The secretory pathway Ca2+/Mn2+ ATPase (SPCA)2 is a Golgi-

localized pump with high affinity for Ca2+ ions*

Jo Vanoevelen, Leonard Dode, Kurt Van Baelen, Rebecca J. Fairclough1, Ludwig Missiaen, Luc Raeymaekers and Frank Wuytack2

From the Laboratorium voor Fysiologie, K.U.Leuven Campus Gasthuisberg O/N, Herestraat 49, B-3000 Leuven, Belgium

Running title: Characterization of human SPCA2

1Present address: Academic Endocrine Unit, Oxford Centre for Diabetes, Endocrinology and Metabolism, Churchill Hospital, Oxford, OX3 7LJ, UK 2To whom correspondence should be addressed: Tel.: +32-16-345936; Fax: +32-16-345991; E-mail: Frank.Wuytack@med.kuleuven.ac.be.

SUMMARY Accumulation of Ca2+ into the Golgi apparatus is mediated by sarco(endo)plasmic-reticulum Ca2+-ATPases (SERCA) and by secretory-pathway Ca2+-ATPases (SPCA). Mammals and birds express besides the housekeeping SPCA1 (human gene name ATP2C1; cytogenetic position 3q22.1) also an homologous SPCA2 isoform (human gene name ATP2C2; cytogenetic position 16q24.1). We show here that both genes present an identical exon/intron layout. We now confirm that hSPCA2 has the ability to transport Ca2+, demonstrate its Mn2+ transporting activity, show its Ca2+- and Mn2+-dependent phosphoprotein intermediate formation and document that these functional activities are insensitive to thapsigargin inhibition. The mRNA encoding hSPCA2 shows a limited tissue expression pattern mainly confined to the gastro-intestinal and respiratory tract, prostate, thyroid-, salivary- and mammary gland. Immunocytochemical localization in human colon sections presents a typical apical juxtanuclear Golgi-like staining. The expression in COS-1 cells allowed the direct demonstration of a 45Ca2+ (K0.5 = 0.27 µM) or 54Mn2+ transport into an A23187-releasable compartiment.

INTRODUCTION In vertebrates, three families of P-type Ca2+-ATPases control intracellular Ca2+ homeostasis:

PMCAs3, SERCAs and SPCAs. All these Ca2+ pumps contribute to the removal of activator Ca2+ from the cytoplasm after stimulation, thus decreasing the cytoplasmic Ca2+ concentration back to baseline levels. PMCAs extrude Ca2+ from the cytoplasm into the extracellular medium whereas SERCAs accumulate Ca2+ into the endoplasmic reticulum. Ca2+ uptake into the Golgi apparatus is mediated by both SERCA and SPCA Ca2+ pumps, reviewed in (1). The high concentration of Ca2+ in the lumen of the intracellular organelles is not only a source of activator Ca2+ for cytosolic processes, but is also indispensable for proper transcription, translation, translocation, folding and processing of secreted proteins (2). In the Golgi apparatus, luminal Ca2+ is also required for intra-Golgi membrane transport, transport between the Golgi and ER and endosome fusion. While the pivotal role of the ER as a Ca2+ store is well established, the view that also the Golgi apparatus can act as an agonist-releasable Ca2+ store is much more recent (3, 4). Also the knowledge of the Golgi-specific SPCA-type Ca2+-ATPases is much more limited than that of the well characterized SERCA and PMCA pumps. The PMR1 gene of Saccharomyces cerevisiae (5) is the first member of the SPCA family that has been described. The Pmr1 protein was localized to the Golgi apparatus or one of its subcompartments (6). Pmr1 is an ion-motive ATPase that supplies the secretory pathway with Ca2+ and Mn2+ ions required for glycosylation, sorting and ER-associated protein degradation (7, 8). Genes homologous to the S. cerevisiae PMR1 have been reported for a number of other fungi.

1

JBC Papers in Press. Published on April 14, 2005 as Manuscript M501026200

Copyright 2005 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

The characterization of the animal PMR1 homologues is more recent. Although the cDNA of the putative rat PMR1 homologue has already been cloned in 1992 using a SERCA-derived probe (9), the authors failed to functionally characterize this protein. Direct evidence that the Pmr1 homologue in animals is able to transport Ca2+ and Mn2+ was presented for the Caenorhabditis elegans homologue by Van Baelen et al.(10). Ton et al. (11) demonstrated the Ca2+-transport activity of the human homologue hSPCA1a. Later studies indicated that hSPCA1d isoform is also capable of transporting Mn2+ ions (12). The ability to transport Mn2+ at an appreciable rate is a characteristic not shared by the SERCA or PMCA Ca2+-transport ATPases. Like Pmr1 in yeast, the SPCA1 protein is localized in the Golgi apparatus. The human ATP2C1 gene which encodes hSPCA1 was recently identified as the defective gene in Hailey-Hailey disease (HHD, OMIM 16960) (13, 14). HHD is an autosomal dominant skin disorder which is characterized by suprabasal acantholysis of keratinocytes, resulting in epidermal blister formation. Up till now a total of 70 different mutations have been described in HHD patients (15). These mutations are scattered throughout the entire gene with no apparent clustering. The symptoms of HHD strongly resemble those of Darier-White disease (OMIM 124200), which is caused by mutations in one of the SERCA2 gene (ATP2A2) alleles (16). The link between defects in intracellular Ca2+ pumps and epidermal pathogenesis indicates that Ca2+ inside the lumen of intracellular stores plays an essential role in preserving skin integrity. The human genome also contains a second gene (ATP2C2) encoding a secretory pathway Ca2+/Mn2+-ATPase (hSPCA2) which has not been characterized except for a recent paper published while the present work was submitted (17). This group revealed hSPCA2 as a Ca2+

pump upon heterologous expression in yeast. Complementation studies in yeast suggested the importance of hSPCA2 in cellular Mn2+ detoxification. It was shown to be expressed in neuronal cells. In contrast to the report of Xiang et al. (17) we show here that hSPCA2 displays a high apparent affinity for Ca2+, similar to those values documented for hSPCA1a expressed in yeast (11) and for hSPCA1d expressed in COS-1 cells (12). Furthermore, the Ca2+- or Mn2+-dependent activation of phosphoenzyme

formation with [γ-32P]-ATP and the ability to pump both Ca2+ and Mn2+ are also demonstrated. Finally, the expression of hSPCA2 at both mRNA and protein levels has been analyzed showing that hSPCA2 has a much more restricted tissue distribution than hSPCA1 and colocalizes with Golgi-specific markers in colon epithelial cells. EXPERIMENTAL PROCEDURES Generation of the expression construct - The hSPCA2-encoding expression construct was generated by PCR using Clontech’s Advantage 2HF-polymerase. Human colon cDNA was used as a template. Two overlapping fragments were amplified separately using primers that introduce a HindIII restriction site at the 5’-end, indicated in lower case in the sequence shown below. The forward primer was NewhSPCA2F 5’-ATGTaagcttCGCCCGCTCACCATGGTCG AG-3’ and the reverse primer: OverlapSPCA2R1 5’-TAATATCACTTAAGT CCATCTTCATCGCCAG-3’. The 3’-fragment was amplified using the forward primer: OverlapSPCA2F1 5’-CTGGCGATGAAGATG GACTTAAGTGATATTA-3’ and the reverse primer SPCA2FullR2 5’-ATCGTgaattcACTA CACATCTTCAGGGTGCATCTG-3’ which introduces an EcoRI site (sequence in lower case) at the 3’-end. All primers were synthesized by Invitrogen. PCR products were subcloned into pGEM-T easy vectors using A/T-cloning. Subsequently, the 5’-fragment was ligated into pcDNA3 vectors using HindIII and EcoRI (present in the multiple cloning site of the pGEM-T easy vector). To complete the coding sequence, the construct containing the 5’-fragment in pcDNA3 was cut with AflII and EcoRI and ligated to AflII/EcoRI-cut 3’-fragments. The AflII site is a unique, endogenous restriction site in the overlapping part of the sequences. The sequence of the expression construct was verified before expression studies were attempted. RNA-extraction and reverse transcription - Total RNA was extracted from tissues or cultured cells using the Trizol™ reagent (Gibco BRL) using the procedure recommended by the manufacturer. PolyA+-RNA was purified from cultured cells with Invitrogen’s micro-FastTrack™ kit. Samples of total RNA of human

2

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

colon tissue and human colon carcinoma tissue were purchased from Stratagene. First-strand cDNA synthesis was performed on RNA samples using the Thermoscript™ RT-PCR System (Invitrogen) with oligo-dT priming. Ratio RT-PCR - cDNA, reverse transcribed from 1 µg of polyA+ mRNA or from total RNA was subjected to PCR using primers that amplify a 176-bp homologous fragment of both ATP2C1- and ATP2C2-derived messengers. The primer sequences were PMRRatioF 5’-CTGAAG(T or G)CTGCAGACATTGG-3’ and PMRRatioR 5’-CTCGTGCTCAGCTGGAATC-3’. The PCR program consisted of a 2 min denaturation step at 94 °C, followed by 27-30 cycles of the sequence 94 °C, 1 min; 53 °C, 1 min; 74 °C, 1 min. The final step was an additional elongation at 74 °C for 2 min. The distinction between ATP2C1 and ATP2C2 amplicons was made by digesting the PCR fragments with MseI and/or ScrFI. MseI cuts the ATP2C1 but not the ATP2C2 amplicon in two fragments of respectively 142 and 34 bp, while ScrFI digests only ATP2C2 and generates two fragments of 100 and 76 bp. After digestion, all PCR samples were subjected to 7.5% PAGE. The gels were run in TBE buffer at 250 V. The different bands were visualised by fluorescent staining with Vistra Green™ (Amersham Pharmacia Biotech) and scanning using a STORM™840 scanner (Amersham Pharmacia Biotech). Quantification was performed using ImageQuant™ software (Molecular Dynamics). Northern dot blot - The DNA hybridization probe was generated using PfuUltra (Stratagene) in 20 µl PCR reactions using 10 ng of plasmid DNA of the hSPCA2 expression construct as template. PCR products were 32P-labeled by the inclusion of 5 µl [α-32P]-dCTP (10 µCi/µl). The primers used for ATP2C2 were: ATP2C2NBlotF 5’-CGCCCGCTCACCATGGTCG-3’ and ATP2C2NBlotR 5’-CGCTCTGGCCAAATC CTCTTTC-3’ and generate a 216 bp-fragment corresponding to positions 78-293 of the published sequence. For ATP2C1, the primers were ATP2C1NBlotF 5’-GGGGGCTTCTCT TCCTTGTC-3’ and ATP2C1NBlotR 5’-CCATG AAAGGCTCGCCTATG-3’. The hybridization probe was separated from unincorporated 32P-labelled nucleotides using mini Quick Spin colums (Roche Molecular Diagnostics). A human MTE™ array was purchased from

Clontech (BD Biosciences). Probes were hybridized according to the manufacturer’s instructions and the labelling was detected by exposure to a Biomax film (Kodak). Cell culture - COS-1 cells were obtained from the ECACC and cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 3.8 mM L-glutamine, 85 IU/ml penicillin and 85 µg/ml streptomycin and 1% of non-essential amino acids. HeLa cervical carcinoma cells were obtained from the ATCC (Rockville, MD) and grown in Ham’s F-12 medium supplemented with 10% fetal calf serum and 85 IU/ml penicillin, 85 µg/ml streptomycin. LNCaP and T47D cells were purchased from ECACC and cultured in RPMI 1640 medium supplemented with 2 mM L-glutamine, 85 IU/ml penicillin, 85 µg/ml streptomycin and 10% fetal calf serum. 16HBE14o- bronchial epithelial cells were a gift from Dr. D.C. Gruenert (University of Vermont, Colchester, VT) and were cultured in a 50%/50% mixture of Dulbecco’s modified Eagle’s medium and Ham’s F-12 medium supplemented with 20% fetal calf serum, 3.8 mM L-glutamine, 85 IU/ml penicillin and 85 µg/ml streptomycin. Cultured human colon carcinoma cells (Caco2) were purchased from the ATCC and maintained in Dulbecco’s modified Eagle’s medium supplemented with 20% fetal bovine serum, 2 mM L-glutamine, 2 IU/ml penicillin and 2 mg/ml streptomycin. Human primary keratinocytes were isolated from human foreskin obtained from circumcision operations. Briefly, the tissue was cut into small pieces and incubated in dispase. The epidermis was peeled from the dermis and incubated in Ca2+- and Mg2+ -free PBS before collecting the cells by centrifugation (18). Keratinocytes were cultured in Keratinocyte-SFM medium (Life Technologies Inc., Merelbeke, Belgium) supplemented with bovine pituitary extract (25 µg/ml) and epidermal growth factor (0.1-0.2 ng/ml). To induce differentiation, the extracellular Ca2+ concentration was raised from 0.09 mM in the normal medium to 1.2 mM and the cells were allowed to grow to post-confluency. For 45Ca2+ fluxes, COS-1 cells were seeded in 12-well gelatin-coated dishes at a density of approximately 30000 cells/cm2 and investigated 8 days later. For microsomal preparation, cells were seeded in 100 mm culture dishes at 2.5 x 106 cells per plate.

3

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

Transfections - COS-1 cells were transiently transfected using the GeneJuice transfection reagent (Novagen). Transfection was performed the day after seeding, except for 45Ca2+ fluxes, in which case the period between seeding and transfection was five days to allow better attachment of the cells to the plates.

SPCA antisera - A hSPCA1 antiserum, designated hSPCA1cytl, has been previously described by our group (19). The hSPCA2-specific antiserum was generated by Eurogentec. Rabbits were immunized with a mixture of two internal peptides, coupled to keyhole limpet hemocyanin as a carrier protein. The internal peptides of hSPCA2 were SLKTEDQEDIY (position 514-524) and VDSVEKGELADRVGK (position 646-659) and were chosen in regions with low sequence homology to hSPCA1. The resulting antiserum was designated XIB. Immunocytochemistry - Cryosections (5 µm) of human colon tissue or cells grown on gelatine-coated coverslips were washed several times with PBS and fixed with 3% paraformaldehyde in PBS for 15 min at room temperature. The cells were permeabilized by incubation in 0.2% Triton X-100 for 2 min at room temperature. After three washes with PBS, non-specific binding was blocked by incubation for 1 h in PBS containing 3% BSA and 1/100 normal goat serum. Primary antibodies were diluted in 1% BSA in PBS and incubated for 1 h, followed by 3 washes with PBS. As negative controls, coverslips were incubated with pre-immune serum at the same dilution as the immune serum. Secondary antibodies were added in 1% BSA in PBS and incubated for 1 h. Secondary antibodies were goat anti-rabbit Alexa Fluor 488 or 594, or goat anti-mouse Alexa Fluor 488 or 594 (Molecular Probes). Finally, cells were washed and the coverslips were mounted in Vectashield (Vector Laboratories) to inhibit photobleaching, sealed with nail polish and examined with a fluorescence microscope (Olympus Cell^R). Subcellular structures could be identified using monoclonal antibodies, directed against components of the ER or the Golgi. A SERCA2 antibody (IID8, Affinity Bioreagents) was used to visualize the ER while the Golgi apparatus was identified using anti-TGN46 antibodies (Serotec).

Preparation of crude cell extracts and isolated membranes - For total protein extractions, cells were washed twice in PBS before lysis in freshly prepared SDS-extraction buffer (150 mM NaCl, 50 mM Tris-HCl pH 8, 5 mM EDTA, 2% SDS, supplemented with the protease inhibitors 0.28 mM PMSF, 0.83 mM benzamidine and 1 µg/ml each of leupeptin, aprotenin and pepstatin A). After a 30 min incubation on ice, the insoluble material was pelleted at 6000 x g and discarded. The supernatant was used for further experiments. Microsomes from transfected COS-1 cells were prepared as described by Verboomen et al. (20). Microsomes from tissues were prepared as follows. Fresh or frozen tissue was rinsed in cold PBS supplemented with 1 mM EDTA and allowed to equilibrate for 10 min on ice in hypotonic Buffer A (10 mM Tris-HCl pH 7.5, 1 mM MgCl2, 2 mM DTT, 1 mM EDTA and the protease inhibitor cocktail given above). Homogenization was performed using a teflon/glass Potter homogeniser. An equal volume of buffer B (10 mM Tris-HCl pH 7.5, 2 mM DTT, 1 mM EDTA, 300 mM KCl and 500 mM sucrose) was added and homogenization was continued. Cellular debris and nuclei were spun down at 4000 x g for 15 min at 4°C. The supernatant was re-centrifuged at 100000 x g for 40 min at 4°C followed by resuspension of the microsomal pellet in Buffer C (10 mM Tris-HCl pH 7.5, 2 mM DTT, 150 mM KCl and 250 mM sucrose) (21). Quantification of total protein was done using the bicinchoninic acid method (Pierce), using the manufacturer’s instructions. When compounds in the samples interfered with this method, the protein was precipitated with trichloroacetic acid prior to quantification by the Lowry method. Western blotting - Microsomes or crude cell extracts were loaded on NuPage™ Bis-Tris 4-12% gradient gels or NuPage™ 7% Tris-acetate gels (Invitrogen). After electrophoresis, the separated proteins were transferred to Immobilon-P membranes. The blots were quenched in TBST, supplemented with 5% (w/v) non-fat dry milk. When necessary, the primary antibody solution was pre-incubated with 10 µg of the immunogenic peptides for 30 min. After 3 washes in TBST, incubation with the primary antibody in TBST containing 1% non-fat dry milk was performed for 1 h at room temperature. After labelling with the secondary antibody coupled to alkaline phosphatase (dilution 1/8000), the labelled bands were detected using

4

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

Vistra ECF (Amersham Pharmacia Biotech) as substrate. Detection of the fluorescence was performed by the STORM™840 scanner (Amersham Pharmacia Biotech) in combination with ImageQuant™ software (Molecular Dynamics). Isotope fluxes - 45Ca2+ fluxes were performed on cells permeabilized with saponin. Cells grown in 12-well, gelatin-coated plates were treated for 10 min with 20 µg/ml saponin in permeabilization buffer (120 mM KCl, 30 mM imidazole-HCl pH 6.8, 2 mM MgCl2, 1 mM ATP, 1 mM EGTA) at 25 °C. After a wash with the same buffer without saponin, the non-mitochondrial stores were loaded with 45Ca2+ (3 µCi/ml) or 54Mn2+ (10 µCi/ml) for 90 min in a buffer containing: 120 mM KCl, 30 mM imidazole-HCl pH 6.8, 5 mM ATP, 10 mM NaN3, 0.44 mM EGTA and 100 nM thapsigargin (SERCA-specific inhibitor). To obtain a free Mg2+ concentration of 0.5 mM and the indicated free Ca2+ and Mn2+ concentration, the total MgCl2, CaCl2 and MnCl2 concentrations were calculated using the CaBuf program (ftp.cc.kuleuven.ac.be/pub/droogmans/ cabuf.zip) developed by G. Droogmans in our lab These calculations are based on the dissociation constants given by Fabiato and Fabiato (22) for Ca2+ and by Martell and Smith for Mn2+ (23). After the loading phase, cells were washed twice with 1 ml of efflux medium (120 mM KCl, 30 mM imidazole-HCl pH 6.8, 1 mM EGTA). Subsequently, the efflux medium was replaced every 2 min to monitor the leak of Ca2+. At the end of each experiment, the remaining 45Ca2+ in the cells was released by the addition of 1 ml of 2% (w/v) SDS for 30 min. 45Ca2+ or 54Mn2+ was quantified by liquid scintillation counting. Formation of the phosphoenzyme intermediate - Microsomes (10 µg) were phosphorylated on ice in a 100 µl reaction mixture containing 1 µl [γ-32P]-ATP of 2 µCi/µl (Amersham Pharmacia Biotech), 160 mM KCl, 17 mM Hepes pH 7.0, 1 mM DTT, 5 mM NaN3 and the appropriate concentrations of Ca2+, Mn2+ and EGTA. The reaction was stopped after 20 s by the addition of ice-cold stop solution (6% trichloroacetic acid, 10 mM phosphoric acid, 1 mM ATP). The protein was allowed to precipitate on ice for 30 min, and pelleted by centrifugation at 4°C for 3 min at 13200 rpm. Pellets are washed two times with stop solution and finally with 0.2 M sodium acetate before solubilization in modified SDS-

PAGE sample buffer (150 mM Tris-HCl pH 7.4, 8 mM EDTA, 3% SDS, 20% sucrose, 0.14 mg/ml bromophenol blue and 10 mM DTT). Phosphorylated proteins were separated on acid 7.5% SDS-PAGE according to (24). After fixation of the gels in 7.5% acetic acid they were dried between sheets of gel drying film (Promega) and exposed to a PhosphoImager™ screen (Molecular Dynamics) for quantification. Background phosphorylation due to endogenous Ca2+ pumps in this expression system is negligible (25, 26). RESULTS ATP2C2 encodes a P-type ATPase The GenBank entry of the cDNA sequence KIAA0703 (27) was the first indication of a putative second isoform of the human secretory-pathway Ca2+-ATPase. The existence of a second hSPCA isoform has previously been mentioned in the literature (11, 28), but no further studies were published until the recent paper by Xiang et al. (17). Our initial studies showed that overexpression of the protein encoded by the KIAA0703 cDNA in COS-1 cells did not result in the formation of a functional Ca2+-ATPase (data not shown). It is now clear that the reason behind this failure lies in the faulty N-terminus. The KIAA0703 sequence contains three putative in-frame ATG start codons at positions 441, 501 and 606. We assembled expression constructs for each of the three putative protein products. All three proteins could be overexpressed, but this did not result in an enhanced Ca2+-transport capacity, indicating that non-functional proteins were formed (data not shown). Moreover, these proteins were not localized in the Golgi but instead were found in the ER and they were susceptible to enhanced degradation since Western blotting showed an intense additional band corresponding to a smaller proteolytic fragment (data not shown). Further analysis of chromosome 16 suggested the existence of two alternative exons located more upstream in the genomic sequence. The first two exons of the KIAA0703 cDNA sequence were replaced by these two new exons, thus generating a new cDNA clone whose nucleotide sequence was deposited under accession no. AY791884. In contrast to the corresponding sequence in KIAA0703, the revised 5’-end of the new cDNA clone was

5

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

highly similar to that of the SPCA2 mRNA sequence from rat found in the database. In addition, the first two exons of ATP2C2 are separated by a large intron of about 30 kb which is similar in size to the corresponding intron of ATP2C1. The complete genomic exon/intron layout is shown in Fig. 1. ATP2C2 spans a region of 95.5 kb on human chromosome 16 (16q24.1) and consists of 27 exons, ranging in size from 36 bp to 393 bp. The introns are between 86 bp and 29785 bp in length. The putative translation initiation ATG codon (ACCATGG) is embedded in a Kozak consensus sequence (29). Gene transcription and processing results in a 3386-bp cDNA of which 2841 bp comprise the open reading frame. The 946 amino acid sequence of the encoded protein is identical to that studied by Xiang et al. (17). Expression pattern of the ATP2C2 mRNA Northern dot-blot hybridization analysis was performed to compare the expression pattern of the novel gene with that of its previously identified family member ATP2C1 (Fig. 2). Whereas there was little variation in the level of the ubiquitous signal of ATP2C1 (Fig. 2B), the new paralogue presented large differences in the relative expression levels among various tissues (Fig. 2A). The strongest signals were detected throughout the gastro-intestinal tract, from the stomach (B5) to e rectum (C6), with the mostintense signal in e rectum (C6). Other positivetissues include trachea (H7), fetal lung (G11),prostate tissue (E8), thyroid gland (D9), salivarygland (E9) and mammary gland (F9). To better quantify the relative expression levelsof ATP2C1 and ATP2C2 mRNA in differenttissues and cell lines, we used a ratio RT-PCRprotocol. In this method, a common set ofprimers is used to co-amplify homologousfragments of related sequences (30). Fig. 3Ashows a typical experiment in which the relativeamount of both essengers is determined byrestriction analys of the PCR products. RatioRT-PCR was performed on mRNA samplesfrom a number of human tissues (lung, colon,colon carcinoma) that tested positive forATP2C2 mRNA expression (Fig. 2A), fromcultured cells derived from positive tissues(Caco2: colon; LNCaP: prostate; 16HBE14o-:lung and T47D: mammary gland), and fromother cultured human cell lines (SH5Y-5Y,HEK, EA, HeLa). The latter cell lines expressedonly the ATP2C1 mRNA (data not shown),

whereas Caco2, LNCaP, 16HBE14o- and T47D cells did co-express the ATP2C2 and ATP2C1 mRNAs. Both transcripts were detected in all above-mentioned human tissues (Fig. 3B). The highest ATP2C2/ATP2C1 ratio (~1:1) was found in colon tissue and in cell lines derived from colon, lung (16HBE14o-) or prostate (LNCaP). Additionally, human keratinocytes were also included in this study because of the central role of hSPCA1 in Hailey-Hailey disease, and because skin tissue was not represented on the master blot (Fig. 2). Human keratinocytes tested negative for the ATP2C2 mRNA when cultured under proliferating conditions (data not shown). However, when keratinocytes were allowed to differentiate (by growing them to confluency in the presence of high extracellular Ca2+ concentration), the expression of both genes could be demonstrated (Fig. 3B). Characterization of the hSPCA2 protein The hSPCA2-specific antibodies (XIB) have been generated against a mixture of two SPCA2-specific peptides that are not conserved in hSPCA1. The antiserum labelled a band on Western blots of hSPCA2-overexpressing COS-1 cells but not of control COS-1 cells (Fig. 4A). The specificity of the antiserum was proven by the absence of the immunoreaction with pre-immune serum and its suppression by pre-incubation of the serum with the immunogenic p pti s. The immunoreactive protein migrated s gh below the predicted Mr value of

by guest on Januarhttp://w

ww

.jbc.org/D

ownloaded from

1hhbdnC4acthmpmbIhFmlucm

6

eli

03,2as aomolotsemoot aa2+-A). juxorree cidd

redoore

otton a S Ci . icrong.

olonRN

detly

93. A similar anomalous faster migration lso been observed for the C. elegans SPCA logue (10). Immunostaining of Western

for other organellar Ca2+ pumps nstrates that overexpression of hSPCA2 did lter the expression levels of the endogenous ATPases SERCA2b and hSPCA1 (Fig. The overexpressed protein was localized in tanuclear region in COS-1 cells. This region sponded to the Golgi apparatus as shown by

localization experiments depicted in the panels of Fig. 4B. SERCA2, which

y 6, 2020

ole

minantly stains the ER, showed a much diffuse, reticular staining pattern (Fig. 4B, m panels). next series of experiments, the endogenous A2 protein was studied in human colon. 5 shows a Western blot performed with

Pg

mis

thth

somes isolated from human colon and The protein could only be demonstrated in , a tissue with one of the highest hSPCA2 A expression levels. The immunoreactive

band migrated slightly higher than the protein overexpressed in COS-1 cells. This could be due to post-translational modifications. An additional band of lower mobility was present in colon microsomes but this band was non-specific since it was not inhibited by pre-incubation of the antibody with the immunogenic peptides (Fig. 5). Immunohistochemical staining of human colon cryosections resulted in hSPCA2-specific labelling in close proximity of the apical pole of the nuclei of colon epithelial cells. This juxtanuclear staining colocalized with the Golgi marker, TGN46 (Fig. 6). hSPCA1 showed a similar juxtanuclear staining, indicating that hSPCA1 and hSPCA2 reside in the same or closely juxtaposed subcellular compartments. The Golgi marker TGN46 appeared to label all epithelial cells because all nuclei were associated with a Golgi-like staining. Additionally, cells stained for the Golgi marker were also positive for both hSPCA1 and hSPCA2. hSPCA2 is a functional Ca2+- and Mn2+-transporting enzyme The hSPCA2 protein was functionally characterized by heterologous overexpression in COS-1 cells. The transient formation of a phosphorylated intermediate is a key feature of all P-type ion-transport ATPases. To visualize the phosphorylated hSPCA2 intermediate, control cells (transfected with empty vector) and hSPCA2-overexpressing COS-1 cells were phosphorylated in vitro using [γ-32P]-ATP. The phosphorylated intermediate can be preserved on SDS polyacrylamide gels in acidic conditions. Overexpressing cells clearly showed a phosphorylated enzyme intermediate in the presence of either Ca2+ or Mn2+ (Fig. 7). This phosphorylated intermediate was also formed when the phosphorylation reactions were performed in the presence of 100 nM of the SERCA-specific inhibitor thapsigargin, thus preventing the phosphointermediate formation from the endogenous SERCA2b. The specificity of the reactions is illustrated by the absence of signals when the experiments were done in the absence of Ca2+ or Mn2+ (EGTA). The bound radioactive phosphate was completely removed in the presence of hydroxylamine, demonstrating that the phosphate group was bound to a carboxyl and not to a hydroxyl group (Fig. 7B). To study the ion-transport activity in overexpressing cells we performed 45Ca2+ fluxes on permeabilized cells (10). In a first series of

experiments we measured the Ca2+ accumulation in the stores of overexpressing COS-1 cells as a function of time. Permeabilized COS-1 cells were incubated for different time periods in a medium resembling the cytosolic ionic composition and containing 45Ca2+. To exclude the interference of the endogenous SERCA2b pump, the experiments were performed in the presence of 100 nM thapsigargin. Fig. 8A shows that the time-dependent loading of 45Ca2+ in hSPCA2-overexpressing COS-1 cells was higher than in control cells. This difference increased with time both in absolute and relative terms. Because the cells are more prone to detach during prolonged incubation time, the loading time was restricted to 90 min as previously described (12). To determine the Ca2+-dependence of the hSPCA2-mediated Ca2+ uptake, permeabilized cells were loaded at different free Ca2+ concentrations for a fixed time of 90 min and the difference in the level of 45Ca2+ uptake of hSPCA2-overexpressing cells and control cells was measured (Fig. 8B). Half-maximal activation (K0.5) of the enzyme was observed at 0.27 µM free Ca2+ concentration. Since previous reports on the SPCA1 pump of humans (12) and C. elegans (10) have shown that SPCA1 can act as a Mn2+ pump, we tested whether hSPCA2 also functions as a Mn2+-transporting enzyme. Control cells and hSPCA2-overexpressing cells were loaded with 54Mn2+ for 90 min in the presence of 100 nM thapsigargin and the efflux of 54Mn2+ was followed for 18 min. The higher uptake of 54Mn2+ in hSPCA2-overexpressing cells provides direct evidence that hSPCA2 can act as a Mn2+-transporting ATPase (Fig. 8C). The efflux of accumulated Mn2+ was accelerated by the ionophore A23187 (10 µM), demonstrating that the Mn2+ ions had been transported into a membrane-delineated compartment. In the presence of A23187, the efflux curves of control cells and hSPCA2-overexpressing cells converge, showing that the amount of background 54Mn2+ binding is identical and that the difference between the two curves is specifically due to an enhanced Mn2+-transporting capacity of hSPCA2-expressing cells. DISCUSSION In this paper we confirm the finding of Xiang et al. (17) that human tissues can express a second

7

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

isoform of secretory pathway Ca2+-ATPase, named hSPCA2, encoded by the ATP2C2 gene. We further explored its tissue-specific and cellular expression pattern and present its functional characteristics upon heterologous expression in COS-1 cells. A TBLASTN search with an SPCA signature peptide sequence (IQEYRSEKSLEELTK) revealed that mammalian genomes (Homo sapiens, Pan troglodytes, Canis familiaris, Bos taurus, Rattus norvegicus and Mus musculus) all contain two different SPCA-encoding genes. Also the chicken genome (Gallus gallus) appears to contain two related genes. We did not find any indication for the presence of a second SPCA isoform in the fish Danio rerio, Fugu rubripes and Tetraodon nigroviridis. This is remarkable since in general, as a result of an additional genome duplication, euteleost fishes seem to have larger gene families than tetrapods (31). Whereas euteleost genomes contain more SERCA genes than mammals, this is apparently not the case for the SPCA genes. Invertebrates (Drosophila melanogaster, Anopheles gambiae, Apis mellifera and Caenorhabditis elegans) also seem to possess only one SPCA1 gene but these lower organisms also contain only a single SERCA gene. Given the complete description of the human genome, the ATP2C2 gene is probably the last functional gene to be identified in the multigene family of type 2 Ca2+-transport ATPases. This superfamily consists of three SERCA-encoding genes (ATP2A1-3), four PMCA genes (ATP2B1-4) and two SPCA-encoding genes (ATP2C1-2). hSPCA1 and hSPCA2 mRNAs present different tissue distributions. hSPCA1 was recognized as a housekeeping enzyme (13) and this observation is confirmed and extended by the result shown in Fig. 2B. The expression pattern of hSPCA2 mRNA is much more heterogeneous. The highest levels occur in the gastro-intestinal tract and in a number of secretory tissues as determined by Northern dot-blot hybridization and by ratio RT-PCR. Also the tissue distribution of EST clones in the database is compatible with this pattern. These results are different from the data published by Xiang et al. who showed abundant hSPCA2 protein expression in brain and testes (17). In our hands, the mRNA expression of hSPCA2 in brain and testes was in the lower range (Fig.2). Because skin tissue was not represented in the RNA blots used and because of the central role of the hSPCA1 pump in Hailey-Hailey disease,

we investigated the co-expression of hSPCA1 and hSPCA2 in a separate set of experiments on isolated keratinocytes using PCR. We demonstrate now that hSPCA2 is expressed at the mRNA level in differentiated keratinocytes. The co-expression of hSPCA2 and hSPCA1 could help to explain the relatively high contribution of thapsigargin-insensitive Ca2+ accumulation in the Golgi of keratinocytes (33) and lung-derived 16HBE14o- cells (34). The proper function of the secretory pathway in keratinocytes may be crucial for correct delivery of cell adhesion components (15). The fact that hSPCA2 is not able to compensate for the decreased level of hSPCA1 in keratinocytes of Hailey-Hailey disease patients may suggest a specific function for this pump distinct from that of hSPCA1. This hypothesis fits with the observation that a complete knockout of the ATP2C1 gene in mice results in an embryonic lethal phenotype (Doetschman T. & Shull G. E., unpublished data, http://www-apps.niehs.nih. gov/centers/public/fac-core/ctr-640-2953.htm, 2004). The yeast Pmr1 pump has been located to the medial-Golgi compartment (6, 7). hSPCA1 is also a Golgi-resident protein (11) and comigrates with markers for the trans-Golgi compartment (32). hSPCA2 shares the property of Golgi localization with hSPCA1, but the exact localization inside the Golgi has not been studied yet. In epithelial cells of human colon tissue, both hSPCA isoforms show the same Golgi-like localization. However, the resolution of immunofluoresence microscopy may not be adequate to determine if both pumps are in the same subcompartment of the Golgi. Reports on the mouse SPCA1 indicate that it can also be present in post-Golgi compartments (35). The co-expression of hSPCA2 with hSPCA1 suggests that hSPCA2 supplements the role of hSPCA1 in the thapsigargin-insensitive Ca2+/Mn2+-accumulation pathway in the Golgi of a number of cell types. The expression in secretory cells suggests that hSPCA2 assists in the function of hSPCA1 or that it performs an as yet unknown specialized function in those cell types in which the Golgi apparatus is important not only for the basic cell-biological maintenance of cell function but also for the specific secretory task these cells perform in the organism. The large number of epithelial cells expressing hSPCA1 and hSPCA2 in colon epithelium cannot be accounted for by the number of mucus-secreting cells only, indicating

8

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

that these Ca2+ pumps also are present in enterocytes and play an important role in their function. Xiang et al. reported a slightly different localization of hSPCA2. They found hSPCA2 in TGN-derived vesicles in rat hippocampal neurons (17). We show a predominantly juxtanuclear Golgi-like distribution in human colon epithelial cells. This difference most probably reflects variations in the architecture of the secretory pathway in different cell types. Our functional studies of the overexpressed protein in mammalian cells confirmed the findings of Xiang et al. that hSPCA2 is a P-type ATPase, capable of transporting both Ca2+ and Mn2+ with high affinity. Half-maximal (K0.5) stimulation of the enzyme occurred at a free Ca2+ concentration of 0.27 µM, which is in the same range as described for the human SPCA1a protein (0.26 µM) (11), hSPCA1d (0.20 µM) (11) and its homologue in C. elegans (0.25 µM) (10). The yeast homologue Pmr1 shows a slightly higher apparent affinity for Ca2+ (≤ 0.1 µM) (8). When hSPCA2 is overexpressed in yeast, a lower half maximal activation of the enzyme for Ca2+ is observed (K0.5 = 1.35 µM) (17). This discrepancy can be explained by the use of different expression systems i.e. yeast vs. mammalian cells. The Golgi apparatus of yeast cells is less structured than that of mammalian cells and the membranes may have a different lipid composition, which may affect transport activity. The involvement of additional essential cofactors present in mammalian cells, but not in yeast, can also not be excluded. Analogous to hSPCA1, Ca2+ transport by the hSPCA2 protein is insensitive to thapsigargin (SERCA-specific inhibitor). We also demonstrated in this study that hSPCA2 is capable of transporting Mn2+. The phosphorylation properties of hSPCA2 were further investigated by testing its ability to be phosphorylated in the presence of Ca2+ and Mn2+. In each case, a phosphoprotein intermediate could be demonstrated. Furthermore, the formation of the phosphointermediate was sensitive to hydroxylamine, showing that the phosphorylation occurs on the catalytic carboxyl group of an aspartate residue, thus excluding the possibility that it was the result of protein kinase activity. In conclusion, we have functionally characterized a second isoform of SPCA-type ATPase. Both hSPCA2 overexpressed in COS-1

cells and hSPCA2 endogenously present in colon epithelial cells display a Golgi-like juxtanuclear distribution. hSPCA2 is capable of transporting Ca2+ with high affinity in a thapsigargin-independent way. Further studies are required to characterize the specific properties of hSPCA2 and its function in cellular divalent ion homeostasis.

9

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

REFERENCES

1. Wuytack, F., Raeymaekers, L., and Missiaen, L. (2002) Cell Calcium 32, 279-305 2. Corbett, E. F., and Michalak, M. (2000) Trends Biochem. Sci. 25, 307-311 3. Missiaen, L., Van Acker, K., Parys, J. B., De Smedt, H., Van Baelen, K., Weidema, A. F.,

Vanoevelen, J., Raeymaekers, L., Renders, J., Callewaert, G., Rizzuto, R., and Wuytack, F. (2001) J. Biol. Chem. 276, 39161-39170

4. Pinton, P., Pozzan, T., and Rizzuto, R. (1998) EMBO J. 17, 5298-5308 5. Rudolph, H. K., Antebi, A., Fink, G. R., Buckley, C. M., Dorman, T. E., LeVitre, J.,

Davidow, L. S., Mao, J. I., and Moir, D. T. (1989) Cell 58, 133-145 6. Antebi, A., and Fink, G. R. (1992) Mol. Biol. Cell 3, 633-654 7. Durr, G., Strayle, J., Plemper, R., Elbs, S., Klee, S. K., Catty, P., Wolf, D. H., and Rudolph,

H. K. (1998) Mol. Biol. Cell 9, 1149-1162 8. Sorin, A., Rosas, G., and Rao, R. (1997) J. Biol. Chem. 272, 9895-9901 9. Gunteski-Hamblin, A. M., Clarke, D. M., and Shull, G. E. (1992) Biochemistry 31, 7600-7608 10. Van Baelen, K., Vanoevelen, J., Missiaen, L., Raeymaekers, L., and Wuytack, F. (2001) J.

Biol. Chem. 276, 10683-10691 11. Ton, V. K., Mandal, D., Vahadji, C., and Rao, R. (2002) J. Biol. Chem. 277, 6422-6427 12. Fairclough, R. J., Dode, L., Vanoevelen, J., Andersen, J. P., Missiaen, L., Raeymaekers, L.,

Wuytack, F., and Hovnanian, A. (2003) J. Biol. Chem. 278, 24721-24730 13. Hu, Z., Bonifas, J. M., Beech, J., Bench, G., Shigihara, T., Ogawa, H., Ikeda, S., Mauro, T.,

and Epstein, E. H., Jr. (2000) Nat. Genet. 24, 61-65 14. Sudbrak, R., Brown, J., Dobson-Stone, C., Carter, S., Ramser, J., White, J., Healy, E.,

Dissanayake, M., Larregue, M., Perrussel, M., Lehrach, H., Munro, C. S., Strachan, T., Burge, S., Hovnanian, A., and Monaco, A. P. (2000) Hum. Mol. Genet. 9, 1131-1140

15. Foggia, L., and Hovnanian, A. (2004) Am. J. Med. Genet. 131, 20-31 16. Sakuntabhai, A., Ruiz-Perez, V., Carter, S., Jacobsen, N., Burge, S., Monk, S., Smith, M.,

Munro, C. S., O'Donovan, M., Craddock, N., Kucherlapati, R., Rees, J. L., Owen, M., Lathrop, G. M., Monaco, A. P., Strachan, T., and Hovnanian, A. (1999) Nat. Genet. 2, 271-277

17. Xiang, M., Mohamalawari D., and Rao, R. (2005) J. Biol. Chem 280, 11608-11614 18. Verstuyf, A., Verlinden, L., Van Baelen, H., Sabbe, K., D'Hallewyn, C., De Clercq, P.,

Vandewalle, M., and Bouillon, R. (1998) J. Bone Miner. Res. 13, 549-558 19. Van Baelen, K., Vanoevelen, J., Callewaert, G., Parys, J. B., De Smedt, H., Raeymaekers, L.,

Rizzuto, R., Missiaen, L., and Wuytack, F. (2003) Biochem. Biophys. Res. Commun. 306, 430-436

20. Verboomen, H., Wuytack, F., De Smedt, H., Himpens, B., and Casteels, R. (1992) Biochem. J. 286, 591-595

21. Reinhardt, T. A., Filoteo, A. G., Penniston, J. T., and Horst, R. L. (2000) Am. J. Physiol. Cell Physiol. 279, C1595-C1602

22. Fabiato, A., and Fabiato, F. (1979) J. Physiol. (Paris) 75, 463-505 23. Martell, A. E., and Smith, R. M. (1974) Critical Stability Constants, Plenum Press, New

York, NY 24. Sarkadi, B., Enyedi, A., Foldes-Papp, Z., and Gardos, G. (1986) J. Biol. Chem. 261, 9552-

9557 25. Sorensen, T. L-M., Dupont, Y., Vilsen, B., and Andersen J. P. (2000) J. Biol Chem. 275,

5400-5408 26. Dode, L., Andersen, J. P., Leslie, N., Dhitavat, J., Vilsen, B., and Hovnanian, A. (2003) J.

Biol. Chem 278, 47877-47889 27. Ishikawa, K., Nagase, T., Suyama, M., Miyajima, N., Tanaka, A., Kotani, H., Nomura, N.,

and Ohara, O. (1998) DNA Res. 5, 169-176 28. Kraev, A., and MacLennan, D. H. (2002) Ann. NY. Acad. Sci. 976, 53-59

10

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

29. Kozak, M. (1987) Nucleic. Acids Res. 15, 8125-8148 30. Wuytack, F., Papp, B., Verboomen, H., Raeymaekers, L., Dode, L., Bobe, R., Enouf, J.,

Bokkala, S., Authi, K. S., and Casteels, R. (1994) J. Biol. Chem. 269: 1410-1416 31. Taylor, J. S., Braasch, I., Frickey, T., Meyer, A., and Van de Peer, Y. (2003) Genome Res. 13,

382-390 32. Behne, M. J., Tu, C-L, Aronchik, I., Epstein, E., Bench, G., Bikle, D. D., Pozzan, T., and

Mauro, T. M. (2003) J. Invest. Dermatol. 121, 688-694 33. Callewaert, G., Parys, J. B., De Smedt, H., Raeymaekers, L., Wuytack, F., Vanoevelen, J.,

Van Baelen, K., Simoni, A., Rizzuto, R., and Missiaen, L. (2003) Cell Calcium 34, 157-162 34. Vanoevelen, J., Raeymaekers, L., Parys, J. B., De Smedt, H., Van Baelen, K., Callewaert, G.,

Wuytack, F., and Missiaen, L. (2004) Cell Calcium 35, 115-121 35. Mitchell, K. J., Tsuboi, T., and Rutter, G. A. (2004) Diabetes 53, 393-40036.

11

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

FOOTNOTES Aknowledgments - We thank I. Willems, M. Schuermans and T. Luyten for excellent technical assistance. A. Floorizone, M. Crabbé and S. De Swaef are acknowledged for maintaining the cell cultures. The Department of Morphology and Molecular Pathology (UZ-Leuven) is acknowledged for providing the human tissue samples. *This work was supported by the Interuniversity Attraction Poles Programme – Belgian Science Policy P5/05, and by the Fonds voor Wetenschappelijk Onderzoek Vlaanderen G.0166.04. The nucleotide sequence reported in this paper has been submitted to the GenBank™/EBI Data Bank with accession number AY791884 (hSPCA2). Other GenBank entries that were mentioned were AB14603 (KIAA0703), AC022165 (human chromosome 16), AF484685 (Rat SPCA2), AF181120 (hSPCA1a), AY268375 (hSPCA1d), M12898 (SERCA1). 3The abbreviations used are: PMCA, plasma-membrane Ca2+-ATPase; SERCA, sarco(endo)plasmic-reticulum Ca2+-ATPase; SPCA, secretory-pathway Ca2+-ATPase; ER, endoplasmic reticulum; TBE, Tris Borate EDTA; MTE, multiple tissue expression array; ECACC, European collection of cell culturing; ATCC, American Type Culture Collection; PBS, phosphate buffered saline; BSA, bovine serum albumin; TBST, tris buffered saline containing 0.05% Tween; TG, thapsigargin; EST, expressed sequence tag

12

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

FIGURE LEGENDS Figure 1 Intron-exon layout of ATP2C2. Nucleotide sequence of the exon-intron boundaries and the size of the introns are shown. Exon sequences are indicated in uppercase while introns are in lowercase. The deduced amino-acid sequence at each junction is displayed below the exon sequences. Exon-intron boundaries are indicated by slashes, and the conserved GT and AG nucleotides flanking the introns are shown in boldface. Figure 2 mRNA expression pattern of the ATP2C2 and ATP2C1 mRNA in human tissues. Human mRNA dot-blot hybridization using an ATP2C2 (A) or ATP2C1 (B) -specific probe. Separate human Multiple Tissue Expression Array blots (Stratagene) were used to obtain the figures in A and B. The localization of the different samples is identical, except for the bone marrow sample (G7), which is absent from the blot in panel A. The origin and positions of RNA samples (columns 1-11) and controls (column 12) are shown (C). Figure 3 Ratio RT-PCR of ATP2C1/ATP2C2-transcripts in different human tissues and cell lines. (A) Typical example of a ratio RT-PCR on LNCaP cells expressing both genes. Amplicons derived from ATP2C1 and ATP2C2 were discriminated by restriction digestion and electrophoresis. Lanes RNA- and RT- show the negative controls. The sample in lane RNA- contained no template. The RT- lane is an RNA sample that was incubated in parallel with the reverse transcription but without added reverse transcriptase. Aliquots of the PCR product were not treated (uncut) or cut with MseI (specific for ATP2C1), ScrFI (specific for ATP2C2) or their combination. The bands were quantified by fluorescent staining of the gel. (B) Relative quantification of ATP2C1 and ATP2C2 messengers by ratio RT-PCR in different human cell types and tissues. Results are the mean ± SE of 3 or 4 independent experiments. A s. igure 4 Overexpress n of hSPCA2 in COS-1 ce

(A) Western blot of extracts of COS-1 cells transfected with empty vector or with a hSPCA2 onstruct. Lane 1 and 2 were loaded with respecti ely 5 µg and 18 µg of SDS-extract of

by guest on January 6, 20http://w

ww

.jbc.org/D

ownloaded from

Ec

control COS-1. Lanes 3 and 4 contain respectively transiently overexpressing hSPCA2. The blots werXIB antibody (1/1000) or hSPCA1 using the hSPCAdetected using a polyclonal antibody (1/2000) descimages of hSPCA2-overexpressing COS-1 cells. stained with the XIB pre-immune serum (1/1000). (stained with the XIB antibody (1/1000), pre-incubatblock specific signals. The middle panels show a Golgi marker TGN46 (d). The bottom panels show ER-resident protein SERCA2 (f). The antibodies u(1/1000), anti-hSPCA2 XIB (1/1000), anti-TGN46pictures were overlayed with the DAPI staining pnuclei. The scale bar represents 20 µm. Figure 5 Western blot of endogenously expressedWestern blot using anti-hSPCA2 (XIB, 1/1000). Firlane: 30 µg lung microsomes; third lane: 15 µg of h

13

BDll

CF

Fv

D

5 µg and 18 µg of extract of COS-1 cells e immunostained for hSPCA2 using the 1cytl antibody (1/3000). SERCA2b was ribed in (35). (B) Immunocytochemical

(a) SPCA2-overexpressing COS-1 cells b) hSPCA2-ov expressing COS-1 cells, ed with 10 µg odouble staininga double staininsed were anti-h (1/300) and aattern of the sa

hSPCA2 in hust lane: 30 µg cSPCA2-overex

20

er

Bio

f the antigenic peptides to for hSPCA2 (c) and the g for hSPCA2 (e) and the SPCA2 XIB pre-immune nti-SERCA2 (1/300). All me field to visualize the

man tissues. olon microsomes; second pressing COS-1 cells as a

positive control. The signal was abolished when the primary antiserum was pre-incubated with 10 µg of the immunogenic peptides (peptide inhibition). Figure 6 Localization of hSPCA2 in human colon tissue sections. (A, B) Double labelling with anti-hSPCA2 (XIB, 1/400) (A) and anti-TGN46 (1/100) (B). (C, D) Distribution of the hSPCA2 protein (XIB, 1/400) (C) and the ER marker SERCA2 (1/100) (D). (E,F) Double-labelling for the hSPCA1 (hSPCA1cytl, 1/400) (E) protein and TGN46 (1/100) (F). All pictures were overlayed with the DAPI-staining pattern to indicate the nuclei. The scale bar represents 20 µm. Figure 7 Formation of the phospho-intermediate. Microsomes of control cells (A, top panel) and hSPCA2-overexpressing cells (A, bottom panel) were phosphorylated in vitro using [γ-32P]-ATP in the presence of 5 µM Ca2+ or 5 µM Mn2+. The effects of 100 nM thapsigargin (TG) (A, top and bottom panel) and 200 mM hydroxylamine (B) on the formation of the phospho-intermediate was also shown. Figure 8 Ion transport by hSPCA2 overexpressed in COS-1 cells. (A) Time-dependent 45Ca2+ loading of permeabilized COS-1 cells transfected with hSPCA2 or with empty vector. Cells were loaded for the indicated periods at 0.2 µM free Ca2+ in the presence of 0.1 µM thapsigargin and 5 mM potassium oxalate. Results represent the mean ± S.E. of four experiments. (B) Ca2+-dependence of 45Ca2+ uptake into COS-1 cells transfected with hSPCA2 or empty vector. hSPCA2-overexpressing and control cells were loaded for 90 min at different free Ca2+ concentrations and the Ca2+ content of the stores was plotted as a function of the free Ca2+ concentration in the loading medium. Data points correspond to the difference between cells transfected with hSPCA2 and control cells transfected with empty vector. The full line is the best fit to the Michaelis-Menten equation. The calculated K0.5 was 0.27 µM with an R2 value of 0.988. The data points are the mean ± S.E. of five experiments. Error bars depict S.E. when larger than the used symbols. (C) 54Mn2+ loading of permeabilized COS-1 cells transfected with hSPCA2 or empty vector as a control. Cells were loaded for 90 min at 1 µM free Mn2+. After two washes with efflux medium, passive Mn2+ efflux was measured for 10 min. The ionophore A23187 (10 µM) was added after 10 min. The graph is representative of four independent experiments.

14

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

FIGURES No. Exon / Intron (bp) / Exon No. 1 ---GAA GCC TTG/gtgagtcccc… 29785 …cttttcacag/ATT GAT GAA--- 2 Glu Ala Leu Ile Asp Glu 2 ---GCG TTT TGT/gtaagaattg… 6519 …ctccttccag/GTG GAC TTA--- 3 Ala Phe Cys Val Asp Leu 3 ---CTG GAT CAG/gtaggaccag… 3161 …tcctcttcag/TTT AAG AAC--- 4 Leu Asp Gln Phe Lys Asn 4 ---ATC GCC ACG/gtgagttccc… 2074 …tgctgtctag/GCA GTG CTT--- 5 Ile Ala Thr Ala Val Leu 5 ---TTC ATC CAG/gtgagtattt… 101 …ttcctgccag/GAG TAC AGG --- 6 Phe Ile Gln Glu Tyr Arg 6 ---TGT AAC TG/gtaagtcaga… 4719 …tcaaatgcag/C CTA AGA GAA --- 7 Cys Asn Cys Leu Arg Glu 7 ---CTC ACT GAG/gtgagtggtt… 6799 …gctctcctag/GTC ACG GAC--- 8 Leu Thr Glu Val Thr Asp 8 ---AGG GGC CAG/gtaagccctg… 90 …ctggacacag/GGG GTC GTG--- 9 Arg Gly Gln Gly Val Val 9 ---GCT GAA GAG/gtaaggggca… 491 …accctttcag/ACA CCT AAA--- 10 Ala Glu Glu Thr Phe Lys 10 ---GGC ATA ATC G/gtgagtgaag… 2472 …tcttccccag/GT CTC ATC--- 11 Gly Ile Ile Gly Leu Ile 11 ---GGG GTC AG/gtaagagtgc… 13364 …ttcgatccag/C CTG GCT GTG--- 12 Gly Val Ser Leu Ala Val 12 ---GAG ACT TTA G/gtgagggact… 137 …gttgtaccag/GT TGC TGC--- 13 Glu Thr Leu Gly Cys Cys 13 ---CGT GCC GAG/gtgagtgcca… 1333 …cgccctgcag/GTC AGC GGA--- 14 Arg Ala Glu Val Ser Gly 14 ---TTA GTG GAG/gtaggtgtca… 1552 …ccccttttag/GCG GGC TGT--- 15 Leu Val Glu Ala Gly Cys 15 ---GCG ATG AAG/gtaggaggtc… 3730 …attatgctag/ATG GAC TTA--- 16 Ala Met Lys Met Asp Leu 16 ---AAG ACT GAG/gtgagacctt… 2103 …cttttctaag/GAT CAG GAA--- 17 Lys Thr Glu Asp Gln Glu 17 ---GGT TTG CGG G/gtcagtgcct… 3232 …ctctcctcag/TG CTG GCC--- 18 Gly Leu Arg Val Leu Ala 18 ---TTG GCC ATA G/gtaactgggt… 1047 tcttctgaag/GA AGA AAC--- 19 Leu Ala Ile Gly Arg Asn 19 ---GTG GGG AAG/gtgggtcccc… 86 …ccgtgtccag/GTG TCC GTG--- 20 Val Gly Lys Val Ser Val 20 ---ATC ATC AAG/gttcgctggg… 1447 …ctttccaaag/GCT CTG CAG--- 21 Ile Ile Lys Ala Leu Gln 21 ---GCC ATC AT/gtaagctgcc… 4136 …ttctcttgcag/G AAT GCA GTG--- 22 Ala Ile Met Asn Ala Val 22 ---CTG AGC AC/gtaagtagag… 199 …tgccccgcag/G AGC ATC TCC--- 23 Leu Ser Thr Ser Ile Ser 23 ---GCG CAG AG/gtgaggcagg… 1158 …ttcggagcag/C TTG GGG GTA--- 24 Ala Gln Ser Leu Gly Val 24 ---TGG AAG GAG/gtgagcgagg… 913 …caccttccag/ATG CCT GAA--- 25 Trp Lys Glu Met Pro Glu 25 ---CGC TCT CAG/gtgagacccg… 176 …ctccctgcag/ACC AAG CTG--- 26 Arg Ser Gln Thr Lys Leu 26 ---GGA GCG CTT G/gtgagtggtg… 1485 …cccttggcag/AT TTG CTG--- 27 Gly Ala Leu Asp Leu Leu 27 ---attaaccat/gtctaactac Figure 1 Intron-exon layout of ATP2C2.

15

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

Figure 2 mRNA expression pattern of ATP2C2 and ATP2C1 in human tissues.

16

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

Figure 3 Ratio RT-PCR of ATP2C1/ATP2C2-transcripts in different human tissues and cell lines.

17

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

A Control hSPCA2 CA2 1 2 3 4

anti- hSPCA2

anti- hSPCA1 anti- SERCA2b

f e

d c

b a

B

n . igure 4 Overexpress of hSPCA2 in COS-1 ceigure 4 Overexpress n of hSPCA2 in COS-1 ce s.

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

ACFF

E

18

BDsllll

F

D

Bioio

B

anti- hSPCA2

peptide

inhibition

colon lung + ← ← Figure 5 Western blot of endogenously expressed hSPCA2 in human tissues.

19

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

B A

F

D C

E

Figure 6 Localization of hSPCA2 in human colon tissue sections.

20

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

EGTA

C

a2+

Ca2+

+TG

Mn2+

+TG

Mn2+

Control

hSPCA2

Ca2+ Mn2+ Ca2+ Mn2+

Control hSPCA2 B

A Hydroxylamine Figure 7 Formation of the phospho-intermediate.

21

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

Figure 8 Ion transport by hSPCA2 overexpressed in COS-1 cells.

22

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from

Missiaen, Luc Raeymaekers and Frank WuytackJo Vanoevelen, Leonard Dode, Kurt Van Baelen, Rebecca J. Fairclough, Ludwig

+ ions2high affinity for Ca+ATPase(SPCA)2 is a golgi-localized pump with2+/Mn2The secretory pathway Ca

published online April 14, 2005J. Biol. Chem. 

  10.1074/jbc.M501026200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

by guest on January 6, 2020http://w

ww

.jbc.org/D

ownloaded from