Transforming growth factor (TGF)-β1-induced miR-133a inhibits myofibroblast differentiation and pulmonary fibrosis

TGF-β1 regulates gene transcriptions via activation of both canonical (Smad2/3) and non-canonical (p38MAPK) pathways^5,7. Indeed, pretreatment of HFL cells with Smad3 inhibitor SIS3 or p38MAPK inhibitor SB203580 partially blocked TGF-β1-induced expression of α-SMA and CTGF (Fig. 2e). Interestingly, Smad3 and p38MAPK inhibitors also attenuated TGF-β1-induced miR-133a expression (Fig. 2f). Although SB203580 in concentration of 10 μM has potential inhibitory effects on the AKT signaling pathway^18,19, pretreatment with 10 μM of LY294002, an inhibitor of the PI3K/AKT signaling pathway only slightly reduced miR-133a upregulation (Supplementary Fig. 2), suggesting that AKT signaling pathway may only play a limited role in TGF-β1-induced miR-133a upregulation.

ECM protein deposition in the lung is the primary cause of death of pulmonary fibrosis patients²⁰. Matrix metalloproteinase 2 and 9 (MMP-2 and MMP-9) are secreted proteins involved in the breakdown of ECM proteins. Quantitative RT-PCR analysis showed that MMP-2 but not MMP-9 was induced by TGF-β1 treatment, which was not affected by overexpression of miR-133a (Fig. 5c). Gelatin zymography confirmed the unchanged activities of MMP-2 and MMP-9 in cells transfected with miR-133a mimic as compared to cells with control mimic (Fig. 5d). Thus, overexpression of miR-133a reduces the expression of ECM proteins without effects on expression and activity of matrix metalloproteinases.

We further examined alterations of TGF-β1 signaling pathways by miR-1, a member of the miR-1/miR-133a cluster²¹ that was also upregulated in TGF-β1-treated HFL cells (Table 1 and Fig. 1a). As shown in Supplementary Fig. 3, compared to control mimic, transfection of miR-1 mimic did not change TGFBR1 protein expression or TGF-β1-induced upregulation of CTGF protein or Col1a1 and Col4a1 mRNAs. This is consistent with data showing no effects of miR-1 on TGF-β1-induced α-SMA expression (Fig. 1b). Thus, miR-133a but not miR-1 specifically targets TGF-β1 profibrogenic signaling pathways to inhibit pulmonary fibroblast differentiation.

Alignment of the 3’UTR of TGFBR1, CTGF and Col1a1 among a wide range of species (mouse, rabbit and human) using the Targetscan bioinformatic tool revealed that the predicted binding sites for miR-133a are highly conserved during evolution (Fig. 6g). To determine whether TGFBR1, CTGF and Col1a1 are direct targets of miR-133a, we inserted the fragment of TGFBR1 3′UTR, CTGF 3′UTR or Col1a1 3′UTR containing the miR-133a putative target site (UTR-133a) or their mutants at the seeding region (UTR-133a/M) into the pmirGLO dual-luciferase reporter vector (Fig. 6h). Compared to control mimic, miR-133a mimic reduced the luciferase activity by about 35% (**P < 0.01), which was completely abolished by mutations at their seeding regions (Fig. 6i). These data suggest a novel function of TGF-β1-induced miR-133a as a negative feedback regulator of TGF-β1 profibrogenic signaling pathways in pulmonary fibroblasts via directly repressing TGFBR1, CTGF and collagen expression.

HEK293 cells and mouse fibroblast NIH3T3 cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Human primary lung fibroblast cells (HFL) were established by Dr. Reynold Panettieri’s laboratory from patients with brain-related disease but no history of pulmonary fibrosis⁴⁰. The cells were frozen at an early passage and cultured for a maximum of fifteen passages. The cells were maintained at 37 °C with 5% CO₂ in Dulbecco’s Modified Eagle’s Medium/Ham’s Nutrient Mixture F-12 (DME/F12, Life Technologies/Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum (GE Healthcare Bio-Sciences, Pittsburgh, PA, USA), 100 U/mL penicillin G and 100 μg/mL streptomycin (Life Technologies/Gibco). Smad3 inhibitor SIS3 and p38MAPK inhibitor SB203580 were purchased from (AdooQ BioScience, Irvine, CA, USA). Recombinant human TGF-β1 and the Smad2/3 Antibody Sampler Kit were obtained from Cell Signaling Technology (Danvers, MA, USA). The mouse anti-α-SMA monoclonal antibody was obtained from Sigma-Aldrich (St. Louis, MO, USA). The rabbit anti-TGFBR1 antibody was obtained from Merck Millipore (Billerica, MA, USA). The mouse anti-CTGF monoclonal antibody, mouse anti-GAPDH monoclonal antibody were obtained from Proteintech (Chicago, IL, USA). The GAPDH levels served as internal controls.

MiRNA sequencing was performed as previously described³³. Briefly, total RNA was isolated from samples using mirVana miRNA Isolation Kit (Invitrogen) according to the manufacturer’s instructions. For small RNA sequencing, RNA quality was check by Fragment Analyzer™ Automated CE System; a NEXTflex^TM Small RNA-Seq Kit v3 (Bioo Scientific #5132-05) was used with 1 µg of total RNA for the construction of small RNA sequencing libraries according to the standard Illumina protocols; miRNA-sequencing was performed with Illumina Nextseq 500 at the University of Nebraska Medical Center Next Generation Sequencing Core Facility (Omaha, NE, USA). miRNAs of HFLs cells treated with 1 ng/mL TGF-β1 or saline were sequenced. All sequencing data including upregulated and down-regulated miRNAs have been deposited in NCBI’s Gene Expression Omnibus and the accession number is GSE125183↗. The quantitative miRNA expression analysis was performed using the app (B&Gu @ University of Torino) in BaseSpace Sequence Hub.

Total RNA was isolated using TRIzol® Reagent (Thermo Fisher Scientific, Waltham, MA). RT-PCR reactions for miRNA detection were determined by TaqMan probe based microRNA assay on total RNA and normalized to RUN6B levels according to a previous report⁴¹. Quantitative RT-PCR was performed using UltraSYBR Mixture (CWBiotech) and normalized by GAPDH levels. Primers and probes used are listed in Supplementary Table 2. The amplification procedure consisted of 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 60 s. The relative expression levels between samples were calculated using the comparative delta CT (threshold cycle number) method (2^−ΔΔCT) with a control sample as the reference point.

MiRNA mimics or inhibitors (mimic negative control: miR01101-1-5, miR-133a mimic: miR10000427-1-5, miR-1 mimic: miR10000416-1-5, miR-143 mimic: miR10000435-1-5, miR-145 mimic: miR10000437-1-5, miR-21 mimic: miR10000076-1-5, inhibitor negative control: miR02101-1-5, miR-133a inhibitor: miR20000427-1-5) were purchased from RiboBio (Guangzhou, China). Small RNA and plasmid transfections were performed with Lipofectamine® RNAiMAX Reagent and Lipofectamine® 3000 Reagent (Thermo Fisher Scientific), respectively, according to the manufacturer’s instructions.

Targetscan algorithms were used to identify the putative miR-133a targets⁴². Annotated 3′-UTRs for upregulated or down-regulated genes in miR-133a overexpressing cells, were scanned for strict miRNA targets (8mer, 7mer-m8, and 7mer-A1). The significance of the measured overlaps was calculated using a hypergeometric test.

HFL cells were lysed in buffer containing 0.2% sodium deoxycholate, 50 mM Tris-HCl, 1% TritonX-100, 15 mM NaCl, 0.1% SDS and 1 × Protease inhibitor cocktail (Sigma) at 4 °C for 20 min. Cell lysates were centrifuged at 12,000 × g for 20 min, the supernatant was collected, and protein concentration was determined with a Pierce BCA Kit (Rockford, IL, USA). Samples were then subjected to western blot analysis as reported^43,44. Briefly, samples were resolved by 10% SDS-PAGE, transferred to Immobilon-FL PVDF membrane (Millipore), and probed with various primary antibodies. Membranes were then incubated with IRDye® 800CW- or 680RD- conjugated secondary antibodies and visualized using a LI-COR Odyssey Imaging System (LI-COR Biosciences, Lincoln, NE, USA).

HFL cell α-SMA was visualized with an anti-α-SMA primary antibody (Sigma) and an Alexa Fluor 488-labeled secondary antibody (Life Technologies). The images were obtained by using Olympus FluoView® FV1200 confocal laser scanning microscope (Olympus Corporation, Center Valley, PA).

To construct luciferase reporter plasmid containing TGF-β receptor 1 (TGFBR1)-3UTR-133a, CTGF-3UTR-133a, or collagen type 1-alpha1 (Col1a1)-3UTR-133a, the 3′UTR fragments containing miR-133a binding sites/mutant sites of human TGFBR1, CTGF or Col1a1 were synthesized at Genewiz (Beijing, China) with XbaI and SacI cutting site and cloned into the pmirGLO reporter vector (Promega, Fitchburg, WI). miR-133a binding site mutants are shown by the underlined nucleotide changes.

The FTS1 promoter that is selectively activated in fibroblasts^45,46 was cloned into the pAAV-CMV-GFP plasmid (Cell Biolabs, San Diego, CA, USA) to replace the CMV promoter. The miR-133a with flanking sequence was amplified from human genomic DNAs and then was cloned into the pAAV-CMV-GFP plasmid or pAAV-FTS1-GFP vector to generate pAAV-CMV-GFP-miR-133a or pAAV-FTS1-GFP-miR-133a. Primers and probes used are listed in Supplementary Table 2. All constructs were validated by Sanger sequencing.

In all, 0.5 µg of the pmirGLO luciferase constructs containing the wild-type or mutants of 3′ UTR fragments of TGFBR1, CTGF or Col1a1, were co-transfected with control mimics or miR-133a mimics into HEK-293 cells using Lipofectamine 3000 according to the manufacturer’s instructions. Cells were harvested 24 h later and then subjected to luciferase activity assays using a Dual-Glo® Luciferase Assay kit (Promega) and Sirius luciferase assay system (Berthold, Germany) as we previously reported³³.

Gelatin zymography was performed as previously described⁴⁷. Briefly, gels (SDS-PAGE, 10%) were co-polymerized with gelatin (1 mg/mL) (Sigma-Aldrich). Equal amounts of samples were loaded onto the wells of the gel. After electrophoresis, gels were washed in renaturing buffer (2.5% Triton X-100 in 50 mM Tris-HCl, pH 7.5) for 3 h at room temperature and followed by 18 h of incubation at 37 °C in developing buffer (10 mM CaCl₂, 200 mM NaCl in 50 mM Tris-HCl, pH 7.5). Gels were then stained with Coomassie blue and de-stained with 30% methanol and 10% acetic acid. Gels were photographed by EPSON Perfection V500 (Long Beach, CA, USA).

Cells were lysed by sonication in 8 M urea/0.1 M Tris-HCl (pH 8.0)/1 × Protease Inhibitor Cocktail (Sigma). Cell lysates were reduced with 10 mM DTT for 2 h at room temperature followed by alkylation with 20 mM iodoacetamide for 30 min in the dark. Samples were trypsinized in 50 mM triethylammonium bicarbonate (TEAB) at 37 °C overnight, then desalted and eluted with 60% acetonitrile. Peptides were lyophilized and re-dissolved with 100 mM TEAB buffer. 100 μg of protein of each sample was labeled with TMT six-plex® (Thermo Fisher Scientific) according to the manufacturer’s instructions. Samples were fractionated using an L-3000 HPLC System (Rigol, Beijing, China) and all nanoLC-MS/MS experiments were performed on a Q Exactive equipped with an Easy n-LC 1000 HPLC system (Thermo Fisher Scientific) as previously described⁴⁸.

The raw data from Q Exactive were analyzed with Proteome Discovery version 2.2.0.388 using Sequest HT search engine for protein identification and Percolator for FDR (false discovery rate) analysis. The Uniprot Human protein database was used and FDR < 0.05 was set for protein identification. Normalization to the protein median of each sample was used to correct for experimental bias and the normalization mode was selected as total peptide amount. Among the 5394 total proteins detected, differentially regulated proteins (P < 0.05) were identified by multiple ANOVA analysis in conjunction with Dunnet’s analysis using R (3.2) and their associations with signaling pathways in HFL cells were explored using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database resource (www.kegg.jp↗) and the R package clusterProfiler that automates the process of biological-term classification and the enrichment analysis of gene clusters⁴⁹. Briefly, genes in each pathway for Homo sapiens were extracted from the KEGG database. Fish’s extract test was further applied to identify enriched pathways (unadjusted P < 0.05). Differentially expressed proteins were further validated by immunoblotting or RT-PCR.

Female C57BL/6JCnc mice (6–8 weeks of age) were obtained from Vital River Experimental Animal Center (Beijing, China) and maintained in a specific pathogen-free environment. All experiments were performed according to the guidelines for experimental animals and approved by the Institutional Animal Care and Use Committee of the Institute of Biophysics, Chinese Academy of Sciences.

The mice were intratracheally administered with either bleomycin (50 mg/kg per mouse in 50 μL of saline; Zhejiang Hisun Pharmaceutical Co., Ltd., Taizhou, China) or vehicle (control) on Day 0^22,40. To overexpress miR-133a specifically in mouse lung fibroblasts, 25 μg of pAAV-FTS1-GFP or pAAV-FTS1-GFP-miR-133a complexed with Entranster^TM-in vivo (Engreen Biosystem Co, Ltd., Beijing, China) was injected via the tail vein on Day 5 and every 4 days thereafter. The mice (n = 4) were killed on Day 15. Lungs were collected and inflated with 4% paraformaldehyde for 2 days and embedded in paraffin before sectioning into 5-μm-thick slices.

Sections of paraformaldehyde-fixed mouse lungs were analyzed by hematoxylin and eosin (H&E) or Masson’s trichrome staining to assess fibrotic changes in the lungs as described⁴⁰. Photos of at least 15 fields from multiple sections of each mouse lung were taken at ×200 magnification and scored separately by the modified Ashcroft method (score range 0–8)^40,50,51. The pulmonary fibrosis histopathology score of each mouse is expressed as the mean score of at least 15 photos.

All values are expressed as the mean ± SEM of at least three independent experiments. Data were analyzed using the GraphPad Prism software (GraphPad Software, Inc, San Diego, CA, USA). The Student’s unpaired t test (two-tailed) was used for comparison between two groups. *P < 0.05 was considered statistically significant; ns, not significant.

Supplementary Figure 1. Supplementary Figure 2. Supplementary Figure 3. Supplementary Table 1. Supplementary Table 2. Supplementary figure legends.