TRAIL agonists rescue mice from radiation-induced lung, skin, or esophageal injury

Abstract

Radiotherapy can be limited by pneumonitis, which is impacted by innate immunity, including pathways regulated by TRAIL death receptor DR5. We investigated whether DR5 agonists could rescue mice from toxic effects of radiation and found that 2 different agonists, parenteral PEGylated trimeric TRAIL (TLY012) and oral TRAIL-inducing compound (TIC10/ONC201), could reduce pneumonitis, alveolar wall thickness, and oxygen desaturation. Lung protection extended to late effects of radiation including less fibrosis at 22 weeks in TLY012-rescued survivors versus unrescued surviving irradiated mice. Wild-type orthotopic breast tumor-bearing mice receiving 20 Gy thoracic radiation were protected from pneumonitis with disappearance of tumors. At the molecular level, radioprotection appeared to be due to inhibition of CCL22, a macrophage-derived chemokine previously associated with radiation pneumonitis and pulmonary fibrosis. Treatment with anti-CCL22 reduced lung injury in vivo but less so than TLY012. Pneumonitis severity was worse in female versus male mice, and this was associated with increased expression of X-linked TLR7. Irradiated mice had reduced esophagitis characterized by reduced epithelial disruption and muscularis externa thickness following treatment with the ONC201 analog ONC212. The discovery that short-term treatment with TRAIL pathway agonists effectively rescues animals from pneumonitis, dermatitis, and esophagitis following high doses of thoracic radiation exposure has important translational implications.

Keywords: Cancer; Innate immunity; Oncology; Radiation therapy.

Figure 1. Experimental design and preliminary results of the mouse strains and treatment cohorts.

(A) Hypothesized outcomes regarding lung protection of C57BL/6, Dr5^–/–, and Trail^–/– mice given a whole-thorax x-ray irradiation dose of 20 Gy and either 100 mg/kg of ONC201 weekly, 10 mg/kg of TLY012 twice weekly, or no treatment. (B) Experimental timeline. Mice received their first treatment an hour before radiation, and were treated with either ONC201 once a week or TLY012 twice weekly until sacrifice on day 13 after irradiation. (Created in BioRender. Strandberg J. 2024. https://BioRender.com/p52s542.) (C) First observations of suppression of radiation pneumonitis in H&E-stained lung tissue of male mice by short-term treatment with TRAIL pathway agonists as indicated (n = 2 per treatment per genotype) (original magnification, ×20).

Figure 2. Protection of mice from radiation pneumonitis and its dependence on genetic strain and TRAIL pathway agonist.

(A and B) Representative H&E stains of lung tissue from male and female mice from C57BL/6, Dr5^–/–, and Trail^–/– backgrounds treated with ONC201, TLY012, or control gavage (n = 2 per sex per genotype per treatment) 13 days after irradiation (original magnification, ×40). (C) Quantification of percentage inflammation of Trail^–/– female mice treated with TLY012 or control (n = 7 per treatment per group) showed significant decrease (P = 0.0385) in inflammation 2 weeks after thoracic irradiation of 20 Gy (1-tailed Mann-Whitney test). (D) Representative images of IHC staining of TLR7 in lung tissue from Trail^–/– female and male mice at 2 weeks after irradiation (original magnification, ×10). Scale bars: 200 μm. (E) Quantification of positive staining of TLR7 in mouse lung separated by treatment group (n = 8 per treatment per group) (1-way ANOVA with Tukey’s post hoc test). (F) When separated by sex (n = 16 per sex), there was a statistically significant increase in female mice (P = 0.0045). Values are mean ± SEM (unpaired 2-tailed t test).

Figure 3. Rescue from long-term effects of radiation pneumonitis and radiation dermatitis by treatment of TLY012 up to 22 weeks after irradiation.

(A–D) Male Trail^–/– mice were treated with a single thoracic x-ray irradiation dose of 18 Gy and treated with TLY012 (B and D) or controls (A and C) for 22 weeks after radiation (n = 3, n = 2). Lung tissue stained with Masson’s trichrome and imaged at ×10 (A and B) or ×20 (C and D) original magnification. Red, muscle fibers; bright blue, collagen; dark red/blue, nuclei. (E) Mice that developed radiation burns 3 weeks after irradiation on the chest were euthanized (control n = 6, TLY012 n = 2). (F) Kaplan-Meier curve of Trail^–/– female mice rescued from radiation dermatitis after a single whole-thorax x-ray irradiation dose of 15 Gy treated with TLY012 compared with control (n = 10 per treatment per group). (G) Representative H&E images (original magnification, ×10) of skin from chest of female Trail^–/– mice 2 weeks after irradiation. Epidermal necrosis (arrows), subepidermal cleft (asterisk), and dermal inflammation involving superficial dermis and focally subcutaneous adipose tissue (stars) were observed in the control group (original magnification, ×20). Scale bars: 200 μm.

Figure 4. Efficacy of chest irradiation using an orthotopic immune-competent breast cancer model with prevention of pneumonitis by TLY012 and reduction of CCL22.

(A) Experimental timeline. Female C57BL/6 mice orthotopically injected with e0771 on day 0 and when tumors reached 2–5 mm in size mice were irradiated with a whole-thorax irradiation dose of 20 Gy and treated with either TLY012, ONC201, or the combination (n = 3–4 per treatment per group). (Created in BioRender. Strandberg J. 2024. https://BioRender.com/c65v064.) (B and C) Tumors were removed after mice were euthanized 18 days after cell injection, and weight and volume were calculated. (D) Pulse oximetry readings before radiation and 9 days after radiation showed oxygen saturation more conserved in the TLY012-treated group compared with irradiated controls. Values are mean ± SEM. (E) Statistical analysis of cytokine fold change showed significant decrease in levels of MDC/CCL22 (P = 0.035) in comparison with irradiated controls (1-way ANOVA with Tukey’s post hoc test).

Figure 5. Treatment with anti-CCL22 in Trail^–/– mice provided partial rescue from radiation pneumonitis.

(A) Experimental timeline of Trail^–/– female mice treated with 20 μg of anti-CCL22 in 500 μL of saline or 20 μg of goat IgG in 500 μL of saline every other day for 2 weeks (n = 5 per treatment per group). (Created in BioRender. Strandberg J. 2024. https://BioRender.com/j52q830.) (B) H&E images of each mouse 2 weeks after thoracic irradiation (original magnification, ×20). Scale bar: 100 μm. (C) Quantification of inflammatory scores provided by a blinded pathologist showed decrease in inflammation in Trail^–/– female mice treated with anti-CCL22 (n = 5 per treatment per group) but not to a significant extent (1-tailed Mann-Whitney test) (P = 0.1032). Values are mean ± SEM.

Figure 6. Respiration-gated imaging in mice irradiated with a whole-thorax x-ray irradiation dose of 15 Gy with or without TLY012 treatment.

(A) 3D reconstruction of mouse lungs from μCT images during exhale and inhale portions of the breathing cycle. Images were subjected to 7% opacity filter in CTVol software. (B) Representative μCT images of female Trail^–/– mouse lungs unirradiated, irradiated with 15 Gy, or irradiated with 15 Gy and rescued with TLY012 treatment during inhale duration of the breathing cycle (n = 2 per group). Mouse weights were 19.6 g, 24.1 g, and 22.4 g, respectively. H&E slides of lungs reinflated postmortem were prepared. Original magnification, ×10.

Figure 7. Reduced severity of radiation esophagitis with ONC212 for 2 weeks after a toxic thoracic x-ray irradiation dose in mice.

(A) Timeline of female C57BL/6 mice irradiated with a single whole-thorax dose of 20 Gy followed by treatment of either control gavage or 25 mg/kg of ONC212 (n = 8 per treatment per group). (Created in BioRender. Strandberg J. 2024. https://BioRender.com/z31h016.) (B) Weights of mice before radiation and 2 weeks after irradiation up until sacrifice at day 13 (n = 8 per treatment per group). (C) Kaplan-Meier curve of mice after 20 Gy thoracic irradiation treated with control gavage or 25 mg/kg of ONC212 (n = 8 mice per treatment per group). Mice were sacrificed before harvest date for greater than 20% weight loss. (D and E) Representative H&E images of cross section of distal esophagus of mice treated with control gavage or 25 mg/kg of ONC212 after 20 Gy thoracic irradiation at ×10 (D) and ×40 (E) original magnification. (F) Representative image of cross section of esophagus at ×10 original magnification highlighting tissue layers examined. (G and H) Quantification of thickness (μm) of different esophageal layers was determined by measurement of 5 random areas around the circumference of the esophagus for each mouse. There was a significant decrease in thickness of the muscularis externa (G) and muscularis mucosa (H) layers in mice treated with ONC212 (P = 0.0099, 0.0366, respectively). (I) Heatmap of cytokine fold change between control and ONC212-treated mice (n = 8 mice per treatment per group).