Toll-like receptor 5 agonist protects mice from dermatitis and oral mucositis caused by local radiation: implications for head-and-neck cancer radiotherapy

Abstract

Purpose: Development of mucositis is a frequent side effect of radiotherapy of patients with head-and-neck cancer. We have recently reported that bacterial flagellin, an agonist of Toll-like receptor 5 (TLR5), can protect rodents and primates from acute radiation syndrome caused by total body irradiation. Here we analyzed the radioprotective efficacy of TLR5 agonist under conditions of local, single dose or fractionated radiation treatment.

Methods and materials: Mice received either single-dose (10, 15, 20, or 25 Gy) or fractioned irradiation (cumulative dose up to 30 Gy) of the head-and-neck area with or without subcutaneous injection of pharmacologically optimized flagellin, CBLB502, 30 min before irradiation.

Results: CBLB502 significantly reduced the severity of dermatitis and mucositis, accelerated tissue recovery, and reduced the extent of radiation induced weight loss in mice after a single dose of 15 or 20 Gy but not 25 Gy of radiation. CBLB502 was also protective from cumulative doses of 25 and 30 Gy delivered in two (10 + 15 Gy) or three (3 × 10 Gy) fractions, respectively. While providing protection to normal epithelia, CBLB502 did not affect the radiosensitivity of syngeneic squamous carcinoma SCCVII grown orthotopically in mice. Use of CBLB502 also elicited a radiation independent growth inhibitory effect upon TLR5-expressing tumors demonstrated in the mouse xenograft model of human lung adenocarcinoma A549.

Conclusion: CBLB502 combines properties of supportive care (radiotherapy adjuvant) and anticancer agent, both mediated via activation of TLR5 signaling in the normal tissues or the tumor, respectively.

Figure 1. The dynamics of mouse body weight changes after radiation treatment of the head and neck area

Mouse body weight changes after single irradiation with 15 (A, n=4), 20 (B, n=6) or 25 Gy (C, n=6) with and without prior CBLB502 injection (2.5 µg/mouse). Results representative of at least two analogous experiments performed with each of the three irradiation regiments are presented. (D). Dependence of the radioprotective effect of CBLB502 on radiation dose is presented as a minimal body weight observed at any time-point in mice irradiated with indicated doses of irradiation. Available results permit calculating DRF equal 1.3 based on 5% body weight loss. (*) The difference between irradiated groups, with and without CBLB502 treatment, is significant (p<0.05).

Figure 2. The protective effect of CBLB502 against radiation damage to tissues in the head and neck area of mice

(A). Mouse body weight changes after 30 Gy local head and neck radiation treatment, with and without CBLB502 (1 µg/mouse, n=6), given in three 10 Gy fractions over three consecutive days. Results representative of at least two analogous experiments are presented. (*) The difference between irradiated groups, with and without CBLB502 treatment, is significant (p<0.05). (B). Upper row. Photographs of irradiated NIH-Swiss mice with or without CBLB502 pretreatment. Lower row. H&E stained skin sections from the neck area. Mice were locally irradiated with 30 Gy X-ray given in three daily 10 Gy fractions. Images were taken on day 14 after the first radiation treatment. Lower row. H&E stained histological sections of the skin sections. (C). H&E stained histological sections of tip, dorsal and ventral epithelium of mouse tongue obtained 11 days after the first of three 10 Gy fractions of 30 Gy total X-ray irradiation. For B (lower panel) and C, H&E stained tissue sections from healthy untreated (u/t) mice were used as a positive control. The pictures of the untreated mice show stratified squamous epithelium with basal layer of small, dark cells from which the mucosa is constatnly renewed and overlaying larger spinous cells and superficial flattened cells. The irradiated and PBS treated mice reveal atrophic and degenerative changes in the basal and overlaying cells with loss of cells, pyknotic and dysplastic nuclei, erosions and hemorrhage. In contrast, the irradiated and CBLB502 treated mice have mucosa with minimal radiation associated changes and morphology closer to the normal, with better preserved, regular basal and overlaying epithelial cells. Tissues from non-irradiated CBLB502-treated animals were histologically indistinguishable from control non-irradiated vehicle-treated animals (data not shown).

Figure 3. Anti-tumor effect of CBLB502 and radiation treatment combined with CBLB502

(A). Visualization of GFP-expressing SCCVII tumors in the neck area of syngeneic C3H mice two weeks post implantation by whole body fluorescence imaging system (Lightools Research, Encinitas, CA) (left) and MRI (left). Arrows point to the location of the tumor on both images. (B). Changes to SCCVII tumor volume after treatment with CBLB502 and radiation both individually and in combination. Representative results of one out of two analogous experiments are presented, n=6. (C). The dynamics of xenogenic control (control shRNA expressing) A549 cells (left panel) and A549-shTLR5 (middle panel) tumor growth in athymic nude mice was recorded after 3 daily injections of CBLB502 or PBS (vehicle control). The tumor volume data are presented relative to the initial tumor volume, n=10. (*) The difference between tumor volumes in groups with and without CBLB502 treatment is significant (p<0.05). Luciferase reporter assay for NF-kappaB activation in A549 and A549-shTLR5 cells was performed 5 hours after in vitro treatment with different doses of CBLB502 (right panel).