GLP-1 and its derived peptides mediate pain relief through direct TRPV1 inhibition without affecting thermoregulation

Abstract

Hormonal regulation during food ingestion and its association with pain prompted the investigation of the impact of glucagon-like peptide-1 (GLP-1) on transient receptor potential vanilloid 1 (TRPV1). Both endogenous and synthetic GLP-1, as well as a GLP-1R antagonist, exendin 9-39, reduced heat sensitivity in naïve mice. GLP-1-derived peptides (liraglutide, exendin-4, and exendin 9-39) effectively inhibited capsaicin (CAP)-induced currents and calcium responses in cultured sensory neurons and TRPV1-expressing cell lines. Notably, exendin 9-39 alleviated CAP-induced acute pain, as well as chronic pain induced by complete Freund's adjuvant (CFA) and spared nerve injury (SNI), in mice without causing hyperthermia associated with other TRPV1 inhibitors. Electrophysiological analyses revealed that exendin 9-39 binds to the extracellular side of TRPV1, functioning as a noncompetitive inhibitor of CAP. Exendin 9-39 did not affect proton-induced TRPV1 activation, suggesting its selective antagonism. Among the exendin 9-39 fragments, exendin 20-29 specifically binds to TRPV1, alleviating pain in both acute and chronic pain models without interfering with GLP-1R function. Our study revealed a novel role for GLP-1 and its derivatives in pain relief, suggesting exendin 20-29 as a promising therapeutic candidate.

Fig. 1. Reduced thermal sensitivity in systemically glucose- or GLP-1 treated- and locally GLP-1 treated-mice and the analgesic effect of GLP-1 on CAP-induced nociceptive behaviors in mice.

a Effects of intraperitoneal (i.p.) administration of 2 g/kg glucose on heat sensitivity according to the hot plate test (mean ± S.E.M., n = 5). Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, compared with the saline group). Effects of intraperitoneal (b) and oral (c) administration of 10 μg/kg GLP-1(7–36) (GLP-1) on heat sensitivity via the hot plate test (mean ± S.E.M., n = 5 ~ 7). Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, ***p < 0.001, compared with the saline group). d Effects of intraplantar administration (i.pl.) of 10 μg of GLP-1(7–36) on heat sensitivity via the Hargreaves test (mean ± S.E.M., n = 5). Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, ***p < 0.001, compared with the saline group). e Effects of intraplantar administration of 10 μg of GLP-1(7-36) on the CAP (1.6 μg)-induced spontaneous licking time (mean ± S.E.M., n = 5). Two-tailed unpaired t test (****p < 0.0001, compared with the saline group). Schematic illustration of the analgesic effect through systemic GLP-1 secretion (f) and local GLP-1 induction (g). ANOVA analysis of variance, CAP capsaicin, GLP-1 glucagon-like peptide-1, S.E.M. standard error of the mean.

Fig. 2. Inhibition of CAP-induced TRPV1 currents and calcium responses in small- to medium-sized mouse DRG neurons following pretreatment with GLP-1 analogs.

a Representative inward currents induced by 100 nM CAP in the absence (left) or presence of 100 nM GLP-1(7–36) (GLP-1; right). Representative traces of calcium influx elicited by 100 nM control CAP (b), GLP-1 (c), exendin-4 (d), or liraglutide (e). The calcium response to high potassium (KCl, 50 mM) was used to identify neurons. f Mean normalized currents of sequential CAP-induced currents (mean ± S.E.M.). Two-tailed unpaired t test (**p < 0.01, compared with each control CAP). g Mean normalized 340/380 ratios of sequential CAP-induced calcium increases (mean ± S.E.M.). One-way ANOVA followed by Dunnett’s multiple comparison test (****p < 0.0001, compared with control CAP). ANOVA analysis of variance, CAP capsaicin, DRG dorsal root ganglia, GLP-1 glucagon-like peptide-1, S.E.M. standard error of the mean, TRPV1 transient receptor potential vanilloid 1.

Fig. 3. Inhibition of CAP-induced TRPV1 currents and calcium responses in small- to medium-sized mouse DRG neurons following pretreatment with exendin 9–39, an antagonist of GLP-1, and its analgesic effects on CAP-induced nociceptive behaviors in mice.

a Effects of intraplantar administration of exendin 9–39 (Exe 9–39) (10 μg) on heat sensitivity via the Hargreaves test. Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, compared with the saline group). b Representative inward currents induced by 100 nM CAP in the presence of 100 nM exendin 9–39 (Exe 9–39; left). Mean normalized currents of sequential CAP-induced currents (mean ± S.E.M.; right). Two-tailed unpaired t test (****p < 0.0001, compared with each control CAP). c Representative traces of calcium influx elicited by 100 nM control CAP and pretreatment with 100 nM Exe 9–39. The calcium response to high potassium (KCl, 50 mM) was used to identify neurons (left). Mean normalized 340/380 ratios of sequential CAP-induced calcium increases (mean ± S.E.M.; right). Two-tailed unpaired t test (****p < 0.0001, compared with each control CAP). d Effects of intraplantar administration of exendin 9–39 (5 and 10 μg) and BCTC (0.5 μg) on CAP (1.6 μg)-induced acute licking time (mean ± S.E.M., n = 5). One-way ANOVA followed by Dunnett’s multiple comparison test (^#p < 0.05, ^####p < 0.0001, compared with the vehicle + CAP group). e Effects of the intraperitoneal administration of exendin 9–39 (50 mg/kg) or BCTC (5 mg/kg) on body temperature (mean ± S.E.M., n = 5). Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, ****p < 0.0001, vehicle group compared with the BCTC group; ^†††p < 0.001, ^††††p < 0.0001, exendin 9–39 compared with the BCTC group). f Effects of intraplantar administration of exendin 9–39 (5 and 10 μg) on CAP-induced acute thermal hyperalgesia (left) and mechanical allodynia (right) in mice (mean ± S.E.M., n = 5 each). Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, compared with the vehicle group; ^#p < 0.05, ^##p < 0.01, ^###p < 0.001, ^####p < 0.0001, compared with the vehicle + CAP group). ANOVA analysis of variance, CAP capsaicin, DRG dorsal root ganglia, GLP-1 glucagon-like peptide-1, S.E.M. standard error of the mean, TRPV1 transient receptor potential vanilloid 1.

Fig. 4. Alleviation of CFA-induced inflammatory and SNI-induced neuropathic pain via exendin 9–39 administration.

a Schematic illustration and timeline of the CFA-induced inflammatory chronic pain model in mice. b Effects of intraplantar injection of exendin 9–39 (Exe 9–39) (5 and 10 μg) on thermal hyperalgesia in a CFA-induced inflammatory chronic pain mouse model via the Hargreaves test (mean ± S.E.M., n = 7 each). Two-way ANOVA followed by the Bonferroni multiple comparison test (****p < 0.0001, compared with the CFA group). c Effects of intraplantar injection of exendin 9–39 (Exe 9–39) (5 and 10 μg) on mechanical allodynia in a CFA-induced inflammatory chronic pain mouse model via the von Frey test (mean ± S.E.M., n = 7 each). Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, **p < 0.01, ***p < 0.001, compared with the CFA group). d Schematic illustration and timeline of the SNI-induced chronic neuropathic pain model in mice. e Effects of intraplantar injection of exendin 9–39 (Exe 9–39) (10 μg) on thermal hyperalgesia in an SNI-induced chronic neuropathic pain mouse model via the Hargreaves test (mean ± S.E.M., n = 7 each). Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, **p < 0.01, compared with the Sham + Vehicle group; ^####p < 0.0001, compared with the SNI + Vehicle group). f Effects of intraperitoneal injection of exendin 9–39 (Exe 9–39) (10 μg) on thermal hyperalgesia in an SNI-induced chronic neuropathic pain mouse model via the Hargreaves test (mean ± S.E.M., n = 7 each). Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, compared with the Sham + Vehicle group; ^#p < 0.05, ^##p < 0.01, compared with the SNI + Vehicle group). ANOVA analysis of variance, CFA complete Freund’s adjuvant, SNI spared nerve injury, GLP-1 glucagon-like peptide-1, S.E.M. standard error of the mean, TRPV1 transient receptor potential vanilloid 1.

Fig. 5. Inhibition of CAP-induced calcium influx via direct interaction with TRPV1 via GLP-1 analogs and exendin 9–39 in HEK293T cells transfected with rat TRPV1 and CHO K1 cells expressing human TRPV1.

a Sequence alignment, molecular weight, and function of GLP-1 analogs and exendin 9–39. b–e Mean normalized calcium influx in HEK293T cells transfected with rat TRPV1. Calcium increases were elicited by CAP (100 nM) in the presence of GLP-1 analogs (100 nM each), GLP-1(7–36) (b), exendin-4 (c), liraglutide (d), or exendin 9–39 (Exe 9–39) (e) (mean ± S.E.M.). Two-tailed unpaired t test (****p < 0.0001, compared with each control CAP). f Mean normalized calcium influx in CHO K1 cells expressing human TRPV1. Calcium increases were elicited by CAP (100 nM) in the absence (control CAP) and presence of GLP-1 analogs (100 nM) or exendin 9–39 (100 nM) (mean ± S.E.M.). One-way ANOVA followed by Dunnett’s multiple comparison test (****p < 0.0001, compared with the control, CAP). g Curves were fitted via a logistic function to show concentration-dependent inhibition of CAP-induced TRPV1 currents by 100 nM GLP-1, exendin-4, liraglutide, or exendin 9–39. h IC₅₀ values of GLP-1, exendin-4, liraglutide, and exendin 9–39 (178.6 ± 23.40 nM, 62.19 ± 5.12 nM, 64.49 ± 10.19 nM, and 28.18 ± 3.92 nM, respectively). i Pull-down assay using His-tagged exendin 9–39 and cell lysates from CHO K1 cells expressing human TRPV1 (‘T’) and naïve CHO K1 cells (‘C’). Detection of the interaction between exendin 9–39 and TRPV1 via western blotting with an anti-human TRPV1 antibody. j Confocal images of FITC-labeled exendin 9–39 in CHO K1 cells expressing human TRPV1 (upper) compared with naïve CHO K1 cells (lower). Overlap of FITC-exendin 9–39 (green), anti-human TRPV1 antibody (red), and Hoechst (blue) in the merged image. Scale bar: 5 µm. ANOVA analysis of variance, CAP capsaicin, CHO K1 Chinese hamster ovary, FITC fluorescein isothiocyanate, GLP-1 glucagon-like peptide-1, S.E.M. standard error of the mean, TRPV1 transient receptor potential vanilloid 1.

Fig. 6. Exendin 9–39 noncompetitively inhibits CAP-induced TRPV1 activation by binding to the extracellular side in CHO K1 cells expressing human TRPV1.

a Currents evoked by 100 nM CAP were partially blocked by sequential application of 10 and 20 nM BCTC (left). The remaining currents between the arrows were CAP-induced currents inhibited by BCTC. Summary analysis of the inward currents reduced by BCTC alone (mean ± S.E.M.). Two-tailed unpaired t test (ns, not significant). b Currents evoked by 100 nM CAP were partially blocked by the sequential application of 10 nM BCTC and 100 nM exendin 9–39 (Exe 9–39). The remaining currents between the arrows were CAP-induced currents inhibited by BCTC or exendin 9–39. Summary of inward currents reduced by BCTC alone and BCTC followed by exendin 9–39 (mean ± S.E.M.). Two-tailed unpaired t test (***p < 0.001). c Representative traces from single-channel recordings of TPRV1 channels in the cell-attached configuration. The cells were held at −60 mV. Exendin 9–39 (100 nM) was delivered via either the external (intracellular) or internal (extracellular) solution in the presence of 10 nM CAP externally. c closed state, o open state. d Schematic illustration of cell-attached patch recordings. e Average single-channel open time (left) and open probability (right) (mean ± S.E.M.). One-way ANOVA followed by Dunnett’s multiple comparison test (***p < 0.001, compared with no exendin 9–39). f Representative traces from single-channel recordings of TPRV1 channels in the inside-out configuration. The cells were held at −60 mV. Exendin 9–39 (100 nM) was delivered via either the external (intracellular) or internal (extracellular) solution in the presence of 10 nM CAP externally. c closed state, o open state. g Schematic of inside-out patch recordings. h Average single-channel open time (left) and open probability (right) (mean ± S.E.M.). One-way ANOVA followed by Dunnett’s multiple comparison test was used (**p < 0.01, compared with no exendin 9–39). ANOVA analysis of variance, BCTC N-(4-tertiarybutylphenyl)-4-(3-cholorphyridin-2-yl) tetrahydropyrazine-1(2H)-carboxamide, CAP capsaicin, CHO K1 Chinese hamster ovary, S.E.M. standard error of the mean, TRPV1 transient receptor potential vanilloid 1.

Fig. 7. Exendin 9–39 had no effect on proton-induced TRPV1 currents or calcium influx in CHO K1 cells expressing human TRPV1.

a Representative proton-induced inward currents evoked at pH 5.5 under control conditions (blue) or with 1 µM exendin 9–39 (blue) or 1 µM BCTC (orange) pretreatment. b Mean normalized current amplitude of TRPV1 currents (nA) after treatment with 1 µM exendin 9–39 or 1 µM BCTC compared with the control, pH 5.5 (blue) (mean ± S.E.M.). One-way ANOVA followed by Dunnett’s multiple comparison test (ns, not significant; ****p < 0.0001, compared with the control, pH 5.5). c Representative proton-induced calcium influx at pH 5.5 under control conditions (blue) or with 1 µM exendin 9–39 (blue) or 1 µM BCTC (orange) pretreatment. d Mean normalized 340/380 ratios of sequential proton-induced calcium increases (mean ± S.E.M.). One-way ANOVA followed by Dunnett’s multiple comparison test (ns, not significant; ****p < 0.0001, compared with the control, pH 5.5). ANOVA analysis of variance, BCTC N-(4-tertiarybutylphenyl)-4-(3-cholorphyridin-2-yl) tetrahydropyrazine-1(2H)-carboxamide, CHO K1 Chinese hamster ovary, S.E.M. standard error of the mean, TRPV1 transient receptor potential vanilloid 1.

Fig. 8. Inhibition of CAP-induced TRPV1 activation by exendin 9–39 fragments and the analgesic effects of exendin 20–29 on CAP-induced nociceptive behaviors in mice.

a Sequence alignment of three truncated small peptides derived from exendin 9–39, their overlapping sequence exendin 20–29, and the molecular weights of these peptides. b Mean normalized calcium influx in CHO K1 cells expressing human TRPV1 with control CAP (100 nM) and each of the four truncated small peptides derived from exendin 9–39 (100 nM each) (mean ± S.E.M.). One-way ANOVA followed by Dunnett’s multiple comparison test (****p < 0.0001, compared with the control, CAP). c Mean normalized currents of sequential CAP-induced currents (mean ± S.E.M.). One-way ANOVA followed by Dunnett’s multiple comparison test (****p < 0.0001, compared with the control, CAP). d Effects of intraplantar administration of exendin 20–29 (20 μg) and BCTC (0.5 μg) on CAP (1.6 μg)-induced acute licking time (mean ± S.E.M., n = 6). One-way ANOVA followed by Dunnett’s multiple comparison test (^##p < 0.01, compared with the vehicle + CAP group). e Effects of intraplantar administration of exendin 20–29 (20 μg) on CAP-induced acute thermal hyperalgesia (left) and mechanical allodynia (right) in mice. Two-way ANOVA followed by the Bonferroni multiple comparison test (*p < 0.05, vehicle + CAP group compared with the exendin 20–29 [20 μg] group; ^#p < 0.05, ^##p < 0.01, vehicle + CAP group compared with the BCTC [0.5 μg] + CAP) group. f Effects of the intraperitoneal administration of 10 μg/kg exendin 20–29 or 10 μg/kg exendin-4 on blood glucose levels following the administration of 2 g/kg glucose (mean ± S.E.M., n = 5). Two-way ANOVA followed by the Bonferroni multiple comparison test (**p < 0.01, ****p < 0.0001, compared with the vehicle group; ^##p < 0.01, ^####p < 0.0001, compared with the glucose control group; ^†p < 0.05, ^††††p < 0.0001, compared with the exendin-4 group). g Schematic illustration of the potential mechanism by which exendin 20–29 does not disrupt the regulation of blood glucose levels in the pancreas, where GLP-1 receptors are present. ANOVA analysis of variance, BCTC N-(4-tertiarybutylphenyl)-4-(3-cholorphyridin-2-yl) tetrahydropyrazine-1(2H)-carboxamide, CAP capsaicin, CHO K1 Chinese hamster ovary, GLP-1 glucagon-like peptide-1, S.E.M. standard error of the mean, TRPV1 transient receptor potential vanilloid 1.