TRPV1 and TRPV4 play pivotal roles in delayed onset muscle soreness

Abstract

Unaccustomed strenuous exercise that includes lengthening contraction (LC) often causes tenderness and movement related pain after some delay (delayed-onset muscle soreness, DOMS). We previously demonstrated that nerve growth factor (NGF) and glial cell line-derived neurotrophic factor (GDNF) are up-regulated in exercised muscle through up-regulation of cyclooxygenase (COX)-2, and they sensitized nociceptors resulting in mechanical hyperalgesia. There is also a study showing that transient receptor potential (TRP) ion channels are involved in DOMS. Here we examined whether and how TRPV1 and/or TRPV4 are involved in DOMS. We firstly evaluated a method to measure the mechanical withdrawal threshold of the deep tissues in wild-type (WT) mice with a modified Randall-Selitto apparatus. WT, TRPV1-/- and TRPV4-/- mice were then subjected to LC. Another group of mice received injection of murine NGF-2.5S or GDNF to the lateral gastrocnemius (LGC) muscle. Before and after these treatments the mechanical withdrawal threshold of LGC was evaluated. The change in expression of NGF, GDNF and COX-2 mRNA in the muscle was examined using real-time RT-PCR. In WT mice, mechanical hyperalgesia was observed 6-24 h after LC and 1-24 h after NGF and GDNF injection. LC induced mechanical hyperalgesia neither in TRPV1-/- nor in TRPV4-/- mice. NGF injection induced mechanical hyperalgesia in WT and TRPV4-/- mice but not in TRPV1-/- mice. GDNF injection induced mechanical hyperalgesia in WT but neither in TRPV1-/- nor in TRPV4-/- mice. Expression of NGF and COX-2 mRNA was significantly increased 3 h after LC in all genotypes. However, GDNF mRNA did not increase in TRPV4-/- mice. These results suggest that TRPV1 contributes to DOMS downstream (possibly at nociceptors) of NGF and GDNF, while TRPV4 is located downstream of GDNF and possibly also in the process of GDNF up-regulation.

Figure 1. Measurement of muscular mechanical withdrawal threshold with Randall-Selitto apparatus in mice.

(A) Schedule for testing effects of EMLA cream treatment on the withdrawal threshold. More than 12 h before these measurements, inflammation was induced by injecting carrageenan into the LGC muscle. VFT: von Frey hair test, RST: Randall-Sellito test. (B) Change in VFT threshold (tip diameter: 0.25 mm) by surface anesthesia. (i) Vehicle cream (n = 9) did not change the threshold. (ii) EMLA treatment (n = 10) significantly raised the threshold compared with before the treatment. Median and interquartile range (IQR) are shown. *** p<0.001 for pre- and post-cream treatment comparison by Mann-Whitney test. Note that pre values are decreased ones after induction of inflammation (same in C). (C) Change after surface anesthesia by EMLA cream treatment in withdrawal threshold measured by RST with a self-made larger probe (tip diameter: 2.6 mm). Filled circles: EMLA treatment (n = 10), open square: vehicle cream treatment (n = 9). EMLA cream did not significantly change the threshold, the same as vehicle cream. Mean ± S.E.M. (n = 6–10 for each group). S.E.M.s are hardly seen because they are small.

Figure 2. Muscular mechanical hyperalgesia induced by lengthening contraction in mice.

(A) Schema of lengthening contraction (LC) application to the lower hindleg flexors, mainly the lateral gastrocnemius (LGC) muscle. LC was induced by electrical stimulation through a pair of needle electrodes inserted near the tibial and sciatic nerves. The ankle joint was dorsi-flexed in synchrony with muscle contraction, and then returned to the starting position over a 3 s resting period. This cycle was repeated 300 times. (B) Change in withdrawal thresholds by RST in WT mice that received LC or sham (stretch only) exercise (n = 9 for each group, mean ± S.E.M.). Vertical axis: withdrawal threshold in mN, horizontal axis: time after exercise. There was a significant difference between the groups, and the threshold decreased 6 to 36 h after exercise in LC group, but not in sham group. ** p<0.01, *** p<0.001 compared with −1 day in LC group; # p<0.05, ### p<0.001 compared with sham group on each time point, two-way repeated measures ANOVA with Bonferroni t-test.

Figure 3. Muscular mechanical hyperalgesia did not develop after LC in TRPV1−/− and TRPV4−/− mice.

(A) Change in the withdrawal thresholds after LC measured by RST in TRPV1−/− (crosses) and TRPV4−/− (open squares) mice. Vertical axis: difference in the threshold from −1 d in mN, horizontal axis: time after LC. (B) Changes in the mechanical hyperalgesia by intramuscular injection of HC-067047, a TRPV4 selective antagonist (100 mg/kg; crosses) or DMSO (open squares) in WT mice. Mean ± S.E.M. (n = 6–10 for each group). ** p<0.01, *** p<0.001 compared with −1 d in WT,+p<0.05 compared with −1 d in TRPV4−/−, ## p<0.01, ### p<0.001 compared with 14 h in HC-067047 group; two-way repeated measures ANOVA with Bonferroni t-test.

Figure 4. NGF-β mRNA was up-regulated in all three genotypes.

(A) Time course of NGF-β mRNA expression in LC-exercised LGC muscle in WT mice. (B) Up-regulation of NGF-β mRNA 3 h after LC in the muscle of three genotypes. Median and interquartile range (IQR). All values were normalized with β-actin mRNA. n = 3–8 for each group (shown in the parentheses under each column). * p≤0.05 compared with pre, and n.s. not different from WT (Kruskal-Wallis one-way analysis of variance on ranks test followed by the Dunn's test).

Figure 5. Muscular mechanical hyperalgesia was not developed after NGF injection in TRPV1−/− mice.

Change in the withdrawal threshold by intramuscular injection of NGF-2.5 S was measured by RST. Vertical axis: difference in the withdrawal threshold from that before NGF-injection, horizontal axis: time after injection. WT mice (closed circles), TRPV1−/− mice (crosses) and TRPV4−/− mice (open squares). *** p<0.001 compared with −1 d (Bef. in the figure) in WT;+p<0.05,++p<0.01 compared with −1 d in TRPV4−/−; two-way repeated measures ANOVA with Bonferroni t-test.

Figure 6. Change in expression levels of COX-2 and GDNF mRNA after LC in TRPV1−/− and TRPV4−/− mice.

(A) COX-2 mRNA level in excised LGC muscle of WT mice after LC. (B) Up-regulation of COX-2 mRNA 3 h after LC in the muscle of three genotypes. (C) GDNF mRNA level in excised LGC muscle of WT mice. (D) Up-regulation of GDNF mRNA 3 h after LC in the muscle of three genotypes. (E) Effect of HC-067047 on the level of GDNF mRNA 3 h after LC. All values were normalized with β-actin mRNA. Median and interquartile range are shown. Number of animals used is shown in the parentheses under each column. * p≤0.05 compared with pre (in C) and with WT (in D); n.s., not different from WT in B and D, or from DMSO group in E, by Kruskal-Wallis one-way analysis of variance on ranks test followed by the Dunn's test except in E. Mann-Whiteny U test was used in E.

Figure 7. GDNF failed to induce muscle mechanical hyperalgesia both in TRPV1−/− and TRPV4−/− mice.

Change in the withdrawal threshold by intramuscular injection of GDNF was evaluated by RST. Vertical axis: difference in withdrawal threshold from that before GDNF-injection, horizontal axis: time after the injection. The threshold in WT mice (closed circles) decreased 1 to 24 h after the injection, while that in TRPV1−/− (crosses) or TRPV4−/− mice (open squares) did not decrease. Mean ± S.E.M. Number of animals used is in the parentheses. *** p<0.001 compared with −1 d (Bef in the graph) in WT; two-way repeated measures ANOVA with Bonferroni t-test.