Deletion of amino acid residues 18-75 inactivates the plasma membrane Ca2+ pump.

A mutant of the plasma membrane Ca2+ pump hPMCA4b(d18-75)(ct120) containing a deletion of the N-terminal amino acid residues 18-75 and lacking the C-terminal 120 amino acid residues was expressed in COS-1 cells. The deletion in the N-terminal region did not significantly affect the level of expression of the Ca2+ pump. Tryptic digestion of the hPMCA4b(d18-75)(ct120) mutant resulted in the appearance of the same fragments obtained by proteolysis of the hPMCA4b(ct120) enzyme, suggesting that deletion of residues 18-75 neither impeded the insertion in the membrane nor extensively affected the folding of the mutant protein. The functional competence of the hPMCA4b(d18-75)(ct120) enzyme was examined by measuring the Ca2+ transport and the Ca2+ ATPase activity of COS-1 cell microsomes expressing the mutant protein. Both tests proved the mutant to be inactive. Under conditions in which hPMCA4b(ct120) becomes phosphorylated, hPMCA4b(d18-75)(ct120) was incapable of reacting with ATP and Ca2+ to form the phosphoenzyme. Taken together these results suggest that the segment of amino acids 18-75 is essential for the activity of the plasma membrane Ca2+ pump.

The Ca 2ϩ pump from plasma membranes (PMCA) 1 is a calmodulin regulated, P 2 -type ATPase (1) responsible for maintaining the intracellular Ca 2ϩ homeostasis in eukaryotic cells by actively extruding Ca 2ϩ to the extracellular space. Four mammalian genes coding for different PMCA isoforms have been identified, each of them producing different mRNAs by alternative splicing processes (2). The PMCA polypeptide would transverse the lipid bilayer about 10 times, and most of the pump, including two major loops and both terminal ends, would be exposed to the cytosol (3). The PMCA contains regions highly conserved in all P-type ATPases and other regions that are distinctive of the PMCA because they are found only in the PMCA isoforms. The regions of the PMCA encompassing the aspartate residue that forms the phosphorylated intermediate (4) and the putative ATP binding site (5) belong to the type of highly conserved regions. In contrast the C-terminal region after the transmembrane segment 10, which is involved in the regulation of the activity of the enzyme by calmodulin (6), and the N-terminal region upstream the first transmembrane segment belong to the regions that are not homologous to that of the other P-type ATPases.
Recently a PMCA mutant called hPMCA4b(ct120) with a deletion of the C-terminal 120 residues including the calmodulin-binding site was expressed in COS-1 cells and found to be fully active, and as expected its activity no longer regulated by calmodulin (7).
We have now investigated the functional relevance of the N-terminal region of the PMCA, which extends from the initial methionine to about the beginning of the first transmembrane segment. With this aim, we constructed a mutant called hPMCA4b(d18 -75)(ct120) by removing the nucleotide sequence coding for amino acids 18 -75 from the cDNA of hPMCA4b(ct120). The measurements of the activity of the hPMCA4b(d18 -75)(ct120) expressed in COS-1 cells indicate that the residues 18 -75 are essential for a functional Ca 2ϩ pump.

Construction of the hPMCA4b Mutant cDNAs and Expression in
COS-1 Cells-The construction of the cDNA of mutants hPMCA4b(ct120) and hPMCA4b(d18 -75) was described previously, and the latter was called Hinm1 (3,7). In addition to the deletion of residues 18 -75, the presence of a new unique restriction site for MluI produced the replacement of serine 17 by threonine in the hPMCA4b(d18 -75).
To obtain the hPMCA4b(d18 -75)(ct120) mutant, the SalI-DraIII fragment was removed from hPMCA4b(d18 -75) and cloned into h4PMCA(ct120). The wild-type and mutant cDNAs were cloned into the pMM 2 vector (7). For protein expression, COS-1 cells (9) were transfected by the DEAE-dextran-chloroquine method (8) and harvested after 48 h. The microsomal fraction was isolated as described previously (7). Protein concentration was estimated by means of the Bio-Rad protein assay, with bovine serum albumin as a standard.
Detection of Expressed Ca 2ϩ Pump Protein-SDS-electrophoresis and immunoblotting were carried out as described previously (10). Proteins were electrophoresed on a 7.5% acrylamide gel according to Laemmli (11) and subsequently transferred to Millipore Immobilon membranes. Nonspecific binding was blocked by incubating the membranes overnight at 4°C in a solution of 160 mM NaCl, 0.05% Tween 20, and 1% nonfat dry milk. The membranes were incubated at 37°C for 1 h with 5F10 monoclonal antibody from ascites fluid (dilution, 1:1000). For staining, biotinylated anti-mouse immunoglobulin G and avidin-horseradish peroxidase conjugate were used.
Ca 2ϩ Transport Assay-Ca 2ϩ uptake assays were performed as described previously (10). The reaction mixture contained 100 mM KCl, 50 mM Tris-HCl (pH 7.3 at 37°C), 5 mM NaN 3 , 400 nM thapsigargin, 20 mM sodium phosphate, 6 mM ATP, 95 M EGTA, and enough MgCl 2 and CaCl 2 to give a concentration of free Mg 2ϩ of 800 M and of free Ca 2ϩ of 1 M. The free concentrations of Mg 2ϩ and Ca 2ϩ were calculated using the program of Fabiato and Fabiato (12). Vesicles (5-10 g of protein) were preincubated at 37°C for 5 min, and the reaction was initiated by the addition of ATP. The reaction was terminated after 5 min by filtering the samples through a 0.45-m filter. The 45 Ca taken up by the vesicles was then determined by counting in a liquid scintillation coun-* This work was supported in part by the Consejo Nacional de Investigaciones Científicas y Tecnológicas, Fundación Antorchas, the National Institutes of Health Grant GM28835, and National Science Foundation Grant INT 93-02981. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
ter. Uptake activities were expressed per mg of COS-1 cell membrane protein. The activity of the expressed Ca 2ϩ pump was estimated after subtracting the activity of the endogenous enzyme from COS-1 cells transfected with the empty plasmid pMM 2 .
Ca 2ϩ -ATPase Assay-Ca 2ϩ ATPase was measured by monitoring the [ 32 P]P i liberated from [␥-32 P]ATP (13). The reaction was carried out at 37°C for 30 min. The reaction mixture contained 50 mM Tris-HCl (pH 7.4 at 37°C), 100 mM KCl, 0.1 mM MgCl 2 , 0.1 mM EGTA, 0.1 mM ATP, 0.1 mM CaCl 2 , 5 mM NaN 3 , 0.5 mM ouabain, 4 g/ml oligomycin, 400 nM thapsigargin, and 10 g of membranes. The Ca 2ϩ ATPase was taken as the difference of ATPase activities in the presence and the absence of Ca 2ϩ .
Detection of the Phosphorylated Intermediate-The phosphorylation reaction was carried out at 4°C in medium containing 10 g of microsomal protein, 100 mM KCl, 25 mM Tris-HCl (pH 7.4 at 4°C), 400 nM thapsigargin, 5 mM MgCl 2 , 0.02 M CaCl 2 , and 0.02 M LaCl 3 in a reaction volume of 0.25 ml. La 3ϩ is known to stabilize the phosphoenzyme of the plasma membrane Ca 2ϩ pump (14). The reaction was initiated by the addition of 1 M [␥-32 P]ATP and terminated after 30 s with 1 ml of stopping solution containing 10% trichloroacetic acid, 10 mM P i , and 1 mM cold ATP at 4°C. After adding 50 l of 1 mg/ml of bovine serum albumin, the denatured proteins were collected by centrifugation at 20,000 ϫ g for 10 min, washed once more with stopping solution, and washed once with distilled water. The precipitated samples were dissolved in sample buffer and separated by SDS-electrophoresis in a 7% acrylamide gel according to Sarkadi et al. (15). After drying the gel, autoradiographs were produced.
Tryptic Digestion of Microsomes-Digestion of COS-1 microsomes with trypsin was done on ice in a medium containing 10 g of microsomes, 0.1 mM EGTA, 50 mM Tris-HCl (pH 7.7 at 4°C), 0.25 mM sucrose, 0.15 mM KCl, 2 mM dithiothreitol. The reaction was initiated by the addition of trypsin to give a final concentration of 20 g/ml and terminated after 15 or 30 s with 10-fold excess of soybean trypsin inhibitor. Controls of undigested microsomes (0 s) were done by adding the inhibitor before the trypsin.

RESULTS
Expression of hPMCA4b(ct120) and hPMCA4b(d18 -75)(ct120) in COS-1 Cells-COS-1 cells were transfected with the vector without insert or with either the hPMCA4b(ct120) or the hPMCA4b(d18 -75)(ct120) DNAs. Different amounts of membranes from the transfected cells were submitted to SDS gel electrophoresis followed by immunoblotting using antibody 5F10, which recognizes an epitope located between amino acids 724 -783 in the central part of the Ca 2ϩ pump protein (3). The immunoblot is shown in Fig. 1. One band of about 140 kDa corresponding to the endogenous Ca 2ϩ pump from COS-1 cells was detected in membranes from cells transfected with the empty vector pMM 2 . An additional band of about 120 kDa was observed in membranes from hPMCA4b(ct120) transfected cells. Membranes from COS-1 cells transfected with hPMCA4b(d18 -75)(ct120) DNA showed a band at about 114 kDa, the expected molecular mass of the hPMCA4b(d18 -75)(ct120) protein. The intensity of the bands of the hPMCA4b(ct120) and hPMCA4b(d18 -75)(ct120) were close, indicating that both proteins were expressed at a similar level.
Partial Proteolysis of the hPMCA4b(d18 -75)(ct120) Protein-As a test for the topology of the hPMCA4b(d18 -75)(ct120) Ca 2ϩ pump, microsomes containing the expressed protein were exposed briefly to the action of trypsin. Fig. 2 shows an immunoblot of the proteolyzed membranes using antibody 5F10. In agreement with previous reports (7), the treatment of COS-1 membranes expressing hPMCA4b(ct120) with 20 g/ml of trypsin for 30 s resulted in the appearance of a major band of approximately 76 kDa.
Because the N-terminal part of the molecule containing amino acid residues 1-315 is removed early during proteolysis and is not recognized by 5F10 (3) providing the insertion and folding of the enzyme was the same, differences are not expected between the proteolytic pattern of the hPMCA4b(ct120) and hPMCA(d18 -75)(ct120). As shown in Fig. 2 this was indeed the case, suggesting that both the hPMCA4b(ct120) and the hPMCA(d18 -75)(ct120) proteins exposed the same sites to trypsin and hence produced similar proteolytic fragments containing the epitope for antibody 5F10.
Activity of the hPMCA4b(d18 -75)(ct120) Enzyme-Results in Table I show that the Ca 2ϩ uptake of microsomes from cells transfected with the hPMCA4b(ct120) DNA was about 12 times higher than that from control microsomes from cells transfected with the empty vector. The Ca 2ϩ transport activity of microsomes expressing the hPMCA4b(d18 -75)(ct120) protein was not significantly different from that of the control, indicating that the mutant enzyme was not able to transport Ca 2ϩ . As also shown in Table I, the Ca 2ϩ ATPase activity of microsomes from COS-1 cells transfected with the hPMCA4b(d18 -75)(ct120) DNA was similar to that of the control, indicating that the hPMCA4b(d18 -75)(ct120) was also not capable of hydrolyzing ATP in a Ca 2ϩ -dependent manner.
In control experiments (not shown), the effects of the deletion of amino acid residues 18 -75 on the activity of the full-length hPMCA4b was investigated. The hPMCA4b(d18 -75) enzyme was also inactive, indicating that the C-terminal regulatory end of the PMCA was not involved in the mechanism leading to inactivation.
Formation of the Phosphorylated Intermediate-The first step of the normal Ca 2ϩ pump cycle involves the reaction of the enzyme with Ca 2ϩ and ATP to form a phosphoenzyme. The ability of the hPMCA4b(d18 -75)(ct120) mutant to form a phosphorylated intermediate was investigated. Fig. 3 shows that two bands corresponding to the endogenous PMCA and sarco/ endoplasmic reticulum Ca 2ϩ pumps were observed when membranes from COS-1 cells transfected with the empty vector were phosphorylated with ATP in the presence of Ca 2ϩ plus La 3ϩ . As expected an additional band in the molecular mass region of 120 kDa was observed in membranes containing the hPMCA4b(ct120) enzyme. The phosphorylation pattern of membranes containing the hPMCA4b(d18 -75)(ct120) protein was similar to that of the control, indicating that no phosphorylation attributable to the hPMCA4b(d18 -75)(ct120) mutant had occurred. DISCUSSION In the present work we have investigated the functional consequences of a deletion in the N-terminal region of the hPMCA4b between the initial methionine and the first transmembrane domain. Mutant hPMCA4b(d18 -75)(ct120) was expressed in amounts comparable with those of the hPMCA4b-(ct120), indicating that the N-terminal deletion did not significantly affect the level of expression of the PMCA. The measurements of Ca 2ϩ transport and ATP hydrolysis indicated that the mutant protein was unable to perform any of these functions. In addition, no detectable amounts of phosphoenzyme were formed by the hPMCA4b(d18 -75)(ct120) enzyme. Thus the lack of Ca 2ϩ transport and ATPase activities of mutant hPMCA4b(d18 -75)(ct120) could be ascribed to its inability to react with Ca 2ϩ and ATP to form the phosphoenzyme.
Deletion of several amino acid residues could result in alterations of the folding or insertion in the membrane of the mutant protein. Although this possibility cannot be discarded, several observations suggest that this may not be the case of the hPMCA4b(d18 -75)(ct120) mutant: (i) the expressed hPMCA4b(d18 -75)(ct120) protein was always found associated with membranes, and after cell lysis it was recovered in the microsomal fraction, (ii) immunoblots of membranes from COS-1 cells transfected with the hPMCA4b(d18 -75)(ct120) DNA showed only one major product of about 114 kDa in amounts similar to those of the hPMCA4b(ct120), indicating that both proteins were equally stable to degradation by intracellular proteases, and (iii) digestion with trypsin and immunoblotting of the hPMCA4b(ct120) and the hPMCA4b(d18 -75)(ct120) revealed similar proteolytic fragments, suggesting that both proteins exposed the same limited number of sites to the protease.
Early analysis by SDS-polyacrylamide gel electrophoresis of the proteolytic fragments produced by treatment of red cell membranes with trypsin has shown that the protease acts on several specific sites of the PMCA resulting in major peptides of 90, 81, and 76 kDa (15)(16)(17). The appearance of the 76-kDa fragment in the SDS-polyacrylamide gel electrophoresis is concomitant with a highly active PMCA. Based on these results (15) concluded that the 76-kDa polypeptide was a fully competent Ca 2ϩ transporter and therefore that the N-and C-terminal amino acid residues missing in the 76-kDa fragment are not essential for activity. However, as mentioned in Ref. 16, the effective separation of the N-terminal fragment containing the transmembrane domains 1 and 2 from the 76-kDa product in the absence of SDS was not proved.
Recently, the expression of N-terminally and C-terminally truncated PMCA mutants allowed to directly assess the activity of the fragments produced by trypsinolysis. Although truncation of the hPMCA4b generating a C terminus similar to that of the 76 kDa fragment produces a highly active enzyme (7), a mutant starting with the N terminus of the 90-kDa polypeptide (PMCA105) is inactive (18). As was suggested previously (18), the lack of activity of the PMCA105 mutant is probably related to the absence of a large portion of the pump containing the first two transmembrane domains and the region highly conserved in all P-ATPases, which has been called the transducing domain. Results in this paper show that a smaller deletion between the N terminus and the first transmembrane domain, a region previously assumed irrelevant for the function of the enzyme, suffices for inactivation. Studies of mutants with smaller deletions and substitutions of amino acids in the Nterminal region will be needed to obtain a more precise knowledge of the residues involved and their role during Ca 2ϩ transport by the PMCA.  3. Autoradiographic detection of the phosphorylated intermediate. Phosphoenzyme formation was carried out as described under "Materials and Methods" in the presence of 1 M ATP, 0.02 mM CaCl 2 , and 0.02 mM LaCl 3 . 5 g of COS-1 membranes were loaded per well. In addition to the PMCA phosphoenzyme, a phosphorylated band of lower molecular mass corresponding to the endogenous sarco/endoplasmic reticulum Ca 2ϩ pump is observed.