All cytokines and growth factors including TGFβ1 were purchased from Peprotech (Rocky Hill, NJ, United States). Hydroxyproline assay kits, and CCl₄ were purchased from Sigma-Aldrich (St. Louis, MO, United States). Bleomycin and target specific pooled siRNAs siSTAT3 and siSTAT6 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, United States). EPRS in pEXPR-103-Strep vector (IBA Lifesciences, Göettingen, Germany) were gifts from Dr. Myung Hee Kim at the Korea Research Institute of Bioscience and Biotechnology (KRIBB, Daejeon, Korea). EPRS (1–1440 amino acids) consists of ERS (1–687 amino acids) and PRS (935–1,440 amino acids) linked via non-catalytic WHEP repeat domains (688–934 amino acids) (Ray and Fox, 2014). The PRS domain of EPRS was cloned into pEXPR-103-Strep vector (IBA Lifesciences). pRc/CMV-WT STAT3 was previously described (Choi et al., 2009) and pCMV-STAT6-IRES-Neo was a gift from Axel Nohturfft (Addgene plasmid # 35482). Adenovirus expressing SMAD2 or SMAD3 were described previously (Lee et al., 2005).

A549 lung adenocarcinoma cells were purchased from the Korean Cell Line Bank (KCLB, Seoul, Korea) and cultured in RPMI (SH30027.01, Hyclone, South Logan, UT, United States). Media were supplemented with 10% fetal bovine serum (FBS, GenDEPOT, Barker, TX, United States) and 1% penicillin/streptomycin (GenDEPOT) and cells were grown at 37°C in 5% CO₂. The SMARTvector shEPRS doxycycline-inducible knockdown cell line was established by treating lentiviral particles (EPRS mCMV-turboGFP V2IHSMCG_687815, 687823, Dharmacon, Lafayette, CO, United States). Positive clones were enriched by treatment of 2 μg/ml puromycin (GenDEPOT) and maintained in complete media supplemented with 1 μg/ml puromycin. siRNAs or cDNA plasmids were transiently transfected using Lipofectamine RNAiMAX or Lipofectamine 3000, respectively, following the manufacturer's instructions (Thermo Fisher Scientific, Waltham, MA, United States).

Subconfluent cells or animal tissues were harvested for whole cell or tissue extracts using RIPA buffer. Proteins in the lysates were separated in Tris-Glycine SDS-polyacrylamide gels at concentrations ranging from 8 to 12%, and transferred to nitrocellulose membranes (Thermo Fisher Scientific). Target-specific antibodies used in this study are summarized in Table 1 (Supplementary Data Sheet 1).

Total RNAs from animal tissues or cells were isolated using Qiazol Reagent (Qiagen, Hilden, Germany), and their cDNAs were synthesized using amfiRivert Platinum cDNA synthesis master mix (GenDEPOT) according to the manufacturer's instructions. Quantitative real time PCR (q-PCR) samples were prepared with LaboPassTM EvaGreen Q Master (Cosmo Genetech, Seoul, Korea) prior to analysis in a CFX Connect™ Real-Time PCR machine (Bio-Rad, Hercules, CA, United States). mRNA levels were normalized against GAPDH and CFX Maestro™ software (Sunnyvale, CA, United States) was used to analyse the data. Primers were purchased from Cosmo Genetech (Seoul, Korea). The primer sequences are shown in Table 2.

Whole-cell lysates were prepared using immunoprecipitation lysis buffer (40 mM HEPES pH7.4, 150 mM NaCl, 1 mM EDTA, 0.5% Triton X-100, and protease inhibitors) and precipitated with Pierce High-Capacity Streptavidin Agarose beads (Thermo Fisher Scientific) overnight at 4°C. Precipitates were washed three times with ice-cold lysis buffer, three times with immunoprecipitation wash buffer (40 mM HEPES pH 7.4, 500 mM NaCl, 1 mM EDTA, 0.5% Triton X-100, and protease inhibitors), and then boiled in 2 × SDS-PAGE sample buffer before immunoblotting.

To analyze promoter activity, LAMC2 (laminin γ2) promoters (encoding regions of −1871 to +388) and COL1A1 (collagen I α1) promoters (encoding regions of −2865 to +89) were amplified by PCR and cloned into the pGL3-basic vector. A549 cells were seeded in 48 well plates and the next day the plasmids were transfected using Lipofectamine 3000 transfection reagent (Thermo Fisher Scientific). β-Gal was co-transfected to allow normalization. One day after transfection, TGFβ1 (2 ng/ml) was added to the culture media. After 24 h, luciferase activity was measured according to the manufacturer's instructions using a luciferase reporter assay kit (Promega, Madison, WI, United States) with a luminometer (DE/Centro LB960, Berthold Technologies, Oak Ridge, TN, United States).

Wildtype (WT) EPRS^+/+ (n = 4 for vehicle and n = 9 for bleomycin) and EPRS^−/+ hetero-knockout (n = 5 for vehicle and n = 7 for bleomycin) C57BL/6 mice were housed in a specific pathogen-free room with controlled temperature and humidity. Mouse protocol and animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Seoul National University (SNU-161201-1-3). For the lung fibrosis model, bleomycin (Santa Cruz Biotechnology) was dissolved in sterilized saline and intratracheal instillation was performed through surgically exposed trachea as a single dose of 1 mg/kg in 100 μl solution per animal. Mice were sacrificed 4 weeks post-intratracheal instillation. Lung tissue samples were snap frozen in liquid nitrogen for western blot, qPCR, and hydroxyproline analysis, or fixed in 4% formaldehyde in PBS for histological analysis.

Paraffin blocks and sections (6-μm thickness) of lung tissues were prepared by Abion Inc. (Seoul, Korea) for immunohistochemistry analysis. Primary antibodies and their dilution ratios are listed in Table 1. Vectastain ABC-HRP kit (Vector Laboratories, Burlingame, CA, United States) were used to visualize the stained samples. Mayer's hematoxylin (Sigma-Aldrich) was used for counter-staining the nuclei.

Statistical analyses were performed using Prism software version 6.0 (GraphPad, La Jolla, CA, United States). Two-way analysis of variance (ANOVA) in group analysis or Student's t-tests were performed to determine statistical significance. A value of p < 0.05 was considered significant.

We studied the regulatory effect of EPRS on the expression of different ECM proteins by introducing doxycycline-inducible EPRS knockdown vectors into the A549 alveolar type II cell line. Expression of ECM proteins such as collagen I, fibronectin, and laminin γ2 were tested by immunoblotting EPRS-knockdown and control A549 cells. All the ECM proteins showed increased expression in control cells treated with TGFβ1 and this effect was abolished in EPRS-knockdown cells (Figure 1A). Expression levels of mesenchymal proteins including α-smooth muscle actin (α-SMA) and snail 1 were also dependent on TGFβ1 treatment and/or EPRS expression (Figure 1A). Since EPRS protein consists of two glutamyl-tRNA-synthetase (ERS) and prolyl-tRNA-synthetase (PRS), it would be reasonable to see whether PRS alone could achieve these effects. Overexpression of PRS enhanced TGFβ1-induced ECM protein expression (Figure 1B). However, overexpression of ERS alone did not increase TGFβ1-mediated ECM protein expression (Figure 1C), indicating that the PRS component of EPRS regulates ECM protein expression. TGFβ1 treatment also increased mRNA levels of COL1A1, COL4A1, FN1, and LAMC2 in control cells, while EPRS suppression inhibited this effect (Figure 1D). Our results suggest that EPRS positively regulated TGFβ1-induced expression of ECM proteins. A previous report (Keller et al., 2012) stated that EPRS suppression increases mRNA levels of DNA damage-inducible transcript 3 (DDIT3, also known as CHOP) to indicate an activation of AAR pathway that supports for tRNA charging processes. However, we found that DDIT3 mRNA levels were unaffected by TGFβ1 treatment, indicating that TGFβ1-induced regulation of ECM protein expression involves alternative mechanism(s) in addition to the role in tRNA charging (Figure 1D).

To investigate potential signaling molecules or pathways involved in EPRS-mediated regulation of ECM protein synthesis following TGFβ1-treatment, we studied the dependency of STAT3 and STAT6, known mediators of IPF (Nikota et al., 2017; Milara et al., 2018), on EPRS expression. We found that TGFβ1 promoted the phosphorylation of STAT3 at Tyr705 (pY⁷⁰⁵STAT3) and STAT6 at Tyr641 (pY⁶⁴¹STAT6), and these effects were abolished by EPRS suppression (Figure 2A). EPRS overexpression increased pY⁷⁰⁵STAT3 and pY⁶⁴¹STAT6 levels upon TGFβ1 treatment in A549 cells (Figure 2B). Our results suggest that EPRS and TGFβ1 signaling regulate STAT3 and STAT6 phosphorylation. We studied the role of EPRS in STAT-mediated expression of ECM proteins. Overexpression of STAT6 indicate greater increases in basal and TGFβ1-induced levels of α-SMA, snail 1, fibronectin, and collagen I in EPRS-positive A549 cells compared with EPRS-knockdown cells (Figure 2C). However, basal and TGFβ1-induced expression levels of fibronectin and collagen I were decreased when STAT6 levels were suppressed (Figure 2D). A similar EPRS-dependent regulation pattern of fibronectin and collagen I was observed when STAT3 was modulated (Figures 2E,F). We also tested the transcriptional activities of COL1A1 and LAMC2 promoters in A549 cells lacking STAT3 or STAT6. COL1A1 or LAMC2 promoters containing STAT-responsive consensus sequences showed increased transcriptional activity in A549 cells treated with TGFβ1. However, EPRS suppression reduced these effects (Figure 2G). Suppression of STAT3 or STAT6 abolished the increased transcriptional activity of COL1A1 or LAMC2 in EPRS-positive A549 cells but not EPRS-suppressed cells (Figure 2G). Together, our results suggested that EPRS regulates ECM protein expression via STAT3 or STAT6 signaling induced by TGFβ1.

We investigated the role of the TGFβ1-mediated SMAD signaling in EPRS-dependent ECM protein expression and STAT3/6 activity. TGFβ1-mediated SMAD2 and SMAD3 phosphorylation was partially inhibited by EPRS suppression (Figure 3A). However, EPRS overexpression did not affect the levels of phosphorylated SMAD2 or SMAD3, which might have already been saturated by TGFβ1 treatment (Figure 3B). TGFβ1-induced levels of pY⁶⁴¹STAT6 were increased by overexpression of SMAD3 but not SMAD2. EPRS suppression abolished that effect (Figures 3C,D).

We then tested for potential protein-protein interactions between TGFβ1 signaling and EPRS that regulate STAT6 phosphorylation. A549 cells containing Streptavidin-tagged EPRS (Strep-EPRS) were treated with or without TGFβ1, prior to precipitation of whole-cell extracts using streptavidin agarose beads for immunoblotting assays. Lysyl-tRNA synthetase (KRS), which forms a multi-aminoacyl-tRNA synthetase complex (MSC) with EPRS (Park et al., 2008), was used as a positive control. In TGFβ1-treated cells, Strep-EPRS transiently precipitated with TGFβR1, JAKs, and STATs, which included STAT6 (Figure 4A). We tested the effect of STAT3 or STAT6 suppression on multi-protein interactions. STAT6 expression was required for the EPRS-mediated multi-protein complex formation (Figure 4B). Specifically, EPRS interaction with TGFβ1R and SMAD2/3 required STAT6 but not STAT3. Interestingly, STAT3 suppression resulted in increased binding of STAT6 to the EPRS/TGFβ1R-containing protein complex (Figure 4B). Interactions between EPRS and JAKs were independent of STAT3 and STAT6 (Figure 4B). Our results suggest that STAT6 is critical for the formation of the multi-protein complex for TGFβ1-induced signaling of ECM protein expression. STAT3 and STAT6 may be involved in parallel signaling pathways to regulate this process.

To investigate the physiological roles of EPRS in pulmonary fibrosis in vivo, WT (Eprs^+/+) and Eprs^−/+ hetero-knockout (KO) mice were treated with bleomycin to induce lung fibrosis by intratracheal instillation before analysis, since homozygous Eprs^−/− is embryonic lethal. Bleomycin-treated WT Eprs^+/+ mice showed the highest increase in expression of ECM proteins, such as fibronectin, collagen I, and laminins, compared with bleomycin-treated Eprs^−/+ hetero-KO mice and untreated WT mice (Figure 5A). Levels of pY⁷⁰⁵STAT3 and pY⁶⁴¹STAT6 were also elevated in bleomycin-treated Eprs^+/+ mice, compared with Eprs^−/+ hetero-KO mice (Figure 5A). Our results suggest that STAT6 activation is part of EPRS-dependent signaling for ECM protein expression in vivo. We observed a slight upregulation in EPRS expression in bleomycin-treated WT Eprs^+/+ mice, compared with other groups, indicating that EPRS might function as a pro-fibrotic molecule.

Hydroxyproline assays to measure collagen I levels in lung extracts showed that bleomycin-treated Eprs^+/+ mice had higher levels compared with bleomycin-treated Eprs^−/+ mice (Figure 5B). In addition to IL8, which is used to characterize IPF (Carre et al., 1991), Col1a1, Fn1, and Lamc2 mRNA levels were also upregulated by bleomycin treatment in Eprs^+/+ lungs but not Eprs^−/+ hetero-KO lungs (Figure 5C).

Lung immunohistochemistry revealed more patchy fibrosis and fibroblastic foci in bleomycin-treated Eprs^+/+ mice compared with Eprs^−/+ mice (Figure 5D). Bleomycin treatment also led to marked increases in collagen I, fibronectin, and laminin γ2 synthesis in Eprs^+/+ mice compared with Eprs^−/+ mice. Levels of phospho-SMAD2, pY⁷⁰⁵STAT3, and pY⁶⁴¹STAT6, which showed intense nuclear staining, were diminished in Eprs^−/+ mice (Figure 5D). The myofibroblasts marker α-SMA was positive in fibroblastic foci of bleomycin-treated mice lungs. These results suggest that the bleomycin-mediated fibrotic phenotypes in animal lungs are dependent on EPRS expression.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The handling editor declared a shared department with several of the authors (D-GS, DK, JWJ, SHN, JEK, H-JK, JHK, SK and JWL) at time of review.