Analysis of the effects of a tricyclic antidepressant on secondary sleep disturbance induced by chronic pain in a preclinical model

Abstract

Chronic pain and sleep have a bidirectional relationship that promotes a vicious circle making chronic pain more difficult to treat. Therefore, pain and sleep should be treated simultaneously. In our previous study, we suggested that hyperactivation of ascending serotonergic neurons could cause secondary sleep disturbance in chronic pain. This study aimed to demonstrate the effects of a tricyclic antidepressant (amitriptyline) and a selective 5-hydroxy-tryptamine 2A (5-HT2A) antagonist (MDL 100907) that adjust serotonergic transmission, on secondary sleep disturbance induced in a preclinical chronic pain model. We produced a chronic neuropathic pain model by partial sciatic nerve ligation in mice, analyzed their electroencephalogram (EEG) and electromyogram (EMG) using the SleepSign software, and evaluated the sleep condition of the pain model mice after administration of amitriptyline or MDL 100907. Amitriptyline improved thermal hyperalgesia and the amount of sleep, especially non-REM sleep. Time change of normalized power density of δ wave in the nerve ligation group with amitriptyline administration showed a normal pattern that was similar to sham mice. In addition, MDL 100907 normalized sleep condition similar to amitriptyline, without improvement in pain threshold. In conclusion, amitriptyline could improve sleep quantity and quality impaired by chronic pain. 5-HT2A receptor antagonism could partially contribute to this sleep improvement, but is not associated with pain relief.

Fig 1. Change of thermal hyperalgesia in neuropathic pain model mice.

Schematic diagram of sciatic nerve ligation (A). Plantar tests were performed before nerve ligation surgery, 7, 8 and 9 days after surgery. Data are expressed as means ± SEM. Data are analyzed using two-way ANOVA followed by Bonferroni’s post hoc test and considered statistically significant at ****P < 0.0001 when comparing the sham (n = 8) and nerve ligation groups (n = 8) (B).

Fig 2. Amitriptyline improved thermal hyperalgesia in neuropathic pain model.

Plantar tests were performed before nerve ligation surgery, test day 7 with vehicle, and test day 8 with amitriptyline administration (n = 8) (A). Schematic diagram of plantar test (B). Paw withdrawal latencies were plotted for every test (C). Data are expressed as means ± SEM. Data are analyzed using two-way ANOVA followed by Bonferroni’s post hoc test and considered statistically significant at ****P < 0.0001 when comparing the sham and ligation groups.

Fig 3. Amitriptyline improved quantity of sleep in neuropathic pain model.

EEG and EMG were recorded for 24 hours with administration of vehicle (test day 7) and amitriptyline (test day 9) (A). Schematic diagram of EEG/EMG implantation (B). The mean value per hour of wake, REM sleep, and non-REM sleep time are represented for the sham (n = 7) and nerve ligation groups (n = 7) (C). Data are expressed as means ± SEM. Data are analyzed using two-way ANOVA followed by Bonferroni’s post hoc test and considered statistically significant at ***P < 0.001, and ****P < 0.0001 when comparing the groups.

Fig 4. Amitriptyline improved quality of sleep in neuropathic pain model.

Distribution of EEG power density in each frequency during non-REM and REM sleep. Data are expressed as means only. *P < 0.05 at 4.2 Hz and 4.7 Hz when comparing sham-vehicle and ligation-vehicle groups (A). Time changes of the normalized power density of δ wave were plotted every 3 hours in sham (n = 5–7) and nerve ligation (n = 7) mice. Data are expressed as means ± SEM (B) and are analyzed using two-way ANOVA followed by Bonferroni’s post hoc test and considered statistically significant at **P < 0.01 when comparing sham-vehicle and ligation-vehicle groups, ^##P < 0.01 when comparing ligation-vehicle and ligation amitriptyline group, and ⁺⁺⁺P < 0.001 when comparing the sham-amitriptyline and ligation-vehicle groups.

Fig 5. Selective 5HT_2A receptor antagonist improved thermal hyperalgesia in neuropathic pain model.

Plantar tests were performed before nerve ligation surgery, on test day 7 with vehicle, and on test day 8 with MDL 100907 administration (n = 8) (A). Paw withdrawal latencies were plotted for every test. Data are expressed as means ± SEM. Data are analyzed using two-way ANOVA followed by Bonferroni’s post hoc test and considered statistically significant at ****P < 0.0001 when comparing the sham and ligation groups (B).

Fig 6. Selective 5HT_2A receptor antagonist improved quantity of sleep in neuropathic pain model.

EEG and EMG were recorded for 24 hours with administration of vehicle (test day 7) and MDL 100907 (test day 9) (A). The mean value per hour of wake, REM sleep, and non-REM sleep time are represented for sham (n = 7) and nerve ligation groups (n = 7). Data are expressed as means ± SEM. Data are analyzed using two-way ANOVA followed by Bonferroni’s post hoc test and considered statistically significant at ***P < 0.001, and ****P < 0.0001 when comparing all groups (B).

Fig 7. Selective 5HT_2A receptor antagonist improved quality of sleep in neuropathic pain model.

Distribution of EEG power density in each frequency during non-REM and REM sleep. Data are expressed as means only (A). Time changes of the normalized power density of δ wave are plotted every 3 hours in sham (n = 6) and nerve ligation (n = 5) mice. Data are expressed as means ± SEM (B). Data are analyzed using two-way ANOVA followed by Bonferroni’s post hoc test and considered statistically significant at *P < 0.05 when comparing the sham-vehicle and ligation-vehicle groups, ^#P < 0.05 when comparing the ligation-vehicle and ligation-MDL 100907 groups, and ⁺P < 0.05 when comparing the sham-amitriptyline and ligation-vehicle groups.