Determination of the Dissociation Constants for Ca2+ and Calmodulin from the Plasma Membrane Ca2+ Pump by a Lipid Probe That Senses Membrane Domain Changes*

The purpose of this work was to obtain information about conformational changes of the plasma membrane Ca2+-pump (PMCA) in the membrane region upon interaction with Ca2+, calmodulin (CaM) and acidic phospholipids. To this end, we have quantified labeling of PMCA with the photoactivatable phosphatidylcholine analog [125I]TID-PC/16, measuring the shift of conformation E2 to the auto-inhibited conformation E1I and to the activated E1A state, titrating the effect of Ca2+ under different conditions. Using a similar approach, we also determined the CaM-PMCA dissociation constant. The results indicate that the PMCA possesses a high affinity site for Ca2+ regardless of the presence or absence of activators. Modulation of pump activity is exerted through the C-terminal domain, which induces an apparent auto-inhibited conformation for Ca2+ transport but does not modify the affinity for Ca2+ at the transmembrane domain. The C-terminal domain is affected by CaM and CaM-like treatments driving the auto-inhibited conformation E1I to the activated E1A conformation and thus modulating the transport of Ca2+. This is reflected in the different apparent constants for Ca2+ in the absence of CaM (calculated by Ca2+-ATPase activity) that sharply contrast with the lack of variation of the affinity for the Ca2+ site at equilibrium. This is the first time that equilibrium constants for the dissociation of Ca2+ and CaM ligands from PMCA complexes are measured through the change of transmembrane conformations of the pump. The data further suggest that the transmembrane domain of the PMCA undergoes major rearrangements resulting in altered lipid accessibility upon Ca2+ binding and activation.

Detailed structural information about the plasma membrane calcium pump (PMCA) 2 is currently lacking. This pump is an integral part of the Ca 2ϩ signaling mechanism (1) and is thus a crucial component of cell function. It is highly regulated by calmodulin (CaM), which activates this protein by binding to an auto-inhibitory region (2) and changes the conformation of the pump from an inhibited state to an activated one (2,3).
Information about the structure and assembly of the transmembrane domain of an integral membrane protein can be obtained from an analysis of the lipid-protein interactions. In this work, we have used a hydrophobic photolabeling method to study the noncovalent interactions between the membrane domain of the PMCA and surrounding phospholipids under different experimental conditions that lead to known conformations. It has been previously demonstrated that both the conformation and the activity of the pump protein are preserved in either solubilized or reconstituted purified preparations compared with the native pump located in the erythrocyte (4).
In recent work, we assessed changes in the overall exposure of PMCA to surrounding lipids by quantifying the extent of protein labeling by the photoactivatable phosphatidylcholine analog [ 125 I]TID-PC/16 under different conditions (5). This showed that labeling of PMCA incubated with Ca 2ϩ and calmodulin decreases by 25% and incubation with Ca 2ϩ alone increases labeling by more than 50% compared with control labeling of the PMCA in the absence of Ca 2ϩ and CaM. These results suggest that the PMCA assumes two different E 1 conformations: one that is auto-inhibited and in which the membrane domain is in contact with a higher amount of lipids (incubating with Ca 2ϩ alone, E 1 I) and one in which the enzyme is fully active (incubating with Ca 2ϩ -calmodulin, E 1 A) and exhibits a more compact transmembrane arrangement with lesser exposure to lipids. These data provide the first evidence that there is an auto-inhibited conformation in P-type ATPases involving both the cytoplasmic regions and the transmembrane segments.
Activation of PMCA is caused by the binding of CaM to the C-terminal tail of the pump, leading to dissociation of the autoinhibitory domain from its close proximity to the active site. This removes the self-inhibition of the enzyme and stimulates the PMCA-mediated Ca 2ϩ transport severalfold (6). Previously, the binding of CaM to PMCA was measured indirectly by determining the PMCA activity (7). Recently, Liyanage et al. (8) reported a fluorescence polarization method to measure the binding of CaM modified with Oregon Green 488. This was an evident improvement because it evaluates CaM binding in equilibrium instead of at steady-state with PMCA activity. However, it does not evaluate the binding of CaM itself but of a modified probe. In this report, we introduce a method that allows calculation of the dissociation constant for binding of non-labeled CaM to PMCA.
We employed the photoactivatable phosphatidylcholine analog 1-palmitoyl-2-[9-[2Ј-[ 125 I]iodo-4Ј-(trifluoromethyldiazirinyl)-benzyloxycarbonyl]-nonaoyl]-sn-glycero-3-phosphocholine [ 125 I]TID-PC/16 that has been previously used to analyze lipid-protein interfaces (9 -11). This reagent partitions in the phospholipidic milieu and upon photolysis reacts indiscriminately with its molecular environment. It is thus possible to directly analyze the interaction between the hydrophobic membrane-spanning domain of a membrane protein and lipids belonging to its immediate environment (12)(13)(14). Applying this technique on the PMCA, we were able to measure equilibrium constants for the dissociation of ligands from PMCA complexes and to draw structural conclusions about the regulation of the transport of Ca 2ϩ in the presence of different modulators.

EXPERIMENTAL PROCEDURES
Reagents-All chemicals used in this work were of analytical grade. Calmodulin was obtained from Calbiochem. Recently drawn human blood for the isolation of PMCA was obtained from the Hematology Section of the Hospital de Clínicas General San Martín and from Fundación Fundosol (Argentina). Blood donation in Argentina is voluntary, and therefore the donor must provide informed consent for the donation of blood and the subsequent legitimate use of the blood by the transfusion service.
Purification of PMCA from Human Erythrocytes-PMCA was isolated from CaM-depleted erythrocyte membranes by the CaM-affinity chromatography procedure (4) and stored in 20% (w/v) glycerol, 0.005% C 12 E 10 , 120 mM KCl, 1 mM MgCl 2 , 10 mM MOPS-K, pH 7.4 at 4°C, 2 mM EGTA, 2 mM dithiothreitol. Protein concentration after purification was about 10 g/ml. No phospholipids were added at any step along the purification procedure. The PMCA preparation thus obtained is devoid of natural phospholipids, as assessed by the failure to detect inorganic phosphate by the method described below (see "Phospholipid Quantification"). Solubilization and purification of PMCA preserve transport activity and maintain the kinetic properties and regulatory characteristics of the enzyme in its native milieu (4 Weber and Brunner (11). After the reaction was completed, the mixture was extracted with chloroform/methanol (2:1, v/v) and [ 125 I]TID-PC/16 was purified by passage through a silica gel column (2.5 ml) using chloroform/ methanol: water/acetic acid (65:25:4:1, v/v) as solvent. The elution was monitored by TLC/autoradiography, and the fractions containing the product were dried and stored at Ϫ20°C.
Phospholipid Quantification-Phospholipid concentration was measured according to Chen et al. (15) with some modifications. Samples and standards containing 10 -100 nmol of phosphorus were dried by heating at 100°C. Mineralization was carried out by adding 0.1 ml of HNO 3 , 0.9 ml of HClO 4 , and incubating at 190°C for 30 min. Inorganic phosphate was determined after Fiske and Subbarow (16).
Measurement of Ca 2ϩ -ATPase Activity-ATPase activity was measured at 37°C by following the release of inorganic phosphate from ATP as described previously (4). The incubation medium was: 80 M DMPC, 120 M C 12 E 10 , 120 mM KCl, 30 mM MOPS-K (pH 7.4), 3.75 mM MgCl 2 , 1 mM EGTA, and enough CaCl 2 to give the desired final free [Ca 2ϩ ]. The reaction was started by the addition of ATP (final concentration 2 mM). Release of P i was estimated according to the procedure of Fiske and Subbarow (16).
Labeling Procedure-A dried film of the photoactivatable reagent was suspended in DMPC/C 12 E 10 mixed micelles (80 and 120 M, respectively) containing 10 g/ml of the membrane protein, 120 mM KCl, 30 mM MOPS-K (pH 7.4), 3.75 mM MgCl 2 , 1 mM EGTA, and enough CaCl 2 to give the desired final free [Ca 2ϩ ]. The samples were incubated for 20 min at 37°C before being irradiated for 15 min with light from a filtered UV source ( Ϸ360 nm).
Radioactivity and Protein Determination-Electrophoresis was performed according to the Tris-Tricine SDS-PAGE method (17). Polypeptides were stained with Coomassie BlueR, the isolated bands were excised from the gel, and the incorporation of radioactivity was directly measured on a ␥-counter. The amount of protein was quantified by eluting each stained band as previously described (18), including bovine serum albumin in each gel as a standard for protein quantification. Specific incorporation was calculated as the ratio between measured radioactivity and amount of protein determined for each band.
Proteolysis of PMCA-Proteolysis of PMCA was performed for 5 min in the presence of 25 mM Tris-HCl, pH 7.4 at 37°C, 2 mM EGTA, and 0.22 g/ml of TLCK-treated chymotrypsin in water. The reaction was stopped by a 10-fold excess of ovomucoid trypsin inhibitor solution at 4°C.
Data Analysis-All measurements were performed in triplicate to quintuplicate unless specified otherwise in the figures.

Incorporation of [ 125 I]TID-PC/16 as a Conformation Marker-
We first wished to determine the extent of [ 125 I]TID-PC/16 labeling of the PMCA in its major known conformational states. The E 2 state is attained by incubating the PMCA in the absence of Ca 2ϩ (1 mM EGTA). The incorporation of [ 125 I]TID-PC/16 in this condition was considered as the control and was set as 100% (Fig. 1). The other two conformers are E 1 I, which is obtained by incubating the enzyme in the presence of Ca 2ϩ and binds the maximum concentration of [ 125 I]TID-PC/16 (180% at optimal concentration of Ca 2ϩ , Fig. 1) and E 1 A, a conformation attainable in the presence of Ca 2ϩ and CaM or with a CaM-like effector such as phosphatidic acid, oleic acid or after removal of the C terminus of PMCA. This conformation binds the least amount of [ 125 I]TID-PC/16 (ϳ75% of the control for E 1 A obtained with Ca 2ϩ and CaM) (Fig. 1). These data are as reported earlier (5) and illustrate the sensitivity of the incorporation of labeled [ 125 I]TID-PC/16 on the different PMCA conformations.
Titration of the E 2 to E 1 I Conformer Shift with Ca 2ϩ -To determine the [Ca 2ϩ ] dependence of the conformational shift from the E 2 to the E 1 I state, we performed a titration experiment ( Fig. 2A) in which we measured the specific incorporation of [ 125 I]TID-PC/16 to PMCA when the enzyme was incubated in the presence of increasing concentrations of Ca 2ϩ . In this experiment, we titrate the concentration of [ 125 I]TID-PC/16 that binds to PMCA as it shifts from the E 2 to the E 1 I conformer, i.e. as [ 125 I]TID-PC/16 incorporation increases from 100% to near 180%. As illustrated in Fig. 2A, [ 125 I]TID-PC/16 incorporation increases hyperbolically with low concentrations of Ca 2ϩ (up to ϳ40 M) and then decreases to a level that cannot be easily evaluated with the available information. However, the data are well described by Equation 1.  It was reported earlier that excess [Ca 2ϩ ] in SERCA also induced a conformation different from that attained at lower concentrations of Ca 2ϩ (19). Fig. 2B shows the Ca 2ϩ -ATPase activity as a function of Ca 2ϩ . The data were quantitatively evaluated using Equation 2, where V max is the maximum velocity, K Ca 2ϩ is the apparent dissociation constant for Ca 2ϩ (Ca 2ϩ concentration at halfmaximal ATPase activity), and K i is the concentration of Ca 2ϩ for half-maximal inhibition of the enzyme at high [  Fig. 2A) and ATPase activity (Fig. 2B) is that the K 0.5 for the binding of [ 125 I]TID-PC/16 is much lower (ϳ0.5 M) than the K Ca 2ϩ for Ca 2ϩ activation  of the PMCA (11 M). This strongly suggests that the intrinsic Ca 2ϩ affinity of the PMCA in equilibrium is very different from its apparent Ca 2ϩ affinity for activation by Ca 2ϩ .

Titration of the Ca 2ϩ Dependence of the E 2 to E 1 A Conformer Shift in the Presence of Calmodulin and Calmodulin-like
Treatments-We next wished to compare the Ca 2ϩ dependence of the E 2 Ϫ E 1 I shift with that of the shift from E 2 to the activated E 1 A conformation of the PMCA. To this end, we performed a series of titrations of [ 125 I]TID-PC/16 incorporation in the presence of various activators of the pump. We chose the well-known "prototypical" PMCA activator CaM, the lipid activators phosphatidic acid (PA) and oleic acid (OA), as well as limited proteolytic treatment using TLCK-chymotrypsin. The latter treatment results in a C-terminally truncated and fully active pump lacking its auto-inhibitory tail (20). These activators thus reflect different mechanisms and pathways of PMCA activation.
Activation   Whereas the values for most parameters are similar for the PA-and CaM-stimulated E 2 -E 1 A transition, the value of n describing the sigmoidicity of the Ca 2ϩ dependence is much higher for PA than for any of the other treatments. Equation 3 is a Hill equation that shows in the denominator term the exponent n for the Ca 2ϩ concentration, where n can be higher than one. In fact the value for n is near 2 for the effect of Ca 2ϩ on [ 125 I]TID-PC/16 binding in the presence of CaM (as well as of OA and for the enzyme that lacks the C-terminal region; see below), indicating that more than one molecule of Ca 2ϩ is needed for the E 2 to E 1 A shift. However, in the presence of PA, n is near 4 for the effect of Ca 2ϩ on [ 125 I]TID-PC/16 binding. This may be caused by a more complex behavior between PA and Ca 2ϩ : (i) PA could be at a non-optimal concentration and (ii) PA could associate with free Ca 2ϩ . However, when we measured the concentration of Ca 2ϩ with a Ca 2ϩ -sensitive electrode in the absence and in the presence of 60 M of PA, we found the value for free [Ca 2 ] to be similar. We also determined the [ 125 I]TID-PC/16 binding to PMCA as a function of PA, and the results indicate that the concentration used in the experi-ments of this work is sufficient to obtain the maximal effect, i.e. to reach PC min (data not shown).
Measuring the PA-stimulated Ca 2ϩ -ATPase activity as a function of concentration of Ca 2ϩ (inset in Fig. 3B) yielded a V max of 11.3 Ϯ 0.2 mol P i mg Ϫ1 ⅐min Ϫ1 , and K Ca 2ϩ ϭ 0.83 Ϯ 0.07 M. These values are comparable to those determined for the activation of the PMCA by CaM and indicate that the apparent Ca 2ϩ affinity for activation by either mechanism approaches the equilibrium Ca 2ϩ affinity of the PMCA. The inset in Fig. 3C shows the Ca 2ϩ -ATPase activity in the presence of 10 M OA as a function of Ca 2ϩ . The experimental points were adjusted to a simple hyperbolic function yielding a V max of 11.1 Ϯ 0.3 mol P i mg Ϫ1 ⅐min Ϫ1 , and K Ca 2ϩ ϭ 0.96 Ϯ 0.14 M.
Removal of the PMCA C-terminal Domain by Proteolysis with TLCK-chymotrypsin-As mentioned above, proteolysis of PMCA by chymotrypsin yields a pump that is fully active, lacks the C-terminal region, and hence does not bind CaM. We used TLCK-chymotrypsin to avoid digestion of the N-terminal domain, which remains intact based on the reaction of the truncated PMCA with the specific antibody JA9 (5). The Ca 2ϩ -ATPase activity as a function of [Ca 2ϩ ] of the C-terminally truncated pump is shown in the inset of Fig. 3D. Adjusting the experimental points to a simple hyperbolic function gave a V max of 13.9 Ϯ 0.4 mol P i mg Ϫ1 ⅐min Ϫ1 , and K Ca 2ϩ ϭ 0.53 Ϯ 0.07 M.
Comparison of Apparent Ca 2ϩ Affinities for PMCA Measured in Equilibrium and in the Steady State- Table 1 summarizes the results of the experiments described above and compares the Ca 2ϩ affinities obtained through binding of [ 125 I]TID-PC/ 16 and through the determination of Ca 2ϩ -ATPase activity in similar experimental conditions. From an inspection of this table we can draw the following conclusions: 1) Treatments

affinities determined by enzyme activity and by specific [ 125 I]TID-PC/16 incorporation under equilibrium conditions
Results represent the mean Ϯ S.E. of 4 -9 independent experiments by triplicate. E 1 A: an activated state of E 1 ; E 1 I: an inhibited state of E 1 . PA: phosphatidic acid; OA: oleic acid; Chym: a 120-kDa PMCA devoid of the C-terminus obtained after chymotrypsin digestion.  1 and 3. b The negative value reflects the fact that for Ca 2ϩ alone, PC max is higher than PC 0 . that alter the equilibrium between E 1 and E 2 are recognized by a change in the incorporation of the probe. 2) Ca 2ϩ affinities correlate well between the two methods employed when CaM or CaM-like treatments modified the PMCA. However, 3) The affinity for Ca 2ϩ alone is similar to that in the presence of CaM or CaM-like treatments only when this parameter is evaluated according to [ 125 I]TID-PC/16 incorporation on PMCA undergoing the shift between E 2 to E 1 A or E 2 to E 1 I. Taken together, these results indicate that the affinity for Ca 2ϩ of the PMCA is not regulated at the Ca 2ϩ site in the membrane but rather through the auto-inhibitory cytoplasmic domain.

Condition Conformation Ca 2؉ -ATPase activity Specific incorporation of [ 125 I]TID-PC/16
Titration of the E 1 I to E 1 A Conformer Shift with Calmodulin-Activation of the PMCA is due to the binding of CaM, which dissociates the auto-inhibitory domain, removes the self-inhibition by the enzyme and thus stimulates PMCA transport severalfold (6). One way to measure the binding of CaM to the PMCA has been to follow pump activity as a function of the CaM concentration (7). We measured the binding of CaM to PMCA by quantifying the amount of bound

DISCUSSION
[ 125 I]TID-PC/16 has previously been used to identify and characterize regions within membrane proteins that interact with lipids (11,14). Its physicochemical behavior in terms of mobility in thin layer chromatography is indistinguishable from PC, and its interaction with the transmembrane region of integral membrane proteins also appears to be identical to that of PC (12,13). PC is generally chosen as the reference lipid against which relative lipid association constants of integral membrane proteins are compared, because these proteins show no selectivity for this phospholipid (21). This fact simplifies the interpretation of our results, as it allows a direct correlation between level of reagent incorporation and amount of protein surface exposed to surrounding lipids.
By using [ 125 I]TID-PC/16 as a sensitive probe of the hydrophobic protein environment, we were able to study the transmembrane region of the PMCA in different conformations. Quantification of the amount of labeling by [ 125 I]TID-PC/16 then allowed us to calculate the apparent affinity of the PMCA for Ca 2ϩ and CaM as the pump shifts between different conformers in equilibrium with surrounding lipids.
Activation of the PMCA by CaM is commonly explained by the binding of CaM to the CaM binding domain in the C-terminal tail of the pump, followed by release of the inhibitory interactions from the cytosolic core. It should be noted that this hypothesis for the mechanism of auto-inhibition and activation does not predict changes in the transmembrane region. However, in a previous report, we demonstrated that the auto-inhibited conformation is distinct in its membrane domain, and that the conformational changes induced by auto-inhibition do expose additional hydrophobic surfaces of the protein to phospholipids. The auto-inhibited conformation was also obtained by adding the CaM-binding peptide C28 to an E 1 CaCaM state of PMCA (corresponding to the activated pump), showing the reversibility of this conformational transition (5). On the basis of these observations we postulated that the PMCA possesses two different E 1 -Ca 2ϩ conformations: one that is auto-inhibited and is in contact with a higher amount of lipids (incubating with Ca 2ϩ alone, E 1 I) and one in which the enzyme is fully active (incubating with Ca 2ϩ -calmodulin, E 1 A) and that exhibits a more compact transmembrane arrangement with a smaller surface area exposed to lipids.
According to kinetic enzyme activity measurements the activated conformer E 1 A can be obtained in the presence of CaM, acidic phospholipids, unsaturated fatty acids, protein oligomerization, or after removing the C-terminal tail by controlled proteolysis, e.g. with chymotrypsin (2,4,6,22). In this report, we used several of these activating modes to compare the apparent affinity of the PMCA for Ca 2ϩ as determined in two entirely different ways: (a) by evaluating the K 0.5(Ca2ϩ) for activation of the Ca 2ϩ -ATPase activity, i.e. in steady-state conditions of the enzyme, and (b) by measuring the equilibrium constants that drive the conformer E 2 to E 1 A as reflected in a change in [ 125 I]TID-PC/16 incorporation. The results clearly show that the PMCA possesses a high-affinity site for Ca 2ϩ (K D ϳ0.5 M) regardless of the presence or absence of activators. Therefore, modulation of the pump activity is exerted through the C-terminal domain, which induces an auto-inhibited conformation but does not modify the affinity for Ca 2ϩ at the site where it binds for transport into the transmembrane domain. This finding is analogous to an earlier study on the cardiac muscle SERCA (SERCA2a), which showed that the equilibrium Ca 2ϩ binding affinity of the pump was unaffected by the inhibitory protein phospholamban (23). Similar to CaM activation of the PMCA, phospholamban phosphorylation stimulates SERCA activity by releasing the inhibitory interaction, which is reflected in an increase in the apparent Ca 2ϩ affinity of the pump.
Ca 2ϩ -ATPase activation yields an "apparent" affinity for Ca 2ϩ , which is not the same as the actual binding affinity under equilibrium conditions. Our approach measures the real substrate affinity for Ca 2ϩ in different conditions through the binding of [ 125 I]TID-PC/16, which is directly proportional to the transmembrane surface of PMCA exposed to surrounding lipids. These two affinities should not be compared directly because they are determined in different states of the system. The experiments of this paper were designed to show that the true affinity of PMCA for Ca 2ϩ is high, a fact that is masked by the steady-state condition, which shows a low affinity site for Ca 2ϩ .
Our findings raise interesting questions concerning the structural features of the transmembrane region of the PMCA in its auto-inhibited versus activated E 1 -Ca conformation. In the SERCA pump, for which atomic-resolution structural information is available for several conformers in both the E 1 and E 2 state (24 -28), calculations of the membrane-embedded accessible surface area revealed only relatively modest differences between different Ca 2ϩ -bound E 1 -conformers (5). Major rearrangements involving the membrane-spanning region are, however, observed upon the E 1 to E 2 (or E 2 -E 1 ) transition of the enzyme cycle. Because of the gross architectural similarity of the PMCA to the SERCA, it is assumed that comparable structural rearrangements accompany the reaction cycle in the PMCA, including large domain movements of the A (actuator) and N (nucleotide-binding) domains with respect to each other and the P (phosphorylation) cytosolic domain. Our data based on the changes in [ 125 I]TID-PC/16 lipid binding to the PMCA clearly suggest that a major difference exists in the membrane domain of the pump in the auto-inhibited E 1 -Ca conformation (E 1 I) compared with the activated E 1 -Ca (E 1 A) conformation where the auto-inhibitory tail is dissociated from the core of the enzyme. The most salient difference between the SERCA and PMCA pumps is the presence of the extended C-terminal tail in the PMCA that encompasses the auto-inhibitory region. How could the presence of this tail, which is thought to be cytosolic, have such a dramatic effect on the membrane portion of the PMCA? Cross-linking studies have shown that in the auto-inhibited state, the tail is in close proximity to sequences in both the A-and N-domain (29,30) and FRET studies indicate that the N and C termini of the pump are separated by less than 50 Å in the inhibited state (3). Thus, in the inhibited E 1 I conforma-tion where the C-tail is "clamped-down" on the A-and N-domains, the transmembrane domain of the PMCA may be uniquely different from that in the E 1 -Ca state of SERCA and of the activated E 1 A state of the PMCA. Indeed, we found that any of the activating treatments (CaM, phosphatidic or oleic acid, and limited chymotrypsin proteolysis) resulted in a comparable (albeit not identical) change in [ 125 I]TID-PC/16 incorporation to PMCA, suggesting that the common denominator is the release of the auto-inhibitory tail to allow a shift to the E 1 A conformer. Although a better understanding of the differences in the membrane domain of the E 1 I and E 1 A conformers of the PMCA will require higher resolution structural information, the data strongly suggest that substantial differences exist in the arrangement of the transmembrane domain in the PMCA in comparison with the SERCA pump, at least in the auto-inhibited E 1 state of the PMCA. We cannot exclude the possibility that CaM, PA, OA, and partial proteolysis affect the membrane domain of the PMCA in a different manner and to a different degree. In fact, incorporation of [ 125 I]TID-PC/16 to PMCA is different in the presence of CaM or PA. Experiments involving [ 125 I]TID-PC/16 labeling and proteolysis with V8 protease show that labeling of PMCA with CaM affects transmembrane segments 3 and 4 and the bundle of TM 5 to 10, whereas in the presence of PA transmembrane segments 1-2 and 3-4 are mostly responsible for the incorporation of [ 125 I]TID-PC/16. 3 Therefore, the definition of the characteristics of the conformer E 1 A is purely operational and involves a series of structurally different conformers with a similar apparent high affinity for Ca 2ϩ .
Using a similar approach as for the determination of Ca 2ϩ affinities we followed the transition from E 1 I to E 1 A by titrating this conformational shift with CaM. This allowed us to calculate the dissociation constant for binding of native nonlabeled CaM to PMCA in equilibrium. The resulting value of 9.6 Ϯ 0.8 nM is similar to the K D for CaM obtained in steady state conditions by evaluating the half-maximal activation of Ca 2ϩ -ATPase (7.2 Ϯ 1.4 nM measured in this study (Fig. 4), 11.6 Ϯ 2.4 nM, reported by Penheiter et al. (31)). This indicates that the apparent affinity for CaM is indistinguishable in steady state and in equilibrium experiments. Recently, Liyanage et al. (8), using a fluorescence polarization assay to evaluate the binding of a modified CaM to the PMCA, determined a K D of 5.8 Ϯ 0.5 nM, a value comparable to the one obtained in this work. However, it is possible that the small difference in these K D values is attributable to different sources of CaM or to the Oregon Green 488 modified CaM used in their method. Regardless, our approach allows an independent calculation of the dissociation constant of nonlabeled CaM from the PMCA.
The data presented here represent the first example where equilibrium constants for the dissociation of ligands from PMCA complexes are measured through the change of transmembrane conformations of the pump. The method is sensitive, precise, and requires a very low amount of protein. This should enable its application to other membrane proteins that cycle through different conformations, such as all members of the P-type ATPase family. Experiments are currently underway using this methodology to evaluate the affinity constants of PMCA for ATP under different experimental conditions.