Protection from Radiation-Induced Pulmonary Fibrosis by Peripheral Targeting of Cannabinoid Receptor-1

Abstract

Radiation-induced pulmonary fibrosis (RIF) is a severe complication of thoracic radiotherapy that limits its dose, intensity, and duration. The contribution of the endocannabinoid signaling system in pulmonary fibrogenesis is not known. Using a well-established mouse model of RIF, we assessed the involvement of cannabinoid receptor-1 (CB1) in the onset and progression of pulmonary fibrosis. Female C57BL/6 mice and CB1 knockout mice generated on C57BL/6 background received 20 Gy (2 Gy/min) single-dose thoracic irradiation that resulted in pulmonary fibrosis and animal death within 15 to 18 weeks. Some C57BL/6 animals received the CB1 peripherally restricted antagonist AM6545 at 1 mg/kg intraperitoneally three times per week. Animal survival and parameters of pulmonary inflammation and fibrosis were evaluated. Thoracic irradiation (20 Gy) was associated with marked pulmonary inflammation and fibrosis in mice and high mortality within 15 to 18 weeks after exposure. Genetic deletion or pharmacological inhibition of CB1 receptors with a peripheral CB1 antagonist AM6545 markedly attenuated or delayed the lung inflammation and fibrosis and increased animal survival. Our results show that CB1 signaling plays a key pathological role in the development of radiation-induced pulmonary inflammation and fibrosis, and peripherally restricted CB1 antagonists may represent a novel therapeutic approach against this devastating complication of radiotherapy/irradiation.

Keywords: AM6545; cannabinoid receptor 1; peripherally restricted CB1 antagonist; radiation-induced pulmonary fibrosis; thoracic irradiation.

Figure 1.

Pharmacological or genetic targeting of cannabinoid receptor-1 (CB₁) decreases radiation-induced pulmonary inflammation in mice administered a single thoracic dose (20 Gy) of irradiation. Here and thereafter, lung tissues from control, irradiated (20 Gy, thoracic, 2 Gy/min), AM6545-treated, and CB₁^−/− mice were taken when solvent-treated irradiated C57BL/6 mice reached radiation-induced pulmonary fibrosis (RIF) stage. (A) Lung tissue immune cell infiltration is decreased in animals that received the peripheral CB₁ antagonist AM6545 (1 mg/kg intraperitoneally three times per week) and in CB₁^−/− mice. Hematoxylin and eosin staining of lung sections (representative images; original magnification: ×200). (B and C) BAL fluid was collected and assessed for total protein and cell counts. Compared with irradiated control animals, CB₁ receptor inhibition with AM6545 or the lack of functional CB₁ receptors in knockout animals substantially decreased bronchoalveolar lavage (BAL) cell count (B) and protein concentration (C). (D and E) Lung tissue TNF-α messenger RNA (mRNA) level (D) and TNF-α BAL level (E) are up-regulated at RIF stage and normalized by CB₁ inhibition with AM6545 or in CB₁^−/− animals. ***P < 0.001 versus nonirradiated control animals. ^##P < 0.01 and ^###P < 0.001 versus irradiated control animals. N.S., not significant.

Figure 2.

Pharmacological or genetic targeting of CB₁ receptors decreases radiation-induced oxidative and nitrative stress in the lung. Lung tissues were stained for markers of oxidative (4-hydroxynonenal [4-HNE]) and nitrative (S-nitrotyrosine) stress. RIF results in a marked increase in lipid peroxidation (4-HNE staining) and protein nitrosylation (nitrotyrosin staining) (original magnification: ×200). Animal treatment with AM6545 or the lack of CB₁ receptors substantially decreases lung tissue reactive oxygen species and reactive nitrogen species formation, as evidenced by normalization of tissue staining for 4-HNE and nitrotyrosine (representative images are shown).

Figure 3.

Pharmacological or genetic targeting of CB₁ normalizes radiation-induced dysregulation in the expression of genes associated with active fibrogenesis and inflammation at RIF stage. (A) Lung tissue mRNA levels for genes linked to active fibrogenesis were quantified by quantitative RT-PCR (qRT-PCR), and their expression was normalized to that of 18S. The normalized expression of each gene was further normalized to the gene expression in solvent-treated control C57BL/6 mice. *P < 0.05, **P < 0.01, and ***P < 0.001 versus nonirradiated control animals. ^#P < 0.05, ^##P < 0.01, and ^###P < 0.001 versus irradiated control animals. (B) Cluster analysis of the effect of pharmacological or genetic targeting of CB₁ receptors on dysregulation in the expression of genes associated with active fibrogenesis and inflammation at RIF stage. Cluster analysis was performed using BioRad software for processing qRT-PCR data. α-SMA, α-smooth muscle actin; AM, AM6545; Coll-1, collagen α2 type I; CTGF, connective tissue growth factor; IRRAD, irradiation; Solv, solvent; TGF-β, transforming growth factor β.

Figure 4.

Pharmacological or genetic targeting of CB₁ decreases radiation-induced pulmonary fibrosis in mice administered a single thoracic dose (20 Gy) of irradiation. (A) Collagen deposition is decreased in animals that received the peripheral CB₁ antagonist AM6545 (1 mg/kg intraperitoneally three times per week) and in CB₁^−/− mice. Sirius red staining of lung sections. (B–D) The lung tissue levels of the markers of fibrogenesis α-SMA (B), Coll α2 (C), acid-soluble collagen (D), and BAL level of TGFβ (E) are decreased or completely normalized in C57Bl/6 mice treated with AM6545 and in CB₁^−/− mice. GAPDH, glyceraldehyde 3-phosphate dehydrogenase. ^#P < 0.05, ^##P < 0.01 versus irradiated control animals.

Figure 5.

Pharmacological or genetic targeting of CB₁ improves survival of mice subjected to thoracic irradiation. Female C57BL/6 (CB₁^+/+) or CB₁^−/− mice weighing 18 to 22 g were subjected to a single 20 Gy (2 Gy/min) dose of thoracic irradiation. Some animals received AM6545 (1 mg/kg intraperitoneally three times per week) as a solution in 2% Tween-80 in saline; control mice received just the solvent. Inhibition of CB₁ receptors with AM6545 or the lack of CB₁ receptors extended animal survival. Statistical difference was calculated using a log-rank (Mantel-Cox) test.