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

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.


Introduction
Nine voltage-gated sodium channel subtypes have been described to date (Na V 1.1-Na V 1.9), several of which are implicated as causative contributors to pain. Of particular interest is Na V 1.7, 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 Na V 1.7 an exciting therapeutic strategy for the treatment of a wide range of pain types, including inherited OD1 was previously described as an α-scorpion toxin that enhances peak Na V 1.7 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 Na V 1.7 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 ( Figure 1A,B), consistent with previously observed effects [14]. OD1 (300 nM) enhanced peak current with a leftward shift to more hyperpolarized potentials ( Figure 1C). OD1 (300 nM) resulted in a small, non-significant shift in the V 50 of inactivation for hNav1.7 peak current, while a significant shift in V 50 of inactivation was observed for the late current 10 ms after depolarization ( Figure 1D; V 50 of inactivation: control,´58.54˘0.23 mV; OD1 peak ,´57.78˘0.37; OD1 late ,´52.71˘0.49). Although the V 50 of inactivation is statistically unchanged at peak current, a significant fraction of hNav1.7 channels remain in the open state in the presence of OD1 (300 nM), even at conditioning potentials that completely inactivate untreated channels, as can be seen by the significantly increased slope of the voltage of inactivation ( Figure 1D; k h control = 6.07˘0.20 mV; k h OD1 peak = 10.54˘0. 33; k h OD1 late = 11.45˘0.44; p < 0.05). Consistent with previous reports of mixed α/β toxin pharmacology on Na V 1.4 and Na V 1.6, a significant hyperpolarizing shift V 50 of activation of´12 mV at Na V 1.7 was observed in the presence of OD1 (300 nM) compared to control conditions ( Figure 1E; V 50 of activation: control,´22.47˘0.47; OD1,´34.50˘0.58; p < 0.05). OD1 (300 nM) also delayed fast inactivation at more depolarized membrane potentials ( Figure 1F). voltage of inactivation ( Figure 1D; kh control = 6.07 ± 0.20 mV; kh OD1peak = 10.54 ± 0.33; kh OD1late = 11.45 ± 0.44; p < 0.05). Consistent with previous reports of mixed α/β toxin pharmacology on NaV1. 4 and NaV1.6, a significant hyperpolarizing shift V50 of activation of −12 mV at NaV1.7 was observed in the presence of OD1 (300 nM) compared to control conditions ( Figure 1E; V50 of activation: control, −22.47 ± 0.47; OD1, −34.50 ± 0.58; p < 0.05). OD1 (300 nM) also delayed fast inactivation at more depolarized membrane potentials ( Figure 1F).

OD1 Causes Spontaneous Action Potential Firing in A-and C-Fibers
To assess the effect of pharmacological NaV1.7 activation on A-and C-fibers, we tested OD1 using the mouse skin-saphenous nerve preparation. Consistent with the crucial role of NaV1.7 in regulating excitability, the application of OD1 to the receptive fields of peripheral sensory neurons led to spontaneous firing of action potentials in some fibers, with 57% of A-fibers tested (Figure 2A,B; control 0 ± 0 and OD1 (30 nM) 13 ± 7 action potentials/2 min; n = 7) and 29% of C-fibers tested firing spontaneously ( Figure 2C,D; control 1 ± 0.6 and OD1 (30 nM) 8 ± 5 action potentials/2 min; n = 7) in the presence of OD1. Figure 1. Activity of OD1 in CHO cells heterologously expressing hNa V 1.7 assessed by automated patch clamping. Representative trace of sodium currents (A) before and (B) after addition of 300 nM OD1 elicited by depolarizing steps between´100 and +70 mV in 10-mV increments. The red trace highlights the depolarizing step to 0 mV. OD1 enhanced peak inward current and delayed inactivation, resulting in persistent current. (C) Current-voltage (IV) relationship before and after the addition of OD1 (300 nM). OD1 enhanced peak current with a leftward shift to more hyperpolarized potentials. (D) The voltage dependence of fast inactivation. OD1 significantly shifted the voltage dependence of fast inactivation for the late current 10 ms after depolarization (V 50 : control,´58.54˘0.23 mV; OD1 peak ,´57.78˘0.37; OD1 late ,´52.71˘0.49). (E) Voltage dependence of activation. OD1 shifted the voltage dependence of activation to a more hyperpolarized potential (V 50 : control,´22.47˘0.47; OD1 (300 nM),´34.50˘0.58). (F) Time of decay (τ) for inactivating current plotted versus test potential. OD1 (300 nM) delays fast inactivation at more depolarized membrane potentials. * p < 0.001 compared to the control. Data are presented as the mean˘SEM, n = 7.

OD1 Causes Spontaneous Action Potential Firing in A-and C-Fibers
To assess the effect of pharmacological Na V 1.7 activation on A-and C-fibers, we tested OD1 using the mouse skin-saphenous nerve preparation. Consistent with the crucial role of Na V 1.7 in regulating excitability, the application of OD1 to the receptive fields of peripheral sensory neurons led to spontaneous firing of action potentials in some fibers, with 57% of A-fibers tested (Figure 2A,B; control 0˘0 and OD1 (30 nM) 13˘7 action potentials/2 min; n = 7) and 29% of C-fibers tested firing spontaneously ( Figure 2C,D; control 1˘0.6 and OD1 (30 nM) 8˘5 action potentials/2 min; n = 7) in the presence of OD1.  Figure 2. Effects of NaV1.7 activation by OD1 on A-and C-fibers using the mouse skin-saphenous nerve preparation. (A) OD1 caused spontaneous firing of action potentials in 57% of the A-fibers tested (action potentials/2 min: control, 0 ± 0; OD1 (30 nM), 13 ± 7; n = 7). (B) Action potentials plotted 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 relative to the preceding action potential. (C) OD1 caused spontaneous firing of action potentials in 29% of the C-fibers tested (action potentials/2 min: control, 1 ± 0.6; OD1 (30 nM), 8 ± 5; n = 7). (D) Action potentials plotted as a function of instantaneous firing frequency from a C-fiber before and after perfusion of OD1. Each point represents a single action potential, with the position on the y-axis indicating the frequency relative to the preceding action potential.

Intraplantar Injection of OD1 Causes Spontaneous Pain Behavior in Mice Mediated Through NaV1.7
To assess the effect of OD1 on pain behaviors in mice, we chose the intraplantar route of administration, as this route delivers the peptide directly to the terminals of peripheral sensory neurons in the skin, allowing simple quantification of unilateral pain behaviors while avoiding systemic adverse effects. Intraplantar injection of the NaV1.7 activator OD1 (10-300 nM) led to the development of spontaneous pain behaviors in a concentration-dependent manner, as evidenced by licking, flinching, lifting and shaking of the injected hind paw, consistent with the spontaneous action potential firing seen in A-and C-fibers ( Figure 3A). These spontaneous pain behaviors developed rapidly, occurring immediately after injection, and persisted for up to 40 min after injection of the highest concentration (300 nM). Intraplantar injection of phosphate-buffered saline/0.1% BSA alone caused no spontaneous pain behaviors. No significant changes in mechanical (paw withdrawal force: OD1 (300 nM), 2.5 ± 0.5 g; control, 2.9 ± 0.5 g; n = 4, p > 0.05) or thermal thresholds (time to withdrawal: 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 (300 nM), the time by which spontaneous pain behaviors had resolved. As OD1 at a concentration of 300 nM was found to elicit robust responses, this concentration was used for further behavioral studies.
Since OD1 has activity at NaV1.6 at higher concentrations (EC50 = 30-47 nM) [15] and activation of NaV1.6 can also cause spontaneous pain behaviors [18], we next assessed the relative contribution of NaV1.7 and NaV1.6 to spontaneous pain behaviors induced by OD1 ( Figure 3B). Co-administration of tetrodotoxin (TTX) with OD1 completely reversed spontaneous pain behaviors, confirming the involvement of only TTX-sensitive NaV subtypes (spontaneous pain behaviors/10 min: control, 105 ± 6; TTX (1 μM), 3 ± 2); p < 0.05). Co-administration with the NaV1.6 inhibitor GIIIA [19] reduced 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 Table 1) completely reversed spontaneous pain behaviors, confirming a major role for NaV1.7 in mediating spontaneous pain behaviors induced by Effects of Na V 1.7 activation by OD1 on A-and C-fibers using the mouse skin-saphenous nerve preparation. (A) OD1 caused spontaneous firing of action potentials in 57% of the A-fibers tested (action potentials/2 min: control, 0˘0; OD1 (30 nM), 13˘7; n = 7). (B) Action potentials plotted 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 relative to the preceding action potential. (C) OD1 caused spontaneous firing of action potentials in 29% of the C-fibers tested (action potentials/2 min: control, 1˘0.6; OD1 (30 nM), 8˘5; n = 7). (D) Action potentials plotted as a function of instantaneous firing frequency from a C-fiber before and after perfusion of OD1. Each point represents a single action potential, with the position on the y-axis indicating the frequency relative to the preceding action potential.

Intraplantar Injection of OD1 Causes Spontaneous Pain Behavior in Mice Mediated
Through Na V 1.7 To assess the effect of OD1 on pain behaviors in mice, we chose the intraplantar route of administration, as this route delivers the peptide directly to the terminals of peripheral sensory neurons in the skin, allowing simple quantification of unilateral pain behaviors while avoiding systemic adverse effects. Intraplantar injection of the Na V 1.7 activator OD1 (10-300 nM) led to the development of spontaneous pain behaviors in a concentration-dependent manner, as evidenced by licking, flinching, lifting and shaking of the injected hind paw, consistent with the spontaneous action potential firing seen in A-and C-fibers ( Figure 3A). These spontaneous pain behaviors developed rapidly, occurring immediately after injection, and persisted for up to 40 min after injection of the highest concentration (300 nM). Intraplantar injection of phosphate-buffered saline/0.1% BSA alone caused no spontaneous pain behaviors. No significant changes in mechanical (paw withdrawal force: OD1 (300 nM), 2.5˘0.5 g; control, 2.9˘0.5 g; n = 4, p > 0.05) or thermal thresholds (time to withdrawal: 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 (300 nM), the time by which spontaneous pain behaviors had resolved. As OD1 at a concentration of 300 nM was found to elicit robust responses, this concentration was used for further behavioral studies.
GpTx-1 inhibited the peak current of hNaV1.7 without significantly shifting the voltage dependence of activation ( Figure 4C

CNV1014802 is a Non-Selective Na V Inhibitor
CNV1014802 is reported to be a state-dependent inhibitor of Na V 1.7; however, information on the potency and selectivity profile has not been reported [8]. Surprisingly, CNV1014802 was found to be a relatively non-selective Na V inhibitor, in the FLIPR membrane potential assay, with a rank order of potency (pIC 50˘S EM): hNa V 1.8 (5.25˘0.1) > hNa V 1.4 (5.09˘0.2) > hNa V 1.2 (4.99˘0.2) > hNa V 1.6 (4.84˘0.1) > hNa V 1.3 (4.82˘0.3) > hNa V 1.1 (4.70˘0.2) > hNa V 1.7 (4.58˘0.2) > hNa V 1.5 (4.18˘0.2) (Figure 7A,B; see also Table 1 for IC 50 values and Figure S1 for a sample trace). We were able to confirm the state-dependence reported for CNV1014802, finding a nine-fold shift in the Na V 1.7 IC 50 between closed/resting state inhibition (54 µM) and open/inactivated state inhibition (6.3 µM), with a preference for the open/inactivated state ( Figure 7C). Compared to PF-04856264, the potency of CNV1014802 at Na V 1.7 is 50-fold weaker; however, the presence of OD1 (300 nM) did not significantly affect the potency of CNV1014802 at Nav1.7 in the closed/resting state (IC 50 35 µM) or the open/inactivated state (IC 50 11 µM) (see Figure S2 for a sample trace).

Discussion
Choosing a relevant animal pain model to evaluate the analgesic efficacy of Na V 1.7-selective inhibitors is an essential step in the pre-clinical development of these compounds. We therefore developed a novel mouse model of Na V 1.7-mediated pain based on intraplantar injection of the scorpion toxin OD1 that allows rapid in vivo characterization of the analgesic efficacy of Na V 1.7 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 Na V 1.7 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 Na V 1.7 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 Na V 1.7, 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 Na V 1.4 and Na V 1.6 [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 Na V 1.7 can be correlated to the functional consequences of Na V 1.7 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 Na V 1.7 to more hyperpolarized potentials [24][25][26][27], corresponding to the β-scorpion toxin-like effects of OD1, whereas paroxysmal extreme pain disorder mutations generally delay the fast inactivation [28][29][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. Na V 1.7 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 Na V 1.7, which would result in depolarization closer to resting membrane potential, leading to spontaneous action potential firing. The spontaneous pain behaviors induced by intraplantar injection 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 Na V 1.7 activator, it also has activity at Na V 1.6, albeit at higher concentrations. Given that intraplantar administration of the Na V 1.6 activator Cn2 also causes spontaneous pain behaviors in mice [18], we assessed the relative contribution of Na V 1.7 and Na V 1.6 to spontaneous pain behaviors induced by OD1. While the Na V 1.6 inhibitor GIIIA partially reversed OD1-induced spontaneous pain by 33%, the Na V 1.7 selective inhibitors GpTx-1 and PF-04856264 were effective after intraplantar injection at concentrations unlikely to significantly inhibit Na V 1.6 in vivo, suggesting that the majority of spontaneous pain behaviors induced by OD1 are mediated through Na V 1.7. This was further confirmed in Na V 1.7´{´mice, where the behavioral phenotype was strongly attenuated, with residual responses likely attributable to the Na V 1.6 activity of OD1. Thus, OD1 elicited a robust and long-lasting behavioral response that was mediated predominantly through Na V 1.7, allowing the efficacy of Na V 1.7 inhibitors to be readily assessed for their potential in the treatment of Na V 1.7-mediated pain.
Consistent with its reported activity at Na V 1.7 [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 Na V 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 Na V 1.7 inhibition is required to cause analgesia, although it is reasonable to assume that complete inhibition of Na V 1.7 current may not be required to prevent Na V 1.7-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 IC 50 may be lower than the observed in vitro IC 50 . 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 Na V 1.7. 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 Na V 1.6, 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 Na V 1.7 over Na V 1.6 seen with GpTx-1 is required to have an acceptable therapeutic window for systemic use; however, off-target activity at other Na V isoforms may also contribute to the adverse effects.
In contrast, the aryl sulfonamide PF-04856264 displayed excellent selectivity (>1000 fold) for Na V 1.7 over Na V 1.4-Na V 1.6. 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 (IC 50 300 nM) and electrophysiology (IC 50 130 nM) assays. However, given that PF-04856264 preferentially binds to Na V 1.7 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 Na V 1.7 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 Na V 1.7 in other pathological pain conditions remains to be established, although such a mechanism might imply lack of efficacy in paroxysmal extreme pain disorder, where Na V 1.7 channels are relatively resistant to fast inactivation.
Consistent with previous reports, CNV1014802 also was a state-dependent inhibitor of Na V 1.7 in our hands, although it lacked the Na V subtype selectivity of PF-04856264. Surprisingly, CNV1014802 lacked in vivo analgesic efficacy when delivered locally at a concentration of 1 mM, perhaps reflecting its relatively poor potency at Na V 1.7 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 Na V 1.7 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 Na V 1.7-mediated responses in the FLIPR membrane potential assay by PF-04856264 and CNV1014802 ( Figure S1), although both compounds completely inhibited currents from Na V 1.7 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 Na V 1.7 inhibitors irrespective of their mechanism of action or binding site.
In conclusion, we have established a Na V 1.7-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 Na V 1.7 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 Na V 1.7, species differences need to be considered carefully when using animal models of human disease. Given the relatively high homology between mouse and human Na V isoforms, significant pharmacological differences have rarely been reported. Indeed, PF-04856264 is equipotent at mouse and human Na V 1.7 [17], and the effects of OD1 on rodent Na V 1.7 closely match the effects on human Na V 1.7 [15]. While Na V 1.7 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 to pain, erythromelalgia and paroxysmal extreme pain disorder, it remains to be established if pharmacological inhibition of Na V 1.7 can achieve global graded analgesia.

Synthesis of GpTx-1
Solvents for reversed-phase high-performance liquid chromatography (HPLC) consisted of 0.05% trifluoroacetic acid (TFA)/H 2 O (A) and 90% acetonitrile/0.043% TFA/H 2 O (B). Analytical HPLC was performed on a Shimadzu LC20AT system (Shimadzu, Kyoto, Japan) using a Thermo Hypersil GOLD 2.1ˆ100 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. 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 (1ˆ1.5 min, then 1ˆ4 min). Couplings were performed in DMF using 5 equivalents of Fmoc-amino acid/HBTU/DIEA (1:1:1) relative to resin loading for 2ˆ20 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% H 2 O at room temperature for 2 h. After most of the cleavage solution was evaporated under a stream of N 2 , the product was precipitated and washed with cold Et 2 O and lyophilized from 50% acetonitrile/0.   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 Na V 1.7 knockout mice were kindly provided by Professor John N. Wood (University College London) [41].

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 (10ˆ10ˆ10 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].

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.

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).

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), MgSO 4 (0.69), NaHCO 3 (26.2), NaH 2 PO 4 (1.67), Na-gluconate (9.64), glucose (5.55), sucrose (7.6) and CaCl 2 (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.

FLIPR Membrane Potential Assays
HEK 293 cells stably expressing hNa V 1.1-1.8 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 FLIPR TETRA , cells were incubated a further 5 min before stimulating Na V using veratridine (60 µM, Na V 1.1-Na V 1.7) or deltamethrin (150 µM, Na V 1.8). Changes in membrane potential were assessed using the FLIPR TETRA (excitation 515-545 nm, emission 565-625 nm) every second for 300 s after adding agonists. To quantitate the effect of test compounds on Na V 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%).

Electrophysiology
CHO cells were used for the electrophysiology, because in our experience, they are more robust to clamp compared to HEK cells. CHO hNa V 1.7 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 hNa V 1.7 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 1ˆ10 6 cells/mL in Ex-Cell ACF CHO Medium with 25 mM HEPES and 1ˆGlutamax (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), CaCl 2 (2), MgCl 2 (1), HEPES (10), CdCl 2 (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 pressuré 20 mbar. Whole-cell currents were filtered at 5 kHz and acquired at 25 kHz. Cells expressing hNa V 1.7 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) 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.

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 (I o ), as defined by: I Na = I/I o . The voltage-dependence of activation was derived from IV curves after normalizing for Na + conductance (G Na ) using: G Na = I Na /(V memb´Vrev ), where V memb is the membrane potential and V rev is the reversal potential. Steady-state kinetic parameters were obtained by fitting the data to Boltzmann equations y = bottom + (top´bottom)/(1 + exp((V 50´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 = y 0ˆe xp(-kˆx) + y p , where y is current, x is time, y 0 is peak current, y p is the plateau and k is the rate constant. τ was calculated separately for each conditioning potential, defined as 1/k.