Cyclic mechanical stretch down-regulates cathelicidin antimicrobial peptide expression and activates a pro-inflammatory response in human bronchial epithelial cells

Cell culture, reagents and cyclic stretch

An E6/E7 viral oncogene immortalized human bronchial epithelial cell line VA10 was cultured as described previously (Halldorsson et al., 2007). Briefly, the cells were maintained in Bronchial/Tracheal Epithelial cell growth medium (Cell Applications, San Diego, CA, USA) with Penicillin-Streptomycin ((20 U/ml, 20 µg/ml, respectively) (Life Technologies, Carlsbad, CA, USA)) at 37 °C and 5% CO₂. BCi. NS 1.1 (henceforth referred to as BCi) is a human bronchial epithelial cell line was a kind gift from Dr. Matthew S. Walters, Weill Cornell Medical College, New York NY, USA (Walters et al., 2013) and was established by immortalization with retrovirus expressing human telomerase (hTERT). The BCi cells were cultured as described above for VA10 cell line. Equal amount of cells were seeded on each well in a 6 well collagen I coated Bioflex plates (Flexcell International Corporation, Burlington, CA, USA), and grown to approximately 80% confluence. These plates were then transferred to a base plate of the cell stretching equipment Flexcell FX-5000TM Tension System (Flexcell International Corporation, Burlington, CA, USA) in a humidified incubator at 37 °C and 5% CO₂. The cells were subjected to cyclic mechanical stretch with the following parameters: a stretching rate of 20% with a square signal, 0.33 Hz frequency (20 cycles/min) and a 1:1 stretch:relaxation ratio, as described previously (Pugin et al., 2008). The cells were stretched for 6 h and 24 h as described in the results. Control Bioflex plates were kept in the same incubator under static conditions as non-stretch controls. Vitamin D3 (1,25D3) and Sodium 4-phenyl butyric acid (PBA) were purchased from Tocris bioscience, UK. Vitamin D3 was reconstituted in 100% ethanol as per manufacturer’s instructions. The final concentration of the solvent was kept at 0.2% v/v and did not affect gene and protein expression of target genes. PBA was reconstituted in ultrapure H₂O.

RNA isolation and quantitative real time PCR

Total RNA was isolated with NucleoSpin RNA kit (Macherey-Nagel, Düren, Germany) and quantified on a spectrophotometer (Nanodrop, Thermo Scientific, Waltham, MA, USA). One µg of total RNA was reverse transcribed into first strand cDNA for each sample with a RevertAid First strand cDNA synthesis kit (Thermo Scientific, Waltham, MA, USA) and modified with 100 unit of reverse transcriptase per reaction. Power SYBR^® green Universal PCR master mix (Life technologies, Carlsbad, CA, USA) was used to quantify the cDNA on a 7500 Real time PCR machine (Life technologies, Carlsbad, CA, USA). The reference gene used for all experiments was UBC (Ubiquitin C) and PPIA (Peptidylprolyl Isomerase A) and an arithmetic mean of reference gene Ct values was used. Primers for TLR1 and TLR6 were designed with Pearl primer and used at a final concentration of 300 nM (Marshall, 2004). All other primers were purchased from Integrated DNA technologies (PrimeTime Predesigned qPCR Assay) and used at a final concentration of 500 nM as per manufacturer’s instructions. The qPCR cycling conditions were as follows; (1) Hold stage: 95 °C for 10 min, followed by 40 cycles of (2) De-natured stage: 95 °C for 15 s and (3) Annealed/extended stage: 60 °C for 1 min. The 2^(−ΔΔCT) Livak method was utilized for calculating fold differences over untreated control (Livak & Schmittgen, 2001). A detailed list of the primers used in the q-RT-PCR assay is shown in Table 1.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and western blot analysis

Supernatants from treated cells were enriched for proteins on Oasis HLB mini columns (Waters, Milford, MA, USA) and lyophilized (Speed Vac; Thermo Scientific, Waltham, MA, USA). Ten µg of total lyophilized protein was loaded on a 4–12% Bis–Tris gradient SDS gel. Protein loading was checked with total protein stain for polyvinylidene fluoride (PVDF) membrane (Pierce MemCode reversible protein stain; Thermo Scientific, Waltham, MA, USA). Total cell lysate was prepared by addition of RIPA lysis buffer (Sigma Aldrich, St. Louis, MO, USA) supplemented with 1× Halt protease and phosphatase inhibitor cocktail (Thermo Scientific, Waltham, MA, USA). Cells were then washed three times with cold 1× PBS, incubated for 30 min on ice with RIPA buffer and centrifuged at 10,000 rpm for 5 min at 4 °C. Supernatants obtained were used for Western blot analysis. The protein content of supernatants was analyzed by utilizing a bio rad protein assay dye reagent based on the Bradford dye binding method (Catalog No. 500-0006, Bio-Rad, USA). SDS-PAGE with subsequent Western blot analysis was performed using the NuPage blotting kit (Life Technologies, Carlsbad, CA, USA). Ten µg of each sample (cell lysates) was loaded on a 4–12% Bis–Tris gradient gel and run at 200 V for 30 min. The proteins were then transferred to a PVDF membrane (Millipore, Billerca, MA, USA) and blocked with 5% non-fat skimmed milk in 1× phosphate buffered saline (PBS) with 0.05% tween (Sigma Aldrich, St. Louis, MO, USA). Antibodies against LL-37 were purchased from Innovagen, Sweden (polyclonal Rabbit, Cat No. PA-LL37-100) and Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) from Santa Cruz Biotechnology, USA (polyclonal Rabbit, Cat No. sc-25778). The primary antibodies were diluted 1:1,000 in 1% non-fat skimmed milk powder (Blotto; Santa Cruz Biotechnologies, Santa Cruz, California, USA) in 0.05% Tween 1× PBS and incubated overnight at 4 °C. Next the membrane was incubated with 1:10.000 Horseradish Peroxidase (HRP)-linked secondary anti-rabbit IgG antibody (Cat No. A0545; Sigma Aldrich, St. Louis, MO, USA) in 0.05% Tween 1× PBS. The protein bands were visualized by chemiluminescence with Pierce ECLPlus Western blotting substrate (Thermo Scientific, Waltham, MA, USA) on Image Quant LAS 4000 station (GE Healthcare, Wauwatosa, WI, USA).

Immunofluorescence

VA10 cells were fixed with 3.5% paraformaldehyde (Sigma Aldrich, St. Louis, MO, USA) prepared in 1× PBS for 15 min. The cells were then washed twice with 1× PBS for 10 min at room temperature and blocked in immunofluorescence (IF) buffer (10% Fetal bovine serum (FBS) and 0.3% triton X-100 in 1× PBS) for 30 min. The cells were incubated overnight with LL-37 primary antibody (polyclonal Rabbit, Cat No. PA-LL37-100; Innovagen, Sweden) diluted 1:100 in the IF buffer, at 4 °C. Following overnight incubation, the cells were washed twice with 1× PBS. Next, the cells were incubated with secondary antibody anti-rabbit IgG Alexa Fluor^® 488 (Catalog No. A11070)/546 conjugate (Catalog No. A11010) from Life Technologies, Carlsbad, CA, USA, 1:1,000 diluted in the IF buffer, for 1 h at room temperature. The cells were counterstained with nuclear stain 4, 6-diamidino-2-phenylindole (DAPI) at 1:5,000 dilution or 3 µM in IF buffer (Catalog No. D9564; Sigma Aldrich, St. Louis, USA). Finally, the cells were washed twice with 1× PBS and ultrapure H₂O, and mounted in Fluormount-G solution (Southern Biotech, Birmingham, AL, USA) for microscopic analysis. The images were captured on Olympus fluoview Fv1200 confocal microscope at a 20× magnification. Olympus fluoview (FV) 1,000 software was used for processing the acquired images.

Oxidative stress measurement

After subjecting VA10 cells to cyclic stretch, CellROX green reagent (Life Technologies, Carlsbad, CA, USA) was added to the medium at a final concentration of 5 µM for 30 min. Next, the cells were washed twice with cold 1× PBS (1,000 rpm for 5 min) and detached with a 1× Accutase solution (Millipore, Billerca, MA, USA). Cells were then harvested and suspended in 100 µl of MACS buffer. (Miltenyi Biotec, San Diego, CA, USA) as per manufacturer’s instructions. The samples were analyzed in MACSQuant flow cytometer (Miltenyi Biotec, San Diego, CA, USA), placing the CellROX green reagent signal in FL1. Intact cells were gated in the Forward Scatter/Side Scatter plot to exclude debris. The resulting FL1 data was plotted on a histogram and is represented as % CellROX positive cells before and after cyclic stretch.

ELISA

Sandwich enzyme-linked immunosorbent assays (ELISAs) were performed utilizing an interleukin 8 (IL-8) and interferon gamma-induced protein 10 (IP-10) assay kit according to manufacturer’s instructions (Peprotech, London, UK). The results are represented from three independent experiments.

Statistical analysis

The q-RT PCR and ELISA results are represented as means ± standard errors of the means (S: E.) from three independent experiments. An unpaired Student’s t-test was used to compare two samples. P < 0.05 was considered statistically significant. All the statistical analysis for q-RT PCR and ELISA experiments was performed with the Prism 6 software (Graph Pad, USA). The Western blot and immunofluorescence data are represented from at least three independent experiments showing similar results.

Cyclic stretch down-regulates the expression of the cathelicidin antimicrobial peptide

We screened for the effect of mechanical stretch on AMP expression. VA10 cells were subjected to stretch for 6 and 24 h to analyze early and late changes in AMP mRNA expression. The mRNA expression of AMPs cathelicidin (CAMP), human beta defensin-1 (DEFB1), Lactoferrin (LTF) and Lysozyme (LYZ) was analyzed with quantitative real time PCR (qRT-PCR). The basal mRNA expression of CAMP was decreased at both 6 h and 24 h after cell stretching (Fig. 1A). DEFB1 mRNA expression was reduced at 24 h after cell stretching but was unaffected after 6 h (Fig. S1). The basal mRNA expression of LTF and LZY was very low (Ct > 32) in the VA10 cells and was excluded from this study. The decrease in cathelicidin gene expression was further confirmed at protein level with Western blot (Fig. 1B) and immunofluorescence analysis (Fig. 1C). Western blot analysis of stretched VA10 cells showed a decrease in secreted pro-LL-37 (encoded by the CAMP gene) levels after 24 h of cyclic stretch (Fig. 1B). Further, immunofluorescence staining of stretched VA10 cells also showed a decrease in LL-37 protein expression at both 6 h and 24 h after stretching (Fig. 1C).

Treatment with vitamin D3 and/or 4-phenyl butyric acid (PBA) counteracts stretch mediated down-regulation of cathelicidin expression

We and others have demonstrated that treatment with vitamin D3 (1,25D3) and PBA enhances cathelicidin expression (Gombart, Borregaard & Koeffler, 2005; Kulkarni et al., 2015a; Kulkarni et al., 2015b). VA10 cells were treated with 100 nM 1,25D3 (Fig. 2A), 2 mM PBA (Fig. 2B) or co-treated with 2 mM PBA and 100 nM 1,25D3 (Fig. 2C). These cells were then stretched for either 6 h or 24 h and gene expression of CAMP was analyzed with q-RT PCR. Treatment of VA10 cells with 1,25D3 and/or PBA before stretch counteracted stretch mediated down-regulation of CAMP mRNA expression (Figs. 2A–2C). This counteraction was further confirmed at protein level with immunofluorescence (Fig. 2E) and Western blot analysis (Fig. 2F). Immunofluorescence staining of LL-37 confirmed that the 1,25D3 treatment prevented stretch mediated decrease in LL-37 protein expression with both at 6 h and 24 h stretch (Fig. 2E). Further, VA10 cells were treated with PBA and 1,25D3 as described above and stretched for 24 h. Protein expression of pro-LL-37 was analyzed with Western blot. Co-treatment with PBA and 1,25D3 predominantly enhanced pro-LL-37 expression in VA10 cells (Fig. 2F). This enhanced expression was lower in the stretched cells (Fig. 2F). Finally, we verified stretch mediated decrease of CAMP mRNA expression in another human bronchial epithelial cell line BCi. Similar to the VA10 cells, the BCi cells were treated with 100 nM 1,25D3 and stretched for 24 h. The basal mRNA expression of CAMP was reduced with 24 stretch, however was not significantly changed after 6 h. Similar to the results in VA10 cells, treatment with 1,25D3 counteracted stretch mediated down-regulation of CAMP mRNA expression in BCi cells (Fig. 2D). Thus, we demonstrate that treatment with 1,25D3 and PBA counteracted cyclic stretch mediated down-regulation of cathelicidin AMP expression.

Cyclic stretch activates a pro-inflammatory response and modulates toll-like receptor expression

Previous studies have demonstrated that VILI and in vitro cyclic stretching of alveolar lung epithelial cells activates a pro-inflammatory response by enhanced secretion of pro-inflammatory cytokines and chemokines (Vlahakis et al., 1999; Halbertsma et al., 2005). We studied the effect of cyclic stretch on inflammation in VA10 cells by screening for stretch mediated changes in pro-inflammatory cytokines and chemokines expression (Figs. 3A–3F). The VA10 cells were stretched for 6 h and 24 h. The mRNA expression of pro-inflammatory cytokines CXCL8 (encoding interleukin 8 or IL-8), IL1B (encoding interleukin 1 beta or IL-1β) and chemokines CXCL10 (encoding interferon gamma induced protein 10 or IP-10), CCL5 (encoding regulated on activation, normal T cell expressed and secreted or RANTES) was analyzed with q-RT PCR (Figs. 3A–3D) after stretching. The mRNA expression of genes encoding IL-8 and IL-1β was enhanced after 6 h and 24 h cyclic stretch (Figs. 3A–3C). Interestingly, mRNA expression of the gene encoding IP-10 was reduced following 6 h and 24 h stretch (Figs. 3A–3C). The mRNA expression of gene encoding RANTES was reduced after 24 h stretching and was not affected after 6 h stretch (Fig. 3D). Analysis of protein expression with ELISA demonstrated enhanced secretion of IL-8 after 24 h of cell stretching (Fig. 3E), whereas the total secreted levels of IP-10 was reduced (Fig. 3F). Increased oxidative stress through enhanced reactive oxygen species (ROS) has been shown to be involved in activation of a pro-inflammatory response (Mittal et al., 2014). We observed increased oxidative stress following cyclic stretch with the ROS detector dye Cell ROX. VA10 cells were stretched for 24 h and Cell ROX dye (5 µM) was added 30 min before end of cyclic stretching. The cells were then harvested and analyzed in a flow cytometer. A significant increase in percentage of Cell ROX positive cells was observed after stretching, indicating increased oxidative stress (Fig. 3G). Thus, we demonstrate that cyclic stretching activates a pro-inflammatory response in VA10 cells.

Next, we screened for stretch mediated changes in toll-like receptor (TLR) expression in VA10 cells (Figs. 4A–4H). TLRs play an important role in activation of pro-inflammatory responses and have been shown to be modulated by mechanical stretching of cells (Takeda & Akira, 2005; Shyu et al., 2010). VA10 cells were stretched for 6 h and 24 h as described above. The mRNA expression of TLR1, TLR2, TLR3, TLR4, TLR5, TLR6, TLR8 and TLR9 was analyzed with q-RT PCR. The mRNA expression of TLR3, TLR5 and TLR8 was reduced 6 h after stretch (Figs. 4C, 4E and 4G). Further, the mRNA expression of TLR2 (Fig. 4B) was increased and gene expression of TLR3 (Fig. 4C) and TLR9 (Fig. 4H) was decreased after 24 h stretch.

Treatment with vitamin D3 and PBA differentially affects stretch mediated changes in pro-inflammatory cytokine IL-8 expression

Vitamin D3 and PBA have been shown to differentially modulate inflammatory responses, mainly having anti-inflammatory effect (Hansdottir et al., 2010; Roy et al., 2012). VA10 cells were treated with 20 nM 1,25D3 (Fig. 5A) and 2 mM PBA (Fig. 5B), followed by stretching for 6 h and 24 h as shown in Fig. 5. The mRNA expression of genes encoding the pro-inflammatory cytokine IL-8 was analyzed with q-RT PCR. Treatment with 1,25D3 and PBA increased basal expression of IL-8 after 6 h treatment in static cells (Figs. 5A and 5B). Interestingly, treatment with PBA significantly enhanced stretch mediated increase in IL-8 gene expression after 6 h stretch (Fig. 5B). This enhancement was not observed at 24 h after stretch (Fig. 5B).