Bacterial voltage-gated sodium channels (BacNa_Vs) from the soil, sea, and salt lakes enlighten molecular mechanisms of electrical signaling and pharmacology in the brain and heart

Abstract

Voltage-gated sodium channels (Na_Vs) provide the initial electrical signal that drives action potential generation in many excitable cells of the brain, heart, and nervous system. For more than 60 years, functional studies of Na_Vs have occupied a central place in physiological and biophysical investigation of the molecular basis of excitability. Recently, structural studies of members of a large family of bacterial voltage-gated sodium channels (BacNa_Vs) prevalent in soil, marine, and salt lake environments that bear many of the core features of eukaryotic Na_Vs have reframed ideas for voltage-gated channel function, ion selectivity, and pharmacology. Here, we analyze the recent advances, unanswered questions, and potential of BacNa_Vs as templates for drug development efforts.

Introduction

Heartbeats, thoughts, and sensations of pleasure and pain begin with the seemingly simple act of the opening of an ion-selective hole in a cell membrane that allows an inward rush of sodium ions. This influx causes a change in the membrane potential within the timescale of milliseconds and initiates the electrical signaling cascade know as the ‘action potential’ that is the signature electrical behavior of excitable cells such as a neurons and muscle ¹. A specialized class of transmembrane proteins, known as voltage-gated sodium channels (Na_Vs), forms the conduits for this rapid ion influx. Biophysical characterization of Na_Vs and elucidation of their functional roles in excitable cells have been a pillar of physiological studies for over 60 years ^{2; 3; 4}. The importance of Na_Vs in human biology is profound. This ion channel class is linked to a multitude of ailments including cardiac arrhythmias, movement disorders, pain, migraine, and epilepsy ⁵ and is the target for a host of pharmaceuticals and ongoing drug development efforts ⁶. Moreover, it is becoming increasingly clear that Na_Vs play a role in many cells that are not traditionally thought of as excitable, such as astrocytes, T cells, macrophages, and cancer cells ⁷. Hence, the need to understand the mechanics of how such channels function, the molecular basis for their activity, and the development of new tools that can probe and control their function remains exceptionally high.

Na_Vs are found in metazoans from jellyfish to humans and are formed by large polytopic transmembrane proteins that are members voltage-gated ion channel (VGIC) signaling protein superfamily ^{1; 2}. This class encompasses voltage-gated channels for sodium, calcium, and potassium, the large family of transient receptor potential (TRP) channels, and a variety of other ion channels (Fig. 1a). The eukaryotic pore-forming Na_V subunit is composed of a single polypeptide chain of ~2,000 amino acids (~260 kDa) comprising four homologous transmembrane domains (Fig. 1b) and, along with that from voltage-gated calcium channels, Ca_Vs, represents the largest pore-forming polypeptide within the superfamily. Nine Na_V isoforms are found in humans (Na_V1.1-Na_V1.9) and have differing pharmacologies, expression patterns, and functional signatures ⁸. In addition to the pore-forming subunit, native channels associate with a class of single-pass transmembrane Na_Vβ subunits. These auxiliary subunits affect function, pharmacology, and can carry mutations can cause disease ⁹.

Figure 1. BacNa_V topology and relationships to voltage-gated ion channel (VGIC) superfamily members.

(a) Unrooted tree showing the amino acid sequence relations of the minimal pore regions of VGIC superfamily members (modified from ¹⁰). Indicated subfamilies are (clockwise): voltage-gated calcium and sodium channels (Ca_V and Na_V), two-pore (TPC) and transient receptor potential (TRP) channels, inwardly rectifying potassium channels (Kir), calcium-activated potassium channels (K_Ca), voltage-gated potassium channels (K_V1–K_V9), K_2P channels, voltage-gated potassium channels from the EAG family (K_V10–K_V12), cyclic nucleotide gated channels (CNG), and hyperpolarization activated channels (HCN). “R” indicates identifiable regulatory domains. (b) Topology diagram comparing eukaryotic Na_V (top) and BacNa_V pore forming subunits. S1-S6 segments are labeled. Individual Na_V six transmembrane repeats of are colored black, blue, orange, and teal. BacNa_V neck and coiled coil (CC) domains are indicated. (c) Sequence alignment for selected BacNa_Vs selectivity filter and pore helices, mammalian Ca_V subtype exemplars, and mammalian Na_V1.4 and Na_V1.7. Selectivity filter numbering is indicated. Position (0) is highlighted in dark orange. Residues involved in selectivity are highlighted light orange. Grey highlights the the conserved Trp (+2) anchor position. (d) Unrooted tree showing a comparison based on the selectivity filter sequences for BacNa_Vs compared with K_V channels, CatSper, Protist one-domain channels, and the individual domains of Na_Vs and Ca_Vs (modified from ⁴³).

Each of the four Na_V transmembrane domains (Domains I-IV) has an architecture shared by many VGIC superfamily members ¹⁰ (Fig. 1a, b). Transmembrane segments S1-S4 form the voltage sensor domain (VSD), whereas transmembrane segments S5-S6 form the pore-forming domain (PD) that houses the element that defines the ion selectivity properties of the channel, the selectivity filter (SF). The intracellular loops that connect the Na_V transmembrane domains have important roles in channel regulation. The best studied are the III-IV loop, which bears a segment known as the inactivation peptide that is essential for the fast inactivation properties of metazoan Na_Vs ^{11; 12; 13}, and the cytoplasmic C-terminal tail, which forms a hub for binding of a number of regulatory factors including calmodulin ¹⁴. Together, these elements endow eukaryotic Na_Vs with complex functional properties and connect them to various regulatory pathways within the cell.

From the standpoint of ion channel biophysics, studies of Na_Vs have set a number of paradigms for understanding channel function including: the importance of the S4 segment of the VSD in voltage sensing, the concept of an intracellular ‘inactivation particle’, ideas that some hydrophobic drugs could access the channel pore by lateral access through the membrane hydrophobic bilayer, and the likely physical dimensions of the selectivity filter ¹. Yet, without structural data, it has been difficult to place such foundational ideas onto a molecular scaffold. Na_Vs, similar to many other eukaryotic membrane proteins, have been difficult to obtain in the quantities and qualities required for high-resolution structure determination. Because of their size and complexity, structural understanding of eukaryotic Na_Vs remains limited to low-resolution electron microscopy images of the complete protein isolated from natural sources, the electric organ of the eel Electrophorus electricus ¹⁵. However, there has been steady progress in obtaining structural information for specific domains including: the inactivation peptide ¹⁶, portions of the C-terminal cytoplasmic tail alone ^{17; 18}, C-terminal tail complexes with regulatory factors such as calmodulin ^{14; 19; 20; 21; 22}, and extracellular domains for two Na_Vβ auxiliary subunit isoforms ^{23; 24}. Elucidation of the architecture of these eukaryotic Na_V elements begins to flesh out key pieces of the Na_V molecular framework but have left the larger question of understanding the molecular structure of the central components of the ion-selective hole unaddressed.

For potassium channels, the biochemical tractability and relative simplicity of bacterial potassium channels was essential for opening the first paths to high-resolution structural studies ^{25; 26; 27; 28}. The discovery of a large family of bacterial Na_Vs (BacNa_Vs) ^{29; 30; 31; 32; 33} gave the Na_V field a simplified scaffold to begin to outline key structural principles of Na_V function and the substrate for the first structural insights into this channel class. BacNa_Vs have ~275 residues, making them ~1/8^th the size of a eukaryotic Na_V pore-forming subunit. Rather than having the twenty-four transmembrane segment architecture of eukaryotic Na_Vs, BacNa_Vs are built from a six transmembrane segment architecture comprising a VSD and PD (Fig. 1b) that assembles into homotetramers ^{34; 35; 36; 37; 38; 39} in a manner similar to many voltage-gated potassium channels ¹. Initial studies demonstrated that BacNa_Vs had an ion selectivity profile that was similar to Na_Vs ^{29; 40}, even though the actual selectivity BacNa_V selectivity filter sequence has more in common with those from Ca_Vs than Na_Vs ^{29; 33; 41} (Fig. 1c). It is interesting that although BacNa_Vs have been posited as ancestors of eukaryotic Na_Vs ⁴², clade analysis places them on a different evolutionary branch that is closer to a calcium channel family found in sperm known as CatSper ⁴³ (Fig. 1d) and in a position consistent with the original identification strategy, which was a CatSper-based database search ²⁹. Regardless of the precise evolutionary connections, the initial report of a functional bacterial homolog of Na_V/Ca_V branch of the VGIC superfamily (Fig. 1a), named NaChBac ²⁹, was a critical turning point for the field and held the promise that it would ultimately yield a high-resolution crystal structure that would enlighten understanding of its eukaryotic relatives ^{29; 44}.

Breakthrough BacNa_v structures

Realization that the BacNa_V family is very large, having >500 identifiable members, together with functional characterization a variety of BacNa_Vs has helped established that these channels share many important features traditionally associated with canonical vertebrate Na_Vs and Ca_Vs including: voltage-dependent activation, slow inactivation, ion selectivity, and drug block ^{29; 30; 32; 40; 45; 46; 47; 48}. These shared functional characteristics imply a significant structural conservation across 3-4 billion years of evolution and suggest that understanding BacNa_V architecture should provide good models for defining core features of the eukaryotic members of the Na_V and Ca_V branch of the VGIC superfamily. Unlike the large, ~2000-3000 residue pore-forming subunits of vertebrate Na_V and Ca_V channels, it has been possible to overexpress and purify large quantities of a variety of bacterial ion channels as stable biochemical samples suitable for crystallization studies ^{25; 26; 49; 50; 51; 52; 53; 54; 55}. The demonstration that some BacNa_Vs shared these biochemical properties ^{35; 36; 56}, together with the possibility to leverage diversity-based strategies ⁵⁷ facilitated by the multitude of BacNa_V sequences elevated the hopes that studies of this family would yield to structural characterization.

In 2011, a landmark study unveiled the first BacNa_V structure, a mutant, I217C, of Na_VAb from Arcobacter butzleri at 2.7Å resolution crystallized from a lipid-based bicelle system ³⁷. A virtual explosion of BacNa_V structures has since followed, fulfilling the promise that these bacterial proteins would shed light on fundamental relationships within the VGIC superfamily. Three additional full-length BacNa_V structures have been reported subsequently: Na_VRh from Rickettsiales sp. HIMB114 crystallized from a detergent-lipid mixture in an asymmetric conformation and determined at 3.05Å resolution ³⁹; wild-type Na_VAb crystallized in distinct and asymmetric conformations and determined at 3.2Å resolution ³⁸; and Na_VCt from Caldalkalibacillus thermarum reconstituted in lipid bilayers and determined at 9Å resolution by electron crystallography ⁵⁸. A novel protein-engineering strategy akin to surgical removal of the VSDs ^{35; 36} has also lead to “pore-only” structures crystallized from detergent solutions for Na_VMs from Magnetococcus marinus MC-1 and solved at 3.49Å ⁵⁹ and 2.9Å resolution ⁶⁰, and Na_VAe1p from Alkalilimnicola ehrlichii determined at 3.46Å resolution ⁴¹. Crystallographic and physiological studies have been further combined to study a highly Ca²⁺-selective form of the parental Na_VAb channel (nicknamed ‘Ca_VAb’), which has provided insight into the structural basis for ion selectivity in calcium channels ⁶¹. Most recently, crystallographic and computationally-derived models of small molecule drugs bound to the Na_VMs channel pore have provided a first glimpse into how some drugs may bind Na_V and Ca_V channels ⁶². Together these studies highlight the versatility and advantage of employing the relatively “simple” BacNa_V channels as a model VGIC system. Considering how far the structural characterization of BacNa_Vs has advanced in recent years, we anticipate many exciting advances in years to come. Here, we review the available BacNav structures in the context of historical physiological data and ask how these structures might help direct future experiments and on-going drug discovery efforts.

Defining the BacNav architecture

The BacNav structures cement the concept that all VGICs share a conserved architecture (Fig. 2a-c) in which four subunits or homologous domains create a central ion-conducting pore domain (PD) surrounded by four voltage sensor domains (VSDs) ^{37; 38; 39; 41; 58; 59; 62}. The VSDs are composed of the S1-S4 segments, where the S4 places highly conserved arginine residues within the membrane electric field that undergo outward movement upon depolarization, giving rise to the phenomena of the “gating currents” ^{63; 64; 65; 66; 67; 68; 69}. The BacNa_V structures also confirm the commonality of the domain-swapped quaternary arrangement, first seen in the K_V1.2/2.1 chimera structures ^{70; 71}, whereby the VSD of one subunit packs alongside the PD of the neighboring subunit (Fig. 2b, c). This domain-swapped organization poses a fantastic topological conundrum that must be solved every time a VGIC folds into the membrane. Mechanistically, it also raises the possibility that movement the S4-S5 linker caused by outward translocation of S4 impact more than one pore domain subunit and enhances cooperativity among the channel subunits during gating. In the BacNa_V PD (Fig. 2a-b), the S5 helices surround the pore-lining S6 helices and are connected through a critical helix-loop-helix motif. Together, these P1 and P2 helices form the selectivity filter and extracellular vestibule that appears to represent a conserved and defining characteristic shared with eukaryotic Na_Vs and Ca_Vs (Fig. 1c).

Figure 2. Overall structure of BacNa_V channels.

(a) Composite full-length BacNa_V structure generated by aligning the Na_VAe1p structure containing the neck and coiled-coil region (PDB: 4LTO) ⁴¹ onto Na_VAb (PDB 3RVY) ³⁷. Key structural and functional features of the BacNa_V channels are labeled including the voltage-sensor domain (green) (VSD), pore domain (slate) (PD), S4/S5 linker (red), C-terminal domain (slate), neck and coiled coil (orange), selectivity filter (yellow), and S6 activation gate. General locations of pharmacologically relevant sites in eukaryotic Na_V channels are also indicated in italics. Black lines indicate approximate boundaries of the membrane bilayer. For clarity, one pore subunit and voltage sensor domain are not shown. (b) Extracellular and (c) intracellular views of the BacNa_V channel highlighting basic functional elements and the domain swapped arrangement of the VSD around the PD of a neighboring subunit.

The cytoplasmic domain (CTD) that follows the pore-lining S6 transmembrane helix has two domains (Figs. 1b and 2a): a membrane proximal region termed the ‘neck’ region ⁴¹ and a C-terminal coiled-coil domain (CCD) ^{33; 41; 72}. Although the entire CTD has been present in the protein constructs used for Na_VAb ^{37; 38}, Ca_VAb ⁶¹, Na_VCt ⁵⁸, one Na_VMs ‘pore-only’ construct ⁶⁰, and Na_VAe1p ⁴¹, electron density revealing its structure and relationship to the BacNa_V PD has only been reported for Na_VAe1p ⁴¹. The CTD is unique to the BacNa_Vs compared to their vertebrate Na_V and Ca_V channel cousins; however analogous structures are seen in other tetrameric channels in the VGIC family, such as Kv7 (KCNQ) ^{73; 74; 75} and TRP channels ⁷⁶.

The BacNa_V channel pore

The pore domain (PD) forms the heart of a VGIC that controls ion selectivity and ion passage across the membrane (Fig. 2a). The large collection of BacNa_V structures all reveal the same basic PD fold (Fig. 3a). Two transmembrane helices, S5 and S6, are bridged by the pore helices P1 and P2 linked by the selectivity filter (SF). The P1-SF-P2 structure forms the channel ‘active site’ required for engaging and selecting the permeant ions, whereas S6 lines the pore and provides the structure that closes the intracellular activation gate of the channel.

Figure 3. Comparisons of BacNa_V PD structures and ion binding sites.

(a) Ribbon diagram of a PD backbone superposition of Na_VAb (3RVY) ³⁷ (black), Na_VAbA/B (4EKW) ³⁸ (light gray), Na_VAbC/D (4EKW) ³⁸ (medium gray), Na_VRh (4DXW) ³⁹ (light green), Na_VMs (4FLF)⁵⁹(dark red), Na_VMs (3ZJZ) ⁶⁰ (magenta), Na_VAe1p (4LTO) ⁴¹ (orange), Na_VCt (4BGN) ⁵⁸(copy A, marine and copy B, slate), and Ca_VAb (4MS2) ⁶¹ (white), Outer ion from Na_VAe1p, inner ion from Na_VRh, and selectivity filter ions from Ca_VAb are shown as orange, light green, and white spheres, respectively. Two subunits are shown. Selectivity filter (SF), P1 (P helix), P2, S5, and S6 elements are labeled. Location of the intracellular gate is indicated. (b) Cylinder diagram of the superposition of a single PD subunit from Na_VAb (3RVY) ³⁷ (grey), Na_VRh (4DXW) ³⁹ (light green), Na_VAe1p (4LTO)⁴¹ (orange), and KcsA (4EFF) ⁷⁷ (blue). Selectivity filter (SF), P helix, P2, S5, and S6 elements are indicated. KcsA M1 and M2 correspond to BacNa_V S5 and S6, respectively. (c) Equivalent views of the (top) Na_VAb (PDB: 3RVY) ³⁷ and (bottom) KcsA (PDB 1K4C) ²⁷ selectivity filters. In contrast to the carbonyl-lined filter in KcsA, the Na_VAb selectivity filter is wider and lined by two sidechains: E177 (or site 0, colored yellow) and S178 (site +1) (green). The highly conserved Thr residue at the end of the P1-helix forms part of the Site ‘4’ K⁺ binding site in potassium channels, but in BacNa_Vs the equivalent Thr sidechain is oriented to interact with a Trp sidechain in the selectivity filter (not shown). For simplicity, only two subunits are shown and all other sidechain residues are omitted. (d) BacNa_V SF crystallographically defined ions. PDs of Na_VRh (4DXW) ³⁹ (light green), Na_VAe1p (4LTO) ⁴¹, and Ca_VAb (4MS2) ⁶¹ (white), are shown. Boxed numbers indicate selectivity filter residue positions. ‘1’, ‘2’, and ‘3’ label the Ca_VAb SF ions. Outer and Inner ion binding sites are labeled.

Comparison among the BacNa_V monomer structures highlights the extremely high similarity in the basic tertiary structure of the individual BacNa_V PD subunits (Table 1). For most superpositions, the differences in the Cα positions are well below 1.0 Å RMSD (Fig. 3a). The biggest deviations are with Na_VRh ³⁹ and are largely due to a slightly different position of the S5 helix in this structure (Fig. 3a). Notably, the deviation of the Na_VRh PD structure from the PD consensus is greater than that observed for PD structures of Na_VMs, which have been suggested as models of an open conformation ^{59; 60}.

Table I.

Comparisons of BacNa_V Pore domain structures

	PDB		NaVAb I217C	NaVAb WT A/B	NaVAb WT C/D	NaVRh G208S	NavMsp1	NavMsp2	NaVAe1p	CaVAb	NaVCt1	NaVCt2	KcsA
NaVAb I217C	3RVY	Monomer-Monomer	*	1.14	0.98	1.57	0.85	1.27	1.04	0.32	0.79	1.12	3.22	Tetramer-Tetramer
NaVAb WT A/B	4EKW		0.51	*	1.84	2.11	1.53	1.98	1.49	1.24	1.34	1.74	3.61
NaVAb WT C/D	4EKW		0.56	0.69	*	1.72	0.87	1.16	1.37	0.91	1.20	1.27	3.09
NaVRh G208S	4DXW		1.09	1.03	1.19	*	1.56	1.96	2.05	1.47	1.70	1.38	2.98
NaVMs1	4F4L		0.72	0.85	0.67	1.20	*	0.83	1.14	0.81	1.14	1.20	3.11
NaVMs2	3ZJZ		1.01	1.19	0.91	1.70	0.59	*	1.43	1.24	1.42	1.29	3.17
NaVAe1p	4LTO		0.92	1.04	1.03	1.36	0.83	0.89	*	1.05	1.19	1.56	3.14
CaVAb	4MS2		0.30	0.48	0.59	1.01	0.70	0.99	0.87	*	0.77	1.05	3.02
NaVCt1	4BGN		0.29	0.56	0.56	1.10	0.70	0.96	0.93	0.24	*	0.83	3.18
NaVCt2	4BGN		0.28	0.50	0.67	1.03	0.72	0.94	0.88	0.70	0.26	*	3.23
KcsA	4EFF		1.86	1.73	1.73	2.08	1.86	1.92	1.86	1.66	1.65	1.62	*

Superpositions of the PDs of the indicated BacNa_Vs based on Na_VAb 130-219, Na_VAe1p 150-239, and the corresponding residues in each of the entries.

Monomer vs. monomer – orange

Tetramer vs. tetramer – blue

Na_VCt1 and Na_VCt2 are molecules ‘A’ and ‘B’⁵⁸, respectively.

Given the presence of the PD in all members of the VGIC superfamily (Fig. 1a), we thought it would be interesting to examine the BacNa_V PD tertiary fold in light of the PD of the full-length structure of the prototypical potassium channel, KcsA ⁷⁷ (Fig. 3b, Table 1). This comparison reveals the striking conservation of the core tertiary fold of an individual PD subunit (Fig. 3b). Although there are some notable differences between BacNa_Vs and potassium channels, such as the backbone conformation of the selectivity filter and presence of a P2 helix, it is clear from the comparison the essential elements and organization of the PD fold are the same. The S5/S6 transmembrane helix pair forms a platform for the P-helix that leads into the loop forming the selectivity filter. This core structure has also recently been described in the first structure of a TRP channel ⁷⁸, further establishing that this basic PD fold should be present in all members of the VGIC superfamily. The crossing angle of the P-helix relative to the two transmembrane segments is different between the BacNa_V and KcsA exemplars and may be related to the requirement for the filter diameters to be different in order to select different cations. Indeed, the BacNa_V extracellular opening is wider than that in potassium channels and its SF is sufficient to hold the potassium channel SF ³⁷ (Fig. 3c). The BacNa_V PD P2 helix follows the selectivity filter and is an element absent from known potassium channel structures. Regardless of this difference, it is striking that the core fold of the monomer is so similar, even though the details of the selectivity filters and how they recognize ions (below) are dramatically different.

The shared features of the PD fold point to a common origin and raise the question: ‘Given the apparent constraints of the basic PD fold, what is the range of structural diversity that can be accommodated in the SF region?’ Exploration of proteins having unconventional selectivity filters, such as the related bacterial potassium transporter family TrkH which has the same basic PD fold ^{79; 80} and channel properties ⁸¹, engineered channels designed to test ideas about selectivity ^{82; 83; 84}, or wholesale replacement of the SF in the context of a genetic selection or computational study may help to answer this question and will be important for establishing the ground rules for eventual de novo design of channels having novel properties.

The PD tertiary architecture appears very robust, as experimental and computational studies of K_V1.3 potassium channel biogenesis indicate that the basic PD fold can adopt a near native-like tertiary fold in the absence of assembling into a quaternary structure ⁸⁵. This tantalizing result opens up questions about what happens to the PDs during biogenesis while an individual PD waits to encounter three other PDs to form a complete pore from either disparate chains, as in BacNa_V, TRP, and K_V channels, or from PDs embedded in very long gene transcripts as in eukaryotic Na_Vs and Ca_Vs. Are there ways the cell can shield this partially formed hole from misfolding, degradation, or aggregation? Does the apparent stability of the PD tertiary fold accelerate assembly? Further, it raises the question about whether there is a ‘non-channel’ ancestor of the PD fold that has some function outside of the now familiar four-fold arrangement. Can this fold act in a monomeric capacity for some yet uncharacterized function?

When assembled around the central axis that forms the ion conduction pathway, the four subunits of the VGIC PD form a central cavity that is bounded by the selectivity filter on the extracellular side and a constriction made by the S6 pore lining helices on the intracellular side. This second region is thought to form the principal barrier that is controlled by the VSDs and that must be opened in order for ions to pass through the channel. The observed PD conformations in the various BacNa_V structures have been suggested to represent closed ^{37; 41; 58}, inactivated ^{38; 39}, and open ^{58; 59; 60} conformations; however, comparison of the PD quaternary structure ⁴¹ (Table 1) reveals that despite some small differences, all of these backbone conformations are very similar, generally having RMSDs for the Cα positions that are ≤1.5Å. By contrast, the strong BacNa_V PD monomer similarities with the KcsA fold are washed out if one considers how the BacNa_V tetramers compare with an intact potassium channel (Table 1). Hence, even though the core PD subunits are very similar, there are clearly distinctly different ways to arrange the S5 and S6 segments around a closed pore.

When one considers that the quaternary structures of all of the BacNa_V PDs are much more similar than they are different, in spite of suggestions about the different possible states that these structures may represent, what is very striking is that none of the channel activation gates are as open as in the K_V1.2/2.1 chimera structure ^{71; 86}. It may be that, unlike K_V channels ⁷⁰, BacNa_V gating involves relatively subtle changes in the PD conformation. However, it should be noted that, when the BacNa_V VSDs are present in the structures, they are in the activated conformation ^{37; 38; 39; 58}. How can activated VSDs yield an open PD in one VGIC case ^{71; 86}, but closed PDs in all the others ^{37; 38; 39; 58; 87}? These observations may indicate something about the stability of the BacNa_V PD conformation, which may have a strong bias to close the intracellular gate, or be related to how the BacNa_V PDs couple to the VSDs. Clearly, there is a need for further studies of the energetic parameters that govern BacNa_V gating to answer these questions. This puzzle also underscores the fact that it maybe hard to identify a truly open BacNa_V PD structure unless it is bound to a known opener or contains well-characterized mutations that stabilize the open state.

Ion binding sites and their influence on ion selectivity

One of the main questions that determination of BacNa_V structures hoped to address was: ‘What is the origin of ion selectivity?’ Further, as the BacNa_V selectivity filter bears features common to both Na_Vs and Ca_Vs ^{29; 33; 40} (Fig. 1c), could the homomeric BacNa_V structures, having four identical selectivity filter segments, serve as prototypes to inform our understanding of eukaryotic Na_Vs or Ca_Vs in which the selectivity filters are necessarily heteromeric? Long standing ideas originating in careful biophysical studies of eukaryotic Na_Vs and Ca_Vs had set the expectation that members of this channel clade should use a selectivity mechanism that was based on sidechain chemistry ^{1; 88; 89; 90; 91} in contrast to the backbone mediated ion recognition mode used by potassium channels ^{1; 25; 27}. To facilitate comparison among Na_V, Ca_V, and BacNa_V SFs, the residue corresponding to the mammalian Na_V SF ‘DEKA’ motif ¹, the conserved SF ‘EEEE’ motif in Ca_V ^{1; 88}, and equivalent glutamate in BacNa_V SFs, which was also described as forming the high-field strength site in Na_VAb ³⁷ , are denoted as position “0” (Fig. 1c). The idea that the sidechains were crucial for selectivity was further supported by the evidence that one could change BacNa_V ion selectivity from sodium to calcium by making a triple-aspartate mutant at selectivity filter positions (0), (+1), and (+4) ^{35; 40; 61} (Fig. 1c). These expectations were all confirmed as the Na_VAb SF ³⁷ showed a structure much wider than that of a potassium channel and lined, in part, by sidechains rather than almost entirely backbone carbonyls ^{25; 27} (Fig. 3C).

Studies of eukaryotic Na_Vs and Ca_Vs provided strong evidence for a multi-ion mechanism ^{1; 92; 93; 94; 95; 96}, raising the prospect that if ions could be identified in the BacNa_V SF structures, there might even be multiple binding sites. However, in contrast to these expectations and the crystallographic results from potassium channel structures ^{25; 27; 71}, the first Na_VAb structure lacked identifiable ions in the pore ³⁷, suggesting the possibility of promiscuous ion coordination in the Na_V SF. The Na_VRh structure, which has a slightly unconventional selectivity filter sequence (Fig. 1c) provided crystallographic evidence for an inner ion binding site (Fig. 3d) formed from the backbone carbonyls of Leu (-1) and Thr (-2) at the C-terminal end of the P-helix that could be occupied by a partly hydrated calcium ion ³⁹, a barium ion ⁹⁷, or a rubidium ion ⁹⁷. Electron density for a partially hydrated calcium ion coordinated by the SF (+1) serine in the Na_VAe1p structure provided the first direct crystallographic evidence for an outer ion binding site at the mouth of the selectivity filter ⁴¹ (Fig. 3d).

The observation of two ion binding sites supports the idea that BacNa_Vs have multi-ion pores, an idea further validated by the recent structures of a Na_VAb mutant bearing the triple aspartate mutation that changes selectivity from sodium to calcium, termed ‘Ca_VAb’ ⁶¹. The Ca_VAb structures identified a series of three ion binding sites within the SF, denoted Site 1, Site 2, and Site 3, as well as two extracellular sites positioned above the Na_VAe1p ‘outer ion’ site (Fig. 3d). The SF in Ca_VAb revealed two high-affinity hydrated Ca²⁺-binding sites followed by a third lower-affinity hydrated site. Four carboxyl side chains from SF residue (+1) form Site 1 and have a critical role in determining Ca²⁺ selectivity ^{40; 41; 61}. Four carboxyls of the ‘DDDD’ motif at SF residue (0) plus four backbone carbonyls from SF residue (-1) form Site 2, a site also targeted by blocking divalent cations (e.g. Mn²⁺ and Cd²⁺). The lower-affinity Site 3 is formed by four backbone carbonyls from SF residue (-2) alone and mediates ion exit into the central cavity. In Ca_VAb, the multi-ion pore architecture is consistent with a conduction occurring by a multi-ion ‘knock-off’ mechanism of ion permeation through a stepwise-binding process that has been suggested to be conserved in Ca_V channels ⁶¹. In addition to this set of crystallographically identified ions in independently determined BacNa_V structures, other diffuse electron density has been reported in the SF of some structures that most likely arises from ions but that could not be assigned due to available resolution limits ^{41; 59}. Besides demonstrating that there are multiple ion binding sites in the BacNa_V filter, these studies suggest that that many, if not all, of the observed ions are at least partially hydrated. This assertion is in line with prior ideas about how Na_V and Ca_V SFs interact with ions ¹, but will require structures to be determined at a much higher resolution than has yet been possible in order to directly visualize these potential permeant ion properties.

The observation of a calcium ion bound to the outer ion site of Na_VAe1p also focused attention on a conserved aspartic acid in the selectivity filter of Domain II in all classes of eukaryotic Ca_Vs. Measurement of the effects of mutation of this position in the human cardiac Ca_V1.2 channel demonstrated that this residue is as important as the (0) position glutamate, which resides deeper in the selectivity filter and is a key determinant of ion selectivity ⁸⁸, strongly suggesting that it may interact directly with the permeant ion ⁴¹. This clear connection between BacNa_Vs and mammalian Ca_Vs highlights the point that BacNa_V filters are actually closer in sequence to Ca_Vs than they are to Na_Vs (Fig. 1c), and importantly, supports the idea that BacNa_Vs should be good model systems for understanding how eukaryotic SFs are built ³³.

The BacNa_V structures have provided an important template for a variety of computational studies directed at trying to understand basic aspects about ion selectivity and permeation behavior. In depth analysis of the many computational studies already reported is beyond the scope of this review, but some key general points are worth noting. Many of the analyses support the multi-ion nature of permeation through the basic BacNa_V SF scaffold ^{98; 99; 100; 101; 102; 103; 104}. In general, these studies have revealed a knock-on mechanism of ion permeation characterized by alternating occupancy of the channel by two or three hydrated sodium ions. In extended MD simulations (>μsec) ^{101; 104}, Na⁺ binding is coupled to an unexpected conformational isomerization of the four key glutamate side chains at position (0) of the selectivity filter to conformations not seen in any of the crystallographic structural studies. The coordination of variable numbers of Na⁺ ions and carboxylate groups leads to their condensation into ionic clusters of variable charge and spatial arrangement. These structural fluctuations result in a myriad of ion binding modes that foster a highly degenerate, energy landscape propitious to the diffusion of Na⁺ over K⁺ and Ca²⁺ ions. Thus, the resulting proposals for ion selectivity and conduction through the BacNa_Vs SF's are markedly distinct from the established mechanisms in highly selective K⁺ channel pores.

BacNa_V PD have lateral openings to the bilayer that facilitate hydrophobic modulator access

The most striking unanticipated feature of BacNa_V PDs was the observation of lateral openings between the PD transmembrane helices that appear to allow access to the hydrophobic core of the lipid bilayer ³⁷ (Fig. 4a). Such openings, which are not present in structures of bacterial potassium channels, inward rectifier potassium channels, or Kv channels, are enticingly consistent with earlier proposals that small molecules such as anesthetics access the channel pore through the lipid bilayer ^{1; 105; 106; 107; 108; 109}. Notably, similar structures have been found in the K_2P potassium channel class, a type of potassium channel that is known to be responsive to various types of anesthetics ^{110; 111; 112}.

Figure 4. BacNa_V PD fenestrations and pharmacology.

(a) Na_VAb PD ³⁷ is shown in surface representation sectioned through the middle. Exemplar side fenestrations are indicated by the arrows. The fenestration “gating” residue (Phe203) is shown as pink sticks. Lipids bound within the central cavity of Na_VAb are shown as cyan spheres and are seen penetrating through the pore fenestrations (arrows). For easy comparison, the Na_VRh PD (PDB 4DXW ³⁹, not shown) was superimposed onto Na_VAb and the bound lipid within the Na_VRh pore is shown as purple and red spheres. Light green background indicates approximate bilayer boundaries. (b) A sectioned view of Na_VAb I217C ³⁷ (left) and Na_VAb WT ³⁸ (right) looking into the central cavity, viewed from below the selectivity filter. Phe203 is shown as pink sticks and select sidechains implicated in drug binding and block in eukaryotic Na_V and Ca_V channels are in space filling color (blue, green and orange). The asymmetric central cavity seen in Na_VAb WT (right) has been suggested to represent a slow inactivated conformation of the pore, where a reshaping of the pore fenestrations and putative drug binding sites are seen. (c) Homology model of human Na_V1.7 based on the Na_VAb (PDB 3RVY). Left: select residues implicated in local anesthetic block in DIV S6 of Na_V channels (Phe1764 green and Tyr1771 orange) and a residue implicated in DHP block in DIII S6 of rat Ca_V1.2 (accession: P22002; Met1187 blue) illustrate a potential composite drug receptor site within the central cavity of eukaryotic Na_V and Ca_V channels. Right: sequence conservation analysis for all human Na_V channels (Na_V1.1- Na_V1.9) is mapped onto the Na_V1.7 homology model and demonstrate regions of high and low conservation in and around the central cavity.

The Na_VAb and Na_VRh structures show that these side portals are filled with lipid or detergent molecules (Fig. 4a) ^{37; 38; 39} and give support to the idea that such cavities have the capacity to be filled by something hydrophobic. Classic studies demonstrate that small hydrophobic pore blockers can gain access to or leave the central cavity site in the PD of a channel that has not opened ^{105; 106; 108; 109}. Similar experimental results have been obtained with BacNa_V block by lidocaine ⁴⁶ and further suggested by MD simulation ^{104; 113; 114}. Notably, Na_VMs has recently been shown to share overlapping pharmacology with human Na_V1.1 and crystallographic analysis of the ‘pore-only’ channel has produced plausible models for drug binding near the pore fenestrations within the central cavity ⁶². Unfortunately, the inherent symmetry of the Na_VMs pore and issues with experimentally resolving the drug conformations leaves considerable limitations in these drug-bound structural models. Nevertheless, an entire membrane phospholipid was seen bound within the Na_VRh channel central cavity ³⁹ and exemplifies how drug molecules might bind asymmetrically in eukaryotic Na_Vs (Fig. 4b-d).

Even though the side portals offer an exciting hypothesis for access to the channel cavity, their presence raises some puzzling questions. Given that they can be filled by lipids or lipid like molecules, is there some means to keep this interaction from happening in a physiological membrane bilayer? If they are filled with lipid, how is it that small molecules can pass through? And importantly, do changes in the conformation of the pore domain remodel these elements such that they could be the targets for natural or designed channel modulators (Fig. 4b-c)? A considerable plasticity is observed within the pore fenestrations of the available BacNa_V structures, particularly if the sidechain rotamers are considered, and suggests that these portals may be dynamic during gating (Fig. 4b). Intriguingly, mutation of specific pore residues in vertebrate Na_Vs allows a charged lidocaine derivate (QX-314) to enter the central cavity when applied extracellularly ¹⁰⁷. The equivalent BacNa_V residues do not obviously connect the central cavity and extracellular milieu, suggesting that pore remodeling must occur during gating. In the absence of an unequivocally open BacNa_V PD structure, these questions are sure to remain at the forefront of research. Nevertheless, the available BacNa_V PD structures clearly provide meaningful starting points for structure-based drug discovery efforts and unprecedented opportunities to design novel, isoform selective small molecule inhibitors for the treatment of Na_V channel related pathologies.

Structure-based sequence comparisons highlight key BacNa_V PD positions

Sequence searches identify the presence of >500 BacNa_Vs in the current sequence databases from many bacteria, including many soil, marine, and salt lake bacteria and some opportunistic pathogens. This sequence diversity together with the common BacNa_V structural framework provides the opportunity to investigate the positional conservation to identify key elements of the structure that have strong conservation and that hence, are important for some aspect of function. We used the positional informational analysis approach ¹¹⁵ to investigate which BacNa_V PD positions had the highest degree of statistical information. Because the Na_VAe1p structure ⁴¹ shows the complete pore, and all BacNa_V PDs have similar structures (Fig. 3, Table 1), we used the Na_VAe1p structure as the reference.

The two most highly conserved positions revealed by the analysis are two tryptophans (Fig. 5). One of these, Trp (+2) of the selectivity filter, Na_VAe1p Trp199, forms the key anchor position for the selectivity filter (Figs. 1c, 5b, 5c). This position makes interactions with the P-helix of the neighboring subunit ³⁷, including a hydrogen bond to the sidechain of Thr195 (-2), which is also highly-conserved (Fig. 5a, b). The Thr195 position is equivalent to the threonine found at the end of the potassium channel P-helix whose sidechain makes important contributions directly to ion binding ^{27; 83; 116; 117}; however, in BacNa_vs rather than interact with the permeant ions, this sidechain is repositioned to make a direct interaction with the conserved Trp (+2) sidechain. This interaction appears to be essential for buttressing the conformation of the SF in a way that permits the sidechains to be displayed in the ion conduction pathway. Both the SF anchor Trp and the hydrogen bonding capability of the (-2) position are strongly conserved in the eukaryotic Na_V and Ca_V SF sequences (Fig. 1c). The interfacial nature of the interaction between SF positions (+2) and (-2) may allow for it to be more than just an essential buttress for the SF conformation, but may allow SF conformational changes to be influenced by neighboring subunits.

Figure 5. BacNa_V PD sequence conservation.

(a) Conservation analysis of the BacNa_V PD, measured by the relative statistical entropy at each position, ¹¹⁵, mapped on the Na_VAe1p sequence. Highly conserved positions in rank order starting with the most conserved are colored dark blue, light blue, and green. Other positions of interest are colored yellow. (b) Conservation analysis depicted on two subunits of Na_VAe1p. Select positions of interest are shown as sticks and are labeled. (c) Spacefilling model of panel (b). (d) Extracellular view of a single PD subunit. Black circle marks the location of the ion conductive pore of the SF. (e) The highly conserved S6 Trp sidechain (Trp195 in Na_VAb, shown in green spheres; Trp215 in Na_VAe1p) is adjacent to a membrane phospholipid that appears to be well ordered as a lipid or detergent molecule in most available BacNa_V structures. Different channel subunits are colored white, yellow, and slate. The fourth subunit is not visible. Red mesh is from an F_O-F_C omit map calculated at 2.7 Å from PDB 3RVY with the “pore lipid” omitted from the calculation, contoured at 2.5σ. This density likely represents a phosphatidylcholine molecule that is present at high concentrations in the Na_VAb crystallization condition. (f) Sequence alignment of S6 segments from BacNa_Vs, Ca_V, and Na_Vs. Colors match those in panels ‘(a)-(d)’. Sequences identities are; Na_VBh1 (NaChBac)(NP_343367.1), 207-240; Na_VAb (YP_001490668.1), 211-244: Na_VRh (PDB:4DXW), 195-228; Na_VAe1 (YP_741167.1), 213-246; Na_VCt (WP_007502948.1), 210-243; Na_VMs (YP_864725.1), 115-129; Na_VSp1 (YP_165303.1), 192-225; Ca_V1.2 (CAA84346), IS6 380-413, IIS6 728-761, IIIS6 1141-1174, IVS6 1451-1484; Ca_V2.1 (NG_011569.1), IS6 335-368, IIS6 689-722, IIIS6 1488-1521, IVS6 1786-1819; Ca_V3.1 (O43497), IS6 370-403, IIS6 939-972, IIIS6 1512-1545, IVS6 1826-1860; Na_V1.4 (NP000325.4), IS6 422-455, IIS6 777-810, IIIS6 1270-1303, IVS6 1573-1606; Na_V1.7 (NM_002977), IS6 377-410, IIS6 932-975, IIIS6 1421-1454, IVS6 1724-1757.

The second most highly conserved PD position is also a Trp, Trp215 in Na_VAe1p (Trp195 in Na_VAb). This residue is at the extracellular top of the S6 segment (Fig. 5b-d). Strikingly, this residue is not involved in any protein-protein contacts. Why then should it be so strongly conserved? Further inspection of the BacNa_V structures shows that the Trp215 equivalent almost invariably has detergent or lipid molecules bound next to it (Fig. 5e). Although this Trp residue not conserved in Na_Vs (Fig. 5e), it is present in IS6 of high-voltage activated Ca_Vs (Fig. 5f), raising the possibility that this site may be involved in the lipid modulation of channel function which as been functionally described for both BacNa_Vs ¹¹⁸ and mammalian Ca_Vs ¹¹⁹. Other points of strong conservation are the residues that contribute to the central part of the PD tertiary core and that form intrasubunit contacts between the S5/S6 scaffold (S5, Phe180; S6, Tyr217, Phe218, Phe221, Ile222), the P-helix (Phe191), and loop (Phe180). These same scaffolding residues have also been implicated in transmitting the structural changes observed between the SF and S6 activation gate of Na_VAb I217C and Na_VAb WT channels, and were suggested to be important in the slow inactivation mechanism of BacNa_V and vertebrate Na_V channels ³⁸.

The initial set of BacNa_V structures did not indicate a common position for the intracellular constriction that should form the channel gate. Resolution of a the full-length Na_VAe1p PD structure revealed an extension of the S6 helix that had been absent from prior structures implicated position Met241 as the site of intracellular pore closure, a position corresponding to the suggested Na_VAb activation gate ³⁷. When tested in full-length Na_VSp1, which is similar to Na_VAe1p but more readily investigated using electrophysiology, functional studies showed that alanine mutation of the pore-lining S6 residues in each of the two helical turns above the Met241 equivalent had no effect on channel activation, whereas the M241A equivalent caused a large negative (~-50 mV) shift in the voltage dependence of activation ⁴¹. It is notable that the Met241 position shows the same degree of conservation as the Glu (0) position in the SF that is the key determinant of ion selectivity. It has been proposed based on state-dependent accessibility studies of residues in eukaryotic Na_V1.4 DIV S6 that the gate is one turn higher (equivalent to Phe233) ¹²⁰. Whether this apparent difference between Na_V1.4 and the BacNa_Vs reflects the dissimilar approaches used to define the gate, structure based versus chemical reactivity of a mutant channel, or indicates genuine differences between homotetrameric and heterotetrameric pores remains an open question.

The PD conservation analysis highlights two other positions of note. One is Asn231, this S6 position is conserved throughout the BacNa_V, Na_V, and Ca_V families (Fig. 5e) but absent from potassium channels ¹²¹ and points towards the S4/S5 linker in some of the structures ³⁸. Mutation of this residue to alanine in IS6 of the eukaryotic Na_V1.4 shifts the voltage dependence of activation to more depolarized potentials and enhances entry in to the slow inactivated state ¹²². Effects of mutations in various S6 segments of Na_V1.2 further support the importance of this position in slow inactivation and impact channel modulation by kinases ¹²³. The other conserved position is S5 Na_VAe1p Tyr162, which makes contacts to the S1 segment of the VSD in Na_VAb and Na_VRh (as noted below). Given the high information content of both of these positions, investigation of their functional roles merits further experimental attention.

Voltage sensor revisited

VSDs are central to the functioning of our nervous system, form the cornerstone of Hodgkin and Huxley's pioneering physiological studies on VGICs ³, and are one of the most thoroughly studied protein domains in biophysical research ^{1; 2; 87; 124; 125; 126; 127}. Since entering the age of ion channel structural biology, several studies have highlighted the VSDs as modular folding units ^{28; 128; 129; 130}, demonstrated the constraints required for electromechanical coupling to the pore by transferring a VSD onto a non-voltage gated channel ^{131; 132; 133}, and shown the functional transferability of the S3-S4 regions between evolutionary distant VSDs ^{134; 135; 136}.

Despite all the fame and attention, the VSD is a four-helix bundle of relatively simple construction (Fig. 6a, b). The S4 segment carries highly conserved positively charged “gating charges” interspaced between two hydrophobic amino acids in a repeating RxxR motif. In response to changes in membrane potential, the gating charges interact with negative charge clusters on the S1-S3 segments to catalyze S4 movement through the electric field ¹²⁵. These extracellular and intracellular negative charge clusters (ENC and INC) are solvent accessible ^{127; 137; 138}, but separated by a hydrophobic constriction site (HCS) that prevents ion leakage through a hydrophobic bottleneck or “gating pore”. A conserved phenylalanine residue in the hydrophobic constriction site and the intracellular negative charge cluster has been implicated as a “gating charge transfer center (CTC)” in K_V¹³⁹ channels . In Na_V1.4, these same residues only appear to be essential for proper VSD folding and membrane trafficking of the channel ¹⁴⁰. Nevertheless, all VSDs form an hourglass-shaped structure in the membrane that essentially functions as a voltage-dependent arginine side-chain transporter.

Figure 6. BacNa_V VSD structural comparison.

(a) VSDs from representative crystal structures in the PDB were freely superimposed onto the Na_VAb VSD (PDB 3RVY) ³⁷. These include VSDs from Na_VRh (4DXW) ³⁹, Na_VCt (4BGN) ⁵⁸, K_V1.2/2.1 chimera (2R9R) ⁷¹, K_VAP (1ORS) ²⁸, VSP (4G7V and 4G80) ¹⁴², and H_V1 (3WKV) ¹⁴¹. Positions of the S1N, S1, S2, S3, and S4 helices and intervening S1-S2, S2-S3, and S3-S4 loops are indicated. Gating charge residues on the S4 segments are shown as purple sticks. Extracellular and intracellular negative charge cluster (ENC and INC) residues are shown as red sticks. Hydrophobic constriction site (HCS) residues are shown as green sticks. In general, structural correspondence between all VSDs is seen, with the largest deviations apparent in the loop (e.g. S1-S2 and S3-S4) and S1N regions. A large sequence insertion in the Kv1.2/2.1 chimera S1-S2 loop has been omitted for clarity. View is from the plane of the membrane bilayer. (b) Taken from the alignment in (a), details of the similarities and differences between the Na_VAb and Na_VRh VSDs can be appreciated. Gating charges, INC and HCS, a conserved Phe sidechain, resides are highlighted. (c) Extracellular view looking onto the Na_VRh VSD and neighboring pore domain. The S5 residue Tyr145 is highly conserved in BacNa_V channels and points directly into the VSD extracellular crevice. S4 gating charge residues are shown as purple sticks. (d) Same view as in (c), looking onto the Na_VAb VSDs and neighboring pore domain. In contrast to Na_VRh, the highly conserved S5 Tyr142 residue points directly at the S1 helix, and not into the VSD extracellular crevice, highlighting the rotation of the Na_VAb VSD around the pore domain is with respect to the VSD position in Na_VRh. (e) Sequence alignment indicates a simple single residue insertion in the S1-S2 linker of Na_VAb (L31) relative to Na_VRh. Structure-based alignment however clearly demonstrates “equivalent” residues to be spatially displaces from one another, by up to ~5 Å. This structural plasticity of the S1-S2 loop in Na_VRh is, in part, responsible for the intracellular movement of its S2 ENC residue Asp48. (f) Superposition of the four VSDs from the Na_VRh structure (4DXW) demonstrates the structural variations seen within the S3-S4 loop and the gating charges of a single VGIC captured in the same crystal. Displacements of equivalent residues up to ~8 Å are seen. (g) A highly conserved non-canonical VSD interaction is observed in Na_VAb. Trp76 found at the foot of the S3 is one of the most highly conserved sidechains in all Na_V and Ca_V VSDs. This residue may help anchor the VSD into the membrane. It also makes a π-cation interaction with a highly conserved Arg residue from in the S4/S5 linker, R117.

The BacNa_V structures provide an important architectural framework and rich diversity of conformational snap-shots to help understand VGICs. When all of the VSD coordinates are extracted from the BacNav structures and compared to the VSDs from Kv channels ^{28; 71; 86}, the proton gated channel (H_V 1) ¹⁴¹, and the voltage-sensitive phosphatase (VSP) ¹⁴², a number of general conclusions can be made. First, at a gross level, all VSD structures are essentially identical and share a conserved structural core independent of their amino acid sequence or functional state (Fig. 6a). Second, the S4-S5 linker has a fixed length amongst diverse, but not all, VGICs, which imposes key constraints on the electromechanical coupling mechanism between the PD and VSD. Third, only four gating charges superimpose within the conserved structural core (Fig. 6a), apparently contradicting the notion that vertebrate VSDs may contain 5-7 functional gating charges per VSD ^{124; 143}. Perhaps the “extra” gating charge residues found on the S4-S5 linker (Fig. 7d) may modulate channel gating through interactions with surrounding lipid headgroups or are important for protein folding (also see conservation analysis below). Fourth, the S4 segment is in an entirely α-helical conformation in VSDs that have been cut away from the rest of their native sequence (i.e. KvAP ²⁸, Hv1 ¹⁴¹, and Ci-VSP ¹⁴²). By contrast, VSDs that are still attached to their natural payloads all display some mixture of α-helix and 3₁₀-helix along their S4 segments. Finally, the structural correspondence between Na_VAb and murine K_V1.2 VSDs is absolutely striking and argues that the folding-pathway and voltage-sensing mechanism have been highly conserved over the course of molecular evolution ³⁷.

Figure 7. BacNa_V VSD sequence conservation.

(a) Conservation analysis of the BacNa_V VSD measured by the relative statistical entropy at each position, $D_i^{\left(a\right)}$ ¹¹⁵, mapped on the Na_VAb sequence. Highly conserved positions are colored dark blue, light blue, and green. (b) and (c) Conservation analysis depicted on the Na_VAb VSD. Select positions of interest are shown as sticks and are labeled. (d) Comparison of S4/S5 linker and (e) S3 segments for the indicated sequences: Na_VBh1(NaChBac)(NP_343367.1) S3 87-113, S4/S5 linker 128-142; Na_VAb (YP_001490668.1) S3 74-90, S4/S5 114-128; Na_VRh (PDB:4DXW), S3 75-91, S4/S5117-133; Na_VAe1 (YP_741167.1), S3 94-110, S4/S5 134-158; Na_VCt (WP_007502948.1) S3 90-105, S4/S5 131-145; Na_VMs (YP_864725.1), S3 75-91; S4/S5 linker 115-129; Na_VSp1 (YP_165303.1), S3 73-89, S4/S5 112-126; Ca_V1.2 (CAA84346), IS3 193-209, IS3 586-602, IIIS3 967-983, IVS3 1281-1297, IS4/S5 252-66, IIS4/S5 265-649, IIIS4/S5 1013-1027, IVS4/S5 1344-1358; Ca_V2.1 (NG_011569.1), IS3 167-183, IIS3 549-565, IIIS3 1311-1327, IVS3 1628-1644, IS4/S5 210-224, IIS4/S5 598-612, IIIS4/S5 1361-1375, IVS4/S5 1676-1690; Ca_V3.1 (O43497), IS3, 150-166, IIS3 805-821, IIIS3 1344-1360, IVS3 1673-1689, IS4/S5 195-209, IIS4/S5 850-864, IIIS4/S5 1396-1410, IVS4/S5 1727-1741; Na_V1.4 (NP_000325.4), IS3 191-207, IIS3 640-656, IIIS3 1095-1111, IVS3 1414-1430, IS4/S5 234-248, IIS4/S5 684-698, IIIS4/S5 1144-1158, IVS4/S5 1466-1480; Na_V1.7 (NM_002977), IS3 186-202, IIS3 795-811, IIIS3 1245-1261, IVS3 1565-1581, IS4/S5 229-243, IIS4/S5 839-853,IIIS4/S5 1294-1308, IVS4/S5 1617-1631.

BacNa_V VSDs

BacNa_Vs are small proteins and their “minimal” VSDs can serve to highlight essential features and pinpoint vertebrate-specific functional elaborations. Structure-based alignments indicate that large sequence insertions occur within the extracellular loop regions of vertebrate VSDs ³⁷ but also demonstrate that the intracellular S2-S3 loop can adopt a common conformation. The BacNa_V structures call attention to a conserved amphipathic helix that precedes the S1 segments (the S1N helix) and that may be dynamic during the gating process ^{128; 130}. Overall, the BacNa_V channels provide important new constraints on this four-helical bundle domain and focus attention on some of the novel details revealed by their structures.

In addition to interactions with negative charge clusters, the Na_VAb ³⁷ and Na_VRh ³⁹ S4 gating charges show coordination to backbone carbonyl groups and other VSD sidechains. These non-traditional gating charge interactions have since been observed in modeling studies, prompting speculation that they contribute to the S4 activation pathway ⁶⁷. Amino acid substitutions at these previously considered benign positions might therefore affect VSD function and could predispose certain individuals to pathophysiological states (e.g. migraine or arrhythmias) ^{124; 143}. The BacNa_V structures also bring the PDs that buttress the VSDs directly into view (Fig. 2b). PD sidechains such as the conserved Tyr145 in Na_VRh, corresponding to Na_VAe1p Tyr162 noted above (Fig. 5), help to directly shape the S4 activation pathway by inserting its sidechain into the extracellular crevice of the VSD (Fig. 6c). In Na_VAb, however, the equivalent S5 residue, Tyr142, lies outside of this extracellular VSD crevice and instead directly engages the S1 helix (Fig. 6d). These structural observations together indicate how otherwise “distant” PD residues might modulate the function or physiological tuning of VSDs in a VGIC.

On closer inspection, the voltage-sensing S4 segments reveal subtle structural differences between Na_VAb and Na_VRh ^{37; 39}. In Na_VAb, the S4 is found almost exclusively in a 3₁₀-helical conformation, placing gating charges from the RxxR motif on the same side of the helix for passage through the gating pore, perhaps through simple translation or tilting. By contrast, the middle portion of the Na_VRh S4 is a 3₁₀-helix and its N-terminal portion is modeled as an α-helix. This raises the possibility that the S4 might undergo a secondary structure transition during charge transfer through the gating pore. Potentially supporting this view, MD simulations suggest that the Na_VAb VSDs have not fully activated ¹⁴⁴. However, this later notion is inconsistent with the physiological characterization of Na_VAb ^{61; 145} and ignores the possibility that all available VGIC crystal structures reasonably represent a fully activated or inactivated VSD conformation, considering that a 0 mV condition is maintained over the course of all crystallization experiments. If we assume that the basic details of the voltage sensing mechanism are universally conserved across all VSDs, one parsimonious conclusion is that the S4s from different VSDs are strained (3₁₀-helix) or relaxed (α-helix) to varying degrees and that this extent depends primarily upon the amino acid sequence of the crystallized protein construct in question. Future experiments are needed to rigorously interrogate these subtle details of VSD structure and function.

A few VSD surprises

Uniquely, the Na_VRh structure presents four simultaneous views of its VSD because one channel tetramer crystallized within the asymmetric unit ^{39; 146}, providing a fortuitous example of the potential structural dynamics that are possible within a VSD. Structural plasticity is seen within the S3-S4 linkers of two subunits (Fig. 6f), and to a lesser extent, in the S1-S2 linkers of all four subunits. These differences highlight the same two focal points of sequence and structural divergence across all VSDs, the S1-S2 and S3-S4 loops (Fig. 6a). The observed structural changes also impact the chemistries available for S4 gating charge interactions (Fig. 6b-f), and the malleability of the Na_VRh VSDs suggests the intriguing possibility that inherent structural heterogeneity exists along the S4 activation pathway.

All four gating charges of Na_VRh are exposed to the extracellular side of the hydrophobic constriction site, with R4 above the so-called CTC, defined here to include the INC and HCS elements. The R4 of Na_VAb remains engaged with the CTC, suggesting that Na_VAb may transfer one less gating charge during activation, although equivalent gating charges are nonetheless found at similar depths within the membrane (Fig. 6b). Na_VRh achieves R4 transfer above the CTC through the concerted intracellular movements of its S1-S3 segments ¹⁴⁶, where its CTC appears “downshifted” within the membrane compared to Na_VAb (Fig. 6b). If relevant, these unprecedented observations hint at a new concept in voltage sensor activation and suggest that there may be more then one moving part in VSDs.

Further comparison between Na_VRh and Na_VAb highlights an unrecognized observation in the S1-S2 linker upon simple sequence alignment and consideration of structurally “equivalent” residues (Fig. 6e). From a structure-based perspective, a portion of the linker sequence appears to toggle between the S1 and S2 segments akin to a slinky toy. The effect in Na_VRh is that the S2 translates towards the intracellular side and results in the displacement of the S2 ENC residue by one helical turn (i.e. Asp48 vs. Asn49; Fig. 6e). This implies that a remarkable degree of freedom exists at the S1-S2 junction, which is perhaps consistent the large sequence insertions that are found in Kv channel linkers ^{71; 86; 147} and the plasticity required for CTC “downshifting” in Na_VRh. Subtle structural transitions in the S1-S2 linker may therefore be more commonplace in VSD function then is currently appreciated.

Key BacNa_V VSD positions revealed by structure-based sequence comparisons

Similar to the PD analysis described above, we used the large collection of available BacNa_V sequences for positional conservation analysis ¹¹⁵ of the VSD residues and mapped these results on the Na_VAb VSD (Fig. 7). This analysis highlights well-known conservation patterns, such as the RxxR motif of the S4 segment (Arg99, Arg102, Arg105, and Arg108), two negatively charged residues (S2 Glu59 and S3 Asp80) that make the intracellular negative charge cluster (INC) ³⁷, and the three components of the hydrophobic constriction site (HCS) (S1 Ile22, S2 Phe56, S3 Val84) ³⁷ (Fig. 7a), and shows features common to an analysis of VSDs that included multiple types of bacterial and eukaryotic cation channels¹⁴⁸. Notably, two previously unanalyzed residues in the S4/S5 linker stood out. One is the proline (Pro114) that makes the bend between S4 and the S4/S5 linker (Fig. 7b-d). The second is the conserved Arg or Lys (Arg117) that appears one turn into the S4/S5 linker helix. Arg117 is within distance to form a cation-π interaction with the conserved Trp at the base of S3 (Trp76). Strikingly, this Trp is the most conserved feature of the BacNa_V VSD (Fig. 7a) and is strongly conserved as an aromatic residue throughout the eukaryotic Na_Vs and Ca_Vs (Fig. 7e). Mutation to cysteine at the conserved Arg117 position in NaChBac shifts the voltage-dependence of activation in the depolarizing direction (~+30 mV) ⁶⁵, consistent with the idea that the Arg117-Trp76 interaction may stabilize the active state of the VSD and the interpretation that the VSDs of the BacNa_V structures represent an activated state. The positive nature of the equivalent S4/S5 linker position is well conserved in many of the Na_V and Ca_V domains (Fig. 7d). Given the striking conservation of the Trp76 position, and ideas about the down-state of the S4 segment under hyperpolarizing conditions, one wonders if there are interactions with the conserved gating charges of S4 and Trp76 in other VSD states that help to shepherd gating charges into or away from the intracellular negative charge cluster.

On the evolution of VSDs

The discovery of isolated (or pore-less) VSD containing proteins such as CI-VSP ¹⁴⁹ and Hv1^{150; 151} raises questions about the evolution of this domain. Did an ion channel sprout a rudimentary S4 segment that subsequently elaborated into today's VSD, and did this VSD later become detached? Or did VSDs evolve independently and then fuse to the PD of an ion channel? The concept that VSDs are dynamic four-helical bundles, taken with the structural diversity seen in their loops (Fig. 6a), begins to suggest that a covalent linkage may not be an absolute requirement for function. In fact, deleting large portions of the S3-S4 linker still produces a functional Kv channel ¹⁵² and genetically cutting the channel into two separate polypeptides through the S3-S4 linker also results in functional expression ¹⁵³. In preliminary experiments, the co-expression of an S1-only construct along with the S2-S6 region of NaChBac was found to produce a functional VGIC, as did S1-S2 co-expression of with a S3-S6 construct (J.Payandeh, unpublished). Therefore, VSDs do not absolutely require a covalent linkage between the transmembrane helices to function. These observations could provide insight into the enigmatic evolution of this essential domain. Although we can clearly cut and paste a VSD onto a non-voltage gated pore to render it voltage-sensitive ^{131; 132; 133}, perhaps recapitulating an important step in of early evolution within the VGIC superfamily, it remains to be determined how extensively one might be able to cut up a VGIC into pieces and still retain function, where the functional cut sites might be, and how many different cuts can be tolerated at once. Such, ‘deconstruction’ experiments should offer interesting new insights, particularly as the available structural data now offers a clear blueprint for candidate sites.

Insights into gating and inactivation

BacNa_Vs, just like other voltage-gated channels, respond to membrane potential changes that cause the channel to inhabit non-conductive (closed and inactivated) and conductive (open) states. Thus far, electrophysiological studies have characterized BacNa_V homologs from twelve different organisms from the soil, sea, and salt lakes, revealing a broad range of activation and inactivation gating properties ^{48; 154}: NaChBac (Bacillus halodurans) ^{29; 45; 46; 63; 155; 156; 157}, Na_VAb (Arcobacter butzleri) ^{37; 38; 145}, Na_VAe1 (Alkalilimnicola erlichii) ⁴¹, Na_VBacL (Bacillus licheniformis) ³², Na_VBP (Bacillus pseudofirmus)³¹, Na_VCt (Caldalkalibacillus thermarum) ⁵⁸, Na_VMs (Magentococcus marinus) ^{62; 98}, Na_VPz (Paracoccus zeaxanthinifaciens) ³⁰, Na_VRosD (Roseobacter denitrificans) ³², Na_VSheP (Shewanella putrefaciens) ³², Na_VSp1 (Silicibacter pomeroyi) ^{30; 35; 41}, Na_VSulP (Sulfitobacter pontiacus) ¹⁵⁸. This rich array of functional diversity and systems for investigation suggests that BacNa_V structural studies have the potential to uncover the main conformations that underlie important conductive and non-conductive states. However, to do so, as with all structural studies, it is essential to connect the structures with the appropriate functional state ⁵⁷. Voltage control over the samples is not possible in the context of a crystallization experiment, unlike conformational transitions in membrane proteins that are driven by chemical transformations such as ATP consumption where specific substrates can trap states ¹⁵⁹. Hence, it is not as straightforward as it might first seem to know what exact functional states the various BacNa_V structures represent absent some explicit tests based on interactions seen in the structures.

For the full-length structures ^{37; 38; 39; 58}, there is agreement that the VSDs are in the activated position due to the long time spent at zero potential once the proteins were purified out of a charged cell membrane; however, despite this general concordance about the VSD state, the structures show conformational variations in both the VSDs and PDs. One common feature of the Na_VAb and Na_VRh structures is that intracellular access out of the aqueous central cavity below the selectivity filter is blocked by a constriction formed by the pore-lining S6 helices. The first Na_VAb structure has been suggested to be ‘pre-open’ ³⁷, whereas the Na_VRh was deemed inactivated ³⁹, particularly as the Na_VRh selectivity filter is in a conformation that blocks ion transit. The asymmetric pore arrangements observed in the second set of Na_VAb structures ³⁸ have been interpreted as inactivated conformations as they match expectation that the so-called ‘slow inactivated’ state should have some conformational change in the pore vestibule ⁴⁵. It is worth noting that the asymmetric filters of these Na_VAb structures are still wide enough for a partial hydrated ion to pass. The two low-resolution conformations described for Na_VCt using electron crystallography show pores that are similar and closed, although the small differences between the conformations have been suggested to represent an open and closed inner gate ⁵⁸. An alternative to these interpretations is that all of the full-length structures are some sort of post-open state, which may not resemble a truly inactivated conformation but may be much closer to how the PD looks when the channels are closed or deactivated.

One other striking difference regarding the VSDs is that, in addition to the differences in S4 interactions described above, the Na_VRh VSDs have essentially a rigid body displacement around the PD of ~30° relative to Na_VAb (Fig. 6 c-d) ^{39; 97}. This difference is reminiscent of the observed VSD disengagement from the PD upon channel closure in simulations of Kv1.2/2.1 chimera in lipid membranes ¹⁶⁰ and would seem to be in line with the independent character of the VSDs with respect to the overall protein structure ^{28; 129; 131; 142; 161}. Such displacements raise interesting questions about the exact functional states of the full length BacNa_V structures, how there can be strong coupling between the internal motions of S4 within the VSD in response to voltage, and whether such global VSD motions have a role in voltage sensing ¹⁴⁶. Furthermore, upon comparing the available Na_VAb I217C and WT structures, significant movements of the VSDs around the PD are also seen that may highlight a novel isoform selective receptor site that is potentially druggable in the eukaryotic Na_V channels.

The conformation of Na_VAe1p, which has been freed from its VSDs, has all the hallmarks of a closed conformation ⁴¹, and moreover, is concordant with the effects of S6 alanine mutations on the energetics of channel opening that were designed to test interactions seen by the structure ⁴¹. The PD subunits in first structure of the Na_VMs ‘pore-only’ channel had four non-identical conformations, the most variant of which was used to generate a possible open conformation model ⁵⁹. This model suggests a twisting displacement around the central axis, a motion akin to the iris-like opening of a camera aperture. A similar conformation has since been reported in other Na_VMs structures ^{60; 62}. It is notable that in simulations, in the absence of applied restraints, the putative open conformation does not remain open ⁹⁸, casting doubt upon the functional state represented by the structures.

If the observations about the similarities of the PD structures are put aside and the conformational assignments of the possible states are taken at face value, one might get the impression that the energetic landscape that separates one PD conformation from another is quite flat. Although such a landscape would explain the small conformational differences among the PDs, this interpretation is inconsistent with the large amount of energy that goes into activating a voltage-gated sodium or potassium channel (~14 kcal mol⁻¹) ^{162; 163}. The functional properties of ‘pore-only’ BacNa_V versions ^{35; 36} demonstrate that the PDs can open stochastically without any energetic input from the VSDs. Further, even though the VSDs in all of the full-length BacNa_V structures are activated, none have a PD as open as that as found in the K_V1.2/2.1 chimera. So, why is it that when the VSDs in the active position one class of channel PDs always looks non-conductive while the other is wide open? Will we ever see a channel structure in which the VSDs are in the ‘down’ state? Additional structures determined using pharmacology that could definitively trap a well-characterized state or of mutants that are known to bias the channel into the open, closed, or inactivated conformations would go a long way to address this puzzle. In this regard it is interesting that PD mutants have been reported that dramatically shift the voltage dependence of opening to more hyperpolarized potentials ¹⁶⁴, reverse the voltage-dependence of gating ¹⁶⁵, and that selectivity affect channel inactivation, but not activation ⁴¹. Investigation of such mutants may help to define conformations of open and inactivated states or of other intermediates in the functional cycle.

The Na_VAe1p structure revealed that BacNa_V CTD has a continuous helix comprising a ‘neck’ and coiled-coil domains ⁴¹. Poly-glycine mutations in the neck shift voltage-dependent opening of Na_VAe1 and Na_VSp1 to more hyperpolarized potentials consistent with the proposal that this domain undergoes an order to disorder transition upon channel opening ⁴¹. In contrast to the order seen in the Na_VAe1p neck, studies of NaChBac ⁷² and Na_VMs have suggested that this region can be largely disordered ⁶⁰. Deletions in the CTDs of different BacNa_V isoforms yield loss of function ^{34; 58; 60}, negative shifts ¹⁵⁸ or small changes ⁶⁰ in the voltage-dependence of activation, and varied effects on channel inactivation rates ^{60; 158}. These results may not be as discordant as they first seem. Given that the BacNa_V neck has varied composition (Fig. 8), the degree of structure and its impact on channel gating may differ among the BacNa_V isoforms. Although exactly how the BacNav neck affects the channel function remains to be understood, one intriguing possibility is that the ‘down’ state of the S4/S5 linker may involve interactions with the neck region. The neck structure has no counterpart in eukaryotic Na_Vs or Ca_Vs; however, many VGIC superfamily members that are assembled from multiple pore-forming subunits have a C-terminal coiled-coil in the cytoplasmic domain that follows S6 similar to that seen in the BacNa_Vs ^{73; 74; 76; 141; 166; 167; 168}. Moreover, there is evidence that conformational changes in the regions between S6 and the coiled coil have profound effects on gating, such as in Kv7 channels ^{75; 169}. Hence, elucidation of the mechanisms by which the BacNa_V CTD influences voltage-dependent channel function might provide generalizable lessons that will impact our understanding of other VGIC superfamily members.

Figure 8. BacNa_V PD and CTD sequence comparison.

S5, selectivity filter (SF), S6, Neck, and coiled-coil regions are indicated. 3-4 hydrophobic repeats of the coiled coils are indicated in grey. Conserved positions in S5, SF, and S6 are highlighted orange. Residues lining the pore are highlighted in blue. SF filter positions are indicated. Blue sequences have structures determined. Purple sequences have reported functional studies.

Pharmacology and fenestrations

Ion channels represent major drug targets in the human body ¹⁷⁰. The rich array of natural and synthetic molecules that comprise the pharmacology of Na_Vs ^{124; 171; 172; 173; 174; 175; 176} is one their most interesting and distinguishing characteristics. In line with their complex assortment of functional states, molecules that affect Na_V function have complex modes of action including pore block, open state stabilization of the pore, and voltage sensor movement alteration (Fig. 9). Despite this impressive pharmacological artillery, traditional drug discovery efforts have historically failed to find highly selective Na_V inhibitors, owing in part to the high sequence identity found amongst the human channel isoforms (Fig. 4c). Despite their inherent limitations, the BacNa_V structures now provide our most accurate templates to understand the physical interaction of Na_V channel modulators and explore the potential for the rational design of new therapeutics.

Figure 9. Eukaryotic Na_V Pharmacology painted onto BacNa_Vs.

Na_VAb subunits are colored to represent the four non-homologous domains of a eukaryotic Na_V channels. One pore domain and one voltage-sensor are removed for clarity. Representative structures of channel modulating toxins, small molecules, and cations are depicted to interact with their best-characterized receptor sites on the channel. All structures are rendered to approximately the same scale.

Targeting the extracellular vestibule

The extracellular vestibule is the site of Na_V modulation by protons, divalent cations, and small toxins (Fig. 9). Proton block of vertebrae Na_Vs has been well characterized and occurs predominantly through the protonation of residues lining the extracellular vestibule above the SF ^{177; 178; 179}. The BacNa_V structures suggest that protons might neutralize key sidechains required to coordinate and conduct Na⁺-water complexes through the SF or destabilize interactions required to maintain the SF in a conductive conformation. Tetrodotoxin (TTX) is one of the most famous ion channel blockers and binds directly to the SF region of sensitive Na_V channels with low nanomolar affinity to cause its neurotoxic effects (Fig. 9). Models of TTX binding have existed for over four decades and generally envision the TTX guanidinium group interacting with the glutamate side-chain from the DEKA motif ^{180; 181; 182; 183; 184; 185}. Because vertebrate Na_Vs contain sequence register shifts around the SF region (Fig. 1c), there is uncertainty about how well the bacterial structures recapitulate the eukaryotic Na_Vs at this site ^{183; 184}. Not surprisingly, BacNa_Vs have proven to be TTX insensitive ²⁹ (F. Abderemane-Ali., M. Krier, and D. L. Minor, Jr., unpublished). Perhaps a protein engineering approach to create a concatenated BacNa_V channel with a heterotetrameric vertebrate-like extracellular vestibule will offer the first tractable structural biology approach to visualize this important pharmacological site.

Access and block in the central cavity

The BacNa_V channel structures provide excellent templates to understand one of the most studied receptor sites in ion channel research, the central cavity of Na_Vs and Ca_Vs. Traditional Na_V channel blockers are used clinically as antiepileptic, antiarrhythmic and analgesic drugs that block in the central cavity through at least two different binding modes. Neutral local anesthetic drugs can access the central cavity in resting Na_V channels and produce “tonic” block in the micromolar to millimolar range ^{1; 186}. Antiepileptic and antiarrhythmic drugs block opened or inactivated Na_V channels in a “use-dependent” way in which affinity is positively correlated with channel activation ^{1; 187}. Although the binding sites for these drugs overlap, different residues contribute to tonic and use-dependent block, possibly reflecting changes in the central cavity associated with channel opening and inactivation ^{188; 189}. A significant structural reshaping of the central Na_VAb cavity has been suggested to reflect the characteristics of a slow inactivated state in eukaryotic Na_Vs (Fig. 4B) and may be relevant to the understanding of these channels ³⁷.

Two highly conserved aromatic residues on the S6 helix of DIV in vertebrate Na_Vs are major determinants of use-dependent block ^{186; 189} and equivalent residues in BacNavs line the central cavity (Fig. 4B, C). Mutagenesis studies in Ca_Vs have further implicated an S6 side-chain on DIII ¹⁹⁰, suggesting a composite drug-binding site is formed across the central cavity ^{191; 192}. We used the BacNav structures as templates to model the central cavity of vertebrate Na_V channels, as illustrated here for human Nav1.7 (Fig. 4C). Homology modeling and conservation analysis immediately rationalizes why this traditional receptor site lacks significant isoform selectivity (Fig. 4C). Remarkably, an adjacent interface between DI and DII represents an intriguing and potentially highly isoform selective site (Fig. 4C). Because the Na_VMs channel has already undergone crystallographic analysis to produce models of drug binding within the central cavity ⁶², the available BacNa_V structures can clearly provide meaningful starting points for structure-based drug discovery efforts and opportunities to design novel, isoform selective small molecule inhibitors for the treatment of Na_V channel related pathologies. One intriguing result from recent pharmacological and MD studies of NaChBac is the conclusion that drug binding may occur at multiple sites, in multiple binding modes¹¹³. There is little doubt that further higher-resolution studies of such compounds bound to BacNa_Vs in different conformations will be essential for defining how the PD can interact with small molecules that perturb function and for generating new ideas about modulator design.

Interfering with the VSD

Diverse organisms in nature target the VSDs of vertebrate Na_Vs by deploying small protein toxins ¹⁹³. These “gating modifying” neurotoxins are classically described to bind to the VSDs of DII or DIV (Fig. 9) and confound the activation or inactivation properties of the channel through different voltage sensor trapping mechanisms, either by enhancing or suppressing activation of the S4 segment ^{135; 171; 176}. Importantly, some gating modifying toxins are known to display significant Na_V channel isoform selectivity ¹⁹⁴ and have major binding determinants found in the extracellular S1-S2 and S3-S4 loops. Feasible structural models of the CssIV toxin-VSD2 interaction and the Lhq2 toxin-VSD4 complex have already appeared using a composite VSD homology model drawn from the S1-S2 and S3-S4 regions of the BacNa_V and K_V channels, respectively ^{195; 196}. It is hoped that these protein toxins may represent suitable scaffolds for pharmaceutical development and that their receptor sites can be exploited for isoform-selective drug discovery.

A recent landmark study has described an exciting new series of small molecules that appear to bind selectively to VSD4 of Na_V1.7 or Na_V1.3 and antagonize these channels ¹⁹⁷. These drug-like molecules seem to stabilize an inactivated state of the VSD. In another breakthrough report, a monoclonal antibody directed against the S3-S4 loop of VSD2 has been shown to selectively stabilize the closed form of Nav1.7 and have profound efficacy in pain studies in vivo ¹⁹⁸. Together, these incredible findings not only validate the pharmacological inhibition of Na_V1.7, but they promise that we are on the verge of truly isoform selective Na_V channel inhibitors showing utility in the clinic.

Prospects

The tractability of BacNa_Vs for structural and functional studies sets the stage for rapid advances in outlining the structural basis for the conformational transitions that underlie VGIC function. The diversity of forms available for investigation will undoubtedly be a major advantage and are likely to reveal the source of functional differences within the family. If we were making a wish list for the next five years, one would hope to see structures at high resolution of ions in the selectivity filter and small molecules bound to the PD or VSD, structures of mutants that trap well-defined functional states, and possibly structures of engineered BacNa_Vs that bear key features of their eukaryotic counterparts, such as an asymmetric selectivity filter. Because there are lipids intimately bound with parts of the channel and some specificity in how lipids modulate BacNa_V function ¹¹⁸, it seems important to further characterize these sites and understand their relationships with the way hydrophobic drugs interact with and affect channel function. Given that native eukaryotic Na_Vs have been the subject of electron microscopy studies for many years¹⁵ and that efforts on expression of complex eukaryotic membrane proteins bear ever more successes, it seems likely that the recent breakthroughs enabling single particle analysis of sub-megadalton complexes at a resolution rivaling X-ray crystallographic structures determined at ~3.0 Å resolution ¹⁹⁹ are poised to give the first view of a eukaryotic Na_V. If one compares the state of the field with that of kinases, it is obvious that even a few structures are only the beginning. As more and more members of the VGIC superfamily are revealed in three-dimensions, the future for connecting the details of channel biophysics with meaningful biological outcomes seems very, very bright indeed.

Highlights.

• Bacterial sodium channels (BacNa_Vs) offer unprecedented structural insights into voltage gated ion channel function

• BacNa_V structures reveal the details of sodium and calcium-selective pores

• BacNa_V structures offer plausible models to understand the action of drugs