Structure-based dynamic arrays in regulatory domains of sodium-calcium exchanger (NCX) isoforms

Giladi, Moshe; Lee, Su Youn; Ariely, Yarden; Teldan, Yotam; Granit, Rotem; Strulovich, Roi; Haitin, Yoni; Chung, Ka Young; Khananshvili, Daniel

doi:10.1038/s41598-017-01102-x

Download PDF

Article
Open access
Published: 20 April 2017

Structure-based dynamic arrays in regulatory domains of sodium-calcium exchanger (NCX) isoforms

Scientific Reports volume 7, Article number: 993 (2017) Cite this article

2540 Accesses
23 Citations
1 Altmetric
Metrics details

Subjects

Abstract

Mammalian Na⁺/Ca²⁺ exchangers, NCX1 and NCX3, generate splice variants, whereas NCX2 does not. The CBD1 and CBD2 domains form a regulatory tandem (CBD12), where Ca²⁺ binding to CBD1 activates and Ca²⁺ binding to CBD2 (bearing the splicing segment) alleviates the Na⁺-induced inactivation. Here, the NCX2-CBD12, NCX3-CBD12-B, and NCX3-CBD12-AC proteins were analyzed by small-angle X-ray scattering (SAXS) and hydrogen-deuterium exchange mass-spectrometry (HDX-MS) to resolve regulatory variances in the NCX2 and NCX3 variants. SAXS revealed the unified model, according to which the Ca²⁺ binding to CBD12 shifts a dynamic equilibrium without generating new conformational states, and where more rigid conformational states become more populated without any global conformational changes. HDX-MS revealed the differential effects of the B and AC exons on the folding stability of apo CBD1 in NCX3-CBD12, where the dynamic differences become less noticeable in the Ca²⁺-bound state. Therefore, the apo forms predefine incremental changes in backbone dynamics upon Ca²⁺ binding. These observations may account for slower inactivation (caused by slower dissociation of occluded Ca²⁺ from CBD12) in the skeletal vs the brain-expressed NCX2 and NCX3 variants. This may have physiological relevance, since NCX must extrude much higher amounts of Ca²⁺ from the skeletal cell than from the neuron.

Structural dynamics of Na+ and Ca2+ interactions with full-size mammalian NCX

Article Open access 16 April 2024

Moshe Giladi, Lukáš Fojtík, … Daniel Khananshvili

Structural mechanisms of the human cardiac sodium-calcium exchanger NCX1

Article Open access 04 October 2023

Jing Xue, Weizhong Zeng, … Youxing Jiang

Structure, mechanism and lipid-mediated remodeling of the mammalian Na+/H+ exchanger NHA2

Article Open access 16 February 2022

Rei Matsuoka, Roman Fudim, … David Drew

Introduction

The properties of Ca²⁺-binding proteins that participate in Ca²⁺-dependent regulation of distinct cell-signaling pathways are controlled by tissue-specific expression and alternative splicing of closely related genes¹. The isoform-dependent and alternative splicing-dependent modification of Ca²⁺-binding regulatory domains are especially important for Ca²⁺-transporting proteins operating in excitable tissues^1,2,3, since feedback interactions of Ca²⁺ with allosteric regulatory domains dynamically modulate the Ca²⁺-transport rates in accordance with dynamic changes in cellular Ca²⁺ oscillations (e.g., in cardiomyocytes during the action potential)^{1, 3,4,5}. Despite the fundamental importance of this issue, the structure-dynamic determinants governing the isoform-dependent and alternative splicing-dependent modification of regulatiory specificity in Ca²⁺-binding domains remain poorly understood^{2, 6}.

Na⁺/Ca²⁺ exchanger (NCX) proteins serve as a major route for Ca²⁺ extrusion from cells^{3, 4}. In mammals, three gene isoforms (SLC8A1, SLC8A2, and SLC8A3) of NCX proteins (NCX1, NCX2, and NCX3) and their splice variants are expressed in a tissue-specific manner, while exhibiting distinct regulatory phenotypes (Fig. 1)^7,8,9,10. The activity of NCX proteins is “secondarily” modulated by cytosolic Na⁺ and Ca²⁺ ^8,9,10. In mammals, an increase in cytosolic [Na⁺] usually inactivates NCX, whereas an increase in cytosolic [Ca²⁺] usually activates NCX^{3, 4, 6}. However, the appearance of these regulatory modes as well as their strength and duration vary in an isoform and splice variant-dependent manner, since the regulatory features of a given isoform/splice variant must match the tissue-specific contributions of NCX to the dynamic handling of Ca²⁺ homeostasis in distinct cell-types^{7, 9, 11}. For example, in some isoforms and splice variants, an increase in the cytosolic [Ca²⁺] can alleviate Na⁺-dependent inactivation, whereas in others, Ca²⁺ cannot alleviate Na⁺-dependent inactivation, or the regulatory mode of Na⁺-dependent inactivation does not exist at all^{7, 9, 11, 12}.

NCX1 is universally distributed, practically in every mammalian cell, although the NCX1 splice variants are selectively expressed in a tissue-specific manner (Fig. 1A)^{8, 10}. Alternative splicing of NCX1 results in at least 17 splice variants, where either exon A or exon B (mutually exclusive exons) can be freely combined with cassette exons (C, D, E, and F) (Fig. 1A)^{3, 4, 8, 10}. In general, exon-A-containing NCX1 splice variants are mainly expressed in excitable tissues (e.g., cardiomyocytes, neurons), whereas exon-B-containing NCX1 splice variants are mainly expressed in non-excitable tissues (e.g., pancreas, kidney) (Fig. 1A)^{3, 4, 8, 10}. NCX2 is predominantly expressed in the brain and spinal cord, but it can also be found in the gastrointestinal tract and kidney tissues. NCX2 does not undergo alternative splicing (Fig. 1B). NCX3 is predominantly expressed in the brain and skeletal muscle, but it can also be found in osseous tissue and the immune system^{8, 10, 11, 13}. Alternative splicing of NCX3 results in at least 5 splice variants and involves the expression of either exon A or exon B (mutually exclusive exons) in the presence or absence of an additional exon C (Fig. 1C)^{8, 10, 11, 13}. Exon-A-containing NCX3 splice variants are mainly expressed in skeletal muscle, whereas exon-B-containing NCX3 splice variants are mainly expressed in the brain (Fig. 1C)¹¹.

Prokaryotic and eukaryotic NCX proteins contain ten transmembrane helices (TM1–10), where the ion-binding pocket selectively binds either one Ca²⁺ or three Na⁺ ions and transports them in separate steps^14,15,16,17. Eukaryotic (but not prokaryotic) NCX proteins contain a large cytosolic loop between TM5 and TM6, mainly composed of two Ca²⁺-binding domains, CBD1 and CBD2¹⁸. The two CBDs form a head-to-tail attached two-domain tandem (termed CBD12) through a very short interdomain linker (Fig. 1)^{18, 19}. The Ca²⁺-dependent activation of NCX1 results from Ca²⁺ binding to CBD1¹², whereas the Ca²⁺-dependent alleviation of Na⁺-induced inactivation results from Ca²⁺ binding to CBD2^{12, 20, 21}. Notably, the Na⁺-dependent inactivation of NCX is due to Na⁺ interaction with the ion-transport domains of NCX and not with CBD12^{22, 23}, although Ca²⁺ binding to NCX1-CBD2 somehow relieves the Na⁺-induced inactivation via propagation of an allosteric signal over a large distance from CBD12 to transport domains⁶.

NMR¹⁸ and X-ray crystallography^{20, 24} studies revealed that CBD1 and CBD2 share a similar structure, exhibiting an immunoglobulin-like β-sandwich structure with seven antiparallel β-strands (Fig. 1). CBD1 contains four Ca²⁺ binding sites, capable of high-affinity Ca²⁺ sensing (K_d < 1 µM) in all NCX orthologs, isoforms, and splice variants^{6, 21, 24,25,26}. The Ca²⁺ binding sites of CBD1 are located at its C-terminal tip, near the CBD1-CBD2 interface (Fig. 1)^{19, 24}. In contrast, the number of Ca²⁺ binding sites at CBD2 varies from zero to three, depending on the specific ortholog, isoform, and splice variant (Fig. 1)^{21, 25, 26}. The affinity for Ca²⁺ binding at CBD2 is rather low, ranging from ~5 to ~200 μM^{21, 23, 25, 26}. Notably, alternative splicing of NCX (described in detail above) takes place only at CBD2 and involves the Ca²⁺ binding sites^{18, 21}. Therefore, in A-exon-containing NCX1 variants, CBD2 binds two Ca²⁺ ions, whereas in B-exon-containing NCX1 variants, CBD2 does not bind Ca²⁺ ions (Fig. 1A)^{21, 25}. Since Ca²⁺ binding to CBD2 alleviates Na⁺-induced inactivation, this effect of Ca²⁺ is only observed in A-exon-containing NCX1 variants^{7, 12, 21}. In NCX2, CBD2 binds only one Ca²⁺ ion with very low affinity (Fig. 1B)^{21, 26}. Since NCX2 does not exhibit any Na⁺-dependent inactivation, the Ca²⁺-dependent alleviation of Na⁺-induced inactivation is functionally irrelevant with NCX2⁹. In NCX3, A-exon-containing variants do not bind Ca²⁺ ions at CBD2, whereas B-exon-containing variants bind three Ca²⁺ ions at CBD2 (Fig. 1C)²⁶. Intriguingly, NCX3-AC (which does not bind Ca²⁺ at CBD2) was shown to exhibit less Na⁺-dependent inactivation compared with NCX3-B (which does bind Ca²⁺ at CBD2) in electrophysiological studies¹¹.

Recent studies using diverse biophysical, biochemical, and structural methods have identified synergistic interactions between the CBDs that shape the dynamic range and kinetic features of Ca²⁺-dependent regulation in splice variants of NCX1, thereby providing the insights into the mechanisms underlying CBD interactions in isolated CBD12 and full-size NCX1^{18,19,20,21, 24,25,26,27,28,29,30,31,32}. These studies have shown that the isolated preparations of CBD12 largely represent the Ca²⁺ sensitivity and kinetics of Ca²⁺-dependent regulation in their matching of full-size NCX isoforms and splice variants^{7, 25, 30}. Therefore, isolated CBD12 serves as an ideal model for investigating the structure-dynamic mechanisms underlying the allosteric regulation specificities in NCX isoforms and splice variants. The interactions between CBD1 and CBD2 in the context of CBD12 results in increased affinity for Ca²⁺ at CBD1 and slow dissociation of an “occluded” Ca²⁺ ion from CBD1^{25, 27}. In addition, alternative splicing at CBD2 not only affects the Ca²⁺ binding affinity and capacity at CBD2—it also affects the Ca²⁺ interactions with CBD1, establishing up to 50-fold differences both in the Ca²⁺ binding affinity and Ca²⁺ off-rates in the cardiac (ACDEF), brain (AD), and kidney (BD) splice variants of NCX1^{25, 27}. Using ⁴⁵Ca²⁺ equilibrium binding and stopped-flow kinetic assays, we have recently demonstrated that NCX2 and NCX3 also exhibit alternative-splicing dependent synergistic interactions between the CBDs, similarly to NCX1²⁶. NCX3-CBD12 splice variants exhibit similar Ca²⁺ affinity at CBD1, but the dissociation rate of the occluded Ca²⁺ ion is ~10-fold slower in skeletal muscle (AC) than in the brain (B) variants²⁶. These differences in Ca²⁺ binding affinity and the Ca²⁺ off-rates at the primary allosteric sensor (CBD1) may have physiological relevance for diversifying regulatory responses in full-size NCX variants expressed in a tissue-specific manner^{1,2,3, 6}.

X-ray crystallography revealed that Ca²⁺ binding to CBD1 results in Ca²⁺ occlusion and tethering of CBD1 and CBD2 through the formation of a hydrogen-bonded salt-bridge network at the two-domain interface, yielding a more stable structure^{6, 19}. The crystallographic structures of CBD12 variants of NCX1, exhibiting different regulatory properties, show nearly identical interdomain angles between CBD1 and CBD2 and thus, global CBD alignment does not seem to shape regulatory specificities in NCX1 splice variants^{6, 19}. In line with this notion, small-angle X-ray scattering (SAXS) studies revealed that Ca²⁺ binding and occlusion at CBD1 of NCX1-CBD12 splice variants induces a shift towards narrowly distributed elongated conformations, as predicted by the population shift mechanism²⁸. This population shift of conformational states is essential for regulation, thereby representing a common mechanism for NCX1 splice variants regardless of their regulatory specificity²⁸. Finally, hydrogen-deuterium exchange mass-spectrometry (HDX-MS) analyses of NCX1-CBD12 splice variants have shown that Ca²⁺ binding to all tested variants mainly rigidifies the backbone dynamics of CBD2 (and not that of CBD1), where the alternative splicing of CBD2 secondarily modifies the strength and expansion of rigidification throughout CBD2^{29, 31}. Importantly, the Ca²⁺-induced effects on CBD2 backbone dynamics correlate well with the regulatory specificity found in the matching splice variants of full-size NCX1s^{29, 31}.

Thus far, NCX1 splice variants share a common mechanism for the initial decoding of the regulatory signal upon Ca²⁺ occlusion and tethering of CBDs (i.e., through the population shift mechanism), whereas alternative splicing of CBD2 controls the propagation and strength of Ca²⁺-dependent rigidification and thereby, diversifies the regulatory response^{28, 29, 31}. Currently it is unclear to what extent (if at all) the above-described mechanisms are also valid for NCX2 and NCX3 splice variants and how the exons control the CBD dynamics as related to shaping the regulatory specificity in a given isoform/splice variant. To overcome this gap, here we analyzed the apo and Ca²⁺-bound forms of isolated NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B proteins with SAXS^{19, 28, 29} and HDX-MS^{29, 31} in order to identify their conformational dynamics and the effect of alternative splicing. The present analysis reveals a general mechanism for regulatory coupling associated with Ca²⁺ interactions with regulatory CBD domains of NCX1, NCX2, and NCX3 isoforms and their splice variants, where the dynamic effects of alternative splicing on conformational dynamics of the two-domain regulatory tandem (CBD12) control the contributions of a given isoform/splice variant to dynamic handling of cytosolic Ca²⁺ oscillations in distinct cell types.

Results

Global SAXS parameters of CBD12 proteins

To determine the low-resolution structure-dynamic features of NCX-CBD12 isoforms in solution as well as to resolve the effect of ligand binding on the global conformational states, SAXS was used (Fig. 2, Table 1)³³. In all the examined proteins (NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B), the Ca²⁺-bound form exhibited similar R_g and D_max values, with PDDF, which represents an elongated conformation (Table 1, Fig. 2B,E,H). The Porod volumes of all the Ca²⁺-bound proteins were similar and matched the volume predicted for a monomeric protein, as previously shown for NCX1-CBD12 (Table 1)²⁸. However, in the apo form, the Porod volume matched a volume that is intermediate between a dimer and a monomer (Table 1). Therefore, we pursued further modelling only for the Ca²⁺-bound forms. All three proteins examined here, display similar, elongated molecular envelopes (Fig. 2C,F,I). The CBD12 homology models, based on the crystal structure of NCX1-CBD12-E454K, fit well in the molecular envelopes in all proteins examined here (Fig. 2C,F,I). Taking this into account, along with the low normalized shape discrepancy (NSD) of the models (Table 1) and the features observed in the PDDFs of all three proteins in the Ca²⁺-bound state (Fig. 2B,E,H), it can be assumed that NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B exhibit a very similar elongated conformation in a relatively rigid state in solution³³.

Table 1 .

Full size table

Ensemble optimization method (EOM) SAXS analysis of CBD12 proteins

To further assess the conformational distributions of the examined proteins, we performed EOM analysis (Fig. 3). The advantage of EOM analysis is that it can describe the conformational distribution of a molecule instead of representing the molecule using global parameters that do not account for conformational heterogeneity^{28, 34}. The selected ensembles fit the experimental data well, as judged by the low χ² values (Table 1). In all three proteins tested here (NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B), the conformational distributions are narrow (Fig. 3B,E,H) and biased towards the most elongated conformations (Fig. 3C,F,I), thereby implying a rigid and elongated conformation of Ca²⁺-bound species for all CBD12 variants in solution. Notably, the effect of Ca²⁺ results from binding to CBD1 and not CBD2, since NCX3-CBD12-AC exhibits these properties despite the fact that it does not bind Ca²⁺ at CBD2²⁶. These features of NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B are somewhat similar to mammalian NCX1-CBD12 splice variants analyzed previously²⁸. Taken together, the data supports the notion that Ca²⁺-binding to the primary (high-affinity) sensor of the CBD12 tandem either in NCX1, NCX2, or NCX3 results in a Ca²⁺-induced population shift of preexisting conformational states, where more elongated conformational states become more populated (Fig. 3)²⁸.

Effect of alternative splicing on HDX profiles in NCX3-CBD12

The SAXS analysis, described above, revealed similar behavior in all NCX-CBD12 isoforms and splice variants tested here, exhibiting a rigid and elongated conformation in the presence of Ca²⁺ (Figs 2 and 3). Therefore, it remains unclear how Ca²⁺-dependent NCX regulation is diversified between the different isoforms and splice variants. To better understand the differential conformational dynamics of NCX-CBD12 isoforms and splice variants, the isolated preparations of NCX2-CBD12 and NCX3-CBD12 proteins were analyzed in the apo and Ca²⁺-bound states by HDX-MS^{29, 31}.

The identified peptic peptides that were used for HDX-MS analysis are shown in Figure S1. The deuterium uptake levels of each protein in the apo state (Figure S2A,C, and E) or in the Ca²⁺-bound state (Figures S2B,D, and F) are overlaid on the protein structures as color-coded heat maps. It is not feasible to directly compare the deuterium exchange levels between proteins with different sequences because amino acid side chains affect the deuterium uptake rates and different peptic peptides are generated. Therefore, a comparison between NCX2-CBD12 and NCX3-CBD12 variants is not feasible, whereas a comparison between NCX3-CBD12-AC and NCX3-CDB12-B is possible within regions with the same sequences (i.e., outside the alternative splicing regions).

To characterize the effect of alternative splicing on the structural dynamics of NCX3-CBD12, we compared the deuterium exchange levels of NCX3-CBD12-AC and NCX3-CDB12-B using peptic peptides denoted as blue bars in Figure S1B. In both the apo and Ca²⁺-bound states, a few regions of the backbone dynamics considerably differ (Fig. 4) despite the predicted structural similarity and the similar global behavior of the analyzed proteins in solution (Figs 1, 2 and 3). In the apo state, NCX3-CBD12-AC exhibits a higher deuterium uptake than does NCX3-CBD12-B in CBD1, CBD1-CBD2 linker, and CBD2 (Fig. 4A, red regions). On the other hand, the CBD1 EF loop (peptide 468–474), which participates in forming the Ca²⁺-binding sites and interacts with the α-helical region of CBD2¹⁹, exhibits less uptake in NCX3-CBD12-AC as compared with NCX3-CBD12-B (Fig. 4A, blue region). This result reveals that the alternative splicing segment (Fig. 1, located on CBD2) affects the folding stability of CBD12 in the apo state and thus, may predefine distinct conformational responses of CBDs to Ca²⁺-binding in a given splice variant^{6, 29, 31}.

Ca²⁺ binding alters the backbone dynamic differences between the two NCX3-CBD12 variants (Fig. 4B). The deuterium uptake levels within the CBD1-CBD2 linker and nearby regions of CBD2 become similar between NCX3-CBD12-AC and NCX3-CDB12-B (Fig. 4B, yellow regions). The CBD1-CBD2 linker is rigidified to a similar extent in both splice variants (Fig. 4B), consistent with similar conformational distributions observed in the EOM SAXS analysis (Fig. 3); the NCX3-CBD12-AC linker is no longer more dynamic than the NCX3-CBD12-B linker upon Ca²⁺ binding.

The α-helical region of CBD2 [peptides 646–655 (AC) or 639–648 (B)], adjacent to the CBD1 Ca²⁺-binding sites, exhibits markedly lower deuterium uptake in NCX3-CBD12-AC as compared with NCX3-CBD12-B in the Ca²⁺-bound state (Fig. 4B). In addition, the C-terminal tip of CBD2 [peptides 538–548, 534–546, 662–669 (AC) and 655–662 (B)] exhibits a lower deuterium uptake in NCX3-CBD12-AC compared with NCX3-CBD12-B in the Ca²⁺-bound state (Fig. 4B). This finding is especially peculiar in light of a previous analysis demonstrating that NCX3-CBD12-AC lacks the capacity for Ca²⁺ binding at CBD2, whereas NCX3-CBD12-B binds three Ca²⁺ ions at the C-terminal tip of CBD2^{30, 34}. Overall, the effect of Ca²⁺ binding on backbone rigidification is more marked in NCX3-CBD12-AC as compared with NCX3-CBD12-B (Fig. 4), which may account for the observed differences in the dissociation kinetics of occluded Ca²⁺ at the interface of CBD1 and CBD2 in NCX3 splice variants²⁶.

HDX profiles of NCX2-CBD12 and NCX3-CBD12

Although a direct comparison of deuterium uptake levels between NCX2-CBD12 and NCX3-CBD12 variants is not feasible, we can compare the regions affected by Ca²⁺ binding.

To better envisage the Ca²⁺-dependent changes in backbone dynamics, we plotted the difference in deuterium uptake as a function of residue at different deuterium uptake durations, and the Ca²⁺-induced deuterium uptake changes were also color-coded onto the CBD12 model structures (Figs 5 and S3). When examining the effect of Ca²⁺-binding, all three tested proteins exhibit markedly reduced deuterium uptake throughout CBD1 and CBD2 (Figs 5 and S3). This is true even for NCX3-CBD12-AC, which does not bind Ca²⁺ at CBD2²⁶. This supports the notion that occupation of the high-affinity primary sensor at CBD1 in NCX2 and NCX3 plays a key role in Ca²⁺-dependent stabilization of CBD2 due to conformational stabilization of the two-domain interface and the interdomain linker upon Ca²⁺ binding and occlusion at the primary sensor at CBD1. A similar structure-dynamic module was suggested for NCX1-CBD12 splice variants based on the HDX-MS analysis^{29, 31}.

The interdomain CBD1-CBD2 linker (peptide 513–525 in NCX2 and 524–533 in NCX3) exhibits lower deuterium uptake upon Ca²⁺-binding in all three proteins analyzed here (Figs 5 and S3). This evidence is consistent with the rigid conformations observed in the EOM SAXS analysis (Fig. 2). Similarly to NCX1-CBD12 splice variants^{29, 31}, the presently analyzed NCX2 and NCX3 splice variants also exhibit the rigidification of the two-domain interface (including the CBD1-CBD2 linker) upon Ca²⁺ binding, meaning that all NCX1, NCX2, and NCX3 isoform/splice variants, tested until now, share a common mechanism for the initial decoding of regulatory Ca²⁺ signals.

Notably, in all the examined proteins, the α-helical region at the domain’s interface (in the vicinity of the alternative splicing segment, Fig. 1) (peptide 639–647 in NCX2 and 646–655 in NCX3) exhibits lower deuterium uptake in the Ca²⁺-bound state (Figs 5 and S3). This could be important for exon-controlled stabilization of Ca²⁺-dependent CBD tethering, which in turn, affects the strength of Ca²⁺ binding and the off-rates of occluded Ca²⁺ dissociation from CBD1.

Discussion

Functional contributions of NCX2 and NCX3 variants to physiological and pathological settings have been documented^{13, 34}, while underscoring the importance of tissue-specific expression and regulatory diversities exhibited by these isoforms and splice variants^{3, 4, 6, 9, 11}. The goal of the present work was to determine the structure-dynamic basis underlying the previously reported regulatory diversities in an isolated CBD12 model and full-size NCX variants in order to formulate the principal mechanism underlying the regulatory diversity in NCX proteins. To this end, the NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B proteins were analyzed here by using SAXS and HDX-MS techniques.

As summarized above, the diversity in allosteric regulation of NCX proteins is puzzling at first glance. Some basic mechanisms seem to apply to all isoforms and splice variants, that is, the synergistic interactions between CBD1 and CBD2 are manifested as increased affinity and slower Ca²⁺ dissociation kinetics at CBD1^{25, 26}. On the other hand, these interactions are clearly modified by alternative splicing to fulfill specific physiological needs in distinct cell types^{25, 26}. In addition, distinct NCX variants exhibit different regulatory responses to Ca²⁺ and Na⁺-dependent regulation despite having high homology and predicted structural similarity⁶. For example, in NCX1, variants that are capable of Ca²⁺ binding at CBD2 exhibit Ca²⁺-dependent alleviation of Na⁺-induced inactivation, whereas in NCX3, NCX3-CBD12-AC, which does not bind Ca²⁺ at CBD2, exhibits less Na⁺-induced inactivation as compared with NCX3-CBD12-B^{11, 26}. NCX2 differs from the rest NCX variants in that that it does not undergo alternative splicing and it does own to the Na⁺-induced inactivation. The results presented here, taken together with extensive previous studies of NCX1 splice variants^{6, 18, 19, 21, 25, 28, 29, 31, 32}, may provide a conceptual framework to resolve a common mechanistic platform and disparities owned by distinct isoform-splice variants.

First and foremost, CBD1 and CBD2 act as Ca²⁺ sensors. Therefore, their binding affinity and kinetics must correlate with the dynamic properties of Ca²⁺ signaling in a given tissue^{1, 2}. In this respect, the importance of alternative splicing is straightforward: NCX1 (NCX1-AD) and NCX3 (NCX3-B, NCX3-BC) splice variants expressed in the brain, as well as NCX2, exhibit faster Ca²⁺ dissociation kinetics from CBD12 compared with NCX1 and NCX3 splice variants expressed in muscles (NCX1-ACDEF in cardiac muscle and NCX3-AC in skeletal muscle)^{25, 26}. This is in line with the prolonged Ca²⁺ transients and action potential durations in muscles compared with neurons^{4, 5, 26}. These hallmark features, assigned to Ca²⁺ interactions with CBD1 and CBD2, strictly correlate with characteristic regulatory modes ascribed to matching (full-size) NCX variants^{4, 7, 9, 11, 13, 25, 26}. The equilibrium and kinetic parameters measured in isolated NCX1-CBD12-ACDEF preparations²⁵ were successfully applied to computer-aided dynamic modeling of allosteric activation of the cardiac NCX1 (ACDEF) during the action and resting potentials, thereby providing indispensable information on the dynamic contributions of NCX in handling the Ca²⁺-dependent Ca²⁺-release from the sarcoplasmic reticulum, the Ca²⁺-spark appearance, and duration within local sub-cellular compartments and the fine-tuning of excitation-contraction coupling in cardiomyocytes³⁵. Similar applications for computer-aided modeling are straightaway required for clarifying the partial contributions of multiple NCX isoforms and their splice variants to the dynamic handling of dynamic Ca²⁺ swings in neuronal tissues. The emerging working hypothesis is that specific conformational variances, raised by exons, diversify the affinity/off-rates of Ca²⁺ at CBD1 (for controlling the Ca²⁺-dependent allosteric activation) and the affinity/capacity of Ca²⁺ binding to CBD2 (for controlling the Ca²⁺-dependent alleviation of Na⁺-induced inactivation)^{25, 26}.

In NCX1, both the brain (AD) and cardiac muscle (ACDEF) splice variants bind Ca²⁺ at CBD2, to alleviate Na⁺-induced inactivation. The ability to counteract Na⁺-induced inactivation is of major importance in excitable tissues where large Na⁺ fluxes (through the voltage-sensitive Na⁺-channels) occur at the initial stage of cell depolarization. NCX2 (expressed in the brain) lacks the Na⁺-induced inactivation, whereas in NCX3 the Na⁺-induced inactivation can be counteracted by increased levels of cytosolic Ca²⁺, but to varying degrees^{9, 11}. It was previously proposed that an electrostatic switch at CBD2 underlies the alleviation of Na⁺-induced inactivation upon Ca²⁺ binding, but this proposal cannot account for the regulatory properties of NCX2 (where the Ca²⁺ binding affinity at CBD2 is extremely low) or for regulatory features of NCX3-AC^{9, 11, 21, 26}. However, the data presented here may resolve the principal structural platform for the dynamic regulation of NCXs, as well as it can pinpoint the exon-governed fine shaping in the backbone dynamics that may account for regulatory diversity.

In recent years, a growing body of evidence has suggested an updated view of allosteric regulation in proteins. More specifically, a greater importance is given to changes in protein dynamics and conformational distributions rather than to global conformational transitions^{36, 37, 38,39,40,41,42}. The accumulating evidence raises several important points. Protein rigidity seems to serve as a common communication route between distant sites in proteins, resulting in structural coupling between the regulatory regions (e.g., CBD12) and the effector site (e.g., the ion-transport domain)⁴². Moreover, in some instances, allosteric interactions are mediated only by dynamic changes, without having any gross effects on the protein conformation^{36, 38, 39}. In other cases, the population shift mechanism describes a shift in the protein ensemble of conformations towards active states, where no obvious global conformational change is observed³⁷. These observations were recently formulated into a unified model of allostery, based on thermodynamics, the structure, and the free energy landscape of the population shift⁴¹. According to this general model, the allosteric mechanism in a specific protein requires three elements⁴¹, allowing us to formulate a general scheme for NCX isoforms and regulation of splice variants.

First, the active conformation responsible for allosteric regulation is required⁴¹. This conformation is provided by the crystal structure of NCX1-CBD12-AD-E454K¹⁹, which also allows us to model NCX2-CBD12 and NCX3-CBD12 in the Ca²⁺-bound conformation. The molecular envelopes calculated here (Fig. 2) verify the relevance of these homology models to the global conformations of Ca²⁺-bound CBD12 isoforms. The EOM analyses performed previously for NCX1-CBD12 and performed here for NCX2-CBD12 and NCX3-CBD12 are in line with the unified model, whereby the allosteric event (Ca²⁺ binding) does not create any new conformational states but instead, only shifts the population among the existing states, towards more stable elongated conformations (Fig. 3)²⁸.

Second, a set of interacting residues is required⁴¹. This set of residues is clearly depicted in the crystal structure of NCX1-CBD12-AD-E454K, where a network of hydrogen-bonded salt bridges and hydrophobic interactions between the α-helical region of CBD2 and F450 from NCX1 form the interdomain interface¹⁹. Notably, the two-domain interface is extremely well conserved among all isoform splice variants, so the use of a crystal structure of NCX1-CBD12-AD-E454K as a universal model is highly justified. According to this model, the mutations of interacting residues must uncouple the regulatory function of CBDs. This prediction has been experimentally demonstrated by showing a lack of Ca²⁺ occlusion and reduced binding affinity of Ca²⁺ in several mutations of residues contributing to the stabilization of the two-domain interface, including the interdomain linker (e.g., G503), the hydrophobic core (e.g., F450), and salt bridge elements (e.g., R532 or D565) tethering CBDs upon Ca²⁺ binding^{6, 19, 23,24,25,26,27,28,29}. For example, the structural evidence revealed by HDX-MS demonstrates the reduced Ca²⁺-dependent CBD2 rigidification in the F450G mutant, thereby confirming the general model for Ca²⁺-dependent tethering of CBDs and the rigidification of the two-domain interface^{19, 29}. Consistent with this statement, this residue is highly conserved among NCX orthologs¹⁹ and F472 in NCX2-CBD12 and F474 in NCX3-CBD12 (corresponding to F450 in NCX1-CBD12) show reduced deuterium uptake upon Ca²⁺ binding (Figs 4 and 5). Moreover, the accumulating HDX-MS data on NCX1-CBD12, NCX2-CBD12, and NCX3-CBD12 proteins depict the allosteric stabilization within CBD2 upon Ca²⁺ binding at CBD1 (Figs 4 and 5) through the population shift mechanism, shared by all tested isoform/splice variants^{6, 28, 29, 31}. Thus, the CBD2 isoforms and splice variants dynamically shape the initial decoding of regulatory signal upon Ca²⁺ binding and thus, the signal message undergoes a transformation toward further propagation of allosteric signal. The underlying mechanisms of signal propagation from the CBDs to the target sites of ion-transport domains remain to be discovered.

Finally, one should consider whether ligand binding at the regulatory site stabilizes the active conformation, destabilizes the inactive conformation, or does both simultaneously⁴³. This question is not fully resolved with NCX proteins because a structure of apo-CBD12 is lacking. However, EOM studies of NCX1-CBD12 splice variants show that the active conformations are also sampled in the absence of Ca²⁺, but with a lower frequency²⁸. The Ca²⁺-bound crystal structure also demonstrates how the presence of Ca²⁺ “locks” the two domains in an elongated conformation¹⁹. Thus, the Ca²⁺-dependent stabilization of the one or more “active conformation states” can explain (at least partially) the mechanism by which the ligand binding modulates the activity of NCX proteins.

Having this general mechanism in mind, it can be concluded that different NCX isoforms and splice variants share a similar “active” conformation, which becomes highly populated upon Ca²⁺ binding (Figs 2 and 3)^{19, 28, 29}. A very limited number of major “conformational state(s)” display some clear and highly conserved interdomain interactions between residues, although the complete dynamic response in CBD2 is different among distinct isoforms and splice variants^{29, 31}. The present (Figs 3, 4 and 5) and previous HDX-MS analyses of the different CBD12 proteins^{29, 31} underscore a general concept according to which the rigidity and allosteric capacity strictly correlate⁴². Intriguingly, the disparate isoforms have different means to control the rigidity of CBD2. That is, in NCX1, alternative splicing using A-exon acts by allowing Ca²⁺ binding at CBD2, which in turn, increases the rigidity of CBD2^{25, 29}. In NCX3, the mere presence of exon A results in increased response to regulatory Ca²⁺ binding at CBD1 (Fig. 4). The concept described here is further supported from previous analyses of CBD12 splice variants from CALX, a Drosophila melanogaster NCX ortholog, which exhibits anomalous Ca²⁺-dependent regulation. In line with the general mechanism described here, Ca²⁺ binding to CBD1 in CALX1.1 does not rigidify CBD2 and indeed, the intact (full-size) CALX1.1 responds to the regulatory Ca²⁺ binding in a negative mode, thereby exhibiting inhibition rather than activation²⁹.

Comparison of HDX-MS profiles of NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B (Figs 4, 5 and S2) revealed that in NCX3, the B and AC exons differentially stabilize CBD1 folding in the apo state; thus NCX3-CBD12-B is less dynamic than is NCX3-CBD12-AC. However, these differences are not prominent in the Ca²⁺-bound states, meaning that Ca²⁺ binding results in greater rigidification of NCX3-CBD12-AC and therefore the apo form predefines the incremental effect of Ca²⁺ binding. The greater dynamic effect may explain the reduced Na⁺-dependent inactivation. Similarly, in NCX1-CBD12 proteins, A-exon-containing variants in which Ca²⁺ can alleviate Na⁺-dependent inactivation exhibit greater rigidification of CBD2 (aided by Ca²⁺-binding). Therefore, the observed deviations in the conformational dynamics of the NCX1 and NCX3 splice variants (Fig. 4) adequately match the CBD2-controlled regulatory variances observed in full-size NCX1 and NCX3 splice variants^{29, 31}. Although NCX2-CBD12 cannot be directly compared to NCX1-CBD12 and NCX3-CBD12, it also undergoes significant rigidification upon Ca²⁺ binding (Fig. 4).

In general, the Ca²⁺-bound forms of all the tested variants of NCX2-CBD12 and NCX3-CBD12 exhibit more rigid backbone dynamics in comparison with the respective apo-forms (Fig. 5). The Ca²⁺-dependent effects are especially remarkable at the α-helical region of CBD2, which from one side is in proximity with the high-affinity sensor of CBD1 (at the two-domain interface) and on the other side, neighbors the splicing segment at CBD2 (Fig. 1). This mode of Ca²⁺-dependent conformational modulation is consistent with rigidification of the two-domain interface upon Ca²⁺-binding to CBD1, where the tethering of CBDs (through the formation of an interdomain salt-bridge network) stabilizes Ca²⁺ occlusion at high-affinity sites of CBD1, thereby allowing a further transformation and propagation of the allosteric signal. Similar conclusions were drawn for Ca²⁺-dependent rigidification of the α-helix in NCX1-CBD12 variants²⁹. Therefore, accumulating evidence strongly supports a general mechanism for the examined NCX1, NCX2, and NCX3 variants, according to which Ca²⁺-binding stabilizes the interdomain linker, the hydrophobic interface between the CBDs, and the nearby α-helix/FG-loop segment on CBD2 next to the splicing segment^{9, 25, 28, 29, 31}. It is quite obvious that the Ca²⁺-dependent rigidification of the two-domain interface detected by HDX-MS and SAXS techniques is coupled with the Ca²⁺ occlusion and two-domain tethering of CBDs, which further stabilizes the active conformation in all tested isoform/splice variants of CBD12 proteins^{19, 23, 25,26,27,28,29,30,31,32, 44}. The experimental methods described here may be used in future studies to resolve the role of alternative splicing in shaping the allosteric regulation of many protein families.

Methods

Overexpression and purification of CBD proteins

The DNA constructs of rat NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B were cloned into the pET28a vector, confirmed by sequencing and expressed in E. coli Rosetta2 (DE3) competent cells (Novagen) as described²⁶. Overexpressed proteins were purified on a TALON-superflow (GE healthcare) column followed by size exclusion chromatography on FPLC by using a Superdex 200 column (GE healthcare)²⁶. The purity of the finally purified protein preparations was >95%, as judged by SDS-PAGE. Proteins were concentrated, flash frozen in liquid nitrogen, and stored at −80 °C until their usage for SAXS or HDX-MS experiments.

Homology modeling of NCX-CBD12 isoforms

The models of NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B were generated based on the crystal structure of NCX1-CBD12-E454K (PDB 3US9)¹⁹. MODELLER was used to generate the models⁴⁵.

SAXS Data Collection and Analysis

SAXS data were measured at the beamline BM29 of the European Synchrotron Radiation Facility (ESRF), Grenoble, France. Data were collected at 4 °C with the X-ray beam at wavelength λ = 1.0 Å, and the distance from the sample to detector (PILATUS 1 M, Dectris Ltd) was 2.867 meters, covering a scattering vector range (q = 4πsinθ/λ) from 0.0025 to 0.5 Å⁻¹. Ten frames of two-dimensional images were recorded for each buffer or sample, with an exposure time of 2 sec per frame. The 2D images were reduced to one-dimensional scattering profiles and the scattering of the buffer was subtracted from the sample profile using the software at the site. The buffers contained 100 mM KCl, 20 mM β-mercaptoethanol, 20 mM Tris-HCl at pH 7.2, and either 10 mM CaCl₂ or 10 mM EDTA. To account for possible inter-particle effects, each sample was measured at four concentrations: ~10, ~5, ~2.5, and ~1.5 mg/mL. In the presence of such effects, the lowest concentration curve was merged with a higher concentration curve at q ~ 0.2 Å⁻¹ to prevent distortion of the low-angle data while preserving a high signal-to-noise ratio at the higher angles, which are far less sensitive to interparticle effects³³. The experimental radius of gyration (R _g) was calculated from data at low q values in the range of qR _g < 1.3, according to the Guinier approximation: lnI(q) ≈ ln(I(0)) − R _g ² q ²/3 using PRIMUS⁴⁶. The D_max values and the Porod volume were derived from the paired-distance distribution function (PDDF or P(r)) calculated using GNOM⁴⁷. Ab initio shape reconstructions were performed with the ATSAS software suite, using scattering data of the q range between 0.02 and 0.30 Å⁻¹ ⁴⁷.

Ensemble optimization method (EOM) analysis

NCX2-CBD12, NCX3-CBD12-AC, and NCX3-CBD12-B were modelled with MODELLER⁴⁵, based on the crystal structure of NCX1-CBD12-E454K (PDB 3US9)¹⁹. Unstructured residues at the N- and C-termini of CBD1 and CBD2 and the large flexible FG-loop of CBD2 were truncated. The software RANCH⁴⁸ was used to generate a pool of 10,000 stereochemically feasible structures with a random CBD1-CBD2 linker, N- and C-termini, and a CBD2 FG-loop. These pools were used as input for GAJOE⁴⁸, which selects an ensemble with the best fit to the experimental data using a genetic algorithm: 50 ensembles of 20 orientations each were “crossed” and “mutated” for 1,000 generations and the process was repeated 50 times.

Deuterium exchange for HDX-MS experiments

Purified proteins were prepared at a concentration of 100 µM in a buffer composed of 20 mM Tris-HCl, pH 7.2, 100 mM KCl, 20 mM β-mercaptoethanol either with 1.42 mM EDTA or 10 mM CaCl_2. The HDX reaction was initiated by mixing 2ul of protein samples with 28 µL of D₂O buffer (20 mM Tris-HCl, pD 7.2, 100 mM KCl, 20 mM β-mercaptoethanol) containing 1.42 mM EDTA, or 10 mM CaCl₂ for 10 sec, 100 sec, 1,000 sec, or 10,000 sec at 4 °C. The reaction was stopped by adding 30 µl of ice-cold quench buffer (0.1 M NaH₂PO₄, 1 M guanidine, 20 mM TCEP, pH 2.01). For non-deuterated (ND) samples, 2 µl of protein samples were mixed with 28 µl of H₂O buffer (20 mM Tris-HCl, pH 7.2, 100 mM KCl, 20 mM β-mercaptoethanol), and 30 µl of ice-cold quench buffer was added.

Mass spectrometry, peptide identification, and HDX-MS data processing

The quenched samples were digested and analyzed by the HDX-UPLC-ESI-MS system (Waters) as previously described²⁹. Mass-spectral analyses were performed with a Xevo G2 Quadruple-time of fly (Q-TOF) equipped with a standard electrospray ionization (ESI) source in MS^E mode in positive ion mode. All setting conditions for the system were reported previously⁴⁹. Identification and HDX-MS data processing of peptic peptides were performed by ProteinLynx Global Server 2.4 (Waters) and DynamX 2.0 (Waters) as previously described³¹. All of the data were derived from at least three independent experiments, and the statistical significance was analyzed by t-test.

References

Berridge, M. J., Bootman, M. D. & Roderick, H. L. Calcium signalling: dynamics, homeostasis and remodelling. Nat Rev Mol Cell Biol 4, 517–529, doi:10.1038/nrm1155 (2003).
Article CAS PubMed Google Scholar
McCue, H. V., Haynes, L. P. & Burgoyne, R. D. The diversity of calcium sensor proteins in the regulation of neuronal function. Cold Spring Harb Perspect Biol 2, a004085, doi:10.1101/cshperspect.a004085 (2010).
Article PubMed PubMed Central Google Scholar
Khananshvili, D. Sodium-calcium exchangers (NCX): molecular hallmarks underlying the tissue-specific and systemic functions. Pflugers Arch 466, 43–60, doi:10.1007/s00424-013-1405-y (2014).
Article CAS PubMed Google Scholar
Blaustein, M. P. & Lederer, W. J. Sodium/calcium exchange: its physiological implications. Physiol Rev 79, 763–854 (1999).
CAS PubMed Google Scholar
Bers, D. M. Cardiac excitation-contraction coupling. Nature 415, 198–205, doi:10.1038/415198a (2002).
Article ADS CAS PubMed Google Scholar
Giladi, M., Tal, I. & Khananshvili, D. Structural features of ion transport and allosteric regulation in sodium-calcium exchanger (NCX) proteins. Front Physiol 7, 30, doi:10.3389/fphys.2016.00030 (2016).
Article PubMed PubMed Central Google Scholar
Dyck, C. et al. Ionic regulatory properties of brain and kidney splice variants of the NCX1 Na(+)-Ca(2+) exchanger. J Gen Physiol 114, 701–711 (1999).
Article CAS PubMed PubMed Central Google Scholar
Kofuji, P., Lederer, W. J. & Schulze, D. H. Mutually exclusive and cassette exons underlie alternatively spliced isoforms of the Na/Ca exchanger. J Biol Chem 269, 5145–5149 (1994).
CAS PubMed Google Scholar
Linck, B. et al. Functional comparison of the three isoforms of the Na+/Ca2+ exchanger (NCX1, NCX2, NCX3). Am J Physiol 274, C415–423 (1998).
CAS PubMed Google Scholar
Quednau, B. D., Nicoll, D. A. & Philipson, K. D. Tissue specificity and alternative splicing of the Na+/Ca2+ exchanger isoforms NCX1, NCX2, and NCX3 in rat. Am J Physiol 272, C1250–1261 (1997).
CAS PubMed Google Scholar
Michel, L. Y. et al. Function and regulation of the Na+-Ca2+ exchanger NCX3 splice variants in brain and skeletal muscle. J Biol Chem 289, 11293–11303, doi:10.1074/jbc.M113.529388 (2014).
Article CAS PubMed PubMed Central Google Scholar
Ottolia, M., Nicoll, D. A. & Philipson, K. D. Roles of two Ca2+-binding domains in regulation of the cardiac Na+-Ca2+ exchanger. J Biol Chem 284, 32735–32741, doi:10.1074/jbc.M109.055434 (2009).
Article CAS PubMed PubMed Central Google Scholar
Michel, L. Y., Hoenderop, J. G. & Bindels, R. J. Towards understanding the role of the Na(2)(+)-Ca(2)(+) Exchanger Isoform 3. Rev Physiol Biochem Pharmacol 168, 31–57, doi:10.1007/112_2015_23 (2015).
Article PubMed Google Scholar
Khananshvili, D. Distinction between the two basic mechanisms of cation transport in the cardiac Na(+)-Ca2+ exchange system. Biochemistry 29, 2437–2442 (1990).
Article CAS PubMed Google Scholar
Liao, J. et al. Structural insight into the ion-exchange mechanism of the sodium/calcium exchanger. Science 335, 686–690, doi:10.1126/science.1215759 (2012).
Article ADS CAS PubMed Google Scholar
Marinelli, F. et al. Sodium recognition by the Na+/Ca2+ exchanger in the outward-facing conformation. Proc Natl Acad Sci USA 111, E5354–5362, doi:10.1073/pnas.1415751111 (2014).
Article ADS CAS PubMed PubMed Central Google Scholar
Ren, X. & Philipson, K. D. The topology of the cardiac Na(+)/Ca(2)(+) exchanger, NCX1. J Mol Cell Cardiol 57, 68–71, doi:10.1016/j.yjmcc.2013.01.010 (2013).
Article CAS PubMed PubMed Central Google Scholar
Hilge, M., Aelen, J. & Vuister, G. W. Ca2+ regulation in the Na+/Ca2+ exchanger involves two markedly different Ca2+ sensors. Mol Cell 22, 15–25, doi:10.1016/j.molcel.2006.03.008 (2006).
Article CAS PubMed Google Scholar
Giladi, M. et al. A common Ca2+-driven interdomain module governs eukaryotic NCX regulation. PLoS One 7, e39985, doi:10.1371/journal.pone.0039985 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Besserer, G. M. et al. The second Ca2+-binding domain of the Na+ Ca2+ exchanger is essential for regulation: crystal structures and mutational analysis. Proc Natl Acad Sci USA 104, 18467–18472, doi:10.1073/pnas.0707417104 (2007).
Article ADS CAS PubMed PubMed Central Google Scholar
Hilge, M., Aelen, J., Foarce, A., Perrakis, A. & Vuister, G. W. Ca2+ regulation in the Na+/Ca2+ exchanger features a dual electrostatic switch mechanism. Proc Natl Acad Sci USA 106, 14333–14338, doi:10.1073/pnas.0902171106 (2009).
Article ADS CAS PubMed PubMed Central Google Scholar
Hilgemann, D. W., Matsuoka, S., Nagel, G. A. & Collins, A. Steady-state and dynamic properties of cardiac sodium-calcium exchange. Sodium-dependent inactivation. J Gen Physiol 100, 905–932 (1992).
Article CAS PubMed Google Scholar
Boyman, L., Mikhasenko, H., Hiller, R. & Khananshvili, D. Kinetic and equilibrium properties of regulatory calcium sensors of NCX1 protein. J Biol Chem 284, 6185–6193, doi:10.1074/jbc.M809012200 (2009).
Article CAS PubMed Google Scholar
Nicoll, D. A. et al. The crystal structure of the primary Ca2+ sensor of the Na+/Ca2+ exchanger reveals a novel Ca2+ binding motif. J Biol Chem 281, 21577–21581, doi:10.1074/jbc.C600117200 (2006).
Article CAS PubMed Google Scholar
Giladi, M. et al. Dynamic features of allosteric Ca2+ sensor in tissue-specific NCX variants. Cell Calcium 51, 478–485, doi:10.1016/j.ceca.2012.04.007 (2012).
Article CAS PubMed Google Scholar
Tal, I., Kozlovsky, T., Brisker, D., Giladi, M. & Khananshvili, D. Kinetic and equilibrium properties of regulatory Ca(2+)-binding domains in sodium-calcium exchangers 2 and 3. Cell Calcium 59, 181–188, doi:10.1016/j.ceca.2016.01.008 (2016).
Article CAS PubMed Google Scholar
Giladi, M., Boyman, L., Mikhasenko, H., Hiller, R. & Khananshvili, D. Essential role of the CBD1-CBD2 linker in slow dissociation of Ca2+ from the regulatory two-domain tandem of NCX1. J Biol Chem 285, 28117–28125, doi:10.1074/jbc.M110.127001 (2010).
Article CAS PubMed PubMed Central Google Scholar
Giladi, M., Hiller, R., Hirsch, J. A. & Khananshvili, D. Population shift underlies Ca2+-induced regulatory transitions in the sodium-calcium exchanger (NCX). J Biol Chem 288, 23141–23149, doi:10.1074/jbc.M113.471698 (2013).
Article CAS PubMed PubMed Central Google Scholar
Giladi, M., Lee, S. Y., Hiller, R., Chung, K. Y. & Khananshvili, D. Structure-dynamic determinants governing a mode of regulatory response and propagation of allosteric signal in splice variants of Na+/Ca2+ exchange (NCX) proteins. Biochem J 465, 489–501, doi:10.1042/BJ20141036 (2015).
Article CAS PubMed Google Scholar
John, S. A., Ribalet, B., Weiss, J. N., Philipson, K. D. & Ottolia, M. Ca2+-dependent structural rearrangements within Na+-Ca2+ exchanger dimers. Proc Natl Acad Sci USA 108, 1699–1704, doi:10.1073/pnas.1016114108 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Lee, S. Y. et al. Structure-dynamic basis of splicing-dependent regulation in tissue-specific variants of the sodium-calcium exchanger. FASEB J 30, 1356–1366, doi:10.1096/fj.15-282251 (2016).
Article CAS PubMed Google Scholar
Salinas, R. K., Bruschweiler-Li, L., Johnson, E. & Bruschweiler, R. Ca2+ binding alters the interdomain flexibility between the two cytoplasmic calcium-binding domains in the Na+/Ca2+ exchanger. J Biol Chem 286, 32123–32131, doi:10.1074/jbc.M111.249268 (2011).
Article CAS PubMed PubMed Central Google Scholar
Koch, M. H., Vachette, P. & Svergun, D. I. Small-angle scattering: a view on the properties, structures and structural changes of biological macromolecules in solution. Q Rev Biophys 36, 147–227 (2003).
Article CAS PubMed Google Scholar
Sisalli, M. J., Annunziato, L. & Scorziello, A. Novel cellular mechanisms for Neuroprotection in Ischemic Preconditioning: A View from inside organelles. Front Neurol 6, 115, doi:10.3389/fneur.2015.00115 (2015).
Article PubMed PubMed Central Google Scholar
Chu, L., Greenstein, J. L. & Winslow, R. L. Modeling Na+-Ca2+ exchange in the heart: Allosteric activation, spatial localization, sparks and excitation-contraction coupling. J Mol Cell Cardiol 99, 174–187, doi:10.1016/j.yjmcc.2016.06.068 (2016).
Article CAS PubMed Google Scholar
Popovych, N., Sun, S., Ebright, R. H. & Kalodimos, C. G. Dynamically driven protein allostery. Nat Struct Mol Biol 13, 831–838, doi:10.1038/nsmb1132 (2006).
Article CAS PubMed PubMed Central Google Scholar
Laskowski, R. A., Gerick, F. & Thornton, J. M. The structural basis of allosteric regulation in proteins. FEBS Lett 583, 1692–1698, doi:10.1016/j.febslet.2009.03.019 (2009).
Article CAS PubMed Google Scholar
Kern, D. & Zuiderweg, E. R. The role of dynamics in allosteric regulation. Curr Opin Struct Biol 13, 748–757 (2003).
Article CAS PubMed Google Scholar
Tzeng, S. R. & Kalodimos, C. G. Dynamic activation of an allosteric regulatory protein. Nature 462, 368–372, doi:10.1038/nature08560 (2009).
Article ADS CAS PubMed Google Scholar
Goodey, N. M. & Benkovic, S. J. Allosteric regulation and catalysis emerge via a common route. Nat Chem Biol 4, 474–482, doi:10.1038/nchembio.98 (2008).
Article CAS PubMed Google Scholar
Tsai, C. J. & Nussinov, R. A unified view of “how allostery works”. PLoS Comput Biol 10, e1003394, doi:10.1371/journal.pcbi.1003394 (2014).
Article ADS PubMed PubMed Central Google Scholar
Rader, A. J. & Brown, S. M. Correlating allostery with rigidity. Mol Biosyst 7, 464–471, doi:10.1039/c0mb00054j (2011).
Article CAS PubMed Google Scholar
Nussinov, R. & Tsai, C. J. Allostery in disease and in drug discovery. Cell 153, 293–305, doi:10.1016/j.cell.2013.03.034 (2013).
Article CAS PubMed Google Scholar
Giladi, M., Friedberg, I., Fang, X., Hiller, R., Wang, Y. X. & Khananshvili, D. G503 is obligatory for coupling of regulatory domains in NCX proteins. Biochemistry 51, 7313–7320, doi:10.1021/bi300739z (2012).
Article CAS PubMed Google Scholar
Eswar, N., Eramian, D., Webb, B., Shen, M. Y. & Sali, A. Protein structure modeling with MODELLER. Methods Mol Biol 426, 145–159, doi:10.1007/978-1-60327-058-8_8 (2008).
Article CAS PubMed Google Scholar
Konarev, P. V., Volkov, V. V., Sokolova, A. V., Koch, M. H. J. & Svergun, D. I. PRIMUS: a Windows PC-based system for small-angle scattering data analysis. J Appl Crystallogr 36, 1277–1282, doi:10.1107/S0021889803012779 (2003).
Article CAS Google Scholar
Svergun, D. I. Determination of the regularization parameter in indirect-transform methods using perceptual criteria. J Appl Crystallogr 25, 495–503, doi:10.1107/S0021889892001663 (1992).
Article Google Scholar
Tria, G., Mertens, H. D., Kachala, M. & Svergun, D. I. Advanced ensemble modelling of flexible macromolecules using X-ray solution scattering. IUCrJ 2, 207–217, doi:10.1107/S205225251500202X (2015).
Article CAS PubMed PubMed Central Google Scholar
Kim, D. K., Yun, Y., Kim, H. R., Seo, M. D. & Chung, K. Y. Different conformational dynamics of various active states of beta-arrestin1 analyzed by hydrogen/deuterium exchange mass spectrometry. J Struct Biol 190, 250–259, doi:10.1016/j.jsb.2015.04.006 (2015).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

This work was supported by the Israel Science Foundation Grant #825/14 to D.K. and the National Research Foundation of Korea, Grant [NRF-2012R1A5A2A28671860 and NRF-2015R1A1A1A05027473] to K.Y.C. The financial support of the Fields Estate foundation to D.K. is highly appreciated. This work was performed by Y.A., R.G., and Y.T. in fulfillment of the M.D. thesis requirements of the Sackler Faculty of Medicine, Tel-Aviv University.

Author information

Moshe Giladi and Su Youn Lee contributed equally to this work.

Authors and Affiliations

Department of Physiology and Pharmacology, Tel-Aviv University, Tel-Aviv, 69978, Israel
Moshe Giladi, Yarden Ariely, Yotam Teldan, Rotem Granit, Roi Strulovich, Yoni Haitin & Daniel Khananshvili
School of Pharmacy, Sungkyunkwan University, Jangan-gu, Suwon, 440-746, South Korea
Su Youn Lee & Ka Young Chung

Authors

Moshe Giladi
View author publications
You can also search for this author in PubMed Google Scholar
Su Youn Lee
View author publications
You can also search for this author in PubMed Google Scholar
Yarden Ariely
View author publications
You can also search for this author in PubMed Google Scholar
Yotam Teldan
View author publications
You can also search for this author in PubMed Google Scholar
Rotem Granit
View author publications
You can also search for this author in PubMed Google Scholar
Roi Strulovich
View author publications
You can also search for this author in PubMed Google Scholar
Yoni Haitin
View author publications
You can also search for this author in PubMed Google Scholar
Ka Young Chung
View author publications
You can also search for this author in PubMed Google Scholar
Daniel Khananshvili
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

M.G., K.Y.C., and D.K. designed the experiments. M.G., S.Y.L., Y.A., R.G., Y.T., R.S., and Y.H. conducted the experiments. M.G., S.Y.L., K.Y.C., and D.K. analyzed the data and prepared the figures. M.G., K.Y.C., and D.K. wrote the manuscript.

Corresponding authors

Correspondence to Ka Young Chung or Daniel Khananshvili.

Ethics declarations

Competing Interests

The authors declare that they have no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Giladi, M., Lee, S.Y., Ariely, Y. et al. Structure-based dynamic arrays in regulatory domains of sodium-calcium exchanger (NCX) isoforms. Sci Rep 7, 993 (2017). https://doi.org/10.1038/s41598-017-01102-x

Download citation

Received: 08 February 2017
Accepted: 24 March 2017
Published: 20 April 2017
DOI: https://doi.org/10.1038/s41598-017-01102-x

This article is cited by

Neuroprotective effect of heparin Trisulfated disaccharide on ischemic stroke
- Gabrielly M. D. Chiarantin
- Lina M. Delgado-Garcia
- Marimélia Porcionatto
Glycoconjugate Journal (2021)
Sodium–Calcium Exchangers of the SLC8 Family in Oligodendrocytes: Functional Properties in Health and Disease
- Samantha A. Spencer
- Edna Suárez-Pozos
- Babette Fuss
Neurochemical Research (2020)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.