Loss of Autoinhibition of the Plasma Membrane Ca 2 (cid:1) Pump by Substitution of Aspartic 170 by Asparagine ACTIVATION OF PLASMA MEMBRANE CALCIUM ATPase 4 WITHOUT DISRUPTION OF THE INTERACTION BETWEEN THE CATALYTIC CORE AND THE C-TERMINAL REGULATORY DOMAIN*

The plasma membrane calcium ATPase (PMCA) ac-tively transports Ca 2 (cid:1) from the cytosol to the extra cel-lular space. The C-terminal segment of the PMCA func-tions as an inhibitory domain by interacting with the catalytic core. Ca 2 (cid:1) -calmodulin binds to the C-terminal segment and stops inhibition. Here we showed that residue Asp 170 , in the putative “A” domain of human PMCA isoform 4xb, plays a critical role in autoinhibition. In the absence of calmodulin a PMCA containing a site-specific mutation of D170N had 80% of the maximum activity of the calmodulin-activated PMCA and a similar high affinity for Ca 2 (cid:1) . The mutation did not change the activation of the PMCA by ATP. Deletion of the C-terminal segment further downstream of the calmodulin-binding site led to an additional increase in the maximal activity of the mutant, which suggests that the mutation did not affect the inhibition because of this portion of the C-terminal segment. The calmodulin-activated PMCA was more sensitive to vanadate inhibition than the autoinhibited enzyme. In contrast, inhibition of the D170N mutant required higher concentrations of vanadate and was not affected by calmodulin.

PMCA1 and PMCA4 are the most widespread isoforms, whereas PMCA2 and PMCA3 are rather cell-specific. The PMCAs belong to the subtype 2B of P-type ATPases that are characterized by their autoregulation by internal sequences. At resting concentrations of Ca 2ϩ , autoinhibition keeps the activity of the PMCA low. When the concentration of Ca 2ϩ in the cytosol increases, Ca 2ϩ -calmodulin binds to the C-terminal portion of the molecule, switching the PMCA to an activated state of higher maximum activity and affinity for Ca 2ϩ (2). The removal by partial proteolysis or deletion mutagenesis of ϳ120 C-terminal residues of human (h)PMCA4xb, including the calmodulin-binding site, results in a constitutively active form insensible to calmodulin (3). More recently, it has been shown that a construct named ct92, truncated after the calmodulinbinding site, displays an apparent affinity for Ca 2ϩ very similar to that of the full-length hPMCA4xb, but its basal activity is 2-3 times higher (4). This fact indicates that the calmodulinbinding sequence suffices to keep the enzyme in a state of low affinity for Ca 2ϩ , whereas additional determinants downstream from the calmodulin-binding sites are required for full inhibition.
In agreement with the mechanism of autoinhibited protein kinases and phosphatases (5,6), the inhibition of the PMCA seems to involve the interaction between the catalytic region of the molecule and the inhibitory sequence (see Fig. 1A). Indeed, it has been shown that a peptide of 28 amino acids (C28) with the sequence of the calmodulin-binding autoinhibitory site binds to segments Ile 206 -Val 271 and Cys 537 -Thr 544 in the minor and large cytoplasmic loops, respectively (7,8). The homologous regions in SERCA are part of domains "A" and "N" (9). The ACA2 from Arabidopsis thaliana is also a calmodulin-regulated Ca 2ϩ pump containing an autoinhibitory region, which, at variance with the PMCA, is located at the N terminus of the molecule. Recently Curran et al. (10) found that mutations of residues in the stalk region of ACA2 disrupt autoinhibition. In particular the residue Asp 219 located in the stalk segment 2 was extremely sensitive, because substitutions by Asn, Glu, or Lys resulted in calmodulin-independent activated pumps. In parallel with ACA2, the stalk portion of the PMCA may also be important for autoinhibition. Indeed, homology modeling of the PMCA using the crystal structures of SERCA1 suggests that residue Asp 170 (homologous to Asp 219 in ACA2) is in a conformationally sensitive portion of M2 (see Fig. 1B).
Here we report on the effects of alterations of Asp 170 of the PMCA. Similar to that described previously in ACA2, we found that the mutation D170N results in a deregulated calmodulininsensitive PMCA. Hence, this residue seems equally critical for autoinhibition in both ion pumps. Nevertheless, we found that the D170N PMCA does not fully resemble the calmodulin-* This work was supported by grants from Agencia Nacional de Promoción Científica y Tecnológica (BID 1201 OC-AR PICT 1-6845), Consejo Nacional de Investigaciones Científicas y Técnicas, Universidad de Buenos Aires, and Ministerio de Salud de la Nación (Argentina). 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.
‡ To whom correspondence should be addressed. Tel.: 5411-4964-8289 (ext. 125); Fax: 5411-4962-5457; E-mail: hpadamo@qb.ffyb.uba.ar. 1 The abbreviations used are: PMCA, plasma membrane Ca 2ϩ ATPase; ct92, mutant of hPMCA4xb lacking the C-terminal 92 amino acids; SERCA, sarcoplasmic/endoplasmic Ca 2ϩ ATPase; MOPS, 3-(Nmorpholino)propanesulfonic acid; h, human; recWT, recombinant wild type. activated enzyme, particularly because in the mutant, activation took place without fully disengaging the C-terminal domain from the catalytic core. These results indicate that the molecular mechanism involved in the activation of the PMCA by D170N is only in part common to that elicited by calmodulin, and the results suggest a direct role of Asp 170 in regulating the access to the calcium-binding residues of the pump.

MATERIALS AND METHODS
Chemicals-Polyoxyethylene 10 lauryl ether (C 12 E 10 ), L-␣-phosphatidyl-choline type X-E (P-5394) from dried egg yolk, phosphodiesterase-3Ј, 5Ј-cyclic nucleotide activator (calmodulin) from bovine brain, calmodulin-agarose, calcimycin (A23187), chymotrypsin, ATP (disodium salt, vanadium-free), sodium dodecyl sulfate, yeast synthetic drop-out media supplement without leucine, yeast nitrogen base without amino acids, dextrose, enzymes and cofactors for the synthesis of [␥-32 P]ATP, and all other chemicals were obtained from Sigma. Tryptone and yeast extract were from Difco. Carrier-free [ 32 P]H 3 PO 4 was provided by PerkinElmer Life Sciences. [␥-32 P]ATP was prepared by the method of Glynn and Chappell (11) except that no unlabeled orthophosphate was added to the incubation medium. Salts and reagents were of analytical reagent grade.
Site-directed Mutagenesis-Portions of hPMCA4xb were amplified by a two-step PCR process to obtain the desired mutation in a clonable fragment using Vent R DNA polymerase (New England BioLabs Inc.) following the manufacturer's instructions. The sequence of the oligonucleotide primers were, HIS (CTGCAGGTCGACCATGGCGCATCACCATCACCA-TCACAACCCATCAGACCGTGTCTTGCCTGCCAACCTCCCCTGTGGA-AGGTCTGTCTG); 3DN, (CTTTGCTCCAATtATTAAAGGCAGTCACTA); 5DN, (GCCTTTAATaATTGGAGCAAAGAGAAGC); and 829, (AAGGAT-GATTCCAGTCTGAGAGTT). The products HIS-3DN and 5DN-829 were isolated from the reaction mixture in a 1% agarose gel and extracted using DNA QIAEX II (Qiagen). The isolated products of the first PCR were used in a subsequent PCR step with primers HIS and 829. The amplified fragment was subcloned in the corresponding position of hPMCA4xb after digestion with SalI and PflmI. A similar strategy was used to obtain the mutant D170K. The mutated PMCA sequence was confirmed by DNA sequencing. The DNA coding for the wild type and mutants was inserted between XhoI and ApaI in the pMP625 vector containing a Leu ϩ marker for expression in Saccharomyces cerevisiae under the control of the PMAI promoter. The resulting plasmids were using for yeast transformation.
Membrane Isolation and Purification of Recombinant PMCA-Yeast was grown at 28°C in 4 liters of YNB Leu Ϫ medium to an approximate A 600 ϭ 1.5 and then it was diluted with 4 liters of complete medium and was grown to an A 600 ϭ 4.0 -5.0. Cells were pelleted, washed with 400 ml of water, and resuspended in 200 ml of homogenization buffer (50 mM MOPS-K, pH 7.40 at 4°C, 500 mM sucrose, 300 mM KCl, 2 mM EGTA, 2 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride). 300 g of glass beads (0.5-mm diameter) were added to the resuspended cells (ϳ80-g dry weight), and cells were disrupted using a bead beater for 15 min on ice. Glass beads and large cell debris were pelleted, and the supernatant was centrifuged at 40,000 ϫ g for 30 min. The microsomal membranes were resuspended in ϳ5 ml of purification buffer (20 mM MOPS-K, pH 7.40 at 4°C, 20% glycerol, 130 mM KCl, 1 mM MgCl 2 , 2 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, and 100 M Ca 2ϩ ), were homogenized in a glass homogenizer, and were solubilized with C 12 E 10 at 5 mg/ml for 10 min. Non-solubilized material was removed by centrifugation at 40,000 ϫ g for 20 min, and the solute was then applied to an agarose-calmodulin column, which was equilibrated in purification buffer with 5 mg/ml C 12 E 10 . The column was washed with 10 bed volumes of the same buffer with 0.5 mg/ml C 12 E 10 and then changed to elution buffer (same composition as washing buffer but with 1 mM EGTA-K instead CaCl 2 ). The entire procedure was performed at 4°C.
Protein Assay-The protein concentration was initially estimated by the method of Bradford (14) using bovine serum albumin as a standard and then corrected by the intensity of the bands in Coomassie Bluestained gels using a standard of PMCA purified from porcine red cells.
Reconstitution of PMCA with Liposomes-The preparation of proteoliposomes was carried out as described by Hao et al. (15) with a slight modification. Briefly, 1 ml of purified PMCA was added to 250 l of a previously sonicated mixture of 29 mg/ml phosphatidylcholine and 57 mg/ml C 12 E 10 . The detergent/protein/phospholipid mixture was kept under gentle stirring for 5 min and then the detergent was removed by adding 0.25 g of prewashed wet Bio-Beads SM-2 (Bio-Rad). The mixture was stirred at room temperature for 30 min, and the Bio-Beads were separated by filtration. After adding 60 M A23187 in ethanol, the proteoliposomes were frozen in liquid nitrogen and stored at Ϫ80°C.
Ca 2ϩ -ATPase Activity-Activity was estimated from the release of [ 32 P]P i from [␥-32 P]ATP at 37°C (16) in 0.3 ml of ATPase reaction medium containing proteoliposomes with 2 g of PMCA, 50 mM HEPES-K, pH 7.00 at 37°C, 100 mM KCl, 4 mM MgCl 2 , 0.5 mM EGTA, 2 M A23187, 3 mM [␥-32 P]ATP, and enough CaCl 2 and calmodulin to yield the Ca 2ϩ and calmodulin concentrations indicated in each experiment. The reaction was initiated by the addition of the proteoliposomes and was terminated by acid denaturation after 30 min. During this period the amount of P i liberated from ATP increased linearly with time. The free Ca 2ϩ concentration in the reaction medium was calculated using the Fabiato and Fabiato program (17) with Blink's constants for EGTA/Ca 2ϩ , EGTA/Mg 2ϩ , ATP/Ca 2ϩ , and ATP/Mg 2ϩ . The ATPase activity of the detergent-solubilized PMCA (before reconstitution) was assayed in a reaction medium containing 20% glycerol, 33 g/ml phosphatidylinositol, and 66 g/ml C 12 E 10 . Ca 2ϩ -independent activity in this experimental condition was 10 -20% of the total ATPase activity.
Electrophoretic Analysis of Purified PMCA, Western Blotting, and Protein Staining-SDS electrophoresis and immunoblotting were carried out as described previously (18). Proteins were electrophoresed on a 7.5% acrylamide gel according to Laemmli (19) and revealed by Coomassie Blue staining or subsequently electrotransferred onto Millipore Immobilon P 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 for 1 h with 5F10 antibody (20, 21) from ascitic fluid (dilution 1:1000). For staining, biotinylated anti-mouse immunoglobulin G and avidin-horseradish peroxidase conjugate were used. The image was acquired using a Foto/Analyst documentation system from Fotodyne, and the intensity of the bands was quantified with the Gel-Pro Image Analyzer software.
Proteolytic Digestion of PMCA-Limited proteolysis of the membranes from yeast transformed with the appropriate construct was carried out at 37°C essentially as described in Padanyi et al. (22). 10 g of membranes isolated from yeast was preincubated for 5 min at 37°C in 100 l of medium containing 100 mM KCl, 25 mM HEPES-K ϩ , 1 mM MgCl 2 , 100 M EGTA-K ϩ , 200 M CaCl 2 (100 M free Ca 2ϩ when added), 5 mM dithiothreitol, 0.05% Triton X-100, and 485 nM calmodulin (when added). The proteolysis was initiated by the addition of 0.05 g/ml ␣-chymotrypsin and was arrested at the indicated times by the addition of ice-cold trichloroacetic acid (6% final concentration). The precipitate was supplement with 100 g of bovine serum albumin, washed twice with distilled water, and then dissolved in electrophoresis sample buffer.

Expression and Purification of the Wild Type PMCA and
Mutants-A novel expression system using yeasts was used for the production of recombinant PMCA. 2 S. cerevisiae DBY2062 cells were transformed with the pMP625 vector containing the human PMCA4xb cDNA under the control of the strong PMA1 promoter. The presence of the PMCA protein in yeast membranes was detected by SDS-PAGE and Western blot (Fig. 1C). Antibody 5F10 recognized a single band at 135 kDa indicating that the full-length polypeptide of the recombinant wild type (recWT) protein was successfully expressed. Mutants D170N and D170K were expressed at a level similar to that of the wild type protein. Membranes from yeast were solubilized with detergent C 12 E 10 , and the recombinant PMCAs were purified by calmodulin affinity chromatography. As shown in Fig. 1D, the EGTA eluate contained highly pure PMCA protein as revealed by electrophoresis and Coomassie Blue staining of the gel.
Functional State of Mutants D170N and D170K-The C 12 E 10 solubilized recombinant PMCA purified from yeasts was supplemented with phosphatidylinositol to obtain a full activation (23). Under these conditions the recWT exhibited a Ca 2ϩ ATPase activity of ϳ2.4 mol P i /mg protein/min, whereas mutant D170N had a somewhat lower activity of 1.5 mol of P i /mg of protein/min. After the reconstitution of the purified proteins in phosphatidylcholine liposomes the maximal activities of the recWT and D170N enzymes in the presence of saturating amounts of Ca 2ϩ -calmodulin were 0.72 mol of P i /mg of protein/min and 0.43 mol of P i /mg of protein/min, respectively. Substitution of Asp 170 by Lys led to a substantial reduction in the activity, which for the soluble enzyme activated by phosphatidylinositol was 0.13 mol of P i /mg of protein/min and thus this mutant was not studied further.
Dependence of Ca 2ϩ ATPase of Mutant D170N on Ca 2ϩ Concentration-The activity of the recWT and D170N enzymes reconstituted in phosphatidylcholine liposomes was measured as function of Ca 2ϩ concentration in both the presence and absence of calmodulin. In the absence of calmodulin the recWT had a low apparent affinity for Ca 2ϩ (K 0.5 Ca 2ϩ ϭ 3.4 M) and a V max at 40% of that attained in the presence of calmodulin ( Fig. 2A). The addition of calmodulin increased the V max and lowered the K 0.5 Ca 2ϩ to ϳ0.5 M. Curves in Fig. 2B show that in the presence of calmodulin the Ca 2ϩ dependence of D170N was identical to that of the recWT. However, in the absence of calmodulin the apparent affinity for Ca 2ϩ of D170N remained as high as in the presence of calmodulin, and its V max was ϳ80% of that observed in the presence of calmodulin. The kinetic parameters of recWT and D170N are summarized in Table I.
Effect of ct92 Truncation on the D170N Mutant-The activation of mutant D170N could reflect the release of the inhibition because of the segment downstream the calmodulin-binding site. To test this idea a mutant D170Nct92 truncated after the calmodulin-binding site was constructed. The maximal specific Ca 2ϩ ATPase activity of C 12 E 10 -solubilized D170Nct92 supplemented with phosphatidylinositol was similar to that of D170N. Fig. 3 shows that as expected, after reconstitution in phosphatidylcholine liposomes D170N and D170Nct92 had similar apparent affinities for Ca 2ϩ . However, the ct92 truncation of D170N increased the V max in the absence of calmodulin, suggesting that the D170N mutation does not affect the inhibition of V max exerted by the portion of the C-terminal regulatory domain downstream from the calmodulin-binding site.
Activation of recWT and D170N by ATP-High concentrations of ATP in the range of the low affinity regulatory site ATP increase the stimulatory effect of calmodulin (24). To test whether the differences in calmodulin activation of recWT and D170N were related to a different dependence on ATP, the activity of the recWT and D170N enzymes was measured as function of ATP concentration in both the presence and absence of calmodulin. Fig. 4 shows that the D170N and recWT enzymes had a similar response to ATP either in the presence or in the absence of calmodulin. At saturating concentrations of ATP calmodulin increased the activity of recWT ϳ3-fold, whereas D170N was stimulated 1.4-fold.
Effect of Vanadate on the Ca 2ϩ ATPase of the recWT and D170N Enzymes-Because vanadate is a transition state anafied from porcine red cells was loaded. D, gel run as in C and stained with Coomassie Blue. Yeast recWT membranes, 3.5 and 7 g of membranes from yeast expressing hPMCA4xb; recWT, 0.6 and 1 g of recombinant hPMCA4xb purified from yeast; D170N, 0.9 and 1.3 g of purified D170N mutant; D170K, 1.5 and 2.3 g of purified D170K mutant; pPMCA, 1.7 and 2.6 g of PMCA purified from porcine red cells.  (35), an automated homology modeling server running at the Geneva GlaxoSmithKline Experimental Research. The coordinates of the structures of SERCA identified with the protein data bank codes 1EUL and 1IWO were used as templates for the E 1 and E 2 conformations, respectively. C, Western blot of yeast membranes containing the expressed recWT hPMCA4xb and mutants. 10 g of total yeast membranes protein was applied on each lane of an SDS-7.5% polyacrylamide gel and subjected to immunoblot analysis with monoclonal antibody 5F10. Yeast transfected with the empty vector (control) or the vector containing the cDNA encoding hPMCA4xb (recWT), D170N, or D170K. pPMCA, 50 ng of PMCA puri-logue of inorganic orthophosphate and competes with the binding of P i to the E 2 conformation of the P-type ATPases, the response to inhibition by vanadate has been frequently used as a probe of the E 2 conformation (25). To assess whether D170N changed the equilibrium between the E 2 and E 1 forms of the enzyme the effect of vanadate was investigated (Fig. 5). In the absence of calmodulin vanadate inhibited the recWT enzyme with K i ϭ 11 M. Notably, in the presence of calmodulin the affinity of recWT for vanadate increased ϳ3-fold. (K i ϭ 2.6 M).
In contrast with the calmodulin-activated enzyme, the D170N mutant had a low apparent affinity for vanadate (K i ϭ 21 M), and it was not affected by calmodulin.
Dependence of recWT and D170N on Calmodulin Concentration-The Ca 2ϩ -ATPase activity of recWT and D170N enzymes was measured as a function of calmodulin concentration. The measurement was performed at 10 M free Ca 2ϩ to ensure the saturation of calmodulin with Ca 2ϩ at all calmodulin concentrations. Under this condition the maximal activation by calmodulin was 2.5-and 1.3-fold for recWT and D170N, respectively. However, as shown in Fig. 6, both enzymes attained half-maximal activation at ϳ13 nM calmodulin.
Accessibility of the C-terminal Region to Degradation by Chymotrypsin-It has been shown recently that the interaction between the C-terminal autoinhibitory domain and the catalytic core of hPMCA4xb affects the sensibility of the C-terminal segment to proteolysis (22). Results in Fig. 7 show the effect of chymotrypsin on the recWT and D170N pumps. In agreement with Ref. 22, the degradation of the full-length recWT at the C-terminal segment producing a 124.5-kDa fragment proceeded slowly in the absence of Ca 2ϩ , slightly faster in the presence of Ca 2ϩ , and much faster in the presence of Ca 2ϩcalmodulin. Similarly, the degradation of D170N was minimally affected by Ca 2ϩ but was significantly enhanced by Ca 2ϩcalmodulin. However, even in the presence of calmodulin, the D170N mutant was more resistant to proteolysis than the recWT enzyme. (26,27) have lead to the identification of several residues from the C-terminal regulatory segment of the PMCA that are critical for calmodulin binding and autoinhibition. In contrast, the role of the catalytic core in autoinhibition is much less known. Ba-Thein et al. (28) reported that activation of the PMCA4 requires the substitution of both its small and large cytoplasmic loops by those of the naturally activated PMCA2. The results presented here show that mutation of residue Asp 170 to Asn suffices for the activation of PMCA4 and thus they provide the first evidence of the importance of a single residue from the core of the PMCA molecule in autoinhibition. Curran

TABLE I Effects of the D170N substitution on the kinetics parameters of recWT
The values of K 0.5 Ca 2ϩ , V max , and n H (Hill coefficient) Ϯ S.D. are from fitting the Hill equation to the curves in Fig. 2. CaM, calmodulin.

RecWT D170N
ϪCaM ϩCaM ϪCaM ϩCaM mutations that constitutively activate the ACA2 pump. The mutation D219N (homologous to D170 in hPMCA4xb) was found to be present in 80% of the potentially deregulated pumps generated by the chemical mutagenesis of ACA2. Therefore, this change is very effective in activating the enzyme. In the PMCA we found that despite its activating effects, mutations D170N and D170K decreased the specific activity to 63 and 5%, respectively, suggesting a role for Asp 170 beyond autoinhibition.
Mutation D170N Activates the PMCA by Increasing the Affinity for Ca 2ϩ -Although the mutation of Asp 170 by Asn duplicated the maximal activity of the PMCA in the absence of calmodulin, its major effect was to increase the apparent affinity for Ca 2ϩ to ϳ7-fold to a value similar to that of the calmodulin-activated enzyme. This effect seems specific for Ca 2ϩ , because neither the affinity of ATP nor that for calmodulin showed significant changes. Despite its high apparent affinity for Ca 2ϩ , the V max of the D170N pump was further increased by ct92 truncation, suggesting that the mutation D170N does not affect the inhibition by determinants downstream from the calmodulin-binding domain.
Sensitivity of the Activated PMCA to Inhibition by Vanadate-Vanadate is potent inhibitor of the PMCA (29); however the effect of calmodulin on the inhibition of the PMCA by vanadate had not been investigated before. We found that calmodulin increases the sensitivity of the PMCA to inhibition by vanadate. Other autoinhibited P-ATPases such as the yeast plasma membrane H ϩ -ATPase also exhibit a higher apparent affinity for vanadate when they are in the activated state (30). The higher apparent affinity for vanadate of the calmodulinactivated PMCA may indicate that in the presence of calmodulin the enzyme is in an open conformation with a more accessible site for vanadate. At variance with the effect of calmodulin, activation by D170N made the PMCA more resistant to inhibition by vanadate. Furthermore inhibition of the D170 mutant required higher concentrations of vanadate than the recWT enzyme in the absence of calmodulin. This result supports the idea that the mechanism underlying the activation of the PMCA by the mutation D170N is not identical to that promoted by calmodulin.
D170N Does Not Disrupt the Interaction between the Autoinhibitor and the Catalytic Core-As in other calmodulin-regulated enzymes, the apparent affinity of PMCA for calmodulin is determined not just by the strength of binding to its site but also by the accessibility of the binding site, which is dependent upon its interaction with the core of the molecule (31). In previous studies (26 -28) mutants with a basal activity higher than that of the recWT enzyme were shown to also have a higher apparent affinity for calmodulin, a fact that was taken as indicative of a weaker interaction between the catalytic core and the C-terminal autoinhibitory domain. We found that mutant D170N despite having a higher basal activity and hence a small degree of calmodulin activation, did not exhibit a significant change in the apparent affinity for calmodulin. This result suggests that this mutation does not sufficiently affect the interaction between the autoinhibitory region and the catalytic core as to modify the accessibility of calmodulin to its site.
An alternative test of the strength of the interaction between the C-terminal regulatory region and the catalytic core of the PMCA has been reported recently (22). Activation of the PMCA either by calmodulin or mutations in the C-terminal segment was shown to increase the proteolytic digestion of the C-terminal region. Consistent with the idea that mutation of Asp 170 to Asn did not abolish the interaction between the C terminus and the core of the PMCA, we found that the chymotryptic digestion of mutant D170N at the C-terminal segment proceeded slowly in the presence of Ca 2ϩ and much faster after the addition of calmodulin. However the mutant appeared to be more resistant to proteolysis at the C terminus than the recWT protein, suggesting the persistence of the interaction with the core even in the presence of calmodulin.
The deregulation by mutation at residue Asp 170 in PMCA and the corresponding residue in ACA2 suggests that different autoinhibited pumps share a similar mechanism for activation. However different residues may be involved in the contacts between the autoinhibitory region and the central catalytic portion of the pumps. Indeed a number of mutations that cause activation have been reported in other pumps (30,32,33). On the other hand, the results presented here point out that mechanisms other than a weaker interaction with the C terminus may result in a calmodulin-like activation. It is noteworthy that our modeling of the PMCA by homology with the SERCA1 pump suggests that in the E 1 conformation Asp 170 is part of a long and continuous M2 helix, which extends up to Cys 182 , whereas in the E 2 conformation a portion of the M2 helix between Phe 168 and Gln 176 becomes unwound. It has been recently proposed (34) that in the SERCA pump the C-terminal end of the M2 helix plays a critical role in the conformational changes leading to the occlusion by closing off the access to the cytosol of the bound ion. Thus, it is conceivable that the activation of the PMCA by the mutation of Asp 170 involves conformational rearrangements in the core of the enzyme that directly modify the accessibility of Ca 2ϩ transport site. FIG. 7. C-terminal digestion by chymotrypsin of recWT and D170N. A, 10 g of purified recWT or D170N protein were digested with chymotrypsin as described under "Materials and methods" in the absence of Ca 2ϩ (EGTA), in the presence of Ca 2ϩ , and in the presence of Ca 2ϩcalmodulin (CaM) for the times indicated. Proteolysis was terminated by the addition of trichloroacetic acid, and the precipitate was loaded in an SDS-polyacrylamide gel and immunoblotted using anti-PMCA antibody 5F10. B, rate of appearance of the 124.5-kDa chymotryptic PMCA fragment. The intensity of the bands from three experiments as that shown in A was quantified using the GelPro image analyzer software. The ratio between the amount of 124.5 kDa fragment and the intact protein is plotted as a function of the time of proteolysis. Empty symbols, without calmodulin; filled symbols, with calmodulin. Circles and triangles, recWT; Squares and inverted triangles, D170N.