GCAP1 is greater than that of wild-type
A titration of RetGC1 activity against increasing GCAP1 concentration in the presence of a high substrate (GTP) concentration demonstrated that the initial rate of cGMP formation reached a maximum (saturated) value at high GCAP1 concentrations (Fig. 3). The R838C mutant had a similar maximum specific activity (Vmax) to wild-type, whereas the remaining mutants, R838A, R838H and R838S, had Vmax values that were substantially higher (150% of wild-type) (Table 1). Estimates of the GCAP1 concentration required for half maximal activation (Kmapp[GCAP1]) obtained by non-linear curve fitting to the relation v = Vmax [GCAP1]/(Kmapp[GCAP1]+[GCAP1]) showed a trend in which the mutants had lower values than wild-type with wild-type > R838C > R838S > R838A > R838H (Table 1). In the case of R838H, the Kmapp[GCAP1] value indicated a 5-fold greater apparent affinity for GCAP1 than wild-type.
|
|
R838 mutants are less sensitive to Ca2+ concentration than wild-type
RetGC1 activity in the retina is controlled by the free Ca2+ concentration in the cell via GCAP1, a Ca2+-binding protein. Wild-type RetGC1 is activated by GCAP1 at Ca2+ concentrations below
300 nM (12,13) and inhibited at micromolar concentrations (14). It was previously reported (6) that, although the R838C mutant behaved similarly to wild-type at free Ca2+ concentrations below 500 nM, it behaved anomalously above 500 nM in showing only a gradual decline in activity and, even at micromolar free Ca2+ concentrations, GCAP1 was still stimulatory. Ca2+ titrations for wild-type RetGC1 and the R838A, R838C, R838H and R838S mutants are shown in Figure 4. For wild-type RetGC1, the free Ca2+ concentration for half-maximal activity (ED50) was estimated to be 420 nM, whereas the value for the R838C mutant was shifted slightly but reproducibly to 530 nM. The R838H mutant showed a rather larger shift to 800 nM, whereas the R838A and R838S mutants were drastically shifted to above 1.0 µM. In all cases, the mutants displayed a significant ‘tail’ of elevated activity at high Ca2+. The actual free Ca2+ concentration range operating in human rods and cones has yet to be determined, but in dark-adapted rods of lower vertebrates the free Ca2+ concentration is near 550 nM and it decreases to near 50 nM after strong illumination (15). Submicromolar concentrations of free Ca2+ are therefore considered to be physiologically relevant. The results presented here indicate that at a free Ca2+ concentration of 1 µM, wild-type RetGC1 is almost completely inhibited, whereas even mutant R838C which shows the smallest shift retains >16% of maximal activity.
|
Comparison with the Y99C mutant of GCAP1
The Y99C mutation of GCAP1 has been linked to autosomal dominant cone dystrophy, a condition related to adCORD but with no detectable rod involvement (16,17). The Y99C mutation has been shown to result in the inactivation of EF3 in GCAP1, leading to similar constitutive RetGC1 activity at high free calcium concentrations. In order to compare the magnitude of the effects of the two classes of mutation, a calcium titration of the activation of wild-type RetGC1 by the Y99C mutant of human GCAP1 is presented in Figure 5. This titration curve is almost identical to those for the R838A and R838S mutations. The GCAP1 mutation does not, however, result in any increase in maximal specific activation of RetGC1 at low calcium concentrations (data not shown) unlike R838A and R838S.
|
Effect of pH on activity of wild-type and R838H RetGC1
The effect of pH on the activity of RetGC1 was investigated in broad terms with both the wild-type and R838H mutant. The objective here was to probe the ionization state of the imidazole group in the R838H mutant. Since the pKa of this group in most proteins is
6.0, raising the pH from below 6.0 to 8.0 should result in a change in the ionization state of the group from positively charged to uncharged and might be expected to alter the pH activity profile. In practice the activity of both wild-type and mutant RetGC1 increased with pH reaching a maximum pH 8.0, but the maximum specific activity of the mutant was much higher (150% compared with that of wild-type) (Fig. 6).
|
Effect of amino acid modifying reagents on wild-type and R838H RetGC1 activity
Chemical modification of the residue at position 838 might be expected to interfere with dimerization of the enzyme and hence reduce activity. Both diethylpyrocarbonate (which specifically modifies histidine residues) and butadione (which specifically modifies arginine residues) reduced RetGC1 activity and yielded typical dose–response curves (data not shown). However, no differences between wild-type and the R838H mutant enzyme were observed. It is concluded from this that critical arginine and histidine residues elsewhere in the protein, most likely at the active site (18), were also chemically modified by the treatment and this masked any effect mediated by residue 838.
Prediction of domain structure
In wild-type RetGC1, the dimerization zone extends from amino acid 817 to 857 and a structure prediction of this region using the computer program Coils (19) indicates that this region is likely to adopt an amphipathic 
-helical coiled-coil structure. The amino acid sequence of such coiled-coil structures consists of a number of heptad repeat units, each representing two turns of an -helix. Hydrophobic interactions between the two -helices occur every fourth residue of the sequence (residues a and d of the heptad a–g), which is frequently leucine, giving rise to a hydrophobic core which maintains the structure. The structure adopted by the dimerization domain in the wild-type and R838 mutants was analysed using Coils. In wild-type RetGC1, a coiled-coil structure is predicted with four turns broken at R838, which occurs at the position a of the heptad. With the R838L mutant, the structure is predicted to continue for three further turns before breaking at T849. Mutants R838A and R838E are likewise predicted to extend the coiled-coil structure, although with a rather lower prediction probability than R838L. The three disease-causing mutations R838C, R838H and R838S, on the other hand, are predicted to have a less drastic effect on the basic wild-type structure, with only a slightly higher probability of the coiled-coil structure continuing for further turns. The probability for the R838C mutant is lower than for the R838H and R838S mutants, which share the same profile.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In this study the effects of different point mutations at position 838 of RetGC1 on the activation by GCAP1 of the cyclase activity have been compared. An earlier characterization of the RetGC1 R838C mutant revealed an increased apparent affinity for activation by GCAP1 and less sensitivity to suppression of its activity by Ca2+/GCAP1 than wild-type (6). That study also showed that both these effects were dominant when examined in vitro in the presence of wild-type RetGC1. The altered Ca2+ sensitivity is predicted to result in residual (constitutive) activity even at elevated Ca2+ concentrations. The resultant change in the equilibrium of Ca2+ and cGMP concentration is thought to be the cause of the cone and subsequent rod degeneration. Cytohistochemistry has revealed the presence of RetGC1 (20) and GCAP1 (9) in both cone and rod photoreceptors, but the concentration appears to be higher in each case in cones than in rods. Thus, any mutation in RetGC1 that affects activity is likely to cause a more serious imbalance in cones than in rods.
The effects of further point mutations at position 838 of RetGC1 can now be compared. The disease-causing mutations R838H and R838S are shown to exert a qualitatively similar effect on enzyme activity to the R838C mutation, but there are quantitative differences between all three mutations. Mutants R838H and R838S increase the maximum specific activation by GCAP1 compared with wild-type, whereas the apparent affinity for activation by GCAP1 for all three mutants was greater than wild-type with R838H > R838S > R838C. In addition, the mutants show a shift in the Ca2+ concentration for half maximal activation to higher Ca2+ concentrations, with the effect of R838S > R838H > R838C. These latter results parallel the effects of the mutations on the phenotype of the disease, since patients with the E837D/R838S mutation are observed to generally display more severe symptoms at an earlier age than those with the R838C and R838H mutations (S.M. Downes et al., in preparation). A previous study of the E837D mutation indicated that on its own it has little effect on activation by GCAP1 (21), although the possibility of it exerting a synergistic effect in the double mutant cannot be discounted. The Y99C mutation in GCAP1 also exerts its effect via a change in Ca2+ sensitivity and the magnitude of the effect is very similar to the more severe retGC1 mutations. However, unlike the RetGC1 mutations, there is no elevation in activity at low Ca2+ concentrations and this difference may go some way towards explaining why the Y99C substitution in GCAP1 affects only cones whereas the R838 mutations of RetGC1 eventually affect rod function also.
Of the artificial mutations studied, only R838A had activity of a similar magnitude to that of wild-type. However, it too showed an increased activation by GCAP1 and was markedly insensitive to inactivation by Ca2+/GCAP1. The mutations R838E, R838K and R838L all resulted in greatly reduced activity. The low activity of the R838K mutant was unexpected as an arginine to lysine substitution is normally regarded as a conservative change. However, it may be noted that the positive charge distribution is more dispersed around the guanidinium group of the arginine residue than on the amino group of lysine and the arginine to lysine substitution may thus result in a loss of directionality in an ion pairing involving the residue. From these results we would predict that the pathology arising from the R838E, K and L mutations might present as a recessive cone–rod dystrophy or LCA.
Activation of both wild-type and R838H RetGC1 by GCAP1 was shown experimentally to vary with pH with maximum activity at pH 
8.0. However, although the activity of wild-type and mutant were approximately equal at pH 6.0, the activity of the mutant at pH 8.0 was 150% of that of wild-type. Notwithstanding the microenvironment of the local protein structure, at pH 6.0 the imidazole group of a histidine residue would be expected to be at least partially ionized and thus mimic an arginine residue fairly effectively. However, at pH 8.0 the imidazole group should carry zero charge with the result that an R838H mutation would then become a non-conservative substitution. This result suggests that enzyme activation by GCAP1 is favoured by an absence of positive charge at position 838. This would fit with a model in which the choice of an arginine residue in the wild-type enzyme represents a compromise between conflicting requirements: the need for maximal enzyme activity (favoured by an absence of charge on residue 838) is balanced by the need for precise regulation by GCAP1 (favoured by the presence of a positive charge).
Analysis of the sequence of the dimerization domain using the structure prediction program Coils reveals that the residue at the 838 position is a key determinant of the extent of the coiled-coil structure responsible for holding together the active RetGC1 dimer. An arginine residue at this position in the wild-type enzyme is predicted to disrupt the structure, limiting it to just four turns of each helix, whereas substitution with other residues results in a higher probability that the structure will continue for further turns. Mild perturbations of the wild-type structure (as exemplified by the R838C, R838H and R838S mutants) are associated with an increased activation by GCAP1 and/or a reduced susceptibility to Ca2+/GCAP1. The R838S and R838H substitutions have the largest impact on structure and R838S the most pronounced effect on enzyme function. On the other hand, substitutions causing a major structural change, like leucine and glutamate, result in substantial reductions in enzyme activity. The Coils program, however, is less successful in predicting the enzyme efficacy of the R838A and R838K mutants. It predicts that the R838A mutant will produce a structure similar to the inactive R838L mutant, yet experimentally it has substantial enzyme activity similar to R838S. R838K on the other hand is predicted to give a structure similar to R838S, yet has low activity.
A model to account for the effect of R838 mutations in adCORD is presented in Figure 7. In the dark when intracellular Ca2+ is high, repulsion between Ca2+/GCAP1 monomer units forces the RetGC1 dimer apart, inhibiting the cyclase activity. After light stimulation when the Ca2+ concentration falls, GCAP1 dimer facilitates formation of the coiled-coil structure in the dimerization domain of RetGC1 and enzyme activation occurs. In the case of the wild-type enzyme, R838 limits the extent of the coiled-coil to just four turns of the helices. Mutations at position 838, however, tend to increase the stability of the coiled-coil and hence alter the position of the equilibrium between the activated and inhibited forms. Inhibition of cyclase activity in the dark becomes a less favourable process, requiring higher Ca2+ concentrations to drive the cyclase to the inactive state. In the case of the R838A, R838H and R838S mutations, the increased stability of the coiled-coil also results in enhanced activation by GCAP1 above that of the wild-type.
|
The mechanism of the photoreceptor degeneration is not understood, but it is known that persistent elevated calcium levels in the cell tend to disrupt the membrane potential of the mitochondrial outer membrane, leading to release of cytochrome c, with subsequent caspase activation and apoptosis (22). In the case of the R838 mutations in RetGC1, the net result might be expected to be the maintenance of cGMP levels in the cell above that required to keep the cGMP-gated cation channels open, resulting in persistently high intracellular Ca2+ levels. However, it is not possible to predict with certainty the effect on cellular cGMP and Ca2+ concentrations because it is unclear to what extent the activity of RetGC2 and other components of the phototransduction cascade might change in a compensatory manner. The generation of R838 transgenic animals would be especially useful in extending our understanding of this aspect of the aetiology of the disease.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Site-directed mutagenesis
Point mutations were introduced into human wild-type RetGC1 using an Altered Sites kit (Promega, Southampton, UK) as follows: a 1.6 kb KpnI–SphI cassette containing the codon for residue 838 was subcloned from pBluescript-RetGC1 into the KpnI–SphI sites of pALTER1. Mutagenesis reactions were performed using the 18–27mer primers listed in Table 2. For the R838A, R838L, R838H and R838S mutations, reactions were performed using the wild-type template. The R838K mutation was generated using the R838S template and the R838E mutation was generated using the R838K template, since in both cases all three nucleotides of the 838 codon needed to be changed from the wild-type R codon. After verification by DNA sequencing, in each case the plasmid was digested with AccI and AatII to yield a 586 bp fragment. This fragment was used to replace the equivalent fragment in the original pBluescript-RetGC1 plasmid. The mutant RetGC1 cDNA (3.6 kb) was then excised with HindIII–XbaI and subcloned into the eukaryotic expression vector pRC-CMV (Invitrogen, Groningen, The Netherlands). The resulting construct was sequenced to confirm the insertion of the mutagenized fragment.
|
Expression of wild-type and mutant RetGC1 in HEK 293-T cells
Constructs were transiently transfected into HEK 293-T cells (ECACC) in 125 mm diameter tissue culture plates using Lipofectamine (Gibco BRL, Paisley, UK). Cells were harvested by scraping after 48 h, washed with Ca2+-free phosphate-buffered saline (PBS) and pelleted gently by centrifugation. The cell pellets were resuspended in low salt homogenization buffer (10 mM MOPS pH 7.3, 5 mM mercaptoethanol, 20 µg/ml leupeptin, 1 mM PMSF) and the cells left to swell on ice for 10 min. The cells were lysed by five passages through a 0.5 mm gauge syringe needle. The salt content of the homogenate was then raised to 0.25 M with 5 M NaCl and the homogenate was spun at 4°C for 5 min at 900 g to remove large debris. Cell membranes were pelleted by further centrifugation of the supernatant at 14 926 g in a pre-cooled 4°C microfuge for 30 min. Membrane pellets were resuspended carefully in 150 µl of low salt homogenization buffer per dish of cells. After assaying the protein content using a BioRad (Hemel Hempstead, UK) protein concentration assay kit, membrane preparations were frozen at –80°C in aliquots. All buffers used were formulated with Analar water and kept in plastic bottles to minimize their free Ca2+ content.
Western analysis of membrane preparations
Aliquots of RetGC1 membrane preparations containing equal amounts of total protein were electrophoresed in 10% SDS–polyacrylamide gels and transferred by electroblotting to Zetaprobe membranes (BioRad) using methanol-free transfer buffer (50 mM Tris–HCl pH 9.1, 390 mM glycine, 0.04% SDS). Blots were blocked with 5% (w/v) milk powder proteins in PBS and then probed with a primary antibody comprising a rabbit polyclonal antibody to the kinase homology domain of RetGC1 (10). The primary antibody was then sandwiched with a secondary antibody comprising an alkaline phosphatase conjugate of goat anti-rabbit IgG (Sigma, Poole, UK) and RetGC1 protein was detected with NBT/BCIP solution (Boehringer, Lewes, UK).
Expression and purification of recombinant human GCAP1
Human N-myristoylated GCAP1 was expressed and purified essentially as described (R.J. Newbold, E.C. Raux, C.E. Walker, S.E. Wilkie, N. Srinivasan, D.M. Hunt, S.S. Bhattacharya and M.J. Warren, in preparation). In brief, the coding sequence for human GCAP1 was amplified from human retinal cDNA via PCR. This was cloned into M13mp18 and an E6S mutation was introduced using the Sculptor mutagenesis system (Amersham Pharmacia Biotech, Little Chalfont, UK) with mutagenesis primer, 5'-CACTGACTTTCCCGACATCACGTTGC-3'. The E6S mutation was introduced to facilitate N-myristoylation, which is required for GCAP1 activation of RetGC1 (23). The Y99C mutation was introduced similarly using the primer 5'-CCGTTGCCATCTACATCACAGAGCTTGAAGTACCAGC-3'. After verification by sequencing, the mutated sequences were inserted into the NdeI–BamHI sites of the vector pET3a(+) (Novagen, Nottingham, UK). The resulting expression plasmids were co-expressed in an Escherichia coli strain of BL21 (DE3) (pLysS) carrying the plasmid pBB131 (a kind gift from Dr J. Gordon, Washington University School of Medicine, St Louis, MO) encoding yeast N-myristoyl transferase. Cells were grown in LB broth (supplemented with 50 µg/ml ampicillin, 50 µg/ml kanamycin and 50 µg/ml myristic acid) to an OD600 of 0.4–0.6, induced with 0.4 mM IPTG and then harvested by centrifugation 2 h later. The cell pellets were stored as necessary at –20°C.
For purification of GCAP1, the cells were partially lysed with 20 µg/ml lysozyme in 20% sucrose buffer [20% sucrose (w/v), 33 mM Tris–HCl pH 8.0, 10 mM 2-mercaptoethanol] to weaken the cell walls and the lysate was centrifuged to remove the periplasm. The pellet was resuspended in low salt Tris buffer (10 mM Tris–HCl pH 8.0, 10 mM mercaptoethanol, 1 mM EDTA, 1 mM PMSF) and completely lysed using a Dounce homogenizer. After centrifugation to remove cytoplasmic components, the pellet was resuspended in 1 M salt buffer (50 mM Tris–HCl pH 8.0, 1 M NaCl, 10 mM 2-mercaptoethanol, 1 mM EDTA, 1 mM PMSF) to release GCAP1 from the membranes. The GCAP1 was then precipitated with 50% (w/v) ammonium sulphate, resuspended in buffer A (50 mM Tris–HCl pH 8.0, 200 mM NaCl, 10 mM 2-mercaptoethanol) and purified by gel filtration through an S300 Sephacryl column. Fractions containing high levels of GCAP1 monomer were identified using a BioRad protein assay and SDS–PAGE, pooled, dialysed against buffer B [10 mM MOPS pH 7.5, 10% (w/v) glycerol, 10 mM 2-mercaptoethanol] and concentrated to 0.5–1.0 mg/ml using a Centriprep spin column (Millipore, Watford, UK). N-myristoylation of the protein was checked by mass spectrometry and final protein purity was estimated to be at least 90% from SDS–PAGE. The protein was observed to retain activity when stored at –20°C for at least 4 weeks.
Guanylate cyclase assays
Measurement of guanylate cyclase activity was carried out essentially as described in Dizhoor et al. (24) but with some modifications as follows. Aliquots of membrane preparation containing equal amounts of total protein were added to 5 µl of 4x GC buffer formulated with the appropriate free Ca2+ concentration (see below). For GCAP1-stimulated reactions, purified recombinant GCAP1 was then added and the mixture was made up to a total volume of 15 µl with 10 mM MOPS pH 7.3. The reaction was started by the addition of 5 µl of a 4x substrate solution (4 mM GTP, 20 mM cGMP, 1.6 mM ATP, 2 µCi [-32P]GTP, 100 000 d.p.m. [8-3H]cGMP) and incubating at 37°C for 30 min. The reaction was stopped by heating to 100°C in a heating block for 2 min. After centrifugation at 10 000 g for 10 min to pellet the heat-denatured proteins, cGMP was separated from GTP by chromatographing 6 µl of the reaction mix on an amino-bonded aluminium F254 thin-layer chromatography plate (Merck, Lutterworth, UK) in 0.2 M LiCl, 30% ethanol. Spots corresponding to cGMP were visualized with a short wavelength UV illuminator, excised and eluted by gentle shaking for 10 min in l ml of 2 M LiCl. Both 3H and 32P were counted in Lumagel Safe scintillant (Canberra Packard, Pangbourne, UK) in a Packard Tri-Carb 1600TR scintillation counter using a channel ratio method.
For the determination of basal activity of wild-type and mutant RetGC1, 4x GC buffer was formulated with zero free Ca2+ (400 mM KCl, 200 mM MOPS pH 7.3, 28 mM mercaptoethanol, 40 mM MgCl2, 32 mM NaCl, 4 mM EGTA pH 7.3). For assays measuring Mn2+/Triton X-100 activity, 4x GC buffer contained 10 mM MnCl2 and 1% Triton X-100 instead of MgCl2. Assays of GCAP1-stimulated activity were performed in 4x GC buffers containing a range of free Ca2+ concentrations. These had the same composition as the zero free Ca2+ buffer except that the EGTA was substituted by Ca2+-EGTA formulated from solutions of EGTA and EGTA saturated with CaCl2 by pH titration in accordance with the method of Tsien and Pozzan (25). Free Ca2+ concentrations were estimated from the Kd values of EGTA for Ca2+ and Mg2+ at 37°C and pH 7.3.
Chemical modification of wild-type and R838H RetGC1 was accomplished by pre-incubating the enzyme in 4x GC buffer at room temperature with either diethylpyrocarbonate (0–100 mM) for 30 min or butadione (0–10 mM) for 10 min. The reactions were started by adding GCAP1 and 4x substrate solution and incubating at 37°C.
Prediction of secondary structure of dimerization zone in wild-type and mutant RetGC1
The structure prediction program Coils (19), which compares the primary sequence of a peptide with a database of known parallel two-stranded coiled-coils and derives a similarity score, was used to analyse peptide sequence from position 815 to 849 of wild-type and R838 mutant forms of RetGC1. Predicted structures were compared based on a window size of 14 using the MTK matrix.
|
ACKNOWLEDGEMENTS |
|---|
This work was supported by grant nos 053405 and 003303 from the Wellcome Trust.
|
FOOTNOTES |
|---|
+ To whom correspondence should be addressed. Tel: +44 20 7608 6820; Fax: +44 20 7608 6863; Email: d.hunt@ucl.ac.uk
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Kelsell, R.E., Gregory-Evans, K., Payne, A.M., Perrault, I., Kaplan, J., Yang, R.-B., Garbers, D.L., Bird, A.C., Moore, A.T. and Hunt, D.M. (1998) Mutations in the retinal guanylate cyclase (RETGC-1) gene in dominant cone–rod dystrophy. Hum. Mol. Genet., 7, 1179–1184.
2 Perrault, I., Rozet, J.-M., Gerber, S., Kelsell, R.E., Souied, E., Cabot, A., Hunt, D.M., Munnich, A. and Kaplan, J. (1998) A retGC-1 mutation in autosomal dominant cone–rod dystrophy. Am. J. Hum. Genet., 63, 651–654.[ISI][Medline]
3 Moore, A.T. (1970) Cone and cone–rod dystrophies. J. Med. Genet., 29, 289–290.[ISI][Medline]
4 Weigell-Weber, M., Fokstuen, S., Torok, B., Niemeyer, G., Schinzel, A. and Hergersberg, M. (2000) Codons 837 and 838 in the retinal guanylate cyclase gene on chromosome 17p: hot spots for mutations in autosomal dominant cone–rod dystrophy. Arch. Ophthalmol., 118, 300.
5 Gregory-Evans, K., Kelsell, R.E., Gregory-Evans, C.Y., Downes, S.M., Fitzke, F.W., Holder, G.E., Simunovic, M., Mollon, J.D., Taylor, R., Hunt, D.M. et al. (2000) Abnormal cone synapses in human cone–rod dystrophy. Ophthalmology, 107, 55–61.[ISI][Medline]
6 Tucker, C.L., Woodcock, S.C., Kelsell, R.E., Ramamurthy, V., Hunt, D.M. and Hurley, J.B. (1999) Biochemical analysis of a dimerization domain mutation in RetGC-1 associated with dominant cone–rod dystrophy. Proc. Natl Acad. Sci. USA, 96, 9039–9044.
7 Perrault, I., Rozet, J.-M., Calvas, P., Gerber, S., Camuzat, A., Dollfus, H., Chatelin, S., Souied, E., Ghazi, I., Leowski, C. et al. (1996) Retinal-specific guanylate cyclase mutations in Leber’s congenital amaurosis. Nature Genet., 14, 461–464.[ISI][Medline]
8 Yu, H., Olshevskaya, T., Duda, T., Seno, K., Hayashi, R.K., Sharma, R.K., Dizhoor, A.M. and Yamazaki, A. (1999) Activation of retinal guanylyl cyclase-1 by Ca2+-binding proteins involves its dimerization. J. Biol. Chem., 274, 15547–15555.
9 Kachi, S., Nishizawa, Y., Olshevskaya, E., Yamazaki, A., Miyake, Y., Wakabayashi, T., Dizhoor A., and Usukura, J. (1999) Detailed localization of photoreceptor guanylate cyclase activating protein-1 and -2 in mammalian retinas using light and electron microscopy. Exp. Eye Res., 68, 465–473.[ISI][Medline]
10 Laura, R.P. and Hurley, J.B. (1998) The kinase homology domain of retinal guanylyl cyclases 1 and 2 specifies the affinity and cooperativity of interaction with guanylyl cyclase activating protein-2. Biochemistry, 37, 11264–11271.[Medline]
11 Potter, L.R. and Garbers, D.L. (1992) Dephosphorylation of the guanylyl cyclase-A receptor causes desensitization. J. Biol. Chem., 267, 14531–14534.
12 Gorczyca, W.A., Gray-Keller, M.P., Detwiler, P.B. and Palczewski, K. (1994) Purification and physiological evaluation of a guanylate cyclase activating protein from retinal rods. Proc. Natl Acad. Sci. USA, 91, 4014–4018.
13 Dizhoor, A.M., Lowe, D.G., Olshevskaya, E.V., Laura, R.P. and Hurley, J.B. (1994) The human photoreceptor membrane guanylyl cyclase, RetGC, is present in outer segments and is regulated by calcium and a soluble activator. Neuron, 12, 1345–1352.[ISI][Medline]
14 Rudnicka-Nawrot, M., Surgucheva, I., Hulmes, J.D., Haeseleer, F., Sokal, I., Crabb, J.W., Baehr, W. and Palczewski, K. (1998) Changes in biological activity and folding of guanylate cyclase-activating protein 1 as a function of calcium. Biochemistry, 37, 248–257.[Medline]
15 Pugh, E.N., Duda, T., Sitaramayya, A. and Sharma, R.K. (1997) Photoreceptor guanylate cyclases: a review. Biosci. Rep., 17, 429–472.[ISI][Medline]
16 Sokal, I., Li, N., Surgucheva, I., Warren, M.J., Payne, A.M., Bhattacharya, S.S., Baehr, W. and Palczewski, K. (1998) GCAP1(Y99C) mutant is constitutively active in autosomal dominant cone dystrophy. Mol. Cell, 2, 1–20.[ISI][Medline]
17 Dizhoor, A.M., Boikov, S.G. and Olshevskaya, E.V. (1998) Constitutive activation of photoreceptor guanylate cyclase by Y99C mutant of GCAP-1. J. Biol. Chem., 273, 17311–17314.
18 Liu, Y., Ruoho, A.E., Rao, V.D. and Hurley, J.H. (1997) Catalytic mechanism of the adenylyl and guanylyl cyclases: modelling and mutational analysis. Proc. Natl Acad. Sci. USA, 94, 13414–13419.
19 Lupas, A. (1996) Prediction and analysis of coiled-coil structures. Methods Enzymol., 266, 513–525.[ISI][Medline]
20 Liu, X., Seno, K., Nishizawa, Y., Hayashi, F., Yamazaki, A., Matsumoto, H., Wakabayashi, T. and Usukura, J. (1994) Ultrastructural localization of retinal guanylate cyclase in human and monkey retinas. Exp. Eye Res., 59, 761–768.[ISI][Medline]
21 Duda, T., Krishnan, A., Venkataraman, V., Lange, C., Koch, K.-W. and Sharma, R.K. (1999) Mutations in the rod outer segment membrane guanylate cyclase in a cone–rod dystrophy cause defects in calcium signalling. Biochemistry, 38, 13912–13919.[Medline]
22 Green, D.R. and Reed, J.C. (1998) Mitochondria and apoptosis. Science, 281, 1309–1312.
23 Palczewski, K., Subbaraya, I., Gorczyca, W.A., Helekar, B.S., Ruiz, C.C., Ohguro, H., Huang, J., Zhao, X., Crabb, J.W., Johnson, R.S. et al. (1994) Molecular cloning and characterisation of retinal photoreceptor guanylyl cyclase-activating protein. Neuron, 13, 395–404.[ISI][Medline]
24 Dizhoor, A.M., Olshevskaya, E.V., Henzel, W.J., Wong, S.C., Stults, J.T., Ankoudinova, I. and Hurley, J.B. (1995) Cloning, sequencing and expression of a 24 kDa Ca2+-binding protein activating photoreceptor guanylyl cyclase. J. Biol. Chem., 270, 25200–25206.
25 Tsien, R. and Pozzan, T. (1989) Measurement of cytosolic free Ca2+ with quin2. Methods Enzymol., 172, 230–262.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. W. Small, R. Silva-Garcia, N. Udar, E. V. Nguyen, and J. R. Heckenlively New Mutation, P575L, in the GUCY2D Gene in a Family With Autosomal Dominant Progressive Cone Degeneration Arch Ophthalmol, March 1, 2008; 126(3): 397 - 403. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Woodruff, E. V. Olshevskaya, A. B. Savchenko, I. V. Peshenko, R. Barrett, R. A. Bush, P. A. Sieving, G. L. Fain, and A. M. Dizhoor Constitutive Excitation by Gly90Asp Rhodopsin Rescues Rods from Degeneration Caused by Elevated Production of cGMP in the Dark J. Neurosci., August 15, 2007; 27(33): 8805 - 8815. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. V. Peshenko and A. M. Dizhoor Ca2+ and Mg2+ Binding Properties of GCAP-1: EVIDENCE THAT Mg2+-BOUND FORM IS THE PHYSIOLOGICAL ACTIVATOR OF PHOTORECEPTOR GUANYLYL CYCLASE J. Biol. Chem., August 18, 2006; 281(33): 23830 - 23841. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Makino, X.-H. Wen, N. Michaud, I. V. Peshenko, B. Pawlyk, R. S. Brush, M. Soloviev, X. Liu, M. L. Woodruff, P. D. Calvert, et al. Effects of Low AIPL1 Expression on Phototransduction in Rods Invest. Ophthalmol. Vis. Sci., May 1, 2006; 47(5): 2185 - 2194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zernant, M. Kulm, S. Dharmaraj, A. I. den Hollander, I. Perrault, M. N. Preising, B. Lorenz, J. Kaplan, F. P. M. Cremers, I. Maumenee, et al. Genotyping Microarray (Disease Chip) for Leber Congenital Amaurosis: Detection of Modifier Alleles Invest. Ophthalmol. Vis. Sci., September 1, 2005; 46(9): 3052 - 3059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Sokal, W. J. Dupps, M. A. Grassi, J. Brown Jr, L. M. Affatigato, N. Roychowdhury, L. Yang, S. Filipek, K. Palczewski, E. M. Stone, et al. A Novel GCAP1 Missense Mutation (L151F) in a Large Family with Autosomal Dominant Cone-Rod Dystrophy (adCORD) Invest. Ophthalmol. Vis. Sci., April 1, 2005; 46(4): 1124 - 1132. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Coleman, Y. Zhang, G. A. J. Brown, and S. L. Semple-Rowland Cone Cell Survival and Downregulation of GCAP1 Protein in the Retinas of GC1 Knockout Mice Invest. Ophthalmol. Vis. Sci., October 1, 2004; 45(10): 3397 - 3403. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. V. Olshevskaya, P. D. Calvert, M. L. Woodruff, I. V. Peshenko, A. B. Savchenko, C. L. Makino, Y.-S. Ho, G. L. Fain, and A. M. Dizhoor The Y99C Mutation in Guanylyl Cyclase-Activating Protein 1 Increases Intracellular Ca2+ and Causes Photoreceptor Degeneration in Transgenic Mice J. Neurosci., July 7, 2004; 24(27): 6078 - 6085. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Michaelides, I. A. Aligianis, J. R. Ainsworth, P. Good, J. D. Mollon, E. R. Maher, A. T. Moore, and D. M. Hunt Progressive Cone Dystrophy Associated with Mutation in CNGB3 Invest. Ophthalmol. Vis. Sci., June 1, 2004; 45(6): 1975 - 1982. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. P. M. Cremers, J. A. J. M. van den Hurk, and A. I. den Hollander Molecular genetics of Leber congenital amaurosis Hum. Mol. Genet., May 15, 2002; 11(10): 1169 - 1176. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Downes, A. M. Payne, R. E. Kelsell, F. W. Fitzke, G. E. Holder, D. M. Hunt, A. T. Moore, and A. C. Bird Autosomal Dominant Cone-Rod Dystrophy With Mutations in the Guanylate Cyclase 2D Gene Encoding Retinal Guanylate Cyclase-1 Arch Ophthalmol, November 1, 2001; 119(11): 1667 - 1673. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M Payne, A. G Morris, S. M Downes, S. Johnson, A. C Bird, A. T Moore, S. S Bhattacharya, and D. M Hunt Clustering and frequency of mutations in the retinal guanylate cyclase (GUCY2D) gene in patients with dominant cone-rod dystrophies J. Med. Genet., September 1, 2001; 38(9): 611 - 614. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||