Effect of Hailey-Hailey Disease Mutations on the Function of a New Variant of Human Secretory Pathway Ca2+/Mn2+-ATPase (hSPCA1)
2003; Elsevier BV; Volume: 278; Issue: 27 Linguagem: Inglês
10.1074/jbc.m300509200
ISSN1083-351X
AutoresRebecca J. Fairclough, Leonard Dode, Jo Vanoevelen, Jens Peter Andersen, Ludwig Missiaen, Luc Raeymaekers, Frank Wuytack, Alain Hovnanian,
Tópico(s)Cancer and Skin Lesions
ResumoATP2C1, encoding the human secretory pathway Ca2+/Mn2+ ATPase (hSPCA1), was recently identified as the defective gene in Hailey-Hailey Disease (HHD), an autosomal dominant skin disorder characterized by persistent blisters and erosions. To investigate the underlying cause of HHD, we have analyzed the changes in expression level and function of hSPCA1 caused by mutations found in HHD patients. Mutations were introduced into hSPCA1d, a novel splice variant expressed in keratinocytes, described here for the first time. Encoded by the full-length of optional exons 27 and 28, hSPCA1d was longer than previously identified splice variants. The protein competitively transported Ca2+ and Mn2+ with equally high affinity into the Golgi of COS-1 cells. Ca2+- and Mn2+-dependent phosphoenzyme intermediate formation in forward (ATP-fuelled) and reverse (Pi-fuelled) directions was also demonstrated. HHD mutant proteins L341P, C344Y, C411R, T570I, and G789R showed low levels of expression, despite normal levels of mRNA and correct targeting to the Golgi, suggesting instability or abnormal folding of the mutated hSPCA1 polypeptides. P201L had little effect on the enzymatic cycle, whereas I580V caused a block in the E1∼P → E2-P conformational transition. D742Y and G309C were devoid of Ca2+- and Mn2+-dependent phosphoenzyme formation from ATP. The capacity to phosphorylate from Pi was retained in these mutants but with a loss of sensitivity to both Ca2+ and Mn2+ in D742Y and a preferential loss of sensitivity to Mn2+ in G309C. These results highlight the crucial role played by Asp-742 in the architecture of the hSPCA1 ion-binding site and reveal a role for Gly-309 in Mn2+ transport selectivity. ATP2C1, encoding the human secretory pathway Ca2+/Mn2+ ATPase (hSPCA1), was recently identified as the defective gene in Hailey-Hailey Disease (HHD), an autosomal dominant skin disorder characterized by persistent blisters and erosions. To investigate the underlying cause of HHD, we have analyzed the changes in expression level and function of hSPCA1 caused by mutations found in HHD patients. Mutations were introduced into hSPCA1d, a novel splice variant expressed in keratinocytes, described here for the first time. Encoded by the full-length of optional exons 27 and 28, hSPCA1d was longer than previously identified splice variants. The protein competitively transported Ca2+ and Mn2+ with equally high affinity into the Golgi of COS-1 cells. Ca2+- and Mn2+-dependent phosphoenzyme intermediate formation in forward (ATP-fuelled) and reverse (Pi-fuelled) directions was also demonstrated. HHD mutant proteins L341P, C344Y, C411R, T570I, and G789R showed low levels of expression, despite normal levels of mRNA and correct targeting to the Golgi, suggesting instability or abnormal folding of the mutated hSPCA1 polypeptides. P201L had little effect on the enzymatic cycle, whereas I580V caused a block in the E1∼P → E2-P conformational transition. D742Y and G309C were devoid of Ca2+- and Mn2+-dependent phosphoenzyme formation from ATP. The capacity to phosphorylate from Pi was retained in these mutants but with a loss of sensitivity to both Ca2+ and Mn2+ in D742Y and a preferential loss of sensitivity to Mn2+ in G309C. These results highlight the crucial role played by Asp-742 in the architecture of the hSPCA1 ion-binding site and reveal a role for Gly-309 in Mn2+ transport selectivity. At present, three distinct classes of phosphorylation-type Ca2+ transport ATPases have been identified in mammalian cells: plasma membrane (PMCA), 1The abbreviations used are: PMCA, plasma membrane calcium adenosine triphosphatase; (h)SPCA1, (human) secretory pathway Ca2+/Mn2+-adenosine triphosphatase; HHD, Hailey-Hailey disease; SERCA, sarco(endo)plasmic reticulum calcium adenosine triphosphatase; PMR, plasma membrane ATPase-related; cDNA, complementary DNA; RACE, rapid amplification of cDNA ends; GSP, gene-specific primers; nt, nucleotide(s); D, donor; Bis-Tris, 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol; MES, 4-morpholineethanesulfonic acid. sarco(endo)plasmic reticulum (SERCA), and Golgi-associated secretory pathway (SPCA) Ca2+ transport ATPases. These proteins serve to actively pump Ca2+ out of the cytoplasm, thus contributing to the maintenance of a low cytosolic Ca2+ concentration in resting conditions (reviewed in Ref. 1Møller J.V. Juul B. le Maire M. Biochim. Biophys. Acta. 1996; 1286: 1-51Google Scholar). Relatively little is known about the SPCA class. ATP2C1, the gene encoding the human secretory pathway Ca2+/Mn2+-ATPase (hSPCA1), was recently identified as the defective gene in Hailey-Hailey disease (HHD, OMIM 16960), an autosomal-dominant skin disorder characterized by abnormal keratinocyte adhesion in the suprabasal layer of the epidermis (2Sudbrak R. Brown J. Dobson-Stone C. Carter S. Ramser J. White J. Healy E. Dissanayake M. Larrègue M. Perrussel M. Lehrach H. Munro C.S. Strachan T. Burge S. Hovnanian A. Monaco A.P. Hum. Mol. Genet. 2000; 9: 1131-1140Google Scholar, 3Hu Z. Bonifas J.M. Beech J. Bench G. Shigihara T. Ogawa H. Ikeda S. Mauro T. Epstein Jr., E.H. Nat. Genet. 2000; 24: 61-65Google Scholar). Sequence analysis suggests that the protein retains the remarkable conservation in structure and function that exists across the Ca2+-ATPase class and that, like PMCA, the SPCA class possesses only one of the two high affinity Ca2+-binding sites present in SERCA. Most of our knowledge of the SPCA class comes from studies of the yeast Saccharomyces cerevisiae homologue, named somewhat confusingly PMR1 for plasma membrane ATPase-related, which shares 49% amino acid identity with hSPCA1. Following the initial identification of PMR1 as a probable Ca2+-ATPase (4Rudolf H.K. Antebi A. Fink G.R. Buckley C.M. Dorman T.E. LeVitre J. Davidow L.S. Mao J. Moir D.T. Cell. 1989; 58: 133-145Google Scholar), the protein was localized to the Golgi or one of its subcompartments (5Antebi A. Fink G.R. Mol. Biol. Cell. 1992; 3: 633-654Google Scholar). Studies of null strains defective in PMR1 illustrated a pleiotropic effect on Golgi function, including impaired proteolytic processing, incomplete glycosylation, and defective pre-, post-, and intra-Golgi translocation of secreted proteins (5Antebi A. Fink G.R. Mol. Biol. Cell. 1992; 3: 633-654Google Scholar). These defects resulted in defective cell wall morphogenesis, which is interestingly reminiscent of abnormal keratinocyte adhesion in HHD. Moreover, in yeast these phenotypes could be reversed by addition of Ca2+ (10 μm) to the medium, implicating a direct role for Ca2+ in Golgi function (5Antebi A. Fink G.R. Mol. Biol. Cell. 1992; 3: 633-654Google Scholar). Evidence suggests that PMR1 can also transport Mn2+ (6Lapinskas P.J. Cunningham K.W. Lui X.F. Fink G.R. Culotta V.C. Mol. Cell. Biol. 1995; 15: 1382-1388Google Scholar). Although being an essential cofactor for a wide range of enzymes (7Dürr G. Strayle J. Plemper R. Elbs S. Klee S.K. Catty P. Wolf D.H. Rudolf H.K. Mol. Biol. Cell. 1998; 9: 1149-1162Google Scholar, 8Cottrell G.S. Hooper N.M. Turner A.J. Biochemistry (Mosc). 2000; 39: 15121-15128Google Scholar, 9Hearn A.S. Stroupe M.E. Cabelli D.E. Lepock J.R. Trainer J.A. Nick H.S. Silverman D.N. Biochemistry (Mosc). 2001; 40: 12051-12058Google Scholar), high concentrations of Mn2+ are toxic, interfering with Mg2+-binding sites on proteins and compromising the fidelity of DNA polymerases (10Beckman R.A. Mildvan A.S. Loeb L.A. Biochemistry (Mosc). 1985; 24: 5810-5817Google Scholar). PMR1-type pumps appear to be the principal route for Mn2+ detoxification, via the secretory pathway, and are important in maintaining both cytosolic and luminal Mn2+ homeostasis. The first functional study on SPCA in animals was conducted on Caenorhabditis elegans. C. elegans SPCA was shown to transport Ca2+ and Mn2+ with high affinity into the Golgi following heterologous expression in COS-1 cells (11Van Baelen K. Vanoevelen J. Missiaen L. Raeymaekers L. Wuytack F. J. Biol. Chem. 2001; 276: 10683-10691Google Scholar). Overexpression of human SPCA1 in yeast and Chinese hamster ovary cells was restricted to the Golgi compartment and, moreover, was able to complement the PMR1 null mutation, as demonstrated by the ability to transport Ca2+ and Mn2+ in yeast (12Ton V.-K. Mandal D. Vahadji C. Rao R. J. Biol. Chem. 2002; 277: 6422-6427Google Scholar). Key insights into structure/function relationships have been provided by functional analysis of over 250 point mutations in SERCA1a. This has led to identification of specific domains important in Ca2+ and ATP binding and phosphorylation, together with residues important in the conformational changes associated with the various stages of the catalytic cycle (13Andersen J.P. Vilsen B. FEBS Lett. 1995; 359: 101-106Google Scholar, 14MacLennan D.H. Rice W.J. Green N.M. J. Biol. Chem. 1997; 272: 28815-28818Google Scholar) (Scheme 1). In contrast, mutagenic studies of the SPCA class have been limited and have only involved the S. cerevisiae protein. Mutations in the active phosphorylation site, Asp-372, had the expected effect of abolishing Ca2+ transport activity (15Sorin A. Rosas G. Rao R. J. Biol. Chem. 1997; 272: 9895-9901Google Scholar). Residues in the putative N-terminal EF hand-like motif were reported to have a modulatory effect on ion transport, through their importance in ion binding and participation in long range interactions involved in ATP binding and phosphorylation (16Wei Y. Marchi V. Wang R. Rao R. Biochemistry (Mosc). 1999; 38: 14534-14541Google Scholar). Other reports have targeted the oxygen-containing side chains of the predicted transmembrane domains M4–M8, likely to be involved in the coordination of Ca2+ and Mn2+ ions (17Wei Y. Chen J. Rosas G. Tompkins D.A. Holt P.A. Rao R. J. Biol. Chem. 2000; 275: 23927-23932Google Scholar), and have defined Gln-783 as crucial in Mn2+ selectivity of PMR1 (18Mandal D. Woolf T.B. Rao R. J. Biol. Chem. 2000; 275: 23933-23938Google Scholar). We have previously described a spectrum of mutations scattered throughout the ATP2C1 gene in HHD patients (2Sudbrak R. Brown J. Dobson-Stone C. Carter S. Ramser J. White J. Healy E. Dissanayake M. Larrègue M. Perrussel M. Lehrach H. Munro C.S. Strachan T. Burge S. Hovnanian A. Monaco A.P. Hum. Mol. Genet. 2000; 9: 1131-1140Google Scholar, 19Dobson-Stone C. Fairclough R. Dunne E. Brown J. Dissanayake M. Munro C.S. Strachan T. Burge S. Sudbrak R. Monaco A.P. Hovnanian A. J. Invest. Dermatol. 2002; 118: 338-343Google Scholar). Although nonsense, frameshift, and splice-site mutations are predicted to cause nonsense-mediated mRNA decay (20Frischmeyer P.A. Dietz H.C. Hum. Mol. Genet. 1999; 8: 1893-1900Google Scholar) or aberrant splicing, the underlying cause of disease in patients carrying missense mutations cannot be predicted a priori. In this study we investigated the molecular and physiological basis of HHD in patients carrying these mutations, providing new insights into SPCA structure/function relationships. Site-directed mutagenesis was used to introduce these disease-causing point mutations into the cDNA sequence of a novel splice variant of hSPCA1, which was identified and functionally characterized in this study. More than 50% of the mutants studied here showed low levels of protein expression, despite normal levels of mRNA and correct localization to the Golgi compartment. Other mutants are characterized by lack of ion transport, caused by specific alterations to the partial reactions of the catalytic cycle, such as defects in Ca2+ and Mn2+ binding and inability of the phosphoenzyme intermediate to undergo the energy-transducing E1∼P → E2-P conformational transition. Materials—Radiochemicals and chemicals were obtained from Amersham Biosciences (UK), PerkinElmer Life Sciences (Boston, MA), and Sigma (Dorset, UK). DNA sequencing was performed using the ABI Prism 377 sequencer and the BigDye Terminator Cycle Sequencing Ready Reaction kit (PE-Applied Biosystems, Cheshire, UK). Unless otherwise stated, PCR was carried out using the Expand™ Long Template PCR system (Roche Applied Science, E. Sussex, UK). Anti-mouse/rabbit alkaline phosphate-conjugated secondary antibodies were from Amersham Biosciences. Sheep anti-TGN46 was purchased from Serotec (Oxford, UK); fluorescein isothiocyanate-conjugated secondary antibodies (donkey anti-sheep, goat anti-rabbit) were from Molecular Probes (Leiden, The Netherlands). Free ion concentrations were calculated using the CaBuf program (available at ftp.cc.kuleuven.ac.be/pub/droogmans/cabuf.zip) based on the stability constants for oxalate, EGTA, and ATP, as described before (11Van Baelen K. Vanoevelen J. Missiaen L. Raeymaekers L. Wuytack F. J. Biol. Chem. 2001; 276: 10683-10691Google Scholar). Reverse Transcriptase-PCR Analyses—Following extraction of total RNA from human keratinocytes with TRIzol (Invitrogen), both reverse transcription of ATP2C1 and 3′ rapid amplification of cDNA ends (RACE)-PCR, were performed using the Marathon™ cDNA amplification kit (Clontech, Palo Alto, CA). The gene-specific primer, ATP2C13′GSP (5′-CCAAGTCTGTGTTTGAGATTGGACTCTGC-3′), corresponded to nucleotides (nt) 2618–2646 in the ATP2C1d sequence. Products were subcloned into the PCR-cloning vector pGEM-T (Promega, Madison, WI) and sequenced in both directions. Splice variants identified by sequencing were subsequently amplified by PCR using ATP2C13′GSP and antisense primers Ex27R (5′-TTGCCCTTCTAAATGATCCTC-3′) or Ex28R (5′-GGAAGAGCTGCAGGAAGATG-3′) in standard reaction conditions. Ex27R and Ex28R, respectively, represent the inverse complement of nt 3039–3059 and 3048–3067 in the ATP2C1a and ATP2C1d sequences. PCR conditions were as follows: 4 min of denaturation at 95 °C and 30 cycles of 30 s at 95 °C,30sat58 °C, and 30 s at 72 °C. First strand cDNA was prepared for LightCycler™-based real-time PCR analysis (Roche Applied Science) following total RNA extraction from transfected COS-1 cells using the Absolutely RNA™ RT-PCR Miniprep kit (Stratagene). A 276-bp PCR product was generated using ATP2C1 primers C7F (5′-TTGGTTGGCTGGTTACTGGG-3′) and C7R (5′-AGCATGCAGACCATCTGAAGT-3′), corresponding to nt 973–992 and the inverse complement of nt 1228–1248, respectively, in the ATP2C1d sequence. GAPDH amplification was performed, to control for equal RNA loading between samples, using primers GAPDHF (5′-ATCATCCCTGCCTCTACTGG-3′) and GAPDHR (5′-TGCTGTAGCCAAATTCGTTG-3′) to generate a 349-bp product. Sequences correspond to nt 120–139 and the inverse complement of nt 449–468 in the GAPDH sequence. Real-time PCR amplification was performed with 1 μl of cDNA product in a 20-μl reaction containing 0.5 μm of each primer and QuantiTect™ SYBR® Green PCR Master Mix (Qiagen, W. Sussex, UK). The PCR cycle consisted of 15 min of denaturation at 95 °C, followed by 45 cycles of 15 s at 95 °C, 20 s at 60 °C, and 20 s at 72 °C. Fluorescence data was acquired at the end of each extension cycle. To confirm amplification specificity, PCR products were subjected to a melting curve analysis, as described previously (21Gutzmer R. Mommert S. Kiehl P. Wittman M. Kapp A. Werfel T. J. Invest. Dermatol. 2001; 116: 926-932Google Scholar). Analysis of real-time PCR data was performed using the LightCycler™ software (Roche Applied Science). Crossing points, defined as the cycle number at which all samples are exponentially amplifying, were recalculated into relative log concentrations of RNA according to PCR efficiency calculated from standard curves generated using serially diluted cDNA for GAPDH and ATP2C1 primer sets. ATP2C1 amplification was then corrected for equal GAPDH expression between samples. PCR products were isolated from the LightCycler™ capillaries after cycling and visualized following electrophoresis on 1.5% agarose gels. cDNA Construction, Site-specific Mutagenesis, COS-1 Cell Culture, and Transfection—A full-length cDNA clone encoding ATP2C1d was constructed using cDNA fragments amplified by 5′ and 3′ RACE-PCR and was subcloned into vector pSPORT1 (Invitrogen). The full-length ATP2C1d cDNA was then transferred into the mammalian expression vector pMT2, with the insert corresponding to the entire coding sequence (nt 125-end) of the deposited ATP2C1d nucleotide sequence. Missense mutations originally reported in HHD patients by Sudbrak et al. (2Sudbrak R. Brown J. Dobson-Stone C. Carter S. Ramser J. White J. Healy E. Dissanayake M. Larrègue M. Perrussel M. Lehrach H. Munro C.S. Strachan T. Burge S. Hovnanian A. Monaco A.P. Hum. Mol. Genet. 2000; 9: 1131-1140Google Scholar) and Dobson-Stone et al. (19Dobson-Stone C. Fairclough R. Dunne E. Brown J. Dissanayake M. Munro C.S. Strachan T. Burge S. Sudbrak R. Monaco A.P. Hovnanian A. J. Invest. Dermatol. 2002; 118: 338-343Google Scholar) were introduced into the wild type ATP2C1 expression clone using the QuikChange™ XL site-directed mutagenesis kit (Stratagene). The full-length cDNA clones were sequenced in both directions to ensure no additional sequence changes had been introduced during mutagenesis. Details of mutations introduced into the hSPCA1 protein sequence are given in Table I. COS-1 cells for immunocytochemistry, microsome preparation, and isotope flux were cultured and transiently transfected with FuGENE™ 6 transfection reagent (Roche Applied Science) as previously described (11Van Baelen K. Vanoevelen J. Missiaen L. Raeymaekers L. Wuytack F. J. Biol. Chem. 2001; 276: 10683-10691Google Scholar). For total RNA or protein extraction 1 × 106 COS-1 cells were seeded in 60-mm culture dishes 24 h before transfection.Table ISummary of mutant proteins studiedhSPCA1 mutationEquivalent SERCA1a residueaDetermined by sequence alignment with SERCA1a.Putative protein domainbPutative protein domain prediction is based on the position of the equivalent residue within the structure of SERCA1a.P201LcP201L, C344Y, and T570I, respectively, represent mutations P185L, C328Y, and T554I, originally reported by Sudbrak et al. (2). Mutation nomenclature has now been updated with respect to the 5′-end sequence published by Hu et al. (3) and the results of our 5′ RACE-PCR experiments.P195ActuatorG309CG310M4L341PL342PhosphorylationC344YcP201L, C344Y, and T570I, respectively, represent mutations P185L, C328Y, and T554I, originally reported by Sudbrak et al. (2). Mutation nomenclature has now been updated with respect to the 5′-end sequence published by Hu et al. (3) and the results of our 5′ RACE-PCR experiments.T345PhosphorylationC411RC420PhosphorylationT570IcP201L, C344Y, and T570I, respectively, represent mutations P185L, C328Y, and T554I, originally reported by Sudbrak et al. (2). Mutation nomenclature has now been updated with respect to the 5′-end sequence published by Hu et al. (3) and the results of our 5′ RACE-PCR experiments.T625ATP bindingI580VI635ATP bindingD742YD800M6G789RA847M7a Determined by sequence alignment with SERCA1a.b Putative protein domain prediction is based on the position of the equivalent residue within the structure of SERCA1a.c P201L, C344Y, and T570I, respectively, represent mutations P185L, C328Y, and T554I, originally reported by Sudbrak et al. (2Sudbrak R. Brown J. Dobson-Stone C. Carter S. Ramser J. White J. Healy E. Dissanayake M. Larrègue M. Perrussel M. Lehrach H. Munro C.S. Strachan T. Burge S. Hovnanian A. Monaco A.P. Hum. Mol. Genet. 2000; 9: 1131-1140Google Scholar). Mutation nomenclature has now been updated with respect to the 5′-end sequence published by Hu et al. (3Hu Z. Bonifas J.M. Beech J. Bench G. Shigihara T. Ogawa H. Ikeda S. Mauro T. Epstein Jr., E.H. Nat. Genet. 2000; 24: 61-65Google Scholar) and the results of our 5′ RACE-PCR experiments. Open table in a new tab Preparation of Antiserum to hSPCA1 Protein—Antibodies were raised essentially as described before (11Van Baelen K. Vanoevelen J. Missiaen L. Raeymaekers L. Wuytack F. J. Biol. Chem. 2001; 276: 10683-10691Google Scholar). ATP2C1 cDNA was amplified using primers ATP2C1CYTF (5′-GCATGCTGTGAAAAAGCTGCCTATTG-3′; ATP2C1d nt 1122–1141) and ATP2C1CYTR (5′-GTCGACGCAACCTTTGGTACTATTTG-3′; ATP2C1d nt 1954–1973) containing, respectively, SphI and SalI sites at their 5′-end. PCR cycling conditions were as follows: 94 °C for 2 min, followed by 30 cycles of 10 s at 94 °C, 30 s at 57 °C, and 60 s at 68 °C. After 10 cycles the extension time was increased to 90 s. The PCR amplification product encoded a fragment of hSPCA1 corresponding to the large cytoplasmic loop between transmembrane segments 4 and 5. The SphI/SalI-cut PCR fragment was ligated into the corresponding sites in the pQE-31 bacterial expression vector (Qiagen). Following expression in Escherichia coli using the QIAexpress Type IV System (Qiagen), the recombinant protein migrated at the expected size of 33 kDa on a 12% SDS-polyacrylamide gel. Protein Preparation and Immunostaining Analyses—For total protein extractions COS-1 cells were washed twice in phosphate-buffered saline before lysis in freshly prepared buffer (150 mm NaCl, 50 mm Tris HCl (pH 8), 5 mm EDTA, 2% SDS, 1 mm phenylmethylsulfonyl fluoride, and 1 μg/ml each of leupeptin, pepstatin, chymostatin, and antipain). After 30-min incubation on ice, insoluble proteins were pelleted at 6000 × g. Soluble protein was retained in the supernatant. Microsomes were isolated from COS-1 cells as described by Verboomen et al. (22Verboomen H. Wuytack F. De Smedt H. Himpens B. Casteels R. Biochem. J. 1992; 286: 591-596Google Scholar). Protein concentrations were determined by using the bicinchoninic acid method (Pierce, Rockford, IL). For Western blotting, proteins were separated and transferred onto Immobilon-P membranes (Millipore, Edinburgh, UK) using the NuPAGE® system with 4–12% Bis-Tris pre-cast gels (Invitrogen). For immunofluorescence, cells were fixed in 4% paraformaldehyde and permeabilized in 0.2% Triton X-100. Cells and membranes were quenched with 3% bovine serum albumin and 1% goat serum, followed by incubation with primary and secondary antiserum as described in the figure legends. Membranes were then incubated with ECF substrate (Amersham Biosciences) and fluorescent imaging was performed using a Storm 840 PhosphorImager in combination with ImageQuaNT™ software (Amersham Biosciences). 45Ca2+/54Mn2+ Fluxes—These radioactive isotope fluxes were essentially performed as described earlier (11Van Baelen K. Vanoevelen J. Missiaen L. Raeymaekers L. Wuytack F. J. Biol. Chem. 2001; 276: 10683-10691Google Scholar). Cells were loaded with 120 mm KCl, 30 mm imidazole-HCl (pH 6.8), 5 mm ATP, 0.44 mm potassium EGTA (pH 6.8), 10 mm NaN3, and 2 μm thapsigargin to inhibit SERCA activity. Unless otherwise stated, cells were loaded for 90 min and 5 mm potassium oxalate was included in the loading medium. MgCl2, CaCl2, and MnCl2 were added to obtain a free Mg2+ concentration of 0.5 mm (unless otherwise stated) and the indicated concentrations of free Ca2+ and Mn2+. Calcium ionophore A23187 (10 μm) was included in the efflux medium as indicated. Analysis of Phospho-intermediates—Phosphorylation from [γ-32P]ATP, processing of the acid-precipitated microsomal proteins, and acidic SDS-PAGE electrophoresis were essentially performed as described before (11Van Baelen K. Vanoevelen J. Missiaen L. Raeymaekers L. Wuytack F. J. Biol. Chem. 2001; 276: 10683-10691Google Scholar). The rate of dephosphorylation of the phosphoenzyme intermediate formed from ATP was analyzed after treatment with EGTA and ADP for serial time intervals prior to acid quenching (23Andersen J.P. Vilsen B. Leberer E. MacLennan D.H. J. Biol. Chem. 1989; 264: 21018-21023Google Scholar, 24Dode L. Vilsen B. Van Baelen K. Wuytack F. Clausen J.D. Andersen J.P. J. Biol. Chem. 2002; 277: 45579-45591Google Scholar). Phosphorylation was performed in 100 μl of a solution containing 160 mm KCl, 17 mm potassium Hepes (pH 7), 5 mm NaN3, 1 mm dithiothreitol, and 2 μm thapsigargin. Phosphorylation from 0.5 mm32Pi was performed at 25 °C as before (23Andersen J.P. Vilsen B. Leberer E. MacLennan D.H. J. Biol. Chem. 1989; 264: 21018-21023Google Scholar). Quantification of the separated phosphoenzyme band was performed by imaging using the Packard Cyclone™ storage phosphor system (Packard Bioscience, Berkshire, UK). Appropriate background phosphorylation levels were subtracted before data analysis as described (24Dode L. Vilsen B. Van Baelen K. Wuytack F. Clausen J.D. Andersen J.P. J. Biol. Chem. 2002; 277: 45579-45591Google Scholar). Alternative Splicing of ATP2C1 Primary Transcript—The primary aim of this study was to explore the stability and functional properties of a selected panel of mutated hSPCA1 proteins (see Table I), with each mutant incorporating one of the documented ATP2C1 missense mutations previously identified in HHD patients in our laboratory (2Sudbrak R. Brown J. Dobson-Stone C. Carter S. Ramser J. White J. Healy E. Dissanayake M. Larrègue M. Perrussel M. Lehrach H. Munro C.S. Strachan T. Burge S. Hovnanian A. Monaco A.P. Hum. Mol. Genet. 2000; 9: 1131-1140Google Scholar, 19Dobson-Stone C. Fairclough R. Dunne E. Brown J. Dissanayake M. Munro C.S. Strachan T. Burge S. Sudbrak R. Monaco A.P. Hovnanian A. J. Invest. Dermatol. 2002; 118: 338-343Google Scholar). We have chosen to assess the effect of these disease mutations when introduced into a new splice variant, ATP2C1d, described here for the first time. This variant was identified following the rapid amplification of cDNA ends (RACE)-PCR analysis, performed on cDNA reverse-transcribed from isolated human keratinocyte total RNA. As shown in Fig. 1A, 3′ RACE-PCR produced three strong bands of 450, 420, and 336 bp. Subcloning and subsequent sequencing confirmed that the 336- and 420-bp fragments, respectively, corresponded to the previously described splice variants ATP2C1a (2Sudbrak R. Brown J. Dobson-Stone C. Carter S. Ramser J. White J. Healy E. Dissanayake M. Larrègue M. Perrussel M. Lehrach H. Munro C.S. Strachan T. Burge S. Hovnanian A. Monaco A.P. Hum. Mol. Genet. 2000; 9: 1131-1140Google Scholar, 3Hu Z. Bonifas J.M. Beech J. Bench G. Shigihara T. Ogawa H. Ikeda S. Mauro T. Epstein Jr., E.H. Nat. Genet. 2000; 24: 61-65Google Scholar) and ATP2C1b (2Sudbrak R. Brown J. Dobson-Stone C. Carter S. Ramser J. White J. Healy E. Dissanayake M. Larrègue M. Perrussel M. Lehrach H. Munro C.S. Strachan T. Burge S. Hovnanian A. Monaco A.P. Hum. Mol. Genet. 2000; 9: 1131-1140Google Scholar), whereas the 450-bp band corresponded to a novel splice variant, designated ATP2C1d. A weaker 325-bp band, corresponding to a variant previously identified by Hu et al. (3Hu Z. Bonifas J.M. Beech J. Bench G. Shigihara T. Ogawa H. Ikeda S. Mauro T. Epstein Jr., E.H. Nat. Genet. 2000; 24: 61-65Google Scholar), was also amplified by PCR and was designated ATP2C1c in this study. However, the corresponding protein, hSPCA1c, is unlikely to play an important functional role in keratinocytes (see "Discussion"). The scheme in Fig. 1B depicts the four modes of alternative splicing, and Fig. 1C illustrates the structure of the corresponding mature mRNAs. ATP2C1d (hSPCA1d), identified in this study, results from activation of a novel internal 5′ donor splice site (designated D2) in exon 27 within codon Val-919 (nt 2756 relative to the ATG start codon). Val-919 is found immediately upstream of the exon 27 TGA translation stop codon, meaning that exon 27 reaches a maximum size of 128 bp. ATP2C1d mRNA is expressed to a high level in keratinocytes and is encoded by the full-length of optional exons 27 and 28, making ATP2C1d the longest alternatively spliced variant. In view of this, all our further experiments were performed using this variant, either in its wild type form or following the introduction of nine missense mutations identified in HHD patients (Table I). Characterization of the Transport Capacity for Ca2+ and Mn2+ of Novel Splice Variant hSPCA1d—For functional characterization, hSPCA1d was transiently expressed in COS-1 cells. First, its ability to transport Ca2+ and Mn2+ was investigated. Because the fraction of Golgi-derived membranes accounts for only a small proportion of total ER in microsomal fractions, conventional techniques used to measure SERCA-mediated Ca2+ transport activity into isolated microsomes (22Verboomen H. Wuytack F. De Smedt H. Himpens B. Casteels R. Biochem. J. 1992; 286: 591-596Google Scholar) could not be employed here. Alternatively, all Ca2+ uptake measurements were performed using detergent-permeabilized whole cells (25Missiaen L. De Smedt H. Parys J.-B. Casteels R. J. Biol. Chem. 1994; 269: 7238-7242Google Scholar). COS-1 cells transfected with empty vector (control) or wild type hSPCA1d cDNA were first permeabilized with saponin in a medium mimicking cytosolic composition and then loaded with 45Ca2+ in the presence of sodium azide (NaN3), to inhibit mitochondrial pump activity, and in the presence of thapsigargin, to inhibit SERCA pump activity. Efflux of the Ca2+ taken up by active pumps was then followed for 16 min in Ca2+-free medium, with the ionophore A23187 being added after 8 min. Fig. 2A shows that hSPCA1d-transfected cells exhibit a significantly higher thapsigargin-insensitive Ca2+-uptake compared with control cells. The addition of A23187 enhanced the release of stored Ca2+, suggesting that the Ca2+ had been transported into a membrane-delineated compartment. These results support the theory that hSPCA1d represents a thapsigargin-insensitive Golgi pump. Furthermore, the addition of 5 mm potassium oxalate to the loading medium stimulated an increase in the Ca2+-uptake capacity of hSPCA1 by more than 50%, when compared with uptake in the absence of oxalate. Oxalate acts to increase the ability of hSPCA1d to transport Ca2+ into the Golgi by precipitating Ca2+ and thus reducing the free Ca2+ concentration within the Golgi lumen. Fig. 2B compares the time-dependent loading of 45Ca2+ in hSPCA1d-expressing COS-1
Referência(s)