Roles of TLR3 in poly(I:C)- and UV-induced gene expressions in human dermal fibroblasts and mouse skin

Abstract

Toll-like receptors (TLRs) are known to recognize not only pathogen-associated molecular patterns (PAMPs) but also danger-associated molecular patterns (DAMPs). Recent studies have characterized the expressions and functions of TLRs in human epidermal cells. However, the characteristics of TLR family members in human dermal fibroblasts have not been thoroughly studied. Ultraviolet (UV) irradiation can result in premature skin aging (photoaging) which is characterized by decreased expression of collagen and increased expression of matrix metalloproteinases (MMPs). Double strand RNA (dsRNA) can be generated at various conditions including virally infected cells or UV-damaged skin cells. TLR3 is one of the receptors for dsRNA. However, little is known about the effect of dsRNA on the expression of procollagen and MMPs in skin fibroblasts. Recently, it is known that UV-irradiated keratinocytes could release damaged self-noncoding RNAs that serve as TLR3 ligands. This finding is suggested to be involved in UV-induced increase of inflammatory cytokines, such as TNF-α and IL-6. However, whether TLR3 is involved in UV-induced reduction of procollagen and induction of MMP expressions in vivo in unknown. In chapter Ⅰ, I systematically investigated the expression of TLRs and their functional responses to each ligand in skin fibroblasts. I found that all 10 TLRs are expressed in skin fibroblasts. Stimulation of skin fibroblasts with each TLR ligand resulted in increase of IL-6, IL-8, and MMP-1 proteins. I also found that the expression level of each TLR was much higher in fibroblasts than in keratinocytes. In particular, I found that the fold-increases in IL-6 and IL-8 mRNA levels upon exposure to a TLR1/2 ligand were much higher in fibroblasts than in keratinocytes, which appears to reflect the difference in expression levels of TLR1 and 2 in between fibroblasts and keratinocytes. In chapter Ⅱ, I observed that treatment of TLR3 ligand, poly(I:C), but not other TLR ligands reduced procollagen and induces IFN-β expression in skin fibroblasts. I found poly(I:C) induced IFN-β expression through interferon regulatory factor 3 (IRF3)-dependent pathway in skin fibroblasts. IRF-3 pathway was not activated by other TLR ligands except poly(I:C). Further experiments revealed that poly(I:C) reduced procollagen expression through induction of IFN-β in skin fibroblasts. I also showed that knockdown or knockout of TLR3 relieved poly(I:C)-induced reduction of procollagen expression. In chapter Ⅲ, I examined the effect of poly(I:C) on MMP-1, -2, and -3 expressions in skin fibroblasts. I found that poly(I:C) treatment induced expressions of MMP-1, -2, and -3, which were dependent on TLR3. Poly(I:C) treatment also induced activations of the mitogen-activated protein kinase (MAPK), the nuclear factor-kappaB (NF-κB) and the IRF3 pathways. By using specific inhibitors, I found that poly(I:C)-induced expressions of MMP-1, -2, and -3 were differentially regulated by these signaling pathways. In particular, inhibition of IRF3 signaling pathways attenuated poly(I:C) induced expressions of MMP-1, -2, and -3. In chapter Ⅳ, I checked whether TLR3 is involved in UV-induced reduction of procollagen and induction of MMP expressions in vivo. I found that the expression of procollagen was not significantly different between UV-irradiated WT and TLR3-/- mice skin. In contrast to procollagen expression, I found that UV radiation-induced expression of MMP-13 was significantly higher in WT than in TLR3-/- mice skin. I suggested that the effect of higher expression of MMP-13 in WT skin fibroblasts might be due to released RNA from necrotic keratinocytes. In summary, I showed that all 10 TLRs were constitutively expressed and functional in skin fibroblasts. Expression levels of TLR family members were higher in skin fibroblasts than in skin keratinocytes. Activation of TLR3 reduced procollagen expression through induction of IFN-β in skin fibroblasts. I also found that MMP-1, -2, and -3 could be induced by dsRNA in skin fibroblasts. Poly(I:C) induced expressions of MMP-1, -2, and -3 through various signaling pathways including TLR3 and IRF3. In the end, by using WT and TLR3-/- mice, I determined that UV-induced increase of MMP-13, but not reduction of procollagen expression, was dependent on TLR3. My results in chapter Ⅰ may provide information for studies of certain skin diseases in which fibroblasts are involved in the pathogenesis, such as keloid, hypertrophic scar and scleroderma. My study in chapter Ⅱ provides the first evidence that dsRNA has potent anti-fibrotic activity in skin. My data in chapter Ⅲ suggests that TLR3 and/or IRF3 may be good targets for regulating the expressions of MMP-1, -2, and -3 induced by dsRNAs. My study in chapter Ⅳ indicates that TLR3 could be viewed as a therapeutic target for UV-induced MMP induction.