The role of toll-like receptor 4 in corneal epithelial wound healing

Abstract

Purpose: We evaluated the role of Toll-like receptor 4 (TLR4) in corneal epithelial wound healing.

Methods: The expression of TLR4 during in vivo corneal epithelial wound healing was examined by immunostaining in mice. The expression and activation of TLR4 was studied in primary or telomerase-immortalized human corneal epithelial cells (HCEC). Scratch assay was performed to evaluate in vitro wound closure using live time-lapse microscopy. Transwell migration assay and Ki67 immunostaining were done to evaluate migration and proliferation, respectively. Lipopolysaccharide (LPS) was used to activate TLR4, whereas CLI-095 was used for its inhibition. The expression of inflammatory cytokines was determined by RT-PCR and ELISA. The activation of p42/44 and p38 was determined by immunoblotting.

Results: In the murine model, TLR4 immunostaining was noted prominently in the epithelium 8 hours after wounding. There was a 4-fold increase in the expression of TLR4 6 hours after in vitro scratch wounding (P < 0.001). Confocal microscopy confirmed the membrane localization of TLR4/MD2 complex. There was a significant increase in migration, proliferation, and wound closure in HCEC treated with LPS (P < 0.05), while there was significant decrease with TLR4 inhibition (P < 0.05). Addition of LPS to wounded HCEC resulted in a significant increase in the expression of IL-6, TNF-α, CXCL8/IL8, and CCL5/RANTES at the mRNA and protein levels. Likewise, LPS increased the activation of p42/44 and p38 in wounded HCEC.

Conclusions: These results suggest that epithelial wounding induces the expression of functional TLR4. Toll-like receptor 4 signaling appears to contribute to early corneal epithelial wound repair by enhancing migration and proliferation.

Keywords: TLR4; Toll-like receptor 4; corneal epithelium; corneal wound healing; innate immunity.

Figure 1

The expression of TLR4/MD2 complex was evaluated in the mouse corneal epithelium at 8 hours after a debridement wound. The expression of TLR4/MD2 complex is increased in the wounded epithelium at 8 hours (B) compared to 0 hours after wounding (A). Arrow indicates the wound edge in (A) and the leading edge in (B). Green, FITC; blue, DAPI,

Figure 2

Primary HCEC were stained (without permeabilization) for TLR4/MD2 complex 6 hours after scratch wounding. Representative confocal images depicting the cells are distinctly positive for TLR4/MD2 complex staining (B) compared to 0 hour after wounding which is mostly negative (A). Green, FITC; blue, DAPI. Dotted lines: scratch wound edge. Primary HCEC were wounded and total lysates were collected in different time-points (C). Western blot analysis demonstrates TLR4 expression increased 6 hours after wounding compared to wound 0 (Chart [D]; *P < 0.001) and later returned to the baseline by 18 hours after wounding. Data shown are representative of three independent experiments. Data were normalized to β-actin as housekeeping protein. Error bars: standard deviation.

Figure 3

To evaluate if the TLR4 is responsive after wounding, primary HCEC were challenged with 100 ng/mL ultrapure LPS, and mRNA expression of IL-6, TNF-α, CXCL8/IL8, and CCL5/RANTES were evaluated in different time-points. There was no significant difference in cytokine expression between unwounded and unwounded + LPS (Chart [A]; P > 0.05). However, there was a significant increase in the cytokine expression in the cells that were challenged with ultrapure LPS at 6 hours after wounding compared to corresponding wounded cells not exposed to LPS (Chart [B]; *P < 0.001). Relative mRNA expression was calculated by normalization of each band signal intensity to GAPDH as a housekeeping gene. The level of each normalized sample was reported as its ratio relative to control. (C) Cytokine production by wounded primary corneal epithelial cell after being challenged with 100 ng/mL LPS. Cell culture supernatants from either wounded cells or wounded + LPS were analyzed for selected cytokine production. The x-axis represents individual cytokines tested. The y-axis represents average absorbance values at 450 nm (**P < 0.01). Data shown are representative of three independent experiments. OD, optical density. Error bars: standard deviation.

Figure 4

The effect of TLR4 exogenous ligand or its internal inhibitor on scratch assay wound closure rate was measured. Mechanically wounded HCLE cells were exposed to either 100 ng/mL ultrapure LPS or pretreated with 5 μM CLI-095. Wound repair was assessed by measuring the remaining wound area (RWA) (*P < 0.01, **P < 0.05). Data shown are representative of three independent experiments. Error bars: standard deviation.

Figure 5

Wounded HCLE cells were trypsinized and plated in 8.0-μm pore size transwells. Then, DMEM, LPS, or CLI-095 were added to the lower compartment. The data are expressed as mean number of migrated cells per micrograph field for each sample well. *P = 0.001, **P < 0.001 (A). Primary HCEC were wounded and treated either with LPS or pretreated with CLI-095. At 14 hours after wounding, they were subjected to Ki67 proliferation marker staining and the percentage of Ki67-positive cell nuclei was quantified (*P < 0.001, **P < 0.05) (B). Data shown are representative of three independent experiments. Error bars: standard deviation.

Figure 6

Primary HCEC were wounded and then challenged with LPS. The ratio of phospho-ERK/ERK increased 6 hours after wounding compared to unwounded cells. The ratio of phospho-ERK/ERK increased at 30 minutes after adding LPS and remained above baseline after 1 hour (Chart [A]; *P < 0.01). Likewise, phospho-P38/P38 increased after wounding compared to unwounded cells. After adding LPS to the 6-hour wounded cells, this ratio increased within 15 minutes (**P < 0.001) and returned to baseline after 1 hour (Chart [B]). Data shown are representative of three independent experiments. All values are normalized to co-stained β-actin signal intensity in each blot. Error bars: standard deviation.