Chemical and Biological Tools for the Study of Voltage-Gated Sodium Channels in Electrogenesis and Nociception

Abstract

The malfunction and misregulation of voltage-gated sodium channels (Na_Vs) underlie in large part the electrical hyper-excitability characteristic of chronic inflammatory and neuropathic pain. Na_Vs are responsible for the initiation and propagation of electrical impulses (action potentials) in cells. Tissue and nerve injury alter the expression and localization of multiple Na_V isoforms, including Na_V1.1, 1.3, and 1.6–1.9, resulting in aberrant action potential firing patterns. To better understand the role of Na_V regulation, localization, and trafficking in electrogenesis and pain pathogenesis, a number of chemical and biological reagents for interrogating Na_V function have been advanced. The development and application of such tools for understanding Na_V physiology are the focus of this review.

1. Voltage-Gated Sodium Channels

1.1. Channel structure

Voltage-gated sodium channels comprise a 240–260 kDa sodium-conducting α-subunit (Figure 1) and one or two auxiliary β-subunits. The α-subunit includes four homologous repeats (DI-DIV), each containing six membrane-spanning α-helices (S1-S6). Helices S5 and S6 compose the so-called pore domain (PD), whereas S1–S4 form the voltage-sensing domains (VSD).^[1] At the base of the extracellular mouth, four residues—aspartate, glutamate, lysine, alanine (DEKA)—form a filter that selects for hydrated sodium ions.^[2] High resolution structures of α-subunits for different mammalian Na_V subtypes, including hNa_V1.2,^[3] 1.4,^[4] 1.5,^[5] and 1.7,^[6] as well as αβ-heteromeric complexes^[3,4,6,7] are now available owing to recent advances in cryogenic electron microscopy (cryo-EM).

Figure 1.

Human voltage-gated sodium channel, hNa_V1.7 (PDB 6J8J).^[6] A) Side view. IFM fast inactivation particle shown in yellow (space filling model).^[4] B) Top view. Domain I (orange), Domain II (red), Domain III (gray), Domain IV (cyan).

The S4 helix of each voltage sensor includes positively charged arginine residues that are sensitive to the membrane potential. As the cell depolarizes, the S4 helices move outward, triggering conformational changes that prompt the opening of the channel pore.^[8] Movement of the DIV S4 helix is the last to occur,^[9] resulting in a synchronous conformational change in the intracellular loop between DIII and DIV. Within this loop, the position of the isoleucine-phenylalanine-methionine (IFM) particle is altered to inhibit Na⁺ conduction—a process termed fast inactivation, which occurs within milliseconds of channel opening.^[10] Slow inactivation, which ensues on the timescale of tens of milliseconds to seconds, arises from additional movement of the VS domains as a result of prolonged and/or repetitive stimulation.^[11] Subsequent membrane repolarization resets the conformation of the voltage-sensing domains, returning the channel to the closed, resting state. Auxiliary proteins, so-called β subunits (β1A, β1B, β2, β3, and β4), are positioned on the extracellular face of the channel and modulate the movement of the voltage sensors, thereby altering the voltage dependence and gating properties of the α-subunit.^[12] The totality of the conformational changes of this heteromeric protein complex in response to membrane voltage regulates the Na⁺ component (i.e., rising phase) of the action potential (AP) (Figure 2).

Figure 2.

A) Na_V component of the action potential. (1) Closed (2) Open (3) Inactivated. B) Limiting conformational states of Na_V.

Voltage dependence and gating properties differ among Na_V isoforms (see Section 1.2) (Table 1).^[8] Subtle variations in channel expression levels and cellular distributions and in the intrinsic gating properties of each subtype give rise to the diversity of AP signal outputs. Specifically, Na_V isoforms display disparate voltage dependences of activation and inactivation, time constants of inactivation, and rate constants for transitioning between conformational states. The initial transition from closed to open channels results in Na⁺ influx, termed transient current, which ceases within milliseconds as channels inactivate. In some cases, incomplete inactivation can produce a longlasting, so-called persistent Na⁺ current.^[13] Membrane repolarization generally returns channels to the reprimed, closed state. Some channels, however, enter a unique nonconducting state after channel opening that, unlike the canonical inactivated state, transitions to the open rather than closed state upon repolarization, producing an aptly named resurgent current.^[14] The composite of these responses contributes to action potential waveform, amplitude, and frequency.^[15]

Table 1.

Characteristic properties of Na_V α-subunit isoforms.

Isoform	V1/2 activation/inactivation [mv][8]		τinactivation[24]	τrepriming	Persistent current	Resurgent current	Ramp current	Window current	Location[8]
1.1	− 33	− 72	Fast	Fast[25]	+ [26]	−	−	−	C/P/H
1.2	− 24	− 63[a]	Fast	Fast[27]	−	+	−	−	C/P[d]/H
1.3	− 24	− 70	Fast	Fast[28]	+ [1]	−	+	−	C/P[d]/H
1.4	− 26	− 56[a]	Fast	Fast[28]	−	− [29]	−	−	M
1.5	− 44	− 87	Slow	Fast[30]	−	− [29]	−	−	H/C/P[d]
1.6	− 29	− 72	Fast	Fast[31]	+	+ [29]	−	−	C/P/H
1.7	− 25	− 74	Fast	Slow[31]	−	− [29]	+	−	P/C
1.8	− 02	− 36[b]	Slow	Fast[32]	+ [33]	+ [22]	+ [33]	−	P/H[34]
1.9	− 54	− 54[c]	Slow	Fast[21]	+	−	+ [21]	+	P

All V_1/2 data were collected in HEK293 unless otherwise noted.+, yes; −, no. Isoform locations are listed in order of decreasing prevalence.

^[a]CHO.

^[b]Co-expressed with β1.

^[c]DRG.

^[d]Embryonic.

Abbreviations: τ, time constant; C, central nervous system; P, peripheral nervous system; H, heart; M, skeletal muscle. Data on current properties (persistent, resurgent, ramp, window) taken from Rush et al.^[20] unless otherwise noted.

1.2. Channel isoforms

Ten gene loci have been identified that encode Na_V α-subunits, nine of which—Na_V1.1 to 1.9—are voltage-dependent, and a tenth, Na_Vx, that is involved in salt-sensing.^[16] Each isoform exhibits distinct voltage dependence and gating properties (Table 1), which are further modulated by alterations in channel splicing,^[17] post-translational modification,^[18] and the binding of auxiliary proteins.^[19] In the peripheral nervous system, for example, isoforms 1.3, 1.6, 1.8, and 1.9 exhibit incomplete inactivation,^[20] producing persistent currents that facilitate action potential initiation. Na_V1.3, 1.7, 1.8, and 1.9^[21] also have the capacity to amplify slow, subthreshold stimuli through so-called ramp currents. Similarly, resurgent currents produced by rapidly recovering Na_V1.6 and 1.8^[22] increase AP firing frequency. The overlapping voltage dependence of activation and inactivation of Na_V1.9 furnishes an open, non-inactivating channel population primed to contribute to AP firing.^[23] Accordingly, the identity and distribution of Na_V isoforms determines signal threshold and dictates electrical responses to generator stimuli.

In both human and animal models of pain there is significant evidence that injury-induced Na_V redistribution produces ectopic AP firing patterns resulting in mechanical allodynia, thermal hyperalgesia, and spontaneous nociception.^[35] In healthy dorsal root ganglion (DRG) cells, Na_Vs are concentrated at the proximal segment and the nodes of Ranvier. Na_V1.2 and 1.3 appear only in embryonic DRGs, whereas adult cells express predominantly Na_V1.7, 1.8, and 1.9, and to a lesser extent Na_V1.1 and 1.6. In animal models of neuropathic pain, however, Na_V1.3 expression is consistently upregulated.^[36–38] Further, Na_V1.8 has been shown to redistribute to the site of injury.^[39] These changes, as well as the concomitant downregulation of Na_V1.9,^[40] alter the excitability of injured neurons. Specifically, the upregulation of Na_V1.3 and Na_V1.8 facilitates channel repriming and generator potential amplification, respectively. Simultaneously, the downregulation of Na_V1.9 produces a hyperpolarized resting potential, enabling the recycling of otherwise inactivated channels (see hyperpolarized V_1/2 of inactivation: Table 1, column 2).^[41] Consistent with animal models, Na_V1.3, 1.7, and 1.8 have been found to accumulate in the axonal tips of human neuromas.^[42–45] Recently, Na_V1.1^[46,47] and 1.6^[48] upregulation has also been implicated in animal models of neuropathic pain, though the underlying pathophysiology of either isoform has not been investigated extensively. By contrast, in models of inflammatory pain, increased expression of Na_V1.3, 1.7, and 1.8 has been noted.^[49,50,51] However, as compared to neuropathic pain, fewer studies of inflammation-induced changes in Na_V expression have been conducted. Similarly, the effect of injury or inflammation on channel splicing, post-translational modification, and trafficking is not well understood at this time.

1.3. Channel modulation in nociception

The dysregulation and redistribution of neuronally-expressed Na_V channels following injury are prompted by signals arising from immune- and/or glial cell-derived mediators.^[52] Briefly, in inflammatory pain or tissue injury, immune cells at the site of injury secrete interleukins and growth factors that bind tyrosine kinase receptors, G-protein-coupled receptors, and cytokine receptors on nociceptors. Subsequent activation of several protein kinases, including protein kinase A, protein kinase C, and extracellular signal-regulated kinases, results in phosphorylation of internal stores of Na_Vs and promotes translocation to the cell membrane (Figure 3).^[53,54] Following nerve injury, similar pathways are activated by Schwann cell-released glial mediators.^[55–57] Increased expression of Na_Vs at the cell membrane, among other changes, facilitates action potential initiation and propagation, which are perceived as painful stimuli. Repetitive AP firing induces changes in the central nervous system (CNS)^[58] that manifest as chronic inflammatory or neuropathic pain.

Figure 3.

Upstream regulators of channel function in inflammatory and neuropathic pain. Created with BioRender.com. Receptors (blue): tyrosine kinase, G-protein coupled receptors, cytokine receptors. Circles (blue): protein kinase A, protein kinase C, extracellular signal-regulated kinases. Channels (red): Na_Vs. Sheath (blue): Schwann cells. Abbreviations: TRP, transient receptor potential channel; ASIC, acid-sensing ion channel; K_V, voltage-gated potassium channel; Ca_V, voltage-gated calcium channel; GABAR, gamma-aminobutyric acid receptor; GluR, glutamate receptors; TACR1, tachykinin receptor 1.

Although the involvement of Na_Vs in inflammatory and neuropathic pain has been studied extensively for the past several decades, there are significant gaps in our understanding of pain pathogenesis. Injury-induced changes in the subcellular distribution of specific Na_V isoforms are poorly characterized, as are the trafficking mechanisms responsible for redistribution. Similarly, the regulatory pathways that control channel localization have not been fully elucidated. The contributions of specific Na_V isoforms to inflammatory and neuropathic pain vary, and how such changes alter APs and the corresponding sensory experience remains unclear. In an effort to answer these outstanding questions, multiple, disparate technologies have been developed to study Na_Vs. Herein, we discuss two categories of tools: chemical tools (derivatized small molecules and proteins) and biological tools (channel tagging, knockdown, and knockout). The full complement of these methods and their advantages and disadvantages for investigations of Na_V function and physiology are detailed.

2. Chemical Tools to Study Na_Vs

A large and diverse number of eukaryotic organisms utilize Na_V-specific small molecules and peptides for the purposes of selfdefense and/or predation.^[59,60] These toxins can induce paralysis, pain, heart arrhythmia, and death depending on their relative potency and mode of action against Na_V subtypes. Research to examine the physiological effects of natural poisons has shaped our current understanding of Na_Vs and their role in electrical signaling. Animal toxins, in particular tetrodotoxin and saxitoxin, were the first tool compounds used to identify and to classify Na_V subtypes and to interrogate channel function.^[61,62] Following these investigations, additional insights into Na_V physiology have been aided with the identification of subtype-selective Na_V modulators, semi-synthetic modification of natural toxins, and small molecule design. Other molecular probes that target Na_Vs include radiolabeled, protein affinity, and fluorescent imaging agents. Although a small number of these tool compounds have been used in specific studies aimed at understanding the action of Na_Vs in electrogenesis and nociception, most remain underutilized and many require additional optimization. The remainder of this section will discuss the historic contributions of chemical reagents to Na_V research, as well as potential applications of these tool compounds to measure Na_V expression, localization, trafficking, and turnover.

2.1. Animal toxins

Tetrodotoxin (TTX) and saxitoxin (STX) were the first small molecule animal toxins discovered to act on Na_Vs.^[61,62] Originally identified as the paralysis-inducing components of poisonous pufferfish and bivalves, respectively, TTX and STX were determined to specifically block sodium currents in electrically excitable cells. These toxins have become mainstays in the field of neuroscience due to their low nanomolar potency and selectivity for a subset of Na_V channel subtypes, so-called TTX-sensitive Na_Vs (Na_V1.1, 1.2, 1.3, 1.4, 1.6, 1.7). Markedly higher concentrations of toxins are needed to block TTX-resistant channels (Na_V1.5, 1.8, 1.9). The classification of Na_Vs as TTX-sensitive (TTX-S) or resistant (TTX-R) is still employed today, and TTX in particular finds considerable use for blocking TTX-S sodium currents in studies to understand neuronal function, including those related to pain perception.^[63,64]

Since the discovery of TTX and STX, numerous toxins have been isolated from venomous animals and toxic algae that are potent towards Na_Vs. No less than eight different receptor sites have been identified on Na_Vs;^[65] small molecules and peptides binding to any one of these locations modulate channel activity (Figure 4). Most interact directly or allosterically with the voltage-sensing domains (Sites 2–7), acting as channel agonists by modifying Na_V gating and voltage dependence. In contrast, simple conductance blockers target Sites 1 and 8 (note: Site 8 and ‘local anesthetic binding site’ have been used interchangeably). Cationic TTX, STX, and μ-conotoxins occlude Na⁺ passage by binding to the extracellular mouth of the pore domain (Site 1), whereas hydrophobic small molecule drugs and anesthetics such as lidocaine, cocaine, and carbamazepine achieve a similar effect by targeting the inner pore (Site 8).^[65] Almost all toxins exhibit selective binding to a particular channel conformer (so-called state dependence) and/or higher affinity upon repeated channel opening (use dependence). The majority of toxins that interact with the voltage sensors are state-dependent, preferentially associating with the closed, open, or inactivated form of the channel. Pore blockers such as TTX and STX tend to be modestly use-dependent, displaying enhanced potency upon more frequent channel opening (Table 2). The multifarious activities of Na_V-targeting toxins have been reviewed extensively.^[65–68]

Figure 4.

Voltage-gated sodium channel animal toxin binding sites. A) Side view of S5–S6 helices and binding sites therein. B) Top view of Na_V and binding sites therein. C) Secondary structure motifs of animal toxin binding sites. Domain I (orange), Domain II (red), Domain III (gray), Domain IV (cyan).

Table 2.

Animal toxins that target Na_Vs.^[65,69–72]

Effect	Binding site	INa+	V1/2 activation	τinactivation/τrepriming	State/use dependence	Toxin
Antagonism	1	↓	−	−/−	Use	saxitoxin tetrodotoxin μ-conotoxin
	8	↓	−	−/↓	State (I) Use	anesthetic anticonvulsant antiarrhythmic antidepressant
	UD[73–75]	↓	UD	UD	UD	kalkitoxin jamaicamides polygodial
Mixed	4	↓	↓	−/−	State (varied)	δ-palutoxin μ O-conotoxin β-scorpion, spider toxins
Agonism	2	↓	↓	↓/−	State (O)	aconitine batrachotoxin grayanotoxin veratridine hoiamide A antillatoxin
	3	−	↓	↓/−	State (C)	δ-atracotoxin α-scorpion, sea anemone, wasp toxins
	5	−	↓	−/↑	State (O)	brevetoxin ciguatoxin
	6	−	↓	↓/−		δ-conotoxin
	7	−	↓	↓/−	State (O) Use	pyrethroid

↑/↓, increase/decrease; –, no effect. These properties are general to the binding site but may vary somewhat for individual toxins. Abbreviations: I_Na ⁺, sodium channel current; τ, time constant; UD, undetermined; O, open; C, closed; I, inactivated.

Despite the structural and functional diversity of existing small molecule and peptide toxins, relatively few specifically target single Na_V subtypes. In general, toxins that bind in the large, hydrophobic inner pore (Sites 2, 5, and 8) tend to be the least selective for different Na_V isoforms. By contrast, those that bind the polar extracellular surface (Sites 3 and 4) or the anionically charged extracellular pore (Site 1) better differentiate among subtypes. TTX and STX show little to no subtype-selectivity among the six TTX-S channels. Conversely, members of the μ-conotoxin family, as well as scorpion and spider toxins, target single Na_V isoforms (vide infra, section 2.5).^[76] These selective channel antagonists have been utilized for single isoform ‘knock-out’ to elucidate the roles of Na_V subtypes in nociception.^[77–80] In the same vein, studies with Na_V agonists give insight into how alterations in channel gating and voltage dependence underlie inflammatory pain.^[81] Recent cryo-EM structures of Na_Vs in complex with animal toxins—including STX,^[6] TTX,^[6] the μ-conotoxin KIIIA,^[3] and the spider toxins HwTx-IV^[6] and ProTx-II^[82]—should facilitate the rational design of more selective agents with potential therapeutic benefits.^[83–85] Future investigations with natural toxins and derivatives thereof present opportunities to quantify, isolate, and/or image select channel subtypes in models of injury or inflammation.

2.2. Radiolabeled probes

Radiolabeled probes offer several advantages for in vitro and in vivo studies of sodium channels. First, due to the minimal structural perturbation of radioisotope incorporation—as compared to the addition of a sterically large, hydrophobic fluorophore, for example—the potency, binding kinetics, and isoform selectivity of a radiolabeled probe typically mirror that of the parent toxin or small molecule investigational drug. Further, because radioactivity can be accurately measured, radioligands make possible precise quantification of protein target concentration. Although the spatial resolution of autoradiography is low, recent advances in positron emission tomography (PET)^[86] and single-photon emission computed tomography (SPECT)^[87] now enable 3D in vivo imaging on the order of a few hundred micrometers resolution. As such, beyond their uses in the quantification of protein receptors in cells, Na_V-specific radiolabeled probes have potential as diagnostics for monitoring channel accumulation at sites of injury or inflammation.

Two classes of radiolabeled probes have been developed for Na_V study. The first class comprises channel pore blockers, most notably saxitoxin (STX) and tetrodotoxin (TTX), both of which were radiolabeled through tritium exchange^[88,89] and used extensively in early experiments to understand the role of Na_Vs in electrogenesis (Figure 5). [¹⁴C]Guanidine^[90] and [³⁵S]gonyautoxin^[91] also fall into this class, although these probes have been employed only sparingly in assays designed for neurotoxin detection and never in the study of electrical signaling. The second class of probes is composed of state-dependent molecules, such as the Site 2 toxin, [³H]batrachotoxinin-A 20-α-benzoate ([³H]BTX-B),^[92,93] and the anti-ischemia drug, [³H]lifarizine.^[94] These molecules have been used almost exclusively for competition experiments to locate ligand binding sites on the channel. The remainder of this section will focus on studies performed and insights gained with the Site 1 radiolabeled TTX and STX probes, as these agents are, by far, the most widely described in the literature.

Figure 5.

Radiolabeled probes targeting voltage-gated sodium channels at different receptor sites. A) Site 1. B) Site 2. C) Site 8. References: [³H]STX,^[88] 11-[³H]TTX,^[114] [¹⁸F]STX,^[125] [³⁵S]GTX II/III,^[91] [³H]BTX-B,^[92] [³H]lifarizine.^[94]

[³H]Saxitoxin ([³H]STX) has been most generally used as a radioprobe for Na_V studies due to its ready access from naturally isolated toxin and tritiated water.^[88] This procedure rapidly and efficiently exchanges the C11 protons of STX, a significant advantage over the previously established, low-yielding method for labeling TTX with tritium gas.^[89] With its high affinity for TTX-S channels (K_i =3.4 nM)^[99] and favorable binding kinetics, [³H]STX is arguably the preeminent reagent for measuring Na_V expression levels in cell membranes. Channels labeled by [³H]STX can be quantified by scintillation counting and/or isolated by protein liquid chromatography and gel electrophoresis.^[95–97] Purification of Na_Vs facilitated by [³H]STX advanced the subsequent design of Na_V-specific antibodies (see Section 3.1),^[98] which have become the de facto proxies for Na_V expression.

Early estimates of single channel conductance and Na_V density in neuronal cells were made by bathing giant squid axons in [³H]STX.^[99] Other experiments measured channel density at the nodes of Ranvier (Na_V density ~12,000/μm²) compared to the internodal space (Na_V density ~25/μm²),^[100] which provided a physical rationale for saltatory conduction and paved the way for subsequent investigations of electrogenesis. Studies with [³H]STX also quantified the effects of upstream regulators of Na_V surface expression, including nerve growth factor (NGF),^[101] PKC,^[102,103] and ERK,^[104] and aided in the characterization of auxiliary proteins such as the beta subunits^{[105,106,107]} and ankyrin G.^[108] In addition, some of the first work describing changes in Na_V expression in response to physical injury^[109,110] and diabetes^[111] was enabled with [³H]STX.

The experimental utility of [³H]STX is limited by washout of the C11 tritium labels in aqueous solution, which necessitates working with this compound at temperatures ≤4°C to prevent errors in channel quantification.^[112] While [³H]tetrodotoxin ([³H]TTX) can be employed for purposes identical to [³H]STX, difficulties in obtaining this material have severely limited its use.^[89] Hence, modified radiolabeled TTX derivatives have been designed with non-exchangeable tritium labels such as [³H]ethylenediamine ditetrodotoxin^[113] and 11-[³H]tetrodotoxin^[114] in order to facilitate experiments in live cells at 25–37°C. However, preparing these probes is also quite challenging and has greatly restricted their utility. Of note, a 2017 report employed 11-[³H]TTX for in vivo determination of ADME (absorption, distribution, metabolism, and excretion) in rats, preclinical work to validate the application of TTX as a low-dose pain relief drug.^[115]

In the past two decades, radiolabeled STX and TTX have witnessed declining application for Na_V studies, passed over in favor of Na_V-specific antibodies (see Section 3.1). For STX, this may be due in part to lack of accessibility, as this compound remains on the Schedule 1 Chemical Weapons list. With the availability of de novo chemical syntheses of STX,^[116–122] it should be possible to generate new tritiated derivatives that are stable to solution exchange. The physicochemical properties of this molecule are optimally suited for its development as a chemical probe.^[123]

De novo synthesis of STX has provided access to a novel [¹⁸F]-labeled STX derivative. Despite structural modification of the parent toxin, [¹⁸F]STX retains high affinity for STX-sensitive Na_Vs (IC₅₀ =46±7 nM vs. Na_V1.4;^[124] 10.6 nM vs. PC12 cells (rNa_V1.2 and 1.7)) and has been used successfully in positron emission tomography (PET) imaging of ectopically expressed Na_Vs in a rat model of neuropathic pain (spared nerve injury). Findings from this work confirm previous reports of elevated Na_V concentrations at a nerve injury site.^[125] Currently, this study with [¹⁸F]STX remains one of a small number of reports highlighting the use of a chemical tool for labeling endogenous TTX-S Na_Vs in vivo. Others describe ¹²⁵I-derived Na_V radioligands for SPECT imaging; however, rapid metabolism of these agents severely limits their use.^[126,127] New radiolabeled small molecules specific for Na_V subtypes have great potential to enable studies of Na_V physiology.

2.3. Crosslinking probes

Photoaffinity probes derived from toxins and other small molecules have been applied to great effect to map Na_V toxin receptor sites^[128–130] and, in select cases, have helped reveal elements of the mechanisms for voltage dependence.^[131] More than a dozen unique photo-crosslinking probes targeting Sites 1–5 of Na_Vs have been described. In general, these tools bear either a phenylazide or phenyldiazirine moiety appended from an alcohol or amine group on the ligand (Figure 6). Incorporation of these photoreactive groups can be synthetically challenging and the sizeable steric modification can markedly alter ligand potency and binding kinetics. Irreversible channel modification with photo-crosslinking agents is typically inefficient and therefore complete inhibition (or activation) of a population of channels is not possible. Nevertheless, photoreactive reagents that bind individual channel subtypes would have value for targeting endogenous channels and for introducing fluorescent labels or cofactors that could enable real-time monitoring of channel turnover in healthy and injured neurons. Some of the more promising probes for these purposes are discussed herein.

Figure 6.

Photo-crosslinking probes for the study of voltage-gated sodium channels. Abbreviations: NAP, 4-azido-2-nitrophenyl; OAB, ortho-azidobenzoate; ANB, azidonitrobenzoyl; AB, azidobenzoate. References: NAP[³H]lysine TTX,^[136] [³C]BTX-OAB,^[131] GTX diazirine,^[129] ANB-¹²⁵I₁ScTx,^[133] HwTx L-photomethionine,^[130] p-AB PbTx,^[132] riluzole azide.^[138]

Seminal investigations describing Na_V identification and purification utilized photoaffinity labeling with scorpion toxinderived tool compounds.^{[133,134,135]} In these studies, an ¹²⁵I-radiolabelled, photoactivatable azidonitrobenzoyl scorpion toxin (ANB-ScTx) was prepared from Leiurus quinquestriatus scorpion venom. Such tools are advantaged over previously developed radioligands (i.e., [³H]STX) given the covalent nature of the photo-crosslinking reaction–Na_Vs can be located and quantified without any concern for probe washout. ANB-ScTx proved instrumental for the initial characterization of Na_Vs. Unfortunately, the low yield of the photolysis reaction limits the application of this probe for protein quantification, and its use has largely fallen out of favor along with the aforementioned radioligands.

Several small molecule-based photoaffinity probes have been prepared for the purpose of inhibiting Na_V activity. 4-Azido-2-nitrophenyl-[³H]lysine tetrodotoxin (NAP[³H]lysine TTX) has been demonstrated to irreversibly inhibit the rising phase of action potentials in the crab giant axon following irradiation.^[136] Other TTX derivatives, including NAP[³H]ethylenediamine TTX and arylazido-β-alanine TTX, have been used for analogous experiments in toad muscle.^[137] No other reports highlighting the application of these reagents have appeared, however, perhaps due to the extreme difficulties in their preparation. Given the reported efficiencies of the photo-crosslinking reactions, the development of subtype-selective variants of these tools warrants further investigation.

Recent research to develop photoaffinity probes for Na_Vs has focused on the design of small molecules capable of modulating channel voltage dependence. For example, riluzole, a state-dependent ligand that blocks sodium conductance and hyperpolarizes the voltage dependence of activation, has been functionalized with an azido group to yield a photoactivatable derivative. Although photo-crosslinking increases the inhibitory effects of riluzole on sodium conduction (~70% block after crosslinking under saturating conditions), the ability of this probe to hyperpolarize voltage dependence is preserved.^[138] This reagent is a first-in-class tool compound given its ability to alter the energetics of Na_V gating rather than simply block channel conduction. These types of probes may hold promise for determining the contributions of sodium channel activation (in the absence of indeterminate upstream signaling pathways) to ectopic electrical signaling.

To date, only one Na_V-specific covalent probe has been reported that does not require photoactivation. Saxitoxin ethyl maleimide (Figure 7A) rapidly and efficiently binds to TTX-S channels (as validated against rNa_V1.2 and 1.4) without the requirement of additional reagents or protein mutation (>80% irreversible channel block in six minutes under saturating conditions).^[139] As with photoaffinity probes, the specific amino acid residue(s) with which STX ethyl maleimide reacts is currently unknown. If properly functionalized, this type of probe presents a potentially unique tool compound for tagging and tracking endogenous TTX-S Na_Vs.

Figure 7.

A) Cysteine-reactive probe for irreversible inhibition of Na_Vs.^[139] B) Photocaged STX derivative for spatiotemporal control of Na_Vs.^[140]

The development of chemical probes that target and covalently react with Na_Vs is an underexplored research area given the potential of such compounds to selectively inhibit channel activity and to facilitate channel isolation and purification. Continued efforts to advance these types of reagents will empower research to understand Na_V physiology.

2.4. Photocaged probes

Modified saxitoxins consisting of a photocleavable ‘caging’ group have been made available through de novo synthesis (Figure 7B).^[140] These compounds enable rapid, precise, spatiotemporal control of Na_Vs through light-induced uncaging. The released inhibitor is a nanomolar potent, reversible binder of TTX-S Na_Vs. Electrophysiological experiments demonstrate the value of these tool compounds for selectively inhibiting action potentials in dissociated neurons and brain slice. Further, recordings in a corpus collosum slice preparation reveal differences in the susceptibility of myelinated vs. unmyelinated axons to Na_V block. These reagents are first-in-class photocaged, Na_V-selective inhibitors, and will enable studies of the dependence of neuronal function on Na_V spatial organization. This class of tools has been recently expanded with the development of the photocaged spider toxin HwTxIV-Nvoc.^[141 The caged peptide toxin has demonstrable in vivo efficacy in both mice and zebrafish after several minutes of ultraviolet light application.

2.5. Fluorescent probes

The development of Na_V-targeting fluorescent probes to monitor changes in channel trafficking and localization in healthy cells and following response to injury has proven particularly challenging. Attachment of large fluorogenic dye molecules to ligands that associate to Na_Vs often adversely affects solubility, potency, and binding kinetics. Partly for this reason, the earliest fluorescent reporters were designed to measure Na⁺ influx^[142] rather than to mark the location and/or expression level of the channel itself. The most common Na⁺ ionophore, benzofuran isophthalate (SBFI), has been used to examine the mechanisms of action of antidepressants, among other studies. Unfortunately, Na⁺-binding dyes^[143] remain plagued by low sensitivity and poor signal-to-noise, and are considerably less effective than Ca²⁺ indicators like Fura-2.^[144] To overcome these limitations, mutation of Na_V to increase Ca²⁺ permeability has been demonstrated, which enables application of a fluorescent Ca²⁺ sensor (e.g., GECO) to measure channel activity.^[145] While potentially valuable for drug discovery, this solution for measuring Na_V response is not suitable for studies in neuronal cells.

In an effort to develop Na_V fluorescent imaging agents, a number of small molecule channel binders have been modified with fluorophores (Table 3). N-Methylanthranilate (NMA), a weakly emissive fluorophore, was affixed to batrachotoxinin A (BTX A) to generate an analogue of bactrachotoxin (BTX), a potent channel agonist. This probe has been utilized for visualizing channel populations at the nodes of Ranvier in a mouse nerve fiber, but is too dim for tracking channels in cells expressing lower Na_V concentrations.^[146] NMA-BTX also has the disadvantage of being time-limited, as prolonged channel activation results in Na_V internalization.^[147,148] Given the development of improved imaging technologies and the relative ease of esterification of the C20-alcohol in batrachotoxinin-A, this tool and others like it may very well warrant a second look. The hydrophobicity of molecules like NMA-BTX can, however, present problems due to nonspecific protein and membrane binding. Other disadvantages of NMA-BTX stem from the suboptimal photophysical properties of NMA; nevertheless, replacing this fluorophore with larger, brighter dyes may come with a serious detriment to ligand-channel affinity and specificity.

Table 3.

Fluorescent probes for the study of voltage-gated sodium channels. IC₅₀ values were collected against listed model system unless otherwise noted.

^[a]Data provided by the manufacturer.

^[b]Modified at W50X.

^[c]Collected in HEK. Similar potencies exhibited vs. hNa_V1.1–1.3, hNa_V1.6.

Toxin-fluorophore conjugates derived from STX, including STX-Cy5 and STX-DCDHF, offer a number of salient advantages over NMA-BTX and related probes, foremost of which is readily reversible channel binding. These probes have been successfully employed for single-molecule and super-resolution imaging experiments to visualize and track channels in PC12 cells.^[149] Unfortunately, extending the application of these molecules to neuronal cells has been plagued by nonspecific ligand binding and poor signal-to-noise. Further optimization of probes such as STX-Cy5 may ultimately give way to a high-precision reagent for real-time, live cell Na_V imaging.

Alternative approaches for generating Na_V-selective imaging agents have exploited lysine modification of peptide toxins (e.g., Tityus serrulatus scorpion β-toxin, Conus geographus and Conus kinoshitai μ-conotoxins) with AlexaFluor and BODIPY dyes or unnatural amino acid incorporation.^[154,155] The availability of these toxins from nature or through de novo peptide synthesis greatly facilitates this research. Thus far, fluorescent peptide toxin conjugates have been primarily employed for proof-of-concept imaging and luminescence resonance energy transfer (LRET) experiments.^[156] However, given the potencies of these toxin derivatives, future experiments to visualize Na_V expression in live cells may be possible. In one such example, Gonzales, et al.^[157] used a modified Haplopelma schmidti spider toxin, Hs1a-Cy7.5, to image Na_V1.7 populations in mouse sciatic nerve ex vivo. With nanomolar potency against neuronal TTX-S Na_V isoforms and an infrared absorption/emission spectrum, Hs1a-Cy7.5 has potential application for in vivo Na_V imaging. To our knowledge, such studies have not yet been reported.

In addition to potent small molecule- and peptide-based fluorophores, chemically modified nanoparticles have been described for Na_V imaging. Europium-doped yttrium vanadate particles coated with guanidine groups represent one such design. Although n-alkylguanidine derivatives are millimolar inhibitors of Na_Vs,^[158] the coated nanoparticles display myriad copies of the guanidine ligand and thus the particle has high avidity for the channel. Using this probe, Na_Vs in cardiac myocytes have been imaged.^[159] Unfortunately, as with other fluorescent agents, follow-up studies in neuronal cells or live tissue have not appeared. Perhaps not surprisingly, the guanidine-studded nanoparticle does not display isoform selectivity. Recent disclosures of a number of isoform-specific Na_V ligands, as discussed below, may portend the coming of age of imaging probes that target Na_V subtypes.

2.6. Isoform-selective, investigational drugs

With the explosion of interest in Na_Vs as therapeutic targets for analgesic development, ignited by the disclosure of genetic data in human subjects incapable of sensing pain, a rush to develop potent, isoform-selective molecules has ensued.^[160] Much of this effort has targeted hNa_V1.7, as defects in this subtype have been implicated in paroxysmal pain disorders (gain-of-function)^[161] and congenital insensitivity to pain (lossof-function).^[162] This portion of the review will briefly highlight a select number of examples of publicly accessible, nanomolar potent tool compounds used for the study of Na_Vs in neuronal signaling and nociception (Table 4). The topic of Na_V drug development has been reviewed extensively, and interested readers are directed to the following works.^[163–165] Disappointingly, the in vitro and preclinical efficacy of investigational analgesic medicines has witnessed little in the way of clinical success, for reasons fully elucidated elsewhere.^[166,167]

Table 4.

IC₅₀ values of Na_V1.7 selective probes, given in μM. All IC₅₀ values were collected in HEK stably expressing Na_Vs unless otherwise noted.

^[a]Data collected in CHO.

Despite the lack of clinical success,^[168,169] studies with Pfizer’s PF-05089771 have greatly informed our understanding of the contributions of Na_V1.7 to electrogenesis and the threshold for signal conduction.^[170] Unfortunately, PF-05089771 differentiates poorly between Na_V1.2 and Na_V1.7 (<10-fold selectivity),^[171] likely resulting in unwanted off-target effects at concentrations necessary to achieve clinical efficacy. The challenge of distinguishing among the nine Na_V isoforms and avoiding other off-target channels and receptors is a recurring problem for Na_V1.7-selective small molecule and protein design.^[172–174] Crystal structure determinations of a Na_V1.7 chimera with similarly potent, yet nonselective, sulfonamides analogous to PF-05089771 have revealed the structural basis for their mechanism of action and enabled the development of subtype-specific inhibitors.^[175,176] Advancement of PF-06456384, which displays more than 300-fold selectivity for Na_V1.7 over other Na_V subtypes, affords the most potent, state-dependent Na_V1.7-specific small molecule known to date (IC₅₀ =0.01 nM).^[177] Alternative sulfonamide-based ligands are also reported to be highly selective, state-dependent Na_V inhibitors, with AM0466 and AM-8379 exhibiting 100- to 1000-fold affinity for Na_V1.7 over any other isoform.^[178–181] Recently, SiteOne Therapeutics has disclosed a first-in-class, state-independent Na_V1.7-selective inhibitor, ST-2262, modeled after the natural product saxitoxin.^[182] This work demonstrates the potential for targeting the outer vestibule of the channel pore to achieve isoform specificity.

In parallel with Na_V1.7 research, Na_V1.8 and Na_V1.9 have also found interest as analgesic targets, studies prompted by human genetic data.^[184,185] Several exquisitely selective, state-dependent investigational drugs have been developed that bind to Na_V1.8, including Pfizer’s PF-04885614^[186] and Abbott’s (now Abbvie) A-803467 (Table 5), both of which are commercially available. A-803467 has been instrumental for elucidating the contributions of Na_V1.8 to nociception in models of spinal injury and osteoarthritis.^[187–189] These experiments were especially important given ambiguous results from Na_V1.8 knockdown and knockout studies (vide infra, section 3). In contrast, while Na_V1.9 has been implicated in several models of neuropathic pain,^[190–192] no Na_V1.9-selective inhibitors have been disclosed to date.

Table 5.

IC₅₀ values of isoform selective probes, given in μM. All IC₅₀ values were collected in HEK stably expressing Na_Vs unless otherwise noted.

^[a]Data collected in Xenopus oocytes; subsequent studies in HEK cells indicate that this compound is also potent against Na_V1.1.^[195]

^[b]EC₅₀.

Unlike the number of selective inhibitors available for investigations of Na_V1.7 and 1.8, few exist that target other potentially interesting channel isoforms (i.e., Na_V1.1, 1.2, 1.3, 1.6). Without compelling genetic evidence that links these subtypes to nociception, there has been less interest in developing selective inhibitors as analgesics. Currently, two molecules have been reported to bind selectively to Na_V1.6 (Table 5): (1) 4,9-anhydrotetrodotoxin, a natural derivative of TTX with >40-fold selectivity for Na_V1.6;^[193] and (2) the Centruroides noxius beta-scorpion peptide (Cn2),^[194] an animal toxin Na_V agonist that evokes resurgent current in this isoform. Neither of these compounds has enjoyed widespread use for channel studies. In fact, a recent report has shown that 4,9anhydroTTX also inhibits Na_V1.1 and 1.3 (particularly the former).^[195] Two additional probes are modestly selective for Na_V1.1 (Table 5): Icagen’s ICA 121431,^[196] a channel blocker that is ~6-fold less potent for Na_V1.3; and Heteroscodra maculata δ-theraphotoxin-Hm1a,^[197] a tarantula peptide toxin that inhibits fast inactivation in Na_V1.1 with ~6-fold selectivity over Na_V1.2 and 1.3. While studies with Hm1a suggest a role for Na_V1.1 in mechanical pain sensation, neither this probe nor the Icagen molecule are selective enough for investigations of single Na_V isoforms in neuronal cells. To the best of our knowledge, no tools capable of selectively targeting Na_V1.2 have been described. Continued efforts to advance potent, subtype-specific Na_V modulators is warranted in order to deconvolute the physiological function(s) of each isoform in electrogenesis and nociception.

3. Biological Tools to Study Na_Vs

The successful development and application of small molecule, Na_V-selective reagents notwithstanding, much of our understanding of Na_V physiology and nociception is owed to the availability of biological techniques and tools. While these methods do not afford the temporal response displayed by most chemical probes, antibody labeling, oligonucleotide-based knockdown, genetic knockout, and protein epitope tagging offer greater precision for targeting Na_Vs. The advent and use of biological methods for Na_V investigations are highlighted below.

3.1. Antibodies

Na_V antibodies (Abs) that bind to large intracellular loop regions of Na_Vs have been employed extensively in studies of Na_V physiology, including nociception.^[36–48] Na_V Abs were originally developed in individual labs and screened for specificity by monitoring heart/CNS/PNS cross-reactivity^{[198,199,200]} or accumulation in Na_V knockout mice.^[201] Several companies (e.g., Alomone and Abcam) have since developed a host of orthologue- and isoform-specific Na_V Abs. While generally useful for immunohistochemistry and Western blot experiments, intracellular antibodies are unfit for live cell imaging due to the necessity of cell permeabilization. In part for this reason, studies of Na_V trafficking in response to cell or tissue injury have been lacking. Recent developments in channel epitope tagging, however, are changing this landscape (Section 3.5).

The preparation of Abs that target extracellular regions of Na_Vs has witnessed some key advances in the past five years. Amgen has synthesized several exquisitely potent Ab-toxin conjugates using peptide toxins GpTx-1^[202] and JzTx-V^[203] from Grammostola porteri and Chilobrachys jingzhao spider venom, respectively. Following optimization studies to improve the potencies and selectivities of these otherwise nonspecific toxins for Na_V1.7, these molecules were ligated to human immunoglobulin G-derived Abs, which exploit binding to the neonatal Fc receptor (FcRn) to reduce lysosomal degradation and extend serum half-life.^[204] The resulting conjugates display nanomolar potency against the target and long circulating half-lives (t_1/2 = 80 hours, as compared to 0.6 h for the unconjugated toxin in the case of GpTx-1), and can be used in vivo for extended block of Na_V function. Derivatives of these novel Ab conjugates could prove quite useful as research tools for the study of channel localization.

Three additional extracellular Abs are available for channel study, targeting Na_V1.3,^[205] 1.5,^[206] and 1.8.^[207] While the Na_V1.5 Ab has been successfully employed for the study of cancer progression,^[208] Na_V1.3 and 1.8 Abs, both from Alomone, have found limited use. The former is nonselective, associating with Na_V1.1 and 1.2 as well as 1.3, and the latter does not bind the human Na_V1.8 orthologue. No reports describing the application of these reagents for cellular imaging have appeared.

3.2. Oligonucleotide-based knockdown

One of the most prominent techniques for studying Na_V physiology is oligonucleotide-based knockdown.^[209] Antisense and RNA interference methods make possible selective depletion of specific Na_V isoforms with a level of precision that is unrivaled by other technologies (genome engineering, notwithstanding). The major drawbacks of such methods are that protein knockdown is never quantitative and the process is slow (i.e., tens of hours are generally required to lower protein levels).^[210] Since the late 1990s, antisense knockdown featured as the preeminent method for deconvoluting the roles of individual Na_V isoforms in various animal models of pain. Early experiments were largely conducted by microinjection of Na_V-specific antisense oligodeoxynucleotides.^[211] Later, as the field of RNA silencing developed, lentiviral delivery of siRNA was popularized.^[212] Depending on the choice of method, model organism, and oligonucleotide sequence, mRNA and protein concentrations can be reduced by 20–80%; however, the exact percent reduction of mRNA and protein is unpredictable and mRNA and protein concentrations are not strictly correlated.

Knockdown of different Na_V isoforms results in pain reduction in a variety of animal models (Table 6). Decreasing expression of Na_V1.3,^[213,214] Na_V1.6,^[48,215,216] and Na_V1.8^[211,217] was found to reduce mechanical allodynia in murine models of neuropathic pain. Knockdown of either Na_V1.3^[218,219] or Na_V1.8^{[212,220–222]} was also shown to reduce thermal hyperalgesia, although the contributions of the latter subtype are better documented than the former. In contrast, Na_V1.6 was only recently implicated in models of neuropathic pain and has been comparatively underexplored. Na_V1.7, while aggressively pursued as a pain target, has not been studied in animal models of neuropathic pain by antisense knockdown or RNA interference. However, knockdown experiments of Na_V1.7 have been performed in models of inflammatory pain and skin burn.^[223–225] Analogous studies of inflammatory pain have examined the role of Na_V1.8^{[211,226–228]} and to a lesser extent Na_V1.6,^[229] Na_V1.9,^[190] and Na_X.^[230] Na_V1.7 also plays a role in acute pain signaling.^[231,232] Although present in sensory neurons, albeit at low levels based on transcriptional profiling, the possible involvement of Na_V1.1 and 1.2 in nociception has not been explored using oligonucleotide-based knockdown methods.

Table 6.

Isoform-selective knockdown of Na_Vs.

Pain Type	Model	1.3	1.6	1.7	1.8	1.9	1.x
Neuropathic	Chronic constriction sciatic nerve	↓[218]	↓[216][b]	−	↓[220,211,217][b]	−	−
	Sciatic nerve entrapment	−	−	−	↓[212][c]	−	−
	Chronic compression DRG	−	↓[216][b]	−	−	−	−
	Spinal cord contusion	↓[213,219]	−	−	−	−	−
	Spared nerve injury	NE[239], ↓[238][a]	−	−	−	−	−
	L5/L6 spinal nerve ligation	−	↓[215][b]	−	↓[211,221,222]	NE[222]	−
	Chemotherapy	−	↓[48]	−	NE[211]	↓[191]	−
	Diabetes	↓[214][a]	−	−	−	−	−
Inflammatory	Complete Freund’s Adjuvant	−	−	↓[224]	↓[211,228]	NE[228]	−
	Zymosan	−	↓[229][b]	−	−	−	−
	Carrageenan	−	−	−	−	↓[190]	−
	Acetic/citric acid	−	−	↓[223][a]	↓[226]	−	−
	PGE2	−	−	−	↓[227]	−	−
	Imiquimod	−	−	−	−	−	↓[230][d]
Other	Incision	−		↓[231][c]	NE[211]	−	−
	Burn	−	−	↓[232][c]	−	−	−
	Cancer	−	−	−	−	−	↓[233][c]
	Venom	−	↓[234][c]	−	↓[235]	−	−

↓ = reduced pain-associated behavior. NE=no effect on pain-associated behavior. Knockdowns were performed with the injection of antisense oligodeoxynucleotides unless otherwise noted.

^[a]Adeno-associated virus (AAV)-shRNA knockdown.

^[b]siRNA transfection.

^[c]lentivirus.

^[d]Dicer-substrate RNAi (DsiRNAs).

Red indicates disagreement among published data.

Due to the number of and variability among animal pain models,^[236,237] knockdown data collected across disparate experimental protocols are often difficult to compare. Even among studies of the identical Na_V isoform in the same model system, conclusions can vary, as is the case for investigations of Na_V1.3 in spared nerve injury. Separate reports offer conflicting evidence as to the involvement,^[238] or lack thereof,^[239] of this subtype in nociception. Whether these inconsistencies are due to variations in experimental setup or off-target effects of the oligonucleotide^[33] remains unknown. With some notable exceptions,^{[231,223,212]} the effects of single isoform knockdown on the expression of off-target Na_Vs is not routinely measured. Compensatory upregulation of alternative Na_V subtypes, which is known to occur in genetic knockouts, may be responsible for some of the ambiguities associated with knockdown experiments.^[240–242] Further, whether knockdown alters the subcellular distribution of Na_Vs—critical for proper electrical signaling^[243]—has not been examined.

3.3. Knockouts

Mouse genetic knockouts (KOs) of every neuronal voltage-gated sodium channel have been described. Most global knockouts, including those for Na_V1.1,^[244] 1.2,^[245] 1.6,^[246] and 1.7,^[247] are lethal within one month of birth. While improvements in cross breeding have facilitated the generation of a Na_V1.7 KO that survives to adulthood,^[248] the study of other isoforms in animal models of pain has proven challenging. One solution involves the development of a Na_V1.8-Cre mouse,^[249] which allows for selective, conditional knockout of individual Na_V subtypes (or other protein targets of interest). Although Na_V1.8-expressing neurons were thought to be almost exclusively nociceptive in nature, it has since been determined that certain mechanoreceptors express Na_V1.8 as well,^[250] thus complicating analysis of the results from experiments with Na_V1.8-Cre lines. Such conditional KOs, as well as null models, have been used extensively to examine the contributions of Na_V isoforms to signaling. Nevertheless, using KO techniques to understand Na_V function and pain pathology is rife with opportunity. Questions regarding the involvement of certain Na_V isoforms in nociception remain outstanding—only a single report of a 1.3 null mouse has appeared^[251] and, as yet, KOs of Na_V1.1 and 1.2 have only been used for investigating specific CNS disorders.

Neuropathic pain can be induced in select Na_V KO models; some of these findings undercut previously established conclusions (from knockdown studies) regarding the involvement of Na_V1.3^[251,252] and 1.8^{[247,253,192]} in nociception. Conversely, results from KO studies of Na_V1.6^[246,254] and 1.9^[191,192] suggest that these subtypes contribute to neuropathic pain signaling. In inflammatory pain models, genetic KO experiments have implicated Na_V1.7,^[248,255] 1.8,^{[241,253,256]} and 1.9.^{[242,257,258]} Unfortunately, there is a considerable amount of disparity in the KO literature (Table 7, red). In one example, conflicting results appear between publications describing the same inflammatory pain model and, more broadly, knockout animal. Comparative analysis of knockout and knockdown data also reveals inconsistencies (Table 7, italicized font). While these differences might be explained by the choice of animal subject (most often rat, in case of knockdown, vs. mouse, in KO models), the discrepancies among KO data are more difficult to rationalize. One probable explanation arises from the fact that Na_V KO, for example of Na_V1.1,^[244] 1.8,^[241] or 1.9,^[242] promotes the concomitant upregulation of alternative isoforms. Moreover, single isoform KO can result in subcellular redistribution of the expressed subtypes.^[240,244] Further, the choice of method for generating KO animals can produce mice with different channel expression profiles.^[242,257] To address these problems, Na_V1.6-floxed mice^[254] have been developed; Na_V deletion by AAV-Cre viral induction in adult mice limits the timeframe during which compensatory upregulation can occur. Similarly, Na_V1.8-Cre-diptheria toxin mice^[256] have been genetically engineered to promote selective apoptosis of Na_V1.8-expressing neurons. Studies with these animals are complicated by the significant biochemical and physiological changes accompanying programmed cell death.

Table 7.

Genetic knockouts of Na_V isoforms.

Pain Type	Model	1.3	1.6	1.7	1.8	1.9
Neuropathic	Chronic constriction sciatic nerve	−	−	−	NE [192]	↓[192]
	L5/L6 spinal nerve ligation	NE[251][a]	−	NE[259,247][d] ↓[259][e]	NE[247][f]	−
	Sciatic nerve ligation	−	NE[254][b] ↓[254][c]	−	NE[253]	NE[242,257]
	Spared nerve injury	−	−	−	NE[192]	NE[192,257]
	Chemotherapy	NE[252]	↓[246]	−	NE[252]	NE [252] ↓[191]
	Diabetes	−	−	−	↓[260]	−
Inflammatory	Complete Freund’s Adjuvant	NE[251]	NE[254][b]	↓[248,255][b] NE [259][d]	NE[192] ↓[256][f]	↓[242,257,258] NE [192]
	Carrageenan	−	−	↓[255][b] NE[259][d]	↓[241] NE[192]	↓[242,258] NE [192]
	Formalin	NE[251]	−	↓[255][b],[248] NE[259][d]	NE[192] ↓[256][f]	↓[242,258] NE[192]
	NGF	−	−	↓[255][b]	↓[253]	NE[257]
	PGE2	−	−	−	NE [253]	↓[242,257]
Other	Itch	−	−	↓[248]	−	↓[261]

↓ = reduced pain-associated behavior. NE=no effect on pain-associated behavior. Whole mouse genetic knockouts were used unless otherwise noted. Red indicates disagreement among published data. Italic font indicates disagreement between knockout (Table 7) and knockdown (Table 6) data.

^[a]Ligation of spinal nerve L5 only.

^[b]Na_V1.8-Cre.

^[c]Floxed mice; AAV-Cre induced KO.

^[d]Advillin-Cre (sensory neuron specific).

^[e]Wnt1-Cre (sensory and sympathetic neuron specific).

^[f]Na_V1.8-Cre-diptheria toxin.

3.4. Optogenetics

One of the most interesting developments to arise from the design of conditional Na_V knockouts is the crossbreeding of loxP-STOP-loxP-rhodopsin mice with the Na_V1.8-Cre mouse line. The selective expression of channelrhodopsin ChR2 or archaerhodopsin Arch3 in Na_V1.8-positive neurons enables light-based control of electrical signaling. Using this optogenetic method, activation of Na_V1.8⁺ neurons has been demonstrated to induce nocifensive behavior,^[262] whereas silencing reduces mechanical allodynia and thermal hyperalgesia in models of both inflammatory and neuropathic pain (Figure 8).^[263] These findings have led to the development of an implantable optoelectronic system for the continuous modulation of bladder pain in mice.^[264] The specific targeting of an electrically excitable cell type, rather than a single Na_V isoform, is a paradigm shift in the study and potential treatment of pain pathogenesis.

Figure 8.

Rhodopsin-modulated electrical signaling.^[265] Created with BioRender.com. A) Blue light activation of non-specific cation channel ChR2 depolarizes the membrane potential, causing Na_V opening and action potential firing. B) Yellow (or green) light activation of proton pump Arch3 prevents Na_V opening thereby reducing excitability.^[266]

3.5. Fluorescent and epitope tagging

Recombinant genetic technologies that enable cellular expression of fluorescent protein- and epitope-tagged Na_Vs are essential for the study of Na_V trafficking. Specific challenges, however, have prevented these methods from realizing their full promise. Na_V1.1, 1.2, and 1.6 plasmids, in particular, mutate and recombine extensively when expressed in E. coli,^[267] rendering the synthesis of functionally tagged channels quite challenging. As such, examples of epitope-tagged channels derived from Na_V1.4 and 1.5—subtypes with comparatively stable plasmid DNA—are most prevalent in the literature (Table 8). Unfortunately, the most broadly applicable of these tools, sodium channel protein-based activity reporting construct (SPARC),^[268] which produces an optical readout of channel opening, was found to aggregate intracellularly, with less than 5% of the total protein reaching the membrane.^[269] Early exploration with Na_V1.5-based constructs demonstrated the importance of protein kinase activation^[270] and β-subunit co-expression^[271] for ensuring proper trafficking and membrane localization. Thus, the use of model cell lines for the purpose of screening Na_V-GFP fusion or epitope-tagged constructs may be intrinsically limited. In light of these problems, much of the work reported with epitope-tagged Na_Vs involves protein isolation rather than live cell imaging studies.

Table 8.

Genetically-modified Na_v α-subunits.

Isoform	Modification	Location (E/I)[a]	Model systems	Use
1.2	FLAG	N-terminus (I)	Transgenic mouse, MDCK	Pull-down[278,279]
	HA	N-terminus (I)	tsA201	Purification[280]
	EGFP	N-terminus (I)	MDCK, mouse neuroglial cocultures	Immunohistochemistry[279]
1.4	GFP	L850/S851 (I)	X. oocytes, HEK, rat E18 hippocampal neurons	Sodium channel protein-based activity reporting construct[268,269]
	EVIFQTPL[b]	H599L/F600 (I)	X. oocytes, tsA201, primary muscle	Structural analysis, live-cell imaging[115]
1.5	GFP	N-terminus, C-terminus (I)	X. oocytes, HEK, dog cardiomyocytes	Live-cell imaging[271, 281, 282,286]
	YFP	N-terminus (I)	X. oocytes, HEK	Live-cell imaging,[278,280] FRET[282]
	CFP	N-terminus (I)	HEK, dog cardiomyocytes	Live-cell imaging,[280] FRET[281]
	FLAG	L299/V300 (I)	HEK	Validation[281]
	FLAG	DI S5-S6 (E)	tsA201, rat cardiomyocytes	Immunohistochemistry[283,284]
	HA or 3xHA	DI S5-S6 (E)	X. oocytes, HEK, rat cardiomyocytes	Immunohistochemistry,[285] Chemiluminescence[270]
1.6 (TTX-	myc	N-terminus (I)	MDCK	Pull-down, Immunohistochemistry[279]
R)	ECFP	N-terminus (I)	MDCK, mouse neuroglial cocultures	Immunohistochemistry[279]
	GFP; BAD	C-terminus (I); S1545/K1546 (E)	ND7/23, rat E18 hippocampal neurons	Live-cell imaging, single molecule tracking[272–274]
1.7	TAP	C-terminus (I)	HEK, transgenic mouse	Protein interaction mapping[276]
1.7 (TTX-	GFP	C-terminus (I)	HEK, ND7/23	Live-cell imaging[286]
R)	Venus; BAD	N-terminus (I);M1529/V1530 (E)	Rat P2–4 DRGs	Live-cell imaging[275]
	3xmyc-Halo-3xHA	N-terminus (I)	Rat P2–4 DRGs	Live-cell imaging, single molecule tracking[275]
1.8	Venus	C-terminus (I)	HeLa, rat superior cervical ganglion neurons	Live cell imaging[287]
	EGFP	N-terminus (I)	HeLa, rat superior cervical ganglion neurons	Live cell imaging[287]
1.9	sfGFP	N-terminus (I)	ND7/23, Nav1.8-Cre transgenic mouse	Immunohistochemistry[277]

^[a](E/I) = (Extracellular/Intracellular).

^[b]Antigen for mAb 142.^[22]

Epitope tags: FLAG, DYKDDDDK; HA, YPYDVPDYA; myc, EQKLISEEDL. Acronyms: G/Y/CFP, green/yellow/cyan fluorescent protein; TAP, tandem affinity purification; sfGFP, superfolder GFP. Cell lines: HEK, human embryonic kidney; CHO, Chinese hamster ovary; ND7/23, mouse neuroblastoma; DRG, dorsal root ganglion.

Recent advances have highlighted the design of dually-tagged sodium channels—affixed with both a biotin-affinity domain (BAD) and a fluorescent protein–for the study of Na_V1.6 and Na_V1.7. Incorporation of the BAD domain in the S1—S2 extracellular loop of domain IV enables fluorescent imaging of channels expressed in the membrane following extracellular application of a streptavidin-dye conjugate (e.g., SACF640, SA-568, SA-594); inclusion of the fluorescent protein facilitates monitoring of total protein concentration. This technology is extremely powerful for live cell monitoring of membrane Na_Vs, and has made possible seminal imaging and single-molecule tracking studies of Na_Vs in neurons. Such experiments have begun to unravel the mechanisms of Na_V insertion into and removal from the cellular membrane, as well as the role of channel nanoclusters in electrical signaling.^[272–274] Studies with Na_V1.7 have, for the first time, measured changes in channel trafficking in response to mediators of inflammatory pain.^[275] Undoubtedly, further application of these unique tools will be instrumental for elucidating the regulatory pathways that underlie Na_V physiology in healthy and aberrant cells.

The rapid evolution of CRISPR technology for gene editing has enabled the design of knock-in (KI) mice that express engineered Na_V constructs. In the case of a Na_V1.7-tandem affinity purification tag (TAP) knock-in, transgenic mice were used to identify several hundred proteins with direct channel interactions.^[276] In a separate study involving a superfolder GFPNa_V1.9 KI model, the GFP tag made possible analysis of channel expression patterns in DRGs by immunohistochemistry.^[277] Compared to other genetic methods (e.g., knockdown, knockout), CRISPR-enabled knock-in has several advantages, foremost of which is the ability to study Na_V function in fully intact systems where compensatory changes in Na_V expression and subcellular distribution are no longer complicating factors. Future studies involving CRISPR-engineered Na_V KI animal subjects are expected to give unprecedented insight into Na_V physiology.

4. Opportunities and Challenges

The study of Na_V function and physiology has been empowered with the invention of chemical and biological tools that make possible the selective manipulation of specific channel subtypes. These advances notwithstanding, all technologies have limitations of some form in terms of efficacy, selectivity, and/or utility for in vitro or in vivo studies. The development of new, high-precision reagents for labeling and modulating Na_V subtypes in primary cells and tissue presents tremendous research opportunities. Such pursuits are challenged by the immutable fact that channels rapidly cycle through different conformational states and that most ligands, be they small molecules or peptides, display state and use dependence. For these reasons, discoveries made in dissociated cells using electrophysiology measurements are often confounded once translated to in vivo applications. A better understanding of channel dynamics, state-dependent ligand binding, and how conformational distributions are influenced under different physiological conditions would accelerate efforts to advance new reagents and therapeutic leads that modulate Na_V activity.

Potential applications for radioligand and fluorescent probes that target single Na_V isoforms in live cells are manifold, particularly given the current limitations of Na_V antibodies and the challenges with engineering epitope- and GFP-tagged channels. Other tool compounds that enable isolation of membrane-expressed Na_Vs would facilitate proteomic studies to examine how post-translational modifications influence channel function in healthy and injured neurons. Improvements in CRISPR genome editing and protein engineering are, undoubtedly, going to usher in new research opportunities to study Na_V expression and trafficking in vivo. Similarly, advances in antibody production will deliver potent, subtype-specific reagents that bind to extracellular sites on the channel. All told, the continued expansion of the Na_V ‘toolbox’ will drive research aimed at illuminating the complex physiological mechanisms that underpin electrogenesis and nociception.

Biography

Anna Elleman received her B.S. in chemical biology from the University of California, Berkeley in 2015 while researching in the laboratory of Professor John Kuriyan. She completed her Ph.D. in Chemistry in 2021 under the direction of Professor Justin Du Bois at Stanford University, where she was a Bio-X Stanford Interdisciplinary Graduate and Center for Molecular Analysis and Design Fellow. She has since returned to the University of California, Berkeley, where she works as a postdoctoral researcher with Professors Richard H. Kramer and Stephen Brohawn.

Justin Du Bois received his B.S. degree from the University of California, Berkeley in 1992, where he conducted undergraduate research with Professor Ken Raymond. In 1997 he earned his Ph.D. from the California Institute of Technology under the direction of Professor Erick Carreira. Following a two-year NIH postdoctoral position with Professor Stephen Lippard at MIT, he joined the faculty at Stanford University. He currently holds the title of Henry Dreyfus Professor of Chemistry, and is a professor, by courtesy, in Chemical and Systems Biology and a Bass University Fellow in Undergraduate Education.