Asymmetric Preorganization of Inverted Pair Residues in the Sodium-Calcium Exchanger

In analogy with many other proteins, Na+/Ca2+ exchangers (NCX) adapt an inverted twofold symmetry of repeated structural elements, while exhibiting a functional asymmetry by stabilizing an outward-facing conformation. Here, structure-based mutant analyses of the Methanococcus jannaschii Na+/Ca2+ exchanger (NCX_Mj) were performed in conjunction with HDX-MS (hydrogen/deuterium exchange mass spectrometry) to identify the structure-dynamic determinants of functional asymmetry. HDX-MS identified hallmark differences in backbone dynamics at ion-coordinating residues of apo-NCX_Mj, whereas Na+or Ca2+ binding to the respective sites induced relatively small, but specific, changes in backbone dynamics. Mutant analysis identified ion-coordinating residues affecting the catalytic capacity (kcat/Km), but not the stability of the outward-facing conformation. In contrast, distinct “noncatalytic” residues (adjacent to the ion-coordinating residues) control the stability of the outward-facing conformation, but not the catalytic capacity. The helix-breaking signature sequences (GTSLPE) on the α1 and α2 repeats (at the ion-binding core) differ in their folding/unfolding dynamics, while providing asymmetric contributions to transport activities. The present data strongly support the idea that asymmetric preorganization of the ligand-free ion-pocket predefines catalytic reorganization of ion-bound residues, where secondary interactions with adjacent residues couple the alternating access. These findings provide a structure-dynamic basis for ion-coupled alternating access in NCX and similar proteins.


FIGURE S1. 45 Ca 2+ -flux assays of Na + /Ca 2+ and Ca 2+ /Ca 2+ exchange reactions. (A)
The time course of Na + i -dependent 45 Ca 2+ -uptake was measured in vesicles containing either overexpressed WT or mutated NCX_Mj. The Na + -loaded (160 mM) vesicles were rapidly diluted (20-50-fold) in the assay medium containing 20 mM Mops/Tris, pH 6.5, 100 mM KCl, and 200 µM 45 CaCl 2 at 35°C and the Ca 2+ -uptake was stopped at the indicated times by injecting a cold EGTA-containing buffer. Quenched solutions were filtered and the radioactivity retained on the filters was assayed using a scintillation solution. (B) The initial rates (t = 5 s) of the Na + /Ca 2+ and Ca 2+ /Ca 2+ exchange reactions were measured by diluting the Na + -loaded (160 mM) or Ca 2+ -loaded (250 µM) vesicles in an assay medium containing 20-2000 µM 45 CaCl 2 . The exchange reactions were stopped and 45 Ca 2+ -uptake was measured as described above. (C) The equilibrium binding of 45 Ca 2+ to purified WT NCX_Mj was measured by using the protein ultrafiltration method 1,2 .
Briefly, an assay medium containing 30-50 µM protein with 100 mM KCl and 10 mM Tris-HCl at pH 7.2 was placed in the upper chamber of an Ultracel-3k concentrator and the assay was performed as previously outlined 1,2 . The concentrations of residual [Ca 2+ ] res in the decalcified assay buffers were measured by using the Fluo-3 assay 2 . The 45 Ca 2+binding titration curves were fit using GraFit software v7.1 (Erithacus Software, Ltd). The K d value was derived by fitting the experimental points obtained by 45 Ca 2+ -titration curves.

FIGURE S2. Quantitative assay of the intrinsic equilibrium (K int ) of bidirectional
Ca 2+ movements -According to the ping-pong mechanism, describing the unidirectional Na + /Ca 2+ exchange cycle or the bidirectional Ca 2+ /Ca 2+ exchange ( Figure 1E), the "apparent binding affinity" of Ca 2+ (K" m or K' m ) at opposite sides of the membrane can be represented as a function of the intrinsic binding constant (K" Ca , K' Ca ) and of the iontranslocation rate constants (l',l",k',k") (Table S1). According to this formalism, the K' m /K" m ratio of the Ca 2+ /Ca 2+ exchange represents the intrinsic equilibrium (K int ) for bidirectional Ca 2+ movements (Equation 1), which is related to Ca 2+ binding at opposite sides of the membrane 4,5 .
According to this formalism, the experimentally measured values of cytosolic K' m =28.2±3.5 µM and extracellular K" m = 183.9±27.7 µM of the Ca 2+ /Ca 2+ exchange ( Figure   3B) result in a K int value of 0.15±0.05 (n=5), thereby revealing an intrinsic asymmetry of bidirectional Ca 2+ movements favoring the extracellular access. By assuming that Ca 2+ binding site of NCX_Mj (S Ca ) has a comparable intrinsic affinity for Ca 2+ binding (K' Ca ≈ K" Ca ) at the cytosolic and extracellular sides, the l' values may be at least five times faster than l". The observed values of V' max and V'' max for Ca 2+ /Ca 2+ exchange are comparable ( Figure   S3) as expected for the ping-pong mechanism ( Figure 1E). were covered by three and eight peptides, respectively. A small portion of TM9 was also found (peptide 259-264). Furthermore, four couples of overlapping peptides (possessing a common extremity) enabled us to locally increase the resolution, giving information on small segments (2-7 amino acids). Interestingly, the interface between the two invertedtopology repeats was covered, including the four putative ion-binding sites.

FIGURE S4
: Deuterium uptake plots for peptides which are not presented in Figure 2C.
Data are presented as the mean ± SD (n=3).