| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Institute of Biochemistry, Swiss Federal Institute of Technology (ETH), 8092 Zurich, Switzerland;
*
Novartis Pharma AG, 4002 Basle, Switzerland; and
Department of Biochemistry and Venetian Institute of Molecular Medicine, University of Padova, 35121 Padova, Italy
1Correspondence: Department of Biochemistry, University of Padova, Viale G. Colombo, 3, 35121 Padova, Italy. E-mail: carafoli{at}civ.bio.unipd.it
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: plasma membrane Ca2+ pump • chimeric proteins • ER/SR Ca2+ pump
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
is a P-type ion motive ATPase. Its primary structure (4)
is similar to that of other P-type ATPases, particularly that of the sarcoplasmic/endoplasmic reticulum (ER/SR) Ca2+ pump (SERCA) (5)
. The most distinctive feature of the PMCA pump is the long (
150 amino acids) intracellular tail that follows the last transmembrane domain. It contains the calmodulin binding domain and some other regulatory sites. The calmodulin binding domain is sandwiched between two acidic sequences that bind 3 Ca2+ ions with extremely high to high affinity (6)
.
The SERCA and PMCA pumps are both predicted to contain 10 transmembrane domains and two large cytosolic loops (located between transmembrane domains 2 and 3, and 4 and 5, respectively). This was confirmed for the SERCA pump by the solution of its tertiary structure (7)
. Most of the mass of the pumps (70%) protrudes into the cytosol, with only 10% in the extracellular space. Besides their similar membrane architecture, components of the catalytic cycle and some of the amino acids involved in the translocation of Ca2+are conserved. Despite these structural similarities, significant functional differences exist: the SERCA pump transports two Ca2+ ions per ATP molecule, the PMCA pump only one. La3+ enhances steady-state concentration of the phosphoenzyme intermediate of the PMCA pump, but not of that of the SERCA pump. These differences have been traced to the lack in transmembrane domain 5 of the PMCA pump (8)
of a conserved Glu residue, which is involved in Ca2+ translocation in the SERCA pump.
In mammalian tissues, the two Ca2+ pumps are housed in separate compartments: the PMCA in the plasma membrane, the SERCA in the endo(sarco)plasmic reticulum membranes (9)
. Since the extracellular concentration of Ca2+ (1–3 mM) is much higher than that in the ER lumen (100–600 µM) (10
, 11)
, the energy demand on the PMCA pump appears to be higher. Little is known about the mechanism regulating the distinctive distribution of the two pumps. The carboxyl-terminal tail of the PMCA pump contains sequence signals that could cause its retention and degradation (12)
in the ER They may play a role in pathology, since their ‘exposure’ requires proteolytic removal of the carboxyl-terminal tail of the pump (12)
, but are unlikely to be involved in normal targeting of the mature PMCA. Previous work had identified putative (ER) retention signals in the first 85 amino acids in the amino-terminal region of the rSERCA1 (rabbit ER/SR Ca2+-ATPase isoform1). We have further investigated this by preparing chimeric constructs of the SERCA with the hPMCA4CI (human plasma membrane Ca2+-ATPase isoform 4, splicing variant CI) (13)
. Chimeric proteins have been used repeatedly to identify functional domains of other P-type pumps—for instance, the ouabain binding domain of the Na+/K+-ATPase or the thapsigargin binding domain of the SERCA pump (14)
. They have also been used to study targeting of H+/K+ and Na+/K+ pumps to different domains of the plasma membrane (15)
and to dissect domains involved in cellular localization of the SERCA and the PMCA pumps (12
, 16)
.
The study has defined more precisely the role of the terminal portion of Ca2+ pumps in their final cellular location. New constructs spanning the region from the NH2 terminus to the first transmembrane domain have indicated that ER retention signals are present in the amino-terminal cytosolic region preceding the first transmembrane domain of the SERCA pump. The formation of large pump oligomers may play a critical role as well in the ER retention of the pump.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Vectors for the expression of chimeric constructs in Cos-7 cells
The cDNA for the rSERCA1 (17)
was cloned in pSG5 (16)
. The pSG5/hPMCA4CI construct has been described (18)
. Construction of the chimeras required the introduction of new restriction sites on rSERCA1 and hPMCA4CI cDNAs. The following primers were used:
SRmut2(XhoI) 5'ATA GAG CAG CTC GAG GAC CTC CTG;
SRmut3(XhoI) 5'CAG GAG GTC CTC GAG CTG CTC TAT;
PMmut2(XhoI) 5'CCC AAA AAG CTC GAG ACT TTC TTA;
PMmut3(XhoI) 5'TAA GAA AGT CTC GAG CTT TTT GGG;
SERCA 163-MunI 5'CTC ACC CCG GAC CAA TTG AAG CGA CAT C-3';
PMCA 524-MunI 5'GAT GCA CTG ACC CAA TTG AAT CTC CAC TAT C-3';
PMCA 398-MunI 5'GGA TCC CTC GAC CCA ATT GAA CCC ATC AGA-3';
PMCA 479-MunI 5'GCA CAG TAA TGC AAT TGA GGA AGC TCA TGG-3';
SERCA 70-BamHI 5'CAA GGA TCC AGC GCA ATG GAG GCC-3'.
Restriction sites are underlined in the oligonucleotide sequences.
Preparation of the chimeric constructs
To generate chimeras (PM)SR, (SR)PM, (sSR)PM, and (sPM)SR, the following strategy was used. A XhoI site was inserted by PCR [using oligonucleotides SRmut2(XhoI), SRmut3(XhoI), PMmut2(XhoI), and PMmut3(XhoI)] in the region corresponding to the beginning of the first TM of the SERCA pump and the corresponding position in the PMCA pump. Vectors containing chimeras C and D (16)
were used. Small fragments could therefore be obtained encompassing the hydrophilic amino-terminal region (between the EcoRI and the XhoI) and the first transmembrane domain (XhoI and NheI) of each pump and ligated together in the desired combination. The final products were transferred to the pSG5 vector.
For chimeras AD, BC, EH, and GF, a MunI restriction site was inserted in pUCBM20 vectors encompassing the amino-terminal region of each pump (subcloned between an EcoRI and XhoI) using the oligonucleotides PMCA 524-MunI and SERCA 163-MunI, respectively. The constructs were digested with EcoRI-MunI, generating a 141 bp (A) and a 2909 bp fragment (containing fragment B and the pUCBM20 vector), and a 227 bp (C) and a 2994 bp fragment (vector+D), respectively. Digestion of the same chimeras with MunI-XhoI generated a 141 bp (E) and 2909 bp fragment (vector+F), and an 87 bp (G) and 3134 bp fragment (vector+H). These fragments (A, C, E, G) and vectors (B, D, F, H) were ligated to form the different pUC AD, BC, EH, and GF constructs. These pUC-based plasmids were digested with EcoRI-EcoRV. The resulting small fragments (AD 490 bp, EH 715 bp, BC 630 bp, GF 490 bp) were ligated with the 3160 bp fragment (EcoRV-XbaI chimera C) in the case of AD and EH and with the 3030 bp fragment (EcoRV-XbaI chimera E) for BC and GF in the pSG5 vector digested with EcoRI-XbaI.
For chimeras EH1 and EH28, pSG5PMCA4CI was digested with EcoRI-SacI and the 730 bp fragment was introduced into the EcoRI-SacI of pUCBM20. A MunI restriction site was inserted by site-directed mutagenesis at position 398 bp (oligo PMCA398-MunI) or position 479 bp (oligo PMCA479-MunI). These mutated constructs were digested with MunI-SacI generating fragments of 714 bp (mutated at position 398) and 633 bp (mutated at position 479). A 90 bp fragment was amplified from pSG5EH by PCR with the SERCA70-BamHI primer to introduce a BamHI site. After digestion with BamHI-MunI, the 90 bp fragment was ligated together with the 714 bp fragment into the BamHI-SacI cut pSG5PMCA4, resulting in chimera EH1. To generate chimera EH28, the 90 bp fragment was ligated with the 633 bp fragment into the BamHI-SacI-digested pSG5PMCA4.
Cell culturing and transfections
Dulbecco’s modification of Eagle’s medium (DMEM), minimal Eagle’s medium (MEM), fetal calf serum (FCS), antibiotics, and other cell culture media supplements were purchased from Life Technologies, Inc. (Basle, Switzerland). The TNM-FH medium was from Sigma (St. Louis, MO).
Cos-7 cells (kidney, SV40-transformed African green monkey) were maintained in DMEM with 5% FCS and 100 µg/ml gentamicin (complete DMEM) in a 5% CO2, incubator at 37°C. Spodoptera frugiperda (Sf9) cells were grown in TNM-FH supplemented with 10% FCS and 100 µg/ml gentamicin at 29 ± 1°C. All routine procedures involving Sf9 cells were performed according to Summers and Smith (19)
. Recombinant baculoviruses were prepared according to the GIBCO protocol (Gibco BRL, Grand Island, NY) (20)
. Transfer vectors needed to generate the recombinant viruses were prepared by cloning the chimeric constructs (generated in pSG5) in BamHI-KpnI sites of pFastBac (Gibco BRL). Transfection with a Maxi Kit (Qiagen AG, Basle Switzerland) -purified supercoiled DNA was performed using SuperFect (Qiagen) according to the supplier’s protocol.
Indirect immunofluorescence microscopy
Cos-7 cells were plated on coverslips, then transfected and analyzed by indirect immunofluorescence microscopy after 48–60 h. The cells were washed twice with PBS (150 mM NaCl, 20 mM NaH2PO4, pH 7.4, 0.1 mM CaCl2, 0.1 mM MgCl2), fixed for 20 min in 3% paraformaldehyde, washed four times with PBS, then incubated for 30 min in 0.1 M glycine. After four washes with PBS the cells were permeabilized in 0.1% Triton X-100 for 3 min, followed by four PBS washes, and incubated in blocking buffer (5% FCS, 0.1% BSA, 5% glycerol, 0.04% NaN3 in PBS) for 1 h.
The coverslips were overlaid for 90 min to overnight with the primary antibodies diluted in blocking buffer (1:50 for the rabbit antisera p94.2 and 4N, 1:50 for the monoclonal antibody 5F10, 10 µg/ml for the affinity-purified mouse antibody A52). After five washes with the blocking buffer, cells were treated with the appropriate secondary antibodies for 1 h (swine anti-rabbit or goat anti-mouse fluorescein-conjugated antibodies diluted 1:30 to 1:50 in blocking buffer, DAKO A/S, Glostrup, Denmark). Coverslips were washed five times in blocking buffer before mounting in a medium containing 80% glycerol, 2.5% DABCO (2,4-diazabicyclo-[2,2,2]-octane) in PBS, pH 8.0. The cells were observed in a microscope (Carl Zeiss, Oberkochen, Germany). Controls with primary and secondary antibodies alone were performed. Only background staining was noted and no differences were observed with cells transfected with vector alone or vectors encoding the pump constructs.
Preparation of membranes from Cos-7 and Sf9 cells
Forty eight to 60 h after transfection, Cos-7 cells were harvested in 50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 1 mM dithiothreitol (DTT), 10 mM EDTA, 75 µg/ml PMSF, and 100 U/ml Trasylol. The cells were disrupted by three cycles of freeze and thaw at -70°C/37°C and the insoluble proteins were sedimented at 15,000 g for 30 min (4°C). The supernatant was discarded and the pellet resuspended in 4 mM Tris-HCl, pH 7.5, and 10% sucrose for storage at -70°C or solubilized in 10 mM Tris-HCl, pH 8.0, 1 mM EDTA, 0.5% SDS for immunoblotting or immunoprecipitation.
Sf9 cells were collected 2 days after infection with the recombinant baculovirus, washed three times in 25 mM Tris-HCl, pH 7.5, 150 mM NaCl, and homogenized in 10 mM Tris-HCl, pH 7.5 (40 strokes of a Dounce homogenizer, on ice) in the presence of 75 µg/ml PMSF, 100 U/ml Trasylol, and 1 mM DTT. After addition of sucrose and KCl to a final concentration of 10% and 150 mM, respectively, the nuclei and large debris were sedimented for 10 min at 750 g; 10 mM EDTA was added to the postnuclear supernatant, which was sedimented at 100,000 g for 45 min. The high-speed pellet was resuspended in 4 mM Tris-HCl, pH 7.5, and 10% sucrose and stored at -70°C.
SDS-polyacrylamide gel electrophoresis (PAGE), Western blotting
Proteins were separated by SDS-polyacrylamide gels essentially according to Laemmli (21)
. The samples were boiled for 5 min after the addition of one volume of 0.5 M DTT, 5% SDS, 6 M urea, 50 mM Tris-HCl, pH 8.0, and 1.5 mM EDTA.
PAGE under acidic conditions was performed according to Sarkadi et al. (22)
. Samples were dissolved at room temperature in 65 mM Tris-phosphate, pH 6.8, 0.5 M DTT, 5% SDS, 6 M urea, 5 mM EDTA. The electrophoresis buffer contained 170 mM MOPS, adjusted to pH 6.0 with Tris-base, and 0.1% SDS.
Proteins separated by SDS-PAGE were first transferred to nitrocellulose (23)
. The nitrocellulose sheets were blocked overnight at room temperature in 2% gelatin (Fluka, St. Louis, MO) in TBS (10 mM Tris-HCl, pH 8.0, 150 mM NaCl). Primary antibodies were applied for 1–2 h at room temperature. The monoclonal antibody A52 (17)
was diluted in TBST (TBS+0.05% Tween 20) at a concentration of 4 µg/ml. The polyclonal antibody p94.2 (raised in rabbit against the COOH-terminal part of the hPMCA4CI) (24)
was used at a 1/500 dilution in TBST and the polyclonal antibody 4N (25)
at a 1/1000–1/2000 dilution in the same buffer. The monoclonal antibody 5F10 (26)
was used at 1/3000 dilution. After washing, the blots were incubated with alkaline phosphatase-coupled secondary antibodies diluted 1/7000 in TBST (Promega, Madison, WI) for 1 h at room temperature. The blots were developed with the color substrates BCIP and NBT (ProtoBlot system, Promega).
Metabolic labeling and pulse and chase
Transfected Cos-7 cells were rinsed with methionine-free MEM (Gibco) and incubated in the same medium for 20 min at 37°C, 5% CO2. Cells were labeled for 2 h at 37°C, 5% CO2 in 150 µl (35 mm plates) of methionine-free MEM containing 150 µCi/ml[35S]-methionine (1000 Ci/mmol, Amersham Pharmacia Biotech UK Ltd., Buckinghamshire, England). After labeling, a portion of the cells was taken as time 0 (pulse sample). The remainder was chased for 2 h in the presence of excess (10,000x) of cold Met. Samples were analyzed by PAGE and autoradiography using a PhosphorImager device (Molecular Dynamics, Sunnyvale, CA).
Cell surface biotinylation
Forty eight hours after transfection, Cos-7 cells were metabolically labeled for 2 h (see above). They were rinsed twice with TBS and once with borate buffer, pH 9.0 (10 mM Na-borate, 154 mM NaCl, 12 mM KCl, 2.25 mM CaCl2). Biotinylation with NHS-SS biotin (sulfosuccinimidyl-2-(biotinamido) ethyl-1,3-dithiopropionate, Pierce, Rockford, IL) at a concentration of 0.5 mg/ml in borate buffer was carried out twice for 15 min (1.5 ml/100 mm plate) (27
, 28)
. After two washes with borate buffer, the free biotin reagent was blocked with 50 mM NH4Cl/PBS for 10 min. The cells were rinsed twice again with borate buffer and solubilized for 1 h with 1.5 ml lysis buffer per plate (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, 0.2% BSA, 75 µg/ml PMSF, 5 µg/ml pepstatin, 5 µg/ml leupeptin). The cell lysate was centrifuged for 30 min at 15,000 g and the supernatant was rocked overnight with 1–5 µg of the affinity-purified monoclonal antibody A52 or 1 µl of the 5F10 monoclonal antibody; 50 µl protein A-Sepharose CL-4B was added to the mixture and the incubation was continued for an additional 8 h before washing the beads six times with lysis buffer. To recover the immunoprecipitated biotinylated antigens, the beads were boiled twice for 3 min with 20 µl 10% SDS, diluted with lysis buffer (1000 µl/tube), and centrifuged for 2 min at 15,000 g. An aliquot of the supernatant (1/10 of the recovered material for the experiment with Cos-7 cells) was taken as the immunoprecipitated, non-avidin-precipitated material. The remainder of the recovered material was incubated overnight with 50 µl of avidin-agarose beads (50% aqueous slurry, 1–2 mg avidin per ml of gel; Pierce). The beads were washed as described above and boiled in SDS-PAGE sample buffer. The samples were analyzed by gel electrophoresis. The dry gels were analyzed and quantified using a PhosphorImager device.
Phosphoenzyme intermediate formation from ATP
Formation of the phosphoenzyme intermediate from ATP was performed with membranes obtained from Cos-7 cells (transient transfection) or Sf9 cells infected with the recombinant viruses; 25–50 µg of membrane proteins was resuspended in 50 µl 20 mM MOPS-KOH, pH 6.8, 100 mM KCl in the presence of 10 µM of CaCl2 and LaCl3. The reaction, carried out on ice for 30 s, was started by the addition of 0.3 µM 
-32P-ATP (300 Ci/mmol) and stopped by adding 6% trichloroacetic acid/1 mM phosphate. The samples were kept on ice for 15 min and then spun down at 15,000 g for 20 min at 4°C. The pellets were washed with 6% trichloroacetic acid/1 mM phosphate and water. They were resuspended in sample buffer and separated on acidic SDS-PAGE. After drying, the gels were analyzed by autoradiography.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In (PM)SR, the first transmembrane domain of the rSERCA1 pump and the region surrounding it were replaced by those of the hPMCA4CI pump. In (sPM)SR, the first 57 amino acids of the rSERCA1 pump were replaced by the first 90 amino acids of the hPMCA4CI pump. In chimeras BC and chimera GF, a portion of the NH2-terminal amino acids of the rSERCA1 was replaced by the corresponding amino acids of the hPMCA4CI.
|
In the (SR)PM chimera, the first transmembrane domain of the hPMCA4CI pump was exchanged for that of the rSERCA1 pump. In chimera (sSR)PM, the first 90 amino acids of the hPMCA4CI were replaced by the first 57 amino acids of the rSERCA1 pump. In chimeras AD and EH, portions of the NH2 domain of the hPMCA4CI pump upstream of transmembrane domain 1 were replaced by amino acids of the rSERCA. Chimera EH1 was constructed by fusing in-frame the first 28 amino acids of the rSERCA1 pump to the NH2 terminus of the hPMCA4CI pump. The amino-terminal Met of the hPMCA4CI had to be replaced by a Glu. In chimera EH28, the first 28 amino acids of the hPMCA4CI pump were exchanged for amino acids 1 to 28 of the rSERCA1 pump.
Expression of hPMCA4CI, rSERCA1, and chimeric proteins in Cos-7 and Sf9 insect cells
The wild-type and mutated proteins were transiently expressed in Cos-7 cells and detected with the help of 3 antibodies. Monoclonal antibody A52 recognizes the catalytic region of the rSERCA1 pump; polyclonal antibody 94.2 and monoclonal antibody 5F10 recognize two different epitopes in the hPMCA4CI pump (see Fig. 1
for details).
Chimeras AD, (sSR)PM, and EH had apparent masses (Fig. 2
A, lanes 2, 4, 5) similar to that of the wild-type pump (between 130 and 133 kDa, Fig. 2A
, lane 1). Thus, Western blots with antibody 5F10 revealed bands of similar size (Fig. 2A
, lanes 2, 4, 5). The 5F10 antibody also reacted, though weakly, with the endogenous PMCA pump (Fig. 2A
, lane 3). The molecular mass of the endogenous pump was slightly larger than that of the chimeras and the expressed wild-type pump. The lower bands are likely to be degradation products of the recombinant proteins (Fig. 2A
, lower panel).
|
Similar results were obtained with chimeras EH1, EH28, and(SR)PM, although the expression level of the latter was lower than that of the other chimeras (not shown). Chimeras GF, BC, and (sPM)SR) were all detected by monoclonal antibody A52 as bands of approximately equivalent intensity and molecular mass of 110 kDa (Fig. 2B
, lanes 7–9), whereas the (PM)SR chimera had a much lower level of expression (not shown). The A52 antibody failed to recognize the endogenous SERCA protein (17)
. PMCA4CI-based chimeras were expressed in amounts equivalent to that of the wild-type pump, but SERCA-based chimeras were expressed at a lower level.
Since it was difficult to measure the activity of the PMCA-based constructs in Cos cells, chimeras EH, AD, EH1, EH28, and (sSR)PM were expressed in Sf9 insect cells using the baculovirus expression system (20)
. Some chimeras were expressed at a high level, and Coomassie brilliant blue staining (Fig. 2C
, upper panel, arrow) could visualize their protein bands in gels. Even proteins not revealed by the staining were expressed at high levels (Fig. 2C
, lower panel). A comparison of the expression in Cos and Sf9 cells showed that even chimeras not visualized by Coomassie blue brilliant staining (EH and (sSR)PM) were expressed at a higher level in Sf9 cells than in Cos cells (not shown). Chimeras AD, EH1, and EH28 were expressed in amounts similar to those of the hPMCA4CI pump whereas chimeras EH and (sSR)PM were expressed at lower levels, suggesting that the 43 NH2-terminal amino acids of the hPMCA4CI pump may influence the protein’s stability in Sf9 cells.
Subcellular localization of the chimeras in Cos-7 cells
Subcellular location of the chimeras was investigated in transfected Cos-7 cells by immunofluorescence staining. The efficiency of the transfections was 10–20% (Fig. 3
A). Cos-7 cells expressing the wild-type hPMCA4CI pump showed staining of the plasma membrane (Fig. 3A
, bottom) (12
, 16)
that was diffused over the surface of the cell, making its borders clearly visible (29
, 30)
. Cells expressing the wild-type rSERCA1 pump (Fig. 3A
) showed instead the reticular pattern typical of proteins retained in the ER. The border of the cells was not visible. The strong staining around the nucleus was probably due to the high amount of the recombinant protein expressed.
|
Cells expressing the SERCA-based chimeras (PM)SR, (sPM)SR, BC, and GF showed staining identical to that of the wild-type pump (Fig. 3B
), indicating that the recombinant proteins had been retained in the ER. In a few cells expressing chimera GF (10%), overlapping PM staining was seen (not shown).
Cells expressing the PMCA-derived chimeras AD, EH, and (sSR)PM yielded strong staining of the ER (Fig. 3C
). A few cells (10%) expressing chimera AD instead showed the plasma membrane staining pattern. Even though the number of AD transfected cells showing plasma membrane staining pattern seemed to be statistically significant, the chimeras were considered as retained in the ER.
The results with chimera EH indicated that the first 28 amino acids of the rSERCA1 pump might cause retention in the ER. Immunocytochemistry with cells expressing chimeras EH1 and EH28 revealed that 60–80% of the cells (the percentage varied among transfection experiments) showed staining of the plasma membrane; the remaining 20–40% instead showed ER staining (Fig. 3D
). The number of ER stained cells was significantly higher than that observed for the PMCA. Cells showing strong signals (those expressing a high level of recombinant protein) showed ER staining.
Formation of the phosphoenzyme intermediate from ATP of the hPMCA4CI, and their chimeras
All the recombinant proteins expressed in Cos-7 cells were tested for their ability to form the phosphoenzyme intermediate. Though it was possible to detect a strong signal for the wild-type SERCA pump, all of the SERCA-based chimeras showed only weak signals (see Fig. 4
A for (sPM)SR). The results on the PMCA-based chimeras using Cos cells were less unambiguous, mainly because the low level of expression of the PMCA pump (and its chimeras) rendered the evaluation of phosphoenzyme formation in such experiments difficult. Therefore, recombinant baculoviruses were prepared for wild-type PMCA4CI and for chimeras AD, EH, EH1, EH28, and(sSR)PM. As shown in Fig. 2C
, high levels of expression were obtained in Sf9 cells. Upon incubation with radioactive ATP, a strong signal was obtained with membrane expressing PMCA4CI, and chimeras EH1 and EH28, whereas no signal was obtained with the other chimeras: Even if chimeras EH and (sSR)PM were expressed at lower level, the amount of recombinant proteins ought to have been sufficient to detect activity. Overexposure of the autoradiograms (Fig. 4B
) failed to reveal any radioactive bands at 130–140 kDa, which was the predicted mass for chimeras AD, EH, and (sPM)SR, in lanes 3, 4, and 7.
|
Tests of the delivery of the chimeras to the cell-surface using biotinylation
Water-soluble sulfo-NHS biotin (sulfonsuccinimidyl-2-(biotinamido)ethyl-1,3-dithiopropionate) was used to biotinylate proteins expressed on the cell surface of Cos-7 cells (27
, 28)
. The NHS group reacts with 
-amino-groups of lysines in proteins exposed to the outside of cells. After solubilization, proteins were subjected to immunoprecipitation with specific antibodies and the biotinylated proteins were enriched with avidin beads. Although only a small proportion of the PMCA amino acids is exposed on the extracellular surface, it has been possible to detect PMCA molecules expressed on the surface of cells (16)
. The percentage of biotinylated chimeras EH1 and EH28 suggested that they behaved like the wild-type PMCA pump (Fig. 5
); a similar portion of the protein was exposed to the extracellular surface. This was consistent with the experiments presented in Fig. 3D
. The percentage of exposed biotinylated chimeras EH and (sSR)PM was instead similar to that of the wild-type SERCA pump, indicating their retention in the ER. Chimera AD showed a higher extent of biotinylation than chimeras EH and (sSR)PM, most likely due to the 10% of cells showing plasma membrane staining in the immunocytochemistry experiments (see above). Since in the immunocytochemistry experiments with chimera GF a portion of the cells reproducibly showed plasma membrane staining (see above), surface labeling was performed with it. Parallel experiments with the SERCA pump and chimera GF (Fig. 5C
) showed a small but reproducible higher surface labeling of chimera GF.
|
Cellular fractionation experiments to investigate whether a portion of chimera EH28 is retained in the ER
Cell membrane proteins transfected with chimera EH28 and the wild-type SERCA and PMCA4CI pumps were separated on a 30%–60% sucrose gradient (Fig. 6
). Under these conditions, the PMCA4CI pump became enriched in fractions 4 to 6 and the SERCA pump in fractions 5 to 8. Chimera EH28 was evenly distributed over fractions 3 to 8, suggesting that a portion of it migrated as the PMCA4CI pump and the remainder as the SERCA pump. This confirmed the indication from immunocytochemistry (Fig. 3D
) and biotinylation (Fig. 5)
experiments that a portion of this chimera was retained in the ER.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Previous experiments on chimeric PMCA proteins in which the first transmembrane domain of the wild-type pump was replaced by that of the SERCA pump had suggested that this could also be the case for transmembrane domain 1 of the SERCA pump (16)
. The experiments described here suggest that additional domains mediate retention of the SERCA pump in the ER. Cos-7 cells expressing chimeras (SR)PM, (sSR)PM, EH, and AD had a staining pattern consistent with that of proteins resident in the ER, which was confirmed by surface labeling experiments. Although this was expected for chimera (SR)PM, it was not for chimeras (sSR)PM, EH, and AD, which only contained portions of the 57-residue amino-terminal cytosolic portion of the SERCA pump. These experiments suggested that in addition to the first transmembrane domain, amino-terminal portions of the SERCA pump were involved in its retention in the ER. The decision to focus on the first 28 amino acids was dictated by the observation that whereas chimera EH (in which the first 28 amino acids of the SERCA pump replaced the first 43 amino acids of the PMCA pump) was always retained in the ER, in some cells chimera AD (which lacked these 28 amino acids) was expressed on the plasma membrane. In chimera EH1, 28 NH2-terminal amino acids of the SERCA pump were added to the NH2 terminus of the PMCA pump; in chimera EH28, the first 28 amino acids of the PMCA pump were replaced by those of the SERCA pump. In both cases ER retention was observed in < 50% of the expressing cells. The variability of the results and the lack of recognizable motifs indicated that no ‘classical’ signal, such as the carboxyl-terminal tetrapeptide KDEL (32)
, the carboxyl-terminal (K(X)KXX) sequence of membrane proteins type I (33)
, or the two arginines at the NH2 terminus of type II membrane proteins (34)
, was active in the first 28 residues of the SERCA pump.
Some of the chimeric constructs were active whereas others had no activity. Although chimera EH28 was active, chimeras AD and EH were not. These results are consistent with previous observations (35)
indicating that the removal of amino acids 18–75 inactivated the plasma membrane Ca2+ pump. In chimeras EH and AD, a fraction of the amino acids of the PMCA pump was replaced by structurally related amino acids, yet the loss of activity was not prevented, suggesting that (some of) these amino acids are not exchangeable. The same apparently was true for the SERCA-based chimeras BC and GF. The finding that they were inactive was unexpected, since extensive amino acid mutations in this region did not to cause loss of activity of the SERCA (36)
. The crystal structure of the SERCA pump has shown that the amino-terminal domain is in close proximity to the first cytosolic loop linking the second and third transmembrane domains and may be an integral part of the actuator domain (7)
. This domain is thought to couple conformational changes in the nucleotide binding center to the Ca2+ translocation unit located within the membrane. It is therefore quite possible that the interaction (at the level of secondary and tertiary structure) of the cytosolic amino-terminal domain of the Ca2+ pump plays a role in its activity and ER retention.
Lack of activity in chimera EH but not in EH28 or EH1 suggests that the region between amino acids 29 and 44 of the PMCA pump is essential for activity and cannot be replaced. Previous results on the PMCA pump obtained with proteases (37)
had emphasized the amino-terminal portion of the pump encompassing the first transmembrane domains, suggesting that this domain may be essential for activity. The difference with the results presented here could be related to the incomplete removal of the proteolytic peptide. More likely, however, the amino-terminal domain could be required for the proper functional assembly of the pump, but would have no function in the mature protein.
Analysis of the amino-terminal portion of the SERCA pump isoforms has revealed some conserved amino acids. The sequence MExAH in the first 10 amino acids and sequences FxV and GLxxxQ in the 11–28 stretch are conserved in all 3 isoforms. Hydrophobic amino acids appear to be preferentially conserved, suggesting that retention in the ER and prevention of activity loss could be related to a secondary structure signal.
The SERCA pump molecules were tightly packed in the crystals used to solve its tertiary structure. The external region of domain A (containing the NH2 terminus of the pump) leaned toward the N domain of the neighboring pump molecule. If this arrangement is not an artifact of crystallization, it suggests that the pump molecules have a tendency to interact with each other, forming large oligomers. The formation of oligomers has been suggested as a mechanism by which membrane proteins are prevented from entering (due to place restrictions) the small transporting vesicles that circulate between the ER, the Golgi complex, and the PM. This would cause their accumulation in the ER. If amino acids in the amino-terminal domain of the SERCA pump, including amino acids 1–28, would be involved in the formation of oligomers, the results with chimeras EH1 and EH28 could be partially explained. The immunocytochemistry observations were performed on transiently transfected cells in which the levels of expression may have varied, thus influencing the efficiency of the oligomerization of the chimeric constructs. This would also be consistent with the observation that 10% of the cells expressing chimeras AD and GF exhibited a plasma membrane staining pattern; amino acids 1–28 of the SERCA pump are absent in both chimeras.
Received for publication May 8, 2001.
Revision received October 2, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Okkenhaug, K.-H. Weylandt, D. Carmena, D. J. Wells, C. F. Higgins, and A. Sardini The human ClC-4 protein, a member of the CLC chloride channel/transporter family, is localized to the endoplasmic reticulum by its N-terminus FASEB J, November 1, 2006; 20(13): 2390 - 2392. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Ebert, G. L. Hume, K. S. Warren, N. P. Cook, C. G. Burns, M. A. Mohideen, G. Siegal, D. Yelon, M. C. Fishman, and D. M. Garrity Calcium extrusion is critical for cardiac morphogenesis and rhythm in embryonic zebrafish hearts PNAS, December 6, 2005; 102(49): 17705 - 17710. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. Reiner, M. L. Hetsko, J. G. Meszaros, C.-H. Sun, H. G. Morrison, L. L. Brunton, and F. D. Gillin Calcium Signaling in Excystation of the Early Diverging Eukaryote, Giardia lamblia J. Biol. Chem., January 17, 2003; 278(4): 2533 - 2540. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
