Formoterol dynamically alters endocannabinoid tone in the periaqueductal gray inducing headache

Abstract

Background: Headache is a pain disorder present in populations world-wide with a higher incidence in females. Specifically, the incidences of medication overuse headache (MOH) have increased worldwide. Comorbidities of MOH include photosensitivity, anxiety, "brain fog", and decreased physical activity. The FDA-approved long-lasting selective β₂-adrenergic receptor agonist, formoterol, is currently approved for use in severe asthma and chronic obstructive pulmonary disease. Recently, interest in repurposing formoterol for use in other disorders including Alzheimer's disease, and neuropathic pain after spinal cord injury and traumatic brain injury has gained traction. Thus, revisiting known side-effects of formoterol, like headache and anxiety, could inform treatment paradigms. The endocannabinoid (eCB) system is implicated in the etiology of preclinical headache, with observed decreases in the circulating levels of endogenous cannabinoids, referred to as Clinical Endocannabinoid Deficiency. As cross-talk between the eCB system and adrenergic receptors has been reported, this study investigated the role of the eCB system and ability of formoterol to induce headache-like periorbital allodynic behavior.

Methods: Female 8-week-old C57Bl/6J mice were treated daily with formoterol (0.3 mg/kg, i.p.) for up to 42-days, during which they were assessed for periorbital allodynia, open field/novel object recognition, and photosensitivity. At the end of the study, the periaqueductal grey (PAG), a brain region known to contribute to both headache induction and maintenance, was collected and subjected to LC-MS to quantify endocannabinoid levels.

Results: Mice exhibited periorbital allodynia at nearly all time points tested and photosensitivity from 28-days onward. Levels of endocannabinoids, anandamide (AEA) and 2-arachidonoylglycerol (2-AG), along with cannabinoid receptor 1 (CB₁R) expression were altered by both age and upon treatment with formoterol. Administration of FAAH/MAGL inhibitors, to target the eCB system, and a non-selective cannabinoid receptor agonist, WIN 55,212 reversed the formoterol-induced periorbital allodynia.

Conclusions: These results suggest that formoterol is dysregulates eCB tone to drive headache-like periorbital allodynic behaviors. These results could help inform preventative treatment options for individuals receiving formoterol, as well as provide information on the interaction between the eCB and adrenergic system.

Keywords: Adrenergic; Endocannabinoid; Formoterol; Headache; In vivo; Mouse; PAG.

Fig. 1

Formoterol induces periorbital allodynia 7 days after drug initiation out through 42 days of treatment. 8-week-old female C57Bl/6J mice were acclimated to the assay chambers and Von Frey filaments 3 times prior to baseline measurements; experimental measurements were taken weekly post-drug initiation for 42-days, formoterol (0.03 mg/kg, i.p.) or vehicle were administered daily for 42 days. A Timeline of experimental setting. B Formoterol treated mice began to exhibit headache-like periorbital allodynic behavior 7-days post drug initiation as compared to vehicle treated and maintained this periorbital sensitivity throughout the 42-day testing period. Data represented as mean of threshold (g) ±SEM. (*denotes p$\le$0.05, **denotes p$\le$0.01, compared to vehicle treated mice)

Fig. 2

Chronic administration of formoterol induces photophobia as assessed via the light/dark box assay. 8-week-old female C57Bl/6J mice were acclimated to the assay chambers twice for 30m prior to baseline measurements, experimental measurements were then taken weekly post drug initiation for 42-days. Mice were tested at each lux (23.8, 53.1, 107.7, 163, 300.42, and 735) for 5m per lux. At all lux tested naïve mice showed no difference in time spent in light (A). At the 107.7 lux, formoterol treated mice spent less time in the light at days 35 and 42 (B), the light aversion index shows moderate habituation ay days 7 and 14 for both the formoterol and vehicle treated mice (C). Formoterol treated mice spent more time in the dark at days 28, 35 and 42 than the vehicle treated mice (D). In the lux curve, starting 28 days post drug initiation (DPI), formoterol treated mice spent significantly less time in the light at 23.8, 53.1, 107.7, 300.42 and 735 lux as compared to vehicle control (E). 35 DPI, formoterol treated mice spent significantly less time in light at lux 107.7 and 735 (F) and at 42 DPI, formoterol treated mice spent significantly less time in the light at lux 53.1, 107.7, 300.42 and 753 as compared to vehicle control (G). Chronic exposure to formoterol does induce light avoidance behaviors starting after 28 days of drug application. Data represented as time in light (s) ±SEM (*denotes p$\le$0.05, **denotes p$\le$0.01, ***denotes p$\le$0.001 compared to vehicle treated mice)

Fig. 3

Anxiety-like behaviors are not observed under long-term treatment with formoterol. 8-week-old female C57Bl/6J mice were baselined for 300s/test prior to drug initiation; experimental measurements were taken weekly post-drug initiation for 42-days, formoterol (0.03 mg/kg, i.p.) or vehicle were administered daily for 42 days. A Throughout the 42-day study no difference between treatment groups was observed in terms of time spent in the closed arm. B Formoterol treated mice spent more time in the open arm at days 14, 35, and 42. Data represented as time spent in arm/baseline ± SEM. (*denotes p$\le$0.05, compared to vehicle control)

Fig. 4

AEA, 2-AG, and CB₁R expression in the PAG changes with age and with formoterol treatment. The PAGs from C57Bl/6J female age-matched naïve mice were harvested and eCB lipids assessed via LC-MS, (A) dynamic age-related changes were observed for both AEA and 2-AG. Data represented as mean ± SEM in pmol/mg unit. B In the naïve mice, protein expression of cannabinoid receptor 1 (R) changes with age. Data represented as mean ± SEM. β-actin was used as loading control. C 7d post drug initiation, AEA levels are significantly decreased in formoterol treated mice versus vehicle treated mice but at 6w post drug initiation AEA levels are significantly increased in formoterol treated mice. D 2-AG level within the PAG is significantly decreased in formoterol treated mice versus vehicle treated mice at 7-days, but by 42-days there is no difference observed between experimental groups. E expression within PAG in formoterol treated mice is significantly decreased as compared to vehicle treated mice both 7d and 6w post drug initiation. Data represented as percentage of naïve ± SEM. (*denotes p$\le$0.05, **denotes p$\le$0.01, ***denotes p$\le$0.001, **** denotes p$\le$0.0001 compared to age-matched or vehicle treated mice)

Fig. 5

Formoterol treatment induced changes in the expression of ADRB2 receptor within the PAG 7-days post-initiation. The PAGs from C57Bl/6J female age-matched naïve mice were harvested and ADRB2 receptor protein expression assessed by Western immunoblotting, (A) age-related changes were not observed. Data represented as mean ± SEM. β-actin was used as loading control. B At 7 DPI, there is a significant increase in the expression of ADRB2 receptor in the formoterol treated mice versus the vehicle treated mice. No significant difference was observed in the expression of ADRB2 receptor 42 DPI, compared to vehicle control. Data represented as percentage of naïve ± SEM. β-actin was used as loading control. (*denotes p$\le$0.05 compared to age-matched or vehicle treated mice)

Fig. 6

MAGL/FAAH and CB₁R inhibitors reversed formoterol induced periorbital allodynia. 8-week-old female C57Bl/6J mice were acclimated to the assay chambers and Von Frey filaments 3 times prior to baseline measurements, followed by treatment with formoterol for 7 days or 42 days. Mice given formoterol for 7 days received either MJN/JNJ, WIN,55,212-2, or vehicle on the last day of formoterol administration. A At time of baseline, mice did not exhibit headache-like periorbital allodynic behaviors; after receiving formoterol for 7 days, increase in periorbital sensitivity was observed. A 30-minutes after administration of MJN/JNJ or WIN, a significant decrease in headache-like periorbital allodynic behavior was observed, with no significant difference in threshold values between day 0 baseline and 1 hour post inhibitor administration for these treatments. B MJN/JNJ administration induced significant increase in the level of AEA within the PAG compared to vehicle control. C There are no significant changes observed in the levels of 2-AG in either treatment group compared to vehicle control. A second cohort of mice received formoterol for 6 weeks and were then given either WIN or vehicle. PAG was harvested 5 hours later. D In the 6w study, 30 minutes after administration of WIN 55,212, a significant decrease in headache-like periorbital allodynic behaviors was observed but values did not reach pre-drug administration baseline values until 2.5 hours post-treatment with WIN55,212. In a separate set, mice received formoterol daily for 7 days, at which point they received either MJN/JNJ, WIN, or vehicle. PAG was harvested after 5 hours and subjected to LC-MS to measure endocannabinoid levels. E WIN treated mice exhibited an increase in levels of AEA versus vehicle treated mice, and (F) levels of 2-AG were not significantly different between the two groups. Data represented as threshold (g) ± SEM or percentage of naive ± SEM. (### denotes p$\le$0.001 compared to Day 0 BL; **denotes p$\le$0.01, **** denotes p$\le$0. compared to form+veh)