Inhibition of the TLR4/RAGE pathway by clearance of extracellular HMGB1 is a potential therapeutic target for radiation-damaged salivary glands

Abstract

Introduction: We recently developed a new therapy using effective-mononuclear cells (E-MNCs) and demonstrated its efficacy in treating radiation-damaged salivary glands (SGs). The activity of E-MNCs in part involves constituent immunoregulatory -CD11b/macrophage scavenger receptor 1(Msr1)-positive-M2 macrophages, which exert anti-inflammatory and tissue-regenerating effects via phagocytic clearance of extracellular high mobility group box 1 (HMGB1). Focusing on the phenomena, this study investigated significance of regulating the HMGB1/toll-like receptor 4 (TLR4)/receptor for advanced glycation end products (RAGE) signaling pathway in the treatment of SG dysfunction caused by radiation damage.

Methods: E-MNCs were transplanted into radiation-damaged mice SGs, and changes of TLR4/RAGE expression were observed. Furthermore, the activation of downstream signals was investigated in both intact SGs and cultured SG epithelial cells after irradiation. Subsequently, TLR4-knock-out (KO) mice were employed to examine how HMGB1/TLR4/RAGE signaling affected damage progression.

Results: Expression of both TLR4 and RAGE was diminished in ductal cells and macrophages/vascular endothelial cells of damaged SGs with E-MNC transplantation, respectively. Meanwhile, expression of TLR2/4 and RAGE in damaged SGs markedly increased in association with extracellular HMGB1 accumulation. Downstream signals were activated, and intranuclear localization of phospho-nuclear factor-kappa B (p-NF-KB) in ductal cells and production of IL-6, tumor necrosis factor-α (TNF-α), and interferon-γ (IFN-γ) were observed. Additionally, culture supernatant of irradiated cultured SG epithelial cells contained damaged associated molecular pattern (DAMP)/senescence-associated secretory phenotype (SASP) factors. Treatment of cultured SG epithelial cells with this supernatant activated TLR4 signaling pathway and induced cellular senescence. In TLR4-KO mice, onset of radiogenic SG dysfunction was markedly delayed. However, TLR2/RAGE signalings were alternatively activated, and SG function was impaired.

Conclusions: Clearance of DAMPs such as HMGB1 may attenuate sterile inflammation in damaged SGs via suppression of the TLR4/RAGE signaling pathway. This cellular mechanism may have significant implications for the development of future cell-based regenerative therapies.

Fig. 1

Schematic diagram describing 5G-culture and the characteristics of mouse E-MNCs. (A) PB-MNCs were cultured for 5 days in serum-free medium supplemented with five recombinant proteins: TPO, VEGF, SCF, Flt-3 ligand, and IL-6. After cultivation, effective-mononuclear cells (E-MNCs) were obtained. (B) Flow cytometry analysis. Percentages of M1 (CD11b⁺/CD206⁻) and M2 (CD11b⁺/CD206⁺) macrophages among PB-MNCs and E-MNCs, and CCR2⁻/galectin 3⁺ cells among the CD11b⁺ population.

Fig. 2

Analysis of the therapeutic mechanism of E-MNC transplantation in mouse SGs. (A) Diagram of cellular mechanism of E-MNC treatment. M2 macrophages among E-MNCs might contribute to the conversion of damaged tissues from a pro-inflammatory to anti-inflammatory state by mediating HMGB1 clearance. (B) Relative mRNA expression of the TLR4 gene in SGs at 4 weeks post-IR. Data are presented as the mean ± standard deviation. (C) TLR4 expression (red, TLR4; blue, DAPI; scale bar, 100 μm) ( × 200) at 4 weeks post-IR. (D) RAGE expression (red, RAGE; blue, DAPI; scale bar, 100 μm) ( × 200) at 4 weeks post-IR.

Fig. 3

Post-IR functional and pathologic changes in SGs. (A) Change in salivary flow rate (SFR) at 0, 2, 4, 8, and 12 weeks post-IR. (B) Hematoxylin and eosin staining (scale bar, 100 μm) ( × 200). (C) SA-β-Gal staining of SGs of 20- and 48-week-old mice (IR-20w; 20-week-old mice at 8 weeks post-IR) compared with non-irradiated mice (Ctrl) (scale bar, 100 μm) ( × 200). (D) Masson's trichrome staining (scale bar, 200 μm) ( × 100). Fibrotic areas are stained blue. (E, F) Relative expression of collagen-1 and elastin mRNAs in SGs of mice at 12 weeks post-IR (IR-12w) compared with non-irradiated mice (Ctrl) (∗∗p < 0.01, ∗∗∗p < 0.001). (G, H) Concentration of extracellular HMGB1 in SG tissues and saliva at 0, 2, 4, 8, and 12 weeks post-IR (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). Data are presented as the mean ± standard deviation.

Fig. 4

Expression of genes related to PRRs for HMGB1. (A–G) Relative expression of HMGB1 receptor mRNAs (TLR2, 3, 4, 7, 8, 9, and RAGE) in SGs from 4 to 12 weeks post-IR compared with non-irradiated mice (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). Data are presented as the mean ± standard deviation.

Fig. 5

Histological observations of TLR4 and RAGE expression in SGs at 12 weeks post-IR. (A) Expression of TLR4 (red, TLR4; blue, DAPI; scale bar, 100 μm) ( × 200). (B) Expression of TLR4 and AQP5 (red, TLR4; green, AQP5; blue, DAPI; scale bar, 100 μm) (upper, × 200; lower, × 400). (C) Expression of TLR4 and KRT14 (red, TLR4; green, KRT14; blue, DAPI; scale bar, 100 μm) (upper, × 200; lower, × 400). (D) Expression of RAGE (red, RAGE; blue, DAPI; scale bar, 100 μm) (upper, × 200; lower, × 400). (E) Expression of RAGE and CD31 (red, CD31; green, RAGE; blue, DAPI; scale bar, 20 μm) ( × 1000). (F) Expression of RAGE and F4/80 (red, RAGE; green, F4/80; blue, DAPI; scale bar, 20 μm) ( × 1000).

Fig. 6

Analysis of NF-κB pathway activity in SGs post-IR. (A) Protein expression of p–NF–κB (p65) in SGs of non-irradiated mice (Ctrl) and irradiated mice (IR) at 1 week post-IR. (B, C) p–NF–κB expression at 4 weeks post-IR compared with non-irradiated mice (Ctrl) (red, p–NF–κB; blue, DAPI; scale bar, 100 μm) ( × 200). (D) NF-κB gene expression in SGs from 4 to 12 weeks post-IR compared with non-irradiated mice (∗∗p < 0.01, ∗∗∗p < 0.001). (E) MyD88 gene expression in SGs from 4 to 12 weeks post-IR compared with non-irradiated mice (∗∗p < 0.01, ∗∗∗p < 0.001). (F) Relative expression of TRAM and TRIF mRNAs post-IR compared with non-irradiated mice (∗∗p < 0.01, ∗∗∗p < 0.001). (G) Relative expression of IL-1β, IL-6, TNF-α, and INF-γ mRNAs (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). Data are presented as the mean ± standard deviation.

Fig. 7

Cultivation of SG epithelial cells and cellular changes in cultured SG cells post-IR. (A) Schematic diagram of the cultivation of SG epithelial cells and phase-contrast imaging (scale bar, 100 μm) ( × 400). (B) Phase-contrast image of cultured SG epithelial cells at 14 days after transfer to coating plate (scale bar, 50 μm) ( × 400). (C) Ep-CAM expression in cultured SG epithelial cells at 14 days after transfer to coating plate (green, Ep-CAM; blue, DAPI; scale bar, 50 μm) ( × 400). (D) Flow cytometry analysis of cultured SG epithelial cells. Percentages of Ep-CAM–positive cells among living cells are shown. (E) Immunocytochemistry analysis of HMGB1 in cultured SG epithelial cells post-IR (red, HMGB1; blue, DAPI; scale bar, 100 μm) ( × 200). (F) Change in concentration of HMGB1 in supernatant of cultured SG epithelial cells post-IR (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). (G) Change in TLR4 expression in Ep-CAM–positive SG epithelial cells cultured in collected supernatant (red, TLR4; green, Ep-CAM; blue, DAPI; scale bar, 100 μm) ( × 400).

Fig. 8

Pathologic changes in SG epithelial cells cultured in collected supernatant. (A) Phase-contrast image of SG epithelial cells cultured in the supernatant at 7 days (scale bar, 400 μm) ( × 100). (B) SA-β-Gal assay of SG epithelial cells cultured in supernatant; cells were assayed at 3-day intervals post-IR (scale bar, 100 μm) ( × 200). (C) Concentrations of HMGB1, p16ink4a, PRDX6, and TNF-α in supernatants collected at 3-day intervals post-IR (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). (D) Expression of TLR genes in SG epithelial cells cultured in supernatant with/without HMGB1-neutralizing antibody at 7 days (Ctrl, cultured with supernatant [Ctrl]; Supernatant, cultured with supernatant [IR]; Anti, cultured with supernatant [IR] and HMGB1-neutralizing antibody) (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). (E) Expression of NF-κB, MyD88, and inflammation-related cytokine genes in SG epithelial cells cultured in supernatant with/without HMGB1-neutralizing antibody at 7 days (Ctrl, cultured with supernatant [Ctrl]; Supernatant, cultured with supernatant [IR]; Anti, cultured with supernatant [IR] and HMGB1-neutralizing antibody) (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). Data are presented as the mean ± standard deviation. (F) p–NF–κB expression in SG epithelial cells cultured in supernatant (IR) at 1 day (red, p–NF–κB; blue, DAPI; scale bar, 100 μm) ( × 400). (G) Protein expression of TRAM in SG epithelial cells cultured in supernatant (IR) at 7 days.

Fig. 9

Analysis of functional and pathologic changes in SGs of TLR4-KO mice at 4 and 8 weeks post-IR. (A) Saliva production (salivary flow rate; SFR) in non-irradiated wild-type (WT) mice (Ctrl), irradiated WT mice (WT), and irradiated TLR4-KO mice (KO) (∗∗p < 0.01). (B) Concentration of HMGB1 in SGs (∗∗∗p < 0.001). (C) Relative expression of TLR2 mRNA post-IR compared with non-irradiated WT mice (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). (D) Relative expression of TLR4 mRNA post-IR compared with non-irradiated WT mice (∗∗∗p < 0.001). (E) Relative expression of RAGE mRNA post-IR compared with non-irradiated WT mice (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). (F) Relative expression of NF-κB mRNA post-IR compared with non-irradiated WT mice (∗∗∗p < 0.001). (G) Relative expression of MyD88 mRNA post-IR compared with non-irradiated WT mice (∗p < 0.05). (H) Relative expression of IL-1β, IL-6, TNF-α, and INF-γ mRNAs post-IR compared with non-irradiated WT mice (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). (I) p–NF–κB expression in SG epithelial cells derived from WT or KO mice cultured in supernatant (IR) at 7 days (red, p–NF–κB; blue, DAPI; scale bar, 100 μm) ( × 200). (J) Number of RAGE-positive cells in SGs post-IR compared with non-irradiated WT mice (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). Data are presented as the mean ± standard deviation.

Fig. 10

Analysis of functional and pathologic changes in SGs of TLR4-KO mice at 12 weeks post-IR. (A) Saliva production (salivary flow rate; SFR) in non-irradiated wild-type (WT) mice (Ctrl), irradiated WT mice (WT), and irradiated TLR4-KO mice (KO) (∗∗p < 0.01). (B) Concentration of HMGB1 in SGs (∗∗∗p < 0.001). (C) Relative expression of TLR4 mRNA post-IR compared with non-irradiated WT mice (∗∗∗p < 0.001). (D) Relative expression of TLR2 mRNA post-IR compared with non-irradiated WT mice (∗∗p < 0.01, ∗∗∗p < 0.001). (E) Relative expression of RAGE mRNA post-IR compared with non-irradiated WT mice (∗∗∗p < 0.001). (F) Relative expression of NF-κB mRNA post-IR compared with non-irradiated WT mice (∗∗p < 0.01, ∗∗∗p < 0.001). (G) Relative expression of MyD88 mRNA post-IR compared with non-irradiated WT mice (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). (H) Relative expression of IL-1β, IL-6, TNF-α, and INF-γ mRNAs post-IR compared with non-irradiated WT mice (∗∗p < 0.01, ∗∗∗p < 0.001). (I) Number of RAGE-positive cells in SGs post-IR compared with non-irradiated WT mice (∗p < 0.05, ∗∗∗p < 0.001). (J) Hematoxylin and eosin staining of SGs of 20-week-old mice (IR 20w; 20-week-old mice at 8 weeks post-IR) (scale bar, 100 μm) ( × 200). (K) SA-β-Gal staining of SGs of 20-week-old mice (IR 20w; 20-week-old mice at 8 weeks post-IR) (scale bar, 100 μm) ( × 200). (L) Number of SA-β-Gal–positive cells in SGs of 20-week-old mice. Data are presented as the mean ± standard deviation.

figs1