Analgesic Effects of Gptx-1, PF-04856264 and CNV1014802 in a Mouse Model of Nav1.7-Mediated Pain

toxins

Article Analgesic Effects of GpTx-1, PF-04856264 and CNV1014802 in a Mouse Model of NaV1.7-Mediated Pain

Jennifer R. Deuis 1,2, Joshua S. Wingerd 1, Zoltan Winter 3, Thomas Durek 1, Zoltan Dekan 1, Silmara R. Sousa 1, Katharina Zimmermann 3, Tali Hoffmann 3, Christian Weidner 3, Mohammed A. Nassar 4, Paul F. Alewood 1, Richard J. Lewis 1 and Irina Vetter 1,2,*

1 Centre for Pain Research, Institute for Molecular Biosciences, The University of Queensland, St Lucia, QLD 4072, Australia; [email protected] (J.R.D.); [email protected] (J.S.W.); [email protected] (T.D.); [email protected] (Z.D.); [email protected] (S.R.S.); [email protected] (P.F.A.); [email protected] (R.J.L.) 2 School of Pharmacy, The University of Queensland, Woolloongabba, QLD 4102, Australia 3 Department of Physiology and Pathophysiology and Department of Anaesthesiology, Friedrich-Alexander University Erlangen-Nuremberg, 91054 Erlangen, Germany; [email protected] (Z.W.); [email protected] (K.Z.); [email protected] (T.H.); [email protected] (C.W.) 4 Department of Biomedical Science, University of Sheffield, Sheffield S10 2TN, UK; m.nassar@sheffield.ac.uk * Correspondence: [email protected]; Tel.: +61-7-3346-2660

Academic Editors: Rong Chen and Yingliang Wu Received: 25 January 2016; Accepted: 10 March 2016; Published: 17 March 2016

Abstract: Loss-of-function mutations of NaV1.7 lead to congenital insensitivity to pain, a rare condition resulting in individuals who are otherwise normal except for the inability to sense pain, making pharmacological inhibition of NaV1.7 a promising therapeutic strategy for the treatment of pain. We characterized a novel mouse model of NaV1.7-mediated pain based on intraplantar injection of the scorpion toxin OD1, which is suitable for rapid in vivo profiling of NaV1.7 inhibitors. Intraplantar injection of OD1 caused spontaneous pain behaviors, which were reversed by co-injection ´{´ with NaV1.7 inhibitors and significantly reduced in NaV1.7 mice. To validate the use of the model for profiling NaV1.7 inhibitors, we determined the NaV selectivity and tested the efficacy of the reported NaV1.7 inhibitors GpTx-1, PF-04856264 and CNV1014802 (raxatrigine). GpTx-1 selectively inhibited NaV1.7 and was effective when co-administered with OD1, but lacked efficacy when delivered systemically. PF-04856264 state-dependently and selectively inhibited NaV1.7 and significantly reduced OD1-induced spontaneous pain when delivered locally and systemically. CNV1014802 state-dependently, but non-selectively, inhibited NaV channels and was only effective in the OD1 model when delivered systemically. Our novel model of NaV1.7-mediated pain based on intraplantar injection of OD1 is thus suitable for the rapid in vivo characterization of the analgesic efficacy of NaV1.7 inhibitors.

Keywords: OD1; NaV1.7; PF-04856264; CNV1014802; raxatrigine; GpTx-1; pain

1. Introduction

Nine voltage-gated sodium channel subtypes have been described to date (NaV1.1–NaV1.9), several of which are implicated as causative contributors to pain. Of particular interest is NaV1.7, as loss-of-function mutations in humans lead to congenital insensitivity to pain, a rare condition that results in the inability to sense pain [1,2]. This makes pharmacological inhibition of NaV1.7 an exciting therapeutic strategy for the treatment of a wide range of pain types, including inherited

Toxins 2016, 8, 78; doi:10.3390/toxins8030078 www.mdpi.com/journal/toxins Toxins 2016, 8, 78 2 of 19 erythromelalgia and paroxysmal extreme pain disorder, two conditions whose pathophysiology arises from NaV1.7 gain-of-function mutations. Accordingly, the development of NaV1.7 inhibitors is being actively pursued by the pharmaceutical industry [3]. NaV1.7 selectivity is key to developing more effective analgesics as activity at major off-targets, including the skeletal muscle isoform NaV1.4, the cardiac isoform NaV1.5, as well as neuronal isoforms NaV1.1, NaV1.2 and NaV1.6, is likely to impact the therapeutic window and cause dose-limiting adverse effects [4–6]. However, achieving sufficient selectivity for NaV1.7 over the other NaV isoforms has proven challenging due to the high sequence homology within the NaV family [7]. Despite this challenge, several small molecules with reported NaV1.7 selectivity are currently undergoing clinical development, including PF-05089771 (Pfizer) and CNV1014802 (Convergence Pharmaceuticals) [8]. In addition to small molecules, venom-derived peptides are proving to be an alternate source for NaV1.7 inhibitors, with several spider peptides, including GpTx-1, Huwentoxin-IV and ProTx-II, reported to have some NaV1.7 selectivity [9–11]. Evaluating the analgesic efficacy of this increasing number of NaV1.7-selective compounds in relevant pre-clinical models of pain is a crucial translational step to clinical success. However, the pathophysiological basis of most commonly used animal models is multi-factorial, making selection of the best disease model for rapid efficacy profiling of NaV1.7 inhibitors challenging. We thus sought to characterize a novel animal model of NaV1.7-mediated pain that is suitable for the rapid in vivo profiling of the analgesic efficacy of NaV1.7 inhibitors. As gain-of-function mutations of NaV1.7 in humans are associated with a range of painful syndromes [12,13], we hypothesized that intraplantar administration of the NaV1.7 activator OD1 could be used as a pharmacological tool to establish a NaV1.7-mediated mouse model of pain. OD1 is a scorpion toxin isolated from the venom of the Iranian yellow scorpion (Odontobuthus doriae) that potently enhances the activity of NaV1.7 by inhibiting channel fast inactivation and increasing peak current with an EC50 between 4 and 8 nM [14,15]. OD1 also has activity at NaV1.4 (EC50 = 10 nM) and NaV1.6 (EC50 = 30–47 nM); however, it is 1000-fold selective (EC50 > 1 µM) over NaV1.1, NaV1.2, NaV1.3, NaV1.5 and NaV1.8 [14,15]. Consistent with its in vitro pharmacological activity, intraplantar administration of OD1 elicited pain behaviors, including licking and flinching of the hind paw, and we have previously used this model to assess the analgesic effects of ProTx-III [16]. The aim of this study was to characterize the OD1 mouse model of pain and to validate the use of this model by testing the efficacy of several reported selective NaV1.7 inhibitors, including the spider peptide GpTx-1, PF-04856264 (as the structure of clinical candidate PF-05089771 is not publicly available) [17] and the clinical candidate CNV1014802 (raxatrigine). As the full pharmacological activity of these inhibitors is not reported, we determined their selectivity at NaV1.1–NaV1.8 and the mode of action at NaV1.7 using functional assays.

2. Results

2.1. OD1 Has Mixed α/β-Scorpion Toxin Activity at NaV1.7 at High Concentrations

OD1 was previously described as an α-scorpion toxin that enhances peak NaV1.7 current expressed in oocytes with a minor effect on the voltage dependence of channel activation or inactivation [14]. However, this effect is difficult to reconcile with the induction of spontaneous pain behavior in vivo, and we thus sought to evaluate the electrophysiological effects at NaV1.7 at concentrations that elicit robust spontaneous pain behaviors in mice (300 nM, see below). At a concentration of 300 nM, OD1 enhanced peak inward current and delayed inactivation, resulting in persistent current (Figure1A,B), consistent with previously observed effects [ 14]. OD1 (300 nM) enhanced peak current with a leftward shift to more hyperpolarized potentials (Figure1C). OD1 (300 nM) resulted in a small, non-significant shift in the V50 of inactivation for hNav1.7 peak current, while a significant shift in V50 of inactivation was observed for the late current 10 ms after depolarization (Figure1D; V 50 of inactivation: control, ´58.54 ˘ 0.23 mV; OD1peak, ´57.78 ˘ 0.37; OD1late, ´52.71 ˘ 0.49). Although the V50 of inactivation is statistically unchanged at peak current, Toxins 2016, 8, 78 3 of 19 a significant fraction of hNav1.7 channels remain in the open state in the presence of OD1 (300 nM), Toxins 2016, 8, 78 3 of 18 even at conditioning potentials that completely inactivate untreated channels, as can be seen by the voltasignificantlyge of inactivation increased slope(Figure of 1D; the voltagekh control of inactivation= 6.07 ± 0.20 (FiguremV; kh1 OD1D; kpeakh control = 10.54 = ± 6.07 0.33;˘ 0.20kh OD1 mV;latek =h 11.45OD1peak ± 0.44;= 10.54 p < 0.05).˘ 0.33; Consistentkh OD1late with= 11.45 previous˘ 0.44; reportsp < 0.05). of mixed Consistent α/β toxin with previouspharmacology reports on of Na mixedV1.4 α β and/ NatoxinV1.6, pharmacology a significant hyperpolarizing on NaV1.4 and Na shiftV1.6, V50 a of significant activation hyperpolarizing of −12 mV at Na shiftV1.7 Vwas50 of observed activation in theof ´ presence12 mV at of Na OD1V1.7 (300 was nM) observed compared in the to presence control ofconditions OD1 (300 (Figure nM) compared 1E; V50 of to activation: control conditions control, −(Figure22.47 ±1 E; 0.47 V 50; OD1,of activation: −34.50 ± control, 0.58; p ´< 22.47 0.05).˘ OD10.47; (300 OD1, nM)´34.50 also˘ delayed0.58; p < fast 0.05). inactivation OD1 (300 nM) at more also depolarizeddelayed fast membrane inactivation potentials at more depolarized (Figure 1F). membrane potentials (Figure1F).

Figure 1. Activity of OD1 in CHO cellscells heterologouslyheterologously expressingexpressing hNahNaVV1.7 assessed by automated patch clamping. Representative trace trace of sodium currents currents ( (A)) before before and and ( B)) after addition of of 300 nM OD1 elicited by depolarizingdepolarizing steps between ´−100100 and and +70 mV mV in in 10 10-mV-mV increments. The The red trace highlights the the depolarizing depolarizing step step to 0 to mV. 0 OD1 mV. enhanced OD1 enhanced peak inward peak current inward and current delayed and inactivation, delayed inactivationresulting in,persistent resulting in current. persistent (C) Current-voltagecurrent. (C) Current (IV)-voltage relationship (IV) relationship before and afterbefore the and addition after the of additionOD1 (300 of nM). OD1 OD1 (300 enhancednM). OD1 peak enhanced current peak with current a leftward with shifta leftward to more shift hyperpolarized to more hyperpolarized potentials. potentials(D) The voltage. (D) The dependence voltage dependence of fast inactivation. of fast inactivation. OD1 signiﬁcantly OD1 significantly shifted the voltage shifted dependence the voltage

dependenceof fast inactivation of fast for inactivation the late current for the 10 ms late after current depolarization 10 ms after (V 50 depolarization: control, ´58.54 (V˘50: 0.23control mV;, −OD158.54peak ± ,0.23´57.78 mV;˘ OD10.37;peak OD1, −57.78late, ´ ±52.71 0.37; ˘OD10.49).late, (−E52.71) Voltage ± 0.49) dependence. (E) Voltage of activation.dependence OD1 of activation. shifted the

OD1voltage shifted dependence the voltage of activation dependence to a moreof activation hyperpolarized to a more potential hyperpolarized (V50: control, potential´22.47 (V˘500.47;: control OD1, −(30022.47 nM), ± 0.47´34.50; OD1˘ (3000.58). nM), (F −)34.50 Time ± of 0.58 decay). (F) ( τTime) for of inactivating decay (τ) for current inactivating plotted currentversus plottedtest potential. versus testOD1 potential. (300 nM) delaysOD1 (300 fast inactivation nM) delays at fast more inactivation depolarized at membrane more depolarized potentials. membrane * p < 0.001 potentials. compared *to p the< 0.001 control. compared Data are to presentedthe control. as Data the meanare presented˘ SEM, nas= the 7. mean ± SEM, n = 7.

2.22.2.. OD1 Causes Spontaneous Action Potential Firing in A A-- and C-FibersC-Fibers

To assess the effect ofof pharmacologicalpharmacological NaV1.71.7 activation activation on on A A-- and C C-fibers,-fibers, we tested OD1 using the mouse skin-saphenousskin-saphenous nerve preparation. Consistent Consistent with with the crucial role of Na NaV1.7 in regulating excitability, the application of OD1 to the receptive fieldsfields of peripheral sensory neurons led to to spontaneous spontaneous firing firing of of action action potentials potentials in in some some fibers, fibers, with with 57% 57% of of A A-fibers-fibers tested tested (Figure (Figure 22AA,B;,B; ccontrolontrol 0 ˘± 00 and and OD1 OD1 ( (3030 nM) 13 ˘± 77 action action potentials/2 potentials/2 min; min; nn == 7) 7) and and 29% 29% of of C C-fibers-fibers tested tested firing firing spontaneously (Figure2 2CC,D;,D; control control 1 1 ˘± 0.6 and OD1OD1 (30(30 nM)nM) 88 ˘± 5 action potentials/2potentials/2 min; min; n = 7) in the presence of OD1.

Toxins 2016, 8, 78 4 of 19 Toxins 2016, 8, 78 4 of 18

A B

n 60

m 50 25 1 min

s 20 l 40

a )

n H 15

30 (

r 10 P 20

i 10 5

A 0 0

Control OD1 Control OD1 C D

n 40

m 4

1 min

)

n 20 H

(

q 2

n 10 1

A 0 0

Control OD1 Control OD1

FigureFigure 2. Effects 2. Effects of Naof VNa1.7V1.7 activation activation byby OD1OD1 on A A-- and and C C-fibers-fibers using using the the mouse mouse skin skin-saphenous-saphenous nervenerve preparation. preparation. (A) OD1(A) OD1 caused caused spontaneous spontaneous firing firing of actionof action potentials potentials in in 57% 57% of of the the A-fibers A-fibers tested (actiontested potentials/2 (action potentials/2 min: control, min: control 0 ˘ 0;, 0 OD1± 0; OD1 (30 (30 nM), nM), 13 13˘ ±7; 7; nn = 7 7).). (B ()B A)ction Action potentials potentials plotted plotted as a function of instantaneous firing frequency from an A-fiber before and after perfusion of OD1. as a function of instantaneous firing frequency from an A-fiber before and after perfusion of OD1. Each point represents a single action potential, with the position on the y-axis indicating the frequency Each point represents a single action potential, with the position on the y-axis indicating the frequency relative to the preceding action potential. (C) OD1 caused spontaneous firing of action potentials in C relative29% to of the the preceding C-fibers tested action (action potential. potentials/2 ( ) OD1 min: causedcontrol, spontaneous 1 ± 0.6; OD1 (30firing nM), of 8 action ± 5; n potentials= 7). in 29%(D) of Action the C-fibers potentials tested plotted (action as a function potentials/2 of instantaneous min: control, firing frequency 1 ˘ 0.6; from OD1 a (30 C-fiber nM), before 8 ˘ 5;andn = 7). (D) Actionafter perfusion potentials of OD1. plotted Each as point a function represents of instantaneous a single action firingpotential, frequency with the fromposition a C-fiber on the beforey-axis and afterindicating perfusion the of OD1.frequency Each relative point to represents the preceding a single action action potential. potential, with the position on the y-axis indicating the frequency relative to the preceding action potential. 2.3. Intraplantar Injection of OD1 Causes Spontaneous Pain Behavior in Mice Mediated Through NaV1.7

2.3. IntraplantarTo assess Injection the effect of OD1 of OD1 Causes on Spontaneous pain behaviors Pain in Behavior mice, we in chose Mice Mediatedthe intraplantar Through route NaV of1.7 administration, as this route delivers the peptide directly to the terminals of peripheral sensory To assess the effect of OD1 on pain behaviors in mice, we chose the intraplantar route of neurons in the skin, allowing simple quantification of unilateral pain behaviors while avoiding administration, as this route delivers the peptide directly to the terminals of peripheral sensory neurons systemic adverse effects. Intraplantar injection of the NaV1.7 activator OD1 (10–300 nM) led to the in thedevelopment skin, allowing of spontaneous simple quantification pain behaviors of unilateral in a concentration pain behaviors-dependent while manner, avoiding as evidenced systemic adverseby effects.licking, Intraplantar flinching, lifting injection and ofshaking the Na of Vthe1.7 injected activator hind OD1 paw, (10–300consistent nM) with led the tospontaneous the development action of spontaneouspotential painfiring behaviors seen in A- in and a concentration-dependent C-fibers (Figure 3A). These manner, spontaneous as evidenced pain behaviors by licking, developed flinching, liftingrapidly, and shaking occurring of immediately the injected after hind injection, paw, consistent and persisted with for the up spontaneous to 40 min after action injection potential of the firing seenhighest in A- and concentration C-fibers (Figure (300 nM).3A). Intraplantar These spontaneous injection of pain phosphate behaviors-buffered developed saline/0.1% rapidly, BSA occurringalone immediatelycaused no after spontaneous injection, pain and behaviors persisted. No for significant up to 40 changes min after in injectionmechanical of (paw the highestwithdrawal concentration force: (300 nM).OD1 (300 Intraplantar nM), 2.5 ± injection0.5 g; cont ofrol, phosphate-buffered 2.9 ± 0.5 g; n = 4, p > 0.05) saline/0.1% or thermal BSAthresholds alone (time caused to withdrawal: no spontaneous OD1 (300 nM), 8.0 ± 2 s; control, 11.0 ± 1 s; n = 4, p > 0.05) were detected 45 min after injection of OD1 pain behaviors. No significant changes in mechanical (paw withdrawal force: OD1 (300 nM), (300 nM), the time by which spontaneous pain behaviors had resolved. As OD1 at a concentration of 2.5 ˘3000.5 nM g; was control, found 2.9to elicit˘ 0.5 robust g; n responses,= 4, p > this 0.05) concentration or thermal was thresholds used for further (time behavioral to withdrawal: studies. OD1 (300 nM),Since 8.0 ˘ OD12 s; has control, activity 11.0 at Na˘V11.6 s; atn higher= 4, p concentrations> 0.05) were detected(EC50 = 30– 4547 minnM) after[15] and injection activation of OD1 (300 nM),of Na theV1.6 timecan also by cause which spontaneous spontaneous pain pain behaviors behaviors [18], hadwe next resolved. assessed As the OD1 relative at a contribution concentration of 300 nMof Na wasV1.7 found and Na toV1.6 elicit to robustspontaneous responses, pain behaviors this concentration induced by OD1 was used(Figure for 3B). further Co-administration behavioral studies.of Sincetetrodotoxin OD1 has(TTX activity) with at OD1 Na V completely1.6 at higher reversed concentrations spontaneous (EC p50ain= 30–47behaviors nM), confirming[15] and activation the involvement of only TTX-sensitive NaV subtypes (spontaneous pain behaviors/10 min: control, 105 ± 6; of NaV1.6 can also cause spontaneous pain behaviors [18], we next assessed the relative contribution TTX (1 μM), 3 ± 2); p < 0.05). Co-administration with the NaV1.6 inhibitor GIIIA [19] reduced spontaneous of NaV1.7 and NaV1.6 to spontaneous pain behaviors induced by OD1 (Figure3B). Co-administration of tetrodotoxinpain behaviors (TTX) by 33%, with indicating OD1 completely partial involvement reversed of spontaneous NaV1.6 (spontaneous pain behaviors, pain behaviors confirming/10 min: the control, 105 ± 6; GIIIA (10 μM), 71 ± 8); p < 0.05). In contrast, co-administration with the NaV1.7 involvement of only TTX-sensitive NaV subtypes (spontaneous pain behaviors/10 min: control, inhibitor GpTx-1 (see the full NaV selectivity in Table 1) completely reversed spontaneous pain 105 ˘ 6; TTX (1 µM), 3 ˘ 2); p < 0.05). Co-administration with the Na 1.6 inhibitor GIIIA [19] reduced behaviors, confirming a major role for NaV1.7 in mediating spontaneousV pain behaviors induced by spontaneous pain behaviors by 33%, indicating partial involvement of NaV1.6 (spontaneous pain behaviors/10 min: control, 105 ˘ 6; GIIIA (10 µM), 71 ˘ 8); p < 0.05). In contrast, co-administration with the NaV1.7 inhibitor GpTx-1 (see the full NaV selectivity in Table1) completely reversed spontaneous Toxins 2016, 8, 78 5 of 18

OD1 (spontaneous pain behaviors/10 min: control, 105 ± 6; GpTx-1 (1 μM), 6 ± 5); p < 0.05). A major contribution of NaV1.7 to OD1-induced pain was also confirmed using NaV1.7−/− mice, which displayed significantly reduced spontaneous pain behaviors compared to Cre and loxP littermate controls (Figure 3C; spontaneous pain behaviors/10 min: control, 129 ± 8; NaV1.7−/−, 41 ± 11; p < 0.05).

Table 1. IC50 values (in μM) at hNaV1.1–1.8 obtained from the Fluorometric Imaging Plate Reader (FLIPR) membrane potential assay.

NaV Subtype GpTx-1 PF-04856264 CNV1014802 Toxins 2016, 8, 78 5 of 19 1.1 6 ± 2 247 ± 27 22 ± 7 1.2 5 ± 2 20 ± 11 13 ± 6 pain behaviors, confirming a1.3 major role for22 ± Na 6 V1.7 in mediating>300 spontaneous23 ± 13 pain behaviors induced by OD1 (spontaneous pain1.4 behaviors/10 326 ± min: 170 control,>300 105 ˘ 6; GpTx-110 ± 5 (1 µM), 6 ˘ 5); p < 0.05). A ´{´ major contribution of NaV1.71.5 to OD1-induced140 ± 40 pain was>300 also confirmed84 ± using 34 NaV1.7 mice, which displayed significantly reduced1.6 spontaneous17 ± 9 pain behaviors309 ± 91 compared15 ± 1 to Cre and loxP littermate ´{´ controls (Figure3C; spontaneous1.7 pain0.58 behaviors/10 ± 0.1 0.30 min: ± 0.07 control, 32 129 ± 14˘ 8; NaV1.7 , 41 ˘ 11; p < 0.05). 1.8 68 ± 32 >300 6 ± 0.2

A B C

120

i 140 300 nM i

60 m

100 nM 100 120

/ s 50 * 30 nM /

s s 80 100

u 40 r 10 nM u

o u

i 80

i v 60

30 v

6 7

h * . * .

. 60 *

1 1

20 s

40 V V

a a B 40

10 a

N N

a 20

a 0 a 20

P 0 0

0 10 20 30 40 o

T .

Time (min) G

Figure 3. Mouse model of NaV1.7-mediated pain based on intraplantar injection of OD1. Figure 3. Mouse model of NaV1.7-mediated pain based on intraplantar injection of OD1. (A) Intraplantar (injectionA) Intraplantar of OD1 injection (10–300 nM)of OD1 caused (10– spontaneous300 nM) caused pain spontaneous behaviors, as pain evidenced behaviors by, licking, as evidenced flinching, by licking,lifting and flinching, shaking lifting of the and injected shaking hind of paw. the Intraplantarinjected hind injection paw. Intraplantar of phosphate-buffered injection of saline/0.1%phosphate- bufferedBSA alone saline/0.1% caused no BSA spontaneous alone caused pain no behaviors. spontaneous (B) pain Intraplantar behaviors injection. (B) Intraplantar of TTX (1 injectionµM) or the of TTX (1 μM) or the NaV1.7 inhibitor GpTx-1 (1 μM) completely reversed spontaneous pain behaviors, NaV1.7 inhibitor GpTx-1 (1 µM) completely reversed spontaneous pain behaviors, while the NaV1.6 while the NaV1.6 inhibitor GIIIA (10 μM) only had a partial´{´ effect. (C) NaV1.7−/− mice displayed inhibitor GIIIA (10 µM) only had a partial effect. (C) NaV1.7 mice displayed significantly reduced significantlyspontaneous painreduced behaviors spontaneous compared pain to Crebehaviors and loxP compared littermate to controls. Cre and Statistical loxP littermate significance controls. was Statisticaldetermined significance using the twas-test determined or one-way using ANOVA the t with-test Dunnett’sor one-way post-test; ANOVA * withp < 0.05 Dunnett’s compared post to-test the; *control. p < 0.05 Data compared are presented to the control. as the meanData are˘ SEM, presentedn = 3–9 as perthe group.mean ± SEM, n = 3–9 per group.

2.4. GpTx-1 is a Selective NaV1.7 Inhibitor that Reverses OD1-Induced Pain Behaviors

Table 1. IC50 values (in µM) at hNaV1.1–1.8 obtained from the Fluorometric Imaging Plate Reader GpTx-1 was recently reported as a NaV1.7-selective inhibitor isolated from the venom of the (FLIPR) membrane potential assay. tarantula Grammostola porteri (identical in sequence to GTx1-15 isolated from Grammostola rosea [20]); however, the mechanism of action of GpTx-1 and selectivity across all NaV subtypes has not been NaV Subtype GpTx-1 PF-04856264 CNV1014802 reported. Using a Fluorometric Imaging Plate Reader (FLIPR) membrane potential assay, GpTx-1 was 1.1 6 ˘ 2 247 ˘ 27 22 ˘ 7 found to inhibit NaV1.7 with 100-fold selectivity over NaV1.4, NaV1.5 and NaV1.8, consistent with the 1.2 5 ˘ 2 20 ˘ 11 13 ˘ 6 selectivity profile 1.3 previously reported 22 ˘ from6 electrophysiological >300 studies [9,20]. 23 ˘ 13The GpTx-1 rank order of potency (pIC1.450 ± SEM) was 326hNa˘V1.7170 (6.27 ± 0.1) > hNa >300V1.2 (5.63 ± 0.3) > 10 hNa˘ 5 V1.1 (5.30 ± 0.1) > hNaV1.6 (4.88 ± 0.2)1.5 > hNaV1.3 (4.69 140 ± 0.1)˘ 40 > hNaV1.8 (4.49 >300 ± 0.5) > hNaV1.5 84(3.89˘ 34 ± 0.1) > hNaV1.4 1.6 17 ˘ 9 309 ˘ 91 15 ˘ 1 (3.61 ± 0.2) (Figure 4A,B; see also Table 1 for IC50 values and Figure S1 for a sample trace). 1.7 0.58 ˘ 0.1 0.30 ˘ 0.07 32 ˘ 14 GpTx-1 inhibited1.8 the peak current 68 ˘ 32 of hNaV1.7 without >300 significantly 6 ˘shifting0.2 the voltage dependence of activation (Figure 4C,D; V50: control, −20.52 ± 0.4 mV; GpTx-1 (10 nM), −21.4 ± 0.5 mV); however, a shift in the voltage dependence of fast inactivation to more hyperpolarized potentials was 2.4. GpTx-1 is a Selective NaV1.7 Inhibitor that Reverses OD1-Induced Pain Behaviors observed (Figure 4D; V50 for control −58.66 ± 0.3 mV and GpTx-1 (10 nM) −64.58 ± 0.5 mV). GpTx-1 GpTx-1 was recently reported as a NaV1.7-selective inhibitor isolated from the venom of the tarantula Grammostola porteri (identical in sequence to GTx1-15 isolated from Grammostola rosea [20]); however, the mechanism of action of GpTx-1 and selectivity across all NaV subtypes has not been reported. Using a Fluorometric Imaging Plate Reader (FLIPR) membrane potential assay, GpTx-1 was found to inhibit NaV1.7 with 100-fold selectivity over NaV1.4, NaV1.5 and NaV1.8, consistent with the selectivity proﬁle previously reported from electrophysiological studies [9,20]. The GpTx-1 rank order of potency (pIC50 ˘ SEM) was hNaV1.7 (6.27 ˘ 0.1) > hNaV1.2 (5.63 ˘ 0.3) > hNaV1.1 (5.30 ˘ 0.1) > hNaV1.6 (4.88 ˘ 0.2) > hNaV1.3 (4.69 ˘ 0.1) > hNaV1.8 (4.49 ˘ 0.5) > hNaV1.5 (3.89 ˘ 0.1) > hNaV1.4 (3.61 ˘ 0.2) (Figure4A,B; see also Table1 for IC 50 values and Figure S1 for a sample trace). Toxins 2016, 8, 78 6 of 18

caused full inhibition of peak current at 1 µM, which was partially reversed after a strong depolarizing pulse to +200 mV for 50 ms (Figure 4E), an effect often seen with gating modifier toxins [21]. GpTx-1 displayed minimal state dependence, with an IC50 of 8 nM in the open/inactivated state and 13 nM in the closed/resting state (Figure 4F). The potency of GpTx-1 was not significantly affected by the presence of OD1 (300 nM) in the open/inactivated state (IC50 3 nM) or in the closed/resting state (IC50 8 nM) (see Figure S2 for a sample trace). Intraplantar injection of GpTx-1 co-administered with OD1 potently and concentration- dependently reduced spontaneous pain behaviors, causing almost complete reversal at the highest concentration (Figure 4G; spontaneous pain behaviors/10 min: control, 105 ± 6; GpTx-1 (1 μM), 6 ± 5; GpTx-1 (300 nM), 17 ± 4; GpTx-1 (100 nM), 34 ± 2; GpTx-1 (30 nM), 68 ± 9; p < 0.05). However, intraperitoneal injection of GpTx-1 at the highest tolerated dose (0.1 mg/kg) had no significant effect on spontaneous pain behaviors (Figure 4H). At a dose of 0.3 mg/kg, GpTx-1 causes motor deficits, as evidenced by a significant increase in the ataxia index (Figure 5A; ataxia index: control, 2.7 ± 0.4; Toxins 2016GpTx, 8-,1 78 (0.3 mg/kg), 9.7 ± 3.4; p < 0.05). Therefore, it was not possible to assess the effect of GpTx-1 at6 of 19 systemic doses above 0.1 mg/kg.

A B C

3 e

s 100 1.1 0.2

o 1.2

p 0.0

s 75

4 e 1.3 -0.2

0 1.4

o 5 50

I -0.4

s I 1.5

p 5 a 25 1.6 -0.6

m r -0.8 o 1.7 0 Control 6 N 1.8 -1.0 10 nM GpTx-1

1 2 3 4 5 6 7 8 -80 -60 -40 -20 0 20 40 60 80 ...... -8 -7 -6 -5 -4 -3 1 1 1 1 1 1 1 1 mV Nav subtype Log [GpTx-1] (M) D E Control 10 nM GpTx-1 1.0 1.0

0.8 0.8

x 0.6

I a 0.6

G / 0.4 0.4

G 0.2 0.2 0.0 0.0

-120 -100 -80 -60 -40 -20 0 20 mV F G H n.s 120 120

i 1.2 –100 mV i

m m 100 100

1.0 – 55 mV

/ –100 mV + OD1 / * 80 0.8 80 s

o – 55 mV + OD1

I 0.6

i I i 60 60

0.4 a

h h 40 * 40

e 0.2 e

i 0.0 i 20 20

a a *

P 0 0

o k

M -10 -9 -8 -7 M M

t µ

n n n t

0 0

o m

3 Log [GpTx-1] (M) 0 0

1 3

. 0

Figure 4. NaV selectivity and in vivo effects of the spider peptide GpTx-1. Activity of GpTx-1 was Figure 4. NaV selectivity and in vivo effects of the spider peptide GpTx-1. Activity of GpTx-1 was assessed in HEK cells heterologously expressing NaV1.1–NaV1.8 using FLIPR membrane potential assessed in HEK cells heterologously expressing NaV1.1–NaV1.8 using FLIPR membrane potential assays. (A) GpTx-1 selectively inhibited NaV1.7, with the following pIC50’s ranked in order of potency assays. (A) GpTx-1 selectively inhibited Na 1.7, with the following pIC ’s ranked in order of potency (hNaV1.7 (6.27 ± 0.1) > hNaV1.2 (5.63 ± 0.3) > hNaV V1.1 (5.30 ± 0.1) > hNaV1.6 (4.8850 ± 0.2) > hNaV1.3 (4.69 ± 0.1) ˘ ˘ ˘ ˘ (hNaV> 1.7hNa (6.27V1.8 (4.490.1) ± 0.5) > hNa > hNaV1.2V1.5 (5.63 (3.89 ±0.3) 0.1) >> hNahNaV1.41.1 (3.61 (5.30 ± 0.2).0.1) Data > hNa are Vpresented1.6 (4.88 as the0.2) mean > hNa ± V1.3 (4.69 SEM˘ 0.1) from > hNa threeV 1.8independent (4.49 ˘ 0.5) experiments > hNaV.1.5 (B) (3.89Representative˘ 0.1) > hNaconcentrationV1.4 (3.61 response˘ 0.2). curves Data of are GpTx presented-1

as theat mean NaV1.1˘–NaSEMV1.8. from (C–F) three Activity independent of GpTx-1 in experiments. CHO cells heterologously (B) Representative expressing concentration hNaV1.7 assessed response curves of GpTx-1 at Na 1.1–Na 1.8. (C–F) Activity of GpTx-1 in CHO cells heterologously expressing V V hNaV1.7 assessed by automated patch clamping. (C) Current-voltage (IV) relationship before and after the addition GpTx-1. (D) Voltage dependence of activation and fast inactivation before and after

the addition of GpTx-1. GpTx-1 (10 nM) had no signiﬁcant effect on the voltage of activation (V50: control, ´20.52 ˘ 0.4 mV; GpTx-1 (10 nM), ´21.4 ˘ 0.5 mV), but shifted the voltage dependence of

fast inactivation to more hyperpolarized potentials (V50: control, ´58.66 ˘ 0.3 mV; GpTx-1 (10 nM), ´64.58 ˘ 0.5 mV). (E) Trace of sodium currents before (black) and after addition of 1 µM GpTx-1 (red). GpTx-1 causes full inhibition of the peak current that was partially reversed (highlighted by the points) by a strong depolarization pulse (+200 mV for 50 ms). (F) Concentration response curve of GpTx-1 elicited by a 20-ms pulse to 0 mV from a holding potential of ´100 mV or an 8-s conditioning voltage step of ´55 mV with and without OD1 (300 nM). GpTx-1 had no signiﬁcant preference for the

open/inactivated state (IC50 8 nM) compared to the closed/resting state (IC50 13 nM). The presence of OD1 (300 nM) did not significantly affect the potency of GpTx-1 in the open/inactivated state (IC50 3 nM) or in the closed/resting state (IC50 8 nM). Data are presented as the mean ˘ SEM, n = 10–14. (G) Intraplantar administration of GpTx-1 (30 nM, 100 nM, 300 nM, 1 µM) concentration-dependently reversed spontaneous pain behaviors in mice evoked by OD1. (H) Intraperitoneal administration of GpTx-1 (0.1 mg/kg) had no significant (n.s.; p > 0.05) effect on spontaneous pain behaviors in mice induced by OD1. Statistical significance was determined using the t-test or one-way ANOVA with Dunnett’s post-test; * p < 0.05 compared to the vehicle control. Data are presented as the mean ˘ SEM, n = 3–9 per group. Toxins 2016, 8, 78 7 of 19

GpTx-1 inhibited the peak current of hNaV1.7 without significantly shifting the voltage dependence of activation (Figure4C,D; V 50: control, ´20.52 ˘ 0.4 mV; GpTx-1 (10 nM), ´21.4 ˘ 0.5 mV); however, a shift in the voltage dependence of fast inactivation to more hyperpolarized potentialsToxins was 2016 observed, 8, 78 (Figure4D; V 50 for control ´58.66 ˘ 0.3 mV and GpTx-17 of 18 (10 nM) ´64.58 ˘ 0.5 mV). GpTx-1 caused full inhibition of peak current at 1 µM, which was partially reversed by automated patch clamping. (C) Current-voltage (IV) relationship before and after the addition after a strongGpTx depolarizing-1. (D) Voltage pulse dependence to +200 of activation mV for and 50 fast ms inactivation (Figure before4E), anandeffect after the often addition seen of with gating modifier toxinsGpTx [21-1. ].GpTx GpTx-1-1 (10 nM) displayed had no significant minimal effect on state the voltage dependence, of activation (V with50: control an, IC −20.5250 of ± 8 nM in the open/inactivated0.4 mV; state GpTx and-1 (10 13 nM), nM −21.4 in ± the 0.5 mV), closed/resting but shifted the statevoltage (Figure dependence4F). of Thefast inactivation potency to of GpTx-1 was more hyperpolarized potentials (V50: control, −58.66 ± 0.3 mV; GpTx-1 (10 nM), −64.58 ± 0.5 mV). not significantly affected by the presence of OD1 (300 nM) in the open/inactivated state (IC 3 nM) or (E) Trace of sodium currents before (black) and after addition of 1 µM GpTx-1 (red). GpTx-1 causes 50 in the closed/restingfull inhibition state of the (IC peak50 8 current nM) that (see was Figure partially S2 reversed for a sample(highlighted trace). by the points) by a strong Intraplantardepolarization injection pulse of GpTx-1 (+200 mVco-administered for 50 ms). (F) Concentration with OD1 response potently curve of and GpTx concentration-dependently-1 elicited by a reduced spontaneous20-ms pulse pain to 0 mV behaviors, from a holding causing potential almost of −100 complete mV or an 8- reversals conditioning at the voltage highest step of concentration −55 mV with and without OD1 (300 nM). GpTx-1 had no significant preference for the µ (Figure4G; spontaneousopen/inactivated pain state (IC behaviors/1050 8 nM) compared min: to the control,closed/resting 105 state˘ (IC6;50 13 GpTx-1 nM). The (1 presenceM), of 6 ˘ 5; GpTx-1 (300 nM), 17 ˘OD14; GpTx-1(300 nM) did (100 not nM), significantly 34 ˘ affect2; GpTx-1 the potency (30 of nM), GpTx 68-1 in˘ the9; open/inactivatedp < 0.05). However, state (IC50 3 intraperitoneal injection of GpTx-1nM) or in at the the closed/resting highest tolerated state (IC50 8 dosenM). Data (0.1 are mg/kg) presented had as the no mean significant ± SEM, n = effect10–14. (G on) spontaneous Intraplantar administration of GpTx-1 (30 nM, 100 nM, 300 nM, 1 μM) concentration-dependently pain behaviorsreversed (Figure spontaneous4H). At pain a dose behaviors of 0.3 in mice mg/kg, evoked GpTx-1 by OD1. ( causesH) Intraperitoneal motor deficits,administration as evidencedof by a significant increaseGpTx-1 in(0.1 the mg/kg) ataxia had indexno significant (Figure (n.s5.; A;p > ataxia0.05) effect index: on spontaneous control, pain 2.7 behaviors˘ 0.4;GpTx-1 in mice (0.3 mg/kg), 9.7 ˘ 3.4; p < 0.05).induced Therefore, by OD1. Statistical it was significance not possible was determined to assess using the the effect t-testof or GpTx-1one-way ANOVA at systemic with doses above Dunnett’s post-test; * p < 0.05 compared to the vehicle control. Data are presented as the mean ± SEM, 0.1 mg/kg. n = 3–9 per group.

A B 15 2.0 *

) 1.5

e 10 m

(

n 1.0

a *

t 5 i

A D 0.5

0 0.0 l l o -1 4 2 o -1 4 2 tr x 6 0 tr x 6 0 n T 2 8 n T 2 8 o p 6 4 o p 6 4 5 1 5 1 C G 8 0 C G 8 0 4 1 4 1 -0 V -0 V F N F N P C P C Figure 5. Motor assessment of systemically (i.p.) delivered GpTx-1, PF-04856264 and CNV1014802. Figure 5. Motor(A) Ataxia assessment index (number of of systemically foot slips per meters (i.p.) deliveredtravelled) assessed GpTx-1, by the PF-04856264 Parallel Rod Floor and Test. CNV1014802. (A) AtaxiaGpTx index-1 (0.3 (number mg/kg; n of = 4) foot significantly slips per increased meters the travelled) ataxia index assessed, while PF by-04856264 the Parallel (30 mg/kg; Rod n = Floor Test. GpTx-1 (0.35) and mg/kg; CNV1014802n = 4) (30 significantly mg/kg; n = 3) had increased no significant the ataxiaeffect compared index, whileto the vehicle PF-04856264 control (n (30= mg/kg; n = 5) and12 CNV1014802). (B) Distance travelled (30 mg/kg; (m) assessedn = 3) by had the Parallel no significant Rod Floor Test. effect CNV1014802 compared (30 to mg/kg; the vehiclen = 3) control significantly reduced the distance travelled, while GpTx-1 (0.3 mg/kg; n = 4) and PF-04856264 (n = 12).(B(30)Distance mg/kg; n travelled= 5) had no (m) significant assessed effect by compared the Parallel to the Rod vehicle Floor control Test. (n CNV1014802 = 12). Statistical (30 mg/kg; n = 3) significantlysignificance reduced was determined the distance using one travelled,-way ANOVA while with GpTx-1Dunnett’s (0.3post- mg/kg;test; * p < 0.05n = compared 4) and PF-04856264 to (30 mg/kg;the nvehicle= 5) control. had no Data significant are presented effect as the comparedmean ± SEM. to the vehicle control (n = 12). Statistical significance was determined using one-way ANOVA with Dunnett’s post-test; * p < 0.05 compared to 2.5. PF-04856264 Is a Selective NaV1.7 Inhibitor that Reverses OD1-Induced Pain Behaviors the vehicle control. Data are presented as the mean ˘ SEM. PF-04856264 is a novel aryl sulfonamide NaV1.7 inhibitor with little effect on NaV1.3 and NaV1.5, although activity at the other NaV subtypes has not been reported [17]. Using the FLIPR membrane 2.5. PF-04856264potential Isassay, a Selective we found Na thatV1.7 PF- Inhibitor04856264 was that a Reversesselective Na OD1-InducedV1.7 inhibitor with Pain >40 Behaviors-fold selectivity over all other NaV subtypes and >1000-fold selective over NaV1.4–NaV1.6. The rank order of potency PF-04856264 is a novel aryl sulfonamide NaV1.7 inhibitor with little effect on NaV1.3 and NaV1.5, (pIC50 ± SEM) was hNaV1.7 (6.53 ± 0.1) > hNaV1.2 (4.87 ± 0.3) > hNaV1.1 (3.61 ± 0.1) > hNaV1.6 (3.53 ± although activity0.1) >> hNa atV1.3 the ~ hNa otherV1.4Na ~ hNaV subtypesV1.5 ~ hNaV1.8 has (Fi notgure been6A,B; see reported also Table [17 1 for]. Using IC50 values the and FLIPR Figure membrane potential assay,S1 for a we sample found trace). that Interestingly, PF-04856264 PF-04856264 was a selective only partially NaV inhibited1.7 inhibitor responses with in >40-fold the FLIPR selectivity assay, leaving a residual 25%–30% change in membrane potential unaffected, despite completely over all other NaV subtypes and >1000-fold selective over NaV1.4–NaV1.6. The rank order of potency inhibiting NaV1.7 current in patch clamp assays (see below). PF-04856264 acts as a gating modifier, (pIC ˘ SEM) was hNa 1.7 (6.53 ˘ 0.1) > hNa 1.2 (4.87 ˘ 0.3) > hNa 1.1 (3.61 ˘ 0.1) > hNa 1.6 50 interacting with theV voltage sensor in Domain IVV and preferentially binding Vto and stabilizing the V (3.53 ˘ 0.1)inactivated >> hNa V state1.3 ~ ofhNa NaVV1.71.4 [17]. ~ hNa We V confirmed1.5 ~ hNa theV 1.8 state (Figure dependence6A,B; of see PF- also04856264 Table inhibition,1 for IC 50 values and Figure S1 for a sample trace). Interestingly, PF-04856264 only partially inhibited responses in the FLIPR assay, leaving a residual 25%–30% change in membrane potential unaffected, despite completely inhibiting NaV1.7 current in patch clamp assays (see below). PF-04856264 acts as a gating modifier, interacting with the voltage sensor in Domain IV and preferentially binding to and stabilizing the Toxins 2016, 8, 78 8 of 18

finding a 20-fold shift in the NaV1.7 IC50 from closed/resting state inhibition (2.7 μM) to open/inactivated state inhibition (134 nM) (Figure 6C). Interestingly, in the presence of OD1 (300 nM), PF-04856264 had a three-fold loss of potency in both the closed/resting state (IC50 11 μM) and the open/inactivated state (IC50 445 nM) (see Figure S2 for a sample trace). Toxins 2016, 8, 78 8 of 19 Intraplantar injection of PF-04856264 concentration-dependently reduced spontaneous pain behaviors induced by co-administered OD1 (300 nM), although surprisingly, the high local concentrations required did not correlate with its in vitro potency (Figure 6D; spontaneous pain inactivated statebehaviors of Na/10 V min:1.7 [c17ontrol]. We, 105 conﬁrmed ± 6; PF-04856264 the state (1 mM), dependence 19 ± 6; PF-04856264 of PF-04856264 (100 μM), 31inhibition, ± 8; ﬁnding PF-04856264 (10 μM), 81 ± 12; p < 0.05). Intraperitoneal administration of PF-04856264 was also a 20-fold shift in the NaV1.7 IC50 from closed/resting state inhibition (2.7 µM) to open/inactivated effective (Figure 6E; spontaneous pain behaviors/10 min: control, 105 ± 6; PF-04856264 (30 mg/kg), state inhibition42 ± (1348; p < nM)0.05) and (Figure well tolerated,6C). Interestingly, with no significant in the motor presence adverse effects of OD1 observed (300 at nM), doses PF-04856264up to had a three-fold30 loss mg/kg, of potency correlating in well both with the the closed/resting in vitro selectivity state profile(IC (Figure50 11 5A;µ M) ataxia and index: the c open/inactivatedontrol, 2.7 ± 0.4; PF-04856264 (30 mg/kg), 4.6 ± 1.3; p > 0.05; Figure 5B; distance travelled: control, 1.49 ± 0.16 m; state (IC50 445 nM) (see Figure S2 for a sample trace). PF-04856264 (30 mg/kg), 1.03 ± 0.17 m; p > 0.05).

Figure 6. NaV selectivity and in vivo effects of the aryl sulfonamide PF-04856264. Activity of

Figure 6. NaPF-V04856264selectivity was assessed and in in vivo HEK effects cells heterologously of the aryl expressing sulfonamide NaV1.1–Na PF-04856264.V1.8 using FLIPR Activity of PF-04856264membrane was assessed potential assays. in HEK (A) PF cells-04856264 heterologously selectively inhibited expressing NaV1.7, with Na theV following1.1–NaV pIC1.850’s using FLIPR ranked in order of potency: hNaV1.7 (6.53 ± 0.1) > hNaV1.2 (4.87 ± 0.3) > hNaV1.1 (3.61 ± 0.1) > hNaV1.6 membrane potential assays. (A) PF-04856264 selectively inhibited NaV1.7, with the following pIC50’s (3.53 ± 0.1). PF-04856264 had no inhibitory activity at the other NaV subtypes up to a concentration of ranked in order300 of μM. potency: Data are hNa presentedV1.7 (6.53 as ˘the0.1) mean > hNa ± SEMV1.2 from (4.87 three˘ 0.3) independent > hNaV1.1 experiments (3.61 ˘ 0.1). > hNaV1.6 (3.53 ˘ 0.1). PF-04856264(B) Representative had concentration no inhibitory response activity curves at of the PF other-04856264 Na atV subtypes NaV1.1, 1.2, up 1.6 to and a concentration 1.7. of 300 µM. Data(C are) Activity presented of PF-04856264 as the mean in CHO˘ cellsSEM heterologously from three expressing independent hNaV1.7 experiments. assessed by automated (B) Representative patch clamping. Concentration response curve of PF-04856264 elicited by a 20-ms pulse to 0 mV from concentrationa holding response potential curves of −100 of PF-04856264mV or an 8 s conditioning at NaV1.1, voltage1.2, step 1.6 of and −55 mV 1.7. with (C) and Activity withoutof OD1 PF-04856264 in CHO cells heterologously(300 nM). PF-04856264 expressing preferentially hNa boundV1.7 to assessed the open/inactivated by automated state (IC patch50 134 nM) clamping. compared to Concentration response curvethe of closed/resting PF-04856264 state elicited (IC50 2.7 by μM). a 20-ms The presence pulse toof 0OD1 mV (300 from nM) a holdingreduced the potential potency of ´100 mV or PF-04856264 in both the open/inactive state (IC50 445 nM) and the closed/resting state (IC50 11 μM). ´ an 8 s conditioningData are voltage presented step as the of mean55 ± mV SEM, with n = 6 and–14. ( withoutD) Intraplantar OD1 administration (300 nM). PF-04856264 of PF-04856264 preferentially bound to the(10 open/inactivated μM, 100 μM, 1 mM) stateconcentration (IC50 134-dependently nM) compared reversed spontaneous to the closed/resting pain behaviors statein mice (IC 50 2.7 µM). The presenceevoked of OD1 by OD1 (300. (E nM)) Intraperitoneal reduced the administration potency of PF PF-04856264-04856264 (30 mg/kg) in both also the significantly open/inactive state reduced spontaneous pain behaviors in mice. Statistical significance was determined using one-way (IC 445 nM) and the closed/resting state (IC 11 µM). Data are presented as the mean ˘ SEM, n = 6–14. 50 ANOVA with Dunnett’s post-test; * p < 0.0550 compared to the vehicle control. Data are presented as (D) Intraplantarthe mean administration ± SEM, n = 3–9 per of group. PF-04856264 (10 µM, 100 µM, 1 mM) concentration-dependently reversed spontaneous pain behaviors in mice evoked by OD1. (E) Intraperitoneal administration of PF-04856264 (30 mg/kg) also signiﬁcantly reduced spontaneous pain behaviors in mice. Statistical signiﬁcance was determined using one-way ANOVA with Dunnett’s post-test; * p < 0.05 compared to the vehicle control. Data are presented as the mean ˘ SEM, n = 3–9 per group.

Intraplantar injection of PF-04856264 concentration-dependently reduced spontaneous pain behaviors induced by co-administered OD1 (300 nM), although surprisingly, the high local concentrations required did not correlate with its in vitro potency (Figure6D; spontaneous pain behaviors/10 min: control, 105 ˘ 6; PF-04856264 (1 mM), 19 ˘ 6; PF-04856264 (100 µM), 31 ˘ 8; PF-04856264 (10 µM), 81 ˘ 12; p < 0.05). Intraperitoneal administration of PF-04856264 was also effective (Figure6E; spontaneous pain behaviors/10 min: control, 105 ˘ 6; PF-04856264 (30 mg/kg), 42 ˘ 8; p < 0.05) and well tolerated, with no signiﬁcant motor adverse effects observed at doses up to 30 mg/kg, correlating well with the in vitro selectivity proﬁle (Figure5A; ataxia index: control, 2.7 ˘ 0.4; PF-04856264 (30 mg/kg), 4.6 ˘ 1.3; p > 0.05; Figure5B; distance travelled: control, 1.49 ˘ 0.16 m; PF-04856264 (30 mg/kg), 1.03 ˘ 0.17 m; p > 0.05). Toxins 2016, 8, 78 9 of 18

2.6. CNV1014802 is a Non-Selective NaV Inhibitor Toxins 2016, 8, 78 9 of 19 CNV1014802 is reported to be a state-dependent inhibitor of NaV1.7; however, information on the potency and selectivity profile has not been reported [8]. Surprisingly, CNV1014802 was found to be a relatively non-selective NaV inhibitor, in the FLIPR membrane potential assay, with a rank 2.6. CNV1014802order of is potency a Non-Selective (pIC50 ± SEM): Na VhNaInhibitorV1.8 (5.25 ± 0.1) > hNaV1.4 (5.09 ± 0.2) > hNaV1.2 (4.99 ± 0.2) > hNaV1.6 (4.84 ± 0.1) > hNaV1.3 (4.82 ± 0.3) > hNaV1.1 (4.70 ± 0.2) > hNaV1.7 (4.58 ± 0.2) > hNaV1.5 (4.18 CNV1014802± 0.2) (Figure is reported 7A,B; see also to beTable a state-dependent1 for IC50 values and Figure inhibitor S1 for a of sample NaV trace).1.7; however, We were able information to on the potencyconfirm and selectivity the state-dependence profile hasreported not for been CNV1014802, reported finding [8]. a Surprisingly, nine-fold shift in CNV1014802 the NaV1.7 IC50 was found to be a relativelybetween non-selective closed/resting state Na inhibitioninhibitor, (54 μM) in and the open/inactivated FLIPR membrane state inhibition potential (6.3 μM), assay, with with a rank a preference for the open/inactivatedV state (Figure 7C). Compared to PF-04856264, the potency of order of potencyCNV1014802 (pIC50 at˘ NaSEM):V1.7 is 50 hNa-fold weaker;V1.8 (5.25 however˘ 0.1), the presence > hNa Vof1.4 OD1 (5.09 (300 nM)˘ 0.2) did not > hNasignificantlyV1.2 (4.99 ˘ 0.2) > hNaV1.6 (4.84affect˘ 0.1) the potency> hNa V of1.3 CNV1014802 (4.82 ˘ 0.3) at Nav1.7 > hNa inV 1.1 the (4.70 closed˘/resting0.2) > state hNa (ICV501.7 35 (4.58 μM) or˘ the0.2) > hNaV1.5 open/inactivated state (IC50 11 μM) (see Figure S2 for a sample trace). (4.18 ˘ 0.2) (Figure7A,B; see also Table1 for IC 50 values and Figure S1 for a sample trace). We Despite in vitro activity at hNaV1.7, CNV1014802 was unable to significantly attenuate were able toOD1 confirm-induced the spontaneous state-dependence pain behaviors reported when delivered for CNV1014802, locally by intraplantar finding injection a nine-fold at a shift in the NaV1.7 IC50 concentrationbetween closed/resting of 1 mM (Figure 7D). state Ho inhibitionwever, CNV1014802 (54 µM) was and effective open/inactivated when delivered by state the inhibition (6.3 µM), withintraperitoneal a preference route for (Figure the open/inactivated7E; spontaneous pain behaviors state (Figure/10 min:7 cC).ontrol Compared, 105 ± 6; CNV1014802 to PF-04856264, the (30 mg/kg), 45 ± 5; CNV1014802 (3 mg/kg), 60 ± 8; p < 0.05), with sedative adverse effects observed at potency of CNV1014802the highest dose at, as Na evidencedV1.7 is by50-fold a significant weaker; reductio however,n in distance the travelledpresence, but of no OD1 significant (300 nM) did not significantlyincrease affect the in the potency ataxia index of CNV1014802 (Figure 5A; ataxia at index: Nav1.7 control in, the 2.7 ± closed/resting 0.4; CNV1014802 (30 state mg/kg), (IC 50 35 µM) or the open/inactivated3.5 ± 1.9; p > state 0.05; (IC Figure 11 5B; µ distanceM) (see travelled: Figure control S2 for, 1.49 a sample ± 0.16 m; trace). CNV1014802 (30 mg/kg), 0.66 ± 0.20 m; p < 0.05). 50

A B

4.0

s 100 1.1

o 1.2

4.5 p 75 s 1.3

1.4

I 5.0 d 50

e 1.5

l 1.6

a 25

5.5 m 1.7

o 0 1.8

N 6.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 -6.0 -5.5 -5.0 -4.5 -4.0 -3.5 Log [CNV1014802] (M) NaV subtype C D E

120

160

n 1.2 –100 mV i

140 m

m 100

–55mV 0

1.0 0

120 1

/ –100 mV + OD1 / s 80 0.8 s

r r 100 *

–55 mV + OD1 u

i I 0.6

i * / 60 I 80

a 0.4 a

h h 60

e 40

0.2 B

i 0.0 i 20

a a 20

P 0 0

-7 -6 -5 -4 -3 g g

n Log [CNV1014802] (M) n

0 3

Figure 7. NaV selectivity and in vivo effects of the clinical candidate CNV1014802. Activity of

Figure 7. NaCNV1014802V selectivity was assessed and in in vivo HEKeffects cells heterologously of the clinical expressing candidate NaV1.1–Na CNV1014802.V1.8 using FLIPR Activity of CNV1014802membrane was assessed potential in assays HEK. (A)cells CNV1014802 heterologously non-selectively expressing inhibited Na NaV channels,V1.1–Na withV1.8 the using FLIPR following pIC50’s ranked in order of potency: hNaV1.8 (5.25 ± 0.1) > hNaV1.4 (5.09 ± 0.2) > hNaV1.2 membrane potential assays. (A) CNV1014802 non-selectively inhibited NaV channels, with the (4.99 ± 0.2) > hNaV1.6 (4.84 ± 0.1) > hNaV1.3 (4.82 ± 0.3) > hNaV1.1 (4.70 ± 0.2) > hNaV1.7 (4.58 ± 0.2) > ˘ ˘ following pIChNa50V’s1.5 ranked (4.18 ± 0.2). in orderData are of presented potency: as the hNa meanV1.8 ± SEM (5.25 from0.1) three > independent hNaV1.4 (5.09experiments.0.2) > hNaV1.2 (4.99 ˘ 0.2) >(B hNa) RepresentativeV1.6 (4.84 ˘ concentration0.1) > hNa responseV1.3 (4.82 curves˘ of0.3) CNV1014802 > hNaV1.1 at Na(4.70V1.1–˘1.80.2). (C) > Activity hNaV 1.7 of (4.58 ˘ 0.2)

> hNaV1.5 (4.18 ˘ 0.2). Data are presented as the mean ˘ SEM from three independent experiments. (B) Representative concentration response curves of CNV1014802 at NaV1.1–1.8. (C) Activity of CNV1014802 in CHO cells heterologously expressing hNaV1.7 assessed by automated patch clamping. Concentration response curve of CNV1014802 elicited by a 20-ms pulse to 0 mV from a holding potential of ´100 mV or an 8-s condition voltage step of ´55 mV with and without OD1 (300 nM). CNV1014802

preferentially bound to the open/inactivated state (IC50 6.3 µM) compared to the closed/resting state (IC50 54 µM). The presence of OD1 (300 nM) did not significantly affect the potency of CNV1014802. Data are presented as the mean ˘ SEM, n = 4–7. (D) Intraplantar administration of CNV1014802 (1 mM) had no significant effect on spontaneous pain behaviors in mice evoked by OD1. (E) However, intraperitoneal administration of CNV1014802 (3 and 30 mg/kg) significantly reduced spontaneous pain behaviors in mice. Statistical significance was determined using the t-test or one-way ANOVA with Dunnett’s post-test; * p < 0.05 compared to the vehicle control. Data are presented as the mean ˘ SEM, n = 3–9 per group. Toxins 2016, 8, 78 10 of 19

Despite in vitro activity at hNaV1.7, CNV1014802 was unable to significantly attenuate OD1-induced spontaneous pain behaviors when delivered locally by intraplantar injection at a concentration of 1 mM (Figure7D). However, CNV1014802 was effective when delivered by the intraperitoneal route (Figure7E; spontaneous pain behaviors/10 min: control, 105 ˘ 6; CNV1014802 (30 mg/kg), 45 ˘ 5; CNV1014802 (3 mg/kg), 60 ˘ 8; p < 0.05), with sedative adverse effects observed at the highest dose, as evidenced by a significant reduction in distance travelled, but no significant increase in the ataxia index (Figure5A; ataxia index: control, 2.7 ˘ 0.4; CNV1014802 (30 mg/kg), 3.5 ˘ 1.9; p > 0.05; Figure5B; distance travelled: control, 1.49 ˘ 0.16 m; CNV1014802 (30 mg/kg), 0.66 ˘ 0.20 m; p < 0.05).

3. Discussion

Choosing a relevant animal pain model to evaluate the analgesic efficacy of NaV1.7-selective inhibitors is an essential step in the pre-clinical development of these compounds. We therefore developed a novel mouse model of NaV1.7-mediated pain based on intraplantar injection of the scorpion toxin OD1 that allows rapid in vivo characterization of the analgesic efficacy of NaV1.7 inhibitors. OD1 was initially classed as a α-scorpion toxin that increases peak current by delaying fast inactivation, but has little effect on the voltage dependence of activation at NaV1.7 expressed in oocytes at 50 nM. [14,22]. Similarly, a small, albeit statistically-significant shift in the voltage dependence of activation was recently reported at NaV1.7 expressed in mammalian cells [23]. Our results confirm that this shift in the voltage dependence of activation is further enhanced at higher concentrations. Thus, OD1 has mixed α/β-scorpion toxin effects at NaV1.7, not only increasing peak current and delaying inactivation, but also shifting the voltage dependence of activation to more hyperpolarized potentials. Thus, the mixed functional activity of OD1, which was also previously reported for NaV1.4 and NaV1.6 [15], appears to be concentration dependent, suggesting that OD1 preferentially interacts with the α-scorpion toxin-binding site with additional interaction with the β-scorpion toxin-binding site at higher concentrations. Both pharmacological effects are consistent with the ability of OD1 to induce spontaneous action potential firing in sensory neurons. While pure α-toxins would be expected to amplify supra-threshold responses due to the inhibition of channel inactivation and would lead to amplification of small generator potentials, a shift in the voltage dependence of activation would additionally result in channel opening closer to resting membrane potential, resulting in greater amplification of sub-threshold responses and induction of spontaneous action potential firing. The functional effects of OD1 on NaV1.7 can be correlated to the functional consequences of NaV1.7 gain-of-function mutations in the painful syndromes inherited erythromelalgia and paroxysmal extreme pain disorder. Typically, mutations present in inherited erythromelalgia patients cause a voltage-dependent shift of activation of NaV1.7 to more hyperpolarized potentials [24–27], corresponding to the β-scorpion toxin-like effects of OD1, whereas paroxysmal extreme pain disorder mutations generally delay the fast inactivation [28–30], corresponding to the α-scorpion toxin activity of OD1. This indicates that the OD1 mouse model may be used to identify new analgesics with the potential to treat the poorly managed conditions of inherited erythromelalgia and paroxysmal extreme pain disorder. NaV1.7 is preferentially expressed in the majority of peripheral sensory neurons that have been functionally characterized as nociceptive A- and C-fibers [31]. Specifically, it is present on the peripheral terminals of nociceptive neurons, where it contributes to the amplification of sub-threshold generator potentials by producing ramp currents that depolarize the membrane to the threshold required to generate an action potential [32,33]. Application of OD1 to the receptive fields of A- and C-fibers caused spontaneous action potential firing, consistent with the effect of OD1 on the biophysical properties of NaV1.7, which would result in depolarization closer to resting membrane potential, leading to spontaneous action potential firing. The spontaneous pain behaviors induced by intraplantar injection Toxins 2016, 8, 78 11 of 19 of OD1 are consistent with the spontaneous action potential firing seen in A- and C-fibers in the skin-nerve preparation. Although OD1 is a selective NaV1.7 activator, it also has activity at NaV1.6, albeit at higher concentrations. Given that intraplantar administration of the NaV1.6 activator Cn2 also causes spontaneous pain behaviors in mice [18], we assessed the relative contribution of NaV1.7 and NaV1.6 to spontaneous pain behaviors induced by OD1. While the NaV1.6 inhibitor GIIIA partially reversed OD1-induced spontaneous pain by 33%, the NaV1.7 selective inhibitors GpTx-1 and PF-04856264 were effective after intraplantar injection at concentrations unlikely to significantly inhibit NaV1.6 in vivo, suggesting that the majority of spontaneous pain behaviors induced by OD1 are mediated ´{´ through NaV1.7. This was further confirmed in NaV1.7 mice, where the behavioral phenotype was strongly attenuated, with residual responses likely attributable to the NaV1.6 activity of OD1. Thus, OD1 elicited a robust and long-lasting behavioral response that was mediated predominantly through NaV1.7, allowing the efficacy of NaV1.7 inhibitors to be readily assessed for their potential in the treatment of NaV1.7-mediated pain. Consistent with its reported activity at NaV1.7 [20], GpTx-1 was found to significantly reduce OD1-induced spontaneous pain behaviors when administered locally. The reasons for the potent in vivo effects of GpTx-1 are not entirely clear. Peptide NaV inhibitors can have particularly slow off-rates, which could contribute to high apparent in vivo potency. In addition, it is currently unclear what level of NaV1.7 inhibition is required to cause analgesia, although it is reasonable to assume that complete inhibition of NaV1.7 current may not be required to prevent NaV1.7-mediated action potential firing. Thus, for compounds that are able to achieve relevant free drug concentrations at peripheral nerve endings and that can achieve full inhibition of current at relevant states, the in vivo IC50 may be lower than the observed in vitro IC50. Although effective locally, we were not able to reproduce this effect with systemic delivery of GpTx-1. This is likely due to dose-limiting side effects, as the maximum tolerated systemic dose (0.1 mg/kg) may not have achieved sufficiently high concentration at peripheral nerve endings in the hind paw to inhibit NaV1.7. Motor adverse effects were observed at doses above 0.1 mg/kg, limiting administration of higher doses, and were most likely attributable to activity at NaV1.6, which is expressed in the periphery on sensory and motor neurons, where loss of function leads to movement disorders and hind limb paralysis [34]. This suggests that greater than the 30-fold selectivity for NaV1.7 over NaV1.6 seen with GpTx-1 is required to have an acceptable therapeutic window for systemic use; however, off-target activity at other NaV isoforms may also contribute to the adverse effects. In contrast, the aryl sulfonamide PF-04856264 displayed excellent selectivity (>1000 fold) for NaV1.7 over NaV1.4–NaV1.6. This selectivity profile was reflected in vivo, with relatively high systemic doses (30 mg/kg) causing no significant motor adverse effects. PF-04856264 was effective at reversing OD1-induced spontaneous pain behaviors when delivered both locally and systemically, although higher than expected doses were required to elicit analgesia by either routes relative to its potency in the FLIPR membrane potential (IC50 300 nM) and electrophysiology (IC50 130 nM) assays. However, given that PF-04856264 preferentially binds to NaV1.7 in the inactivated state, it is possible that the delayed inactivation produced by OD1 promoted an unfavorable channel state for PF-04856264 binding. Indeed, the presence of OD1 led to a three-fold loss of potency in the electrophysiological assays. State-dependent inhibition of NaV1.7 may also explain the inability of PF-04856264 to fully block membrane potential changes using the fluorescence assay, where residual responses could be mediated by veratridine-modified channels resistant to inactivation that in turn are unaffected by PF-04856264 [35]. The physiological relevance of a preference for the inactivated state of NaV1.7 in other pathological pain conditions remains to be established, although such a mechanism might imply lack of efficacy in paroxysmal extreme pain disorder, where NaV1.7 channels are relatively resistant to fast inactivation. Consistent with previous reports, CNV1014802 also was a state-dependent inhibitor of NaV1.7 in our hands, although it lacked the NaV subtype selectivity of PF-04856264. Surprisingly, CNV1014802 Toxins 2016, 8, 78 12 of 19 lacked in vivo analgesic efficacy when delivered locally at a concentration of 1 mM, perhaps reflecting its relatively poor potency at NaV1.7 in membrane potential (32 µM) and electrophysiology (6.3 µM) assays. In addition, CNV1014802 (formally GSK1014802 and now raxatrigine) is reported to be highly (94%) protein bound, which could also lead to a decrease in free effective drug at the site of action after local administration [36]. In contrast to local administration, CNV1014802 was effective when administered systemically, although some sedative adverse effects were observed at the highest dose (30 mg/kg). These effects are consistent with the reported ability of CNV1014802 to penetrate the central nervous system, which likely is crucial for its efficacy in animal models of seizures and schizophrenia [36,37], suggesting that analgesic activity after systemic dosing may also be due to central effects. Based on our pharmacological validation, the OD1-induced spontaneous pain model can be used to assess the in vivo activity of compounds when delivered by both the intraplantar and a systemic route. An advantage of the intraplantar route is that it uses considerably less material and thus may be the preferable route of administration for screening compounds for in vivo activity in the early stages of lead identification. However, as was the case for PF-04856264 and CNV1014802, in vivo activity in the OD1 model did not correspond well with the in vitro potency found in the electrophysiology and FLIPR membrane potential assays. At least in the case of PF-04856264, species-specific effects are unlikely to contribute to this discrepancy, as equipotency at human and mouse NaV1.7 has been previously reported [17]. In addition, although PF-04856264 lost potency in the presence of OD1, consistent with preferential binding to the inactivated state, this change was relatively small and would not explain the relative lack of potency in vivo. Interestingly, we observed incomplete inhibition of NaV1.7-mediated responses in the FLIPR membrane potential assay by PF-04856264 and CNV1014802 (Figure S1), although both compounds completely inhibited currents from NaV1.7 patch-clamp recordings. While it is unclear whether this effect correlates to the ability of these compounds to prevent activity-induced membrane depolarization at peripheral sensory nerve endings in vivo, such effects may contribute to the poor in vivo activity observed. In addition, it is plausible that other factors, such as the pharmacokinetic properties of these compounds, affect in vivo efficacy. Nonetheless, these results highlight the need for a simple animal model, such as the OD1 model described here, which permits rapid evaluation of in vivo on-target effects, as the in vitro properties do not always correspond well to in vivo activity. Scorpion α-toxins, such as OD1, are known to bind to Domain IV [38]; however, we have demonstrated that efficacy in the OD1 model is not dependent on competitive binding at this site. For example, TTX, which binds to the pore of the channel, and GpTx-1, which binds to Domain II [39], were both able to completely abolish OD1-induced spontaneous pain behaviors. Although not as potent in vivo, the state-dependent inhibitors PF-04856264, predicted to share an overlapping binding site with scorpion α-toxins on Domain IV [17], and CNV1014802 (binding site remains to be determined) were also effective. Thus, the OD1 model is suitable to evaluate in vivo activity of NaV1.7 inhibitors irrespective of their mechanism of action or binding site. In conclusion, we have established a NaV1.7-mediated mouse model of pain based on intraplantar injection of the scorpion toxin OD1 and pharmacologically characterized the activity of the spider peptide GpTx-1, the small molecule PF-04856264 and the clinical candidate CNV1014802. The model is suitable to evaluate in vivo on-target activity of NaV1.7 inhibitors; however, it is unclear if activity in OD1-induced pain translates to anti-nociceptive or anti-allodynic effects in other rodent models or analgesia in humans. While efficacy in the OD1 model is useful to validate in vivo on-target activity at NaV1.7, species differences need to be considered carefully when using animal models of human disease. Given the relatively high homology between mouse and human NaV isoforms, significant pharmacological differences have rarely been reported. Indeed, PF-04856264 is equipotent at mouse and human NaV1.7 [17], and the effects of OD1 on rodent NaV1.7 closely match the effects on human NaV1.7 [15]. While NaV1.7 is arguably a well-validated pain target in humans based on genetic evidence from loss-of-function and gain-of-function conditions, including congenital insensitivity Toxins 2016, 8, 78 13 of 19 to pain, erythromelalgia and paroxysmal extreme pain disorder, it remains to be established if pharmacological inhibition of NaV1.7 can achieve global graded analgesia.

4. Experimental Section

4.1. Chemicals OD1 was synthesized as previously described [15]. PF-04856264 was from SYNthesis Med Chem (Parkville, VIC, Australia) and CNV1014802 was from Axon MedChem (Groningon, The Netherlands). GIIIA was synthesized by Boc chemistry using methods previously described [40]. TTX was from Tocris Bioscience (Bristol, UK), and veratridine was from Alomone Labs (Jerusalem, Israel). All other reagents were from Sigma Aldrich (Castle Hill, NSW, Australia), unless otherwise stated. Peptides were routinely diluted in 0.1% bovine serum albumin (BSA).

4.2. Synthesis of GpTx-1 Solvents for reversed-phase high-performance liquid chromatography (HPLC) consisted of 0.05% trifluoroacetic acid (TFA)/H2O (A) and 90% acetonitrile/0.043% TFA/H2O (B). Analytical HPLC was performed on a Shimadzu LC20AT system (Shimadzu, Kyoto, Japan) using a Thermo Hypersil GOLD 2.1 ˆ 100 mm C18 column (ThermoScientific, Brendale, QLD, Australia) heated at 40 ˝C with flow rate of 0.3 mL/min. A gradient of 10–55% B over 30 min was used, with detection at 214 nm. Preparative HPLC was performed on a Vydac 218TP1022 column (Grace, Columbia, MD, USA) running at a flow rate of 16 mL/min using a gradient of 5%–45% B over 40 min. Mass spectrometry was performed on an API2000 (ABI Sciex, Mt Waverley, VIC, Australia) mass spectrometer in positive ion mode. Protected 9-fluorenylmethyloxycarbonyl (Fmoc) amino acid derivatives and 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU) were purchased from Iris Biotech (Marktredwitz, Germany); Rink-amide resin, N,N-dimethylformamide (DMF), N,N-diisopropylethylamine (DIEA) and TFA from Auspep (Tullamarine, VIC, Australia). All other reagents were obtained from Sigma-Aldrich (Castle Hill, NSW, Australia). GpTx-1 was assembled on a Symphony (Protein Technologies Inc., Tucson, AZ, USA) automated peptide synthesizer on a Rink-amide (loading 0.67 mmol/g) polystyrene resin on a 0.1 mmol scale. Fmoc deprotections were achieved using 30% piperidine/DMF (1 ˆ 1.5 min, then 1 ˆ 4 min). Couplings were performed in DMF using 5 equivalents of Fmoc-amino acid/HBTU/DIEA (1:1:1) relative to resin loading for 2 ˆ 20 min. Amino acid side-chains were protected as Asn(Trt), Asp(OtBu), Arg(Pbf), Cys(Trt), His(Trt), Lys(Boc), Ser(tBu), Thr(tBu), Trp(Boc), Tyr(tBu). Cleavage from the resin and removal of side-chain protecting groups was achieved by treatment with 95% TFA/2.5% TIPS/2.5% H2O at room temperature for 2 h. After most of the cleavage solution was evaporated under a stream of N2, the product was precipitated and washed with cold Et2O and lyophilized from 50% acetonitrile/0.1% 4+ TFA/H2O 223 mg; ESI-MS (m/z): calc. (average) 1021.0 [M + 4H] , found 1020.8. Purified reduced peptide (41 mg), reduced glutathione (100 equiv) and oxidized glutathione (10 equiv.) were dissolved in 6 M GnHCl (21 mL) then added to a solution of 0.36 M NH4OAc (pH 8.0, 230 mL) and stirred at room temperature with exposure to air for 48 h. The single major product was isolated by preparative HPLC: 13.5 mg; ESI-MS (m/z): calc. (average) 1019.5 [M + 4H]4+, found 1019.4. The crude product was purified by preparative HPLC to give 42 mg of hexathiol GpTx-1 (99% purity).

4.3. Animals Ethical approval for in vivo experiments in animals was obtained from The University of Queensland animal ethics committee (PHARM/261/13/ARC and IMB/PACE/326/15). Experiments involving animals were conducted in accordance with the Animal Care and Protection Regulation Qld (2012), the Australian Code of Practice for the Care and Use of Animals for Scientiﬁc Purposes, 8th edition (2013, and the International Association for the Study of Pain Guidelines for the Use of Toxins 2016, 8, 78 14 of 19

Animals in Research. All efforts were made to minimize animal suffering and to reduce the number of animals used. For behavioral assessment, we used adult male C57BL/6J mice aged 6–8 weeks. Animals were housed in groups of 3 or 4 per cage, under 12-h light-dark cycles and had standard rodent chow and water ad libitum. Advillin-Cre constitutive NaV1.7 knockout mice were kindly provided by Professor John N. Wood (University College London) [41].

4.4. Behavioral Assessment Following Injection of OD1 OD1 was diluted in phosphate-buffered saline/0.1% BSA (10–300 nM) and administered by shallow subcutaneous (intraplantar) injection to the left hind paw of mice in a volume of 40 µL under light isoﬂurane (3%) anesthesia. Mice were then placed individually into polyvinyl boxes (10 ˆ 10 ˆ 10 cm), and spontaneous pain behavior (licks and ﬂinches) was counted by an investigator unaware of the treatments received from video recordings for 40 min post-injection at 5-min intervals. Once spontaneous pain behavior had ceased, mechanical and thermal allodynia was assessed using an electronic von Frey apparatus (MouseMet Electronic von Frey, TopCat Metrology, Ely, UK) and the Hargreaves method (Plantar Analgesia Meter, IITC, Woodland Hills, CA, USA), as previously described [42].

4.5. Effect of Pharmacological Inhibitors and Clinical Compounds OD1 (300 nM) was administered as described above, and cumulative spontaneous pain behaviors were quantiﬁed for 10 min immediately post-injection of OD1 by a blinded observer. Compounds delivered by the intraplantar route were co-injected with OD1 at the concentrations stated. Compounds delivered by the intraperitoneal route were administered 15 min prior to injection of OD1 at the doses stated in a volume of 10 µL/g. All compounds were diluted in phosphate-buffered saline, except PF-04856264 and CNV1014802, which were diluted in phosphate-buffered saline/10% DMSO due to poor solubility.

4.6. Motor Performance Assessment Motor performance was assessed using the Parallel Rod Floor Test and analyzed using ANY-Maze software (Stoelting Co., version 4.70, Wood Dale, IL, USA). GpTx-1 (0.3 mg/kg), PF-04856264 (30 mg/kg) and CNV1014802 (30 mg/kg) were administered by the intraperitoneal route, as described above, 15 min prior to assessment of motor performance. Mice were then placed in the Parallel Rod Floor Test apparatus, and the distance travelled (m) and number of foot slips were recorded over 1 min using the ANY-Maze software. The ataxia index was calculated by dividing the number of foot slips by the distance travelled (m).

4.7. Skin-Nerve Preparation The effect of OD1 on A- and C-fibers was assessed using single fiber recordings from isolated mouse skin-saphenous nerve preparations, as previously described [43]. Briefly, the saphenous nerve along with the skin of the dorsal hind paw and lower leg of adult C57BL/6J mice was removed and placed in an organ chamber, perfused with carbogenated synthetic interstitial fluid containing the following (in mM): NaCl (107.8), KCl (3.5), MgSO4 (0.69), NaHCO3 (26.2), NaH2PO4 (1.67), Na-gluconate (9.64), glucose (5.55), sucrose (7.6) and CaCl2 (1.53); pH 7.3. The saphenous nerve was placed in a separate recording chamber immersed in paraffin oil and was desheathed and teased apart until a single fiber recording was identified by mechanically probing the corresponding receptive field. Fibers were then classified based on conduction velocity assessed by electrical stimulation of the receptive field with a microelectrode (C-fiber < 1 m/s, A-fiber 1.6–12 m/s). The receptive field was isolated using a plastic ring, and OD1 (30 nM) was continuously perfused at a rate of 8 mL/min at 32 ˝C. Data were recorded and analyzed using DAPSYS, Version 8. Toxins 2016, 8, 78 15 of 19

4.8. Cell Culture

Human embryonic kidney (HEK 293) cells stably expressing hNaV1.1, hNaV1.2, hNaV1.3, hNaV1.4, hNaV1.5, hNaV1.7 and hNaV1.8 (SB Drug Discovery, Glasgow, UK) were cultured in MEM containing 10% v/v fetal bovine serum supplemented with L-glutamine (2 mM) and selection antibiotics as recommended by the manufacturer. HEK 293 cells stably expressing hNaV1.6 (kind gift from Dr. Frank Lehmann-Horn, University of Ulm, Germany) were cultured in DMEM containing 10% v/v fetal bovine serum and geneticin 0.5 mg/mL. CHO cells stably expressing hNav1.7 (Chantest, OH, USA) were cultured in Ham’s F12 containing 10% v/v fetal bovine serum and selection antibiotics, as recommended ˝ by the manufacturer. Cells were grown in a humidiﬁed 5% CO2 incubator at 37 C, grown to 70%–80% conﬂuence and passaged every 3–4 days using TrypLE Express (Invitrogen, Scoresby, VIC, Australia).

4.9. FLIPR Membrane Potential Assays

HEK 293 cells stably expressing hNaV1.1–1.8 were plated 48 h before the assay on 384-well black-walled imaging plates at a density of 10,000–15,000 cells per well and were loaded with red membrane potential dye (Molecular Devices, San Francisco, CA, USA) according to the manufacturer’s instructions for 30 min at 37 ˝C. After the addition of GpTx-1, PF-04856264 and CNV1014802 using the TETRA FLIPR , cells were incubated a further 5 min before stimulating NaV using veratridine (60 µM, NaV1.1–NaV1.7) or deltamethrin (150 µM, NaV1.8). Changes in membrane potential were assessed using the FLIPRTETRA (excitation 515–545 nm, emission 565–625 nm) every second for 300 s after adding agonists. To quantitate the effect of test compounds on NaV responses, the area under the curve (AUC) corresponding to 300 s after the addition of veratridine was computed using ScreenWorks (Molecular Devices, Version 3.2.0.14) and normalized to buffer responses (0%) and veratridine control responses (100%).

4.10. Electrophysiology CHO cells were used for the electrophysiology, because in our experience, they are more robust to clamp compared to HEK cells. CHO hNaV1.7 were passaged 48 h prior to the assay in a T-175 flask and cultured in selection-free media at 37 ˝C or CHO hNav1.7 EZCells™ were thawed the day of the experiment according to the manufacture’s specifications. To induce expression in the CHO hNaV1.7 cells, tetracycline (1 µg/mL) was added to culture media 24 h prior to the assay. Cells were harvested at 70%–80% confluence and resuspended to 1 ˆ 106 cells/mL in Ex-Cell ACF CHO Medium with 25 mM HEPES and 1ˆ Glutamax (Gibco, Thermo Fisher Scientic, Scoresby, VIC, Australia) before being transferred to the QPatch QStirrer and allowed to recover for 30 min. Extracellular solution (EC) was used to resuspend all test compounds and contained (in mM): NaCl (140), KCl (4), CaCl2 (2), MgCl2 (1), HEPES (10), CdCl2 (0.1) and glucose (10). The pH was adjusted to 7.4 with NaOH, and osmolarity was adjusted with sucrose to 315 mOsm. The intracellular solution contained (in mM): CsF (140), EGTA/CsOH (1/5), HEPES (10) and NaCl (10). The pH was adjusted to 7.4 with CsOH and osmolarity adjusted with sucrose to 320 mOsm. OD1 and GpTx-1 were diluted in EC with the addition of 0.1% BSA. Whole-cell patch-clamp experiments were performed at room temperature on a QPatch-16 automated electrophysiology platform (Sophion Bioscience, Ballerup, Denmark) using 16-channel planar patch chip plates (QPlates; Sophion Bioscience) with a patch hole diameter of 1 µm and a resistance of 2 ˘ 0.02 MΩ. Cell positioning and sealing parameters were set as follows: positioning pressure ´60 mbar, minimum seal resistance 0.1 GΩ, holding potential ´100 mV, holding pressure ´20 mbar. Whole-cell currents were filtered at 5 kHz and acquired at 25 kHz. Cells expressing hNaV1.7 were maintained with a holding potential of ´100 mV between voltage protocols. Current-voltage (IV) relationships were determined using a family of 500-ms conditioning pulses from ´120 mV to +70 mV in 5-mV steps, followed by depolarization to 0 mV to assess the voltage dependence of fast inactivation (repetition interval: 20 s). Both peak and late current (10 ms post-peak) Toxins 2016, 8, 78 16 of 19 of the IV family were measured. State-dependence was assessed after a 10-min compound incubation (˘OD1 300 nM) to ensure steady-state inhibition for each concentration. In order to assess compound activity at the partially inactivated/open state, a 50-ms depolarization to 0 mV was measured after a conditioning pulse to ´55 mV for 8 s, with a 50-ms recovery period between (repetition interval 12 s). To assess possible interaction with the voltage sensor domain, a triple-pulse protocol was used, comprising two steps to 0 mV for 50 ms separated by a strong depolarization step to +200 mV for 50 ms with 20-ms recovery to the ´100-mV holding potential between each step.

4.11. Data Analysis and Statistics Data were plotted and analyzed using GraphPad Prism (GraphPad Softare, Version 6.0, La Jolla, CA, USA). Statistical significance was defined as p < 0.05 and was determined by the t-test or one-way ANOVA with Dunnett’s post-test, as appropriate. Data are expressed as the mean ˘ standard error of the mean. For electrophysiology experiments, the voltage dependence of steady-state inactivation was calculated by dividing the amplitude of the test current (I) by the maximal current elicited (Io), as defined by: INa = I/Io. The voltage-dependence of activation was derived from IV curves after + normalizing for Na conductance (GNa) using: GNa = INa/(Vmemb ´ Vrev), where Vmemb is the membrane potential and Vrev is the reversal potential. Steady-state kinetic parameters were obtained by fitting the data to Boltzmann equations y = bottom + (top ´ bottom)/(1 + exp((V50 ´ x)/slope)). Sophion QPatch Assay Software (Sophion, Version 5.0, Ballerup, Denmark) was used to determine the time of decay (τ) for inactivating current using the equation for first-order exponential decay: y = y0 ˆ exp(- k ˆ x) + yp, where y is current, x is time, y0 is peak current, yp is the plateau and k is the rate constant. τ was calculated separately for each conditioning potential, defined as 1/k.

Supplementary Materials: The following are available online at www.mdpi.com/2072-6651/8/3/78/s1. Acknowledgments: This work was supported by an Australian Postgraduate Award and a Graduate School International Travel Award (J.R.D.); an Australian Future Fellowship (FT130101215) (I.V.); the German Research Foundation (RE 704/2-1 (AOBJ) 568743) (T.H.); NHMRC Principal Research Fellowships (R.J.L. and P.F.A); an NHMRC Program Grant (569927) (R.J.L. and P.F.A.); and an Australian Research Council Linkage Grant (LP130101131) (J.S.W., S.R.S., R.J.L., I.V.). Author Contributions: J.R.D., J.S.W., R.J.L. and I.V. conceived of and designed the experiments. J.R.D., J.S.W., Z.W., S.R.S. and I.V. performed the experiments. J.R.D. and J.S.W. analyzed the data. T.D., Z.D., K.Z., T.H., C.W., M.A.N. and P.F.A. contributed reagents/materials/analysis tools. J.R.D., J.S.W., R.J.L. and I.V. wrote the paper. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Ahmad, S.; Dahllund, L.; Eriksson, A.B.; Hellgren, D.; Karlsson, U.; Lund, P.E.; Meijer, I.A.; Meury, L.; Mills, T.; Moody, A.; et al. A stop codon mutation in SCN9A causes lack of pain sensation. Hum. Mol. Genet. 2007, 16, 2114–2121. [CrossRef][PubMed] 2. Cox, J.J.; Reimann, F.; Nicholas, A.K.; Thornton, G.; Roberts, E.; Springell, K.; Karbani, G.; Jafri, H.; Mannan, J.; Raashid, Y.; et al. An SCN9A channelopathy causes congenital inability to experience pain. Nature 2006, 444, 894–898. [CrossRef][PubMed] 3. Sun, S.; Cohen, J.C.; Dehnhardt, C.M. Inhibitors of voltage-gated sodium channel Nav1.7: Patent applications since 2010. Pharm. Pat. Anal. 2014, 3, 509–521. [CrossRef][PubMed] 4. Trimmer, J.S.; Cooperman, S.S.; Agnew, W.S.; Mandel, G. Regulation of muscle sodium channel transcripts during development and in response to denervation. Dev. Biol. 1990, 142, 360–367. [CrossRef] 5. Rogart, R.B.; Cribbs, L.L.; Muglia, L.K.; Kephart, D.D.; Kaiser, M.W. Molecular cloning of a putative tetrodotoxin-resistant rat heart Na+ channel isoform. Proc. Natl. Acad. Sci. USA 1989, 86, 8170–8174. [CrossRef][PubMed] 6. Caldwell, J.H.; Schaller, K.L.; Lasher, R.S.; Peles, E.; Levinson, S.R. Sodium channel Na(v)1.6 is localized at nodes of ranvier, dendrites, and synapses. Proc. Natl. Acad. Sci. USA 2000, 97, 5616–5620. [CrossRef] [PubMed] Toxins 2016, 8, 78 17 of 19

7. Catterall, W.A.; Goldin, A.L.; Waxman, S.G. International union of pharmacology. XLVII. Nomenclature and structure-function relationships of voltage-gated sodium channels. Pharmacol. Rev. 2005, 57, 397–409. [CrossRef][PubMed] 8. Bagal, S.K.; Chapman, M.L.; Marron, B.E.; Prime, R.; Storer, R.I.; Swain, N.A. Recent progress in sodium channel modulators for pain. Bioorg. Med. Chem. Lett. 2014, 24, 3690–3699. [CrossRef][PubMed] 9. Murray, J.K.; Ligutti, J.; Liu, D.; Zou, A.; Poppe, L.; Li, H.; Andrews, K.L.; Moyer, B.D.; McDonough, S.I.; Favreau, P.; et al. Engineering potent and selective analogues of GpTx-1, a tarantula venom peptide antagonist of the Na(v)1.7 sodium channel. J. Med. Chem. 2015, 58, 2299–2314. [CrossRef][PubMed] 10. Revell, J.D.; Lund, P.E.; Linley, J.E.; Metcalfe, J.; Burmeister, N.; Sridharan, S.; Jones, C.; Jermutus, L.; Bednarek, M.A. Potency optimization of huwentoxin-IV on hNav1.7: A neurotoxin TTX-s sodium-channel antagonist from the venom of the chinese bird-eating spider selenocosmia huwena. Peptides 2013, 44, 40–46. [CrossRef][PubMed] 11. Schmalhofer, W.A.; Calhoun, J.; Burrows, R.; Bailey, T.; Kohler, M.G.; Weinglass, A.B.; Kaczorowski, G.J.; Garcia, M.L.; Koltzenburg, M.; Priest, B.T. ProTx-II, a selective inhibitor of Nav1.7 sodium channels, blocks action potential propagation in nociceptors. Mol. Pharmacol. 2008, 74, 1476–1484. [CrossRef][PubMed] 12. Hoeijmakers, J.G.; Faber, C.G.; Merkies, I.S.; Waxman, S.G. Painful peripheral neuropathy and sodium channel mutations. Neurosci. Lett. 2015, 596, 51–59. [CrossRef][PubMed] 13. Waxman, S.G.; Merkies, I.S.; Gerrits, M.M.; Dib-Hajj, S.D.; Lauria, G.; Cox, J.J.; Wood, J.N.; Woods, C.G.; Drenth, J.P.; Faber, C.G. Sodium channel genes in pain-related disorders: Phenotype-genotype associations and recommendations for clinical use. Lancet Neurol. 2014, 13, 1152–1160. [CrossRef] 14. Maertens, C.; Cuypers, E.; Amininasab, M.; Jalali, A.; Vatanpour, H.; Tytgat, J. Potent modulation of the voltage-gated sodium channel Nav1.7 by OD1, a toxin from the scorpion odonthobuthus doriae. Mol. Pharmacol. 2006, 70, 405–414. [CrossRef][PubMed] 15. Durek, T.; Vetter, I.; Wang, C.I.; Motin, L.; Knapp, O.; Adams, D.J.; Lewis, R.J.; Alewood, P.F. Chemical engineering and structural and pharmacological characterization of the alpha-scorpion toxin OD1. ACS Chem. Biol. 2013, 8, 1215–1222. [CrossRef][PubMed] 16. Cardoso, F.C.; Dekan, Z.; Rosengren, K.J.; Erickson, A.; Vetter, I.; Deuis, J.R.; Herzig, V.; Alewood, P.F.; King, G.F.; Lewis, R.J. Identiﬁcation and characterization of ProTx-III [mu-TRTX-Tp1a], a new voltage-gated sodium channel inhibitor from venom of the tarantula thrixopelma pruriens. Mol. Pharmacol. 2015, 88, 291–303. [CrossRef][PubMed] 17. McCormack, K.; Santos, S.; Chapman, M.L.; Krafte, D.S.; Marron, B.E.; West, C.W.; Krambis, M.J.; Antonio, B.M.; Zellmer, S.G.; Printzenhoff, D.; et al. Voltage sensor interaction site for selective small molecule inhibitors of voltage-gated sodium channels. Proc. Natl. Acad. Sci. USA 2013, 110, E2724–E2732. [CrossRef][PubMed] 18. Deuis, J.R.; Zimmermann, K.; Romanovsky, A.A.; Possani, L.D.; Cabot, P.J.; Lewis, R.J.; Vetter, I. An animal model of oxaliplatin-induced cold allodynia reveals a crucial role for Nav1.6 in peripheral pain pathways. Pain 2013, 154, 1749–1757. [CrossRef][PubMed] 19. Wilson, M.J.; Yoshikami, D.; Azam, L.; Gajewiak, J.; Olivera, B.M.; Bulaj, G.; Zhang, M.M. Mu-conotoxins that differentially block sodium channels Nav1.1 through 1.8 identify those responsible for action potentials in sciatic nerve. Proc. Natl. Acad. Sci. USA 2011, 108, 10302–10307. [CrossRef][PubMed] 20. Cherki, R.S.; Kolb, E.; Langut, Y.; Tsveyer, L.; Bajayo, N.; Meir, A. Two tarantula venom peptides as potent and differential Na(v) channels blockers. Toxicon 2014, 77, 58–67. [CrossRef][PubMed] 21. Leipold, E.; DeBie, H.; Zorn, S.; Borges, A.; Olivera, B.M.; Terlau, H.; Heinemann, S.H. MuO conotoxins inhibit Nav channels by interfering with their voltage sensors in domain-2. Channels (Austin Tex.) 2007, 1, 253–262. [CrossRef] 22. Catterall, W.A.; Cestele, S.; Yarov-Yarovoy, V.; Yu, F.H.; Konoki, K.; Scheuer, T. Voltage-gated ion channels and gating modiﬁer toxins. Toxicon 2007, 49, 124–141. [CrossRef][PubMed] 23. Motin, L.; Durek, T.; Adams, D.J. Modulation of human Nav1.7 channel gating by synthetic alpha-scorpion toxin OD1 and its analogs. Channels (Austin Tex.) 2015, 10, 139–147. [CrossRef][PubMed] 24. Cheng, X.; Dib-Hajj, S.D.; Tyrrell, L.; Waxman, S.G. Mutation I136V alters electrophysiological properties of the Na(v)1.7 channel in a family with onset of erythromelalgia in the second decade. Mol. Pain 2008, 4, 1. [CrossRef][PubMed] Toxins 2016, 8, 78 18 of 19

25. Estacion, M.; Harty, T.P.; Choi, J.S.; Tyrrell, L.; Dib-Hajj, S.D.; Waxman, S.G. A sodium channel gene SCN9A polymorphism that increases nociceptor excitability. Ann. Neurol. 2009, 66, 862–866. [CrossRef][PubMed] 26. Lampert, A.; Dib-Hajj, S.D.; Eastman, E.M.; Tyrrell, L.; Lin, Z.; Yang, Y.; Waxman, S.G. Erythromelalgia mutation L823R shifts activation and inactivation of threshold sodium channel nav1.7 to hyperpolarized potentials. Biochem. Biophys. Res. Commun. 2009, 390, 319–324. [CrossRef][PubMed] 27. Stadler, T.; O'Reilly, A.O.; Lampert, A. Erythromelalgia mutation Q875Estabilizes the activated state of sodium channel Nav1.7. J. Biol. Chem. 2015.[CrossRef][PubMed] 28. Fertleman, C.R.; Baker, M.D.; Parker, K.A.; Moffatt, S.; Elmslie, F.V.; Abrahamsen, B.; Ostman, J.; Klugbauer, N.; Wood, J.N.; Gardiner, R.M.; et al. SCN9A mutations in paroxysmal extreme pain disorder: Allelic variants underlie distinct channel defects and phenotypes. Neuron 2006, 52, 767–774. [CrossRef] [PubMed] 29. Jarecki, B.W.; Sheets, P.L.; Jackson, J.O.; Cummins, T.R. Paroxysmal extreme pain disorder mutations within the D3/S4-S5 linker of nav1.7 cause moderate destabilization of fast inactivation. J. Physiol. 2008, 586, 4137–4153. [CrossRef][PubMed] 30. Dib-Hajj, S.D.; Estacion, M.; Jarecki, B.W.; Tyrrell, L.; Fischer, T.Z.; Lawden, M.; Cummins, T.R.; Waxman, S.G. Paroxysmal extreme pain disorder M1627K mutation in human Nav1.7 renders DRG neurons hyperexcitable. Mol. Pain 2008, 4, 37. [CrossRef][PubMed] 31. Djouhri, L.; Newton, R.; Levinson, S.R.; Berry, C.M.; Carruthers, B.; Lawson, S.N. Sensory and electrophysiological properties of guinea-pig sensory neurones expressing Nav 1.7 (PN1) Na+ channel alpha subunit protein. J. Physiol. 2003, 546, 565–576. [CrossRef][PubMed] 32. Black, J.A.; Frezel, N.; Dib-Hajj, S.D.; Waxman, S.G. Expression of Nav1.7 in DRG neurons extends from peripheral terminals in the skin to central preterminal branches and terminals in the dorsal horn. Mol. Pain 2012, 8, 82. [CrossRef][PubMed] 33. Rush, A.M.; Cummins, T.R.; Waxman, S.G. Multiple sodium channels and their roles in electrogenesis within dorsal root ganglion neurons. J. Physiol. 2007, 579, 1–14. [CrossRef][PubMed] 34. Meisler, M.H.; Kearney, J.; Escayg, A.; MacDonald, B.T.; Sprunger, L.K. Sodium channels and neurological disease: Insights from SCN8A mutations in the mouse. Neuroscientist 2001, 7, 136–145. [CrossRef][PubMed] 35. Zhu, H.L.; Wassall, R.D.; Takai, M.; Morinaga, H.; Nomura, M.; Cunnane, T.C.; Teramoto, N. Actions of veratridine on tetrodotoxin-sensitive voltage-gated Na currents, Nav1.6, in murine vas deferens myocytes. Br. J. Pharmacol. 2009, 157, 1483–1493. [CrossRef][PubMed] 36. Large, C.H.; Bison, S.; Sartori, I.; Read, K.D.; Gozzi, A.; Quarta, D.; Antolini, M.; Hollands, E.; Gill, C.H.; Gunthorpe, M.J.; et al. The efﬁcacy of sodium channel blockers to prevent phencyclidine-induced cognitive dysfunction in the rat: Potential for novel treatments for schizophrenia. J. Pharmacol. Exp. Ther. 2011, 338, 100–113. [CrossRef][PubMed] 37. Large, C.H.; Kalinichev, M.; Lucas, A.; Carignani, C.; Bradford, A.; Garbati, N.; Sartori, I.; Austin, N.E.; Ruffo, A.; Jones, D.N.; et al. The relationship between sodium channel inhibition and anticonvulsant activity in a model of generalised seizure in the rat. Epilepsy Res. 2009, 85, 96–106. [CrossRef][PubMed] 38. Stevens, M.; Peigneur, S.; Tytgat, J. Neurotoxins and their binding areas on voltage-gated sodium channels. Front. Pharmacol. 2011, 2, 71. [CrossRef][PubMed] 39. Murray, J.K.; Biswas, K.; Holder, J.R.; Zou, A.; Ligutti, J.; Liu, D.; Poppe, L.; Andrews, K.L.; Lin, F.F.; Meng, S.Y.; et al. Sustained inhibition of the Nav1.7 sodium channel by engineered dimers of the domain ii binding peptide gptx-1. Bioorg. Med. Chem. Lett. 2015, 25, 4866–4871. [CrossRef][PubMed] 40. Nielsen, K.J.; Watson, M.; Adams, D.J.; Hammarstrom, A.K.; Gage, P.W.; Hill, J.M.; Craik, D.J.; Thomas, L.; Adams, D.; Alewood, P.F.; et al. Solution structure of mu-conotoxin PIIIA, a preferential inhibitor of persistent tetrodotoxin-sensitive sodium channels. J. Biol. Chem. 2002, 277, 27247–27255. [CrossRef][PubMed] 41. Minett, M.S.; Nassar, M.A.; Clark, A.K.; Passmore, G.; Dickenson, A.H.; Wang, F.; Malcangio, M.; Wood, J.N. Distinct Nav1.7-dependent pain sensations require different sets of sensory and sympathetic neurons. Nat. Commun. 2012, 3, 791. [CrossRef][PubMed] Toxins 2016, 8, 78 19 of 19

42. Deuis, J.R.; Whately, E.; Brust, A.; Inserra, M.C.; Asvadi, N.H.; Lewis, R.J.; Alewood, P.F.; Cabot, P.J.; Vetter, I. Activation of kappa opioid receptors in cutaneous nerve endings by conorphin-1, a novel subtype-selective conopeptide, does not mediate peripheral analgesia. ACS Chem. Neurosci. 2015, 6, 1751–1758. [CrossRef] [PubMed] 43. Zimmermann, K.; Hein, A.; Hager, U.; Kaczmarek, J.S.; Turnquist, B.P.; Clapham, D.E.; Reeh, P.W. Phenotyping sensory nerve endings in vitro in the mouse. Nat. Protoc. 2009, 4, 174–196. [CrossRef] [PubMed]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).