Human Host Defense Peptide LL-37 Suppresses TNFα-Mediated Matrix Metalloproteinases MMP9 and MMP13 in Human Bronchial Epithelial Cells

Abstract

Introduction: TNFα-inducible matrix metalloproteinases play a critical role in the process of airway remodeling in respiratory inflammatory disease including asthma. The cationic host defense peptide LL-37 is elevated in the lungs during airway inflammation. However, the impact of LL-37 on TNFα-driven processes is not well understood. Here, we examined the effect of LL-37 on TNFα-mediated responses in human bronchial epithelial cells (HBECs).

Methods: We used a slow off-rate modified aptamer-based proteomics approach to define the HBEC proteome altered in response to TNFα. Abundance of selected protein candidates and signaling intermediates was examined using immunoassays, ELISA and Western blots, and mRNA abundance was examined by qRT-PCR.

Results: Proteomics analysis revealed that 124 proteins were significantly altered, 12 proteins were enhanced by ≥2-fold compared to unstimulated cells, in response to TNFα. MMP9 was the topmost increased protein in response to TNFα, enhanced by ∼10-fold, and MMP13 was increased by ∼3-fold, compared to unstimulated cells. Furthermore, we demonstrated that LL-37 significantly suppressed TNFα-mediated MMP9 and MMP13 in HBEC. Mechanistic data revealed that TNFα-mediated MMP9 and MMP13 production is controlled by SRC kinase and that LL-37 enhances related upstream negative regulators, namely, phospho-AKT (T308) and TNFα-mediated TNFAIP3 or A20.

Conclusions: The findings of this study suggest that LL-37 may play a role in intervening in the process of airway remodeling in chronic inflammatory respiratory disease such as asthma.

Keywords: Airway remodeling; Cathelicidin; Host defense peptides; Inflammation; LL-37; Matrix metalloproteinases; TNFα.

Fig. 1.

Alteration of the bronchial epithelial cell proteome by TNFα. HBEC-3KT cells were stimulated with TNFα (20 ng/mL) for 24 h. Cell lysates (14 μg total protein per sample) obtained from five independent experiments were probed using the high-content aptamer-based proteomic array. Pairwise differential analysis was conducted on normalized log2 protein expression values, and Welch’s t test with a cut-off of p < 0.05 was used to select protein abundance changes that were significantly altered in response to TNFα, compared to unstimulated cells. The volcano plot demonstrates differentially abundant proteins in response to TNFα compared to unstimulated cells.

Fig. 2.

TNFα induces secretion of MMP9 and MMP13 from HBEC cultures. HBEC-3KT cells (a) or submerged human primary bronchial epithelial cells (PBECs) (b) were stimulated with either TNFα (20 ng/mL) or IFNγ (30 ng/mL) for 24 h. Cell culture supernatants were examined for the abundance of MMP9 and MMP13 by ELISA. Each symbol represents an independent experiment, using cells from independent donors for PBEC, and bars show the median and min-max range. One-way ANOVA with Fisher’s least significant difference test was used for statistical analysis (*p < 0.05, ****p < 0.0001). c PBEC was differentiated at the air-liquid interface (ALI) and stimulated with either TNFα (10–40 ng/mL) or IFNγ (15–60 ng/mL) for 24 h, after which cell culture medium from the basal side was examined for the abundance of MMP9 and MMP13 by ELISA. Each symbol represents an independent experiment using cells from an independent donor and bars show the median and min-max range.

Fig. 3.

LL-37 and citLL-37 suppress TNFα-mediated MMP9 and MMP13 production. HBEC were stimulated with LL-37, citLL-37, or sLL-37 (0.25 μm each) in the presence/absence of TNFα (20 ng/mL) for 24 h. TC supernatants were examined for the abundance of MMP9 and MMP13 by ELISA in HBEC-3KT (N = 5 independent experiments) (a) and human PBEC (independent experiments with N = 3 independent donors with n = 2 technical replicates each) (b). Y-axis represents % change compared to paired unstimulated cells within each independent experimental replicate. Each dot represents an independent experiment, and bars show the median and IQR, whereas whiskers show min-max range. Two-way analysis of variance with Fisher’s least significant difference test was used for statistical analysis (**p ≤ 0.001, ****p ≤ 0.0001).

Fig. 4.

SRC kinase inhibitors suppress TNFα-mediated MMP9 and MMP13 production. HBEC-3KT cells were pre-treated with pharmacological inhibitors SRC1 inhibitor (SRCi) and dasatinib for 1 h prior to stimulation with TNFα (20 ng/mL). TC supernatants were collected 24 h after TNFα stimulation and examined for the abundance of MMP9 (a) and MMP13 (b), by ELISA. Each data point represents an independent replicate (N = 4 independent experiments), and the line represents the average values. Two-way ANOVA with Dunnett’s test for multiple comparisons was used to determine statistical significance (***p < 0.001, ****p < 0.0001).

Fig. 5.

LL-37 and citLL-37 differentially alter AKT phosphorylation in bronchial epithelial cells. HBEC-3KT (N = 3 independent experiments) was stimulated with LL-37, citLL-37, or sLL-37 (0.25 μm each). Total cell lysate (25 μg) obtained 30 min post-stimulation was used to examine the abundance of phospho-AKT (T308) (a) and phospho-AKT (S473) (b) by Western blots. Densitometry was performed after normalization with β-actin as a paired loading control for each sample. Y-axis represents % change compared to paired unstimulated controls. Each symbol represents densitometry assessment from an independent experiment, and bars show the median and min-max range. Repeated-measures one-way ANOVA with Fisher’s least significant difference test was used for statistical analysis (*p ≤ 0.05).

Fig. 6.

LL-37 enhances TNFα-mediated negative regulator TNFAIP3 or A20 in bronchial epithelial cells. HBEC-3KT (N = 4 independent experiments) was stimulated with LL-37, citLL-37, or sLL-37 (0.25 μm each), in the presence and absence of TNFα (20 ng/mL). mRNA isolated 2 h post-stimulation was used to examine the expression of TNFAIP3 by qRT-PCR. Each symbol represents an independent experiment. Fold changes (Y-axis) for TNFAIP3 were normalized to 18S RNA and calculated compared to unstimulated cells normalized to 1, using the comparative ΔΔCt method. Results are shown as boxplots, wherein bars show median and IQR, and whiskers show minimum and maximum values. One-way ANOVA was used for statistical analysis (***p ≤ 0.001). The dashed line represents normalized baseline value of 1 for unstimulated cells.