Secreted Phospholipase A2-IIA Modulates Transdifferentiation of Cardiac Fibroblast through EGFR Transactivation: An Inflammation–Fibrosis Link

Secreted phospholipase A2-IIA (sPLA2-IIA) is a pro-inflammatory protein associated with cardiovascular disorders, whose functions and underlying mechanisms in cardiac remodelling are still under investigation. We herein study the role of sPLA2-IIA in cardiac fibroblast (CFs)-to-myofibroblast differentiation and fibrosis, two major features involved in cardiac remodelling, and also explore potential mechanisms involved. In a mice model of dilated cardiomyopathy (DCM) after autoimmune myocarditis, serum and cardiac sPLA2-IIA protein expression were found to be increased, together with elevated cardiac levels of the cross-linking enzyme lysyl oxidase (LOX) and reactive oxygen species (ROS) accumulation. Exogenous sPLA2-IIA treatment induced proliferation and differentiation of adult rat CFs. Molecular studies demonstrated that sPLA2-IIA promoted Src phosphorylation, shedding of the membrane-anchored heparin-binding EGF-like growth factor (HB-EGF) ectodomain and EGFR phosphorylation, which triggered phosphorylation of ERK, P70S6K and rS6. This was also accompanied by an up-regulated expression of the bone morphogenic protein (BMP)-1, LOX and collagen I. ROS accumulation were also found to be increased in sPLA2-IIA-treated CFs. The presence of inhibitors of the Src/ADAMs-dependent HB-EGF shedding/EGFR pathway abolished the CF phenotype induced by sPLA2-IIA. In conclusion, sPLA2-IIA may promote myofibroblast differentiation through its ability to modulate EGFR transactivation and signalling as key mechanisms that underlie its biological and pro-fibrotic effects.


Introduction
Inflammation and autoimmunity are involved in the progression of many heart diseases. Myocarditis is a precursor of dilated cardiomyopathy (DCM) and represents the most common cause of chronic heart failure or even sudden death; nevertheless, little is known about the mechanisms

Animals and Immunization
BALB/c mice from Charles River Laboratories were housed in the animal care facility at the Medical School of the University of Valladolid (UVa) and were provided food and water ad lib, under standard conditions. All experimental protocols were reviewed and approved by the Animal Ethics Committee of the UVa (Project number 6203828) and were in accordance with European legislation (86/609/EU).
Disease was induced in 6-8 week-old male mice by immunisation at day 0 with 50 µg of the murine specific α-myosin-heavy chain-derived acetylated peptide (MyHCα 614-629 ), as was previously described [25]. MyHCα 614-629 was generated in the peptide synthesis laboratory of Dr. F. Barahona (CBM, Madrid, Spain). After terminal anesthesia with xylazine/ketamine, mice were sacrificed either on day 21 or 65. The heart was removed and weighed.

Histological and Immunohistochemical Studies
Hearts were obtained on day 65 from control and EAM mice. One-half was fixed in 4% paraformaldehyde and embedded in paraffin and the other half was frozen at −80 • C. Embedded tissues were cut in 5 µm thick sections, stained with hematoxylin-eosin (H&E) and Masson's trichrome (Sigma-Aldrich, St Louis, MO, USA), and examined by light microscopy. For the purposes of this study, each specimen was evaluated qualitatively with a Nikon Eclipse 90i microscope connected to a DS-Ri1 digital camera (Nikon Instruments Inc., Amstelveen, the Netherlands) with a 20× objective lens. Sections from 4-10 segments per mouse were examined blindly by two investigators.
Immunohistochemistry was carried out on 5 µm sections mounted on lysine-coated glass. Tissue was permeabilized with Tween 20 for 15 min and blocked with 5% serum for 20 min at room temperature; antigen retrieval was by heat mediation in a citrate buffer. Samples were incubated with anti-LOX antibody (1/100 in 10% serum in TBS + 0.05% Tween) for 14 h at 4 • C. An FITC anti-rabbit IgG polyclonal (1/500) was used as the secondary antibody. Images were obtained on a Leica TCS SP5X confocal microscope (TCS Leica Microsystems, Mannheim, Germany). Bars 50 µm).

In Situ Detection of Superoxide Production
To evaluate in situ superoxide production from hearts, unfixed frozen 8 µm thick cross-sections were stained with 2 µM dihydroethidium (DHE; Molecular Probes, Eugene, OR, USA) at 37 • C for 30 min in a light-protected humidified chamber. Images were obtained with a Nikon Eclipse 90i inverted fluorescence microscope using 2× or 20× objective lenses. Red fluorescence was collected through a 590 nm filter after excitation of cells at 488 nm.
2.5. Measurements of sPLA 2 -IIA by an Enzyme-Linked Immunosorbent Assay (ELISA) sPLA 2 -IIA levels were determined in both serum samples and heart tissue using a commercial ELISA (Cusabio Biotech Co, Wuhan, China), according to the manufacturer's protocols. Heart tissue homogenates were prepared with the apical part of the heart, homogenised in 1 mL of ice-cold PBS, supplemented with a protease inhibitor cocktail (Sigma-Aldrich, St Louis, MO, USA). Samples were centrifuged at 800× g for 15 min at 4 • C. Total protein concentration in the supernatants was determined by using the Bradford method with bovine serum albumin (BSA) as standard. Data were processed and expressed as concentration of cytokine/mg of tissue or concentration of cytokine/mL for serum samples.

Cell Culture
Primary cultures of CFs were obtained from male Wistar rats weighing 250-300 g by differential centrifugation of cardiac cells released after mechanical and enzymatic digestion of the hearts [25]. Cells were grown on poly-L-lysine coated flasks in DMEM supplemented with 10% FCS, 10 mM Cells 2020, 9,396 4 of 20 l-glutamine, 100 U/mL penicillin/streptomycin, 10 mM pyruvate and 2 mM HEPES. These cells were labeled as P1 and used at 2-3 passages to minimise changes in phenotype associated with culture. None variation between individual fibroblast preparations was observed using routine phenotyping methods with anti-vimentin and anti-α-SMA antibodies.

Cell Proliferation Assay
Cell proliferation was quantified by using the Promega kit, Cell Titer 96 ® Aqueous One Solution Cell Proliferation Assay, according to the manufacturer's recommendations. Briefly, CFs were seeded on 96-well culture plates (20 × 10 3 cells/well) and serum-starved for 24 h. Cells were pre-treated with either vehicle or the indicated inhibitor for 30 min before the addition of 1 µg/mL of sPLA 2 -IIA or 1 µM of angiotensin II (AngII). After 24 h of incubation, the proliferative response was quantified by recording the absorbance at 490 nm in a microplate reader (OD value). Formazan product formation is measured as an assessment of the number of metabolically active cells. Assays were each performed in quintuplicate, n = 3.

Quantification of Collagen Deposition
In vitro fibrotic activity of sPLA 2 -IIA was assessed by picro-Sirius Red (pSR) staining and quantitative analysis, as previously was reported [26]. In brief, CFs cultured in 12-well plates were treated with 1 µg/mL of sPLA 2 -IIA at 37 • C for 72 h. Afterwards, cells were fixed in methanol overnight at 4 • C, carefully washed twice with PBS and incubated in the 0.1% pSR staining solution (Sigma-Aldrich) at room temperature for 1 h. The staining solution was removed and cells were washed three times with 0.1% acetic acid. Then, pSR was eluted in 0. To evaluate mitochondrial transmembrane potential (∆ψm), CFs were loaded with 4 µM rhodamine 123. After washing, cells were treated with 1 µg/mL of sPLA2-IIA for 6 or 24 h at 37 • C and analysed by flow cytometry.

Flow Cytometric Analysis
CFs, 5 × 10 6 /flask, were treated with 1 µg/mL of sPLA 2 -IIA for the indicated times in the absence or presence of selected inhibitors at 37 • C. Cells to be analysed for expression of HB-EGF were fixed in a mixture of 4% paraformaldehyde in PBS for 15 min at room temperature before incubation with FITC-conjugated anti-HB-EGF antibody (Calbiochem, San Diego, CA, USA) for 1 h at 4 • C. For cellular LOX, BMP-1, paxillin, α-smooth muscle actin (α-SMA), TFGβ, phospho-EGFR and phospho-Src analysis, CFs were fixed in 4% paraformaldehyde for 15 min, washed with PBS and permeabilised with 0.3% Triton X-100 for 5 min. After that, cells were incubated with specific primary antibodies for 1 h at 4 • C, and then with a FITC-labelled secondary antibody for 45 min at 4 • C. After washing, Cells 2020, 9,396 5 of 20 the cells were analysed with a Flow Cytometer (Gallios TM ; Beckman Coulter, Brea, CA, USA). Data analysis was performed using WinMDI 2.7 software.
2.6.6. Measurements of LOX Enzymatic Activity LOX enzyme activity was measured in the supernatant from CFs cultures by using the Amplite™ Fluorimetric Lysyl Oxidase Assay Kit (AAT Bioquest Inc., Sunnyvale, CA, USA), according to the manufacturer's instructions. Parallel assays were prepared with 500 µM beta-aminopropionitrile fumarate (BAPN) to inhibit LOX activity and ensure assay specificity. Fluorescence was measured using a fluorescence spectrophotometer (Ex/Em = 535/590). Data are expressed as fold-change when compared with control conditions.

Western Blot Analysis
After treatment, CFs were washed twice with PBS and harvested in Laemmli SDS sample buffer. Equal amounts of protein extracts were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes; after blocking, they were incubated for 18 h at 4 • C with primary antibodies against actin (Santa Cruz Biotechnology Inc, Santa Cruz, CA, USA), ERK 1/2 (Zymed Laboratories, San Francisco, CA, USA), p-ERK1/2 (Thr202/Tyr204), phospho-P70S6 kinase (p-P70S6K, Thr389), and phospho-S6 ribosomal protein (p-rS6, Ser235/236) (Cell Signaling Technology, Inc, Danvers, MA, USA). After washing, membranes were incubated with the corresponding secondary antibody (1:2000, v/v) conjugated with horseradish peroxidase at room temperature for 30 min. The blots were developed using enhanced chemiluminescence. Experiments shown here were repeated at least 3 times with reproducible results and a representative one is presented.

Immunofluorescence Studies
Resting, AngII-and sPLA 2 -IIA-treated CFs, on 18-mm poly-l-lysine coated cover glass coverslips, were fixed in cold 4% paraformaldehyde for 10 min, permeabilised in 0.3% Triton X-100 for 10 min and blocked with 3% BSA for 1 h. Immunostaining with anti-α-SMA, LOX or BMP-1 antibodies (Abcam, Cambridge, UK) was performed at room temperature for 30 min, followed by incubation with FITC-conjugated secondary antibody. Cells were then analysed in a Nikon Eclipse 90i. Objective lens 20×.

Statistical Analyses
Differences between 2 groups were analysed by unpaired Student's t-test. Specific differences between more groups were analysed using 1-way ANOVA followed by Bonferroni's test where appropriate. Results are described as mean ± SD. p < 0.05 was considered statistically significant. Statistical analyses were performed using the GraphPad Prism Version 4 software (San Diego, CA, USA). The expression of sPLA 2 -IIA was investigated in hearts from EAM-affected mice. MyCHαimmunised mice show an increase in HW/BW index of 40% and 51% higher, at days 21 and 65, respectively, compared to those of the healthy un-induced group (p < 0.001, Figure 1A). In addition, and confirming the myocarditis development, the histological analysis shows inflammatory infiltrates and fibrotic areas in the myocardium, evidenced by H&E and Masson's trichrome staining, respectively, in EAM, but not in sham-immunised mice, at day 65 post-immunisation ( Figure 1B). Moreover, and consistent with previously reported, ROS cardiac levels were higher in EAM mice than in control ones, as suggested by the stronger brightness (fluorescent intensity) ( Figure 1C). Likewise, immunofluorescence staining of the key collagen cross-linking enzyme, LOX, showed a higher expression in the cardiac tissue of EAM mice than in control mice ( Figure 1D). Figure 1E shows that the sPLA 2 -IIA protein was up-regulated in the hearts and serum from mice, while it was barely detectable in control mice.

In Vivo Findings
The expression of sPLA2-IIA was investigated in hearts from EAM-affected mice. MyCHα-immunised mice show an increase in HW/BW index of 40% and 51% higher, at days 21 and 65, respectively, compared to those of the healthy un-induced group (p < 0.001, Figure 1A). In addition, and confirming the myocarditis development, the histological analysis shows inflammatory infiltrates and fibrotic areas in the myocardium, evidenced by H&E and Masson's trichrome staining, respectively, in EAM, but not in sham-immunised mice, at day 65 post-immunisation ( Figure 1B). Moreover, and consistent with previously reported, ROS cardiac levels were higher in EAM mice than in control ones, as suggested by the stronger brightness (fluorescent intensity) ( Figure 1C). Likewise, immunofluorescence staining of the key collagen cross-linking enzyme, LOX, showed a higher expression in the cardiac tissue of EAM mice than in control mice ( Figure 1D). Figure 1E shows that the sPLA2-IIA protein was up-regulated in the hearts and serum from mice, while it was barely detectable in control mice.

sPLA 2 -IIA Induces CF Proliferation and Collagen Synthesis
Next, we hypothesized that sPLA 2 -IIA might act as a cytokine-like modulator of cardiac damage in EAM. To test this possibility, we focused on CFs because of its key role during the pathological remodelling of the heart and we examined whether sPLA 2 -IIA could induce the transition of quiescent fibroblasts towards synthetic and proliferative myofibroblasts.
Primary cultures of adult rat CFs were treated with different doses of sPLA 2 -IIA for 24 h. These doses were previously selected according to those detected in human plasma under pathological conditions [19]. Our results revealed that CFs proliferation was stimulated in a dose-dependent manner and reached a maximal change at 1 µg/mL, the dose which was used for all subsequent experiments. AngII (1 µM) and FCS (10 %) were used as references of proliferative agonists in CFs.
Differentiation of CFs towards activated myofibroblasts was assessed by expression of α-SMA and paxillin, as well as by collagen production. As shown in Figure 2B,C, unstimulated CFs exhibited Cells 2020, 9, 396 7 of 20 a low and diffuse α-SMA signal, while sPLA 2 -IIA-treated cells appeared to have high levels of α-SMA in organized filaments spanning the cell. Likewise, paxillin staining of CFs showed high-intensity fluorescence in sPLA 2 -IIA-treated cells, whereas untreated fibroblasts showed diffuse low-intensity staining. These increased expression levels of α-SMA and paxillin were also confirmed by flow cytometer analysis ( Figure S1).

sPLA2-IIA Induces Oxidative Stress in Cardiac Fibroblast
To determine whether sPLA2-IIA induces oxidative stress, which has been reported to be involved in the progression of EAM to DCM, we analyzed the intracellular ROS levels in CF exposed to sPLA2-IIA (1 μg/mL) at different times. As shown in Figure 3A,C, sPLA2-IIA enhanced Quantitative analysis of collagen accumulation on CFs monolayer cultures showed an augmented collagen deposition in sPLA 2 -IIA-treated CFs ( Figure 2D). These results were in accordance with the time-dependent enhanced collagen I protein expression observed in CFs upon sPLA 2 -IIA stimulation (Figure 2Ei). Flow cytometry data also revealed increased levels of collagen I in CFs stimulated for 24 h with sPLA 2 -IIA (1 µg/mL) or AngII (1 µM) (Figure 2Eii). In addition, the expression levels of the profibrotic cytokine transforming growth factor-β (TGFβ), measured by flow cytometric analysis, were increased in CFs treated with sPLA 2 -IIA ( Figure 2F).

sPLA 2 -IIA Induces Oxidative Stress in Cardiac Fibroblast
To determine whether sPLA 2 -IIA induces oxidative stress, which has been reported to be involved in the progression of EAM to DCM, we analyzed the intracellular ROS levels in CF exposed to sPLA 2 -IIA (1 µg/mL) at different times. As shown in Figure 3A,C, sPLA 2 -IIA enhanced intracellular ROS production on CFs in a time-dependent manner (p < 0.001). Representative microphotographs with fluorescence microscopy show the enhancement of H 2 O 2 and O 2 − production ( Figure 3B,D).
Pretreatment with either antioxidant NAC or with the flavoprotein inhibitor DPI normalised ROS values ( Figure 3E). Mitochondrial ROS production, measured by MitoSOX fluorescence using fluorescence microscopy, was also induced upon sPLA 2 -IIA stimulation ( Figure 3F and Figure S2C). However, sPLA 2 -IIA treatment for up to 24 h did not induce a substantial decline in Rd123 fluorescence ( Figure 3G), indicating that mitochondrial membrane potential was unaffected. A similar effect was observed in response to the reference AngII (1 µM), whereas, H 2 O 2 (500 µM) exposure expectedly triggered a dramatic decrease in CFs Rd123 staining compared with untreated control cells (p < 0.05).

sPLA 2 -IIA-Activated Intracellular Signaling Cascades in Cardiac Fibroblast Requires EGFR Transactivation and ProHB-EGF Shedding
Subsequently, we explore some of the signal transduction molecules involved in sPLA 2 -IIAmediated phenotypic and functional changes in CF. As shown in Figure 4A,B, CFs stimulated with 1 µg/mL of sPLA 2 -IIA exhibited a rapid, transient and time-dependent pattern of phosphorylation in Src, ERK1/2, P70S6K and rS6 proteins.
Next, considering that EGFR and its ligands serve as a switchboard for the regulation of multiple cellular processes, we investigated whether sPLA 2 -IIA affected EGFR activity on CFs. As shown in Figure 4C, flow cytometry analysis revealed that EGFR phosphorylation at Tyr1176 increased after the addition of sPLA 2 -IIA and was suppressed by pretreatment with the selective inhibitor of EGFR tyrosine phosphorylation AG1478 (2.5 µM) ( Figure 4D). sPLA 2 -IIA-induced phosphorylation of ERK1/2, P70S6K and rS6 was also inhibited by the pretreatment with AG1478 ( Figure 4E), whereas this inhibitor had no effect on sPLA 2 -IIA-induced phosphorylation of Src ( Figure 4F). Moreover, sPLA 2 -IIA-induced ROS production was also inhibited by the pretreatment with AG1478 ( Figure S2).
Mitochondrial ROS production, measured by MitoSOX fluorescence using fluorescence microscopy, was also induced upon sPLA2-IIA stimulation ( Figures 3F and S2C). However, sPLA2-IIA treatment for up to 24 h did not induce a substantial decline in Rd123 fluorescence ( Figure 3G), indicating that mitochondrial membrane potential was unaffected. A similar effect was observed in response to the reference AngII (1 μM), whereas, H2O2 (500 μM) exposure expectedly triggered a dramatic decrease in CFs Rd123 staining compared with untreated control cells (p < 0.05).  After that, we determined whether EGFR transactivation induced by sPLA 2 -IIA involves shedding of heparin-binding epidermal growth factor (HB-EGF). As illustrated in Figure 5A, flow cytometry analysis of CFs confirmed that the cell membrane-associated pool of HB-EGF decreased in response to sPLA 2 -IIA in a time-dependent manner. CFs pre-treatment for 30 min with GM6001 (50 µM), a broad-spectrum matrix metalloproteinase inhibitor, resulted in complete inhibition of HB-EGF shedding ( Figure 5B). A similar pattern was observed in the presence of TAPI-1 (10 µM), an inhibitor of proteases containing a disintegrin and metalloproteinase (ADAM) domain/tumor necrosis factor-α-converting enzyme (TACE). These results suggest that sPLA 2 -IIA induces the shedding of pro-HB-EGF on CFs through an ADAMs-mediated mechanism. Inhibition of Src kinase with PP2 also prevented sPLA 2 -IIA-induced HB-EGF release, while pre-treatment with the MEK inhibitor and suppressor of the MEK-ERK signaling PD098059 had no effect (data not shown). Likewise, as shown in Figure 5C,D, the phosphorylation of EGFR and/or the signaling mediators ERK, P70S6K and rS6 was abrogated by CFs pretreatment with GM6001 (50 µM) or TAPI-1 (10 µM), as well as with the anti-HB-EGF-neutralizing antibody (10 µg/mL). This further confirms the role of pro-HB-EGF ectodomain in sPLA 2 -II-induced EGFR transactivation and signaling. None of the above-mentioned inhibitors affected phosphorylation of Src kinase (data not shown).
Finally, we also observed that sPLA 2 -IIA-induced ROS increase in CFs was abrogated through the use of the inhibitors GM6001 and TAPI-1, as well as the anti-HB-EGF neutralizing antibody ( Figure S3) On resting cells, the presence of these inhibitors or the neutralizing antibody did not affect constitutive pro-HB-EGF cellular levels (data not shown).

sPLA 2 -IIA-Induced Transactivation of EGFR Regulates Collagen Production and Maturation in CFs
Next, we investigated whether EGFR transactivation was involved in sPLA 2 -IIA-induced collagen production. The results showed that 24 h after sPLA 2 -IIA exposure, CFs increased collagen I expression (up to~5-fold) measured by flow cytometry analysis ( Figure 6A). The mean fluorescence intensity was 59.6 ± 8 in non-treated cells, while it was 307.3 ± 80 (p < 0.05) after sPLA 2 -IIA treatment. This response was completely prevented by the presence of inhibitors that block EGFR transactivation and signaling (AG1478, GM6001, TAPI-1), as well as by inhibitors of the intracellular protein kinase cascades Src-MEK/ERK-mTOR/P70S6K/rS6 (PP2, PD098059, Rapamycin). In contrast, molecules inhibiting oxidative stress (DPI, NAC) did not alter the up-regulated collagen I synthesis. We also measured collagen accumulation by a Sirius Red assay on CFs exposed to sPLA 2 -IIA for 3 days. As shown in Figure 6B, collagen deposition was significantly higher in stimulated cells compared with resting control, and the presence of the above-mentioned kinase inhibitors also abrogated collagen build-up. The antioxidants, NAC and DPI, did not affect the increased collagen expression levels triggered by sPLA 2 -IIA Next, to further investigate the relationship among sPLA 2 -IIA and factors involved in the maturation of collagen fibers, we studied the expression of BMP-1 and LOX in CFs treated with sPLA 2 -IIA for 24 h. The data indicated that sPLA 2 -IIA induces BMP-1 and LOX upregulation, reaching similar expression levels to those observed in response to the reference agonists AngII (Figures 7A and  8A, respectively), supporting that sPLA 2 -IIA not only favours collagen production but also collagen cross-linking. These findings were also verified by examination of the cells under a fluorescence microscope ( Figures 7B and 8B, respectively). In addition, LOX up-regulation was also confirmed by Western blot analysis ( Figure 8C).

Discussion
The involvement of sPLA2-IIA in chronic and acute inflammatory diseases has been well documented, but its precise role in myocardial disorders remains to be determined [14][15][16][17][18]. The present study explored molecular mechanisms regulated by the phospholipase that directly affects cardiac fibroblast phenotype and functions. Our data show that cellular infiltration and interstitial fibrosis in hearts from EAM mice were accompanied by an up-regulation of sPLA2-IIA protein levels. These high sPLA2-IIA levels were also reflected in serum. More interestingly, we observed that CFs, the predominant secretory cells producing ECM components, may be an important target of sPLA2-IIA. In cultures of adult rat CFs, sPLA2-IIA induced cell proliferation and increased As already mentioned for collagen synthesis, sPLA 2 -IIA-induced BMP-1 and LOX expression were also prevented by the EGFR inhibitor AG1478 (Figures 7C and 8D, respectively). Similarly, CFs pretreatment with inhibitors that blocked EGFR transactivation, as well as with inhibitors of the intracellular protein kinase cascades such as PP2, PD098059 and Rapamycin, also abolished the induction of these proteins ( Figure S4A,B,D,E). In contrast, the presence of the ROS inhibitors, NAC or DPI, did not affect their increased expression levels promoted by sPLA 2 -IIA ( Figure S4C,F).
Additionally, LOX enzymatic activity was measured in the conditioned medium of CFs treated with sPLA 2 -IIA for 24 h, in the absence or presence of the above-mentioned inhibitors ( Figure 8E,F). sPLA 2 -IIA increased LOX activity into the cell culture media 5-fold compared to control levels (p < 0.001), which was similar to that observed with 1 µM of AngII, our reference agonist. Pre-treatment with the selective protein-kinase inhibitors abolished this up-regulation. The antioxidant NAC did not affect sPLA 2 -IIA augmented LOX activity in the cell-conditioned medium.

Discussion
The involvement of sPLA 2 -IIA in chronic and acute inflammatory diseases has been well documented, but its precise role in myocardial disorders remains to be determined [14][15][16][17][18]. The present study explored molecular mechanisms regulated by the phospholipase that directly affects cardiac fibroblast phenotype and functions. Our data show that cellular infiltration and interstitial fibrosis in hearts from EAM mice were accompanied by an up-regulation of sPLA 2 -IIA protein levels. These high sPLA 2 -IIA levels were also reflected in serum. More interestingly, we observed that CFs, the predominant secretory cells producing ECM components, may be an important target of sPLA 2 -IIA. In cultures of adult rat CFs, sPLA 2 -IIA induced cell proliferation and increased expression of collagen I, TGFβ, LOX and BMP-1 protein, which are major contributing factors to the development of cardiac fibrosis.
Fibrosis results from ECM synthesis/degradation balance. LOX plays a crucial role in the maintenance of extracellular matrix stability and could participate in cardiac remodeling associated with the pathogenesis of myocardial diseases. Several studies have indicated that BMP-1 activates LOX precursor (pro-LOX) to mature active form, which is responsible for the cross-linking of collagen fibers. This allows for the formation of the mature and insoluble ECM, which is less prone to degradation [12]. LOX activity is required for normal processing of collagen, yet over-activation of LOX is associated with fibrosis. Thus, an increased expression and activity of LOX has been demonstrated in the myocardium of patients with heart failure, as well as with dilated cardiomyopathy, which correlated with increased collagen content and collagen cross-linking [27,28]. Although experimental rodent models of several cardiac disorders have enabled the assessment of the role of LOX activity in the progression of cardiac dysfunction and adverse ECM alterations, further studies are needed to identify accurately injurious stimuli, as well as cellular mechanisms responsible for the increased LOX expression and LOX-dependent damage in the heart [29,30].
In this study, in addition to the enhanced expression of structural (i.e., collagen) and matricellular (i.e., LOX, BMP-1) ECM proteins found on sPLA 2 -IIA-treated CFs, higher LOX activity was detected in the supernatants of sPLA 2 -IIA-treated CFs cultures, which was consistent with the increased collagen deposition observed in the monolayers of stimulated CFs. These findings agree with the hypothesis that sPLA 2 -IIA may act as a cardiac profibrotic factor, potentiating not only ECM production but also its cross-linking and consequently making it more difficult to degrade.
The potential role of sPLA 2 -IIA in fibrotic disorders has been suggested in the past but based mainly on indirect evidence. Mice transfected with the human sPLA 2 -IIA gene have increased collagen levels in atherosclerotic lesions, and spontaneously hypertensive rats treated with a specific sPLA 2 -IIA inhibitor showed a reduction in cardiac fibrosis during the development of hypertension [20,21]. However, to our knowledge, these data are the first to directly demonstrate the regulation of structural and matricellular ECM proteins in adult CFs by the inflammatory protein sPLA 2 -IIA, thus suggesting a pathogenic role in inflammatory cardiac disorders. However, sPLA 2 -IIA might not be the only one in the family, since recent evidence in a model of myocardial ischemia has pointed out that other isoforms, sPLA 2 -IB and sPLA 2 -IIE and its receptor PLA 2 R1, might mediate collagen-dependent biological effects in the infarcted myocardium [31].
Regarding the mechanism involved in sPLA 2 -IIA pro-fibrotic activity, another significant finding of the present study was that this phospholipase enhanced phosphorylation/activation of signal transduction molecules such as MEK/ERK-mTOR/P70S6K/rS6 and production of ROS through transactivation of EGFR.
The EGFR family and its ligands serve as a switchboard for the regulation of multiple cellular processes. EGFR-via transactivation-has the potential to mediate signaling of non-EGFR ligands and thereby serve as a heterologous transducer of cellular signaling. EGFR transactivation in cardiac cells has been observed following stimulation with several vasoactive substances and well-known inducers of cardiomyocyte hypertrophy such as AngII, phenylephrine, endothelin-1 and aldosterone [32][33][34][35]. In CFs, β-adrenergic receptor-dependent changes in cytokine expression are also predominantly mediated through an EGFR-sensitive manner [36]. Moreover, it has recently been shown that EGFR transactivation is also involved in ROS generation and cell apoptosis induced by high concentrations of glucose in rat cardiomyocytes, thus pointing to its participation in diabetes-induced cardiac injury [37]. Here, our data reinforce the role of sPLA 2 -IIA in inflammatory heart disorders where EGFR might serve as a signaling hub, engaging in cross-talk with multiple pathways.
To determine some of the mechanisms that link sPLA 2 -IIA signaling to EGFR transactivation on CFs, studies using a neutralizing antibody for HB-EGF, the metalloproteinase inhibitor Batimastat/GM6001, and the TACE inhibitor TAPI-1, have suggested that processing of the membrane-anchored pro-HB-EGF via matrix metalloproteinases activity (most likely TACE) is required for sPLA 2 -IIA-induced tyrosine phosphorylation of the EGFR. These data are in agreement with a series of studies, suggesting that the ADAM/MMP protease-mediated shedding of membrane-tethered ligands is a key step to activating EGFR and downstream signaling pathways. HB-EGF is known to induce cell activation in various cell types, including cardiac cells via transactivation of EGFR. In CFs and cardiomyocytes, it has been demonstrated that cellular signaling induced by inflammatory agonists and proteases, closely related to cardiac remodeling, is dependent upon HB-EGF shedding and subsequent EGFR activation [33,38,39]. Therefore, further approaches, such as knockdown EGFR or HB-EGF expression, might be considered to be future strategies for validating the results obtained with pharmacological interventions.
Upon ligand binding, EGFR transactivation elicits downstream activation of several signalling cascades involving classic second messengers, protein kinases, non-receptors tyrosine kinases and ROS, among others [40]. In our study, we found that the selective inhibitor of the Src tyrosin kinase, PP2, blocked sPLA 2 -IIA-induced HB-EGF shedding and EGFR phosphorylation, which indicates that Src phosphorylation is an upstream signal required for sPLA2-IIA-induced CF activation. However, we observed that MEK/ERK-mTOR/P70S6K/rS6 are essential intracellular signalling molecules downstream of EGFR because CFs treated with selective inhibitors of these pathways did not modify the HB-EGF membrane levels nor blunt EGFR transactivation. sPLA 2 -IIA-induced ROS generation was also downstream of EGFR phosphorylation and required its transactivation-dependent MMP shedding of EGF-like ligands.
Moreover, we observed that on CFs stimulated with sPLA 2 -IIA, the increased synthesis of LOX, BMP-1 and collagen I proteins requires activation of EGFR after the shedding of HB-EGF from the cell surface, based on the complete abrogation of these changes by a neutralizing antibody specific for HB-EGF or metalloproteinase specific inhibitors. Additionally, active MEK/ERK-mTOR/P70S6K/rS6 signalling in CFs was crucial to orchestrate the profibrotic effects of sPLA 2 -IIA, since pharmacologic inhibition of these pathways significantly decreased the production of ECM compounds in response to sPLA 2 -IIA. However, although several in vitro studies have demonstrated that oxidative stress is involved in the profibrotic actions of agonist such as TGF-β1 or leptin, our data indicate that ROS production was not necessary for the enhanced matrix protein components production triggered by sPLA 2 -IIA on CFs [35,41].
Moreover, the data show that sPLA 2 -IIA not only stimulates ECM production but also favours the transformation of CFs into cardiomyofibroblasts with more capacity for ECM production and proliferation, a normal process that occurs in response to myocardial injury [12,13].
The present study reveals some of the molecular mechanism involved in the phenotypic changes induced by sPLA 2 -IIA in adult CFs, although there are still many missing steps and uncertainties that need to be studied in depth, e.g., membrane molecular effectors acting upstream HB-EGF shedding and connecting signalling events and the biological effect of sPLA 2 -IIA. Moreover, based on existing findings, a direct sPLA 2 -IIA-integrin interaction is an interesting hypothesis to address CFs in future studies [42].
In summary, in this study, we highlight the complexity of the molecular mechanisms involved in myocardial fibrosis, providing data to understand the role of sPLA 2 -IIA in pathological conditions affecting heart function. According to the pivotal role of CFs in cardiac disease and remodelling, we believe that these observations will raise significant clinical interest.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4409/9/2/396/s1, Figure S1: sPLA 2 -IIA modulates cardiac fibroblasts phenotype, Figure S2: Effect of the EGFR inhibitor AG1478 on sPLA 2 -IIA-induced ROS accumulation in cardiac fibroblasts, Figure S3: Effect of inhibitors that block EGFR transactivation on sPLA 2 -IIA-induced ROS increase in cardiac fibroblasts, Figure S4: Effect of selected inhibitors on sPLA 2 -IIA-induced BMP-1 and LOX expression in cardiac fibroblasts. Funding: This work was supported by grants from Plan Estatal I+D+i: MINECO SAF2012-34460 and SAF2016-81063; ISCII PI18/010257729; and CIBERCV. The study was co-funded by Fondo Europeo de Desarrollo Regional (FEDER), a way to make Europe. RM was supported by the Sara Borrell Program from ISCIII; BG and CC were funded by the FPI Program from the Government of Castilla y León (all co-funded by FSE).