Hydrogen Peroxide Induces Muscle Nociception via Transient Receptor Potential Ankyrin 1 Receptors

Abstract

Background: H2O2 has a variety of actions in skin wounds but has been rarely studied in deep muscle tissue. Based on response to the transient receptor potential ankyrin 1 antagonists after plantar incision, we hypothesized that H2O2 exerts nociceptive effects via the transient receptor potential ankyrin 1 in muscle.

Methods: Nociceptive behaviors in rats (n = 269) and mice (n = 16) were evaluated after various concentrations and volumes of H2O2 were injected into the gastrocnemius muscle or subcutaneous tissue. The effects of H2O2 on in vivo spinal dorsal horn neuronal activity and lumbar dorsal root ganglia neurons in vitro were evaluated from 26 rats and 6 mice.

Results: Intramuscular (mean ± SD: 1,436 ± 513 s) but not subcutaneous (40 ± 58 s) injection of H2O2 (100 mM, 0.6 ml) increased nociceptive time. Conditioned place aversion was evident after intramuscular (-143 ± 81 s) but not subcutaneous (-2 ± 111 s) injection of H2O2. These H2O2-induced behaviors were blocked by transient receptor potential ankyrin 1 antagonists. Intramuscular injection of H2O2 caused sustained in vivo activity of dorsal horn neurons, and H2O2 activated a subset of dorsal root ganglia neurons in vitro. Capsaicin nerve block decreased guarding after plantar incision and reduced nociceptive time after intramuscular H2O2. Nociceptive time after intramuscular H2O2 in transient receptor potential ankyrin 1 knockout mice was shorter (173 ± 156 s) compared with wild-type mice (931 ± 629 s).

Conclusions: The greater response of muscle tissue to H2O2 may help explain why incision that includes deep muscle but not skin incision alone produces spontaneous activity in nociceptive pathways.

Fig. 1. Nociceptive behavior in rats as total time spent flinching, lifting and licking of the hind leg during a 60-minute period

(A) Time-course of the nociceptive behavior after various volumes of 100 mM H₂O₂ were injected into the gastrocnemius muscle. Data were collected in 5-minute bins from six animals in each group. Data points show the average nociceptive time in 5-minute bins, and error bars were omitted for clarity. (B) Total time of nociceptive behavior over 60 minutes after various volumes of 100 mM H₂O₂ were injected into the gastrocnemius muscle (summarized data from Fig. 1A). Each group contained six rats. * P = 0.0035, † P = 0.0004, ‡ P = 0.0005, # P < 0.0001 compared with the SIF injection group by two-way ANOVA (interaction factor: F_{3, 40} = 2.227, P = 0.0999, Injection volume factor: F_{3, 40} = 2.022, P = 0.1263, Group factor: F_{1, 40} = 91.61, P < 0.0001) followed by post-hoc t-test with Bonferroni’s correction. (C) Spontaneous nociceptive behavior after various concentrations of 0.6 ml H₂O₂ were injected subcutaneously (SQ) or intramuscularly (IM). Each group contained six rats. * P < 0.0001 compared with the IM SIF injection group, † P < 0.0001 compared with IM 10 mM H₂O₂ injection group, ‡ P < 0.0001 compared with IM 30 mM H₂O₂ injection group, # P < 0.0001 compared with SQ 100 mM H₂O₂ injection group by one-way ANOVA (F_{7, 40} = 34.92, P < 0.0001) followed by post-hoc Tukey’s test. (D) Effects of local pre-injection of a TRPA1 antagonist AP-18 (50 mM, 0.3 ml) on nociceptive behavior caused by intramuscular injection of H₂O₂ (100 mM, 0.3 ml). Therefore, the total injection volume was 0.6 ml, and the final concentration was 25 mM for AP-18, and 50 mM for H₂O₂. Vehicle + SIF group (n=7); Vehicle + H₂O₂ group (n=7); AP-18 + H₂O₂ group (n=6). * P < 0.0001 compared with vehicle + SIF injection group, † P = 0.0001 compared with AP-18 + H₂O₂ injection group by one-way ANOVA (F_{2, 17} = 25.65, P < 0.0001) followed by post-hoc Tukey’s test. Vehicle + SIF and Vehicle + H₂O₂ control groups are the same control data presented in Fig. 5B in our previous study. (E) Total nociceptive time during the 60-minute period after injection of H₂O₂ into the gastrocnemius muscle or subcutaneous tissue overlying gastrocnemius muscle in sodium azide-treated rats. Sodium azide (10 mg/kg) was administered intraperitoneally 30 minutes before the injection of H₂O₂. Each group contained six rats. *, † P < 0.0001 compared with vehicle group by unpaired t-test. All data are expressed as means ± SEM. (F) Total nociceptive time during the 60-minute period after cinnamaldehyde (CM, 30 mM, 0.6 ml) injection into the gastrocnemius muscle (IM CM) or subcutaneous tissue (SQ CM) overlying gastrocnemius muscle in rats. Each group contained seven rats. * P = 0.0129 compared with the SQ CM group, † P = 0.0121 compared with IM vehicle group by one-way ANOVA (F_{2, 18} = 6.923, P = 0.0059) followed by post-hoc Tukey’s test. All data are expressed as means ± SD.

Fig. 2. The effects of intramuscular (IM) or subcutaneous (SQ) injection of H₂O₂ (100 mM, 0.6 ml) on conditioned place aversion (CPA) in rats

Each column represents the time spent in the preferred and non-preferred chambers during the pre- and post-conditioning sessions (A), and the CPA scores (B). For the intramuscular co-injection of H₂O₂ and AP-18, sequential injections of AP-18 (50 mM, 0.3 ml) followed by H₂O₂ (200 mM, 0.3 ml), were made into the gastrocnemius muscle. Therefore, the total injection volume was 0.6 ml, and the final concentration of H₂O₂ was 100 mM. Each group contained six rats. All data are expressed as means ± SD. * P = 0.0075 by paired t-test. † = 0.0199 compared with the SIF group, ‡ P = 0.0194 compared with the subcutaneous injection of H₂O₂ group, # P = 0.0479 compared with the intramuscular co-injection of H₂O₂ and AP-18 group by one-way ANOVA (F_{3, 20} = 4.884, P = 0.0104) followed by post-hoc Tukey’s test.

Fig. 3. Effects of intramuscular (IM) or subcutaneous (SQ) injection of 30 mM H₂O₂ on conditioned place aversion (CPA) in rats

Each column represents the time spent in the preferred and non-preferred chambers during the pre- and post-conditioning sessions (A), and the CPA scores (B) of the intramuscular injection of SIF group (n = 8), subcutaneous injections of H₂O₂ (30 mM, 0.6 ml) group (n = 8), and intramuscular injection of H₂O₂ (30 mM, 0.6 ml) group (n = 8), intramuscular sequential-injection of AP-18 (50 mM, 0.3 ml) followed by H₂O₂ (60 mM, 0.3 ml) group (n = 9). All data are expressed as means ± SD. * P = 0.0011 by paired t-test. † = 0.0036 compared with the SIF group, ‡ P = 0.0026 compared with the subcutaneous injection of H₂O₂ group, # P = 0.0048 compared with the intramuscular co-injection of H₂O₂ and AP-18 group by one-way ANOVA (F_{3, 29} = 7.212, P = 0.0009) followed by post-hoc Tukey’s test.

Fig. 4. Effects of intramuscular injection of H₂O₂ on activity of dorsal horn neurons (DHNs)

(A) Example recordings of neurons after subcutaneous (SQ) injection of H₂O₂ (upper panel), intramuscular (IM) injection of SIF (middle panel), and IM injection of H₂O₂ (lower panel). Arrows represent pinching the skin (SP), squeezing gastrocnemius muscle (MS), and the time of injections. Bin width = 1 second. Unit represents each single action potential. (B) The average depth from the surface of spinal cord in which DHNs were recorded in each groups. Data are expressed as median with interquartile range. (C) Time-course trend of changes in activity of the DHNs after injection of H₂O₂ or vehicle. Data points show the median impulse/second for six neurons in each group in 5-minute bins. (D) Total activity in 5 minutes of the DHNs prior to injection. Data are expressed as median with interquartile range. Each group contained 6 neurons. (E) Total activity of the DHNs during 60 minutes after injection. Data are expressed as median with interquartile range. Each group contained 6 neurons. * P = 0.0482, # P = 0.0261 by Kruskal-Wallis test.

Fig. 5. Effect of sciatic nerve block with capsaicin on behavior after skin + deep tissue plantar incision and intramuscular injection of H₂O₂

(A) Effect of sciatic nerve block with capsaicin on guarding behavior after skin + deep tissue plantar incision. Percutaneous sciatic nerve block was performed using 0.05% capsaicin plus 0.5% bupivacaine or 0.5% bupivacaine in vehicle. The results are presented as means ± SD for eight rats in each group. Two-way ANOVA with repeated measures on one factor (interaction factor: F_{7, 98} = 7.913, P < 0.0001) followed by Bonferroni’s post hoc test was used to compare the cumulative pain score at each time point between the groups. * P = 0.0202, † P < 0.0001 compared with the vehicle + 0.5% bupivacaine group at each time point. POD = postoperative day. (B) Effect of peripheral nerve block with 0.05% capsaicin on nociceptive behavior induced by injection of H₂O₂ into the gastrocnemius muscle. H₂O₂ (100 mM, 0.6 ml) was injected into the gastrocnemius muscle at various time points (1, 3, 5, and 7 days) following the nerve block, and the total time spent flinching, lifting, and licking was recorded for 60 minutes. Data are presented as mean ± SD. Capsaicin (0.05%) + 0.5% Bupivacaine group contained 5 rats and vehicle group contained 6 rats. * P = 0.0004, † P = 0.0029, ‡ P = 0.0168 compared with the vehicle + 0.5% bupivacaine group by two-way ANOVA with repeated measured on one factor (interaction factor: F_{3, 27} = 0.9405, P = 0.4348, Time factor: F_{3, 27} = 1.554, P = 0.2233, Group factor: F_{1, 9} = 22.56, P = 0.0010) followed by Bonferroni’s post-hoc test.

Fig. 6. DRG neuronal responses to H₂O₂ (1 mM), capsaicin (0.5 μM) and allyl isothiocyanate (AITC; 100 μM)

(A) Example traces of response to H₂O₂ and capsaicin (CAPS; 20-second application of each) of individual, dissociated lumbar 3–5 DRG neurons during Fura-2 Ca²⁺ imaging. (B) Venn diagrams illustrating the overlap of cells exhibiting Ca²⁺ transients in response to H₂O₂ and capsaicin. (C) Percentage of cells that responded both to H₂O₂ and capsaicin (blue column), that responded to H₂O₂ but not to capsaicin green column), and that responded to capsaicin but not to H₂O₂ (red column). (D) Example traces in response to H₂O₂ and AITC (20-second duration of each) of individual, dissociated lumbar 3–5 DRG neurons during Fura-2 Ca²⁺ imaging. (E) Venn diagrams illustrating the overlap of cells exhibiting Ca²⁺ transients in response to H₂O₂ and AITC. (F) Percentage of cells that responded to both H₂O₂ and AITC (blue column), that responded to H₂O₂ but not to AITC (green column), and that responded to AITC but not to H₂O₂ (red column).

Fig. 7. H₂O₂-induced nociceptive behavior and Ca²⁺ transients in dorsal root ganglia (DRG) neurons induced by H₂O₂ in TRPA1 knockout (TRPA1 −/−) and wild-type (TRPA1 +/+) mice

(A) Total time of nociceptive behavior over 60 minutes after H₂O₂ (0.05 ml, 30 mM) was injected into the gastrocnemius muscle. Data were obtained from eight animals in each group. * P = 0.0051 compared with the trpa1 +/+ group by unpaired t-test. Data are expressed as means ± SD. (B) Example traces in response to H₂O₂ (1 mM; 20-second duration) of individual, dissociated lumbar 3–5 DRG neurons from TRPA1 −/− mice during Fura-2 Ca²⁺ imaging. (C) Example traces in response to H₂O₂ (1 mM; 20-second duration) of individual, dissociated lumbar 3–5 DRG neurons from TRPA1 +/+ mice during Fura-2 Ca²⁺ imaging. (D) Summary of the percentage of cells responding to 1 mM H₂O₂. Numbers in parentheses represent the number of cells responding and total number of cells tested. * P < 0.0001 compared with the TRPA1 +/+ mice group by Chi-square test.