TREM-2 promotes acquired cholesteatoma-induced bone destruction by modulating TLR4 signaling pathway and osteoclasts activation

Abstract

Triggering receptor expressed on myeloid cells (TREM) has been broadly studied in inflammatory disease. However, the expression and function of TREM-2 remain undiscovered in acquired cholesteatoma. The expression of TREM-2 was significantly higher in human acquired cholesteatoma than in normal skin from the external auditory canal, and its expression level was positively correlated with the severity of bone destruction. Furthermore, TREM-2 was mainly expressed on dendritic cells (DCs). In human acquired cholesteatoma, the expression of proinflammatory cytokines (IL-1β, TNF-α and IL-6) and matrix metalloproteinases (MMP-2, MMP-8 and MMP-9) were up-regulated, and their expression levels were positively correlated with TREM-2 expression. Osteoclasts were activated in human acquired cholesteatoma. In an animal model, TREM-2 was up-regulated in mice with experimentally acquired cholesteatoma. TREM-2 deficiency impaired the maturation of experimentally acquired cholesteatoma and protected against bone destruction induced by experimentally acquired cholesteatoma. Additional data showed that TREM-2 up-regulated IL-1β and IL-6 expression via TLR4 instead of the TLR2 signaling pathway and promoted MMP-2 and MMP-8 secretion and osteoclast activation in experimentally acquired cholesteatoma. Therefore, TREM-2 might enhance acquired cholesteatoma-induced bone destruction by amplifying the inflammatory response via TLR4 signaling pathways and promoting MMP secretion and osteoclast activation.

Figure 1. TREM-2 expression is up-regulated in human acquired cholesteatoma.

(A) TREM-2 expression was estimated by quantitative real-time polymerase chain reaction. (B) Representative pictures of Immunohistochemical analysis in human acquired cholesteatoma and normal skin. TREM-2⁺ cells are depicted in brown. Panels are representative of 11 samples of human acquired cholesteatoma and 11 samples of normal skin. Images were acquired with a biomicroscope (CX21, Olympus, Japan) and images of 100× were acquired with an oil immersion objective. *Represents a TREM-2⁺ cell. (C) TREM-2⁺ cells were counted in 5 independent 40× fields and the average values were calculated. Then, correlation analysis was applied between the amount of TREM-2⁺ cells and the level of bone destruction (n = 11).

Figure 2. TREM-2 expression origin in human acquired cholesteatoma.

(A,B) Representative pictures of immunefluorescence analysis of human acquired cholesteatoma (n = 10). TREM-2⁺ cells are depicted in green, CD11c⁺ and CD11b⁺ in red and merge pictures shows antigen colocalization in yellow. Images were acquired with a biomicroscope (CX21, Olympus, Japan). (C) The histogram graph is representative of 5 different cell counts in 40× fields (mean ± SEM). ***p < 0.001. “n.d.” means not detected.

Figure 3. The expression of proinflammatory cytokines in human acquired cholesteatoma was up-regulated and was positively correlated with level of TREM-2.

(A–C) The expression levels of IL-1β, TNF-α and IL-6 were up-regulated in human acquired cholesteatoma (n = 11) compared with in normal skin (n = 11). Data are presented as the mean ± SEM and represent three independent experiments. (D) The expression levels of IL-1β, TNF-α and IL-6 were positively correlated with level of TREM-2. **p < 0.01; ***p < 0.001.

Figure 4. The expression of MMPs in human acquired cholesteatoma was up-regulated and was positively correlated with level of TREM-2.

(A–C) The expression levels of MMP-2, MMP-8 and MMP-9 were up-regulated in human acquired cholesteatoma (n = 10) compared with in normal skin (n = 10). Data are presented as the means ± SEM and represent three independent experiments. (D) The expression levels of MMP-2, MMP-8 and MMP-9 were positively correlated with level of TREM-2. **p < 0.01; ***p < 0.001.

Figure 5. TRAP staining in human acquired cholesteatoma compared with in normal skin.

(A–F) Representative pictures of TRAP staining in human acquired cholesteatoma (n = 10) and normal skin (n = 10) and TRAP⁺ cells were detected in cholesteatoma (A–C). *Represents a typical osteoclast (C). Images were acquired with a biomicroscope (CX21, Olympus, Japan) and images of 100× were acquired with an oil immersion objective. (G) TRAP⁺ cells were counted in 5 independent 40x fields and the average values were calculated. Then, correlation analysis was applied between the amount of TRAP⁺ cells and the level of bone destruction (n = 10).

Figure 6. Cathepsin K was detected in infected ossicles.

(A–F) Representative pictures of TRAP staining in normal (n = 10) and infected ossicles (n = 10). (A–C) no cathepsin K⁺ cells were found in normal ossicles. However, many cathepsin K⁺ cells were found in infected ossicles (D–F). *Represents a typical osteoclast (F). Images were acquired with a biomicroscope (CX21, Olympus, Japan) and images of 100× were acquired with an oil immersion objective.

Figure 7. The evaluation of TREM-2 knockout effect in TREM-2−/− mice (n = 5) compared with WT mice (n = 5).

TREM-2 expression was estimated by quantitative real-time polymerase chain reaction. Data are presented as the means ± SEM and represent three different experiments. “n.d.” means not detected; ***p < 0.001.

Figure 8. Assessment of animal model. Representative pictures of HE staining in WT group (n = 5) and TREM-2−/− group (n = 5).

(A–C) The graft developed into a mature acquired cholesteatoma in WT mice. However, it atrophied in TREM-2−/− mice (D–F). (G) The thickness of grafts was measured using Photoshop CS6 and (H) cells were counted in 40× graphs using Image J and analyzed using Graphpad Prism. The histogram graphs are representative of 5 different measurement of thickness and 5 different cell counts (mean ± SEM). **p < 0.01; ***p < 0.001.

Figure 9. Bone destruction area was observed on the calvarium surface of WT mice (n = 5) compared with TREM-2−/− mice (n = 5).

(A–D) The circle shows a bone destruction area, (E–H) and no bone destruction area was observed in TREM-2−/− mice. (I) Quantitative assessment of bone destruction area showed significant difference between two groups. *Area ratio = area of bone destruction/area of grafted skin. ***p < 0.001.

Figure 10. TREM-2−/− mice (n = 5) showed no significant up-regulation of TLR4 and proinflammatory cytokines compared with WT mice (n = 5).

The expression of TLR2 (A), TLR4 (A), IL-1β (B), IL-6 (C) and TNF-α (D) were estimated by quantitative real-time polymerase chain reaction. Data are presented as the means ± SEM and represent of 3 different experiments. “ns” means no significance; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 11. The expression of MMPs was not up-regulated in TREM-2−/− mice (n = 5) compared with in WT mice (n = 5).

The expression of MMP-2 (A), MMP-8 (B) and MMP-9 (C) were estimated by quantitative real-time polymerase chain reaction. Data are presented as the means ± SEM and represent of 3 different experiments. “ns” means no significance; **p < 0.01; ***p < 0.001.

Figure 12. Osteoclasts were detected only in grafts of WT mice.

(A–P) Representative pictures of TRAP staining in normal skin and grafts from WT (n = 5) and TREM-2−/− group (n = 5). *Represents typical osteoclasts (H). Images were acquired with a biomicroscope (CX21, Olympus, Japan) and images of 100× were acquired with an oil immersion objective. (Q) TRAP⁺ cells were counted in 40× field using Image J. The histogram graphs are representative of 5 different cell counts (mean ± SEM). ***p < 0.001. “n.d.” means not detected.

Figure 13. The diagrammatic sketch of how TREM-2 most likely completes the bone destruction in acquired cholesteatoma.