Reverse pharmacogenomics: carbamazepine normalizes activation and attenuates thermal hyperexcitability of sensory neurons due to Nav 1.7 mutation I234T

Abstract

Background and purpose: Pharmacotherapy for pain currently involves trial and error. A previous study on inherited erythromelalgia (a genetic model of neuropathic pain due to mutations in the sodium channel, Na_v 1.7) used genomics, structural modelling and biophysical and pharmacological analyses to guide pharmacotherapy and showed that carbamazepine normalizes voltage dependence of activation of the Na_v 1.7-S241T mutant channel, reducing pain in patients carrying this mutation. However, whether this approach is applicable to other Na_v channel mutants is still unknown.

Experimental approach: We used structural modelling, patch clamp and multi-electrode array (MEA) recording to assess the effects of carbamazepine on Na_v 1.7-I234T mutant channels and on the firing of dorsal root ganglion (DRG) sensory neurons expressing these mutant channels.

Key results: In a reverse engineering approach, structural modelling showed that the I234T mutation is located in atomic proximity to the carbamazepine-responsive S241T mutation and that activation of Na_v 1.7-I234T mutant channels, from patients who are known to respond to carbamazepine, is partly normalized with a clinically relevant concentration (30 μM) of carbamazepine. There was significantly higher firing in intact sensory neurons expressing Na_v 1.7-I234T channels, compared with neurons expressing the normal channels (Na_v 1.7-WT). Pre-incubation with 30 μM carbamazepine also significantly reduced the firing of intact DRG sensory neurons expressing Na_v 1.7-I234T channels. Although the expected use-dependent inhibition of Na_v 1.7-WT channels by carbamazepine was confirmed, carbamazepine did not enhance use-dependent inhibition of Na_v 1.7-I234T mutant channels.

Conclusion and implications: These results support the utility of a pharmacogenomic approach to treatment of pain in patients carrying sodium channel variants.

Linked articles: This article is part of a themed section on Recent Advances in Targeting Ion Channels to Treat Chronic Pain. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v175.12/issuetoc.

Figure 1

Structural modelling of the I234T of human Na_v1.7 channels. (A) Diagram of the human Na_v1.7 channel topology showing the location of I234T, S241T and V400M, all carbamazepine‐responsive mutations, in the domain I (D I). (B) Cytosolic view of the 3D structural model of human Na_v1.7 transmembrane domains. (C) Close‐up cytosolic view of the boxed area of panel (B). I234, S241 and V400 are shown as stick model. VSD, voltage‐sensing domain; PM, pore module. L1/2/3, intracellular loop 1/2/3.

Figure 2

Carbamazepine depolarizes the voltage dependence of activation of the Na_v1.7‐I234T mutant channel. (A, B) Representative current traces recorded from HEK293 cells expressing Na_v1.7‐I234T mutant channels with either (A) DMSO or (B) carbamazepine (CBZ) pre‐incubation. (C) The voltage dependence of activation curves of Na_v1.7‐I234T mutant channels treated with DMSO or carbamazepine were plotted and fitted with Boltzman equation. A depolarizing shift of activation of ~6 mV was observed when Na_v1.7‐I234T mutant channels was pre‐incubated with carbamazepine. (D) The voltage dependence of activation curves of Na_v1.7‐WT channels treated with DMSO or carbamazepine were plotted and fitted with Boltzman equation. No significant difference was found between the curves from Na_v1.7‐WT channels incubated with DMSO and those incubated with carbamazepine.

Figure 3

DRG sensory neurons expressing Na_v1.7‐I234T mutant channels display increased firing. (A–F) Heatmap of representative MEA recordings from DRG sensory neurons expressing Na_v1.7‐I234T (A–C) or WT channels (D–F). The firing frequency of each active electrode is colour‐coded: white/red represents high firing frequency; blue/black represents low firing frequency. Each circle represents an active electrode within an 8 × 8 electrode array. (G) Mean firing frequency of neurons expressing Na_v1.7‐WT channels and Na_v1.7‐I234T mutant channels at 33, 37 and 40°C. *P <0.05, significantly different from WT; I234T, n = 4 experiments with eight rats; WT, n = 5 experiments with 10 rats. (H) Averaged number of active electrodes from neurons expressing Na_v1.7‐WT or Na_v1.7‐I234T channels at these three temperatures. *P <0.05, significantly different from WT.

Figure 4

Carbamazepine reduces the firing of DRG sensory neurons expressing Na_v1.7‐I234T mutant channels. (A–F) Heatmap of a representative MEA recording of DRG sensory neurons expressing Na_v1.7‐I234T with DMSO or carbamazepine (CBZ) pre‐incubation. Carbamazepine produces a pronounced reduction in the number of active electrodes and mean firing frequency. (G) mean firing frequency of neurons expressing Na_v1.7‐I234T channels with DMSO or carbamazepine pre‐incubation at all three temperatures (33, 37 and 40°C). *P <0.05, significantly different from DMSO; n = 6 experiments with 12 rats. (H) Averaged number of active electrodes from neurons expressing Na_v1.7‐I234T channels with DMSO or carbamazepine pre‐incubation at all three temperatures (33, 37 and 40°C). *P <0.05, significantly different from DMSO.

Figure 5

Carbamazepine does not reduce the firing of DRG sensory neurons expressing Na_v1.7‐WT channels significantly. (A–F) Heatmap of a representative MEA recording of DRG sensory neurons expressing Na_v1.7‐WT channels with DMSO or carbamazepine (CBZ) pre‐incubation. Carbamazepine did not significantly affect the firing frequency of these neurons. (G) Mean firing frequency of neurons expressing Na_v1.7‐WT channels with pre‐incubation with DMSO or carbamazepine, at all three temperatures (33, 37 and 40°C). n = 5 experiments with 10 rats. (H) Averaged number of active electrodes from neurons expressing Na_v1.7‐WT channels with DMSO or carbamazepine pre‐incubation at all three temperatures (33, 37 and 40°C).

Figure 6

The effect of carbamazepine is not mediated via a use‐dependent inhibition of Na_v1.7‐I234T mutant channels. The use‐dependent block, defined as the ratio of the peak from the 30th pulse normalized to the peak of the first pulse, at a frequency of 20 Hz recording, is determined for each cell. (A, B) Representative first and last (30th) current traces recorded from HEK293 cells expressing Na_v1.7‐WT channels with either (A) DMSO or (B) carbamazepine (CBZ) pre‐incubation. (C, D) Representative first and last (30th) current traces recorded from HEK293 cells expressing Na_v1.7‐I234T mutant channels with either DMSO (C) or carbamazepine (D) pre‐incubation. (E, F) Use‐dependent block curves of Na_v1.7‐WT and Na_v1.7‐I234T mutant channels at a frequency of 20 Hz recording were plotted. (E) Use‐dependent fall‐off of peak Na_v1.7‐WT current with the pre‐incubation of DMSO or 30 μM carbamazepine.*P <0.05, significantly different from DMSO; n=9. (F) Use‐dependent fall‐off of peak Na_v1.7‐I234T current with the pre‐incubation of DMSO or 30 μM carbamazepine.