α2- and β2-Adrenoreceptor-Mediated Efficacy of the Atypical Antidepressant Agomelatine Combined With Gabapentin to Suppress Allodynia in Neuropathic Rats With Ligated Infraorbital or Sciatic Nerve

Abstract

Previous data showed that neuropathic pain induced by mechanical lesion of peripheral nerves has specific characteristics and responds differently to alleviating drugs at cephalic versus extracephalic level. This is especially true for tricyclic antidepressants currently used for alleviating neuropathic pain in humans which are less effective against cephalic neuropathic pain. Whether this also applies to the antidepressant agomelatine, with its unique pharmacological properties as MT₁/MT₂ melatonin receptor agonist and 5-HT_2B/5-HT_2C serotonin receptor antagonist, has been investigated in two rat models of neuropathic pain. Acute treatments were performed 2 weeks after unilateral chronic constriction (ligation) injury to the sciatic nerve (CCI-SN) or the infraorbital nerve (CCI-ION), when maximal mechanical allodynia had developed in ipsilateral hindpaw or vibrissal pad, respectively, in Sprague-Dawley male rats. Although agomelatine (45 mg/kg i.p.) alone was inactive, co-treatment with gabapentin, at an essentially ineffective dose (50 mg/kg i.p.) on its own, produced marked anti-allodynic effects, especially in CCI-ION rats. In both CCI-SN and CCI-ION models, suppression of mechanical allodynia by 'agomelatine + gabapentin' could be partially mimicked by the combination of 5-HT_2C antagonist (SB 242084) + gabapentin, but not by melatonin or 5-HT_2B antagonist (RS 127445, LY 266097), alone or combined with gabapentin. In contrast, pretreatment by idazoxan, propranolol or the β₂ antagonist ICI 118551 markedly inhibited the anti-allodynic effect of 'agomelatine + gabapentin' in both CCI-SN and CCI-ION rats, whereas pretreatment by the MT₁/MT₂ receptor antagonist S22153 was inactive. Altogether these data indicate that 'agomelatine + gabapentin' is a potent anti-allodynic combination at both cephalic and extra-cephalic levels, whose action implicates α₂- and β₂-adrenoreceptor-mediated noradrenergic neurotransmission.

Keywords: agomelatine; chronic constriction injury; gabapentin; infraorbital nerve; mechanical allodynia; neuropathic rats; sciatic nerve; α2- and β2-adrenoreceptors.

FIGURE 1

Effects of agomelatine, gabapentin and their combination on mechanical allodynia in CCI-SN (A) and CCI-ION (B) rats. Left panels: Agomelatine (45 mg/kg), gabapentin (50 mg/kg), agomelatine + gabapentin and/or respective vehicles (saline, HEC) were injected i.p. 2 weeks after nerve ligation. Pressure threshold values (as g) were determined using von Frey filaments applied onto the ipsilateral hindpaw (A-CCI-SN) or vibrissal pad (B-CCI-ION) at various times after injections (abscissa). Each point is the mean ± SEM of n independent determinations. ^∗P < 0.05, compared with pressure threshold values determined just prior to drug injection (0 on abscissa), one way ANOVA with repeated measures, Dunnett’s test. C on abscissa: intact healthy rats before surgery. Right panels: AUC values calculated from the respective time-course curves: (1) saline + HEC [n = 25 (A), n = 13 (B)]; (2) agomelatine + saline (n = 7/5); (3) gabapentin + HEC (n = 9/6); (4) agomelatine + gabapentin (n = 40/28). A- CCI-SN : one way ANOVA [F(3,77) = 92.39, P < 0.0001] followed by Tukey’s test (^∗P < 0.05, ^∗∗∗P < 0.001); B- CCI-ION: one way ANOVA [F(3,48) = 24.19, P < 0.0001] followed by Tukey’s test (^∗∗∗P < 0.001).

FIGURE 2

Effect of pretreatment with S22153 on mechanical allodynia inhibition by ‘agomelatine + gabapentin’ in CCI-SN (A) and CCI-ION (B) rats. Left panels: (1) S22153 (20 mg/kg) was injected i.p. 30 min before administration of agomelatine (45 mg/kg i.p.) + gabapentin (50 mg/kg i.p.) – at time 0 on abscissa – in rats whose right SN (A) or ION (B) nerve had been ligated 2 weeks before. Pressure threshold values (as g) were determined as described in the legend to Figure 1. Each point is the mean ± SEM of n independent determinations. ^∗P < 0.05 compared with pressure threshold values determined just prior to drug injection, one way ANOVA with repeated measures, Dunnett’s test. C on abscissa: intact healthy rats before surgery. In both CCI-SN (A) and CCI-ION (B) rats, ‘saline+HEC’-treated groups were also included, but, for the sake of clarity, corresponding data are not represented as they superimposed with those obtained in rats treated with S22153 plus these vehicles. Right panels: AUC values calculated from the respective time-course curves: (1) S22153 + saline + HEC [n = 5 (A), n = 5 (B)]; (2) agomelatine + gabapentin (n = 6/7); (3) S22153 + agomelatine + gabapentin (n = 11/14). A- CCI-SN : one way ANOVA [F(2,18) = 17.40, P < 0.0001] followed by Tukey’s test (^∗∗P < 0.01, ^∗∗∗P < 0.001); B- CCI-ION : one way ANOVA [F(2,25) = 5.913, P = 0.0079] followed by Tukey’s test (^∗P < 0.05, ^∗∗P < 0.01).

FIGURE 3

Effects of SB 206553 alone or co-administered with gabapentin on mechanical allodynia in CCI-SN (A) and CCI-ION (B) rats. Comparison with the anti-allodynic effect of ‘agomelatine + gabapentin.’ Left panels: SB 206553 (2.5 or 10 mg/kg) + saline, SB 206553 (2.5 or 10 mg/kg) + gabapentin (50 mg/kg), agomelatine (45 mg/kg) + gabapentin and/or respective vehicles (saline, HEC) were injected i.p. 2 weeks after nerve ligation. Pressure threshold values (as g) were determined as described in the legend to Figure 1. Each point is the mean ± SEM of n independent determinations. ^∗P < 0.05, compared with pressure threshold values determined just prior to drug injection (0 on abscissa), one way ANOVA with repeated measures, Dunnett’s test. C on abscissa: intact healthy rats before surgery. In both CCI-SN (A) and CCI-ION (B) rats, ‘saline+HEC’-treated groups were also included, but, for the sake of clarity, corresponding data are not represented as they superimposed with those obtained in rats treated with S206553 plus saline. Right panels: AUC values calculated from the respective time-course curves: (1) SB 206553 + saline [n = 6 (A), n = 6 (B)]; (2) SB 206553 (2.5 mg/kg) + gabapentin (n = 6/6); (3) SB 206553 (10 mg/kg) + gabapentin (n = 6/13); (4) agomelatine + gabapentin (n = 8/6). A- CCI-SN: one way ANOVA [F(3,22) = 10.97, P < 0.0001] followed by Tukey’s test (^∗P < 0.05, ^∗∗∗ P < 0.001); B- CCI-ION: one way ANOVA [F(3,27) = 6.972, P = 0.0013] followed by Tukey’s test (^∗∗P < 0.01).

FIGURE 4

Effects of RS 127445 alone or co-administered with gabapentin on mechanical allodynia in CCI-SN (A) and CCI-ION (B) rats. Comparison with the anti-allodynic effect of ‘agomelatine + gabapentin.’ Left panels: RS 127445 (20 mg/kg), RS 127445 (5 or 20 mg/kg) + gabapentin (50 mg/kg), agomelatine (45 mg/kg) + gabapentin and/or respective vehicles (saline, HEC; not shown) were injected i.p. 2 weeks after nerve ligation. Pressure threshold values (as g) were determined as described in the legend to Figure 1. Each point is the mean ± SEM of n independent determinations. ^∗P < 0.05, compared with pressure threshold values determined just prior to drug injection (0 on abscissa), one way ANOVA with repeated measures, Dunnett’s test. C on abscissa: intact healthy rats before surgery. In both CCI-SN (A) and CCI-ION (B) rats, ‘saline+HEC’-treated groups were also included, but, for the sake of clarity, corresponding data are not represented as they superimposed with those obtained in rats treated with RS 127445 plus saline. Right panels: AUC values calculated from the respective time-course curves: (1) RS 127445 (20 mg/kg) + saline [n = 6 (A), n = 5 (B)]; (2) RS 127445 (5 mg/kg) + gabapentin (n = 6/7); (3) RS 127445 (20 mg/kg) + gabapentin (n = 6/8); (4) agomelatine + gabapentin (n = 7/7). A- CCI-SN : one way ANOVA [F(3,21) = 25.20, P < 0.0001] followed by Tukey’s test (^∗∗∗P < 0.001); B- CCI-ION : one way ANOVA [F(3,23) = 11.86, P < 0.0001] followed by Tukey’s test (^∗∗P < 0.01, ^∗∗∗P < 0.001).

FIGURE 5

Effects of SB 242084 alone or co-administered with gabapentin on mechanical allodynia in CCI-SN (A) and CCI-ION (B) rats. Comparison with the anti-allodynic effect of ‘agomelatine + gabapentin.’Left panels: SB 242084 (10 mg/kg), SB 242084 (2.5 or 10 mg/kg) + gabapentin (50 mg/kg), agomelatine (45 mg/kg) + gabapentin, and/or respective vehicles (saline, HEC; not shown) were injected i.p. 2 weeks after nerve ligation. Pressure threshold values (as g) were determined as described in the legend to Figure 1. Each point is the mean ± SEM of n independent determinations. ^∗P < 0.05, compared with pressure threshold values determined just prior to drug injection (0 on abscissa), one way ANOVA with repeated measures, Dunnett’s test. C on abscissa: intact healthy rats before surgery. In both CCI-SN (A) and CCI-ION (B) rats, ‘saline+HEC’-treated groups were also included, but, for the sake of clarity, corresponding data are not represented as they superimposed with those obtained in rats treated with SB 242084 plus saline. Right panels: AUC values calculated from the respective time-course curves: (1) SB 242084 (10 mg/kg) + saline [n = 7 (A), n = 5 (B)]; (2) SB 242084 (2.5 mg/kg) + gabapentin (n = 8/6); (3) SB 242084 (10 mg/kg) + gabapentin (n = 10/8); (4) agomelatine + gabapentin (n = 7/7). A- CCI-SN : one way ANOVA [F(3,28) = 16.76, P < 0.0001] followed by Tukey’s test (^∗P < 0.05, ^∗∗P < 0.01, ^∗∗∗P < 0.001); B- CCI-ION: one way ANOVA [F(3,22) = 7.98, P = 0.0009] followed by Tukey’s test (^∗∗P < 0.01).

FIGURE 6

Effects of idazoxan alone or co-administered with agomelatine, gabapentin or ‘agomelatine + gabapentin’ on mechanical allodynia in CCI-SN (A) and CCI-ION (B) rats. Left panels: Idazoxan (2 mg/kg i.p.) or its vehicle (saline) was administered 30 min before other treatments in rats whose right SN (A) or ION (B) had been ligated 2 weeks before. Five different treatment groups were made: (1) idazoxan + HEC + saline, (2) idazoxan + agomelatine (45 mg/kg) + saline, (3) idazoxan + HEC + gabapentin (50 mg/kg), (4) agomelatine + gabapentin, (5) idazoxan + agomelatine + gabapentin. Pressure threshold values (as g) were determined as described in the legend to Figure 1. Each point is the mean ± SEM of independent determinations in n rats for each condition. ^∗P < 0.05, compared with pressure threshold values determined just prior to treatments (0 on abscissa), one way ANOVA with repeated measures, Dunnett’s test. C on abscissa: intact healthy rats before surgery. In both CCI-SN (A) and CCI-ION (B) rats, ‘saline+HEC’-treated groups were also included, but, for the sake of clarity, corresponding data are not represented as they superimposed with those obtained in rats treated with idazoxan + HEC + saline. Right panels: AUC values calculated from the respective time-course curves: (1) idazoxan + HEC + saline [n = 7 (A), n = 5 (B)]; (2) idazoxan + agomelatine + saline (n = 6/6); (3) idazoxan + HEC + gabapentin (n = 6/5); (4) agomelatine + gabapentin (n = 7/6); 5 = idazoxan + agomelatine + gabapentin (n = 7/14). A- CCI-SN : one way ANOVA [F(4,28) = 16.33, P < 0.0001] followed by Tukey’s test (^∗∗P < 0.01, ^∗∗∗P < 0.001); B- CCI-ION : one way ANOVA [F(4,35) = 8.267, P < 0.0001] followed by Tukey’s test (^∗P < 0.05, ^∗∗P < 0.01, ^∗∗∗P < 0.001).

FIGURE 7

Effects of propranolol alone or co-administered with agomelatine, gabapentin or ‘agomelatine + gabapentin’ on mechanical allodynia in CCI-SN (A) and CCI-ION (B) rats. Left panels: Propranolol (10 mg/kg i.p.) or its vehicle (saline) was administered 30 min before other treatments in rats whose right SN (A) or ION (B) had been ligated 2 weeks before. Five different treatment groups were made: (1) propranolol + HEC + saline, (2) propranolol + agomelatine (45 mg/kg) + saline, (3) propranolol + HEC + gabapentin (50 mg/kg), (4) agomelatine + gabapentin, (5) propranolol + agomelatine + gabapentin. Pressure threshold values (as g) were determined as described in the legend to Figure 1. Each point is the mean ± SEM of independent determinations in n rats for each condition. ^∗P < 0.05, compared with pressure threshold values determined just prior to treatments (0 on abscissa), one way ANOVA with repeated measures, Dunnett’s test. C on abscissa: intact healthy rats before surgery. In both CCI-SN (A) and CCI-ION (B) rats, ‘saline+HEC’-treated groups were also included, but, for the sake of clarity, corresponding data are not represented as they superimposed with those obtained in rats treated with propranolol + HEC + saline. Right panels: AUC values calculated from the respective time-course curves: (1) propranolol + HEC + saline [n = 5 (A)/5 (B)]; (2) propranolol + agomelatine + saline (n = 5/5); (3) propranolol + HEC + gabapentin (n = 5/5); (4) agomelatine + gabapentin (n = 8/7); 5 = propranolol + agomelatine + gabapentin (n = 14/11). A- CCI-SN : one way ANOVA [F(4,32) = 73.51, P < 0.0001] followed by Tukey’s test (^∗∗∗P < 0.001); B- CCI-ION: one way ANOVA [F(4,28) = 25.39, P < 0.0001] followed by Tukey’s test (^∗∗∗P < 0.001).

FIGURE 8

Effects of ICI 118551 alone or co-administered with agomelatine, gabapentin or ‘agomelatine + gabapentin’ on mechanical allodynia in CCI-SN (A) or CCI-ION (B) rats. Left panels: ICI 118551 (5 mg/kg i.p.) or its vehicle (saline) was administered 30 min before other treatments in rats whose right SN (A) or ION (B) had been ligated 2 weeks before. Five different treatment groups were made: (1) ICI 118551 + HEC + saline, (2) ICI 118551 + agomelatine (45 mg/kg) + saline, (3) ICI 118551 + HEC + gabapentin (50 mg/kg), (4) agomelatine + gabapentin, (5) ICI 118551 + agomelatine + gabapentin. Pressure threshold values (as g) were determined as described in the legend to Figure 1. Each point is the mean ± SEM of independent determinations in n rats for each condition. ^∗P < 0.05, compared with pressure threshold values determined just prior to drug injection, one way ANOVA with repeated measures, Dunnett’s test. C on abscissa: intact healthy rats before surgery. In both CCI-SN (A) and CCI-ION (B) rats, ‘saline+HEC’-treated groups were also included, but, for the sake of clarity, corresponding data are not represented as they superimposed with those obtained in rats treated with ICI 118551 + HEC + saline. Right panels: AUC values calculated from the respective time-course curves: 1 = ICI 118551 + HEC + saline [n = 5 (A), n = 5 (B)]; 2 = ICI 118551 + agomelatine + saline (n = 5/5); 3 = ICI 118551 + HEC + gabapentin (n = 5/6); 4 = agomelatine + gabapentin (n = 6/7); 5 = ICI 118551 + agomelatine + gabapentin (n = 13/7). A- CCI-SN : one way ANOVA [F(4,29) = 310.9, P < 0.0001] followed by Tukey’s test (^∗∗∗P < 0.001); B- CCI-ION : one way ANOVA [F(4,25) = 39.35, P < 0.0001] followed by Tukey’s test (^∗∗∗P < 0.001).