Research ArticlePain

Palmitoylation of δ-catenin promotes kinesin-mediated membrane trafficking of Nav1.6 in sensory neurons to promote neuropathic pain

See allHide authors and affiliations

Sci. Signal.  27 Mar 2018:
Vol. 11, Issue 523, eaar4394
DOI: 10.1126/scisignal.aar4394

Palmitoylation and pain

Chronic pain is associated with inflammation and increased synaptic activity in sensory neurons. Zhang et al. found that induction of the inflammatory cytokine TNF-α after chemotherapy or nerve injury in rats promoted the formation of a complex between the cell adhesion protein δ-catenin, the voltage-gated sodium channel Nav1.6, and the kinesin motor protein KIF3A, which increased the trafficking of the sodium channel to the cell membrane. Formation of this complex was dependent on the palmitoylation of δ-catenin, and inhibiting the activity of palmitoyl acyltransferases prevented the increase in both Nav1.6 surface abundance in sensory neurons and pain sensitivity in rats. These findings reveal potential therapeutic targets for treating chronic, neuropathic pain in patients.

Abstract

Palmitoylation of δ-catenin is critical to synapse plasticity and memory formation. We found that δ-catenin palmitoylation is also instrumental in the development of neuropathic pain. The abundances of palmitoylated δ-catenin and the palmitoyl acyltransferase DHHC3 were increased in dorsal root ganglion (DRG) sensory neurons in rat models of neuropathic pain. Inhibiting palmitoyl acyltransferases or decreasing δ-catenin abundance in the DRG by intrathecal injection of 2-bromopalmitate or shRNA, respectively, alleviated oxaliplatin or nerve injury–induced neuropathic pain in the rats. The palmitoylation of δ-catenin, which was induced by the inflammatory cytokine TNF-α, facilitated its interaction with the voltage-gated sodium channel Nav1.6 and the kinesin motor protein KIF3A, which promoted the trafficking of Nav1.6 to the plasma membrane in DRG neurons and contributed to mechanical hypersensitivity and allodynia in rats. These findings suggest that a palmitoylation-mediated KIF3A/δ-catenin/Nav1.6 complex enhances the transmission of mechanical and nociceptive signals; thus, blocking this mechanism may be therapeutic in patients with neuropathic pain.

INTRODUCTION

Sorting and transportation of intracellular proteins to specific subcellular compartments are highly regulated precise processes, which serve as two of the essential methods to regulate the neuron and brain function in physiological and pathological scenarios. Palmitoylation is a reversible attachment of a 16-carbon acyl chain to cysteine residues through a thioesteric bond, which may enhance the hydrophobicity of targeting proteins and its affinity with intracellular lipid organelles, thus modifying the sorting and trafficking, as well as the function, of intracellular proteins (1, 2). Studies have demonstrated that protein palmitoylation is critically involved in the modulation of neuronal functions during neurodevelopment and in neuropsychiatric disease (3, 4). δ-Catenin, a component of the cadherin-catenin cell adhesion complex, is highly expressed in the brain and contributes to cognitive function. Marked deficits in learning and memory are observed in mice with δ-catenin mutations (5, 6), and palmitoylation of δ-catenin mediates activity-induced synaptic plasticity in central neurons (7). In mammalian cells, palmitoylation is catalyzed by a family of palmitoyl acyltransferases (PATs), each containing a conserved Asp-His-His-Cys (DHHC) motif (8). PATs are highly expressed in neurons (9, 10) and are, like δ-catenin, involved in the modulation of neuronal function. For example, the PAT DHHC5 interacts with the scaffolding protein postsynaptic density-95 (PSD-95), which has a critical role in learning and memory (11), and genetic variation in the gene encoding the PAT DHHC8 is involved in the pathogenesis of schizophrenia (12). Furthermore, the hyperexcitability of sensory neurons contributes to chronic pain (13). However, whether and how palmitoylation of δ-catenin, through its regulation of dorsal root ganglion (DRG) neuronal activity, participates in chronic pain is unclear.

Voltage-gated sodium (Nav) channels mediate the rising phase of action potentials and remarkably regulate the excitability and firing pattern in central and peripheral neurons. To date, nine isoforms of the α subunit (Nav1.1 to Nav1.9) have been identified (14). Among them, Nav1.6 is widely expressed in neurons in the DRG (15) and contributes to the propagation and generation of action potentials (16). Results obtained from knockout and antisense studies support a pivotal role for Nav1.6 in the development of neuropathic pain (17, 18). It is recognized that several Nav isoforms, including Nav1.6, are mainly localized in the cytoplasm of primary sensory neurons and modulation of the sorting and trafficking of intracellular Navs may significantly modify channel function and neuronal excitability (19). Among several known factors, kinesin superfamily motor proteins play a critical role in regulating the intracellular trafficking, transport, and localization of Navs in afferent fibers (19). Kinesin superfamily proteins (KIFs), which contain 15 members [kinesin 1 (KIF1) to KIF14B], mediate the microtubule-dependent transport of cargo proteins in various cell types (20). Previous studies showed that overexpression of KIF5B increased the cell surface and axonal distribution of Nav1.8, and knockdown of KIF5B decreased the current density of Nav1.8 in DRG neurons, indicating that the anterograde axonal transport of Nav1.8 occurs through a mechanism involving motor protein (21). In addition, the expression of mutant KIF3A blocked vesicle transport from the rough endoplasmic reticulum (ER) to the Golgi apparatus in cultured Xenopus cells (22). Notably, reports suggest that palmitoylation of the N-methyl-d-aspartate (NMDA) acid glutamate receptor subunit NR2B in central neurons participates in various physiological contexts, such as cognition and chronic pain (23, 24). Here, we investigated the role of palmitoylation of δ-catenin and its potential interaction with motor proteins KIFs in Nav1.6 membrane trafficking and their functional significance in the pathogenesis of chronic pain.

RESULTS

TNF-α–dependent palmitoylation of δ-catenin in DRG neurons contributes neuropathic pain

Consistent with the previous studies (25), administration of oxaliplatin (4 mg/kg) led to substantial mechanical allodynia, as evidenced by a progressive reduction in the hindpaw withdrawal threshold in rats (fig. S1A). By considering the critical role of palmitoylation of δ-catenin in synaptic plasticity and memory formation (7), we then examined the palmitoylation of δ-catenin in spinal dorsal horn and DRG tissues after oxaliplatin treatment using the acyl-biotin exchange (ABE) assay. δ-Catenin palmitoylation in DRG tissues, but not the spinal dorsal horn, was significantly increased after oxaliplatin treatment (Fig. 1, A and B), which coincided with the time course of mechanical hypersensitivity induced by oxaliplatin. Meanwhile, suppression of δ-catenin palmitoylation by consecutive intrathecal injection of broad-spectrum palmitoylation inhibitor 2-bromopalmitate (2-BP; fig. S1B) significantly reduced mechanical allodynia induced by oxaliplatin (Fig. 1C). Notably, we also showed that lumbar 5 ventral root transection (L5-VRT), a rodent model of neuropathic pain induced by nerve injury, significantly increased the abundance of palmitoylated δ-catenin in the DRG, but not the spinal dorsal horn (Fig. 1, D and E), and the application of 2-BP significantly ameliorated the L5-VRT–induced mechanical allodynia (Fig. 1F). Furthermore, the recombinant adeno-associated virus (AAV) encoding δ-catenin short hairpin RNA (shRNA) (pAAV-CAG-eGFP-U6-shRNA) was intrathecally injected into the subarachnoid space of L4 to L6 spinal cord of rats. Twenty-one days after virus injection, marked green fluorescence in DRG neurons suggested a high efficiency of transfection (fig. S1C). The analysis of polymerase chain reaction (PCR) and Western blotting results indicated that the expression of mRNA encoding δ-catenin (fig. S1D) and the abundance of δ-catenin protein (fig. S1E) in the DRG were significantly reduced on day 21 after δ-catenin shRNA injection. Knockdown of δ-catenin in the DRG significantly attenuated the mechanical allodynia induced by oxaliplatin (Fig. 1G) or L5-VRT (Fig. 1H). Furthermore, we examined the distribution of δ-catenin in DRG. Immunostaining study revealed the expression of δ-catenin in calcitonin gene-related peptide–positive; isolectin B4 (IB4)–positive, small-diameter; and neurofilament 200–positive, large-diameter DRG neurons (Fig. 1I). Collectively, these results demonstrated the critical role of palmitoylation of δ-catenin in DRG neurons in the induction of neuropathic pain induced by oxaliplatin or nerve injury.

Fig. 1 The palmitoylation of δ-catenin contributes to the neuropathic pain.

(A and B) δ-Catenin palmitoylation was examined in the dorsal root ganglion (DRG) (A) and spinal dorsal cord (B) isolated on days 1, 4, 7, and 10 after rats were treated with oxaliplatin (Oxal). n = 6 rats per time point. One-way analysis of variance (ANOVA) followed by Tukey post hoc test, *P < 0.05, **P < 0.01 versus vehicle. (C) Mechanical allodynia in the rats treated with oxaliplatin and intrathecal injection of various doses of palmitoylation inhibitor 2-bromopalmitate (2-BP) for 10 consecutive days before oxaliplatin. n = 12 rats per group. Two-way repeated-measures ANOVA followed by Bonferroni post hoc test, F5,264 = 120.0, P < 0.0001. **P < 0.01 versus vehicle and ##P < 0.01 versus the corresponding oxaliplatin group. (D and E) δ-Catenin palmitoylation in the DRG (D) and spinal dorsal cord (E) from rats isolated on days 1, 5, 10, and 15 after lumbar 5 ventral root transection (L5-VRT). n = 6 rats per time point. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the sham group. (F to H) Effects of intrathecal application of 2-BP (F) or δ-catenin short hairpin RNA (shRNA) (G and H) on mechanical allodynia induced by L5-VRT (F and H) or oxaliplatin (G). n = 12 rats in per group. Two-way repeated-measures ANOVA followed by Bonferroni post hoc test, F2,99 = 235.4 (F), F3,176 = 550.8 (G), F3,132 = 246.5 (H), P < 0.0001. **P < 0.01 versus sham/scramble and ##P < 0.01 versus the corresponding L5-VRT/oxaliplatin group. (I) Immunohistochemistry assessing the colocalization of δ-catenin with neurofilament 200 (NF200), calcitonin gene–related peptide (CGRP), and IB4 (neuronal markers) and Nav1.6 in DRG tissues. n = 3 rats. Scale bar, 50 μm.

The inflammatory cytokine tumor necrosis factor–α (TNF-α) has played an extensive role in the development and maintenance of neuropathic pain (26). Here, we observed the increased abundance of TNF-α in the DRG in the rats that received oxaliplatin injection (Fig. 2A) or L5-VRT (Fig. 2B), which exhibited a time course similar to that of the increase in mechanical hypersensitivity and δ-catenin palmitoylation in these rodents. Notably, intraperitoneal injection of thalidomide, an inhibitor of TNF-α synthesis, blocked the increase in oxaliplatin-induced δ-catenin palmitoylation (Fig. 2C). Meanwhile, incubation of isolated, cultured DRG neurons with TNF-α significantly enhanced the palmitoylation of δ-catenin (Fig. 2D). These results suggest the potential role of TNF-α signaling in the increased δ-catenin palmitoylation in DRG neurons in the setting of neuropathic pain.

Fig. 2 TNF-α mediates the δ-catenin palmitoylation in DRG.

(A and B) Tumor necrosis factor–α (TNF-α) abundance in DRG tissues from rats after oxaliplatin treatment (A) or L5-VRT (B). n = 6 rats per time point. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus controls. (C) Effect of intraperitoneal application of thalidomide before oxaliplatin for 10 consecutive days on oxaliplatin-induced palmitoylation of δ-catenin assessed 10 days after oxaliplatin treatment. n = 6 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus vehicle, ##P < 0.01 versus the corresponding oxaliplatin group. (D) Effect of TNF-α (1 ng/ml for 4 hours) on the abundance of palmitoylated δ-catenin in cultured DRG neurons. n = 6 per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus vehicle.

DHHC3 played a critical role in palmitoylation of δ-catenin

In mammalian cells, palmitoylation is catalyzed by a family of PATs, each containing a conserved DHHC motif. Here, we examined the mRNA expression of all PATs in DRG in rodent models of neuropathic pain. The results showed that the expression of mRNA encoding DHHC3, DHHC8, and DHHC9 was significantly increased in the DRG of rats that received oxaliplatin treatment or L5-VRT (Fig. 3A). To define which DHHC palmitoylates δ-catenin in chronic pain, we then performed co-immunoprecipitation (Co-IP) with δ-catenin antibody to determine the potential interaction between δ-catenin and DHHCs in the DRG. The results showed that the potential binding of δ-catenin with DHHC3, but not DHHC8 or DHHC9, was increased after oxaliplatin treatment (Fig. 3B). Consistently, increased content of δ-catenin was recovered in the immunocomplex precipitated by DHHC3 antibody (Fig. 3C), which suggests a potential interaction between δ-catenin and DHHC3 in the DRG lysates from the rats injected with oxaliplatin. We also observed a significantly increased abundance of DHHC3, with a similar time-dependent increase as seen for the palmitoylation of δ-catenin, in the DRG of rats injected with oxaliplatin (Fig. 3D). In addition, high-resolution images from structured illumination microscopy (SIM) suggested that DHHC3 can colocalize with the δ-catenin in DRG neurons and that the colocalization significantly increased 10 days after oxaliplatin treatment (Fig. 3E and movie S1). Intrathecal injection of DHHC3 shRNA, which decreased the mRNA and protein expression of DHHC3 (fig. S2, A and B), significantly inhibited δ-catenin palmitoylation in the DRG and mechanical allodynia induced by oxaliplatin treatment or L5-VRT (Fig. 3, F to I). Furthermore, incubation with TNF-α significantly increased the abundance of DHHC3 (Fig. 3J) and its colocalization with δ-catenin (Fig. 3K) in cultured DRG neurons. These results demonstrated that the palmitoylation of δ-catenin by DHHC3 in DRG played a critical role in the development of neuropathic pain induced by oxaliplatin or nerve injury.

Fig. 3 DHHC3 was involved in the δ-catenin palmitoylation in the development of neuropathic pain.

(A) Expression of mRNA encoding palmitoyl acyltransferases in DRG tissues after oxaliplatin (day 10) or L5-VRT (day 15) treatment. n = 5 rats per group. Two-tailed, independent Student’s t test. (B) Co–immunoprecipitation (IP) of DHHC3 (Asp-His-His-Cys 3), DHHC8, and DHHC9 with antibody to δ-catenin in lysates from DRG tissues isolated from rats treated with vehicle or oxaliplatin (day 10). n = 6 rats per group. IgG, immunoglobulin G; IB, immunoblot. (C) Significant increase in δ-catenin was immunoprecipitated with DHHC3 antibody on day 10 after oxaliplatin treatment. n = 6 rats per group. (D) Representative blots and histogram showed the up-regulation of DHHC3 induced by oxaliplatin treatment. n = 5 rats per time point. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group. (E) High-resolution image from structured illumination microscopy (SIM) assessing the colocalization between δ-catenin and DHHC3 in DRG tissues of rats treated with vehicle or oxaliplatin (day 10). n = 4 rats per group. Two-tailed, independent Student’s t test, **P < 0.01 versus the vehicle group. White box, zoomed right. Scale bar, 1 μm. (F) δ-Catenin palmitoylation in DRG tissues from rats treated with oxaliplatin (day 10) after intrathecal injection of DHHC3 shRNA. n = 5 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus vehicle and ##P < 0.01 versus the corresponding scramble + oxaliplatin group. (G) δ-Catenin palmitoylation in DRG tissues from L5-VRT rats (day 15) after intrathecal injection of DHHC3 shRNA. n = 5 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the sham group and ##P < 0.01 versus the corresponding scramble + L5-VRT group. (H and I) Mechanical allodynia induced by oxaliplatin treatment (H) or L5-VRT (I) after intrathecal injection of DHHC3 shRNA. n = 12 rats in per group. Two-way repeated-measures ANOVA followed by Bonferroni post hoc test, F3,176 = 463.2 (H), F3,132 = 349.4 (I), P < 0.0001. **P < 0.01 versus the control group and ##P < 0.01 versus the corresponding oxaliplatin or L5-VRT group. (J) Representative blots and histogram assessing DHHC3 abundance in cultured DRG cells cultured with TNF-α (1 ng/ml for 4 hours). n = 7 per group. Two-tailed, independent Student’s t test, **P < 0.01 versus vehicle. (K) High-resolution image assessing the colocalization between DHHC3 and δ-catenin in cultured DRG neurons incubated with vehicle or TNF-α (1 ng/ml for 4 hours). White box, zoomed right. n = 3 per group. Two-tailed, independent Student’s t test, **P < 0.01 versus the vehicle group. Scale bar, 1 μm.

Palmitoylation of δ-catenin mediated the membrane trafficking of Nav1.6

Palmitoylation serves as one of the posttranslational modification to increase protein hydrophobicity and facilitate protein interactions with lipid bilayers, thereby substantially altering the sorting and function of proteins (2). To identify the underlying mechanism by which palmitoylated δ-catenin mediated the chronic pain, we examined the potential interactions between δ-catenin and Nav channels in the DRG (namely Nav1.6, Nav1.7, Nav1.8, and Nav1.9) using the Co-IP method. We found that δ-catenin antibody pulled down a substantial amount of Nav1.6 and Nav1.7 but barely any Nav1.8 or Nav1.9 (Fig. 4A). This potential interaction between δ-catenin and Nav1.6, but not Nav1.7, was significantly attenuated by intrathecal injection of broad-spectrum palmitoylation inhibitor 2-BP (Fig. 4A). Immunostaining revealed that δ-catenin colocalized with Nav1.6 in DRG tissues (Fig. 1I), and high-resolution imaging demonstrated an enhanced colocalization in response to oxaliplatin, which was significantly attenuated by intrathecal injection of broad-spectrum palmitoylation inhibitor 2-BP (Fig. 4B). Furthermore, we observed increased membrane and total abundance of Nav1.6 in DRG in the rats injected with oxaliplatin (Fig. 4C), which exhibited a similar time-dependent increase as that of δ-catenin palmitoylation and mechanical allodynia. Note that the proportion of Nav1.6 trafficking to the membrane was increased in the DRG of rats injected with oxaliplatin (Fig. 4C). To verify the role of membrane-localized Nav1.6 in the chronic pain, we intrathecally injected Nav1.6 shRNA, which induced marked green fluorescence in DRG neurons (fig. S3A) and decreased abundance of Nav1.6 mRNA (fig. S3B) and protein (fig. S3C) 21 days later. Intrathecal injection of Nav1.6 shRNA reduced the surface accumulation of Nav1.6 10 days after oxaliplatin treatment (Fig. 4D) and significantly attenuated oxaliplatin- or L5-VRT–induced mechanical allodynia (Fig. 4, E and F). Next, we further examined whether membrane trafficking of the Nav1.6 isoform in the DRG-mediated chronic pain by applying the reagent brefeldin A (BFA), which generally blocks protein transport from the ER to the Golgi apparatus, thus inhibiting membrane-directed trafficking of intracellular proteins (27). The results showed that consecutive intrathecal injection of BFA (15 nmol/10 μl) inhibited the surface accumulation of Nav1.6 10 days after oxaliplatin treatment (Fig. 4G). Note that 2-BP (Fig. 5A), as well as δ-catenin shRNA (by intrathecal injection) (Fig. 5B), attenuated the surface accumulation of Nav1.6 and the proportion of Nav1.6 trafficking but did not change the total expression of Nav1.6 in DRG neurons 10 days after oxaliplatin treatment. Intrathecal injection of 2-BP (Fig. 5C) or δ-catenin shRNA (Fig. 5D) also decreased the surface accumulation of Nav1.6 in DRG neurons 15 days after L5-VRT. Consistently, incubation with TNF-α significantly increased the membrane accumulation of Nav1.6 in cultured DRG neurons (Fig. 5E). Notably, preincubation of 2-BP significantly disrupted the colocalization of Nav1.6 with δ-catenin (Fig. 5F) and inhibited the membrane accumulation of Nav1.6 (Fig. 5E) in cultured DRG neurons incubated with TNF-α. These results suggest a critical role of palmitoylated δ-catenin in the membrane trafficking of Nav1.6 in DRG neurons in the setting of neuropathic pain.

Fig. 4 Nav1.6 membrane trafficking contributed to the development of neuropathic pain.

(A) Co-IP of Nav1.6, Nav1.7, Nav1.8, and Nav1.9 with antibody to δ-catenin in lysates from DRG tissues isolated from rats treated with vehicle, oxaliplatin (day 10), or 2-BP + oxaliplatin (day 10). n = 6 rats per group. (B) High-resolution image showing the colocalization between Nav1.6 and δ-catenin in DRG tissues from rats treated with vehicle, oxaliplatin (day 10), or 2-BP + oxaliplatin (day 10). n = 4 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding oxaliplatin group. Scale bar, 1 μm. (C) Abundance of membrane and total Nav1.6 in DRG tissues from vehicle or oxaliplatin rats and the proportion of Nav1.6 trafficking in vehicle or oxaliplatin rat’s DRG tissues. n = 6 rats per time point. One-way ANOVA followed by Tukey post hoc test, *P < 0.05, **P < 0.01 versus the vehicle group. TfR, transferrin receptor. (D) Surface expression of Nav1.6 induced by oxaliplatin (day 10) after intrathecal injection of Nav1.6 shRNA. n = 6 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group; #P < 0.05, ##P < 0.01 versus the corresponding oxaliplatin group. (E and F) Mechanical allodynia induced by oxaliplatin (E) and L5-VRT (F) after intrathecal injection of Nav1.6 shRNA. n = 12 rats per group. Two-way repeated-measures ANOVA followed by Bonferroni post hoc test, F3,176 = 343.8 (E), F3, 132 = 392.8 (F), P < 0.0001. **P < 0.01 versus the corresponding control group and ##P < 0.01 versus the corresponding oxaliplatin or L5-VRT group. (G) Surface accumulation of Nav1.6 in DRG tissues of oxaliplatin (day 10) rats after intrathecal injection of brefeldin A (BFA) (15 nmol/10 μl). n = 6 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding oxaliplatin group.

Fig. 5 δ-Catenin palmitoylation-mediated Nav1.6 membrane trafficking.

(A and B) Membrane, total Nav1.6 expression, and proportion of Nav1.6 trafficking in oxaliplatin (on day 10) rat DRG tissues after 2-BP (A) or δ-catenin shRNA (B) intrathecal application. n = 5 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding oxaliplatin group. (C and D) Abundance of Nav1.6 in membrane of L5-VRT (day 15) rat DRG tissues after intrathecal application of 2-BP (C) or δ-catenin shRNA (D). n = 5 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the sham group and ##P < 0.01 versus the corresponding L5-VRT group. (E) Abundance of Nav1.6 in membrane of DRG neurons after incubation of vehicle, TNF-α (1 ng/ml for 4 hours) or 2-BP (100 nM) 20 to 30 min before TNF-α. n = 6 per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the TNF-α group. (F) High-resolution images showing the colocalization between Nav1.6 and δ-catenin in cultured DRG neurons described in (E). n = 4 per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the TNF-α group. Scale bar, 1 μm.

We further defined the modulation of Nav1.6 isoform function by palmitoylated δ-catenin in human embryonic kidney (HEK) 293T cells transfected with Nav1.6. Incubation with TNF-α significantly increased the palmitoylation of δ-catenin (Fig. 6A) and current density of Nav1.6 (meaning that the enhanced membrane accumulation led to the increase of Nav1.6 channel currents) (Fig. 6, B and C) in transfected HEK293T cells, which was attenuated by pretreatment with 2-BP (Fig. 6, A to C). Neither TNF-α nor 2-BP altered the activity or recovery characteristics of Nav1.6 channels in the transfected HEK293T cells (Fig. 6, D to F). Next, we transfected the plenti-CMV-Ctnnd2-mCherry-3FLAG into the Nav1.6-contained HEK293T cells, which was validated by the immunoblotting results that transfection of plenti-CMV-Ctnnd2-mCherry-3FLAG significantly increased δ-catenin abundance (fig. S4). It was observed that palmitoylated δ-catenin abundance (Fig. 6G) and Nav1.6 current density (Fig. 6H) were also increased in HEK293T cells that overexpressed δ-catenin, but these increases were prevented by incubation with 2-BP (Fig. 6, H and I). These results further confirmed the significance of palmitoylated δ-catenin to modulate the function of Nav1.6 isoform.

Fig. 6 Palmitoylation of δ-catenin increased the current density of Nav1.6.

(A) δ-Catenin palmitoylation in the human embryonic kidney (HEK) 293T cells after incubation of vehicle, TNF-α (1 ng/ml for 4 hours), or 2-BP (100 nM) 20 to 30 min before TNF-α. n = 6 per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding TNF-α group. (B) Representative current-voltage relation curve of currents recorded in the HEK293T cells described in (A). (C) Current density of Nav1.6 in cells as described in (A). n = 10 cells per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding TNF-α group. (D to F) Activation (D), inactivation (E), or recovery (F) curves of Nav1.6 in cells as described in (A). (G) Abundance of palmitoylated δ-catenin in HEK293T cells overexpressing δ-catenin. n = 6 per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the mCherry group. (H) Nav1.6 current density in HEK293T cells overexpressing δ-catenin and incubated with 2-BP (100 nM). n = 9 cells in mCherry group and n = 10 cells in Ctnnd2-mCherry and Ctnnd2-mCherry + 2-BP group, respectively. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the mCherry group and ##P < 0.01 versus the corresponding Ctnnd2-mCherry group. (I) δ-Catenin palmitoylation in cells described in (H). n = 5 experiments per group. Two-tailed, independent Student’s t test, **P < 0.01 versus the Ctnnd2-mCherry + 2-BP group.

KIF3A was involved in palmitoylated δ-catenin–mediated membrane trafficking of Nav1.6

A previous report showed that motor protein kinesins played an important role in neuronal cargo trafficking (20). To determine whether motor protein kinesins contributed to palmitoylated δ-catenin–mediated Nav1.6 trafficking, we first examined the interaction between δ-catenin and several motor proteins by Co-IP. The results showed that δ-catenin showed a potent potential interaction with KIF3A and KIF3B whereas δ-catenin barely potential interacted with KIF5A, KIF5B, KIF17, and KIFC2 precipitated by δ-catenin antibody in DRG (Fig. 7A). In addition, consecutive intrathecal injection of 2-BP significantly inhibited the formation of immunocomplex containing δ-catenin and KIF3A, but not KIF3b, in DRG of the rats injected with oxaliplatin (Fig. 7A). Similarly, oxaliplatin markedly increased the contents of δ-catenin in the immunocomplex precipitated by KIF3A antibody in the DRG, and this was also significantly attenuated by intrathecal injection of 2-BP (Fig. 7B). High-resolution imaging further confirmed that the colocalization of δ-catenin and KIF3A was significantly increased in DRG neurons from oxaliplatin-treated rats compared with those from vehicle-treated rats, and this was blocked by intrathecal injection of 2-BP (Fig. 7C). Consistently, in the cultured DRG neurons, incubation with TNF-α significantly increased the colocalization of δ-catenin and KIF3A in the cytoplasm of cultured DRG neurons, which was also prevented by preincubation with 2-BP (Fig. 7D). These results suggested that palmitoylated δ-catenin can function potentially as an adaptor protein for KIF3A in DRG neurons in the setting of neuropathic pain.

Fig. 7 KIF3A was involved in palmitoylated δ-catenin–mediated membrane trafficking of Nav1.6.

(A) DRG tissue lysates from rats treated with vehicle, oxaliplatin, or 2-BP + oxaliplatin (on day 10) were immunoprecipitated with δ-catenin–specific antibody, and the immunocomplex was further analyzed with blotting for KIF3A (kinesin 3A), KIF3B, KIF5A, KIF5B, KIF17, or KIFC2. n = 6 rats per group. (B) Blotting assessment of δ-catenin Co-IP with KIF3A antibody in DRG tissue lysates described in (A). n = 6 rats per group. (C) SIM triple-fluorescence staining of δ-catenin and KIF3A in DRG tissues from rats treated as described in (A). n = 3 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding oxaliplatin group. Scale bar, 1 μm. (D) Colocalization of δ-catenin and KIF3A in cultured DRG neurons after incubation with vehicle, TNF-α (1 ng/ml for 4 hours), or 2-BP (100 nM) 20 to 30 min before TNF-α. n = 4 per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding TNF-α group. Scale bar, 1 μm. (E and F) Co-IP analysis of KIF3A and Nav1.6 in DRG tissue lysates from oxaliplatin-treated rats (on day 10) after intrathecal application of 2-BP (E) or δ-catenin shRNA (F). n = 6 rats per group. (G) SIM triple-fluorescence staining of Nav1.6 and KIF3A in DRG tissues from rats treated as described in (E) and (F). n = 4 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding oxaliplatin group. Scale bar, 1 μm. (H) SIM triple-fluorescence staining of KIF3A and Nav1.6 in cultured neurons after incubation with vehicle, TNF-α (1 ng/ml for 4 hours), or 2-BP (100 nM) 20 to 30 min before TNF-α. n = 4 per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding TNF-α group. Scale bar, 1 μm. (I and J) Abundance of surface Nav1.6 in DRG tissues intrathecally injected with KIF3A shRNA and treated with oxaliplatin (day 10) (I) or subjected to L5-VRT (day 15) (J). n = 5 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the corresponding control group and ##P < 0.01 versus the corresponding oxaliplatin or L5-VRT group. (K and L) Mechanical allodynia in rats induced by oxaliplatin (K) or L5-VRT (L) after intrathecal injection of KIF3A shRNA. n = 12 rats per group. Two-way repeated-measures ANOVA followed by Bonferroni post hoc test, F3,176 = 346.3 (K), F3,132 = 251.6 (L), P < 0.0001. **P < 0.01 versus the corresponding control group and ##P < 0.01 versus the corresponding oxaliplatin or L5-VRT group. (M) Surface abundance of Nav1.6 in cultured DRG neurons transfected with KIF3A shRNA and incubated with TNF-α (1 ng/ml for 4 hours). n = 5 per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding TNF-α group. (N) Nav1.6 current density in HEK293T cells transfected with KIF3A shRNA and incubated with TNF-α (1 ng/ml for 4 hours). n = 10 cells per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the scramble group and ##P < 0.01 versus the corresponding TNF-α group.

Next, we explored whether KIF3A participated in palmitoylated δ-catenin–mediated Nav1.6 trafficking and contributed to the chronic pain. Co-IP results showed that the potential interaction between KIF3A and Nav1.6 was enhanced in the DRG after oxaliplatin treatment, which was abolished by intrathecal injection of 2-BP (Fig. 7E) or δ-catenin shRNA (Fig. 7F). SIM also showed that 2-BP or δ-catenin shRNA prevented the oxaliplatin-induced colocalization of KIF3A and Nav1.6 in DRG neurons (Fig. 7G). In addition, SIM study also showed that incubation with TNF-α increased the colocalization of KIF3A and Nav1.6 in cultured DRG neurons, and this was inhibited by 2-BP (Fig. 7H). These results suggest that δ-catenin palmitoylation is a critical modification to promote interaction (direct or indirect, unclear) between KIF3A and Nav1.6 in the DRG of rats with neuropathic pain.

We further found that intrathecal injection of KIF3A shRNA, knocking down the expression of KIF3A in DRG (fig. S5), significantly prevented oxaliplatin-induced surface accumulation of Nav1.6 in DRG neurons (Fig. 7I) and exhibition of mechanical allodynia–associated behavior in rats (Fig. 7K); intrathecal injection of KIF3A shRNA also prevented these effects of L5-VRT (Fig. 7, J and L). Western blotting analysis also showed that KIF3A shRNA inhibited the increased surface expression of Nav1.6 induced by TNF-α in cultured DRG neurons (Fig. 7M) and ameliorated the increase of Nav1.6 current density induced by TNF-α in HEK293T cells (Fig. 7N). These results suggested the critical role of motor protein KIF3A in δ-catenin palmitoylation-mediated Nav1.6 surface accumulation in the setting of neuropathic pain.

Palmitoylation mediated the formation of a complex comprising KIF3A, Nav1.6, and δ-catenin

By considering the enhanced potential interaction among δ-catenin, KIF3A, and Nav1.6 in DRG neurons in the rats with neuropathic pain, we performed three-color fluorescent staining to DRG neurons. High-resolution images revealed that the colocalization of KIF3A, Nav1.6, and δ-catenin was increased in DRG neurons 10 days after oxaliplatin treatment (Fig. 8A) or in cultured DRG neurons incubated with TNF-α (Fig. 8B and movie S2), and this colocalization was significantly reduced by intrathecal injection or preincubation with 2-BP (Fig. 8, A and B). These results suggest that palmitoylation of δ-catenin was a critical step for the interaction of KIF3A, Nav1.6, and δ-catenin and subsequent Nav1.6 membrane trafficking in response to mechanical sensation.

Fig. 8 Palmitoylation mediates the formation of a complex comprising KIF3A, Nav1.6, and δ-catenin.

(A) SIM triple-fluorescence staining of δ-catenin, KIF3A, and Nav1.6 in DRG tissues from rats treated with vehicle, oxaliplatin, or 2-BP + oxaliplatin (at day 10). n = 4 rats per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding oxaliplatin group. Scale bar, 1 μm. (B) SIM triple-fluorescence staining of δ-catenin, KIF3A, and Nav1.6 in cultured DRG neurons incubated with vehicle, TNF-α (1 ng/ml for 4 hours), or 2-BP (100 nM) 20 to 30 min before TNF-α. n = 5 per group. One-way ANOVA followed by Tukey post hoc test, **P < 0.01 versus the vehicle group and ##P < 0.01 versus the corresponding TNF-α group. Scale bar, 1 μm.

DISCUSSION

The trafficking of Nav1.6 toward plasma membrane is an essential step for its function to mediate the neuronal excitability and repetitive firing, thereby critically involved in the development of neurological disorders including chronic pain (17). Previous studies reported that knockdown of Nav1.6 expression in sensory neurons substantially attenuated the DRG neuronal excitability and mechanical allodynia in the rodent models of chronic pain induced by local inflammation (17) or nerve injury (18). It was also reported that in vitro exposure to oxaliplatin induced remarkable after-potentials in the A-fiber compound action potential response to electrical stimulation at 30°C in human and mouse myelinated axons, which was absent in Nav1.6 knockout mice (28). Our findings in this study confirmed the critical role of membrane-directed trafficking of Nav1.6 in sensory neurons and its induction by inflammation in the development of mechanical hypersensitivity in rodent models of neuropathic pain.

Our data further uncovered a mechanism that enhances Nav1.6 abundance at the membrane, namely through protein-protein interactions promoted by the palmitoylation of δ-catenin. δ-Catenin is a neuron-specific catenin that functions as a bridge between cadherins and PDZ-containing scaffolding proteins, thus substantially regulating the physiological process of learning and memory (5). For example, δ-catenin binds to the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR)–binding protein and glutamate receptor interaction protein and promotes the insertion of AMPAR subunit 2 (GluA2) in hippocampal synapses (29). Palmitoylation regulates the sorting and trafficking of intracellular proteins including ligand- or voltage-gated ion channels [such as the NMDA receptor (30), AMPAR (31), and Nav1.2 (32)] in neurons. Here, we found that increased palmitoylation of δ-catenin in DRG neurons promoted membrane-directed trafficking of the Nav1.6 isoform in the DRG and oxaliplatin- or L5-VRT–induced mechanical allodynia in rats. Considering the potential that the drugs dispersed to both the DRG and spinal cord after intrathecal injection, the present study cannot exclude the possibility that the anti-allodynia effect of intrathecal delivery of the δ-catenin shRNA or the PAT inhibitor 2-BP might partially result from drug action in the spinal dorsal horn, but we, nonetheless, demonstrated their effect to modulate Nav1.6 membrane trafficking and current density in DRG neurons. TNF-α, a proinflammatory cytokine critically involved in the pathogenesis of neuropathic pain (26, 33), was an inducing signal, enhanced by oxaliplatin, to promote increased abundance of the palmitoylase DHHC3 and its interaction/colocalization with δ-catenin in DRG sensory neurons and mechanical allodynia in rats. Hence, these results established that palmitoylation of δ-catenin, potentially by DHHC3, prompted the membrane-directed trafficking of Nav1.6 in sensory neurons and contributed to the development of neuropathic pain.

Our data also suggest a critical role for the motor protein KIF3A in this mechanism. Previous studies demonstrated that various motor proteins of the KIFs exhibit differential function to form intracellular cargo apparatus and regulate protein trafficking, including that of the vascular endothelial growth factor receptor 2 in endothelial cells (34), the glutamate receptor in the ventral nerve cord of Caenorhabditis elegans (worms) (35), and voltage-gated potassium (Kv) channels in mouse hippocampal neurons (36). Specifically, KIF3A mediates the polarized trafficking of Kv1.2 channel to axons (37, 38) and the transport of vesicles containing β-catenin toward the plasma membrane in the developing neuroepithelium (39). Through both direct and indirect association, kinesins also promote the polarized trafficking of cytoplasmic p120 catenin along microtubules toward their plus ends (40). Here, triple-fluorescence SIM analysis demonstrated substantial colocalization of KIF3A, Nav1.6, and δ-catenin in DRG neurons from rats treated with oxaliplatin and was blocked by the palmitoylation inhibitor 2-BP. These data suggest an interaction of δ-catenin, KIF3A, and Nav1.6 that is driven by (presumably δ-catenin’s) palmitoylation to promote the trafficking of Nav1.6 to the membrane in sensory neurons. However, the assays used do not prove direct protein interaction; thus, palmitoylated δ-catenin, KIF3A, and Nav1.6 may interact indirectly through another protein(s). Further work will certainly refine this mechanism that we propose underlies the hyperexcitability of sensory neurons and mechanical hypersensitivity in rats and whether this mechanism is relevant to pain signaling in humans. The findings reveal molecular targets to explore for the treatment of patients with neuropathic pain.

MATERIALS AND METHODS

Animals

Male Sprague-Dawley rats, weighing 200 to 220 g, were obtained from the Institute of Experimental Animals of Sun Yat-sen University. All animals were housed in separated cages, and the room was kept at 24°C temperature and 50 to 60% humidity under a 12:12-hour light/dark cycle and with ad libitum access to food and water. All experimental procedures were approved by the Local Animal Care Committee and were carried out in accordance with the guidelines of the National Institutes of Health (NIH) on animal care and the ethical guidelines.

Reagents and L5-VRT

Antibodies for all methods are listed in table S3. Oxaliplatin (Sigma-Aldrich) was freshly diluted with 5% glucose/H2O as a stock solution of 1 mg/ml and intraperitoneally injected for five consecutive days to induced mechanical allodynia for rats. The L5-VRT model was performed as described previously (41, 42). Briefly, surgery was performed on rats under deep anesthesia after intraperitoneal administration of sodium pentobarbital (50 mg/kg) (Sigma-Aldrich). An L5 hemilaminectomy was performed to expose the L5 nerve root. The ventral root was gently pulled out with fine forceps and transected 2 to 3 mm proximal to the DRG, and a small portion (2 mm) of the root was dissected. In the sham group, an identical operation was performed for exposure of the L5 ventral root, but the nerve was not transected.

Drug administration and behavioral test

Intrathecal injection of 2-BP (10 μl; Sigma-Aldrich) was performed according to our previously described method (25). Briefly, a polyethylene-10 catheter (Smiths) was inserted into the rat’s subarachnoid space through L5 to L6 spinal cord segmental intervertebral space, and the tip of the catheter was located at the L5 spinal segmental level. 2-BP was intrathecally injected 30 min before the first dose of oxaliplatin or L5-VRT surgery once per day.

The 50% withdrawal threshold was assessed using von Frey hairs (Ugo Basile) as described previously (43). Briefly, each animal was loosely restrained in a plastic box on a metal mesh and allowed to acclimate for at least 15 min per day for three consecutive days. On the first testing day, the animals were reintroduced to the testing environment and allowed to accommodate, and then series of von Frey filaments with different bending force were presented alternately from the underneath to the mid-plantar surface of each hind paw. In the absence of a paw withdrawal response to the initially selected hair, a stronger stimulus was presented; in the event of paw withdrawal, the next weaker stimulus was chosen. Optimal threshold calculation by this method requires six responses in the immediate vicinity of the 50% threshold. A nociceptive response was defined as a brisk paw withdrawal or flinching of the paw after von Frey filament application. All the experiments were performed by investigators that were blinded to the treatments/conditions.

Injection of AAV

pAAV-CAG-eGFP-U6-shRNA(Ctnnd2), pAAV-CAG-eGFP-U6-shRNA(Zdhhc3), pAAV-CAG-eGFP-U6-shRNA(Scn8a), pAAV-CAG-eGFP-U6-shRNA(Kif3a), and pAAV-CAG-eGFP-U6-MCS(NC) were intrathecally injected according to our previously described method (44). Twenty-one days after the virus injection, the experiments were performed.

shRNA preparation and screening

Targeted shRNAs were applied to knock down the abundance of Nav1.6, DHHC3, KIF3A, and δ-catenin. Briefly, shRNAs (three each) targeting rat Nav1.6, DHHC3, KIF3A, or δ-catenin mRNA were designed and synthesized by Obio Biotechnology, respectively; sequences are listed as table S1. HBZY-1 cells (Obio Biotechnology) were transfected with 2 μl of shRNA using AAV (Obio Biotechnology). Nav1.6, DHHC3, KIF3A, or δ-catenin expression levels were determined using quantitative PCR (qPCR). Our pilot in vivo experiments showed that intrathecal injection of 10 μl of Nav1.6 shRNA1, DHHC3 shRNA1, KIF3A shRNA1, or δ-catenin shRNA3 markedly suppressed the expression of their respective protein in the DRG. Hence, the synthesized Nav1.6 shRNA1, DHHC3 shRNA1, KIF3A shRNA1, or δ-catenin shRNA3 were chosen for the subsequent experiments.

Western blot

Animals were anesthetized with sodium pentobarbital [50 mg/kg, intraperitoneally (i.p.)] at various time points. The L4 to L6 DRGs were immediately removed and frozen at −80°C. Samples were homogenized on ice in 15 mM tris containing a cocktail of proteinase inhibitors (Roche) and phosphatase inhibitors (Roche). For membrane protein, samples were homogenized on ice with the Plasma Membrane Protein Extraction Kit (Invent Biotechnologies Inc.). Protein samples were separated by gel electrophoresis [SDS–polyacrylamide gel electrophoresis (PAGE)] and transferred onto a polyvinylidene difluoride membrane. The blots were placed in the block buffer for 1 hour at room temperature and incubated with primary antibody. Then, the blots were incubated with horseradish peroxidase (HRP)–conjugated secondary antibody. Enhanced chemiluminesence (ECL) (Pierce) was used to detect the immunocomplex. The band was quantified with a computer-assisted imaging analysis system (ImageJ, NIH). The experimenter was blinded to the treatments.

RNA extraction and real-time qPCR

Total RNA was extracted from the rat L4 to L6 DRGs with TRIzol reagent (Invitrogen). Reverse transcription was performed with oligo-dT primers and Moloney murine leukemia virus reverse transcriptase (Promega) according to the manufacturer’s protocol. PCR primer sequences are listed in table S2. Real-time qPCR was performed with SYBR Green qPCR SuperMix (Invitrogen) and an ABI PRISM 7500 Sequence Detection System. The reactions were set up on the basis of the manufacturer’s protocol. PCR conditions were incubation at 95°C for 3 min followed by 40 cycles of thermal cycling (10 s at 95°C, 20 s at 58°C, and 10 s at 72°C). The relative expression ratio of mRNA was quantified via the 2−ΔΔCT method.

Palmitoylation assay

Palmitoylation of δ-catenin in rat L4 to L6 DRG extracts was detected using the related ABE methods as described in the studies (45, 46). Briefly, DRG tissues were lysed with the lysis buffer [1% IGEPAL CA-630, 20 mM N-ethylmaleimide, 50 mM tris-HCl (pH 7.5), 150 mM NaCl, and 10% glycerol], and the sepharose beads (GE Healthcare) were added into the lysate sample after incubation with δ-catenin antibody for overnight. The precipitate was rinsed with tris buffer and treated with hydroxylamine (HAM) [tris buffer plus 1-m HAM: NH2OH (pH 7.0 to pH 7.2)] for 1 hour at room temperature. Proteins were then rinsed with tris buffer (pH 6.2) and incubated with biotin-BMCC [(1-biotinamido)-4-[4′-(maleimidomethyl) cyclohexanecarboxamido]hexane] solution [tris buffer (pH 6.2) and 1-μm biotin-BMCC] for 1 hour at 4°C. Proteins were washed in tris buffer (pH 6.2) and then analyzed by SDS-PAGE. Palmitoylation was detected using HRP-conjugated streptavidin (Thermo Fisher Scientific) by ECL detection. After the nitrocellulose membrane was stripped, δ-catenin was detected by immunoblotting with ECL detection.

Co-immunoprecipitation

The dissected tissues were lysed in cold IP–radioimmunoprecipitation assay buffer [20 mM tris-HCl (pH 7.5), 150 mM NaCl, 0.1% Triton X-100, 1% sodium deoxycholate, 10 mM NaF, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and leupeptin (1 mg/ml)]. The lysate was centrifuged and took 5% of the supernatant for input sample. The remaining supernatant was precipitated with 1 to 5 μg of antibody at 4°C overnight and afterward protein A/G beads (GE Healthcare) at 4°C for 4 hours. The immunoprecipitated sample was denatured and prepared for immunoblotting.

Immunohistochemistry and SIM

Rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and perfused through the ascending aorta with 4% paraformaldehyde. The L4 to L6 DRGs were removed, postfixed in the same fixative for 3 hours, and transferred to 30% sucrose overnight. Cryostat sections (8 μm) were cut and processed for immunohistochemistry using primary antibodies. After incubation overnight at 4°C, the sections were incubated with secondary antibodies for 1 hour at room temperature. Images were captured using an SIM super-resolution microscope system with 78-nm resolution in the x and y axes and 300-nm resolution in the z axis (Nikon).

DRG neuron preparation

DRG neurons were dissociated via enzyme digestion as previously described with slight modifications (47). Briefly, L4 to L6 DRGs were freed from their connective tissue sheaths and broken into pieces with a pair of sclerotic scissors in Dulbecco’s modified Eagle’s medium/F12 medium (Gibco) under low temperature (in a mixture of ice and water). DRG neurons were plated on glass coverslips coated with poly-l-lysine (Sigma-Aldrich) in a humidified atmosphere (5% CO2, 37°C) after enzymatic and mechanical dissociation. The cells were used for electrophysiological recordings about 4 to 24 hours after plating.

Patch-clamp recording

Whole-cell voltage-clamp recordings of transfected HEK293T cells were made using an EPC 10 amplifier (HEKA Elektronik). The pipette solution contained the following: 145 mM CsCl, 0.1 mM CaCl2, 2 mM MgCl2, 10 mM NaCl, 0.5 mM Na2–guanosine triphosphate, 2 mM Mg–adenosine triphosphate, 1.1 mM EGTA, and 10 mM Hepes (pH 7.2). The external solution contained the following: 140 mM NaCl, 4 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 5 mM d-glucose, and 10 mM Hepes (pH 7.4). Nav current was measured using a stimulus voltage pattern consisting of a 50-ms test pulse from −120 to 100 mV, separated by a 500-ms test interval at −90 mV. The value at each potential was plotted to form current-voltage relation curves. The activation, inactivation, or recovery characteristics of Nav1.6 current were examined generally following the protocols as previously described (48). The data were analyzed using Igor Pro 4.01 (WaveMetrics) and PULSEFIT (HEKA Elektronik).

Data analysis

All data were expressed as means ± SD and analyzed with SPSS 13.0. Western blot, immunohistochemistry, and electrophysiological data were analyzed by two-tailed, independent Student’s t test and one-way analysis of variance (ANOVA), followed by Tukey post hoc test. For behavioral tests, two-way repeated-measures ANOVA, followed by Bonferroni post hoc test, was carried out. Threshold for statistical significance was P < 0.05. Although no power analysis was performed, the sample size was determined according to previous publications in pain-associated behavior and molecular studies.

SUPPLEMENTARY MATERIALS

www.sciencesignaling.org/cgi/content/full/11/523/eaar4394/DC1

Fig. S1. shRNA knockdown of δ-catenin in DRG tissues.

Fig. S2. shRNA knockdown of DHHC3 in DRG tissues.

Fig. S3. shRNA knockdown of Nav1.6 in DRG tissues.

Fig. S4. Overexpression of δ-catenin in HEK293T cells.

Fig. S5. shRNA knockdown of KIF3a in DRG tissues.

Table S1. shRNA nucleotide sequences.

Table S2. PCR primer sequences.

Table S3. Antibodies.

Movie S1. Colocalization between δ-catenin and DHHC3 in oxaliplatin-treated DRG neurons.

Movie S2. Colocalization among δ-catenin, KIF3A and Nav1.6 in DRG neurons incubated with TNF-α.

REFERENCES AND NOTES

Acknowledgments: We thank M. Xie (Sun Yat-sen University, Guangzhou, China) for excellent technical support. Funding: This study was supported by National Natural Science Foundation of China (31671090 and 31470990 to W.-J.X., 81600959 to C.-C.L., 81671088 to S.-L.W, and 81771201 to C.M.), Natural Science Foundation of Guangdong (2016A030308003 to W.-J.X. and 2016A030311045 to C.M.), China Postdoctoral Science Foundation (2016M590833 and 2017T100651 to C.-C.L.), Science and Technology Project in Guangzhou (201607010254 to C.M.), and the Fundamental Research Funds for the Central Universities (15ykjco4b to W.-J.X. and 17ykzd20 to S.-L.W.). Author contributions: X.-L.Z. performed the Western blotting, Co-IP, microscopy, electrophysiology, and palmitoylation assays; isolated the DRG tissues; analyzed the data; and assisted in drafting and revising the manuscript. H.-H.D. performed the Western blotting, Co-IP, drug administration, and behavioral tests and analyzed the data. T.X. performed the Western blotting, shRNA preparation, and screening experiments and analyzed the data. M.L. performed the Western blotting, RNA extraction, and real-time qPCR and analyzed the data. S.-L.W. performed the operations. J.-Y.W., C.-C.L., and S.-B.Z. analyzed the data. C.M. analyzed the data and coordinated the experiments. W.-J.X. conceived the project and drafted and revised the manuscript. Competing interests: The authors declare that they have no competing interests. Data and materials availability: Materials will be made available upon request (contact W.-J.X., xinwj{at}mail.sysu.edu.cn). All data needed to evaluate the conclusions in the paper are present in the paper or the Supplementary Materials.

Correction: Antibody sources are now provided in the Supplementary Materials (table S3). The PDF and HTML (full text) were updated 6 April 2018.

View Abstract

Navigate This Article