Background

Acid-sensing ion channel 3 (ASIC3) upregulation has been reported in dorsal root ganglion neurons after incision and contributes to postoperative nociception. This study hypothesized that upregulation of ASIC3 in incised tissues is induced by nerve growth factor through the phosphoinositide 3–kinase/protein kinase B signaling pathway.

Methods

A plantar incision model was established in adult male and female Sprague–Dawley rats. ASIC3 was inhibited by APETx2 treatment, small interfering RNA treatment, or ASIC3 knockout. Sciatic nerve ligation was performed to analyze ASIC3 transport. A nerve growth factor antibody and a phosphoinositide 3–kinase inhibitor were used to investigate the mechanism by which nerve growth factor regulates ASIC3 expression.

Results

Acid-sensing ion channel 3 inhibition decreased incisional guarding and mechanical nociception. ASIC3 protein levels were increased in skin and muscle 4 h after incision (mean ± SD: 5.4 ± 3.2-fold in skin, n = 6, P = 0.001; 4.3 ± 2.2-fold in muscle, n = 6, P = 0.001). Sciatic nerve ligation revealed bidirectional ASIC3 transport. Nerve growth factor antibody treatment inhibited the expression of ASIC3 (mean ± SD: antibody 2.3 ± 0.8-fold vs. vehicle 4.9 ± 2.4-fold, n = 6, P = 0.036) and phosphorylated protein kinase B (mean ± SD: antibody 0.8 ± 0.3-fold vs. vehicle 1.8 ± 0.8-fold, n = 6, P = 0.010) in incised tissues. Intraplantar injection of nerve growth factor increased the expression of ASIC3 and phosphorylated protein kinase B. ASIC3 expression and incisional pain–related behaviors were inhibited by pretreatment with the phosphoinositide 3–kinase inhibitor LY294002.

Conclusions

Acid-sensing ion channel 3 overexpression in incisions contributes to postoperative guarding and mechanical nociception. Bidirectional transport of ASIC3 between incised tissues and dorsal root ganglion neurons occurs through the sciatic nerve. Nerve growth factor regulates ASIC3 expression after plantar incision through the phosphoinositide 3–kinase/protein kinase B signaling pathway.

Editor’s Perspective
What We Already Know about This Topic
  • The low pH of surgical wounds contributes to postoperative pain

  • Acid-sensing ion channels such as acid-sensing ion channel 3 are structures that sensitize nociceptive neurons after soft tissue incision

What This Article Tells Us That Is New
  • Using a rat model of surgical incision, it was shown that both male and female animals demonstrated less wound area sensitization when acid-sensing ion channel 3 channels were pharmacologically blocked or knocked down using small interfering RNA

  • Anti–nerve growth factor treatment blocked the upregulation of acid-sensing ion channel 3 expression after incision and reduced the pain-related behaviors

Postoperative pain is a significant clinical concern, and appropriate management of postoperative pain is beneficial for short-term effects, such as patient outcomes and clinical expenses.1  Additionally, sufficient treatment of pain after surgery may decrease long-term complications, including postoperative delirium and persistent postoperative pain.2,3  However, 41% of patients still experience moderate-to-severe pain on the day of surgery, and 15% experience such pain on the fourth postoperative day.4 

The mechanisms of postoperative pain may differ from those of inflammatory and neuropathic pain.5  In animal studies, postoperative nociception is associated with peripheral and central sensitization.6  An increase in lactate concentration and a decrease in pH occur in incised tissues corresponding to incisional nociception and are observed in the periphery.7,8  Furthermore, nociceptive afferent fibers innervating incised skin or muscle show enhanced chemosensitivity to lactic acid.9,10  Acid-sensing ion channels (ASICs), which are members of the epithelial sodium channel/degenerin family activated by extracellular protons, might participate in this process.11  Among these channels, acid-sensing ion channel (ASIC3) is the most sensitive to protons and is primarily expressed in the peripheral nervous system as well as some nonneuronal tissues.12  The role of ASIC3 in mediating postoperative nociception has been confirmed through application of a specific ASIC3 blocker, APETx2, in an incision model.13  However, ASIC3 expression, distribution, and transportation in incised tissues and the underlying modulatory mechanism of ASIC3 expression remain poorly understood.

We have previously shown that nerve growth factor is released in incised tissues and contributes to postoperative nociception.14–16  Nerve growth factor can regulate a number of ligand-gated ion channels, including transient receptor potential vanilloid 117  and G protein–coupled receptors.18  A previous study has shown that nerve growth factor increases the ASIC-like current amplitude of dorsal root ganglion neurons and the number of ASIC-expressing neurons.19  Several intracellular signaling pathways, including the p38 mitogen-activated protein kinase and phospholipase C/protein kinase C pathways, are involved in these processes.20  Recent studies have suggested that the phosphoinositide 3–kinase/protein kinase B signaling pathway plays an important role in nociceptive transmission. It has been shown that binding of nerve growth factor to its receptor, tropomyosin-related kinase A (TrkA), activates the phosphoinositide 3–kinase signaling pathway and increases transient receptor potential vanilloid 1 expression.21,22 

In the current study, we hypothesized that upregulation of ASIC3 in incised tissues is induced by nerve growth factor through the phosphoinositide 3–kinase/protein kinase B signaling pathway and participates in postoperative nociception. We established the rat plantar incisional pain model and used APETx2 and ASIC3-knockout rats to confirm the effect of ASIC3 on postoperative nociception. Local ASIC3 expression and distribution were examined. The sciatic nerve was ligated to investigate ASIC3 transport between incised tissues and dorsal root ganglion neurons. Finally, a nerve growth factor antibody, exogenous nerve growth factor, and a phosphoinositide 3–kinase inhibitor were used to investigate the role of nerve growth factor in regulating ASIC3 expression after incision. The findings reported in this study enhance our understanding of the contribution of ASIC3 to postoperative nociception.

Materials and Methods

Animals

All animal procedures were approved by the institutional animal care and use committee of West China Hospital of Sichuan University (Chengdu City, China). ASIC3-knockout rats on a Sprague–Dawley background were generated using a clustered regularly interspaced short palindromic repeats (CRISPR)/clustered regularly interspaced short palindromic repeats–associated protein 9 (Cas9) system by Bioray Laboratories (China). Briefly, two guide RNAs were designed to target the regions upstream of exon 1 and downstream of exon 11 in the ASIC3-201 splice variant. A mixture of purified transcribed Cas9 and guide RNAs was injected into Sprague–Dawley rat embryos. Polymerase chain reaction analysis with the primers 5´-TACATCCCGTGCCCTACTC-3´ and 5´-GATAGCCTCCAAAGACAGC-3´ was used to determine the genotypes of the rats. Wild-type Sprague–Dawley rats were provided by Chengdu Dossy Experimental Animal Co., Ltd. (China). Male and female rats were 7 to 8 wk and 9 to 11 wk old, respectively, with weight 220 to 260 g. The rats were maintained on a 12-h light/dark cycle with ad libitum access to food and water. The animals were randomly assigned to groups using a random number table and were tested or treated in sequential order. All experiments were performed in a double-blinded manner. A statistical power calculation was not conducted before the study. The sample size was based on our previous experience with this design.14,15,23 

Incision and Drug Administration

An animal model of plantar incision was established as previously described by Brennan et al.24  Rats were anesthetized with 2.0 to 2.5% isoflurane (Shanghai Hengrui, China) and prepared in a sterile manner. Then a 1-cm longitudinal incision was made on the plantar aspect of the right hind paw 0.5 cm from the end of the heel. The skin, fascia, and underlying flexor muscle were incised, and the wound was closed with 5-0 nylon sutures after adequate hemostasis. Subsequently, topical antibiotics were administered to prevent wound infection. Sham-incised rats were anesthetized without incision.

To investigate the effects of APETx2 on pain-related behaviors, we used a micropipette to perform intraplantar injection of APETx2 (Abcam, USA; dissolved in 0.9% sodium chloride, 20 µM, 20 µl) or 0.9% sodium chloride (20 µl) around the wound immediately after performing the plantar incision. Pain-related behavior tests were then performed at different time points (n = 6 per group). Intraplantar injection of nerve growth factor antibody (Abcam; 0.1 mg/ml, 50 µl) or vehicle was performed under isoflurane anesthesia 4 h before incision, and the incised tissues were collected 4 h after incision to examine whether nerve growth factor is involved in the upregulation of ASIC3 (n = 6 per group). Intraplantar injection of exogenous nerve growth factor (R&D Systems, USA; 1 µg/10 µl, 50 µl) or vehicle was performed under isoflurane anesthesia, and the incised tissues were collected at different time points (n = 4 per group). In addition, intraplanar injection of the selective phosphoinositide 3–kinase inhibitor LY294002 (Sigma, USA; 100 μM, 100 µl) was performed 1 h before incision. The incised tissues were collected 4 h after incision (n = 4 per group). Pain-related behavior tests were performed 4 h and 1 d after incision (n = 6 to 9 per group) to examine whether phosphoinositide 3–kinase/protein kinase B was associated with the upregulation of ASIC3 and to identify changes in pain-related behaviors induced by incision.

Intraplantar Injection of Small Interfering RNA

Intraplantar injection of a mixture of small interfering RNA and transfection reagent (50 µl, containing 3 µg of small interfering RNA) was performed for 3 d (once per day), after which plantar incision was performed. The ASIC3 small interfering RNA sequences were based on a previous study13  and were as follows: ASIC3, CUACACGCUAUGCCAAGGAdtdt, and a corresponding scrambled sequence, GCUCACACUACGCAGAGAUdtdt. ASIC3 expression in incised tissues was examined by Western blot analysis 1 d after plantar incision.

Sciatic Nerve Ligation

To examine the transport of ASIC3 between incised tissues and dorsal root ganglia, sciatic nerve ligation was performed. Rats were anesthetized with 2.0 to 2.5% isoflurane. After the local site was shaved and disinfected, the skin was incised, and the biceps femoris was bluntly dissected until the sciatic nerve was clearly exposed. A moderately tight ligature was made with a 5-0 silk thread. The sham group underwent blunt dissection without ligation of the sciatic nerve. A plantar incision was made as described above immediately after sciatic nerve ligation to investigate the transport of ASIC3.

Pain-related Behaviors

Pain-related guarding behaviors were assessed by calculating a score based on a previous study.16  Briefly, rats were placed on a small plastic mesh floor, and after the rats had acclimated to the testing environment, the incised paw was viewed with a magnifying mirror. We observed the paw of each animal for a 1-min period every 5 min for 1 h. A score of 0, 1, or 2 was given depending on the paw position during the observation period. A score of 0 was recorded if the wounded area was blanched or distorted by the mesh upon full weight bearing. When the wounded area was just touching the mesh without blanching or distortion, a score of 1 was given, and a score of 2 was given when the wounded area was kept completely off the mesh. The sum of the recorded 12 scores (0 to 24) during the 1-h observation time was calculated for each paw.

Mechanosensitivity was measured by using von Frey filaments (DanMic Global, USA). Rats were placed on an elevated plastic mesh floor covered with a transparent plastic cage and acclimated until they remained on the mesh. Both hind paws of the rats were measured by using calibrated von Frey filaments that were applied adjacent to the wound from underneath the cage. We started by applying a 0.4-g filament and subsequently applied filaments of gradually increasing force until the rat withdrew its paw or a 15-g filament was applied. If the 15-g filament did not induce a withdrawal response, the next filament at 26 g was recorded as the threshold according to the methods of our previous study.23  Each rat was tested three times with an interval of 10 min, and the average of the three results was recorded as the paw withdrawal threshold.

Thermal sensitivity was measured with a heat stimulus test. Rats were individually placed in transparent plastic boxes on an elevated glass floor and allowed to acclimate for 1 h before the experiment. A radiant heat source (Ugo-Basile SRL, Italy) was adjusted to induce paw withdrawal in normal rats within 10 to 20 s, placed under the glass floor in the middle of the incised foot, and kept there until the rat withdrew its foot. The cutoff value in the absence of a response was 30 s. The test was performed three times with an interval of 10 min, and the paw withdrawal latency was recorded as the average of the three trials.

Western Blot Analysis

Plantar skin, muscle, and lumbar 4-6 dorsal root ganglia were collected at the corresponding time points in the different experiments after the rats were decapitated after sodium pentobarbital anesthesia (60 mg/kg) and then transferred to -80°C until use. Radioimmunoprecipitation assay lysis buffer (Beyotime, China) containing protease inhibitors was used to extract total protein after the samples were pulverized, and then the protein concentration was quantified using a BCA Protein Assay Kit (Beyotime) according to the instructions.

Twenty micrograms of protein was separated by 10% polyacrylamide gel electrophoresis and then transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories, USA). To reduce nonspecific binding, the membranes were incubated in Tris-buffered saline/Tween-20 containing 5% nonfat milk at room temperature for 2 h. Subsequently, the membranes were incubated with a rabbit anti–ASIC3 polyclonal antibody (1:2,000, GeneTex), a rabbit anti–phospho–Akt monoclonal antibody (1:2,000, Cell Signaling Technology, USA) and a rabbit anti–Akt monoclonal antibody (1:1,000, Cell Signaling Technology) at 4°C overnight and then with a peroxidase-conjugated goat anti–rabbit IgG secondary antibody (1:5,000, Proteintech, China) for 2 h. Immunoreactivity was visualized with Clarity Western ECL Substrate (Bio-Rad Laboratories) and an Amersham Imager 600 (General Electric, USA) according to the manufacturer’s instructions. ImageJ software (National Institutes of Health, USA) was used for densitometry measurements. Actin was detected as a loading control for every sample.

Quantitative Real-time Polymerase Chain Reaction

Sham-incised or incised skin and muscle were obtained at different time points after the rats were decapitated after sodium pentobarbital anesthesia (60 mg/kg). Total RNA was extracted using a TRIzol RNA extraction kit (Thermo Fisher Scientific, USA) according to the manufacturer’s instructions. After the RNA quantity and quality were determined, complementary DNA was prepared with a RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific) using 2 μg of RNA. Twenty micrograms of total complementary DNA was used for quantitative real-time polymerase chain reaction with specific primers and SYBR Green Supermix (Bio-Rad Laboratories) according to the manufacturer’s protocol. A Bio-Rad CFX96 Real-Time RT-PCR system (Bio-Rad Laboratories) was used to analyze ASIC3 messenger RNA levels. The following primers were used: ASIC3 forward (5′-CACCCAATGACTTGCACTGG-3′), ASIC3 reverse (5′-TAGGCAGCATGTTCAGCAGG-3′), glyceraldehyde-3-phosphate dehydrogenase forward (5′-GACATGCCGCCTGGAGAAAC-3′), and glyceraldehyde-3-phosphate dehydrogenase reverse (5′-AGCCCAGGATGCCCTTTAGT-3′). Thermal cycling was performed with an initial incubation of 5 min at 95°C for polymerase activation, 40 cycles of polymerase chain reaction consisting of denaturation at 95°C for 15 s and annealing and extension at 60°C for 1 min, and melting curve analysis for determination of amplicon specificity. The sample expression levels were normalized to the level of the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase as an internal control and analyzed by the standard 2-ΔΔCT method. Samples without the complementary DNA template were used as controls.

Immunohistochemistry

After undergoing anesthesia with sodium pentobarbital (60 mg/kg), rats were transcardially perfused with 0.1 M phosphate-buffered saline containing heparin followed by 4% paraformaldehyde. Skin, deep tissue, and flexor muscle samples were obtained from sham-incised and incised hind paws and then postfixed in the same precooled fixative overnight. The samples were incubated with increasing concentrations of sucrose (20 to 30%), embedded in Tissue-Tek compound (Sakura Finetek, Japan), and rapidly frozen in liquid nitrogen. Consecutive sections (12 μm) were prepared and used for subsequent analysis or were frozen at –80°C until use. The sections were refixed in 4% paraformaldehyde for 10 min; then, endogenous peroxidase was inactivated by incubating the sections with 30% methanol containing 0.3% hydrogen peroxide (v/v) for 45 min. The sections were blocked with 2% horse serum and then incubated with an anti–ASIC3 rabbit polyclonal antibody (1:400, GeneTex, USA) at 4°C overnight. After six washes in phosphate-buffered saline, the sections were incubated with secondary antibodies at room temperature for 2 h. ASIC3 expression was visualized by using ABC kits (Vector Laboratories, USA) according to the manufacturer’s instructions. Sections without primary or secondary antibodies were processed as controls.

Immunofluorescence

To examine the subsets of ASIC3-immunoreactive neurons in dorsal root ganglia, we performed immunofluorescence double staining for ASIC3 and calcitonin gene-related peptide (CGRP), isolectin B4 (IB4), or neurofilament 200 (NF200). Samples were procured, and sections were obtained as described above. The sections were incubated in 0.1 M phosphate-buffered saline containing 10% goat serum and 0.3% Triton-100 for 2 h to block nonspecific binding and then with ASIC3 and CGRP, IB4, or NF200 primary antibodies overnight for double labeling. The following primary antibodies were used: an anti–ASIC3 rabbit polyclonal antibody (1:400, GeneTex), an anti–CGRP mouse monoclonal antibody (1:1,000, Abcam), an isolectin GS–IB4 antibody (1:500, Thermo Fisher Scientific), and an anti–hypophosphorylated neurofilament H antibody (N52; 1:400, Abcam). After six washes in phosphate-buffered saline, the sections were incubated with secondary antibodies at room temperature for 2 h. The following secondary antibodies were used: goat anti–rabbit IgG heavy & light chain (Alexa Fluor 594–conjugated; 1:1,000, Abcam), preadsorbed goat anti–mouse IgG1 heavy chain (FITC–conjugated; 1:500, Abcam), and goat anti-mouse IgG heavy & light chain (FITC–conjugated; 1:1,000, Abcam). Sections without primary or secondary antibodies were processed as controls.

Statistical Analysis

All data are expressed as the mean ± SD. There were no missing data, and outliers were not tested. The Shapiro–Wilk test was performed to check the normality of the data. At P > 0.05, the data were assumed to be normally distributed and thus eligible for a parametric test. Otherwise, for data that were not normally distributed, the Mann–Whitney U test was used. For Western blot analysis and quantitative real-time polymerase chain reaction, ANOVA followed by Dunnett’s post hoc test for paired comparisons was used to determine the differences between means (incision vs. sham group). A two-tailed independent-samples t test was performed to compare the percentage of ASIC3-positive neurons colocalized with the indicated markers between the two groups. Pain-related behavior data were analyzed with one-way ANOVA or two-way repeated-measures ANOVA followed by Tukey’s post hoc test to compare the treatment or knockout group with the corresponding vehicle or sham group. All statistical analyses were performed by using Prism version 7.04 (GraphPad Software, Inc., USA). All effects were considered statistically significant at P < 0.05.

Results

The primary experimental outcome was pain-related behavioral changes. The secondary experimental outcomes included ASIC3 expression level and location and protein kinase B activation.

ASIC3 Contributes to Postoperative Nociception

Compared with vehicle administration, intraplantar APETx2 administration led to a statistically significant decrease in the cumulative pain score (11.2 ± 2.3 vs. 8.8 ± 3.1, n = 6, P = 0.004) 4 h after incision (fig. 1A) and increased the withdrawal threshold to von Frey filaments from 3.4 ± 1.1 g to 6.5 ± 2.6 g (n = 6, P = 0.021) at postoperative hour 4 and from 5.5 ± 2.9 g to 9.7 ± 1.7 g (n = 6, P = 0.001) on postoperative day 1 (fig. 1B). The latency of withdrawal from radiant heat after incision was not significantly greater after APETx2 administration than after vehicle administration (n = 6, P = 0.999 at postoperative hour 4, P = 0.353 on postoperative day 1, fig. 1C).

Fig. 1.

Effects of APETx2 treatment and ASIC3 knockout on incisional pain-related behaviors. Plantar incision increased the cumulative pain score (A and D), decreased the withdrawal threshold to von Frey filaments (B and E), and decreased the withdrawal latency to heat (C and F). Compared with vehicle injection, intraplantar administration of APETx2 decreased the cumulative pain score at postoperative hour 4 (A), increased the withdrawal threshold at postoperative hour 4 and on postoperative day 1 (B), and had no effect on withdrawal latency to heat (C). Compared to wild-type rats, ASIC3-knockout rats exhibited a decreased cumulative pain score at postoperative hour 4 and on postoperative day 1 (D) and increased withdrawal thresholds at postoperative hour 4 and on postoperative days 1 and 2 (E). There was no difference in withdrawal latency to heat between wild-type and ASIC3-knockout rats after incision (F). The data are expressed as the mean ± SD. n = 6 per group; #P < 0.05, ##P < 0.01, ###P < 0.001 versus sham; *P < 0.05, **P < 0.01, ***P < 0.001 versus vehicle or wild-type; as determined by one-way ANOVA or two-way repeated-measures ANOVA followed by Tukey’s post hoc test. ASIC3, acid-sensing ion channel 3.

Fig. 1.

Effects of APETx2 treatment and ASIC3 knockout on incisional pain-related behaviors. Plantar incision increased the cumulative pain score (A and D), decreased the withdrawal threshold to von Frey filaments (B and E), and decreased the withdrawal latency to heat (C and F). Compared with vehicle injection, intraplantar administration of APETx2 decreased the cumulative pain score at postoperative hour 4 (A), increased the withdrawal threshold at postoperative hour 4 and on postoperative day 1 (B), and had no effect on withdrawal latency to heat (C). Compared to wild-type rats, ASIC3-knockout rats exhibited a decreased cumulative pain score at postoperative hour 4 and on postoperative day 1 (D) and increased withdrawal thresholds at postoperative hour 4 and on postoperative days 1 and 2 (E). There was no difference in withdrawal latency to heat between wild-type and ASIC3-knockout rats after incision (F). The data are expressed as the mean ± SD. n = 6 per group; #P < 0.05, ##P < 0.01, ###P < 0.001 versus sham; *P < 0.05, **P < 0.01, ***P < 0.001 versus vehicle or wild-type; as determined by one-way ANOVA or two-way repeated-measures ANOVA followed by Tukey’s post hoc test. ASIC3, acid-sensing ion channel 3.

Acid-sensing ion channel 3–knockout rats were used to further examine the role of ASIC3 in incisional nociception. Compared to wild-type rats, ASIC3-knockout rats showed a statistically significant decrease in cumulative pain score at postoperative hour 4 and on postoperative day 1 (n = 6, P < 0.001; fig. 1D). The withdrawal threshold of ASIC3-knockout rats to von Frey filaments was higher than that of wild-type rats at postoperative hour 4 (n = 6, P = 0.001), on postoperative day 1 (n = 6, P < 0.001), and on postoperative day 2 (n = 6, P < 0.001; fig. 1E). Statistically significant differences in the withdrawal latency to radiant heat after incision were not observed between ASIC3-knockout and wild-type rats (n = 6, P = 0.215 at postoperative hour 4; n = 6, P = 0.748 on postoperative day 1, P = 0.881 on postoperative day 2; fig. 1F).

ASIC3-Knockout Rat Generation and ASIC3 Antibody Specificity

A CRISPR/Cas9 system was used to generate ASIC3-knockout rats. The rat ASIC3 gene contains 11 coding exons, which were all deleted by two guide RNAs targeting exon 1 and exon 11 (fig. 2A). Heterozygous male and female rats were intercrossed to obtain homozygous ASIC3-knockout rats, and their genotypes were determined by assessment of polymerase chain reaction amplification products. Homozygous, heterozygous, and wild-type genotypes were obtained (fig. 2B). Incised tissues at 4 h after plantar incision and slices of lumbar 5 dorsal root ganglion were obtained from knockout and wild-type rats and subjected to Western blot analysis and immunofluorescence staining to confirm the specificity of the ASIC3 primary antibody. We found that the predicted ASIC3 bands were absent in knockout rats (fig. 2C) and observed ASIC3-positive neurons in wild-type rats but not in knockout rats (fig. 2D).

Fig. 2.

ASIC3-knockout rat generation and ASIC3 antibody specificity. (A) Schematic of the guide RNAs targeting the ASIC3 loci. (B) Gels used for genotype identification after polymerase chain reaction amplification. The upper band represents the amplification products of the wild type (+/+), and the lower band represents the amplification products of the knockout (–/–). Gels containing both the upper and lower bands represent heterozygotes (+/–). (C) The tissues of two knockout rats and two wild-type rats were subjected to Western blot analysis of ASIC3 four hours after plantar incision. Two bands were found in the incised tissues from the wild-type rats, with one corresponding to the predicted molecular weight of ASIC3 (approximately 59 kd); the other was a nonspecific band. Although ASIC3 bands were not observed in knockout rats, nonspecific bands were still observed, suggesting that the predicted bands were ASIC3. (D) The specificity of the ASIC3 antibody was examined by immunofluorescence staining of lumbar 5 dorsal root ganglia from wild-type and knockout rats. ASIC3-positive neurons were observed in wild-type rats but not in knockout rats. Scale bar: 50 μm. ASIC3, acid-sensing ion channel 3; bp, base pair; Cas9, clustered regularly interspaced short palindromic repeats–associated protein 9; kd, kilo dalton.

Fig. 2.

ASIC3-knockout rat generation and ASIC3 antibody specificity. (A) Schematic of the guide RNAs targeting the ASIC3 loci. (B) Gels used for genotype identification after polymerase chain reaction amplification. The upper band represents the amplification products of the wild type (+/+), and the lower band represents the amplification products of the knockout (–/–). Gels containing both the upper and lower bands represent heterozygotes (+/–). (C) The tissues of two knockout rats and two wild-type rats were subjected to Western blot analysis of ASIC3 four hours after plantar incision. Two bands were found in the incised tissues from the wild-type rats, with one corresponding to the predicted molecular weight of ASIC3 (approximately 59 kd); the other was a nonspecific band. Although ASIC3 bands were not observed in knockout rats, nonspecific bands were still observed, suggesting that the predicted bands were ASIC3. (D) The specificity of the ASIC3 antibody was examined by immunofluorescence staining of lumbar 5 dorsal root ganglia from wild-type and knockout rats. ASIC3-positive neurons were observed in wild-type rats but not in knockout rats. Scale bar: 50 μm. ASIC3, acid-sensing ion channel 3; bp, base pair; Cas9, clustered regularly interspaced short palindromic repeats–associated protein 9; kd, kilo dalton.

ASIC3 Expression Profiles in Incised Skin and Muscle

Acid-sensing ion channel 3 expression was increased at both the messenger RNA and protein levels in skin and muscle after plantar incision (fig. 3). Quantitative real-time polymerase chain reaction showed that ASIC3 messenger RNA levels were increased 6.1 ± 5.2-fold in skin (n = 6, P = 0.044; fig. 3A) and 8.1 ± 5.0-fold in muscle (n = 6, P = 0.005; fig. 3B) 4 h after incision in the incision group compared with the sham group and returned to baseline levels within 1 d. ASIC3 protein expression was assessed by Western blot analysis. A band corresponding to the predicted molecular mass of 59 kd was detected in sham-operated and incised skin (fig. 3C, top) and muscle (fig. 3D, top). ASIC3 levels were approximately 5.4 ± 3.2-fold (n = 6, P = 0.001) higher at postoperative hour 4 and 6.1 ± 3.1-fold (n = 6, P < 0.001) higher on postoperative day 1 in incised skin than in sham-operated skin (fig. 3C, bottom). The expression of ASIC3 was 4.3 ± 2.2-fold higher (n = 6, P = 0.001) at postoperative hour 4, 5.7 ± 2.3-fold higher (n = 6, P < 0.001) on postoperative day 1, and 3.8±1.4-fold higher (n = 6, P = 0.006) on postoperative day 2 in incised muscle than in sham-operated muscle (fig. 3D, bottom).

Fig. 3.

Expression and distribution of ASIC3 in skin and muscle after incision. ASIC3 messenger RNA levels in skin (A) and muscle (B) were increased 4 h after plantar incision. The fold changes relative to the levels in the sham group after normalization to glyceraldehyde-3-phosphate dehydrogenase are shown. Western blot analysis was performed to determine the time course of ASIC3 expression in skin (C) and muscle (D) after plantar incision (top). The quantified normalized ASIC3 levels are shown in the bottom panel. ASIC3 levels increased in skin (C) and muscle (D) from 4 h after incision, peaked on day 1, and returned to baseline on day 7 after incision. The fold changes relative to the levels in the sham group after normalization to actin are shown. (E) Schematic of different areas of the skin around the incision. Samples were collected from area A (around the incision) and area B (1 mm outside area A). (F) Representative Western blot results and the quantified ASIC3 expression levels in different areas of incised skin on postoperative day 1 are shown. (G) Immunohistochemistry did not show ASIC3 immunoreactivity in sham-operated skin (a1) or muscle (b1), but ASIC3-like staining (arrows) was detected in nerve-like structures in skin (a2) and muscle (b2) on postoperative day 1. Higher-magnification views of the rectangular areas in a2 and b2 are shown in a3 and b3. Scale bars: 100 μm for a1 and b1; 50 μm for b1 and b2; 25 μm for a3 and b3. (H) Immunofluorescence staining showed that ASIC3 colocalized with calcitonin gene-related peptide (a3, arrows), isolectin B4 (b3, arrows), and neurofilament 200 (c3, arrows) in nerve fibers in cutaneous axons 1 d after incision. Scale bars: 25 μm. The data are shown as the mean ± SD. n = 6 per time point; *P < 0.05, **P < 0.01, ***P < 0.001 versus sham, as determined by one-way ANOVA followed by Dunnett’s test. ASIC3, acid-sensing ion channel 3; CGRP, calcitonin gene-related peptide; IB4, isolectin; NF200, neurofilament.

Fig. 3.

Expression and distribution of ASIC3 in skin and muscle after incision. ASIC3 messenger RNA levels in skin (A) and muscle (B) were increased 4 h after plantar incision. The fold changes relative to the levels in the sham group after normalization to glyceraldehyde-3-phosphate dehydrogenase are shown. Western blot analysis was performed to determine the time course of ASIC3 expression in skin (C) and muscle (D) after plantar incision (top). The quantified normalized ASIC3 levels are shown in the bottom panel. ASIC3 levels increased in skin (C) and muscle (D) from 4 h after incision, peaked on day 1, and returned to baseline on day 7 after incision. The fold changes relative to the levels in the sham group after normalization to actin are shown. (E) Schematic of different areas of the skin around the incision. Samples were collected from area A (around the incision) and area B (1 mm outside area A). (F) Representative Western blot results and the quantified ASIC3 expression levels in different areas of incised skin on postoperative day 1 are shown. (G) Immunohistochemistry did not show ASIC3 immunoreactivity in sham-operated skin (a1) or muscle (b1), but ASIC3-like staining (arrows) was detected in nerve-like structures in skin (a2) and muscle (b2) on postoperative day 1. Higher-magnification views of the rectangular areas in a2 and b2 are shown in a3 and b3. Scale bars: 100 μm for a1 and b1; 50 μm for b1 and b2; 25 μm for a3 and b3. (H) Immunofluorescence staining showed that ASIC3 colocalized with calcitonin gene-related peptide (a3, arrows), isolectin B4 (b3, arrows), and neurofilament 200 (c3, arrows) in nerve fibers in cutaneous axons 1 d after incision. Scale bars: 25 μm. The data are shown as the mean ± SD. n = 6 per time point; *P < 0.05, **P < 0.01, ***P < 0.001 versus sham, as determined by one-way ANOVA followed by Dunnett’s test. ASIC3, acid-sensing ion channel 3; CGRP, calcitonin gene-related peptide; IB4, isolectin; NF200, neurofilament.

Incised tissues were collected from different areas around the incision (fig. 3E). Compared to that in the sham group, ASIC3 protein expression in the incision group was 2.5 ± 1.0-fold higher (n = 6, P = 0.009) in area A, which was adjacent to the incision, and 1.5 ± 0.9-fold higher (n = 6, P = 0.501) in area B, which was outside of area A, on postoperative day 1 (fig. 3F).

Sections of skin and muscle samples were collected on postoperative day 1, when ASIC3 expression was most significantly increased according to Western blot analysis. ASIC3 immunoreactivity was not detected in sham-operated skin or muscle (fig. 3G, a1 and b1). ASIC3 immunoreactivity was found in nerve-like structures of incised skin (fig. 3G, a2, arrows) and muscle (fig. 3G, b2, arrows).

Immunofluorescent double labeling was used to further confirm ASIC3 expression in cutaneous axons. ASIC3 was unequally distributed in nerve fibers in different cutaneous axons (fig. 3H, a1, b1, and c1). We also detected CGRP-positive peptidergic C-fibers, IB4-positive nonpeptidergic C-fibers, and NF200-labeled myelinated nerve fibers in nerve axons (fig. 3H, a2, b2, and c2). The merged image shows that ASIC3 was localized in CGRP, IB4, and NF200 fibers (fig. 3H, a3, b3, and c3, arrows, yellow).

ASIC3-Immunoreactive Neuron Phenotype Changes in the Dorsal Root Ganglion after Plantar Incision

Acid-sensing ion channel 3 expression in different subtypes of dorsal root ganglion neurons is shown in figure 4. Immunofluorescent double labeling showed that ASIC3 was present in lumbar 5 dorsal root ganglion neurons in both the sham-operated rats and the incised rats 1 d after incision and that it localized (arrows) in CGRP-, IB4-, and NF200-positive neurons (fig. 4, A and B).

Fig. 4.

Expression of ASIC3 in dorsal root ganglion neurons. (A and B) Immunohistochemistry indicated that ASIC3 was localized in CGRP-, IB4-, and NF200-expressing neurons in lumbar 5 dorsal root ganglia from both sham-operated rats and incised rats 1 d after incision (a3, b3, and c3; arrows). Scale bars: 50 μm. Western blot analysis revealed no statistically significant increases in ASIC3 in lumbar 4-6 dorsal root ganglia at different time points after plantar incision (C, top). The quantified normalized ASIC3 levels are shown in the bottom panel (C, n = 6 per time point). The fold changes relative to the levels in the sham group after normalization to actin are shown. Compared with that in the sham group, the percentage of ASIC3-CGRP coexpressing neurons in the lumbar 5 dorsal root ganglion was increased in the incision group 1 d after incision, but the percentages of ASIC3-IB4 and ASIC3-NF200 coexpressing neurons in the lumbar 5 dorsal root ganglion were decreased 1 d after incision (D; one slice from each rat, n = 5 per group; **P < 0.01, ***P < 0.001 versus sham, as determined by a two-tailed independent-samples t test). The data are shown as the mean ± SD. ASIC3, acid-sensing ion channel 3; CGRP, calcitonin gene-related peptide; IB4, isolectin; NF200, neurofilament.

Fig. 4.

Expression of ASIC3 in dorsal root ganglion neurons. (A and B) Immunohistochemistry indicated that ASIC3 was localized in CGRP-, IB4-, and NF200-expressing neurons in lumbar 5 dorsal root ganglia from both sham-operated rats and incised rats 1 d after incision (a3, b3, and c3; arrows). Scale bars: 50 μm. Western blot analysis revealed no statistically significant increases in ASIC3 in lumbar 4-6 dorsal root ganglia at different time points after plantar incision (C, top). The quantified normalized ASIC3 levels are shown in the bottom panel (C, n = 6 per time point). The fold changes relative to the levels in the sham group after normalization to actin are shown. Compared with that in the sham group, the percentage of ASIC3-CGRP coexpressing neurons in the lumbar 5 dorsal root ganglion was increased in the incision group 1 d after incision, but the percentages of ASIC3-IB4 and ASIC3-NF200 coexpressing neurons in the lumbar 5 dorsal root ganglion were decreased 1 d after incision (D; one slice from each rat, n = 5 per group; **P < 0.01, ***P < 0.001 versus sham, as determined by a two-tailed independent-samples t test). The data are shown as the mean ± SD. ASIC3, acid-sensing ion channel 3; CGRP, calcitonin gene-related peptide; IB4, isolectin; NF200, neurofilament.

Western blot analysis was used to detect the expression of ASIC3 in lumbar 4-6 dorsal root ganglia after incision (fig. 4C, top). Quantitative analysis indicated no statistically significant changes in ASIC3 expression in dorsal root ganglia at different time points after incision (P = 0.941 at postoperative hour 4, P = 0.600 on postoperative day 1, P = 0.777 on postoperative day 2; fig. 4C, bottom).

However, the proportion of ASIC3-CGRP-colabeled neurons relative to the total population of ASIC3-positive lumbar 5 dorsal root ganglion neurons was 36.4 ± 8.5% in the sham group but 66 ± 9.0% in the incision group 1 d after incision (five slices from five rats, P < 0.001; fig. 4D). The proportions of ASIC3-IB4- and ASIC3-NF200-colabeled neurons were lower in the incision group (21.6 ± 5.9% and 41.2 ± 4.7%, respectively) than in the sham group (42.2 ± 4.1% and 49.0 ± 3.5%, respectively) 1 d after the procedure (five slices from five rats, P < 0.001 and P = 0.009; fig. 4D).

Transport of ASIC3 between Incised Tissues and Dorsal Root Ganglia

Because there was an increase in ASIC3 in the tissue after incision, sciatic nerve ligation was performed to investigate the transport of ASIC3 between incised tissues and dorsal root ganglia through the sciatic nerve (fig. 5). The sciatic nerve was ligated, and ASIC3 expression in different parts of the sciatic nerve distal or proximal to the ligature was examined (fig. 5A, c). CGRP and ASIC3 expression were not detected in sham-incised rats (fig. 5A, a1, a2). CGRP and ASIC3 accumulated near the ligation site on both sides of the sciatic nerve 1 d after plantar incision (fig. 5A, b1, b2). Western blot analysis indicated that the total expression levels of ASIC3 in the sciatic nerve were equal on both sides of the ligation site 1 d after plantar incision (fig. 5B). We further ligated the sciatic nerve at two different sites (fig. 5C, b). ASIC3 expression accumulated on the incision side and the dorsal root ganglion side but not in the axons between the two ligations (fig. 5C, a1, a2).

Fig. 5.

Transportation of ASIC3 in the periphery. (A) Immunofluorescence staining for CGRP and ASIC3 in the sciatic nerves of the sham group (a1 and a2) and sciatic nerve ligation + plantar incision group (b1 and b2) on postoperative day 1. The circle depicts the ligature (c). Scale bars: 50 μm. (B) Western blot analysis of ASIC3 expression in the sciatic nerve proximal or distal to the ligature 1 d after plantar incision. (C) Immunofluorescence staining for ASIC3 in the sciatic nerves after plantar incision and sciatic nerve ligation on postoperative day 1 (a1 and a2). The two circles depict ligatures at two different sites (b). Scale bars: 50 μm. (D) Schematic of sciatic nerve ligation and different sections of the sciatic nerve distal or proximal to the ligature, with 1 to 4 indicating the different sections of the sciatic nerve. (E) Western blot analysis of ASIC3 in four different sections of the sciatic nerve after sciatic nerve ligation followed by plantar incision on postoperative day 1. (F) Western blot analysis of ASIC3 expression in the plantar skin after incision, sciatic nerve ligation, or sciatic nerve ligation + incision on postoperative day 1. (G) Schematic of transversal plantar incision. The solid horizontal line indicates the transversal incision; the broken square on either side of the solid horizontal line represents the harvested skin samples proximal and distal to the incision. (H) Representative Western blot results and quantified ASIC3 expression distal and proximal to the incision on postoperative day 1. (I) Protocol of ASIC3 small interfering RNA administration. (J) Western blot analysis of ASIC3 expression in incised skin on postoperative day 1 after ASIC3 small interfering RNA preinjection. The data are shown as the mean ± SD. The fold changes relative to the levels in the sham group or segment 1 group after normalization to actin as the control are shown. n = 5 per group for the ASIC3 small interfering RNA experiment, and n = 6 per group for the other experiments; *P < 0.05 versus sham or segment 1; **P < 0.01 versus sham; ***P < 0.001 versus sham; #P < 0.05 versus scramble; as determined by one-way ANOVA followed by Dunnett’s test. ASIC3, acid-sensing ion channel 3; CGRP, calcitonin gene-related peptide.

Fig. 5.

Transportation of ASIC3 in the periphery. (A) Immunofluorescence staining for CGRP and ASIC3 in the sciatic nerves of the sham group (a1 and a2) and sciatic nerve ligation + plantar incision group (b1 and b2) on postoperative day 1. The circle depicts the ligature (c). Scale bars: 50 μm. (B) Western blot analysis of ASIC3 expression in the sciatic nerve proximal or distal to the ligature 1 d after plantar incision. (C) Immunofluorescence staining for ASIC3 in the sciatic nerves after plantar incision and sciatic nerve ligation on postoperative day 1 (a1 and a2). The two circles depict ligatures at two different sites (b). Scale bars: 50 μm. (D) Schematic of sciatic nerve ligation and different sections of the sciatic nerve distal or proximal to the ligature, with 1 to 4 indicating the different sections of the sciatic nerve. (E) Western blot analysis of ASIC3 in four different sections of the sciatic nerve after sciatic nerve ligation followed by plantar incision on postoperative day 1. (F) Western blot analysis of ASIC3 expression in the plantar skin after incision, sciatic nerve ligation, or sciatic nerve ligation + incision on postoperative day 1. (G) Schematic of transversal plantar incision. The solid horizontal line indicates the transversal incision; the broken square on either side of the solid horizontal line represents the harvested skin samples proximal and distal to the incision. (H) Representative Western blot results and quantified ASIC3 expression distal and proximal to the incision on postoperative day 1. (I) Protocol of ASIC3 small interfering RNA administration. (J) Western blot analysis of ASIC3 expression in incised skin on postoperative day 1 after ASIC3 small interfering RNA preinjection. The data are shown as the mean ± SD. The fold changes relative to the levels in the sham group or segment 1 group after normalization to actin as the control are shown. n = 5 per group for the ASIC3 small interfering RNA experiment, and n = 6 per group for the other experiments; *P < 0.05 versus sham or segment 1; **P < 0.01 versus sham; ***P < 0.001 versus sham; #P < 0.05 versus scramble; as determined by one-way ANOVA followed by Dunnett’s test. ASIC3, acid-sensing ion channel 3; CGRP, calcitonin gene-related peptide.

In addition, the ligated sciatic nerve was divided into four sections 1 d after plantar incision (fig. 5D). The expression of ASIC3 in the sciatic nerve was greater closer to the ligation site (2 and 3) after plantar incision on postoperative day 1 (fig. 5E). Furthermore, ASIC3 expression was higher in the incised skin of rats that underwent sciatic nerve ligation followed by plantar incision on postoperative day 1 than in the skin of sham incision rats (3.8 ± 2.8-fold, n = 6, P = 0.028); however, sciatic nerve ligation itself did not cause ASIC3 upregulation in the skin in incised rats compared with sham-incised rats (n = 6, P = 0.937) on postoperative day 1 (fig. 5F). A transverse plantar incision was made (fig. 5G), and ASIC3 was upregulated equally on both sides of the transverse incision (fig. 5H).

In response to peer review, an inhibitor of ASIC3 messenger RNA translation was added at the incision to further confirm the local synthesis of ASIC3. We found that compared to injection of corresponding scrambled small interfering RNA, intraplantar injection of ASIC3 small interfering RNA (with the injection procedure in fig. 5I) decreased the expression of ASIC3 one day after incision (6.8 ± 3.1-fold vs. 2.9 ± 0.9-fold, n = 5, P = 0.018; fig. 5J).

Nerve Growth Factor Upregulates the Expression of ASIC3 through the Phosphoinositide 3–Kinase/Protein Kinase B Signaling Pathway

Acid-sensing ion channel 3 expression was increased in incised skin compared to sham-operated skin 4 h after incision (4.7 ± 1.7- and 4.9 ± 2.4-fold in the incision and vehicle groups, n = 6, P = 0.003 and P = 0.001 vs. sham, respectively; fig. 6A), and ASIC3 overexpression was inhibited by intraplantar injection of a nerve growth factor antibody 4 h before incision (2.3 ± 0.8-fold, n = 6, P = 0.036 vs. vehicle; fig. 6A). The nerve growth factor antibody also inhibited incision-induced phosphorylated protein kinase B upregulation 4 h after incision (1.9 ± 0.6- and 1.8 ± 0.8-fold in the incision and vehicle groups, n = 6, P = 0.025 and P = 0.048 vs. sham, respectively; 0.8 ± 0.3-fold in the anti–nerve growth factor group, n = 6, P = 0.011 vs. vehicle; fig. 6B). However, compared with that in the sham group, protein kinase B expression in the skin in the incised group was not significantly different after incision (n = 6, P = 0.479 vs. sham; fig. 6B) or nerve growth factor antibody pretreatment (n = 6, P = 0.220 vs. sham; fig. 6B).

Fig. 6.

Nerve growth factor mediates the incision-induced upregulation of ASIC3 via phosphoinositide 3–kinase/protein kinase B. Western blot analysis indicated that increases in ASIC3 (A) and phosphorylated protein kinase B (B) expression occurred at 4 h after incision. The overexpression of ASIC3 and phosphorylated protein kinase B was blocked by intraplantar administration of a nerve growth factor antibody. The expression of total protein kinase B was not changed (B). n = 6 per group; *P < 0.05, **P < 0.01 versus sham, #P < 0.05 versus vehicle; as determined by one-way ANOVA followed by Dunnett’s test. The expression of ASIC3 and phosphorylated protein kinase B in the skin was increased at 4 h and 1 d after intraplantar administration of nerve growth factor (C). n = 5 per group; *P < 0.05, **P < 0.01, ***P < 0.001 versus vehicle; as determined by one-way ANOVA followed by Dunnett’s test. Western blot analysis indicated that the increase in ASIC3 four hours after incision was inhibited by intraplantar administration of the phosphoinositide 3–kinase inhibitor LY294002 (D). The data are shown as the mean ± SD. n = 4 per group; *P < 0.05 versus sham; as determined by one-way ANOVA followed by Dunnett’s test. Pretreatment with LY294002 decreased the cumulative pain score (E) and increased the withdrawal threshold (F) to von Frey filaments at postoperative hour 4 and on postoperative day 1 after plantar incision. LY294002 increased the withdrawal latency to radiant heat at postoperative hour 4 only (G). n = 6 in the incision group, n = 9 in the incision + LY294002 group; *P < 0.05, **P < 0.01 versus incision, as determined by two-way repeated-measures ANOVA followed by Tukey’s post hoc test. Akt, protein kinase B; p-Akt, phosphorylated protein kinase B; ASIC3, acid-sensing ion channel 3.

Fig. 6.

Nerve growth factor mediates the incision-induced upregulation of ASIC3 via phosphoinositide 3–kinase/protein kinase B. Western blot analysis indicated that increases in ASIC3 (A) and phosphorylated protein kinase B (B) expression occurred at 4 h after incision. The overexpression of ASIC3 and phosphorylated protein kinase B was blocked by intraplantar administration of a nerve growth factor antibody. The expression of total protein kinase B was not changed (B). n = 6 per group; *P < 0.05, **P < 0.01 versus sham, #P < 0.05 versus vehicle; as determined by one-way ANOVA followed by Dunnett’s test. The expression of ASIC3 and phosphorylated protein kinase B in the skin was increased at 4 h and 1 d after intraplantar administration of nerve growth factor (C). n = 5 per group; *P < 0.05, **P < 0.01, ***P < 0.001 versus vehicle; as determined by one-way ANOVA followed by Dunnett’s test. Western blot analysis indicated that the increase in ASIC3 four hours after incision was inhibited by intraplantar administration of the phosphoinositide 3–kinase inhibitor LY294002 (D). The data are shown as the mean ± SD. n = 4 per group; *P < 0.05 versus sham; as determined by one-way ANOVA followed by Dunnett’s test. Pretreatment with LY294002 decreased the cumulative pain score (E) and increased the withdrawal threshold (F) to von Frey filaments at postoperative hour 4 and on postoperative day 1 after plantar incision. LY294002 increased the withdrawal latency to radiant heat at postoperative hour 4 only (G). n = 6 in the incision group, n = 9 in the incision + LY294002 group; *P < 0.05, **P < 0.01 versus incision, as determined by two-way repeated-measures ANOVA followed by Tukey’s post hoc test. Akt, protein kinase B; p-Akt, phosphorylated protein kinase B; ASIC3, acid-sensing ion channel 3.

In response to peer review, intraplantar injection of exogenous nerve growth factor was performed to further explore the effects of nerve growth factor on ASIC3 and phosphorylated protein kinase B expression (fig. 6C). We observed overexpression of ASIC3 in plantar skin at 4 h (2.8 ± 0.9-fold, n = 5, P = 0.010 vs. vehicle) and 1 d (2.5 ± 1.2-fold, n = 5, P = 0.030 vs. vehicle) after intraplantar administration of nerve growth factor. Similarly, compared to vehicle administration, intraplantar administration of nerve growth factor increased the expression of phosphorylated protein kinase B in skin at 4 h (8.5 ± 2.7-fold, n = 5, P < 0.001 vs. vehicle) and 1 d (5.8 ± 2.9-fold, n = 5, P = 0.008 vs. vehicle) after the procedure.

Interestingly, ASIC3 overexpression in incised skin 4 h after incision was inhibited by intraplantar injection of LY294002 1 h before incision (n = 4, P = 0.410 vs. sham; fig. 6D). Compared to no treatment, LY294002 treatment significantly decreased the cumulative pain score at 4 h (9.8 ± 4.5 vs. 16.2 ± 3.6, n = 6 in the incision group, n = 9 in the incision + LY294002 group, P = 0.003 vs. incision; fig. 6E) and 1 d (8.0 ± 3.0 vs. 12.8 ± 5.3, n = 6 in the incision group, n = 9 in the incision + LY294002 group, P = 0.0285 vs. incision; fig. 6E) after incision. LY294002 statistically significantly increased the withdrawal threshold to von Frey filaments at 4 h (4.5 ± 1.5 g vs. 1.8 ± 1.1 g, n = 6 in the incision group, n = 9 in the incision + LY294002 group, P = 0.020 vs. incision; fig. 6F) and 1 d (6.2 ± 1.5 g vs. 2.4 ± 1.8 g, n = 6 in the incision group, n = 9 in the incision + LY294002 group, P = 0.001 vs. incision; fig. 6F) after incision. In addition, LY294002 statistically significantly increased the withdrawal latency to radiant heat only at 4 h (7.9 ± 1.2 s vs. 14.3 ± 5.8 s, n = 6 in the incision group, n = 9 in the incision + LY294002 group, P = 0.028 vs. incision; fig. 6G) after incision.

In response to peer review, key experiments were repeated in female rats (fig. 7). The effects of APETx2 on pain-related behaviors after plantar incision in female rats were similar to those in male rats (fig. 7A–C). ASIC3 overexpression in the incised tissues of female rats at different time points was also similar to that in the incised tissues of male rats (fig. 7D). The expression of phosphorylated protein kinase B, but not that of protein kinase B, was also increased in the incised tissues after plantar incision in female rats (fig. 7E). In addition, we found overexpression of ASIC3 and phosphorylated protein kinase B in skin after intraplantar administration of nerve growth factor in females (fig. 7F). The effects of LY294002 on pain-related behaviors after plantar incision in female rats were similar to those in males (fig. 7G-I).

Fig. 7.

Supplementary data from some key experiments on female rats. (A–C) Intraplantar administration of APETx2 decreased the cumulative pain score (A) at postoperative hour 4 (n = 6 per group; **P < 0.01 versus vehicle; two-way ANOVA followed by Tukey’s post hoc test); increased the withdrawal threshold (B) at postoperative hour 4, on postoperative day 1 and on postoperative day 2 (n = 6 per group; *P < 0.05 versus vehicle; two-way ANOVA followed by Tukey’s post hoc test); and had no effect on withdrawal latency to heat (C, n = 9 per group; two-way ANOVA followed by Tukey’s post hoc test). (D and E) Time course of ASIC3 (n = 6 per time point; *P < 0.05, ***P < 0.001 versus sham; one-way ANOVA followed by Dunnett’s test), p-Akt (n = 5 per time point; *P < 0.05, **P < 0.01, ***P < 0.001 versus sham; one-way ANOVA followed by Dunnett’s test), and Akt (n = 5 per time point; P > 0.05 versus sham; one-way ANOVA followed by Dunnett’s test) expression in incised tissues. (F) The expression of ASIC3 and phosphorylated protein kinase B in skin was increased 4 h and 1 d after intraplantar administration of nerve growth factor (n = 4 per group; *P < 0.05, **P < 0.01 versus vehicle; one-way ANOVA followed by Dunnett’s test). (G–I) Pretreatment with LY294002 decreased the cumulative pain scores (G) and increased the withdrawal threshold to von Frey filaments (H) at postoperative hour 4 and on postoperative day 1. The withdrawal latency to radiant heat was increased at postoperative hour 4 (I; n = 6 per group; *P < 0.05, **P < 0.01, ***P < 0.001 versus incision; one-way ANOVA followed by Dunnett’s test). The data are shown as the mean ± SD. Akt, protein kinase B; ASIC3, acid-sensing ion channel 3; p-Akt, phosphorylated protein kinase B.

Fig. 7.

Supplementary data from some key experiments on female rats. (A–C) Intraplantar administration of APETx2 decreased the cumulative pain score (A) at postoperative hour 4 (n = 6 per group; **P < 0.01 versus vehicle; two-way ANOVA followed by Tukey’s post hoc test); increased the withdrawal threshold (B) at postoperative hour 4, on postoperative day 1 and on postoperative day 2 (n = 6 per group; *P < 0.05 versus vehicle; two-way ANOVA followed by Tukey’s post hoc test); and had no effect on withdrawal latency to heat (C, n = 9 per group; two-way ANOVA followed by Tukey’s post hoc test). (D and E) Time course of ASIC3 (n = 6 per time point; *P < 0.05, ***P < 0.001 versus sham; one-way ANOVA followed by Dunnett’s test), p-Akt (n = 5 per time point; *P < 0.05, **P < 0.01, ***P < 0.001 versus sham; one-way ANOVA followed by Dunnett’s test), and Akt (n = 5 per time point; P > 0.05 versus sham; one-way ANOVA followed by Dunnett’s test) expression in incised tissues. (F) The expression of ASIC3 and phosphorylated protein kinase B in skin was increased 4 h and 1 d after intraplantar administration of nerve growth factor (n = 4 per group; *P < 0.05, **P < 0.01 versus vehicle; one-way ANOVA followed by Dunnett’s test). (G–I) Pretreatment with LY294002 decreased the cumulative pain scores (G) and increased the withdrawal threshold to von Frey filaments (H) at postoperative hour 4 and on postoperative day 1. The withdrawal latency to radiant heat was increased at postoperative hour 4 (I; n = 6 per group; *P < 0.05, **P < 0.01, ***P < 0.001 versus incision; one-way ANOVA followed by Dunnett’s test). The data are shown as the mean ± SD. Akt, protein kinase B; ASIC3, acid-sensing ion channel 3; p-Akt, phosphorylated protein kinase B.

Discussion

In the current study, we found that incision-induced mechanical hypersensitivity and pain-related guarding behavior were partially reversed by the ASIC3 blocker APETx2 or by ASIC3 knockout. The ASIC3 level was increased in incised skin and muscle. Sciatic nerve ligation revealed bidirectional transport of ASIC3 between incised tissue and the dorsal root ganglia. ASIC3 overexpression at the incision site was regulated by nerve growth factor through the phosphoinositide 3–kinase/protein kinase B signaling pathway.

ASIC3 in Nociception

Neither APETx2 treatment nor ASIC3 knockout had an effect on heat responses, which is not consistent with a previous study.13  The effect of ASIC3 on heat hyperalgesia is not consistent in pain models. ASIC3-knockout in mice did not influence heat hyperalgesia in repeated intramuscular acid injection or carrageenan-induced muscle inflammation.25,26  Thus, role of ASIC3 in mechanical responses is more evident in mechanical hyperalgesia than in heat hyperalgesia. A previous study also detected the expression of ASIC3 in cutaneous mechanosensory structures, such as Merkel nerve endings, Meissner corpuscles, and free nerve endings.27 

Acid-sensing ion channel 3 may also contribute to the detection of an ischemic-like signal. The time course of ASIC3 expression was in accordance with changes in pain-related behaviors, lactate concentration, and pH in local tissues after plantar incision.7,8  ASIC3 in peptidergic C-fibers and nonpeptidergic C-fibers is believed to be associated with nociceptive transmission. Increased ASIC3 in CGRP immunoreactive dorsal root ganglia neurons is consistent with greater responses of muscle C-fibers to lactic acid.10  Together, these findings support the idea that peripheral ASIC3 upregulation is involved in incisional nociception.

ASIC3 Changes in Dorsal Root Ganglia

Similar to our findings, the results of a previous study indicate that the proportion of ASIC3-positive neurons is not significantly different between spinal nerve ligation model animals and sham animals.28  In the current study, the proportions of IB4- and NF200-positive neurons expressing ASIC3 were decreased, but the expression of ASIC3 in CGRP-positive neurons was increased. Previous studies have also found that injury, nerve damage, and inflammation may change ASIC3 immunoreactivity from fewer large neurons and nonpeptidergic neurons to more expression in peptidergic neurons.13,28,29  Changes in ASIC3-immunoreactive dorsal root ganglion neurons may also contribute to postincisional nociception.

ASIC3 Transport

Previous studies have indicated that ASICs are mainly expressed in sensory neurons and may be transported exclusively to peripheral sensory terminals.12,30  In this study, the sciatic nerve was ligated to examine the transport of ASIC3, and equal accumulation of ASIC3 was found on both sides of the ligature. To ensure that ligation could completely block potential axonal transport, we performed double ligation of the sciatic nerve. ASIC3 accumulated at the incision side and the dorsal root ganglion side. Since ASIC3 was not observed in the axons between the two ligations, we believe that the transport was blocked. The sciatic nerve around the ligature was further divided into four distinct segments, and the accumulation of ASIC3 was higher on both sides near the ligature than in remote segments. These results indicate that ASIC3 transport occurs between incised tissues and dorsal root ganglia and might be bidirectional.

ASIC3 Expression in Dorsal Root Ganglia

Acid-sensing ion channel 3 expression in the dorsal root ganglia was not significantly different between incised and sham-operated rats; this finding is not consistent with a previous study showing that ASIC3 is upregulated in sensory neurons innervating the hind paw plantar muscle.13  This difference is likely a consequence of the different methods used. We examined ASIC3 in the lumbar 4-6 dorsal root ganglia by Western blot analysis, whereas the previous study examined retrograde-labeled lumbar 5 dorsal root ganglion neurons using immunofluorescence staining in specific muscle-innervating dorsal root ganglion neurons. As shown in figure 4A, ASIC3 is normally observed in rat dorsal root ganglion neurons. When ASIC3 is upregulated in dorsal root ganglion neurons after incision, ASIC3 is then moved distally via axonal transport. Although ASIC3 levels should be increased under conditions of bidirectional transport of ASIC3, we do not know exactly the hierarchy of the transport from incisions or dorsal root ganglia.

ASIC3 Is Synthesized in Incisions

A transverse incision was made on the paw to cut the nerve fibers from the tibial nerve coursing from the heel toward digits and thus block the axoplasmic transport of ASIC3 from the proximal to the distal region. We found equal upregulation of ASIC3 on both sides of the transversely incised skin, suggesting that ASIC3 is expressed locally. To further confirm this hypothesis, we simultaneously made an incision on the paw and ligated the sciatic nerve; subsequently, ASIC3 upregulation was observed in the incised tissues. In addition, intraplantar injection of ASIC3 small interfering RNA decreased ASIC3 expression at the incision. A growing body of evidence suggests that nonneuronal cells such as keratinocytes, immune cells, and glial cells play important roles in the pathogenesis of nociception.31  We suggest that nerve growth factor may also be produced by some nonneuronal cells in skin and muscle after incision, and affects the ASIC3 expression in afferents in the incision.

Nerve Growth Factor Regulates ASIC3 through Phosphoinositide 3–Kinase/Protein Kinase B Signaling Pathway

We have previously shown that nerve growth factor is markedly upregulated in the skin and muscle after plantar incision.14,15  Nerve growth factor binds to TrkA and triggers the expression of intracellular signal transduction–inducing genes. TrkA is highly expressed in the peripheral nervous system, mast cells, immune tissue, and skin.32  In this study, we found that intraplantar administration of nerve growth factor upregulated ASIC3 expression and that pretreatment with a nerve growth factor antibody significantly reversed plantar incision–induced ASIC3 expression, indicating that nerve growth factor may be involved in ASIC3 expression at the incision site. Recent studies have suggested that phosphoinositide 3–kinase plays an important role in the induction of thermal and mechanical hyperalgesia by capsaicin and nerve growth factor.33,34  Increased expression of phosphorylated protein kinase B has been found in the spinal cord in plantar incision models, and phosphoinositide 3–kinase inhibition significantly ameliorates plantar incision–induced postoperative pain-related behavior, suggesting that the phosphoinositide 3–kinase/protein kinase B signaling pathway is involved in central sensitization after incision.35,36  In the current study, we found that intraplantar injection of nerve growth factor upregulated phosphorylated protein kinase B expression and that phosphorylated protein kinase B upregulation in incised tissues was abolished by a nerve growth factor antibody, suggesting activation of the phosphoinositide 3–kinase/protein kinase B signaling pathway by nerve growth factor. Furthermore, phosphoinositide 3–kinase inhibition by LY294002 significantly reversed ASIC3 overexpression and mainly decreased incisional guarding and mechanical hyperalgesia. These data suggest that ASIC3 overexpression in incised tissues is regulated by nerve growth factor via the phosphoinositide 3–kinase/protein kinase B pathway.

Phosphoinositide 3–Kinase/Protein Kinase B Signaling Pathway Is Associated with Both Mechanical and Heat Nociception

As shown in figure 6E–G, LY294002 also temporarily decreased incisional heat hyperalgesia 4 h after incision. This occurred because some heat nociception likely utilized the phosphoinositide 3–kinase/protein kinase B signaling pathway, perhaps via transient receptor potential vanilloid 1, and nerve growth factor is involved in this process.22  Inhibition of transient receptor potential vanilloid 1 and nerve growth factor decreases heat hyperalgesia after plantar incision.23 

Sex Differences in Protein Kinase B

Sex differences in nociception occur under certain conditions. A previous study showed sexual differences in protein kinase B activation in the spinal dorsal horns of mice subjected to plantar incision.35  In the current study, sex differences in protein kinase B activation were not observed, which may be related to the different sample tissues and species examined in our study. We examined the expression of phosphorylated protein kinase B in the skin of rats, whereas the previous study examined phosphorylated protein kinase B in the spinal cords of mice.

Benefaction from the Current Study

The incisional pain model is widely used to study the mechanism of postincisional nociception, and its features are highly consistent with the symptoms of clinical postoperative pain. Acidification occurs in the surgical wounds.7,8  Since ASIC3 is sensitive to acid, the results from this study might be generalizable to human conditions.

Nerve growth factor binds to TrkA with high affinity and to p75 with low affinity. Trk and p75 receptors show independent signaling properties, and the phosphoinositide 3–kinase/protein kinase B signaling pathway is activated when nerve growth factor binds to Trk receptors.18,37  Based on our observations, inhibition of nerve growth factor seems to have a better effect on postoperative nociception than inhibition of ASIC3. However, nerve growth factor and its receptors are widely expressed in the peripheral and central nervous systems. The negative effects of nerve growth factor inhibition, like osteonecrosis,38  in treating pain remain a concern. The clinical applicability of ASIC3 or nerve growth factor inhibition on postoperative pain therapy remains to be further studied.

Conclusions

Acid-sensing ion channel 3 contributes to incisional guarding and mechanical pain–related behaviors in rats. Nerve growth factor produced after incision in skin and muscle14,15  increases ASIC3 in incised skin and muscle, and bidirectional transport of ASIC3 between incised tissues and dorsal root ganglia occurs. ASIC3 expression at the incision site is regulated by nerve growth factor through the phosphoinositide 3–kinase/protein kinase B pathway. Thus, local ASIC3 inhibition may be a novel treatment strategy for postoperative pain.

Acknowledgments

The authors thank Timothy J. Brennan, M.D., Ph.D., Department of Anesthesia, University of Iowa Hospitals and Clinics, Iowa City, Iowa, for his valuable advice in revising this article.

Research Support

This study was supported by grant Nos. 81471141 and 81271238 from the National Natural Science Foundation of China (Beijing, China) and the Shenzhen Key Medical Discipline Construction Fund (Shenzhen, China).

Competing Interests

The authors declare no competing interests.

References

1.
Kehlet
H
,
Holte
K
:
Effect of postoperative analgesia on surgical outcome.
Br J Anaesth
.
2001
;
87
:
62
72
2.
Vaurio
LE
,
Sands
LP
,
Wang
Y
,
Mullen
EA
,
Leung
JM
:
Postoperative delirium: The importance of pain and pain management.
Anesth Analg
.
2006
;
102
:
1267
73
3.
Wu
CL
,
Raja
SN
:
Treatment of acute postoperative pain.
Lancet
.
2011
;
377
:
2215
25
4.
Sommer
M
,
de Rijke
JM
,
van Kleef
M
,
Kessels
AG
,
Peters
ML
,
Geurts
JW
,
Gramke
HF
,
Marcus
MA
:
The prevalence of postoperative pain in a sample of 1490 surgical inpatients.
Eur J Anaesthesiol
.
2008
;
25
:
267
74
5.
Pogatzki-Zahn
EM
,
Zahn
PK
,
Brennan
TJ
:
Postoperative pain–Clinical implications of basic research.
Best Pract Res Clin Anaesthesiol
.
2007
;
21
:
3
13
6.
Brennan
TJ
,
Zahn
PK
,
Pogatzki-Zahn
EM
:
Mechanisms of incisional pain.
Anesthesiol Clin North Am
.
2005
;
23
:
1
20
7.
Woo
YC
,
Park
SS
,
Subieta
AR
,
Brennan
TJ
:
Changes in tissue pH and temperature after incision indicate acidosis may contribute to postoperative pain.
Anesthesiology
.
2004
;
101
:
468
75
8.
Kim
TJ
,
Freml
L
,
Park
SS
,
Brennan
TJ
:
Lactate concentrations in incisions indicate ischemic-like conditions may contribute to postoperative pain.
J Pain
.
2007
;
8
:
59
66
9.
Kang
S
,
Brennan
TJ
:
Chemosensitivity and mechanosensitivity of nociceptors from incised rat hindpaw skin.
Anesthesiology
.
2009
;
111
:
155
64
10.
Kido
K
,
Gautam
M
,
Benson
CJ
,
Gu
H
,
Brennan
TJ
:
Effect of deep tissue incision on pH responses of afferent fibers and dorsal root ganglia innervating muscle.
Anesthesiology
.
2013
;
119
:
1186
97
11.
Dubé
GR
,
Lehto
SG
,
Breese
NM
,
Baker
SJ
,
Wang
X
,
Matulenko
MA
,
Honoré
P
,
Stewart
AO
,
Moreland
RB
,
Brioni
JD
:
Electrophysiological and in vivo characterization of A-317567, a novel blocker of acid sensing ion channels.
Pain
.
2005
;
117
:
88
96
12.
Deval
E
,
Lingueglia
E
:
Acid-sensing ion channels and nociception in the peripheral and central nervous systems.
Neuropharmacology
.
2015
;
94
:
49
57
13.
Deval
E
,
Noël
J
,
Gasull
X
,
Delaunay
A
,
Alloui
A
,
Friend
V
,
Eschalier
A
,
Lazdunski
M
,
Lingueglia
E
:
Acid-sensing ion channels in postoperative pain.
J Neurosci
.
2011
;
31
:
6059
66
14.
Wu
C
,
Boustany
L
,
Liang
H
,
Brennan
TJ
:
Nerve growth factor expression after plantar incision in the rat.
Anesthesiology
.
2007
;
107
:
128
35
15.
Wu
C
,
Erickson
MA
,
Xu
J
,
Wild
KD
,
Brennan
TJ
:
Expression profile of nerve growth factor after muscle incision in the rat.
Anesthesiology
.
2009
;
110
:
140
9
16.
Banik
RK
,
Subieta
AR
,
Wu
C
,
Brennan
TJ
:
Increased nerve growth factor after rat plantar incision contributes to guarding behavior and heat hyperalgesia.
Pain
.
2005
;
117
:
68
76
17.
Ji
RR
,
Samad
TA
,
Jin
SX
,
Schmoll
R
,
Woolf
CJ
:
p38 MAPK activation by NGF in primary sensory neurons after inflammation increases TRPV1 levels and maintains heat hyperalgesia.
Neuron
.
2002
;
36
:
57
68
18.
Pezet
S
,
McMahon
SB
:
Neurotrophins: Mediators and modulators of pain.
Annu Rev Neurosci
.
2006
;
29
:
507
38
19.
Mamet
J
,
Baron
A
,
Lazdunski
M
,
Voilley
N
:
Proinflammatory mediators, stimulators of sensory neuron excitability via the expression of acid-sensing ion channels.
J Neurosci
.
2002
;
22
:
10662
70
20.
Mamet
J
,
Lazdunski
M
,
Voilley
N
:
How nerve growth factor drives physiological and inflammatory expressions of acid-sensing ion channel 3 in sensory neurons.
J Biol Chem
.
2003
;
278
:
48907
13
21.
Zhang
X
,
Huang
J
,
McNaughton
PA
:
NGF rapidly increases membrane expression of TRPV1 heat-gated ion channels.
EMBO J
.
2005
;
24
:
4211
23
22.
Coelho
A
,
Wolf-Johnston
AS
,
Shinde
S
,
Cruz
CD
,
Cruz
F
,
Avelino
A
,
Birder
LA
:
Urinary bladder inflammation induces changes in urothelial nerve growth factor and TRPV1 channels.
Br J Pharmacol
.
2015
;
172
:
1691
9
23.
Wu
C
,
Gavva
NR
,
Brennan
TJ
:
Effect of AMG0347, a transient receptor potential type V1 receptor antagonist, and morphine on pain behavior after plantar incision.
Anesthesiology
.
2008
;
108
:
1100
8
24.
Brennan
TJ
,
Vandermeulen
EP
,
Gebhart
GF
:
Characterization of a rat model of incisional pain.
Pain
.
1996
;
64
:
493
501
25.
Sluka
KA
,
Price
MP
,
Breese
NM
,
Stucky
CL
,
Wemmie
JA
,
Welsh
MJ
:
Chronic hyperalgesia induced by repeated acid injections in muscle is abolished by the loss of ASIC3, but not ASIC1.
Pain
.
2003
;
106
:
229
39
26.
Sluka
KA
,
Radhakrishnan
R
,
Benson
CJ
,
Eshcol
JO
,
Price
MP
,
Babinski
K
,
Audette
KM
,
Yeomans
DC
,
Wilson
SP
:
ASIC3 in muscle mediates mechanical, but not heat, hyperalgesia associated with muscle inflammation.
Pain
.
2007
;
129
:
102
12
27.
Price
MP
,
McIlwrath
SL
,
Xie
J
,
Cheng
C
,
Qiao
J
,
Tarr
DE
,
Sluka
KA
,
Brennan
TJ
,
Lewin
GR
,
Welsh
MJ
:
The DRASIC cation channel contributes to the detection of cutaneous touch and acid stimuli in mice.
Neuron
.
2001
;
32
:
1071
83
28.
Omori
M
,
Yokoyama
M
,
Matsuoka
Y
,
Kobayashi
H
,
Mizobuchi
S
,
Itano
Y
,
Morita
K
,
Ichikawa
H
:
Effects of selective spinal nerve ligation on acetic acid-induced nociceptive responses and ASIC3 immunoreactivity in the rat dorsal root ganglion.
Brain Res
.
2008
;
1219
:
26
31
29.
Ikeuchi
M
,
Kolker
SJ
,
Sluka
KA
:
Acid-sensing ion channel 3 expression in mouse knee joint afferents and effects of carrageenan-induced arthritis.
J Pain
.
2009
;
10
:
336
42
30.
García-Añoveros
J
,
Samad
TA
,
Zuvela-Jelaska
L
,
Woolf
CJ
,
Corey
DP
:
Transport and localization of the DEG/ENaC ion channel BNaC1alpha to peripheral mechanosensory terminals of dorsal root ganglia neurons.
J Neurosci
.
2001
;
21
:
2678
86
31.
Ji
RR
,
Chamessian
A
,
Zhang
YQ
:
Pain regulation by non-neuronal cells and inflammation.
Science
.
2016
;
354
:
572
7
32.
Skaper
SD
:
Nerve growth factor: A neuroimmune crosstalk mediator for all seasons.
Immunology
.
2017
;
151
:
1
15
33.
Zhuang
ZY
,
Xu
H
,
Clapham
DE
,
Ji
RR
:
Phosphatidylinositol 3-kinase activates ERK in primary sensory neurons and mediates inflammatory heat hyperalgesia through TRPV1 sensitization.
J Neurosci
.
2004
;
24
:
8300
9
34.
Malik-Hall
M
,
Dina
OA
,
Levine
JD
:
Primary afferent nociceptor mechanisms mediating NGF-induced mechanical hyperalgesia.
Eur J Neurosci
.
2005
;
21
:
3387
94
35.
Xu
B
,
Mo
C
,
Lv
C
,
Liu
S
,
Li
J
,
Chen
J
,
Wei
Y
,
An
H
,
Ma
L
,
Guan
X
:
Post-surgical inhibition of phosphatidylinositol 3-kinase attenuates the plantar incision-induced postoperative pain behavior via spinal Akt activation in male mice.
BMC Neurosci
.
2019
;
20
:
36
36.
Xu
B
,
Guan
XH
,
Yu
JX
,
Lv
J
,
Zhang
HX
,
Fu
QC
,
Xiang
HB
,
Bu
HL
,
Shi
D
,
Shu
B
,
Qin
LS
,
Manyande
A
,
Tian
YK
:
Activation of spinal phosphatidylinositol 3-kinase/protein kinase B mediates pain behavior induced by plantar incision in mice.
Exp Neurol
.
2014
;
255
:
71
82
37.
Chao
MV
:
Neurotrophins and their receptors: A convergence point for many signalling pathways.
Nat Rev Neurosci
.
2003
;
4
:
299
309
38.
Gimbel
JS
,
Kivitz
AJ
,
Bramson
C
,
Nemeth
MA
,
Keller
DS
,
Brown
MT
,
West
CR
,
Verburg
KM
:
Long-term safety and effectiveness of tanezumab as treatment for chronic low back pain.
Pain
.
2014
;
155
:
1793
801