Spider venom-derived peptide induces hyperalgesia in Nav1.7 knockout mice by activating Nav1.9 channels

Abstract

The sodium channels Na_v1.7, Na_v1.8 and Na_v1.9 are critical for pain perception in peripheral nociceptors. Loss of function of Na_v1.7 leads to congenital insensitivity to pain in humans. Here we show that the spider peptide toxin called HpTx1, first identified as an inhibitor of K_v4.2, restores nociception in Na_v1.7 knockout (Na_v1.7-KO) mice by enhancing the excitability of dorsal root ganglion neurons. HpTx1 inhibits Na_v1.7 and activates Na_v1.9 but does not affect Na_v1.8. This toxin produces pain in wild-type (WT) and Na_v1.7-KO mice, and attenuates nociception in Na_v1.9-KO mice, but has no effect in Na_v1.8-KO mice. These data indicate that HpTx1-induced hypersensitivity is mediated by Na_v1.9 activation and offers pharmacological insight into the relationship of the three Na_v channels in pain signalling.

Fig. 1. HpTx1 rescues the pain response in Na_v1.7-KO mice.

a RP-HPLC profile of the venom from the spider H. venatoria. The F3 fraction contains HpTx1 (pink). b Sequence alignment of HpTx1 with several ICK toxins; red lines show the disulfide linkage. c Comparison of nocifensive behaviors (licking or biting) following intraplantar injection of vehicle (10 μl 0.9% saline, n = 6) versus HpTx1 (1 μM or 10 μM in 10 μl saline, n = 6) (two-way ANOVA followed by Tukey’s multiple comparisons test, treatment × genotype: F_(2,30) = 3.447, P = 0.0449; treatment: F_(2,30) = 24.05, P < 0.0001; genotype: F_(1,30) = 23.09, P < 0.0001). d Mechanical response thresholds measured in paws in response to vehicle (black circles, n = 6), 1 μM HpTx1 (yellow squares, n = 6 for fNa_v1.7 mice; n = 5 for Na_v1.7-KO mice) or 10 μM HpTx1 (red triangles, n = 6) injections (two-way ANOVA followed by Tukey’s multiple comparisons test, treatment × genotype: F_(2,29) = 10.52, P = 0.0004; treatment: F_(2,29)  = 54.72, P < 0.0001; genotype: F_(1,29) = 150.2, P < 0.0001). e Latency of paw withdrawal (WD) to a noxious thermal stimulus measured after intraplantar injection of vehicle (black circles, n = 6), 1 μM HpTx1 (yellow squares, n = 6 for fNa_v1.7 mice; n = 5 for Na_v1.7-KO mice) or 10 μM HpTx1 (red triangles, n = 6) (two-way ANOVA followed by Tukey’s multiple comparisons test, treatment ×  genotype: F_(2,29) = 2.857, P = 0.0737; treatment: F_(2,29) = 36.83, P < 0.0001; genotype: F_(1,29) = 17.71, P = 0.0002). f Images of hind paws with Evans blue staining. Ipsi ipsilateral paws, Contra contralateral paws. g Quantification of Evans blue staining in ipsilateral and contralateral hind paws (n = 3, two-way ANOVA followed by Tukey’s multiple comparisons test, treatment × paw: F_(2,12) = 96.1, P < 0.0001; treatment: F_(2,12) = 87.79, P < 0.0001; paw: F_(1,12) = 128.9, P < 0.0001). h Relative thickness of injected hind paws normalized to that of uninjected ones (n = 5 for vehicle, n = 6 for formalin, n = 8 for HpTx1, one-way ANOVA followed by Tukey’s multiple comparisons test: F_(2,16) = 160.9, P < 0.0001). Data are represent the mean ± S.E.M. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001. Exact P (c–e, g, h) are presented in Supplementary Data 1. Source data are provided as a Source Data file. Source Data file.

Fig. 2. HpTx1 activates some small DRG neurons in WT and Na_v1.7-KO mice.

a–d Current-clamp recording shows that HpTx1 enhances the excitability of small (<30 μm) DRG neurons from WT and fNa_v1.7 mice. Bars show significant changes for RMP (a, two-way repeated measures ANOVA followed by Bonferroni’s multiple comparisons test, treatment × genotype: F_(1,39) = 0.02197, P = 0.8829; treatment: F_(1,39) = 22.12, P < 0.0001; genotype: F_(1,39) = 0.9114, P = 0.3456) and rheobase (b, two-way repeated measures ANOVA followed by Bonferroni’s multiple comparisons test, treatment × genotype: F_(1,41) = 0.01839, P = 0.8928; treatment: F_(1,41) = 15.47, P = 0.0003; genotype: F_(1,41) = 0.02329, P = 0.8795), but no effect on AP amplitude (c) in the presence of 0.75 μM HpTx1. d Statistics plots show significant increases in AP spike number in the presence of 0.75 μM HpTx1 (n = 15, two-way repeated measures ANOVA followed by Bonferroni’s multiple comparisons test, treatment × inject current: F_(7,98) = 2.228, P = 0.0382; treatment: F_(1,14) = 7.716, ^#P = 0.0148; inject current: F_(7,98) = 8.916, P < 0.0001). e–g Current-clamp recordings show that HpTx1 increases the excitability of small DRG neurons from Na_v1.7-KO mice. e Bars show significant changes for RMP (left, n = 28, parametric paired two-tailed t test: t₂₇ = 3.0, P = 0.006) and rheobase (middle, n = 29, nonparametric Wilcoxon matched-pairs signed-rank test: P = 0.0003), but no effect for AP amplitude (right, n = 29, parametric paired two-tailed t test: t₂₈ = 1.7, P = 0.093) in the presence of 0.75 μM HpTx1. f AP traces recorded from a representative Na_v1.7-KO mouse DRG neuron before (black) and after (red) the application of HpTx1. The dashed lines indicate 0 mV. g Statistics plots show significant increases in AP spike number in the presence of 0.75 μM HpTx1 (n = 19, two-way repeated measures ANOVA followed by Bonferroni’s multiple comparisons test, treatment × inject current: F_(7,126) = 2.313, P = 0.0298; treatment: F_(1,18) = 17.69, ^###P = 0.0005; inject current: F_(7,126) = 14.2, P < 0.0001). All DRG neurons recorded were held at −53 ± 2 mV. Error bars represent the mean ± S.E.M. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001. Exact P (a, b, d, g) are presented in Supplementary Data 1. Source data are provided as a Source Data file. Source Data file.

Fig. 3. HpTx1 inhibits Na_v1.7 currents and enhances Na_v1.9 activity.

a The dose-dependent inhibition of hNa_v1.7 currents by HpTx1 (n = 7). The inset shows representative current traces in the presence (red) or absence (black) of 2 μM HpTx1. b Voltage-dependent steady-state activation (G/G_max, n = 4) and fast inactivation (I/I_max, n = 7) of hNa_v1.7 are not altered by 1 μM HpTx1. c Representative currents show the effect of 1 μM HpTx1 on TTX-S Na_vs in mouse small DRG neurons. d, e HpTx1 increases TTX-R Na_v currents in mouse small DRG neurons and inhibits their fast inactivation, as shown by representative current traces (d) and current density (e, n = 6, two-way repeated measures ANOVA followed by Bonferroni’s multiple comparisons test, treatment × voltage: F_(6,30) = 9.099, P < 0.0001; treatment: F_(1,5) = 15.41, ^#P = 0.0111; voltage: F_(6,30) = 16.25, P < 0.0001). f Bars show noninactivated components observed in the steady-state inactivation (SSI) curve (at −20 mV) of multiple mouse TTX-R channels in the presence of 0.75 μM HpTx1 (unpaired two-tailed t test, WT mice: t₁₂ = 4.09, P = 0.0015, n = 7; Na_v1.7-KO mice: t₂₀ = 6.447, P < 0.00001, n = 11; Na_v1.8-KO mice: t₂₂ = 5.905, P < 0.00001, n = 12; Nav1.9-KO mice: t₆ = 1.999, P = 0.0925, n = 4). Note that 1 μM TTX was applied in these experiments (d–f). g The dose–response curves for the HpTx1-induced inhibition of the fast inactivation of hNa_v1.9 expressed in ND7/23 cells (n = 5). The inset shows representative current traces (left) and normalized current traces (right) in the absence (black) and presence of 0.75 μM HpTx1 (red). h HpTx1 significantly slows the fast inactivation time of hNa_v1.9 (n = 6 for control, n = 5 for HpTx1, two-way ANOVA followed by Bonferroni’s multiple comparisons test, treatment × voltage: F_(7,66) = 6.386, P < 0.0001; treatment: F_(7,66) = 27.35, ^####P < 0.0001; voltage: F_(1,66) = 411.9, P < 0.0001). i Voltage dependence of the steady-state activation (G/G_max) and inactivation (I/I_max) of hNa_v1.9 for the control (black dots, n = 5 for activation, n = 6 for inactivation) and with 0.75 μM HpTx1 application (red diamonds, n = 5 for activation, n = 9 for inactivation). Data are presented as the mean ± S.E.M. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001, NS not significant. Exact P (e, h-i) are presented in Supplementary Data 1. Source data are provided as a Source Data file. Source Data file.

Fig. 4. HpTx1-evoked pain hypersensitivity relies on Na_v1.9 activity.

a–c Current-clamp recordings show that HpTx1 decreases the membrane excitability of small DRG neurons from Na_v1.9-KO mice. a Bars show no significant changes in RMP (left, n = 29) or AP amplitude (right, n = 25), but a significant increase in rheobase (middle, n = 25, nonparametric Wilcoxon matched-pair signed-rank two-tailed test: P = 0.008) in the presence of 0.75 μM HpTx1. b AP traces recorded from a representative small Na_v1.9-KO DRG neuron before (black) and after (red) application of 0.75 μM HpTx1. The dashed lines indicate 0 mV. c Statistics plots show significant decreases in AP spike number in the presence of 0.75 μM HpTx1 (n = 25, two-way repeated measures ANOVA followed by Bonferroni’s multiple comparisons test, treatment × inject current: F_(7,168) = 8.834, P < 0.0001; treatment: F_(1,24) = 25.49, ^####P < 0.0001; inject current: F_(7,168) = 25.28, P < 0.0001). d Comparison of nocifensive behaviors (licking or biting) following intraplantar injection of vehicle (10 μl 0.9% saline, n = 6) versus HpTx1 (1 μM or 10 μM in 10 μl saline, n = 6) (two-way ANOVA followed by Tukey’s multiple comparisons test, treatment × genotype: F_(2,30) = 8.551, P = 0.0012; treatment: F_(2,30) = 11.04, P = 0.0003; genotype: F_(1,30) = 24.37, P < 0.0001). e Mechanical response thresholds measured in paws in response to vehicle (black circles, n = 6), 1 μM HpTx1 (yellow squares, n = 6) or 10 μM HpTx1 (red triangles, n = 6) injections (two-way ANOVA followed by Tukey’s multiple comparisons test, treatment × genotype: F_(2,30) = 18.68, P < 0.0001; treatment: F_(2,30) = 0.0356, P = 0.9651; genotype: F_(1,30) = 67.3, P < 0.0001). f Latency of WD to noxious heat stimuli measured after intraplantar injection of vehicle (black circles, n = 6), 1 μM HpTx1 (yellow squares, n = 6) or 10 μM HpTx1 (red triangles, n = 6) (two-way ANOVA followed by Tukey’s multiple comparisons test, treatment × genotype: F_(2,30) = 44.54, P < 0.0001; treatment: F_(2,30) = 9.701, P = 0.0006; genotype: F_(1,30) = 113.5, P < 0.0001). All DRG neurons recorded were held at −53 ± 2 mV. Data are presented as the mean ± S.E.M. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001. Exact P (c–f) are presented in Supplementary Data 1. Source data are provided as a Source Data file. Source Data file.

Fig. 5. HpTx1 has no effect on pain responses in Na_v1.8-KO mice.

a–e Current-clamp recordings show the effects of HpTx1 on the excitability of small DRG neurons from Na_v1.8-KO mice and Na_v1.7/Na_v1.8-DKO mice. a RMP (two-way repeated measures ANOVA followed by Bonferroni’s multiple comparisons test, treatment × genotype: F_(1,35) = 5.792, P = 0.0215; treatment: F_(1,35) = 52.44, P < 0.0001; genotype: F_(1,35) = 3.389, P = 0.0741), b rheobase (two-way repeated measures ANOVA followed by Bonferroni’s multiple comparisons test, treatment × genotype: F_(1,35) = 3.465, P = 0.0711; treatment: F_(1,35) = 5.092, P = 0.0304; genotype: F_(1,35) = 2.656, P = 0.1121), and c AP amplitude (Na_v1.8-KO mice, n = 18; Na_v1.7/Na_v1.8-DKO mice, n = 19). d, e HpTx1 has no effect on the AP firing frequency of small DRG neurons from Na_v1.8-KO mice (d, n = 18) or Na_v1.7/Na_v1.8-DKO mice (e, n = 19, two-way repeated measures ANOVA followed by Bonferroni’s multiple comparisons test, treatment × inject current: F_(7,126) = 3.141, P = 0.0043; treatment: F_(1,18) = 4.275, P = 0.0534; inject current: F_(7,126) = 79.17, P < 0.0001). f–h 10 μM HpTx1 (red diamonds) has no effect on nocifensive behaviors (f, n = 6) or mechanical (g, n = 6) or thermal pain (h, n = 6). All DRG neurons recorded were held at −53 ± 2 mV. Data are presented as the mean ± S.E.M. ^*P < 0.05, ^**P < 0.01, ^****P < 0.0001. Exact P (a, b, e) are presented in Supplementary Data 1. Source data are provided as a Source Data file. Source Data file.

Fig. 6. The molecular mechanism of HpTx1 action on Na_v1.9 and Na_v1.7 channels.

a Sequence alignments corresponding to the DIV s3b-s4 region of Na_v subtypes. The highlighted sequences show the regions swapped between Na_v1.8 and Na_v1.9. b Representative current traces from Na_v1.9/1.8 DIV s3b-s4 P1 (top) and Na_v1.8/1.9 DIV s3b-s4 P1 (bottom) chimaera channels in the absence (black) and presence (red) of HpTx1. c Effects of HpTx1 on WT and mutant hNa_v1.9 channels. Dot plots display the effect of 0.75 μM HpTx1 on the peak current (top, n = 14 for WT; n = 4 for T1444L, M1445L, I1446F, and T1448A; n = 5 for L1449I and E1450L; n = 3 for N1451K) and the persistent current (bottom, n = 14 for WT; n = 4 for T1444L, I1446F, T1448A, and N1451K; n = 5 for M1445L, L1449I, and E1450L). Key residues involved in the interaction between HpTx1 and hNa_v1.9 are labeled (one-way ANOVA with Dunnett’s multiple comparison test, I₉₅/I_peak: F_(7,35) = 17.72, P < 0.0001; I/I_max: F_(7,38) = 8.157, P < 0.0001). d (top) Sequence alignments corresponding to the DII s3b-s4 region of Na_v subtypes. The highlighted sequences show the regions swapped between Na_v1.7 and Na_v1.8. Representative current traces from Na_v1.7/1.8 DII s3b-s4 (bottom left) and Na_v1.8/1.7 DII s3b-s4 (bottom right) chimaera channels in the absence (black) or presence of 5 μM HpTx1 (red). e Dose-dependent inhibitory curves show the effect of HpTx1 on WT (n = 7) and mutant hNa_v1.7 channels (n = 4 for F813S, n = 6 for L814A and A815S, n = 3 for D816K, n = 6 for V817K, n = 7 for E818G, n = 4 for E818R, n = 5 for G819S and n = 3 for Na_v1.7/1.8 DII s3b-s4) and the Na_v1.8/1.7 DII s3b-s4 chimaera channel (n = 5). f Bars show the fold changes in IC₅₀ values of HpTx1 for mutant channels compared with that for the WT hNa_v1.7 channel. Data are presented as the mean ± S.E.M. Exact P (c) are presented in Supplementary Data 1. Source data are provided as a Source Data file. Source Data file.