The Na⁺/Ca²⁺ exchanger: molecular structure, functional regulation, and mode-selective modulation

Abstract

The Na⁺/Ca²⁺ exchanger (NCX) is a key regulator of intracellular Ca²⁺ homeostasis, mediating bidirectional ion transport in response to electrochemical gradients. NCX plays essential roles in diverse physiological processes, including cardiac contraction, neuronal signaling, and hormone secretion, while its dysregulation is implicated in a wide range of pathophysiological conditions such as heart failure, ischemia, and neurodegenerative diseases. Recent advances in structural biology, particularly cryo-electron microscopy, have provided detailed insights into the molecular architecture of eukaryotic NCX, revealing the mechanisms underlying ion transport, regulation, and inactivation. These studies have highlighted the functional importance of conserved structural elements, including the α1- and α2-repeats and cytosolic regulatory domains, in governing exchanger activity. In parallel, pharmacological studies have identified a variety of NCX modulators with distinct mode and isoform selectivity. While classical inhibitors predominantly target the reverse mode, emerging compounds—including highly selective inhibitors and novel activators—demonstrate diverse mechanisms of action, often involving allosteric modulation rather than direct interaction with the ion-binding site. This review integrates current knowledge of NCX structure, function, and pharmacology, with a particular focus on the relationship between structural determinants and transport mode selectivity. We also discuss the challenges that have limited the clinical translation of NCX-targeting drugs and outline future perspectives for the rational design of next-generation NCX modulators.

INTRODUCTION

Since the discovery of calcium as a second messenger, it has been implicated in an increasing number of biological functions [10]. Calcium ions (Ca²⁺) serve as ubiquitous signaling molecules that orchestrate diverse physiological processes across all cell types. Ca²⁺ regulates fundamental cellular processes such as proliferation, protein synthesis, and differentiation, as well as specialized functions including muscle contraction, neurotransmitter release, electrical excitability, and synaptic plasticity [4]. This broad involvement highlights the remarkable versatility of Ca²⁺ as a signaling molecule. A defining feature of Ca²⁺ signaling is the steep concentration gradient across the plasma membrane: while extracellular Ca²⁺ is maintained at relatively high concentrations (1 to 2 mM) [11], cytosolic Ca²⁺ is kept at extremely low concentrations (~100 nM at rest), creating a gradient of approximately 20,000-fold. This gradient enables Ca²⁺ to function as a rapid and reversible signaling molecule, whereby even subtle changes in intracellular Ca²⁺ concentration can trigger profound cellular responses. Consequently, precise spatiotemporal regulation of intracellular Ca²⁺ levels is essential for maintaining normal cellular function [11].

Ca²⁺ homeostasis refers to the maintenance of stable intracellular Ca²⁺ concentration across different cellular compartments [60]. This process is governed by Ca²⁺ fluxes between the extracellular space and the cell, as well as Ca²⁺ transport within intracellular organelles [60]. Cells can be broadly classified into excitable cells, which generate action potentials, and non-excitable cells, which do not. At the molecular level, cytosolic Ca²⁺ is regulated by channels, adenosine triphosphate (ATP)-driven pumps, and ion exchangers located on both the plasma membrane and intracellular membranes, as well as by Ca²⁺-binding proteins in the cytosol [53]. At the plasma membrane, voltage-gated Ca²⁺ channels (VGCCs) and receptor-operated Ca²⁺ channels (ROCCs) mediate Ca²⁺ influx. In contrast, intracellular Ca²⁺ stores—primarily the endoplasmic reticulum (ER) and sarcoplasmic reticulum (SR)—release Ca²⁺ via inositol 1,4,5-trisphosphate receptors (IP₃Rs) and ryanodine receptors (RyRs). In addition, organelles such as the ER and mitochondria contribute to the regulation of cytosolic Ca²⁺ through coordinated interactions [74, 79, 104]. Following cellular signaling events, cytosolic Ca²⁺ must be rapidly restored to basal levels. This is achieved by Ca²⁺ extrusion and sequestration mechanisms, the sarcoplasmic/endoplasmic reticulum Ca²⁺-ATPase (SERCA), which pumps Ca²⁺ into intracellular stores, and plasma membrane transport systems such as the plasma membrane Ca²⁺-ATPase (PMCA) and the Na⁺/Ca²⁺ exchanger (NCX), which extrude Ca²⁺ from the cell [14, 15].

Among Ca²⁺ extrusion mechanisms, NCX plays a particularly important role, especially in excitable cells such as cardiomyocytes and neurons, where large and rapid Ca²⁺ fluxes occur. NCX exhibits low affinity but high capacity for Ca²⁺, whereas PMCA displays high affinity but low capacity [16]. This complementary relationship reflects their distinct functional roles: NCX removes large elevations in cytosolic Ca²⁺, whereas PMCA finetunes Ca²⁺ concentration near basal levels [16]. The physiological significance of NCX is supported by quantitative studies; for example, it accounts for approximately 70% and 90% of Ca²⁺ extrusion in pancreatic β-cells and the heart, respectively [6, 108]. Beyond Ca²⁺ removal, NCX couples Ca²⁺ transport to the transmembrane Na⁺ gradient, exchanging three Na⁺ ions for one Ca²⁺ ion in an electrogenic process that is regulated by ionic gradients and membrane potential.

A defining characteristic of NCX is its bidirectional transport capability. NCX mediates Ca²⁺ fluxes in exchange for Na⁺, operating either in the forward mode (Ca²⁺ efflux) or reverse mode (Ca²⁺ influx), with directionality determined by ionic gradients and membrane potential [56]. Under normal physiological conditions, NCX predominantly operates in the forward mode, extruding Ca²⁺ from the cytosol. Although forward-mode activity represents the primary physiological function of NCX, transient reversal may contribute to excitation-contraction (EC)-coupling, vascular constriction, and synaptic transmission [56]. However, when the Na⁺ electrochemical gradient is diminished or reversed—such as during membrane depolarization or elevated intracellular Na⁺—NCX switches to the reverse mode, resulting in Ca²⁺ influx. In pathological conditions, reverse-mode activity often becomes predominant [56]. For example, in ischemia, heart failure, and stroke, elevated intracellular Na⁺ drives sustained Ca²⁺ influx via reverse-mode NCX, leading to Ca²⁺ overload, mitochondrial dysfunction, and ultimately cell death. These observations highlight the central role of NCX in the dysregulation of Ca²⁺ homeostasis under disease conditions. Accordingly, NCX is increasingly recognized as a key therapeutic target in cardiovascular and neurological disorders.

Mammals express three NCX isoforms—NCX1, NCX2, and NCX3—each exhibiting distinct tissue distribution and functional properties. These isoforms, together with their splice variants, are expressed in a tissue-specific manner and contribute to a wide range of physiological processes, including cardiac contractility, neuronal plasticity and learning, blood pressure regulation, renal Ca²⁺ reabsorption, immune responses, neurotransmitter and insulin secretion, apoptosis, cell proliferation, and mitochondrial bioenergetics [56]. NCX1 is widely expressed in both excitable tissues (e.g., heart and brain) and non-excitable tissues (e.g., kidney and pancreas), where it plays critical roles in cardiac contractility, renal Ca²⁺ reabsorption, and insulin secretion [38, 94]. NCX2 is predominantly expressed in the brain and spinal cord, contributing to neuronal Ca²⁺ homeostasis and synaptic plasticity [58, 94], while NCX3 is expressed in the brain and skeletal muscle [77, 78]. The existence of multiple splice variants for each isoform further expands the functional diversity of NCX-mediated Ca²⁺ regulation, enabling fine-tuned control of Ca²⁺ dynamics in different cellular contexts. Understanding the molecular basis of NCX function—including its three-dimensional structure, ion transport mechanism, and regulatory properties—is essential for the developing isoform-selective therapeutic strategies.

This review provides a comprehensive overview of the NCX, integrating recent structural insights with functional and pharmacological perspectives to better understand its roles in physiology and disease. We highlight how advances in structural biology have elucidated the molecular mechanisms underlying the unique bidirectional transport and regulatory properties of NCX, and how this knowledge is being leveraged to develop selective NCX inhibitors and activators as potential therapeutic agents. Given the growing recognition of NCX as a central regulator of Ca²⁺ dysregulation in major human diseases—including heart failure, arrhythmias, stroke, and neurodegenerative disorders—a deeper understanding of this transporter holds significant promise for future drug discovery.

CALCIUM HOMEOSTASIS AND PHYSIOLOGICAL ROLES OF NCX

Intracellular Ca²⁺ concentration varies widely depending on cellular location [3]. At rest, cytosolic Ca²⁺ is maintained at extremely low concentrations (~100 nM) [34], approximately 20,000-fold lower than extracellular Ca²⁺ concentrations. This steep concentration gradient is fundamental to Ca²⁺ signaling, enabling rapid and reversible cellular responses. However, dysregulation of Ca²⁺ homeostasis can have deleterious effects, including apoptosis and cell death [3, 34, 89]. Because Ca²⁺ regulates nearly all aspects of cellular function, its concentration and signaling dynamics must be tightly controlled (Fig. 1). The plasma membrane plays a central role in regulating Ca²⁺ entry and exit through various ion channels and transport systems [3]. Ca²⁺ channels are broadly classified into two major types based on their activation mechanisms. VGCCs are activated by changes in membrane potential and are primarily expressed in excitable cells, where they serve as the major pathways for Ca²⁺ influx [20, 21, 35, 115]. Excitable cells—including neurons, muscle cells (skeletal, smooth, and cardiac), and certain endocrine and germ cells—are capable of generation action potentials that enable electrical signaling [41]. In contrast, most other cell types are considered non-excitable [61] and generally express low levels of VGCCs [61]. Therefore, Ca²⁺ influx in non-excitable cell types is primarily mediated by ROCCs, which are activated downstream of receptor stimulation, typically via G protein-coupled signaling pathways [61]. In contrast to ligand-gated ion channels (ionotropic receptors), which are directly activated by ligand binding, ROCCs are indirectly regulated through intracellular signaling cascades. Several types of ROCCs have been identified [110], including members of the transient receptor potential (TRP) channel family [83]. In addition, ligand-gated ion channels such as N-methyl-D-aspartate receptors (NMDARs) [50, 102] and α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid receptors (AMPARs) mediate Ca²⁺ influx through direct ligand binding, particularly in the central nervous system [50]. Furthermore, store-operated Ca²⁺ entry is mediated by Orai (CRACM) channels, which are activated by stromal interaction molecule (STIM) proteins following depletion of intracellular Ca²⁺ stores.

Fig. 1.

Schematic representation of membrane proteins regulating intracellular Ca²⁺ homeostasis. G protein-coupled receptors (GPCRs) regulate intracellular signaling through heterotrimeric G proteins. Ca²⁺ influx is mediated by plasma membrane channels (Orai, voltage-gated Ca²⁺ channels (VGCC), and receptor-operated Ca²⁺ channels (ROCC)), while Ca²⁺ extrusion is controlled by transporters such as Na⁺/Ca²⁺ exchanger (NCX) and plasma membrane Ca²⁺-ATPase (PMCA). Intracellular Ca²⁺ dynamics are further regulated by endoplasmic reticulum (ER)-localized proteins including 1,4,5-trisphosphate receptors (IP₃R), ryanodine receptors (RyR), and sarcoplasmic/endoplasmic reticulum Ca²⁺-ATPase (SERCA). Stromal interaction molecule (STIM) 1 acts as an ER Ca²⁺ sensor that detects Ca²⁺ depletion and subsequently activates Orai channels to induce store-operated Ca²⁺ entry.

Increases in intracellular Ca²⁺ arise either from Ca²⁺ influx across the plasma membrane or from release of Ca²⁺ from intracellular stores, primarily the ER or, in muscle cells, the SR [96]. The ER is the largest intracellular organelle and extends throughout much of the cell, serving as a high-capacity reservoir for Ca²⁺ [5, 103]. Two major Ca²⁺ release channels mediate signaling-induced Ca²⁺ release from the ER lumen to the cytosol [7, 27]. The first is the IP₃R, an intracellular ligand-gated Ca²⁺ channel localized to the ER membrane [7, 27]. Its ligand, IP₃, is a second messenger generated by activation of phospholipase C (PLC) downstream of G_q-coupled or receptor tyrosine kinase signaling pathways [51]. PLC hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP₂) to produce IP₃ and diacylglycerol (DAG) [7, 27]. Ca²⁺ itself acts as an allosteric modulator of the IP₃R, playing a key role in shaping the IP₃R-mediated Ca²⁺ signals [27, 107]. The second major ER Ca²⁺ release channel is the RyR, a high-conductance and relatively nonspecific cation channel (~100–150 pS for Ca²⁺) [70]. The mammalian genome encodes three RyR isoforms (RyR1, RyR2, and RyR3), which share approximately 70% sequence homology [25]. Ca²⁺ is the principal activator of all three isoforms [28, 39], a process known as calcium-induced calcium release (CICR), in which Ca²⁺ promotes its own release from intracellular stores [96]. Both IP₃Rs and RyRs exhibit CICR, enabling signal amplification and spatiotemporal patterning of Ca²⁺ signals.

G protein-coupled receptors (GPCRs) are membrane receptors that play a central role in intracellular signaling pathways, including those regulating Ca²⁺ dynamics [23]. GPCRs are seven-transmembrane (heptahelical) receptors located on the plasma membrane that coupled to heterotrimeric guanine nucleotide-binding proteins (G-proteins) [23]. G proteins consist of three subunits—α, β, and γ. In the inactive state, the α subunit is bound to guanosine diphosphate (GDP). Upon receptor activation, GDP is released and replaced by guanosine triphosphate (GTP), leading to activation of the G protein [1]. G proteins are classified into four major families based on their α subunit: Gα_i, Gα_s, Gα_12/13, and Gα_q [55]. In the Gα_q pathway, PLCβ-mediated hydrolysis of PIP₂ generates IP₃ and DAG [67]. IP₃ induces Ca²⁺ release from the ER via IP₃ receptors, while DAG activates protein kinase C (PKC).

In the Gα_s pathway, Gα_s activates a family of adenylyl cyclases (ACs), leading to an increase in intracellular levels of the second messenger cyclic adenosine 3′,5′-monophosphate (cAMP) [12, 22, 81, 82, 106]. Elevated cAMP activates protein kinase A (PKA), which phosphorylates multiple target proteins, including L-type Ca²⁺-channel (LCC), RyRs, and myosin binding protein C [23]. In the Gα_i pathway, Gα_i inhibits AC activity, resulting in reduced intracellular cAMP levels [109]. In the Gα_12/13 pathway, Gα_12/13 activates RhoA and its downstream effector Rho-associated kinase (ROCK), leading to inhibition of Ca²⁺-dependent exocytosis [113].

Ca²⁺ stored in the SR/ER plays a critical role in regulating cytosolic Ca²⁺ concentration. A key mechanism that prevents store depletion is the STIM1-ORAI1 interaction, which mediates store-operated Ca²⁺ entry (SOCE). STIM1 is a type I membrane protein localized in the ER and functions as a luminal Ca²⁺ sensor through its EF-hand domain [40]. Orai proteins are Ca²⁺-permeable channels located in the plasma membrane that mediate Ca²⁺ influx under the control of STIM proteins [73]. Upon depletion of ER Ca²⁺ stores and disassociation of Ca²⁺ from the EF-hand domain, STIM1 undergoes conformational changes, oligomerizes, and translocates to ER-plasma membrane junctions, where it directly interacts with ORAI1 [62, 92]. This interaction activates Orai channels, resulting in Ca²⁺ influx from extracellular space, which subsequently contributes to replenishment of ER Ca²⁺ stores [62, 92].

Following cellular signaling events that elevate cytosolic Ca²⁺, Ca²⁺ must be rapidly cleared to restore basal conditions. This is achieved by active transport mechanisms that move Ca²⁺ against its concentration gradient. The SERCA, a membrane transport protein ubiquitously expressed in the ER of eukaryotic cells [93], pumps Ca²⁺ from the cytosol into the SR/ER lumen through ATP hydrolysis. By maintaining low cytosolic Ca²⁺ levels, SERCA enables a wide range of signaling pathways and physiological processes, including synaptic transmission, muscle contraction, and fertilization [93]. At the plasma membrane, Ca²⁺ extrusion is mediated by two major systems: PMCA and NCX [14]. Although Ca²⁺ is essential for cellular function, its dysregulation can lead to deleterious effects, highlighting the importance of tight control of intracellular Ca²⁺ homeostasis.

PMCA exhibits high affinity and low capacity for Ca²⁺ transport [14, 17, 18], pumping Ca²⁺ from the cytosol against its electrochemical gradient using energy derived from ATP hydrolysis. This high-affinity, low-capacity transport is well suited for fine-tuning cytosolic Ca²⁺ concentrations near resting levels. PMCA activity is regulated by multiple signaling pathways: it is stimulated by protein kinase A (PKA) [36] and Ca²⁺/calmodulin kinase (CaMK), and inhibited by PKC [14, 98]. This complex regulation enables PMCA to integrate multiple cellular signals in the control of Ca²⁺ homeostasis.

In contrast to PMCA, NCX utilizes the Na⁺ gradient to drive Ca²⁺ extrusion as a secondary active transporter [26, 114]. As illustrated in Fig. 1, NCX plays a central role in Ca²⁺ extrusion across the plasma membrane within the broader network of Ca²⁺ regulatory systems. NCX is localized to the plasma membrane and is notably absent from SR/ER membranes [8]. NCX exhibits low affinity but high capacity for Ca²⁺, making it particularly important for removing large Ca²⁺ loads during periods of intense cellular activity. Three isoforms, NCX1, NCX2, and NCX3, are expressed with distinct tissue distributions throughout the body [66]. Each isoform undergoes alternative splicing that further diversifies its functional and regulatory properties. A key functional feature of NCX is its bidirectional transport capability. NCX exchanges three Na⁺ ions for one Ca²⁺ ion in an electrogenic process. Under physiological conditions with a favorable Na⁺ gradient, NCX predominantly operates in forward mode, mediating Ca²⁺ extrusion [9, 52]. However, when the electrochemical gradient is reduced or reversed—such as during membrane depolarization or under pathological conditions associated with intracellular Na⁺ accumulation—NCX switches to the reverse mode, resulting in Ca²⁺ influx [9, 52]. This mode switching, governed by the thermodynamic balance between Na⁺ and Ca²⁺ gradients and membrane potential, distinguishes NCX from unidirectional transporters such as SERCA and PMCA and underpins its dual role in both physiological regulation and pathological Ca²⁺ dysregulation.

STRUCTURE ARCHITECTURE AND REGULATORY MECHANISMS OF NCX

NCX proteins are highly conserved from prokaryotes, including archaea, to eukaryotes, reflecting their fundamental importance in Ca²⁺ homeostasis across all domains of life. Structural studies of NCX from Methanococcus jannaschii (NCX_Mj) have provided foundational insights into the ion transport mechanisms, owing to its suitability for X-ray crystallography. In recent years, the advent in cryo-electron microscopy (cryo-EM) have enabled high-resolution structural analysis of eukaryotic NCX proteins in near-native states. These studies have revealed critical regulatory features absent in prokaryotic NCX, which are essential for understanding mammalian physiology and disease. In this section, we focus on the structures of eukaryotic NCX proteins, with particular emphasis on how structural insights elucidate the mechanisms underlying bidirectional ion transport and regulatory control.

The mammalian NCX1 protein comprises approximately 938 amino acids, with a theoretical molecular mass of ~120 kDa, and contains ten transmembrane (TM) segments (TMs 1–10) (Fig. 2). The overall architecture of eukaryotic NCX consists of two major structural domains: a TM domain responsible for ion translocation and a large intracellular regulatory domain that modulates transporter activity in response to cytosolic Ca²⁺. Notably, the intracellular regulatory domain interrupts the primary sequence between TM5 and TM6, effectively dividing the TM domain into two homologous halves (TMs 1–5 and TMs 6–10) [19, 76, 91, 95, 101]. This structural organization is characteristic of all eukaryotic NCX isoforms and distinguishes them from their simpler prokaryotic counterparts.

Fig. 2.

Schematic structural model of the eukaryotic Na⁺/Ca²⁺ exchanger (NCX). The transmembrane (TM) domain consists of ten helices (TMs 1–10), organized into two homologous halves (TMs 1–5 and TMs 6–10) separated by a large intracellular regulatory loop (5L6 loop). The ion translocation pathway is formed by a four-helix bundle comprising the conserved α1- (TMs 2–3) and α2-repeats (TMs 7–8). Peripheral helices (TM1 and TM6) are positioned at the outer region of the TM domain and contribute to conformational transitions during the transport cycle. The cytosolic regulatory domain contains key elements including the exchanger inhibitory peptide (XIP), the two-helix bundle (THB), and the Ca²⁺-binding domains (CBD) 1 and CBD2, which mediate Ca²⁺-dependent regulation of exchanger activity. The THB is a structural component of the cytosolic regulatory domain. The β-hub (B-hub), formed through interaction between the XIP region and a linker β-hairpin, contributes to NCX inactivation. The spatial organization highlights the structural coupling between the ion translocation core and regulatory modules that underlie NCX function and mode switching.

Within the TM domain, eight of the ten TM helices (TMs 2–5 and 7–10) form a tightly packed core that constitutes the functional center of the transporter (Fig. 2). At the core of this structure are two highly conserved α-repeats: TMs 2–3 from the α1 repeat and TMs 7–8 as the α2 repeat. These α-repeats assemble into a four-helix bundle (TM2, TM3, TM7, and TM8) that constitutes the ion translocation pathway. In contrast to the compact core, TMs 1 and 6 are longer helices located at the periphery of the TM domain, where they are more loosely packed against the core at angle of approximately 45° [112].

These peripheral helices play a crucial role in the conformational changes required for the alternating access mechanism of ion transport and serve as key interaction sites for the regulatory domain. The ion-binding pocket at the center of the α-repeats comprises four sites arranged in a diamond-shaped configuration: S_ext, S_mid, S_int, and S_Ca [33]_. Twelve amino acid residues contribute to Na⁺ and Ca²⁺ coordination within this pocket—four residues from TM2, four from TM7, and two each from TM3 and TM8 [33]. Our understanding of ion occupancy at these sites has evolved significantly through combined structural and computational studies. Initial crystallographic analyses of NCX_Mj suggested that S_ext, S_mid, and S_int were occupied by three Na⁺ ions, while S_Ca bound one Ca²⁺ ion [64]. However, subsequent studies demonstrated that simultaneous occupation of all four sites by three Na⁺ ions and one Ca²⁺ ion is thermodynamically unfavorable [72]. Recent molecular dynamics (MD) simulations and ion flux analyses have led to a revised model in which three Na⁺ ions occupy S_ext, S_int, and S_Ca, whereas Ca²⁺ binds to S_Ca when present [72]. Importantly, S_mid does not directly bind Na⁺ or Ca²⁺; instead, it coordinates a water molecule associated with a protonated aspartate residue (D240 in NCX_Mj), which may contribute to ion selectivity and transport coupling [33].

Structural analyses of eukaryotic NCX have confirmed that conserved residues from both α-repeats form the central ion-binding sites, with functional assignments analogous to those in NCX_Mj: S_ext and S_int serve as dedicated Na^+-binding sites, S_Ca functions as the primary binding site for either Ca²⁺ or Na⁺ (depending on the transport cycle), and S_mid coordinates a water molecule [64, 65]. Cryo-EM structures of human NCX reveal three Na⁺ ions and a water molecule modeled based on the well-defined ion-ligand geometry established in NCX_Mj structures [64, 65]. The high degree of conservation in the ion-binding pocket architecture between prokaryotic and eukaryotic NCX underscores the fundamental mechanism of 3Na⁺:1Ca²⁺ exchange that has been preserved throughout evolution.

The most significant structural difference between prokaryotic and eukaryotic NCX is the presence of a large cytoplasmic regulatory domain in eukaryotic variants. This domain is inserted between TM5 and TM6 and comprises an extensive intracellular loop of approximately 520 residues (termed the 5L6 loop), which contains multiple regulatory elements [57, 88]. The principal components of this regulatory domain include the exchanger inhibitory peptide (XIP), a key auto-inhibitory regulatory element, a two-helix bundle (THB), Ca²⁺ binding domains 1 (CBD1) and 2 (CBD2) [32, 57, 88], and a short palmitoylation helix (TMH2) [24, 112]. The two homologous CBD1 and CBD2 constitute the major structural mass of the cytosolic domain and play distinct yet complementary roles in Ca²⁺-dependent regulation of NCX activity [112].

CBD1 exhibits high Ca²⁺ affinity [112] and contains four Ca²⁺-binding sites that are typically occupied under physiological conditions. However, due to its considerable distance from the TM domain, Ca²⁺ binding at CBD1 is unlikely to directly influence ion transport activity [112]. Instead, structural studies of isolated CBD1–CBD2 domains suggest that Ca²⁺ binding at CBD1 primarily serves a structural role by stabilizing the hinge region between the two domains and maintaining the overall architecture of the regulatory domain [30,31,32, 97]. In contrast, CBD2 contains two Ca²⁺-binding sites that are often observed in the apo state in structural studies, indicating a lower Ca²⁺ affinity compared to CBD1 [112].

Eukaryotic NCX possesses a sophisticated Ca²⁺-dependent inactivation mechanism in which the Ca²⁺-binding properties of CBD2 play a central role. Inactivation typically occurs when NCX operates in the reverse mode (Na⁺ efflux, Ca²⁺ influx) under conditions of low cytosolic Ca²⁺, resulting in a Na⁺-loaded, inward-facing conformation. In this state, a linker β-hairpin (β1–β2) interacts with the XIP (auto-inhibitory peptide) region to form a structural assembly termed the β-hub [112]. The β-hub subsequently engages the cytosolic domain—primarily via the CH2 helix of CBD2—to generate a stable inactivation complex [112]. This complex constrains TMs 1 and 6 in an inward-facing orientation, preventing the conformational transitions required for ion transport and thereby halting exchanger activity. Importantly, this inactivation mechanism is reversible and highly sensitive to changes in cytosolic Ca²⁺ levels, with Ca²⁺ binding to CBD2 promoting the dissociation of the β-hub and restoration of transport activity.

As cytosolic Ca²⁺ levels rise, Ca²⁺ binding to CBD2 induces a pronounced conformational change. This shift promotes a direct interaction between the CH2 helix and the β-hub structure, leading to the destabilization and subsequent disassembly of the β-hub. The resulting rearrangement of the cytosolic domain relative to the TM module [112] likely displaces the XIP region from its inhibitory site. This release relieves conformational constraints on TMs 1 and 6, thereby restoring a TM configuration capable of normal ion transport cycling [112].

PHARMACOLOGICAL MODULATION OF NCX

Several classes of NCX inhibitors have been identified, differing in their specificity and isoform selectivity. Among the earliest compounds, the isothiourea derivative 2-[2-[4-(4-nitrobenzyloxy) phenyl]ethyl]isothiourea methanesulfonate (KB-R7943) was introduced in 1996 as a selective inhibitor of reverse-mode NCX1 [43, 111]. Subsequent studies demonstrated that KB-R7943 also inhibits NCX2 and NCX3, with a higher affinity for NCX3 [2] (Table 1). The inhibitory profile of KB-R7943 is concentration-dependent: it selectively inhibits the reverse mode at low concentrations (≤10 μM), whereas at higher concentrations (≥10 μM) it affects both forward and reverse modes [13, 43, 111]. Amino acid residues Val820, Gln826, and Gly833 within the α2-repeat are at least partly responsible for the interaction between NCX1 and KB-R7943 [49] (Fig. 3). Pharmacologically, KB-R7943 has been shown to attenuate increases in cytosolic Ca²⁺ and reduce neuronal injury in models of ischemia, mechanical brain trauma, oxygen-glucose deprivation, and glutamate excitotoxicity [13, 63, 68, 69, 99]. However, it should be noted that KB-R7943 also exhibits nonspecific effects on other ion channels and receptors [44].

Table 1. Isoform and transport mode selectivity of Na⁺/Ca²⁺ exchanger (NCX) modulators.

NCX inhibition/activation	Compound name	Isoform selectivity	Mode selectivity	Determinants of NCX modulator sensitivity
Inhibitor	KB-R7943	NCX3 >NCX1/NCX2 (3-fold more effective on NCX3 than on NCX1 and NCX2.) [90, 91]	Reverse mode >> forward mode (KB-R7943 inhibits the reverse mode by NCX much more effectively than the forward mode.) [87, 88, 92, 93]	Gly833 (α2-repeat; critical), Val820 (α-2 repeat; contributing), Gln826 (α2-repeat; contributing) [99]
	ORM-10962	NCX1 >> NCX2/NCX3 (ORM-10962 inhibits NCX1 specifically.) [94]	Both mode (ORM-10962 equally inhibited both the reverse and forward mode NCX.) [95]	?
	ORM-11372	NCX1.1 (ORM‐11372 is the most potent and selective NCX 1.1 inhibitor.) [96]	Both mode (ORM‐11372 inhibited NCX 1.1 reverse and forward mode currents with a similar potency in both human iPS and rat CMs.) [96]	?
	SAR296968	NCX1/NCX2/NCX3 (SAR296968 exhibited a potent inhibitory activity of NCX1 in CHO cells and similar effects were observed also for NCX2 and NCX3.) [97]	Both mode (SAR296968 inhibited both the forward and reverse mode of the NCX current in a concentration-dependent manner.) [98]	?
	SEA0400	NCX1 >> NCX2 (SEA0400 predominantly blocks NCX1, only mildly blocks uptake by NCX2, and exerts almost no influence upon uptake by NCX3.) [99]	Both mode (SEA0400 suppresses both forward and reverse mode operation of NCX.) [100, 101]	Phe213 (TM5; critical), Gly833 (α2-repeat; indirect contribution) [99]
	SM-15811	?	Reverse mode (YM-281956 inhibits NCX reverse mode and is more potent than SEA0400.) [102]	?
	SN-6	NCX1 >NCX2/NCX3 (SN-6 inhibits NCX1 more potently than NCX2 and NCX3.) [103]	Reverse mode >> forward mode (SN-6 inhibits the unidirectional outward INCX more potently than the unidirectional inward INCX.) [104]	Val227 (XIP; critical), Tyr228 (XIP; critical), Gly833 (α2-repeat; critical) Asn839 (α2-repeat; critical), Phe213 (TM5; contributing) [103]
	YM-244769	NCX3 >> NCX1/NCX2 (YM-244769 preferentially inhibits NCX3 rather than NCX1 or NCX2.) [105]	Reverse mode (YM-244769 is a potent and highly selective NCX blocker that preferentially inhibits the reverse mode of NCX3.) [106]	Gly833 (α2-repeat; critical) [48]
Activator	Neurounina-1	NCX1/NCX2 (Neurounina-1 stimulated NCX1 and NCX2 activities with an EC50 in the picomolar to low nanomolar range, whereas it did not affect NCX3 activity.) [107]	Both modes (Neurounina-1 displayed a potent and reversible stimulatory effect on NCX1 and NCX2 in both forward and reverse modes of operation.) [107]	Val118 (α1-repeat; critical), Asn125 (α1-repeat; critical), Leu808 (α2-repeat; critical) [107]

XIP: exchanger inhibitory peptide. SN-6: (2-[4-(4-nitrobenzyloxy) benzyl] thiazolidine-4-carboxylic acid ethyl ester). SEA0400: (2-[4-[(2,5-difuorophenyl) methoxy]phenoxy]-5-ethoxyaniline).

Fig. 3.

Structural determinants underlying isoform and transport mode selectivity of Na⁺/Ca²⁺ exchanger (NCX) modulators. Amino acid sequences and residue numbering are based on the human NCX1 isoform. Key residues involved in inhibitor and activator binding are mapped onto the α1- and α2-repeat regions and the exchanger inhibitory peptide (XIP)-associated regulatory domain. These regions are structurally linked to conformational transitions between inward- and outward-facing states, suggesting that ligand binding may influence mode selectivity by stabilizing specific conformations of the exchanger.

In 2001, SEA0400 (2-[4-[(2,5-difuorophenyl) methoxy]phenoxy]-5-ethoxyaniline) was developed as a more potent and selective NCX1 inhibitor [75]. SEA0400 potently inhibits Na⁺-dependent Ca²⁺ influx via the reverse mode of NCX1 in cultured neuronal cells [75] and cardiomyocytes [105], with IC₅₀ values of 5–33 nM and 92 nM, respectively. This represents an inhibitory potency approximately 80–100 times greater than that of KB-R7943 [49]. SEA0400 exhibits excellent specificity with minimal effects on other ion channels, transporters, and receptors. The amino acid residue Phe213 in TM5 has been identified as a critical determinant of SEA0400-mediated inhibition [49], whereas Gly833 in the α2-repeat is thought to contribute indirectly to inhibitor sensitivity [99]. Structural analyses indicate that SEA0400 does not directly bind to the XIP region but instead inhibits Ca²⁺ transport through an allosteric mechanism involving the TM1-2 region [24]. Recent studies have shown that SEA0400 reduces cerebral infarct volume in a transient middle cerebral artery occlusion model [75] and attenuates ischemia-reperfusion injury in the heart [71, 105] and kidneys [85]. These findings suggest that the protective effects of SEA0400 against ischemic injury may be superior to those of KB-R7943 [49].

Consistent with this mechanism, benzyloxyphenyl inhibitors such as SEA0400 are thought to exert their effects through allosteric modulation of the TM domain rather than direct interaction with the XIP region [24]. In 2002, SN-6 (2-[4-(4-nitrobenzyloxy) benzyl] thiazolidine-4-carboxylic acid ethyl ester) was developed as a derivative of KB-R7943 [47]. SN-6 has a chemical structure similar to that of KB-R7943 and inhibits Ca²⁺ influx via NCX1 with an IC₅₀ of 2.9 µM, showing 3- to 5-fold grater potency compared with NCX2 or NCX3 in NCX-transfected fibroblasts [47]. At concentrations of 0.3–30 µM, SN-6 does not significantly affect the Ca²⁺ efflux, indicating preferential inhibition of the reverse mode. Accordingly, SN-6 exhibits greater selectivity than KB-R7943 [47]. Amino acid residues Val227 and Tyr228 in XIP region, as well as Gly833 and Asn839 in the α2-repeat, are critical determinants for SN-6 inhibition, whereas Phe213 in the TM5 domain acts as an auxiliary site [47]. The benzyloxyphenyl NCX inhibitors—KB-R7943, SEA0400 and SN-6—display distinct isoform selectivity profiles [84]. KB-R7943 is more effective against NCX3 than NCX1 or NCX2 [42], whereas SEA0400 preferentially inhibits NCX1 over NCX2 and has minimal effect on NCX3 [45]. SN-6 inhibits NCX1 more potently than NCX2 or NCX3 [47].

In 2006, YM-244769 was identified as a novel inhibitor with improved selectivity [48], preferentially inhibiting Na⁺-dependent Ca²⁺ influx via NCX3 (3.8- to 5.3-fold greater than for NCX1 or NCX2) [48]. Amino acid residue Gly833 in the α2-repeat is a key determinant for YM-244769-mediated inhibition [48]. All these inhibitors share a benzyloxyphenyl moiety, suggesting that this structural feature is important for NCX affinity [46]. These compounds preferentially inhibit the reverse mode of NCX1 under unidirectional conditions [44]. In contrast, the NCX activator Neurounina-1, which preferentially enhances forward-mode NCX1 activity, require distinct structural determinants, including Val118 and Asn125 in the α1-repeat and Leu808 in the α2-repeat [80]. These features may underlie their preferential activity under pathological conditions characterized by elevated intracellular Na⁺ levels [46].

Other highly selective inhibitors include ORM-10103 [54], ORM-10962 [59], and GYKB-6635 [29, 59]. Although many NCX inhibitors have been limited by poor solubility for in vivo use [87], ORM-10962 represents an exception, exhibiting improved solubility [87]. Its inhibitory action is functionally asymmetrical, with a more pronounced effect on the reverse mode [86]. Furthermore, ORM-11372, identified through virtual screening of flavan derivatives [87], is currently the most potent and selective NCX1.1 inhibitor reported to date. It inhibits both reverse- and forward-mode currents with similar potency in human iPS cell-derived and rat cardiomyocytes [87]. Another significant development is SAR340835, a water-soluble prodrug of SAR296968 [90]. Following intravenous administration, SAR340835 is rapidly and completely converted to its active moiety, SAR296968, within 30 min [90]. SAR296968 has been identified as a selective NCX inhibitor across species, including humans, with no detectable effects on native voltage-dependent Ca²⁺ and Na⁺ currents in vitro [90]. In CHO cells expressing different NCX isoforms, SAR296968 exhibits high potency against NCX1, with comparable efficacy toward NCX2 and NCX3 [90]. Thus, the NCX-inhibitory activity of SAR340835 is attributable to its active metabolite SAR296968.

Parallel to the development of SAR340835, structure-activity relationship (SAR) studies have led to the identification of SM-15811, a structurally novel and highly potent NCX inhibitor [37]. SM-15811 was developed from a 3,4-dihydro-2 (1H)-quinazolinone derivative (compound 1a), as the lead structure [37]. Although the initial lead compound exhibited only moderate inhibitory activity, optimization of this scaffold resulted in SM-15811, which shows approximately two orders of magnitude increase in inhibitory potency compared with the patent compound. Experimental studies demonstrate that SM-15811 concentration-dependently attenuates increase in fura-2 fluorescence ratios induced by Na⁺-free conditions, indicating direct inhibition of Na⁺-dependent Ca²⁺ influx via NCX in cardiomyocytes [37]. Furthermore, SM-15811 exhibits high selectivity for NCX over other essential ion channels, including voltage-gated Na⁺ and K⁺ channels [100]. Collectively, these advances in NCX inhibition development provide important insights for the design of future therapeutic strategies targeting Ca²⁺-related pathologies.

Although a diverse array of organic NCX inhibitors has been developed, research on NCX activators remains limited [100]. Neurounina-1 (7-nitro-5-phenyl-1-(pyrrolidin-1-ylmethyl)-1H-benzo[e] [1, 4]diazepin-2 (3H)-one), a derivative generated through structural optimization of SM-15811, has demonstrated promising pharmacodynamic properties [80]. Neurounina-1 potently and reversibly enhances the activity of NCX1 and NCX2 at low nanomolar concentrations [100]. Notably, it more effectively stimulates the forward mode (Ca²⁺ efflux) of NCX1 than that of NCX2 and exhibits minimal activity toward the NCX3 isoform [100]. Accordingly, Neurounina-1 represents a promising lead compound for future therapeutic development.

STRUCTURAL DETERMINANTS OF MODE AND ISOFORM SELECTIVITY IN NCX MODULATORS

Despite the growing number of NCX modulators identified to date, the structural basis underlying their mode and isoform selectivity remains incompletely understood. A systematic integration of pharmacological and structural findings is therefore required to elucidate how ligand binding influences NCX function.

As illustrated in Fig. 3, determinant residues for NCX modulators are primarily localized within the α-repeat regions and the XIP-associated regulatory domain, both of which are structurally coupled to conformational transitions of the exchanger. These regions undergo dynamic rearrangements during transitions between inward- and outward-facing states, suggesting that ligand binding at these sites may shift the equilibrium between transport modes. This distribution is consistent with observations in other ion transporters, where small molecules typically interact with structural elements that regulate ion translocation rather than directly targeting the ion-binding site itself. In particular, many benzyloxyphenyl inhibitors, such as KB-R7943 and SEA0400, depend on residues within the α2-repeat, which is closely associated with the Na⁺-binding and inward-facing conformations. This may explain their preferential inhibition of the reverse mode, which involves Na⁺-loaded inward-facing states and suggests functional asymmetry between the α1- and α2-repeats. In contrast, activators such as Neurounina-1 require determinant residues in both α1- and α2-repeats, indicating that activation involves coordinated modulation of multiple structural elements rather than stabilization of a single conformational state. The isoform specificity and transport mode selectivity of representative NCX modulators (Table 1) further support this structural framework, providing a basis for interpreting the mechanistic diversity of NCX modulation. Collectively, these findings suggest that the α1- and α2-repeats play distinct yet cooperative roles in regulating ion exchange activity and represent key structural determinants for pharmacological modulation of NCX.

SN-6 and SEA0400 primarily interact with the XIP region, which is involved in the regulation of NCX activity, suggesting an allosteric mechanism rather than direct interaction with the ion translocation pathway. These compounds are proposed to stabilize an inactivated state, potentially through interactions that couple the XIP region with other structural elements, such as the CH2 helix. Taken together, these findings support a model in which NCX modulators target multiple, functionally coupled structural modules—including both α-repeats and regulatory domains—highlighting the importance of conformational regulation rather than direct blockade of the ion-binding site. However, the precise structural basis linking ligand binding to mode selectivity remains incompletely understood, and further studies integrating structural and functional analyses will be required to establish a comprehensive structure–activity relationship for NCX modulators.

CONCLUSIONS AND FUTURE PERSPECTIVES

Calcium signaling plays a central role in a wide range of physiological and pathological processes, and dysregulation of intracellular Ca²⁺ homeostasis is closely associated with various diseases. Given its unique bidirectional transport properties, NCX represents an attractive therapeutic target for modulating intracellular Ca²⁺ dynamics. However, despite extensive research efforts, the development of clinically effective NCX-targeting drugs remains limited.

Despite more than three decades of research on NCX inhibitors, no clinically approved drugs have been established, highlighting several fundamental pharmacological challenges. First, reverse-mode inhibitors such as KB-R7943 and SEA0400 have demonstrated protective effects in pathological conditions associated with Ca²⁺ overload. However, their limited selectivity and potential off-target effects remain major obstacles. In addition, because NCX operates bidirectionally, inhibition of the reverse mode may also interfere with physiological Ca²⁺ extrusion under certain conditions, raising concerns about safety and context-dependent effects. Second, NCX activators represent a promising but underexplored strategy. In contrast to inhibitors, activators such as Neurounina-1 may enhance forward-mode activity and promote Ca²⁺ extrusion, thereby preventing intracellular Ca²⁺ accumulation. This pharmacological profile suggests a fundamentally different therapeutic approach from reverse-mode blockade. However, the mechanisms underlying activation, as well as isoform selectivity and long-term effects, remain poorly understood. Third, recent structural insights into NCX inactivation mechanisms provide a potential new direction for drug development. The identification of regulatory elements such as the XIP region, β-hub, and CBD2 suggests that modulators targeting the inactivated state could achieve more precise functional control. Such compounds may regulate NCX activity by stabilizing or destabilizing specific conformational states, rather than directly blocking ion transport.

Overall, future development of NCX modulators will require an integrated understanding of structural dynamics, mode selectivity, and physiological context. In particular, the relationship between conformational states and pharmacological modulation remains a key unresolved issue, and addressing this will be essential for translating NCX-targeting strategies into clinically effective therapies.

CONFLICTS OF INTEREST

The authors declare no conflicts of interest.