Gas6 Prevents Epithelial-Mesenchymal Transition in Alveolar Epithelial Cells via Production of PGE2, PGD2 and Their Receptors

The epithelial-mesenchymal transition (EMT) is important in organ fibrosis. We hypothesized that growth arrest-specific protein 6 (Gas6) and its underlying mechanisms play roles in the prevention of EMT in alveolar epithelial cells (ECs). In this study, to determine whether Gas6 prevents TGF-β1-induced EMT in LA-4 and primary alveolar type II ECs, real-time PCR and immunoblotting in cell lysates and ELISA in culture supernatants were performed. Migration and invasion assays were performed using Transwell chambers. Pretreatment of ECs with Gas6 inhibited TGF-β1-induced EMT based on cell morphology, changes in EMT marker expression, and induction of EMT-activating transcription factors. Gas6 enhanced the levels of cyclooxygenase-2 (COX-2)-derived prostaglandin E2 (PGE2) and PGD2 as well as of their receptors. COX-2 inhibitors and antagonists of PGE2 and PGD2 receptors reversed the inhibition of TGF-β1-induced EMT, migration, and invasion by Gas6. Moreover, knockdown of Axl or Mer reversed the enhancement of PGE2 and PGD2 and suppression of EMT, migration and invasion by Gas6. Our data suggest Gas6-Axl or -Mer signalling events may reprogram ECs to resist EMT via the production of PGE2, PGD2, and their receptors.


Introduction
The epithelial to mesenchymal transition (EMT) events have been very well defined during embryogenesis, organ development, wound healing and stem cell behaviors [1]. This process has also been shown to contribute pathologically to organ fibrosis and cancer progression. Emerging evidence suggests that EMT is an important event in idiopathic pulmonary fibrosis (IPF) [2][3][4]. IPF likely results from recurrent alveolar epithelial cell (EC) injury coupled with accumulation and differentiation of fibroblasts into myofibroblasts, leading to the accumulation of extracellular matrix (ECM) and destruction of the lung parenchyma [5]. Alveolar ECs undergoing epithelial-mesenchymal transition (EMT), a process frequently mediated by TGF-β [6], are a major source of fibroblasts [7,8]. Although several drugs are currently used to treat IPF, no proven, efficacious therapies currently exist [9].

Immunoblot Analysis
ECs were lysed in 0.5% Triton X-100-containing lysis buffer and resolved on a 10% SDS-PAGE gel. Separated proteins were electrophoretically transferred onto nitrocellulose and blocked for 1 h at room temperature with Tris-buffered saline containing 3% BSA. The membranes were then incubated at room temperature for 1 h with primary antibodies at 4 • C overnight and incubated with HRP-conjugated secondary antibodies for 1 h at room temperature. Proteins were visualized using enhanced chemiluminescence, according to the manufacturer's instructions.

Migration and Invasion Assays
Cell migration and invasion were tested using Transwell chambers (Corning Inc., Corning, NY, USA) coated with 10 µg/mL fibronectin and 300 µg/mL Matrigel matrix according to the manufacturer's instruction, respectively. In brief, pre-incubated primary mouse AT II cells or LA-4 cells (5 × 10 4 cells/well for the migration assay and 5 × 10 4 cells/well for the invasion assay) in the absence or presence of TGF-β1 (10 ng/mL) were plated in replicate wells in serum-free RPMI in the upper chambers and in RPMI 1640 supplemented with 10% FBS placed in the bottom wells at 37 • C for 48 h migration time or 48 h invasion time. The nonmigrated or noninvaded cells on the upper surface of the membrane were removed with a cotton swap. Migrated cells on the lower surface were fixed with 4% paraformaldehyde and stained using 0.1% crystal violet. Three random microscopic fields (10× magnification) were photographed and counted. All experiments were performed in triplicates.

Enzyme-Linked Immunosorbent Assay (ELISA)
Culture supernatants were collected. The levels of PGE 2 and PGD 2 concentration were measured using an EIA kit according to the manufacturer's instructions.

Statistical Analysis
Data are expressed as the mean ± S.E. Analysis of variance was used for multiple comparisons, and Tukey's post hoc test was used where appropriate. The Student's t test was used to compare two sample means. A P value less than 0.05 was considered statistically significant. All data were analysed using JMP software (SAS Institute, Cary, NC, USA).

Gas6 Inhibits TGF-β1-Induced EMT in Lung and Kidney ECs
Pretreatment with 400 ng/mL Gas6 prevented a spindle-like morphology ( Figure 1A) and changes in EMT markers, such as decreased E-cadherin and increased N-cadherin, and α-SMA, at both the protein and mRNA levels after a 48-or 72-h stimulation with TGF-β1 in LA-4 ECs ( Figure 1B,C). We also observed this inhibitory effect in ATII ECs ( Figure 1B,C), A549 human non-small lung cancer cells, and HEK293 human kidney cells (Supplementary Figure S1A). However, EMT marker protein expression was not inhibited when pretreatment occurred 2 h before TGF-β1 treatment or the culture medium was replaced 20 h after Gas6 pretreatment prior to TGF-β1 stimulation for 72 h (Supplementary Figure S1B,C).

Gas6 Inhibits TGF-β1-Induced EMT in Lung and Kidney ECs
Pretreatment with 400 ng/mL Gas6 prevented a spindle-like morphology ( Figure 1A) and changes in EMT markers, such as decreased E-cadherin and increased N-cadherin, and α-SMA, at both the protein and mRNA levels after a 48-or 72-h stimulation with TGF-β1 in LA-4 ECs ( Figure  1B,C). We also observed this inhibitory effect in ATII ECs ( Figure 1B,C), A549 human non-small lung cancer cells, and HEK293 human kidney cells (Supplementary Figure S1A). However, EMT marker protein expression was not inhibited when pretreatment occurred 2 h before TGF-β1 treatment or the culture medium was replaced 20 h after Gas6 pretreatment prior to TGF-β1 stimulation for 72 h (Supplementary Figure S1B,C).

Gas6 Enhances COX-2-Derived Production of PGE 2 , PGD 2 , and Their Receptors
COX-2 mRNA abundance peaked at 1 h and returned to resting levels 20 h after Gas6 treatment in LA-4 and ATII ECs ( Figure 3A). COX-2 protein expression in LA-4 ECs increased up to 24 h in LA-4 ECs ( Figure 3B). PGE 2 and PGD 2 production increased in LA-4 ECs 20 h after Gas6 treatment ( Figure 3C) but was blocked by COX-2 siRNA ( Figure 3D). Interestingly, mRNA and protein levels of EP2 and DP2 were enhanced 20-24 h after Gas6 treatment, whereas EP4 and DP1 mRNA and protein levels were unaffected, in LA-4 ECs ( Figure 3E,F). However, transfection of LA-4 ECs with COX-2 siRNAs inhibited Gas6-indued Ptger2 and Dp2 mRNA expression ( Figure 3G). Increases in Ptger2 and Dp2 mRNA expression by Gas6 were also shown in ATII ECs ( Figure 3H). ERK1/2, and -Akt protein antibodies. Beta-actin or alpha-tubulin was used as a loading control. Densitometric analysis of the indicated protein abundances. Data in all bar graphs are the mean ± S.E. of three independent experiments. * P < 0.05 compared with control; + P < 0.05 as indicated.

Axl and Mer Receptor Tyrosine Kinases are Involved in Anti-EMT Effects of Gas6 in LA-4 ECs
Axl and Mer receptors were rapidly activated in LA-4 ECs after Gas6 stimulation ( Figure 6A,B). Knockdown of Axl or Mer reversed the enhanced induction of COX-2 mRNA expression by Gas6 as well as PGE 2 and PGD 2 production, and Ptger2 and Dp2, but not Ptger4 and Dp1, mRNA expression ( Figure 6C-G).

Axl and Mer Receptor Tyrosine Kinases are Involved in Anti-EMT Effects of Gas6 in LA-4 ECs
Axl and Mer receptors were rapidly activated in LA-4 ECs after Gas6 stimulation ( Figure 6A,B). Knockdown of Axl or Mer reversed the enhanced induction of COX-2 mRNA expression by Gas6 as well as PGE2 and PGD2 production, and Ptger2 and Dp2, but not Ptger4 and Dp1, mRNA expression ( Figure 6C-G). The inhibitory effects of Gas6 on EMT marker changes in LA-4 ECs at the protein and gene levels, and the downregulation of the mRNA expression of EMT transcription factors, were reversed by their specific siRNAs ( Figure 6H,I; Supplementary Figure S5A,B). Axl-or Mer-specific siRNAs also reversed the reduction of TGF-β1-induced phosphorylation of ERK1/2 and Akt by Gas6 ( Figure 6J; Supplementary Figure S5C).

Gas6 Inhibits TGF-β1-Induced Migration and Invasion in ATII Cells
Pretreatment of LA-4 or ATII ECs with Gas6 inhibited TGF-β1-induced cell migration and invasion ( Figure 7A-D; Supplementary Figure S6A-D). However, COX-2 inhibitor NS398 or antagonists of EP2, and DP2, but not DP4 and DP1, reversed Gas6-induced inhibition of cells migration and invasion. Similarly, siRNAs of Axl or Mer also reversed the inhibitory effects of Gas6 on TGF-β1-induced migration and invasion of LA-4 ECs ( Figure 7E,F).

Discussion
Gas6/Axl signaling plays an important role in the control of epithelial cell growth and survival [32]. For example, Gas6/Axl signaling mediates a survival or anti-apoptotic response in quiescent lens ECs under conditions of growth factor deprivation and proliferating lens ECs from TGFβ1-or TNFα-induced apoptosis [26]. Recently, it was proposed that Gas6/Axl signaling plays a critical role in oral epithelial cells as a key immunological regulator of host-commensal interactions [32]. These data suggest a key homeostatic role of Gas6 in the maintenance of tissue epithelial cells. In the present study, we focused on a new role for Gas6 in preventing EMT, cell migration and invasion in lung ATII ECs. Gas6 treatment 20 h prior to TGF-β1 inhibited changes in EMT markers in lung ECs, including LA-4 ECs, primary ATII ECs, A549 cells, and HEK-293 kidney cells. The master regulators of EMT, including Snail, Zeb, and Twist, are upregulated by TGF-β via both Smad-dependent and -independent mechanisms [33]. We demonstrated that Gas6 blocked non-Smad TGF-β1 signalling and downregulated the mRNA expression of EMT transcription factors. Notably, the anti-EMT effect of Gas6 was not observed after the replacement of fresh culture media prior to TGF-β addition. These data strongly suggest that Gas6 inhibits EMT through the induction and secretion of bioactive mediators to block TGF-β signalling in an autocrine/paracrine manner.
The COX-2/PGE 2 and PGD 2 pathways result in the inhibition of EMT in lung and renal cells [25,34]. Thus, we hypothesized that PGE 2 and PGD 2 secretion by Gas6 mediates anti-EMT effects in an autocrine/paracrine manner. We found that Gas6 treatment enhances COX-2 expression with secretion of these mediators from LA-4 ECs. We confirmed, using specific siRNAs for COX-2, the secretion of PGE 2 and PGD 2 after Gas6 stimulation. Moreover, either knockdown of COX-2 or pharmacologic inhibition of COX-2 activity by NS398 reversed the inhibition of TGF-β1-induced EMT and EMT transcription factor expression in LA-4 and ATII ECs by Gas6. In particular, COX-2 siRNAs also reversed the inhibitory effect of Gas6 on non-Smad TGF-β1 signaling in LA-4 ECs. Therefore, these data indicate that Gas6 prevents non-Smad TGF-β signaling and EMT in ATII ECs via PGE 2 and PGD 2 secretion.
Considering the effects of the agonists of EP1-4, the suppressive effects of PGE 2 on TGF-β1-induced EMT may be mediated via the EP2 and EP4 receptors in A549 cells [35]. In the current study, we found that antagonists of EP2 and DP2, but not of EP4 and DP1, reversed Gas6-induced anti-EMT effects in LA-4 and ATII ECs. Interestingly, the mRNA and protein expression levels of EP2 and DP2, but not of EP4 and DP1, were enhanced up to 24 h in a time-dependent manner by Gas6 in LA-4 ECs. Taken together, these data suggest that the increases in EP2 and DP2 abundance and activity contribute, at least in part, to the anti-EMT effect of Gas6 in ATII ECs.
The functions of Gas6 appear limited to those caused by the activation of TAM receptors (Axl>Tyro3>>>Mer). In the present study, Gas6 induced higher levels of Axl phosphorylation than Mer phosphorylation in LA4 ECs. However, siRNA knockdown of Axl or Mer reversed the inhibition of changes in TGF-β1-induced EMT markers at the gene and protein levels by Gas6, restoring Snai1, Zeb1, and Twist1 mRNA expression to similar levels. Of note, we found that PGE 2 and PGD 2 production as well as reduction in transforming growth factor (TGF)-β1-induced ERK1/2 and Akt phosphorylation levels by Gas6 were blocked by specific siRNAs for Axl or Mer. These data suggest that the Gas6-Axl and -Mer pathways mediate PGE 2 and PGD 2 production, maintain EC homeostasis, and prevent pathologic process EMT in LA-4 ECs after TGF-β1 stimulation. There may be an interdependence of these related receptors in LA-4 ECs. In platelets, all three TAM family receptors are expressed; the targeted knockout of a single receptor was shown to result in a phenotype of decreased platelet aggregation and clot instability [36,37]. In addition, the inhibition of related receptor tyrosine kinases was found to cause a significant decrease in astrocytoma survival and growth [38]. Further studies are needed to fully determine how receptor tyrosine kinases interact under specific biological conditions. EMT is characterized by the enhancement of cellular migration and invasive potential. Fibroblasts and myofibroblasts from IPF patients have been shown to have distinct properties [39], including the ability to invade the ECM similar to metastatic cancer cells [40]. In addition to anti-EMT effects, we demonstrated that Gas6/Axl or Mer signaling inhibits TGF-β1-induced migration and invasion of LA-4 and ATII ECs via PGE 2 and PGD 2 production.
Lee et al. [41] reported that Gas6 enhanced the migration and invasion of various hepatocellular carcinoma (HCC) cell lines in the absence of TGF-β1. The Gas6-Axl pathway enhanced Snai1 mRNA expression. However, we demonstrated that Gas6 itself did not affect mRNA or protein expression of EMT markers or transcription factors in a variety of ECs; Gas6 also did not affect the migration or invasion of LA-4 or ATII ECs in the absence of TGF-β1. We believe that the diversification of Gas6 function via Mer and Axl activation may be due to differences in the experimental design of our study and those of other studies.
Gas6/Axl signaling induces proliferation of mesangial cells in kidney fibrosis [42], vascular smooth muscle cells response to intimal vascular injury [43], and cardiac fibroblasts during prolonged serum starvation. [44]. Thus, this pathway has been implicated in growth and survival processes during regeneration, and tissue repair. The selective targeting of Gas6-Axl specific antibodies or small-molecule inhibitors of TAM receptors was shown to modulate the activation of fibroblasts in patients with IPF [45]. On the other hand, an exogenous treatment with protein S, a TAM receptor ligand, significantly decreased the levels of inflammatory and profibrotic markers, decreased lung fibrosis, and blocked apoptosis in alveolar ECs [46]. These controversial data suggest that the roles of Gas6 and TAM receptors are complex; thus, future studies are necessary to more fully understand the diverse roles of Gas6-TAM signaling in certain cell types and/or in vivo disease models, considering different cellular microenvironments.
TAM receptors have been implicated in human cancers by virtue of their pathological overexpression [47][48][49][50]. Gas6 is also concomitantly overexpressed in human cancers, implying that, besides receptor overexpression, TAM participate in autocrine circuits by overexpressing both receptors and ligands [51][52][53]. Profound experimental evidence supports the unanimity that TAM receptors, in a cell-autonomous manner, act as pro-oncogenes enhancing the growth, survival, migration, and EMT of cancer cells [54]. Paradoxically, experimental evidence has demonstrated that the marked susceptibility to experimentally induced colitis observed in mutant mice lacking Gas6 as well as Axl and Mer correlates with increased incidence of colon cancer, resulting in larger tumors and reduced overall survival [55,56]. These studies indicate the complicated implication of Gas6 and TAM in cancer and underscore the importance of understanding their tissue and cell type-specific functions in cancer. To end this, further mechanistic insights into TAM-mediated regulation of anti-tumor immune responses are needed.

Conclusions
In summary, our study reveals a critical role for Gas6-Axl or -Mer signaling in the prevention of TGF-β1-induced EMT, migration, and invasion in ATII ECs. The production of COX-2-derived PGE 2 and PGD 2 and their receptors is essential for the effects of Gas6. Considering that EMT is has been implicated in the pathogenesis of fibrosis in response to epithelial stress and injury in multiple organs, including the lung, kidney, liver, and peritoneum, the present data suggest that Gas6 could be used to develop preventive and therapeutic strategies to limit pathologic fibrosis in diverse organ diseases.

Conflicts of Interest:
The authors have nothing to disclose.