New Insights into the Role of the N Terminus in Conformational Transitions of the Na,K-ATPase*

The deletion of 32 residues from the N terminus of the α1 catalytic subunit of the rat Na,K-ATPase (mutant α1M32) shifts the E1/E2 conformational equilibrium toward E1, and the combination of this deletion with mutation E233K in the M2-M3 loop acts synergistically to shift the conformation further toward E1 (Boxenbaum, N., Daly, S. E., Javaid, Z. Z., Lane, L. K., and Blostein, R. (1998)J. Biol. Chem. 273, 23086–23092). To delimit the region of the cytoplasmic N terminus involved in these interactions, the consequences of a series of N-terminal deletions of α1 beyond Δ32 were evaluated. Criteria to assess shifts in conformational equilibrium were based on effects of perturbation of the entire catalytic cycle ((i) sensitivity to vanadate inhibition, (ii) K+ sensitivity of Na-ATPase measured at micromolar ATP, (iii) changes in K′ATP, and (iv) catalytic turnover), as well as estimates of the rates of the conformational transitions of phospho- and dephosphoenzyme (E1P → E2P and E2(K+) → E1 + K+). The results show that, compared with α1M32, the deletion of up to 40 residues (α1M40) further shifts the poise toward E1. Remarkably, further deletions (mutants α1M46, α1M49, and α1M56) reverse the effect, such that these mutants increasingly resemble the wild type α1. These results suggest novel intramolecular interactions involving domains within the N terminus that impact the manner in which the N terminus/M2-M3 loop regulatory domain interacts with the M4-M5 catalytic loop to effect E1 ↔ E2 transitions.

The Na,K-ATPase or sodium pump is a ubiquitous integral membrane protein that catalyzes the glycoside sensitive, ATPcoupled exchange of three intracellular Na ϩ for two extracellular K ϩ ions across the plasma membrane of all animal cells. The sodium pump is essential to the maintenance of the electrochemical gradients of Na ϩ and K ϩ across the cell membrane, providing the driving force for the transport of nutrients into the cell and maintaining the cellular resting membrane potential. It comprises two essential subunits: a large catalytic ␣ subunit (ϳ100 kDa), containing the ligand binding and phos-phorylation sites and a smaller, highly glycosylated ␤ subunit (ϳ35-55 kDa), which acts as a chaperone for ␣ (for review see Refs. 1 and 2). A third subunit, ␥ (ϳ7 kDa), was found in the kidney where it functions as a regulator of the pump (see Refs. 3 and 4).
The sodium pump is a member of a family of transporters known as P-type ATPases that are directly phosphorylated and dephosphorylated on a conserved aspartate residue within their cytoplasmic domain during the course of the reaction cycle. Both the phosphorylated and dephosphorylated forms of the enzyme can exist in at least two states that undergo conformational transitions (E 1 P 3 E 2 P and E 2 3 E 1 ) that are coupled to the ion-translocating steps. Definitive evidence for distinct E 1 and E 2 conformational states and a role of the N terminus in effecting E 1 7 E 2 transitions was first obtained in 1975 by Jorgensen (5,6) using tryptic cleavage of the renal enzyme in the presence of different ligands. Later confirmatory evidence was obtained using N-terminal deletion mutants expressed in cultured cells (7). This study showed that whereas deleting up to and including the lysine-rich cluster is without effect, removal of 32 residues (mutant ␣1M32) alters the enzyme kinetics by shifting the E 1 /E 2 conformational equilibrium toward E 1 forms. Interestingly, a similar shift toward E 1 is caused by Glu 233 3 Lys substitution in the first (M2-M3) cytoplasmic loop. Furthermore, the combined removal of the N-terminal 32 residues and replacement of Glu 233 3 Lys (mutant ␣1M32E233K) results in a remarkably synergistic shift in poise toward E 1 state(s) as evidenced in analyses of several characteristic properties including, for examples, extraordinary insensitivity to vanadate, high affinity for ATP (5 M), and slow (Յ500 min Ϫ1 ) catalytic turnover of ␣1M32E233K. These findings were interpreted to indicate that interactions between these two cytoplasmic regions and the M4-M5 catalytic loop are critical for conformational coupling (8).
The present study was carried out to define more precisely the role of the N terminus in conformational transitions. Accordingly, we investigated the consequences of further deletions of the N terminus of the Na,K-ATPase beyond the first 32 residues to near the beginning of the first transmembrane segment, using several criteria to assess shifts in the steadystate E 1 7 E 2 poise. The results of this analysis reveal a progressively increased shift in the E 1 /E 2 poise toward E 1 , followed by reversal toward the wild type ␣1 E 1 /E 2 distribution. Combined with secondary structure predictions of the N terminus, this behavior identifies a self-regulatory domain within the N terminus of the Na,K-ATPase that modulates conformational transitions via novel intramolecular interactions.

EXPERIMENTAL PROCEDURES
Determination of Helicity-Predictions of the secondary structure of the N-terminal sequence were evaluated using the following algorithms: PREDATOR (9), Sspro (10), SOPMA (11), GOR IV (12), Deleage * This work was supported by operating grants from the Canadian Institutes of Health Research (Grant MT-3876) and the Quebec Heart and Stroke Foundation (to R. B.), the National Institutes of Health (Grant HL 49204) (to L. K.), and a predoctoral fellowship from the Heart and Stroke Foundation of Canada (to L. S.). 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.
Mutagenesis, Transfection, Selection, and Cell Culture-The desired mutations were introduced into the 5Ј SacI 230 -SalI 875 restriction fragment cassette of the rat ␣1 cDNA as described previously (7). The mutant cassettes were then ligated into the rat ␣1 cDNA in the place of the wild type SacI-SalI cassette. The full-length mutant cDNAs were released form the shuttle vector by digestion with HindIII, ligated into the expression plasmid pCDNA3.1 (Invitrogen), and orientation of the cDNA was determined by restriction analysis. HeLa cells were transfected with the pCDNA-␣1 mutant constructs using either the calcium phosphate method (18) or the LipofectAMINE technique (Invitrogen), and cells expressing the relatively ouabain-resistant rat ␣1 enzymes and mutants thereof were selected as described previously (19,20). HeLa cells expressing the mutant ␣1 enzymes were amplified in Dulbecco's modified Eagle's medium plus 10% newborn calf serum, 100 units/mg penicillin G, 100 g/ml streptomycin, and 1 M ouabain as described previously (7). The above medium was initially supplemented with an additional 15 mM KCl (20 mM total; see Ref. 21), which was removed after several passages.
Membrane Preparation-NaI-treated microsomal membranes were prepared from the mutant cells as described earlier (19,20). Protein content was determined with a detergent-modified Lowry assay (22).
Enzyme Assays-Na,K-ATPase activity was measured as the release of 32 P i from [␥-32 P]ATP as previously described (23). Briefly and unless indicated otherwise, the membranes were preincubated for 10 min at 37°C with all reactants added except [␥-32 P]ATP. The reaction was initiated by the addition of [␥-32 P]ATP. Final concentrations for Na,K-ATPase activity measurements were 100 mM NaCl, 10 mM KCl, 3 mM MgSO 4 , 20 mM histidine (pH 7.4), 5 mM EGTA (pH 7.4), and 5 M ouabain (SIGMA). 5 mM ouabain was used to determine baseline hydrolysis activity. As in earlier studies and unless indicated otherwise, assays of Na,K-ATPase activity were carried out using 1 mM ATP to maintain close to saturating ATP concentration and also maximize sensitivity of assays of the relatively low activity cultured cells (see Refs. 20,24,and 25). Na-ATPase activity was measured at 1 M ATP as described previously (7), with varying amounts of added KCl and choline chloride to maintain constant chloride (40 mM) concentration, and baseline activity was determined with 40 mM KCl. For studies of vanadate sensitivity, inorganic orthovanadate (Fisher) solutions were prepared prior to the experiment and added with the [␥-32 P]ATP solution to initiate the reaction. Na,K-ATPase activities obtained at various vanadate concentrations and expressed as percentage of that obtained in the absence of vanadate, were analyzed by fitting the data to a one-compartment model using a non-linear least-square analysis of a general logistic function, as described elsewhere (26).
Phosphoenzyme Determination-Phosphoenzyme levels were determined as described earlier (25). Briefly, membranes (ϳ0.01 mg/assay) were preincubated with ouabain (20 M ouabain, 0.4 mM MgSO 4 , and 10 mM glycylglycine-Tris (pH 7.4)) for 10 min at 37°C to inhibit endogenous HeLa Na,K-ATPase. The suspension was then treated with either oligomycin (20 g/ml) for 1 min at room temperature or with vehicle alone (ethanol) for baseline measurements (see below). Phosphorylation was then carried out for 10 s at 0°C in a final volume of 150 l in medium comprising (final concentrations) 100 mM NaCl, 10 mM glycylglycine-Tris (pH 7.4), 5 mM EGTA, 1 mM MgSO 4 , and 1 M [␥-32 P]ATP (specific activity 10,000 -20,000 cpm/pmol). Baseline EP levels were determined by replacing 100 mM NaCl with 50 mM KCl and 50 mM choline chloride, and phosphorylating the enzyme in the absence of oligomycin.
Formation of the E 1 from E 2 (K ϩ )-The rate of K ϩ deocclusion (E 2 (K ϩ ) 3 E 1 ϩ K ϩ ) was measured indirectly by determining the rate of E 1 formation from E 2 (K ϩ ) as described elsewhere (25), with the modifica-tions described by Therien and Blostein (27).
Rate of E 1 P 3 E 2 P-Following formation of E 1 P in the presence of high chloride concentration (28), the rate of E 1 P 3 E 2 P was determined by measuring the rate of disappearance of total phosphoenzyme following rapid dilution of the salt (to allow normal relaxation of E 1 P 3 E 2 P) plus addition of KCl to catalyze rapid hydrolysis of E 2 P (28, 29). Accordingly, the enzyme was first phosphorylated in medium containing 600 mM NaCl to stabilize E 1 P, 1 mM MgCl 2 , 1 mM EGTA, and 20 mM Tris-HCl (pH 7.4) with 1 M [␥-32 P]ATP for 30 s at 0°C to obtain maximal phosphoenzyme. Dephosphorylation was then initiated by 6-fold dilution with a chase medium containing final concentrations of 20 mM KCl, 10 M unlabeled ATP, 1 mM EGTA, and 20 mM Tris-HCl (pH 7.4), which simultaneously lowered the NaCl concentration to 100 mM. Samples were taken for measurement of E 32 P for periods of up to 30 s. Background phosphoenzyme levels were obtained by allowing the chase to continue for 60 s.
At least two different membrane preparations obtained from at least two different clones were assayed. The data presented are representative of at least three independent experiments. Each value shown is the mean Ϯ S.D. of triplicate determinations.

RESULTS
To determine more precisely the structural basis and mechanism underlying the role of the cytoplasmic N terminus of the ␣1 catalytic subunit of the Na,K-ATPase on the steady-state E 1 3 E 2 conformational poise, we first considered the nature of the secondary structure of this 85-residue N-terminal domain. This region is a highly charged, flexible structure with a high degree of helicity. A comparison of the results of several computational analyses (see "Experimental Procedures") reveals the presence of three putative ␣-helical domains encompassing residues 27-33, 42-50 and 61-68 ( Fig. 1). To remove or disrupt these regions, wild type HeLa cells were transfected with N-terminal deletion mutants of the ␣1 subunit of the rat Na,K-ATPase corresponding to either disruption or loss of the first putative helix (␣1M32 and ␣1M40, respectively) and disruption (␣1M46 and ␣1M49) or loss (␣1M56) of the second helix (Fig. 1). All five of the mutant constructs yielded functional enzyme capable of sustaining HeLa cell growth in 1 M ouabain.
Several criteria were used to analyze shifts in the E 1 /E 2 distribution: some that reflect the operation of the overall catalytic cycle and others that reflect partial reactions relevant to the major E 1 7 E 2 and E 1 P 7 E 2 P transitions. The former include: (i) sensitivity of Na,K-ATPase activity to inhibition by vanadate, (ii) response of Na-ATPase to varying K ϩ concentrations relevant to the normally rate-limiting E 2 (K ϩ ) 3 3 E 1 ϩ K ϩ , (iii) catalytic turnover, and (iv) KЈ ATP relevant to low affinity ATP binding to E 2 (K ϩ ). The latter include:  1. N terminus of the rat ␣1 Na,K-ATPase and deletion mutants thereof. Computational analysis of the N-terminal sequence predicts three helices in the N-terminal of the ␣1 Na,K-ATPase within, at least, residues 27-33, 42-50, and 61-68 (for details see "Experimental Procedures"). Numbering is based on the mature rat ␣1 amino acid sequence (60).
deletions on conformational equilibrium, we investigated the effect of vanadate on Na,K-ATPase activity. Inorganic orthovanadate is a transition state analog of inorganic phosphate that binds to P-type ATPases in the E 2 conformation (30). Consequently, sensitivity of an enzyme to inhibition by vanadate is a measure of the proportion of enzyme in the E 2 conformation. As shown by the representative experiment in Fig. 2 and summarized in Table I, ␣1M32 and ␣1M40 are both less sensitive to vanadate inhibition than is ␣1, suggesting that the E 1 /E 2 equilibrium of these enzymes progressively shifts toward E 1 as up to 40 residues are deleted from the N terminus. However, this shift is reversed to a more wild type-like sensitivity by deleting Ն46 residues (mutants ␣1M46, ␣1M49, and ␣1M56).
At micromolar ATP concentrations, sufficient to saturate only the high affinity phosphorylation site, the response of Na-ATPase to K ϩ is a sensitive means to characterize mutantspecific differences in the K ϩ deocclusion pathway of the reaction cycle (E 2 (K ϩ ) 3 3 E 1 ϩ K ϩ ], which becomes ratelimiting under these conditions (7,31). Thus, as shown previously and in Fig. 3, K ϩ inhibits the Na-ATPase activity of ␣1 but stimulates that of ␣1M32. The deletion of 40 residues, ␣1M40, results in an even greater K ϩ stimulation, up to ϳ400%. Interestingly, further deletion, i.e. ␣1M46, ␣1M49, and ␣1M56, yields ␣1-like K ϩ inhibition. These results, summarized in Table I, suggest a progressive shift in the E 1 /E 2 conformational equilibrium toward E 1 upon removal of up to 40 residues from the N terminus, which is reverted by deleting Ն46 residues.
The catalytic turnover of the Na,K-ATPase is estimated as the ratio of V MAX to EP MAX , the latter measured at 0°C in the  Fig. 2. b Activity with 1 mM K ϩ presented as percent of control activity without K ϩ (data from representative experiment in Fig. 3). c Catalytic turnover was determined as the ratio of maximal Na,K-ATPase activity measured at 100 mM NaCl, 10 mM KCl, and 1 mM ATP to maximal phosphoenzyme measured at 0°C in the presence of 100 mM NaCl and oligomycin (see " presence of ATP, Na ϩ , and oligomycin to trap the enzyme in the (Na ϩ )E 1 P state (see "Experimental Procedures") (32). Table I shows the catalytic turnover of ␣1 and the mutant enzymes. As previously shown, ␣1M32 has ϳ50% reduction in turnover, consistent with a shift in the E 1 7 E 2 poise toward E 1 . ␣1M40 has a significantly lower turnover, only 20% that of ␣1. In contrast, deletion of Ն46 residues restores the catalytic turnover to near that of ␣1.
Differences in K ϩ sensitivity of Na-ATPase at low ATP concentration (Fig. 3) suggest a change in the rate of a step in the K ϩ deocclusion phase of the Albers-Post reaction mechanism, which can be modeled by a branched pathway (see Scheme 1 in Ref. 33). In one branch, pathway A, ATP first binds with low affinity denoted by KЈ ATP(L) , to the K ϩ occluded enzyme, followed by rapid deocclusion (ATP⅐E 2 (K ϩ ) 3 ATP⅐E 1 ⅐K 3 ATP⅐E 1 ϩ K ϩ ). In the other branch, pathway B, the slow release of K ϩ from E 2 (K ϩ ) via E 2 (K ϩ ) 3 E 1 ⅐K ϩ 3 E 1 ϩ K ϩ is followed by high affinity ATP binding to E 1. Accordingly, a mutation causing a change in K ϩ stimulation of Na-ATPase at micromolar ATP concentration may reflect a change in KЈ ATP(L) (pathway A) and/or a change in rate of the second pathway (pathway B). Fig.  4 shows that both ␣1M32 and ␣1M40 have significantly higher apparent affinities for low affinity ATP binding compared with ␣1 (see KЈ ATP(L) values in Table I). In contrast, ␣1M46 and ␣1M49 have KЈ ATP(L) values similar to that of the wild type ␣1 enzyme, consistent with the observed reversal of the K ϩ effect on Na-ATPase at 1 M ATP. It was noted, however, that the KЈ ATP(L) of ␣1M56 is lower that that of ␣1 even though this enzyme is inhibited by K ϩ at low ATP. This characteristic of ␣1M56 is considered further in the "Discussion." (ii) Analysis of Partial Reactions Relevant to Conformational Transitions-In another series of experiments we compared the slow release of K ϩ from E 2 (K ϩ ) via the branch E 2 (K ϩ ) 3 E 1 ⅐K ϩ 3 E 1 ϩ K ϩ for each of the mutants. As in our earlier studies (25,27) we used an indirect approach whereby the relatively slow rate of E 1 formation from E 2 (K ϩ ) is estimated at various periods following dilution of preformed E 2 (K ϩ ) in Na ϩ medium containing [␥-32 P]ATP at 10°C. Assuming that the ensuing phosphorylation of E 1 is rapid, the amount of phosphoenzyme (presumably Na⅐E 1 P since oligomycin is present to trap this intermediate) is thus a measure of E 1 formed from E 2 (K ϩ ) . As shown in Fig. 5, the changes in E 2 (K ϩ ) deduced from the increases in E 1 P, fit well to single exponentials. The rate constants for ␣1M32 and ␣1M40 shown in Table II are significantly higher than that of ␣1. (Rate constants are similar to our previously published values for ␣1 and ␣1M32, Ref. 25). This result suggests that for both ␣1M32 and ␣1M40 the E 2 (K ϩ ) 7 7 E 1 ϩ K ϩ poise is shifted to the right. On the other hand, values for ␣1M46, ␣1M49, and ␣1M56 are similar to, or even lower than, those for ␣1M32 and ␣1M40, suggesting that deletion of Ն46 residues reverses the above right-shift to a more wild type-like E 1 /E 2 distribution. It should be noted, however, that whereas the K ϩ sensitivity of Na-ATPase at 1 M ATP reflects overall turnover via both branch pathways of K ϩ deocclusion, with the contribution of pathway A increasing as KЈ ATP(L) decreases, the time course of formation of E 1 from E 2 (K ϩ ) reflects only deocclusion via pathway B. It is not known whether changes in each of the two pathways are quantitatively equivalent. They may not be equivalent. Thus, K ϩ stimulation of the Na-ATPase is notably greater for ␣1M40 compared with ␣1M32 despite their similar rates of E 2 (K ϩ ) 3 E 1 ; K ϩ inhibition profiles of ␣1M46 and ␣1M49 are similar to that of ␣1, yet their rates of E 2 (K ϩ ) 3 E 1 are slower than that of ␣1. Despite these issues and the fact that this reaction was analyzed at 10°C, the generally consistent pattern of increasing followed by decreasing rates of E 2 (K ϩ ) 3 E 1 associated with progressive deletions, is noteworthy.
To investigate the contribution of the conformational transition of the phosphoenzyme to the E 1 7 E 2 shifts observed above, we examined the E 1 P 3 E 2 P transition at 0°C as previously described (28,29). Accordingly, the enzyme is first phosphorylated by [␥-32 P]ATP at high salt concentration (600 mM NaCl) to stabilize E 1 P. Dephosphorylation is then initiated by a downward jump in NaCl concentration to 100 mM with the simultaneous addition of 20 mM KCl and 10 M unlabeled ATP.  5. Rate of formation of E 1 from E 2 (K ؉ ). The rate of K ϩ deocclusion was measured indirectly as the rate of formation of E 1 from E 2 (K ϩ ) at 10°C as described under "Experimental Procedures." Results shown are taken from a representative experiment; each value is the mean Ϯ S.D. of triplicate determinations. Symbols are as in the legend to Fig. 2.
Since the rate of K ϩ -activated dephosphorylation of E 2 P is much faster than that of the preceding formation of E 2 P from E 1 P, the time course of the E 1 P decay reflects primarily E 1 P 3 E 2 P. As shown in Fig. 6 and summarized in Table II, compared with the wild type ␣1 enzyme, the rate of this transition is ϳ6-fold slower for ␣1M32 and ␣1M40. Mutants ␣1M46, ␣1M49, and ␣1M56 have similar or only slightly slower dephosphorylation rates relative to the wild type enzyme. Taken together, these partial reaction analyses of the mutants and wild type enzyme reflect the behavior seen in assays of the overall catalytic cycle summarized in Table I. DISCUSSION The primary sequences of the N and C termini are the least conserved domains among the various P-type ATPases found in lower organisms and throughout the plant and animal kingdom. These extramembranous termini contain regions that inhibit pump activity, as well as binding motifs for regulatory proteins that may release pump inhibition by these autoinhibitory sequences (34). Examples include the plasma membrane Ca 2ϩ -ATPase, which contains auto-inhibitory domains in its N terminus in plants or in its C terminus in animals (35)(36)(37). Despite their low homology, a structural paradigm does exist for the N terminus of P-type ATPases, whereby the region comprises two domains. These are: (i) a variable domain, which differs markedly in sequence and length among the ATPases and (ii) a homologous domain that continues into the S1 stalk region of the N terminus (38). The variable domain, as the name indicates, varies greatly in size from the very long termini of metal-transporting ATPases such as the Cu-ATPase (39,40) to being virtually absent in the sarcoplasmic reticulum Ca 2ϩ -ATPase (SERCA). 1 The homologous domain consists of about 33 amino acids and has a high propensity to form two short helices in H,K-ATPase, SERCA, and Na,K-ATPase (38), which, in the latter pump, comprises helices H2 and H3 described below.
The N terminus of the Na,K-ATPase is a highly charged and flexible structure, with a high degree of helicity (see Fig. 1). Although the primary sequence of this region diverges greatly among isoforms and species, the following features persist: (i) a lysine-rich cluster of about 5-9 residues and (ii) a highly conserved 10 amino acid sequence circumscribed by two methionines (M(D/E)ELKKE(I/V)(S/T)M) (41) and containing a proteolytically sensitive lysine residue (5).
In his landmark studies, Jorgensen showed that the lysinerich domain of the N terminus of the Na,K-ATPase precedes a proteolytically sensitive site (T2) that can be selectively cleaved by trypsin when the enzyme is in the E 1 conformation (5,6). Cleavage at this residue increases KЈ K and decreases KЈ ATP , with associated effects on the phosphorylation and dephosphorylation reactions of the enzyme (42)(43)(44), providing evidence for a shift in the E 1 /E 2 equilibrium toward E 1 forms (45,46). Several studies have demonstrated that although the lysine cluster is not essential to pump function (47)(48)(49)(50)(51), the N terminus may play a role in Na ϩ sensitivity (52), K ϩ affinity, and in the dependence on membrane potential (53,54) likely resulting from changes in rates of Na ϩ translocation (41) and K ϩ deocclusion reactions (55). 1 The abbreviation used is: SERCA, sarcoplasmic reticulum Ca 2ϩ -ATPase. We have previously shown that deletion of up to 27 residues preceding the first putative helix has no effect on the conformational equilibrium of the enzyme (7,8,25). Only upon deletion of 32 residues, which corresponds to T2 described above, is there a shift toward E 1 (8,25). Here we show that disruption (␣1M32) or loss (␣1M40) of H1 results in a progressive shift toward E 1 forms of the enzyme. We also show that disruption (␣1M46 and ␣1M49) or loss (␣1M56) of H2 (as well as H1) reverses the shift to a more wild type-like E 1 7 E 2 equilibrium.
The aforementioned progressive changes notwithstanding, the behavior of ␣1M56 is slightly peculiar. Unlike ␣1M46 and ␣1M49, ␣1M56 retained a lower KЈ ATP(L) even though all three mutants resemble wild type ␣1 with respect to the following: (i) vanadate sensitivity, (ii) sensitivity to K ϩ inhibition at low ATP, (iii) rate of E 1 formation from E 2 (K ϩ ), and (iv) rate of E 1 P 3 E 2 P. These observations may indicate some change or destabilization of structure that occurs with extensive truncation. For example, truncation beyond the first 49 residues may perturb a heretofore undefined interaction of the N terminus with another region, which may cause ligandspecific effects independent of the conformational transitions, such as an increased access of nucleotide substrate to the binding site.
Cytoplasmic Interactions Involved in Conformational Transitions-Insights into cytoplasmic region(s) that interact with the N terminus were first obtained by Jorgensen and Collins (56) who proposed salt-bridge formation between the N terminus and other cytoplasmic domains. In a later study (8), the remarkable synergistic effects of the ⌬32 deletion and the E233K mutation implicated the N terminus/M2-M3 loop in such a salt-bridge formation. It is noteworthy that the cytoplasmic region encompassing the N terminus/M2-M3 loop is analogous to the Activator or A domain identified in the crystal structure of SERCA1a (57). In the Na,K-ATPase, interaction of the A domain with the catalytic loop was observed in the studies of Karlish and co-workers who used metal catalyzed cleavage to study domain interactions of the Na,K-ATPase in E 1 versus E 2 conformations (for review, see Ref. 58). Accordingly, in the E 1 conformation, the nucleotide binding domain in the catalytic M4-M5 loop docks onto the phosphorylation domain in M4-M5, and domain A moves aside. In E 2 (K ϩ ), domain A docks onto the phosphorylation domain, and the nucleotide binding domain is displaced. It is noteworthy that all of the data are consistent with the published SERCA1a crystal structure and with further structural evidence that E 1 7 7 E 2 P transitions of SERCA involve large domain movements (57). The emerging picture is that the structural changes underlying conformational transitions are generally similar for SERCA and the Na,K-ATPase, and much can be extrapolated about their analogous structures (for a detailed comparison, see Ref. 59). A main structural distinction is the N terminus, which is encompassed by domain A. In SERCA, the N terminus is a much shorter extension from the M1 transmembrane helix, lacking the first helical domain, and that which remains has a uniquely different primary sequence from that of the Na,K-ATPase.
The Distinct Modulatory Role of the N Terminus of the Na,K-ATPase-The present mutagenesis study reveals a specific role of the N terminus of the Na,K-ATPase. The model shown in Fig. 7 depicts a region of the N terminus that acts as an autoregulatory domain modulating the E 1 /E 2 conformational transitions. In the case of the wild type enzyme (Fig. 7, panel A), helical region H2 can interact with H1 either through helix-helix interactions and/or salt-bridge formation in the E 2 conformation, allowing the M2-M3 loop and the catalytic loop to come together (shaded ovals). H2 can also interact with a domain of the M2-M3 loop in E 1 to keep it apart from the catalytic loop as in the E 1 conformation. When H1 is disrupted or lost (␣1M32 or ␣1M40, see Fig. 7, panel B), FIG. 7. Hypothetical model for N-terminal regulation of conformational equilibrium. Based on evidence from earlier work and the present study, a model is proposed whereby an autoregulatory region of the N terminus modulates the poise in E 1 7 E 2 . For details see "Discussion." H2 preferentially interacts with the M2-M3 loop, placing a constraint on the latter, thereby weakening the M2-M3/ M4-M5 interaction and consequently stabilizing the E 1 conformation. The result is a shift in the conformational equilibrium in favor of E 1 . Lastly, the disruption (␣1M46 and ␣1M49) or loss of H2 (␣1M56) (see Fig. 7, panel C) alleviates the constraint on the M2-M3 loop allowing it once again to freely interact with the large catalytic M4-M5 loop, thus resembling the wild type ␣1 enzyme.
It is also noteworthy that the ␣2 and ␣3 isoforms of the catalytic subunit of the rat Na,K-ATPase have similar helical structures in their N termini. Furthermore, analogous disruptions of the first helix of ␣2 and ␣3 (mutants ␣2M30 and ␣3M26, respectively) 2 shift the conformational equilibrium for each enzyme toward E 1 forms, suggesting that the N terminus of the Na,K-ATPase may carry an autoregulatory domain present in all three isoforms.
In conclusion, we have identified an autoregulatory domain within the N terminus of the sodium pump, which modulates conformational transitions, suggesting novel intramolecular interactions within the cytoplasmic domains of the enzyme. Studies are currently underway to further pinpoint the residues in the N terminus that are involved in these interactions.