FoxO1 mediates TGF-beta1-dependent cardiac myofibroblast differentiation

    FoxO1 mediates TGF-beta1-dependent cardiac myofibroblast differentiation Ra´ul Vivar, Claudio Humeres, Claudia Mu˜noz, P´ıa Boza, Samir Bolivar, Felipe Tapia, Sergio Lavandero, Mario Chiong, Guillermo Diaz-Araya PII: DOI: Reference:

S0167-4889(15)00372-9 doi: 10.1016/j.bbamcr.2015.10.019 BBAMCR 17707

To appear in:

BBA - Molecular Cell Research

Received date: Revised date: Accepted date:

8 August 2015 2 October 2015 26 October 2015

Please cite this article as: Ra´ ul Vivar, Claudio Humeres, Claudia Mu˜ noz, P´ıa Boza, Samir Bolivar, Felipe Tapia, Sergio Lavandero, Mario Chiong, Guillermo Diaz-Araya, FoxO1 mediates TGF-beta1-dependent cardiac myofibroblast differentiation, BBA - Molecular Cell Research (2015), doi: 10.1016/j.bbamcr.2015.10.019

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

PT

BBAMCR-05October1 2015

SC

RI

FoxO1 mediates TGF-beta1-dependent cardiac myofibroblast differentiation

Advanced Center for Chronic Diseases (ACCDiS), Faculty of Chemical & Pharmaceutical

MA

a

NU

Raúl Vivara,b, Claudio Humeresb, Claudia Muñozb, PíaBozab, Samir Bolivarb, Felipe Tapiab, Sergio Lavanderoa,c, Mario Chionga,* , Guillermo Diaz-Arayaa,b,*

Sciences & Faculty of Medicine, University of Chile; Santiago, Chile. b

Department of Pharmacological & Toxicological Chemistry, Faculty of Chemical &

D

Pharmaceutical Sciences & Faculty of Medicine, University of Chile; Santiago, Chile c

TE

Departments of Internal Medicine (Division of Cardiology) and Molecular Biology, University of

AC CE P

Texas Southwestern Medical Center, Dallas, TX, USA.

Short title: FoxO1 is necessary for cardiac fibroblast differentiation

*

Corresponding authors at: Advanced Center for Chronic Diseases (ACCDiS), Faculty of Chemical

& Pharmaceutical Sciences & Faculty of Medicine, University of Chile; 8380492 Santiago, Chile. Tel: +562 29782975; fax: +562 297829192 E-mail address: [email protected] (G Diaz-Araya) or [email protected] (M. Chiong).

ACCEPTED MANUSCRIPT

RI

SC

Cardiac fibroblasts Cardiac myofibroblasts Fetal bovine serum Transforming Growth Factor-β1 α-smooth muscle actin Forkhead box type O Connective tissue growth factor p21-cyclin-dependent kinase inhibitor p15-cyclin-dependent kinase inhibitor

AC CE P

TE

D

MA

NU

CF CMF FBS TGF-β1 α-SMA FoxO CTGF p21(WAF1/CIP1) p15Ink

PT

Abbreviations

ACCEPTED MANUSCRIPT Abstract Cardiac fibroblast differentiation to myofibroblast is a crucial process in the development of

PT

cardiac fibrosis and is tightly dependent on transforming growth factor beta-1 (TGF-β1). The

RI

transcription factor forkhead box O1 (FoxO1) regulates many cell functions, including cell death by apoptosis, proliferation, and differentiation. However, several aspects of this process remain

SC

unclear, including the role of FoxO1 in cardiac fibroblast differentiation and the regulation of

NU

FoxO1 by TGF-β1. Here, we report that TGF-β1 stimulates FoxO1 expression, promoting its dephosphorylation, nuclear localization and transcriptional activity in cultured cardiac fibroblasts.

MA

TGF-β1 also increases differentiation markers such as α-smooth muscle actin, connective tissue growth factor, and pro-collagen I, whereas it decreases cardiac fibroblast proliferation triggered by

D

fetal bovine serum. TGF-β1 also increases levels of p21waf/cip-cycle inhibiting factor protein, a

TE

cytostatic factor promoting cell cycle arrest and cardiac fibroblast differentiation. In addition, TGFβ1 increases cardiac fibroblast contractile capacity as assessed by collagen gel contraction assay.

AC CE P

The effect of TGF-β1 on cardiac fibroblast differentiation was prevented by FoxO1 downregulation and enhanced by FoxO1 overexpression. Thus, our findings reveal that FoxO1 is regulated by TGF-β1 and plays a critical role in cardiac fibroblast differentiation. We propose that FoxO1 is an attractive new target for anti-fibrotic therapy. Keywords: TGF-β1; FoxO1; Cardiac Fibroblast; Myofibroblast, Cell Differentiation

ACCEPTED MANUSCRIPT 1. Introduction

PT

Diabetes mellitus, hypertension, and myocardial infarction promote the development of cardiac fibrosis[1-6]. In these pathological conditions, normally quiescent cardiac fibroblasts (CF)

RI

are activated, differentiating to cardiac myofibroblasts (CMF). The physiological role of CMF has

SC

been associated with tissue healing [7]. However, in pathological conditions CMF secrete extracellular matrix protein (mainly collagen type I), which favors progressive cardiac fibrosis

NU

increasing myocardial stiffness leading to diastolic dysfunction and decreasing myocardial function

MA

[8]. In hypertensive rats, thiazolidinediones, relaxin, and antioxidants reduce the presence of positive alpha-smooth muscle actin (-SMA) cells, which is a marker of CF to CMF differentiation, different to those founded in vascular smooth muscle cells own to blood vessels;

TE

D

this decrease correlates with reduced cardiac tissue fibrosis and improved myocardial function[810]. This observation highlights the importance of CF differentiation into CMF in cardiac fibrosis.

AC CE P

The cytokine TGF-β1 plays a crucial role in the development of fibrosis, by inducing the differentiation of CF into CMF [11, 12], whereas in rats with myocardial infarction, reduced levels of this cytokine are related to decrease α-SMA positive cells and collagen I expression which correlates with ameliorates cardiac remodeling[13, 14]. TGF-β1 is a potential target for anti-fibrotic therapies. However, its multiple pathophysiological actions render its therapeutic useimpractical. TGF-β1 activates two different membrane receptors, TGF-β1 type I and type II, leading to serine phosphorylation of SMAD2/3. Phosphorylated SMADs form a heterotrimeric complex with SMAD4 and translocate into the nucleus to promote specific gene transcription[15-17]. SMADs require heterodimerization with other transcription factors, such as CBP and AP-1[18, 19]. Recently, it has been shown that FoxO family members and the TGF-β1 signaling pathway act together to promote gene expression of many targets, such as inhibitors of cyclin/CDK complex master regulators of cell cycle, such as p21waf/cip, p15ink, and connective tissue growth factor (CTGF) which is a main regulator of extracellular matrix protein[20, 21].

ACCEPTED MANUSCRIPT FoxO1 is a transcription factor involved in apoptosis, oxidative stress, and cell differentiation [22, 23]. FoxO1 is also regulated by post-transcriptional modifications such as

PT

phosphorylation and ubiquitination that control its stability and activity. Phosphorylation of FoxO1 in serine and threonine residues leads to its inactivation and degradation [24]. FoxO1

RI

dephosphorylation by PP2A increases its activity, promoting nuclear translocation, which induces

SC

gene transcription [25], whereas TGF-β1 increases PP2A activity [26]. However, whether TGF-β1 regulates FoxO1 expression or transcriptional activity in CF is unknown.

NU

FoxO1 has shown to be an important cell cycle regulator [27, 28], through p21waf/cip

MA

regulation [20, 21, 29]. In addition, p21waf/cip is also required for CF differentiation into CMF [30]. CTGF is another protein involved in CF differentiation that is transcriptionally regulated by TGF-β1 and FoxO1 [20, 31, 32].

TE

D

Based on this evidence, we hypothesized that TGF-1 regulates CF differentiation through a

AC CE P

mechanism involving FoxO1regulation.

2.Material and methods

2.1. Materials

TGF-β1 and Accutase® were obtained from Millipore. 488 Alexa Fluor® conjugated secondary antibodies, actinomycin D, cycloheximide, MG132, Bradford solution, propidium iodide, and primary antibodies (GAPDH, β-tubulin, α-SMA) were obtained from Sigma-Aldrich. Primary antibodies (FoxO1 and p-FoxO1) were obtained from Cell Signaling Technologies. Primary antibodies (p21waf/cip, lamin A/C, pro-collagen I, and CTGF) were obtained from Abcam. Secondary antibodies conjugated with horseradish peroxidase were obtained from Calbiochem. Collagen rat tail solution, RNAse, Trypan blue, lipofectamine, Opti-MEM®, and siRNAs were obtained from Life Technologies. All adenoviruses used in this work were a kind gift

ACCEPTED MANUSCRIPT from Dr. Joseph A Hill, University of Texas, Southwestern Medical Center (Dallas, Texas). ECL

PT

Advance Western Blotting Detection Kit was obtained from GE Healthcare Europe GmbH.

RI

2.2. Cell culture and treatments

SC

CF were isolated from the left ventricle (LV) of 3-day-old Sprague-Dawley rats. Briefly, LVs were harvested, cut into small pieces (~1-2 mm), and placed in a Petri dish. A total of 10mL

NU

Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (FBS) and antibiotics

MA

were added to Petri dishes. Petri dishes were maintained at 37°C under 5% CO2 in a humidified incubator until reaching 70% confluence. Adherent cells were cultivated. Cells from passage 1 were used for all experiments. Cells were maintained in M199 medium for 18 h. CF were

D

stimulated with TGF-β1 (1-10 µg/mL) for different experimental times (2 to 72 h). For the

TE

inhibition experiments, cells were incubated with actinomycin D (transcription inhibitor, 2 µg/mL),

AC CE P

cycloheximide (translation inhibitor, 1 µM), and MG132 (proteosome inhibitor, 10 nM) for 1 h before TGF-β1 incubation.

2.3. Western blot analysis

After respective treatments, cells were lysed in 50mM Tris, 300 mMNaCl, 1 mM MgCl 2, 0.5 mM EDTA, 0.1mM EGTA, 20% glycerol, 1% NP40, 0.5 mM DTT, and inhibitor cocktail. Lysates were vigorously vortexed for 10 s and centrifuged at 15,000 rpm for 10 min, and total protein content was determined using Bradford assay. Equivalent amounts of protein were subjected to SDS-PAGE. Western blotting was performed by transferring proteins to nitrocellulose membranes and blocking with 10% fat-free milk (w/v) in TBS-tween for 1 h at room temperature. Membranes were probed with the appropriate primary antibody overnight at 4°C and then with peroxidaseconjugated secondary antibody for 2 h at room temperature. Finally, the ECL Advance Western

ACCEPTED MANUSCRIPT Blotting Detection Kit was used for immunodetection. Protein levels were determined by densitometric analysis using Image J (NIH, Bethesda, MD, USA) and normalized to the

PT

corresponding GAPDH level.

SC

RI

2.4. Immunofluorescence assay

CF were fixed in 4% paraformaldehyde solution for 20 min at room temperature and

NU

permeabilized in 0.1% Triton X100 for 10 min at room temperature. Non-specific proteins were

MA

blocked with 3% bovine serum albumin solution for 30 min at room temperature. Cells in coverslips were incubated with the appropriate primary antibody overnight at 4°C and an appropriate fluorophore-conjugated secondary antibody for 2 h at room temperature. Images were

AC CE P

2.5. Subcellular fractionation

TE

D

obtained using a spinning-disk microscope.

Cells were lysed in 10 mM HEPES, 1.5 mM MgCl2, 10 mMKCl, 0.5 mM DTT, and 0.05% NP40 (pH 7.9), kept on ice for 10 min, and centrifuged at 300 rpm for 10 min at 4°C. Supernatants corresponded to cytosol samples. Pellets were resuspended in 5 mM HEPES, 1.5 mM MgCl 2, 0.2 mM EDTA, 0.5 mM DTT, and 26% glycerol (v/v) (pH 7.9). NaCl was added to obtain a 300 mMNaCl final concentration; the mixture was maintained on ice for 30 min and then centrifuged at 24,000 g for 20 min at 4°C. Pellets were discarded, and supernatants corresponded to nuclear extracts. Protein content was analyzed using Bradford assay. The purity of the subcellular fractions was assessed using β-tubulin for the cytosol and lamin A/C for the nucleus.

2.6. Cell proliferation

ACCEPTED MANUSCRIPT CF proliferation was induced with FBS at concentrations ranging from 1% to 10% in M199 medium for 24 h. To evaluate cell proliferation, two techniques were used: i) Trypan blue assay:

PT

Cells were detached from culture dishes with Accutase®. After detachment, 100% FBS was added to ensure cell integrity and inhibit further Accutase® action. Cell number was determined by the

RI

trypan blue exclusion method on a Neubauer chamber. Cell count was performed in quadruplicate.

SC

ii) Crystal violet assay: Cells were stained with 10% crystal violet in ethanol for 20 min at room temperature. Stained cells were then lysed with 50% ethanol. Crystal violet absorbance was

NU

measured at 590 nM on a microplate. A plot of crystal violet absorbance versus cell number (from

MA

10,000 to 200,000 cells) was used to determine cell number in each sample.

D

2.7. Cell cycle analysis

TE

Trypsin-EDTA 1x-detached cells were centrifuged at 2000 rpm for 5 min at 4°C,

AC CE P

resuspended in cold methanol, and stored at -20°C overnight. Cell were centrifuged at 2000 rpm for 5 min at 4°C, and the pellet was resuspended in PBS solution containing RNAse (1 µg/mL) and kept at room temperaturefor 2 h. Cells were incubated with propidium iodide solution (10 µg/mL) for 1 min at room temperature. Finally, cell cycle was analyzed in a flow cytometer (Becton Dickinson). Data were expressed as % of cells in G0/G1 and S/G2/M.

2.8. FoxO1 and FoxO3a silencing

CF were seeded in 35mm dishes at 80% confluence and serum-deprived overnight. Lipofectamine + Opti-MEM® (100 L) and specific siRNA + Opti-MEM® (100 L) were prepared, incubated for 15 min at room temperature, mixed, and then incubated for an additional 15 min. 800 µL of Opti-MEM was added and transferred to the cells. After incubation for 16 h at 37°C, the medium was changed to M199 for an additional 24 h. Scramble siRNA (Silencer®

ACCEPTED MANUSCRIPT Negative Control) was used as a control. The sequences used were: FoxO1: 5´-> 3´ (sense) GCACCGACUUUAUGAGCAAtt; (antisense) UUGCUCAUAAAGUCGGUGCtg. FoxO3a: 5´-

PT

>3´(sense) AGCUCUUGGUGGAUCAUCtt; (antisense) UGAUGAUCCACCAAGAGCUct.

SC

RI

2.9. Adenoviral transduction

CF were plated at 70% confluence and serum-deprived overnight. Three different adenoviruses

NU

were used, wild-type FoxO1 (Ad-FoxO1-wt), constitutively active FoxO1 (Ad-FoxO1-ca), and

MA

adenovirus green fluorescent protein (Ad-GFP) as a control. Adenoviruses were added at 50 and 100 multiplicity of infection (MOI) and incubated for 24 h.MOI 100 was chosen to realize our experiments because it showed no loss of cell viability (measured by cell count and amount of

D

protein) and this concentration induced FoxO1 overexpression greater than MOI 50.

AC CE P

TE

2.10. Collagen gel contraction assay

Twenty-four well dishes were coated with 3% BSA solution overnight. In each well, 400 µL of collagen gel solution (1 mg/mL in PBS) and 100 µL of cell suspension (50,000 cells per well) were added, gently homogenized, and maintained at 37°C for 2 h. Gels were detached from the edge of the well with a sterile pipette tip, and 500 µL of M199 medium was added. Gel contraction was determined after 48 h and expressed as the ratio between initial and final area of collagen gel.

2.11. Cell adhesion assay

After different assays, the CF weretrypsinized, counted and seeded (1x105 cell/well) into 12well plate.The plate was maintained at 37°C for 1 h. Then the adherent cells were then fixed with 4% paraformaldehyde for 15 min at room temperature. Fixed CF were washed with PBS and stained with crystal violet (0.5% w/v crystal violet in 20% ethanol) for 10 min. Plates were washed

ACCEPTED MANUSCRIPT with water and allowed to dry completely. Finally, 2% SDS was added for 30 min at room

PT

temperature, to solubilize the fixed and stained cells and the absorbance was read at 550μm.

RI

2.12. Statistical analysis

SC

Data are presented as mean ± SEM from at least four independent experiments. Statistical analysis was performed using one-way ANOVA and Tukey's test for multiple comparisons with

AC CE P

TE

D

MA

NU

GraphPad Prism 5.0 software. p<0.05 was considered statistically significant.

ACCEPTED MANUSCRIPT 3. Results

PT

3.1. TGF-β1 increases FoxO1 protein levels

RI

Treating CF with TGF-β1 for 24 h increased FoxO1and -SMA protein levels in a dose-

SC

dependent manner, with significant effects from 5 ng/mL (Fig. 1A). Furthermore, TGF-β1 also increased FoxO1 and -SMA protein levels in a time-dependent manner, with significant

NU

effectsfrom 24 h (Fig. 1B). Similar results were obtained with immunofluorescence, thus the

MA

treatment with TGF-β1 for 48 h increased FoxO1and in addition,the increased of levels and cytoskeleton incorporation of α-SMA (Fig. 1C). To assess the mechanism involved in the increased

D

FoxO1 levels induced by TGF-1, CF were pretreated with actinomycin D (gene transcription

TE

inhibitor), cycloheximide (protein translation inhibitor), and MG132 (proteosome inhibitor). Actinomycin D and cycloheximide abolished the TGF-1-induced FoxO1 protein level increase,

AC CE P

whereas MG132 had no effect (Fig. 1D). These results show that TGF-1 increases FoxO1 gene transcription and translation.

3.2. TGF-β1 increases FoxO1 activity

Phosphorylation at serine 256 promotes FoxO1 inactivation and degradation. TGF-β1 induced FoxO1 dephosphorylation in a time-dependent manner (Fig. 2A). In parallel to FoxO1 dephosphorylation, TGF-1 induced an increase in FoxO1 nuclear localization as assessed by immunofluorescence (Fig. 2B). This result was corroborated by subcellular fractionation. TGF-1 treatment increased FoxO1 levels in the nucleus while decreasing its content in the cytosol (Fig. 2C). In order to verify that FoxO1 dephosphorylation and nuclear translocation were associated with increased transcriptional activity, two FoxO1-regulated genes, antioxidant enzyme SOD2 and

ACCEPTED MANUSCRIPT cell cycle regulator p21waf/cip, were analyzed. TGF-β1 increased both SOD2 (Fig. 2D) and p21waf/cip (Fig. 2E) protein levels in a time-dependent manner. Taken together, these results

PT

suggest that TGF-β1 promotes FoxO1 transcriptional activity.

RI

3.3. FoxO1 is involved in the TGF-1-induced differentiation of cardiac fibroblasts into

SC

myofibroblasts

NU

TGF-β1 promotes CF differentiation. TGF-1 induced differentiation of CF into CMF, as

MA

visualized by increases in total protein content, α-SMA, pro-collagen I, CTGF, p-SMAD3, YB-1, collagen gel contractile ability, and cell adhesion (Suppl.Fig. 1-2). FoxO1's participation in TGF-β1-induced CF differentiation was evaluated using FoxO1

TE

D

siRNA (si-FoxO1). si-FoxO1 reduced FoxO1 but not FoxO3 protein levels, while siRNA for FoxO3 reduced FoxO3 but not FoxO1 protein levels (Suppl.Fig. 3). FoxO1 knockdown prevented

AC CE P

the TGF-1-dependent increase in FoxO1 protein levels. This effect was specific to si-FoxO1, because siRNA for FoxO3 did not reproduce the effect (Fig. 3A). Moreover, siRNA for FoxO1 also abolished the TGF-1-dependent increases in total protein content, α-SMA, pro-collagen I, CTGF protein levels and assembly of -SMA on stress fibers(Fig. 3B-F). Furthermore, FoxO1 knockdown prevented TGF-β1-dependent collagen gel contraction (Fig. 3G). On the contrary, the cell adhesion, SMAD3 activity, and YB-1 protein level increases promoted by TGF-β1 were not affected by FoxO1 silencing (Suppl.Fig. 4). These results suggest that FoxO1is involved in TGF1-dependent differentiation of CF into CMF.

3.4. FoxO1 is involved in thecytostatic effects of TGF-β1 in cardiac fibroblasts

The cytostatic effect of TGF-β1 was evidenced by the prevention of FBS-induced CF

ACCEPTED MANUSCRIPT proliferation and cell cycle arrest (Suppl.Fig. 5-6A). FoxO1 knockdown abolished the TGF-β1dependent inhibition of CF proliferation induced by FBS (Fig. 4A and Suppl.Fig. 6B). The same

PT

effect was observed when cell cycle was analyzed. FBS decreased the number of cells in G0/G1 while increasing the number of cells in S/G2/M. TGF-1 completely abolished the FBS-induced

RI

effects on cell cycle. However, pretreatment with siRNA for FoxO1 completely prevented the

SC

effects of TGF-1 (Fig. 4B) (the representative histograms of this assays is showed in Suppl. Fig. 7A). These effects could be associated with the control of cyclins, because FoxO1 knockdown

NU

abolished the TGF-1-dependent increase in p21waf/cip(Fig. 4C). These results show that FoxO1

MA

is involved in the cytostatic effect of TGF-1 on CF, probably through the control of p21waf/cip.

TE

D

3.5. FoxO1 overexpression enhances the effects of TGF-β1 in cardiac fibroblasts

We demonstrated that TGF-β1 controls FoxO1 expression and that FoxO1 knockdown

AC CE P

prevented the effects of TGF-β1 in CF. In order to verify the participation of FoxO1 in the TGFβ1-dependent differentiation of CF into CMF, we overexpressed a wild-type (Ad-FoxO1-wt) and a constitutive active form (Ad-FoxO1-ca) of FoxO1 with adenoviruses (at MOI 100 we did not observe effects of virus on cell viability). Ad-GFP was used as a transduction control. CF transduced for 24 h with both Ad-FoxO1-wt and Ad-FoxO1-ca showed increased FoxO1 protein levels (Suppl.Fig. 8). The increases in total protein content, α-SMA, pro-collagen I,CTGF protein levels and assembly of -SMA on stress fiber induced by TGF-β1 were potentiated by both FoxO1-wt and FoxO1-ca overexpression (Figs. 5A-E). As expected, both FoxO1-wt and FoxO1-ca enhanced TGF-β1-induced collagen gel contraction (Fig. 5F). On the contrary, the CF adhesion, SMAD3 activity, and YB-1 protein level increases induced by TGF-β1 were not affected by FoxO1 overexpression (Suppl.Fig. 4). These results show that FoxO1 is involved in differentiation of CF to CMF.

ACCEPTED MANUSCRIPT

PT

3.6. FoxO1 overexpression enhances the cytostatic effects of TGF-β1 in cardiac fibroblasts

TGF-β1 inhibited the CF proliferation induced by FBS. FoxO1-wt and FoxO1-ca

RI

overexpression enhanced the cytostatic effect of TGF-β1 (Fig. 6A and Suppl.Fig. 6C). The same

SC

result was observed when cell cycle was assessed. FBS decreased the CF population in G0/G1 while increasing the population in S/G2/M. FoxO1-wt and FoxO1-ca overexpression enhanced the

NU

effect of TGF-β1 as indicated by a further increase in CF population in G0/G1 and a further

MA

decrease in S/G2/M (Fig. 6B) (the representative histograms of this assays is showed in Suppl. Fig. 7B). FoxO1 promotes the expression of cell cycle regulator p21waf/cip, which induces cell cycle arrest and decreases cell proliferation. TGF-β1 increased p21waf/cip protein levels at 48 h, whereas

D

FoxO1-wt and FoxO1-ca overexpression further increased p21waf/cip protein levels (Fig. 6C).

TE

Taken together, these results show that FoxO1 overexpression enhances the cytostatic effects of

AC CE P

TGF-β1.

4. Discussion

The data presented in this paper show that FoxO1 is a key regulator of the effects of TGF-β1 on differentiation of CF into CMF. FoxO1 regulates the expression of several differentiation markers (α-SMA, pro-collagen I, CTGF, and p21waf/cip) and controls both CF proliferation and collagen gel contraction. Our results show that TGF-β1 regulates FoxO1 expression. Similarly, TGF-β1 increases FoxO1 expression in primary chondrocyte cultures, which correlates with fibrosis and cellular aging[33]. Moreover, FoxO1 induces increased TGF-β1 expression in skin wound healing [34]. However, regulation of FoxO1 expression by TGF-β1 seems to be cell type-dependent; in mesenchymal cells such as chondrocytes or fibroblasts, or epithelial cells such as keratinocytes

ACCEPTED MANUSCRIPT [20], TGF-β1 induces FoxO1 expression, whereas in malignant cells, TGF-β1 inhibits FoxO1 expression[35].

PT

TGF-β1 regulates FoxO1 expression in CF, demonstrated using actinomycin D and cycloheximide. In rat hepatocytes, fasting activates the histone acetyltransferase p300 to induce

RI

FoxO1 expression. p300 recognizes a specific promoter site in FoxO1 and promotes its gene

SC

transcription and protein translation[36, 37]. Interestingly, TGF-β1 activates and increases the expression of p300, which is associated with collageninduction[38, 39], suggesting that TGF-β1

NU

could promote FoxO1 transcription by activating p300.

MA

FoxO1 undergoes post-translational modifications that regulate its nuclear localization, where it exerts its transcriptional activity[40]. FoxO1 phosphorylation controls its transcriptional activity and stability[41]. In fact, AKT activation induced by insulin promotes FoxO1

D

phosphorylation at Ser256, which favors its cytosolic localization, decreases its transcriptional

TE

activity, and promotes FoxO1 degradation[42]. In contrast, oxidative stress, fasting, and insulin

AC CE P

resistance decrease FoxO1 phosphorylation increasing FoxO1 nuclear localization and transcriptional activity[43-45]. Our results demonstrate that in CF, TGF-β1 decreases FoxO1 phosphorylation and increases FoxO1 nuclear localization. The phosphatases involved in FoxO1 dephosphorylation are yet to be conclusively identified. However, PP2A appears to be the main phosphatase involved in this process[25]. Hepatocyte-specific ablation of PP2A attenuates progression of liver fibrosis via TGF-β1/SMAD signaling [26], which indicates a possible relationship between TGF-β1 and PP2A. This evidence suggests that TGF-β1 activates PP2A to induce FoxO1 dephosphorylation, promoting expression of SOD2 and p21waf/cip, two specific FoxO1 gene targets. Furthermore, the p21waf/cip promoter has several recognition sequences, including FoxO (FHBE) and SMAD (SBE) element response[46], suggesting that both proteins are required for p21waf/cip expression. In fact, our results indicate that FoxO1 down-regulation decreased p21waf/cip expression induced by TGF-β1. These data suggest that TGF-β1 induces FoxO1 activity in CF.

ACCEPTED MANUSCRIPT CF differentiation induced by TGF-β1 is related to the expression of many proteins associated with fibrosis, such as collagen I (the main MEC protein)[47], CTFG (a major TGF-β1

PT

signaling effector)[48], and α-SMA (the main CMF marker)[49]. In skin fibroblasts, ultraviolet light induces FoxO1 activation and increases collagen I, whereas FoxO1 down-regulation decreases

RI

this effect[47]. Furthermore, in HUVEC cells, FoxO1 is necessary for CTGF expression[48],

SC

whereas in primary keratinocytes, FoxO1 is necessary for TGF-β1-induced CTGF expression [20]. There is no evidence indicating that α-SMA is regulated by FoxO1. Our results show that FoxO1

NU

up-regulation enhanced the effects of TGF-β1 on pro-collagen I, CTGF, and α-SMA expression,

MA

whereas FoxO1 knockdown inhibited the effects of TGF-β1 on differentiation markers, suggesting that FoxO1 is necessary for the CF differentiation induced by TGF-β1. The mechanism by which FoxO1 regulates the effects of TGF-β1 in CF is unknown.

D

SMAD3 activity and YB-1 protein levels are regulated by TGF-β1; both proteins are important in

TE

TGF-β1-induced differentiation in the epithelial-mesenchymal transition [49-51]. However, our

AC CE P

results indicate that although TGF-β1 increased both SMAD3 activity and YB-1 expression, these effects were not controlled by FoxO1. These data suggest that FoxO1 likely regulates the effects of TGF-β1 by a different mechanism. SMAD7 and c-Ski are other important proteins that regulate cardiac fibroblast differentiation induced by TGF-β1[52, 53]. It would be interesting to analyze whether FoxO1 regulates these proteins and whether such regulation could control the effects of TGF-β1 in CF. Cell differentiation generally leads to decreased cell proliferative capacity, as is the case for cardiomyocytes[54]. The evidence regarding the proliferative effects of TGF-β1 on CF differentiation is controversial. Drissenet al., showed that TGF-β1 triggers CF differentiation and decreased cell proliferation[55], whereas other authors indicated that TGF-promotes proliferation and differentiation of CF[56]. Moreover, our results show that TGF-β1 inhibited CF proliferation induced by FBS. However, in human kidney fibroblasts, TGF-β1 induced cell proliferation through a FGF2-dependent mechanism[57]. These data suggest that the effects of TGF-β1 on fibroblast

ACCEPTED MANUSCRIPT proliferation depend on both organ and species. Our results indicate that the antiproliferative effect of TGF-β1 was enhanced by FoxO1 overexpression and decreased by FoxO1 knockdown,

PT

suggesting that FoxO1 is essential to the decrease in TGF-β1 induced CF proliferation. FoxO family proteins induce the expression of several proteins involved in cell cycle control, such as

RI

p15ink, p27kip, and p21waf/cip. These proteins are cyclin/CDK complex inhibitors that induce cell

SC

cycle arrest and decrease cell proliferation[58]. In mesenchymal cells, dexamethasone via the glucocorticoid receptor inhibits cell proliferation through increased p21waf/cip expression[59]. In

NU

addition, p21waf/cip is a key element in the CF differentiation to CMF induced by TGF-β1[30].

β1 and inhibition of CF proliferation.

MA

Our data suggest that FoxO1 is important in the increased p21waf/cip expression induced by TGF-

An important function of the CF is its ability to contract the extracellular matrix (ECM),

D

which plays a role in healing damaged tissue. TGF-β1 enhances the contractile ability of CF [60].

TE

In addition, TGF-β1 is important in cardiac tissue healing after deleterious events such as

AC CE P

myocardial infarction [61, 62]. Our results show that FoxO1 in necessary for the collagen contraction induced by TGF-β1. There is no direct evidence linking FoxO1 to the contractile ability of CF or to their role in cardiac healing. However, it has been reported that FoxO1 promotes skin wound healingin hyperglycemic rats, while diabetic rats exhibit decreased FoxO1 activity[40, 63]. Our data would imply that TGF-β1 could promote cardiac tissue healing through FoxO1 regulation. Finally, abundant data show that many molecules with demonstrable anti-fibrotic effects increase cAMP/PKA activity, including prostacyclin and natriuretic peptides [64, 65]. In CF, cAMP elevations induced by isoproterenol (ISO) and forskolin (FSK) inhibit TGF-β1-induced collagen I and α-SMA expression[27]. Furthermore, the cAMP/PKA signaling pathway promotes FoxO1 phosphorylation and induces its degradation and inactivation[66, 67]. This evidence suggests that ISO and FSK increase the cAMP-inhibiting CF differentiation induced by TGF-β1, by a mechanism related to PKA activation and FoxO1 degradation. Taken together, the data from the literature and our experimental results suggest that TGF-β1 regulates FoxO1 activity and

ACCEPTED MANUSCRIPT expression, which is crucial to the process of TGF-β1-induced CF differentiation to CMF.

PT

Limitations While most heart diseases occur in adults, we used neonatal rat CF, only as a model to explain our

RI

findings to explain the mechanism why Foxo1 regulates CF to CMF differentiation. In our

SC

laboratory, there is considerable evidence that the phenomena that occur in the neonate can be extrapolated to adults, just change the magnitude of the effects.

NU

Finally, the main limitation of this study was that our studies were conducted in vitro.

MA

Undoubtedly, our in vitro findings would have had greater significance if they had conducted similar assays in vivo; however, in these assays we cannot determine the participation of FoxO1 in

TE

5. Conclusion

D

the differentiation of CF to CMF.

AC CE P

TGF-β1 increases FoxO1 transcriptional activity and expression. FoxO1 is required for the antiproliferativeeffects of TGF-β1 and for TGF-β1-induced promotion of CF differentiation, favoring the cellular function associated to cardiac fibrosis.

Acknowledgements This work was funded by ComisionNacional de Ciencia y Tecnologia (CONICYT), Chile: FONDAP 15130011 (SL, MC, GDA), FONDECYT 1130300 (GDA), FONDECYT 3130657 (RV).

References. [1] L. Yang, J.Y. Gao, J. Ma, X. Xu, Q. Wang, L. Xiong, J. Yang, J. Ren, Cardiac-specific overexpression of metallothionein attenuates myocardial remodeling and contractile dysfunction in

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

l-NAME-induced experimental hypertension: Role of autophagy regulation, Toxicology letters, 237 (2015) 121-132. [2] S.J. Holditch, C.A. Schreiber, R. Nini, J.M. Tonne, K.W. Peng, A. Geurts, H.J. Jacob, J.C. Burnett, A. Cataliotti, Y. Ikeda, B-Type Natriuretic Peptide Deletion Leads to Progressive Hypertension, Associated Organ Damage, and Reduced Survival: Novel Model for Human Hypertension, Hypertension, 66 (2015) 199-210. [3] D. Zheng, S. Dong, T. Li, F. Yang, X. Yu, J. Wu, X. Zhong, Y. Zhao, L. Wang, C. Xu, F. Lu, W. Zhang, Exogenous Hydrogen Sulfide Attenuates Cardiac Fibrosis Through Reactive Oxygen Species Signal Pathways in Experimental Diabetes Mellitus Models, Cellular physiology and biochemistry : international journal of experimental cellular physiology, biochemistry, and pharmacology, 36 (2015) 917-929. [4] W. Qiao, C. Wang, B. Chen, F. Zhang, Y. Liu, Q. Lu, H. Guo, C. Yan, H. Sun, G. Hu, X. Yin, Ibuprofen attenuates cardiac fibrosis in streptozotocin-induced diabetic rats, Cardiology, 131 (2015) 97-106. [5] C.H. Ju, X.P. Wang, C.Y. Gao, S.X. Zhang, X.H. Ma, C. Liu, Blockade of KCa3.1 Attenuates Left Ventricular Remodeling after Experimental Myocardial Infarction, Cellular physiology and biochemistry : international journal of experimental cellular physiology, biochemistry, and pharmacology, 36 (2015) 1305-1315. [6] F. See, W. Thomas, K. Way, A. Tzanidis, A. Kompa, D. Lewis, S. Itescu, H. Krum, p38 mitogen-activated protein kinase inhibition improves cardiac function and attenuates left ventricular remodeling following myocardial infarction in the rat, Journal of the American College of Cardiology, 44 (2004) 1679-1689. [7] V.V. Petrov, J.F. van Pelt, J.R. Vermeesch, V.J. Van Duppen, K. Vekemans, R.H. Fagard, P.J. Lijnen, TGF-beta1-induced cardiac myofibroblasts are nonproliferating functional cells carrying DNA damages, Experimental cell research, 314 (2008) 1480-1494. [8] A. Leask, Getting to the heart of the matter: new insights into cardiac fibrosis, Circulation research, 116 (2015) 1269-1276. [9] T. Shinzato, Y. Ohya, M. Nakamoto, A. Ishida, S. Takishita, Beneficial effects of pioglitazone on left ventricular hypertrophy in genetically hypertensive rats, Hypertension research : official journal of the Japanese Society of Hypertension, 30 (2007) 863-873. [10] W. Zhao, T. Zhao, Y. Chen, R.A. Ahokas, Y. Sun, Oxidative stress mediates cardiac fibrosis by enhancing transforming growth factor-beta1 in hypertensive rats, Molecular and cellular biochemistry, 317 (2008) 43-50. [11] T. Chen, J. Li, J. Liu, N. Li, S. Wang, H. Liu, M. Zeng, Y. Zhang, P. Bu, Activation of SIRT3 by resveratrol ameliorates cardiac fibrosis and improves cardiac function via the TGF-beta/Smad3 pathway, American journal of physiology. Heart and circulatory physiology, 308 (2015) H424-434. [12] H. Lei, D. Wu, J.Y. Wang, L. Li, C.L. Zhang, H. Feng, F.Y. Fu, L.L. Wu, C1q/tumor necrosis factor-related protein-6 attenuates post-infarct cardiac fibrosis by targeting RhoA/MRTF-A pathway and inhibiting myofibroblast differentiation, Basic research in cardiology, 110 (2015) 35. [13] Y. Lu, J.J. Liu, X.Y. Bi, X.J. Yu, S.S. Kong, F.F. Qin, J. Zhou, W.J. Zang, Pyridostigmine ameliorates cardiac remodeling induced by myocardial infarction via inhibition of the transforming growth factor-beta1/TGF-beta1-activated kinase pathway, Journal of cardiovascular pharmacology, 63 (2014) 412-420. [14] Q. He, W. Zhou, C. Xiong, G. Tan, M. Chen, Lycopene attenuates inflammation and apoptosis in post-myocardial infarction remodeling by inhibiting the nuclear factor-kappaB signaling pathway, Molecular medicine reports, 11 (2015) 374-378. [15] F. Jiang, G.S. Liu, G.J. Dusting, E.C. Chan, NADPH oxidase-dependent redox signaling in TGF-beta-mediated fibrotic responses, Redox biology, 2 (2014) 267-272. [16] J. MacLean, K.B. Pasumarthi, Signaling mechanisms regulating fibroblast activation, phenoconversion and fibrosis in the heart, Indian journal of biochemistry & biophysics, 51 (2014) 476-482.

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

[17] M.R. Zeglinski, M. Hnatowich, D.S. Jassal, I.M. Dixon, SnoN as a novel negative regulator of TGF-beta/Smad signaling: a target for tailoring organ fibrosis, American journal of physiology. Heart and circulatory physiology, 308 (2015) H75-82. [18] G. Chen, T. Wang, L. Uttarwar, R. vanKrieken, R. Li, X. Chen, B. Gao, A. Ghayur, P. Margetts, J.C. Krepinsky, SREBP-1 is a novel mediator of TGFbeta1 signaling in mesangial cells, Journal of molecular cell biology, 6 (2014) 516-530. [19] R. Oleggini, N. Gastaldo, A. Di Donato, Regulation of elastin promoter by lysyl oxidase and growth factors: cross control of lysyl oxidase on TGF-beta1 effects, Matrix biology : journal of the International Society for Matrix Biology, 26 (2007) 494-505. [20] R.R. Gomis, C. Alarcon, W. He, Q. Wang, J. Seoane, A. Lash, J. Massague, A FoxO-Smad synexpression group in human keratinocytes, Proceedings of the National Academy of Sciences of the United States of America, 103 (2006) 12747-12752. [21] J. Seoane, H.V. Le, L. Shen, S.A. Anderson, J. Massague, Integration of Smad and forkhead pathways in the control of neuroepithelial and glioblastoma cell proliferation, Cell, 117 (2004) 211223. [22] C.B. Chiribau, L. Cheng, I.C. Cucoranu, Y.S. Yu, R.E. Clempus, D. Sorescu, FOXO3A regulates peroxiredoxin III expression in human cardiac fibroblasts, The Journal of biological chemistry, 283 (2008) 8211-8217. [23] S.M. Ronnebaum, C. Patterson, The FoxO family in cardiac function and dysfunction, Annual review of physiology, 72 (2010) 81-94. [24] H. Matsuzaki, H. Daitoku, M. Hatta, K. Tanaka, A. Fukamizu, Insulin-induced phosphorylation of FKHR (Foxo1) targets to proteasomal degradation, Proceedings of the National Academy of Sciences of the United States of America, 100 (2003) 11285-11290. [25] X. Duan, Z. Kong, Y. Liu, Z. Zeng, S. Li, W. Wu, W. Ji, B. Yang, Z. Zhao, G. Zeng, betaArrestin2 Contributes to Cell Viability and Proliferation via the Down-Regulation of FOXO1 in Castration-Resistant Prostate Cancer, Journal of cellular physiology, 230 (2015) 2371-2381. [26] N. Lu, Y. Liu, A. Tang, L. Chen, D. Miao, X. Yuan, Hepatocyte-specific ablation of PP2A catalytic subunit alpha attenuates liver fibrosis progression via TGF-beta1/Smad signaling, BioMed research international, 2015 (2015) 794862. [27] D.F. Han, J.X. Zhang, W.J. Wei, T. Tao, Q. Hu, Y.Y. Wang, X.F. Wang, N. Liu, Y.P. You, Fenofibrate induces G0/G 1 phase arrest by modulating the PPARalpha/FoxO1/p27(kip) pathway in human glioblastoma cells, Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine, 36 (2015) 3823-3829. [28] F. Liu, X.J. Ma, Q.Z. Wang, Y.Y. Zhao, L.N. Wu, G.J. Qin, The effect of FoxO1 on the proliferation of rat mesangial cells under high glucose conditions, Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association European Renal Association, 29 (2014) 1879-1887. [29] J. Palazuelos, M. Klingener, A. Aguirre, TGFbeta signaling regulates the timing of CNS myelination by modulating oligodendrocyte progenitor cell cycle exit through SMAD3/4/FoxO1/Sp1, The Journal of neuroscience : the official journal of the Society for Neuroscience, 34 (2014) 7917-7930. [30] S. Roy, S. Khanna, T. Rink, J. Radtke, W.T. Williams, S. Biswas, R. Schnitt, A.R. Strauch, C.K. Sen, P21waf1/cip1/sdi1 as a central regulator of inducible smooth muscle actin expression and differentiation of cardiac fibroblasts to myofibroblasts, Molecular biology of the cell, 18 (2007) 4837-4846. [31] F. Accornero, J.H. van Berlo, R.N. Correll, J.W. Elrod, M.A. Sargent, A. York, J.E. Rabinowitz, A. Leask, J.D. Molkentin, Genetic Analysis of Connective Tissue Growth Factor as an Effector of Transforming Growth Factor beta Signaling and Cardiac Remodeling, Molecular and cellular biology, 35 (2015) 2154-2164. [32] J.L. Guo, Y. Yu, Y.Y. Jia, Y.Z. Ma, B.Y. Zhang, P.Q. Liu, S.R. Chen, J.M. Jiang, Transient receptor potential melastatin 7 (TRPM7) contributes to H2O2-induced cardiac fibrosis via

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

mediating Ca(2+) influx and extracellular signal-regulated kinase 1/2 (ERK1/2) activation in cardiac fibroblasts, Journal of pharmacological sciences, 125 (2014) 184-192. [33] Y. Akasaki, A. Hasegawa, M. Saito, H. Asahara, Y. Iwamoto, M.K. Lotz, Dysregulated FOXO transcription factors in articular cartilage in aging and osteoarthritis, Osteoarthritis and cartilage / OARS, Osteoarthritis Research Society, 22 (2014) 162-170. [34] B. Ponugoti, F. Xu, C. Zhang, C. Tian, S. Pacios, D.T. Graves, FOXO1 promotes wound healing through the up-regulation of TGF-beta1 and prevention of oxidative stress, The Journal of cell biology, 203 (2013) 327-343. [35] K. Miyazono, Tumour promoting functions of TGF-beta in CML-initiating cells, Journal of biochemistry, 152 (2012) 383-385. [36] V. Perrot, M.M. Rechler, The coactivator p300 directly acetylates the forkhead transcription factor Foxo1 and stimulates Foxo1-induced transcription, Molecular endocrinology, 19 (2005) 2283-2298. [37] A.R. Wondisford, L. Xiong, E. Chang, S. Meng, D.J. Meyers, M. Li, P.A. Cole, L. He, Control of Foxo1 gene expression by co-activator P300, The Journal of biological chemistry, 289 (2014) 4326-4333. [38] A. Bugyei-Twum, A. Advani, S.L. Advani, Y. Zhang, K. Thai, D.J. Kelly, K.A. Connelly, High glucose induces Smad activation via the transcriptional coregulator p300 and contributes to cardiac fibrosis and hypertrophy, Cardiovascular diabetology, 13 (2014) 89. [39] J. Ni, Y. Shen, Z. Wang, D.C. Shao, J. Liu, Y.L. Kong, L.J. Fu, L. Zhou, H. Xue, Y. Huang, W. Zhang, C. Yu, L.M. Lu, P300-dependent STAT3 acetylation is necessary for angiotensin IIinduced pro-fibrotic responses in renal tubular epithelial cells, Acta pharmacologica Sinica, 35 (2014) 1157-1166. [40] C. Zhang, B. Ponugoti, C. Tian, F. Xu, R. Tarapore, A. Batres, S. Alsadun, J. Lim, G. Dong, D.T. Graves, FOXO1 differentially regulates both normal and diabetic wound healing, The Journal of cell biology, 209 (2015) 289-303. [41] A.O. Scheimann, S.K. Durham, A. Suwanichkul, M.B. Snuggs, D.R. Powell, Role of three FKHR phosphorylation sites in insulin inhibition of FKHR action in hepatocytes, Hormone and metabolic research = Hormon- und Stoffwechselforschung = Hormones et metabolisme, 33 (2001) 631-638. [42] D.V. Skarra, V.G. Thackray, FOXO1 is regulated by insulin and IGF1 in pituitary gonadotropes, Molecular and cellular endocrinology, 405 (2015) 14-24. [43] J.A. Hall, M. Tabata, J.T. Rodgers, P. Puigserver, USP7 attenuates hepatic gluconeogenesis through modulation of FoxO1 gene promoter occupancy, Molecular endocrinology, 28 (2014) 912924. [44] Z.Q. Liu, M. Shen, W.J. Wu, B.J. Li, Q.N. Weng, M. Li, H.L. Liu, Expression of PUMA in Follicular Granulosa Cells Regulated by FoxO1 Activation During Oxidative Stress, Reproductive sciences, 22 (2015) 696-705. [45] Y. Qi, Q. Zhu, K. Zhang, C. Thomas, Y. Wu, R. Kumar, K.M. Baker, Z. Xu, S. Chen, S. Guo, Activation of Foxo1 by insulin resistance promotes cardiac dysfunction and beta-myosin heavy chain gene expression, Circulation. Heart failure, 8 (2015) 198-208. [46] K.C. Arden, FoxO: linking new signaling pathways, Molecular cell, 14 (2004) 416-418. [47] H. Tanaka, Y. Murakami, I. Ishii, S. Nakata, Involvement of a forkhead transcription factor, FOXO1A, in UV-induced changes of collagen metabolism, The journal of investigative dermatology. Symposium proceedings / the Society for Investigative Dermatology, Inc. [and] European Society for Dermatological Research, 14 (2009) 60-62. [48] J. Samarin, I. Cicha, M. Goppelt-Struebe, Cell type-specific regulation of CCN2 protein expression by PI3K-AKT-FoxO signaling, Journal of cell communication and signaling, 3 (2009) 79-84.

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

[49] H. Yang, L. Zhan, T. Yang, L. Wang, C. Li, J. Zhao, Z. Lei, X. Li, H.T. Zhang, Ski prevents TGF-beta-induced EMT and cell invasion by repressing SMAD-dependent signaling in non-small cell lung cancer, Oncology reports, 34 (2015) 87-94. [50] B. Ha, E.B. Lee, J. Cui, Y. Kim, H.H. Jang, YB-1 overexpression promotes a TGF-beta1induced epithelial-mesenchymal transition via Akt activation, Biochemical and biophysical research communications, 458 (2015) 347-351. [51] J.Y. Wang, Y.B. Gao, N. Zhang, D.W. Zou, P. Wang, Z.Y. Zhu, J.Y. Li, S.N. Zhou, S.C. Wang, Y.Y. Wang, J.K. Yang, miR-21 overexpression enhances TGF-beta1-induced epithelial-tomesenchymal transition by target smad7 and aggravates renal damage in diabetic nephropathy, Molecular and cellular endocrinology, 392 (2014) 163-172. [52] Z. Chen, W. Li, Y. Ning, T. Liu, J. Shao, Y. Wang, Ski diminishes TGF-beta1-induced myofibroblast phenotype via up-regulating Meox2 expression, Experimental and molecular pathology, 97 (2014) 542-549. [53] X. Yan, J. Zhang, L. Pan, P. Wang, H. Xue, L. Zhang, X. Gao, X. Zhao, Y. Ning, Y.G. Chen, TSC-22 promotes transforming growth factor beta-mediated cardiac myofibroblast differentiation by antagonizing Smad7 activity, Molecular and cellular biology, 31 (2011) 3700-3709. [54] C. Jopling, E. Sleep, M. Raya, M. Marti, A. Raya, J.C. Izpisua Belmonte, Zebrafish heart regeneration occurs by cardiomyocyte dedifferentiation and proliferation, Nature, 464 (2010) 606609. [55] R.B. Driesen, C.K. Nagaraju, J. Abi-Char, T. Coenen, P.J. Lijnen, R.H. Fagard, K.R. Sipido, V.V. Petrov, Reversible and irreversible differentiation of cardiac fibroblasts, Cardiovascular research, 101 (2014) 411-422. [56] S. Rosenkranz, TGF-beta1 and angiotensin networking in cardiac remodeling, Cardiovascular research, 63 (2004) 423-432. [57] F. Strutz, M. Zeisberg, A. Renziehausen, B. Raschke, V. Becker, C. van Kooten, G. Muller, TGF-beta 1 induces proliferation in human renal fibroblasts via induction of basic fibroblast growth factor (FGF-2), Kidney international, 59 (2001) 579-592. [58] X. Zhang, N. Tang, T.J. Hadden, A.K. Rishi, Akt, FoxO and regulation of apoptosis, Biochimica et biophysica acta, 1813 (2011) 1978-1986. [59] Y. Terada, T. Okado, S. Inoshita, S. Hanada, M. Kuwahara, S. Sasaki, T. Yamamoto, F. Marumo, Glucocorticoids stimulate p21(CIP1) in mesangial cells and in anti-GBM glomerulonephritis, Kidney international, 59 (2001) 1706-1716. [60] P. Lijnen, V. Petrov, R. Fagard, Transforming growth factor-beta 1-mediated collagen gel contraction by cardiac fibroblasts, Journal of the renin-angiotensin-aldosterone system : JRAAS, 4 (2003) 113-118. [61] M. Bujak, N.G. Frangogiannis, The role of TGF-beta signaling in myocardial infarction and cardiac remodeling, Cardiovascular research, 74 (2007) 184-195. [62] M. Dobaczewski, M. Bujak, N. Li, C. Gonzalez-Quesada, L.H. Mendoza, X.F. Wang, N.G. Frangogiannis, Smad3 signaling critically regulates fibroblast phenotype and function in healing myocardial infarction, Circulation research, 107 (2010) 418-428. [63] E. Xiao, D.T. Graves, Impact of Diabetes on the Protective Role of FOXO1 in Wound Healing, Journal of dental research, 94 (2015) 1025-1026. [64] H. Francois, K. Athirakul, D. Howell, R. Dash, L. Mao, H.S. Kim, H.A. Rockman, G.A. Fitzgerald, B.H. Koller, T.M. Coffman, Prostacyclin protects against elevated blood pressure and cardiac fibrosis, Cell metabolism, 2 (2005) 201-207. [65] T. Ichiki, J.A. Schirger, B.K. Huntley, F.V. Brozovich, J.J. Maleszewski, S.M. Sandberg, S.J. Sangaralingham, S.J. Park, J.C. Burnett, Jr., Cardiac fibrosis in end-stage human heart failure and the cardiac natriuretic peptide guanylyl cyclase system: regulation and therapeutic implications, Journal of molecular and cellular cardiology, 75 (2014) 199-205.

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

[66] J.W. Lee, H. Chen, P. Pullikotil, M.J. Quon, Protein kinase A-alpha directly phosphorylates FoxO1 in vascular endothelial cells to regulate expression of vascular cellular adhesion molecule-1 mRNA, The Journal of biological chemistry, 286 (2011) 6423-6432. [67] M. Shen, Z. Liu, B. Li, Y. Teng, J. Zhang, Y. Tang, S.C. Sun, H. Liu, Involvement of FoxO1 in the effects of follicle-stimulating hormone on inhibition of apoptosis in mouse granulosa cells, Cell death & disease, 5 (2014) e1475.

ACCEPTED MANUSCRIPT Figure legends Fig. 1.TGF-β1 increases FoxO1 and -SMA protein levels. A) CF were incubated for 48 h with

PT

TGF-β1 (0 to 20 ng/mL). B) CF were incubated with TGF-β1 (10 ng/mL) for 0 to 72 h. C) CF were

RI

incubated with TGF-β1 (10 ng/mL) for 0 to 72 h. D) CF were preincubated with actinomycin D (ACT, 2 µg/mL), cycloheximide (CHX, 10 µM), or MG132 (MG, 10 nM) for 1 h; cells were then

SC

incubated with TGF-β1 (10 ng/mL) for 24 h. In A), B), and D), FoxO1 and -SMA protein levels

NU

were quantified by Western blotting. GAPDH was used as a load control. In C), FoxO1 was detected by immunofluorescence using anti-FoxO1 antibody and Alexa Fluor® 488-conjugated

MA

secondary antibody (green staining). α-SMA was detected using anti-α-SMA antibody and Alexa Fluor® 566-conjugated secondary antibody (red staining). **p<0.01 and ***p<0.001 vs. control

TE

D

(C); ##p<0.01 vs. TGF-1 (-). Data are mean ± SEM of four independent experiments.

AC CE P

Fig. 2.TGF-β1 promotes FoxO1 activity. A) CF were incubated with TGF-β1 (10 ng/mL) for 0 to 8 h. p-FoxO1 and FoxO1 protein levels were determined by Western blotting. GAPDH was used as a load control. In B), CF were seeded on coverslips and then stimulated with TGF-β1 (10 ng/mL) for 2, 4, and 8 h. Nuclear localization of FoxO1 was detected by immunofluorescence using antiFoxO1 antibody and Alexa Fluor® 488-conjugated secondary antibody (green staining). α-SMA was detected using anti-α-SMA antibody and Alexa Fluor® 566-conjugated secondary antibody (red staining). C) CF were stimulated with TGF-β1 (10 ng/mL) for 4 and 8 h. FoxO1 protein levels in nuclei were determined by Western blotting. Nuclear isolation was performed by centrifugation as mentioned in the Methods section. FoxO1 cytosolic protein levels were determined from the ratio of FoxO1 to β-tubulin, and nuclear FoxO1 content was determined from the ratio of FoxO1 to lamin A/C, both reported as -fold over control. D) and E) CF were incubated with TGF-β1 (10 ng/mL) for different experimental times (0 to 72 h). SOD2 (D) and p21waf/cip (E) protein levels were

ACCEPTED MANUSCRIPT determined by Western blotting. GAPDH was use as a load control. *p<0.01 **p<0,01 and ***p<0.001 vs. control (C); #p<0.01 and ###p<0.001 vs. control nuclear (C). Data are mean ± SEM

RI

PT

of four independent experiments.

SC

Fig. 3.FoxO1 knockdown decreases the effects of TGF-β1 in cardiac fibroblasts. CF were transfected with FoxO1 siRNA (si-FoxO1), FoxO3a siRNA (si-FoxO3a), or siRNA negative control

NU

(Scr) for 16 h and then stimulated with TGF-β1 (10 ng/mL) for 48 h. (A) FoxO1 protein levels were determined by Western blotting. GAPDH was used as load control. (B) Total protein content was

MA

quantified by Bradford assay. α-SMA was evaluated by Western blotting (C) and immunofluorescence (D). pro-collagen I (E), and CTGF (F) protein levels were determined by

D

Western blotting. GAPDH was used as a load control. (G) CF contractile ability was evaluated by

TE

collagen gel contraction assay. The panel shows representatives images of collagen gel contraction,

AC CE P

and the bar graph shows the percentage of gel area reduction as compared to the initial area. The final area was evaluated at 48 h. **p<0.01 and ***p<0.001 vs.Scr; #p<0.05,

##

p<0.01 and

p<0.001 vs.Scr + TGF-β1. Data are mean ± SEM of four independent experiments.

###

Fig. 4.FoxO1 knockdown decreases the cytostatic effects of TGF-β1 on cardiac fibroblasts. CF were transfected with si-FoxO1, si-FoxO3a, or Scr for 16 h and then stimulated with TGF-β1 (10 ng/mL) for 48 h and incubated with FBS 10% for 24 h. A) Cell proliferation was evaluated by cell count in a Neubauer chamber. B) CF distribution in G0/G1 phase and S/G2/M phase was evaluated by flow cytometry. (C) p21waf/cip protein levels were determined by Western blotting. GAPDH was used as a load control. *p<0.05, **p<0.01 and ***p<0.001 vs.Scr; #p<0.05,

##

p<0.01 and

p<0.001 vs.Scr + FBS; &p<0.05 vs.Scr + FBS + TGF-β1. Data are mean ± SEM of four

###

independent experiments.

ACCEPTED MANUSCRIPT

Fig. 5.FoxO1 up-regulation enhances the effects of TGF-β1 on cardiac fibroblasts. CF were

PT

transduced with either FoxO1-wild-type adenovirus (Ad-FoxO1-wt), FoxO1 constitutive active adenovirus (Ad-FoxO1-ca), or GFP adenovirus (Ad-GFP) for 24 h and then stimulated with TGF-

RI

β1 (10 ng/mL) for 48 h. A) Total protein amount was analyzed by Bradford assay. α-SMAwas

SC

evaluated by Western blotting (B) and immunofluorescence (C). pro-collagen I (D), and CTGF (E)

NU

protein levels were determined by Western blotting. GAPDH was used as a load control. (F) CF contractile ability was evaluated by collagen gel contraction assay. The panel shows representatives

MA

images of collagen gel contraction, and the bar graph shows the percentage of gel area reduction as compared to the initial gel area. The final area was evaluated at 48 h. **p<0.01 and ***p<0.001 vs. p<0.01 vs. GFP + TGF-β1. Data are mean ± SEM of four independent

##

D

GFP; #p<0.05 and

AC CE P

TE

experiments.

Fig. 6.FoxO1 up-regulation enhances the cytostatic effects of TGF-β1 on cardiac fibroblasts. CF were transduced with Ad-FoxO1-wt, Ad-FoxO1-ca, or Ad-GFP for 24 h and then stimulated with TGF-β1 (10 ng/mL) for 48 h. CF were then incubated with FBS 10% for 24 h. A) Cell proliferation was evaluated by cell count in a Neubauer chamber. B) CF distribution in G0/G1 phase and S/G2/M phase was evaluated by flow cytometry. (C) p21waf/cip protein levels were determined by Western blotting. GAPDH was used as a load control. **p<0.01 and ***p<0.001 vs. GFP; #p<0.05, ##p<0.01 and

p<0.001 vs. GFP + FBS; &p<0.05 vs. GFP + FBS + TGF-β1. Data are mean ± SEM of four

###

independent experiments.

Figure 1

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 2

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 3

Figure 4

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 5

Figure 6

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Highlights. TGF-β1 induces the activation of FoxO1 and increases FoxO1 protein level.



TGF-β1 promotes cardiac fibroblasts differentiation into cardiac myofibroblasts.



FoxO1 silencing prevents TGF-β1-dependent cardiac fibroblasts differentiation.



FoxO1 overexpression enhances TGF-β1-dependent cardiac fibroblasts differentiation.

AC CE P

TE

D

MA

NU

SC

RI

PT

