Inhibition of Signal Transducer and Activator of Transcription 3 (STAT3) Attenuates Interleukin-6 (IL-6)-induced Collagen Synthesis and Resultant Hypertrophy in Rat Heart

IL-6 has been shown to play a major role in collagen up-regulation process during cardiac hypertrophy, although the precise mechanism is still not known. In this study we have analyzed the mechanism by which IL-6 modulates cardiac hypertrophy. For the in vitro model, IL-6-treated cultured cardiac fibroblasts were used, whereas the in vivo cardiac hypertrophy model was generated by renal artery ligation in adult male Wistar rats (Rattus norvegicus). During induction of hypertrophy, increased phosphorylation of STAT1, STAT3, MAPK, and ERK proteins was observed both in vitro and in vivo. Treatment of fibroblasts with specific inhibitors for STAT1 (fludarabine, 50 μm), STAT3 (S31-201, 10 μm), p38 MAPK (SB203580, 10 μm), and ERK1/2 (U0126, 10 μm) resulted in down-regulation of IL-6-induced phosphorylation of specific proteins; however, only S31-201 and SB203580 inhibited collagen biosynthesis. In ligated rats in vivo, only STAT3 inhibitors resulted in significant decrease in collagen synthesis and hypertrophy markers such as atrial natriuretic factor and β-myosin heavy chain. In addition, decreased heart weight to body weight ratio and improved cardiac function as measured by echocardiography was evident in animals treated with STAT3 inhibitor or siRNA. Compared with IL-6 neutralization, more pronounced down-regulation of collagen synthesis and regression of hypertrophy was observed with STAT3 inhibition, suggesting that STAT3 is the major downstream signaling molecule and a potential therapeutic target for cardiac hypertrophy.

Cardiac fibrosis is considered to be a major player during hypertrophy, where excess synthesis and a disproportionate accumulation of extracellular matrix proteins in the myocardium leads to its stiffness and diastolic dysfunction, leading to heart failure (1-3). Fibrillar collagen 1 and collagen 3 are the most abundant extracellular matrix proteins in the myocardium that maintain myocardial structural integrity (4). Collagen biosynthesis process is mainly compartmentalized in cardiac fibroblasts with a continuous turnover of newly synthesized collagen and degradation of old existing collagen, which involves post-transcriptional as well as post translational events. A physiological balance exists between collagen synthe-sis and degradation that is necessary to maintain tissue integrity of the myocardium. Extracellular catalytic cleavage of collagen is mediated by matrix metalloproteinases (MMPs) 2 that are eventually regulated by their naturally occurring endogenous inhibitors, i.e. tissue inhibitors of metalloproteinases (TIMPs) (1,5).The fine balance between MMPs and TIMPs plays a crucial role in regulation of cardiac collagen turnover (5,6). Furthermore, lysyl oxidase (LOX) is also instrumental during collagen biogenesis that catalyzes lysine-derived cross-bridges between two or more lysine residues of nascent procollagen peptides during their maturation into collagen molecules (5,7). Altered and deregulated expression of these factors has been demonstrated in failing myocardium (5).
Involvement of various neurohumoral and growth factors has also been illustrated in the development of cardiac fibrosis both in vitro and in vivo (3,8). Cellular cross-talk between AngII-treated cardiomyocytes and fibroblasts has been shown to play an important role during collagen up-regulation with involvement of IL-6 during fibrosis (8 -10). Recently, Meléndez et al. (11) showed that IL-6 infusion in rats resulted in cardiac hypertrophy with considerable fibrosis within the myocardium. But the precise mechanism and involvement of key signaling molecules for collagen up-regulation process during hypertrophy is still unclear.
It is known that the IL-6 family of cytokines mediates their action primarily through the activation of Janus kinase signal transducers and activators of transcription (JAK-STAT) pathway along with other intracellular signaling pathways (12). The IL-6 family of cytokines mostly activates STAT3 and to a lesser extent STAT1 via a common receptor subunit (13). STAT proteins are also shown to be activated by interleukins indirectly through the activation of mitogen-activated protein kinases (MAPKs). STAT proteins have been shown to induce transcription of several genes involved in ischemia/reperfusion injury and pressure-overload hypertrophy, as well as in cardioprotection, angiogenesis, cell survival, and apoptosis (14 -20). The present study was, therefore, designed to study the functional role of STAT proteins in IL-6-mediated collagen gene up-regulation in vitro and in vivo. This is the first report showing involvement of STAT3 in the IL-6-induced collagen up-regulation process during cardiac hypertrophy.

EXPERIMENTAL PROCEDURES
28-Week-old Wistar male rats used in this study were obtained from the National Institute of Nutrition, Hyderabad, India. This investigation was approved by the Institutional Animal Ethics Committee, University of Calcutta (Registration #885/ac/05/CPCSEA), registered under "Committee for the Purpose of Control and Supervision of Experiments on Laboratory Animals," Ministry of Environment and Forests, Government of India and conforms with the Guide for the Care and Use of Laboratory Animals published by the United States National Institutes of Health Publication 85-23, revised 1996.
Cardiac Fibroblast Culture and Treatment-Fibroblast cells were isolated from 28-week-old male rat hearts by the collagenase dispersion method (21). Briefly, animals were euthanized in a prefilled CO 2 chamber with 100% concentration of CO 2 , and the hearts were cut into small pieces and digested by collagenase (80 units/ml DMEM; Worthington). The cells were pelleted by centrifugation and resuspended in fresh complete DMEM supplemented with 10% fetal bovine serum and plated in cell culture flask. Cells were maintained at 37°C with 5% CO 2 and were subsequently passaged. 75-80% confluent serum-starved cells were treated for 24 h with 10 Ϫ8 mol/liter (Sar 1 )-Angiotensin II (Bachem) and 50 ng/ml murine IL-6 (Peprotech). In another set of experiments IL-6 was preincubated with neutralizing IL-6 antibody (Abcam; catalog #ab9770) for 1 h before using it to treat cultured cardiac fibroblasts. Time-dependent phosphorylation status of different signaling proteins by AngII or IL-6 treatment was also studied starting from 5 min onwards until 12 hr. AngII-treated cells were used as positive controls for all subsequent experiments. Untreated cells were used as controls.
Generation of Animal Model of Cardiac Hypertrophy-In vivo cardiac hypertrophy model was generated by ligating right renal artery of 28-week-old male rats (250 -300 g; n ϭ 10) as described earlier (22). Rats were anesthetized with an intraperitoneal injection of a mixture of ketamine (80 mg/kg) and xylazine (12 mg/kg). The right renal artery was tied with a silk suture after placing a stainless wire (outer diameter, 0.34 mm) along the artery. The wire was then removed from the knot, causing constriction of the artery equivalent to the diameter of the wire. Rats that underwent a similar procedure without aortic ligation were termed as sham-operated control group (n ϭ 10). Animals were maintained in optimum condition for 14 days and were sacrificed on the 15th day after surgery. Hearts were taken out and either fixed in Karnovsky's fixative for histological study or stored in liquid nitrogen for future use. Hypertrophy was measured by the heart weight (HW in mg) to body weight (BW in g) ratio (23). For neutralization experiment, one group of ligated rats was administered with neutralizing IL-6 antibody (10 g/day; Abcam; catalog #ab9770) intraperitoneally on alternate days after ligation until the day of sacrifice (24).
Determination of Cardiac Function-Two-dimensional echocardiography was performed to determine cardiac function in vivo. Lightly sedated renal artery-ligated and sham-op-erated rats on the 15th day after surgery were evaluated using the M-mode views on a transthoracic study, measuring left ventricular systolic and diastolic dimensions and fractional shortening (%). Data were correlated with timing of the QRS complex. Digitized images were obtained using an ultrasound system (VividS5 system, GE Healthcare).
Estimation of IL-6 in Fibroblast Culture Supernatant and Rat Blood Serum by ELISA-Secreted IL-6 content in the AngIItreated (24 h) fibroblast culture supernatant and circulatory IL-6 level in rat blood serum (on the 15th day after surgery) was estimated by ELISA using rat IL-6 ELISA Kit (Thermo-Scientific, IL) following the manufacturer's procedure. The amount of IL-6 in each sample was determined by interpolating the absorbance value against a standard curve, and values were expressed as picograms of IL-6 per ml sample (pg/ml).
Reverse Transcription PCR (RT-PCR) and Real Time RT-PCR-Total RNA from cells (24 h treatment) and tissues (on the 15th day after surgery) was isolated using TRIzol reagent (Invitrogen) following manufacturer's instructions. 1 g of total RNA was used to make cDNA using the cloned avian myeloblastosis virus (AMV) first-strand cDNA synthesis kit (Invitrogen). RT-PCR was done for collagen 1, collagen 3, MMP-2, MMP-9, TIMP-1, TIMP-2, LOX, GAPDH, atrial natriuretic factor (ANF), ␤-myosin heavy chain (␤-MHC), c-fos, c-jun, and c-myc using specific oligonucleotide primers (supplemental Table 1). Relative quantification of PCR-amplified products was also done by real-time PCR with Power SYBR Green TM PCR Master Mix using ABI7500 (Applied Biosystems). Collagen 1, collagen 3, MMP-2, MMP-9, TIMP-1, TIMP-2, and LOX gene expressions were studied by real time PCR, and GAPDH was used as a reference gene to normalize expression of other genes (supplemental Table 2). All real time PCR reactions were done in triplicate after an initial incubation at 95°C for 10 min and then a PCR cycle at 95°C for 30s, 59°C for 30s, and 72°C for 30s for 40 cycles. Relative gene expression was quantified by comparative "ct" (2 Ϫ⌬⌬ct ) method.
Estimation of Total Collagen by Hydroxyproline Assay-Hydroxyproline assay was performed to measure total collagen content both in renal artery-ligated ventricular tissue (on the 15th day after surgery) as well as in fibroblast culture supernatant (24 h treatment) (25). Briefly, the tissue samples and fibroblast culture supernatants were digested with 6 N hydrochloric acid overnight at 110°C followed by vacuum drying of the samples. After resuspending the samples in citrate acetate buffer, a colored reaction was done by adding isopropyl alcohol, chloramine T, and Ehrlich's reagent. The samples were incubated at 25°C for 18 h, and intensity of the red color was measured at 558 nm using Varioskan Multimode Reader (Thermo Fisher). With the help of a standard curve, hydroxyproline content in the unknown samples was calculated. The amount of collagen was calculated by multiplying hydroxyproline content by a factor of 8.2.
Treatment of Cardiac Fibroblasts with Inhibitors to STAT and MAPK Proteins-Treatment of cultured fibroblasts with IL-6 was done in the presence or absence of different STAT and MAPK inhibitors. Fludarabine (50 M) (Aldrich), a specific STAT1 inhibitor, and S31-201 (10 M) (Calbiochem), a STAT3 inhibitor, were used in this study (27,28). SB203580 (Calbiochem) at a concentration of 10 M and U0126 (Calbiochem) at a concentration of 10 M were used to block p38 MAPK and ERK1/2 respectively (29). All the inhibitors were added 45 min before the treatment, except fludarabine, which was added 2 hours before treatment (27)(28)(29). Cells treated with equivalent concentration of DMSO were used as controls. IL-6 induction of cultured cells after inhibitor treatment was done for 20 min to check phosphorylation status of STAT3, STAT1, p38 MAPK, and ERK1/2.
Treatment of Cardiac Fibroblasts with STAT3 siRNA-STAT3 activity in cultured cardiac fibroblasts was also blocked by using a STAT3-specific siRNA. Cultured fibroblasts were transfected with 300 pmol of STAT3 siRNA (SI 02040731; Qiagen) using Lipofectamine 2000 reagent (Invitrogen) following the manufacturer's guidelines. 24 h after transfection cells were used for treatment with IL-6. Cells transfected with nonspecific siRNA (NSsiRNA) (SI03650325; Qiagen) were used as controls. 20 min of IL-6 treatment was done to study protein phosphorylation, and treatment for 24 h was performed to study mRNA expression.
Administration of STAT3 Inhibitor (S31-201) and STAT3 siRNA to Renal Artery-ligated Rats-STAT3 inhibitor (S31-201) was administered intraperitoneally in renal artery-ligated rats at a dose of 5 mg/kg of body weight from the 1st day of ligation until the 14th day (30). A separate group of renal arteryligated animals (vehicle group) was also injected with an equivalent concentration of the solvent (DMSO) in which S31-201 was dissolved. STAT3 siRNA for in vivo delivery (In vivo Direct siRNA; s129047; Ambion) was dissolved in RNase free 1ϫ sterile PBS and was directly injected to the ventricles of ligated animals (n ϭ 10) following the manufacturer's instructions at a concentration of 10 nmol from the 1st day of ligation until the 14th day (22). Another group of renal artery-ligated rats was also treated with nonspecific control siRNA (NSsiRNA) (4404020; Ambion). After 14 days animals were sacrificed, and heart tissues were collected in liquid N 2 and stored in Ϫ80°C for future use.
Statistical Analysis-All results are expressed as ϮS.E. Data were analyzed with one way analysis of variance followed by Student's t test using SPSS software (Version 14.0). Experiments were repeated at least three times before analysis. Results with p value Ͻ0.05 were considered significant.

Estimation of Circulating IL-6 Level during Hypertrophy-
The role of IL-6 in hypertrophy has been reported by several authors (8 -11). To confirm release of IL-6 during renal artery ligation-induced hypertrophy, serum IL-6 was measured by ELISA. A significant increase in the circulating IL-6 level was observed in the serum of renal artery-ligated rats (721.416 Ϯ 34.45 pg/ml) compared with sham-operated control animals (39.33 Ϯ 12.96 pg/ml; Fig. 1A). Treatment of renal artery-ligated animals with neutralizing IL-6 antibody showed a 48.1% decrease in circulating IL-6 level compared with ligated rats (Fig. 1A). Western blotting also showed a significant decrease in IL-6 levels in rat ventricular tissue after IL-6 neutralization (data not shown) together with a significant decrease in expression of hypertrophy marker genes as revealed by RT-PCR (Fig.  1B), suggesting that increased circulating IL-6 after renal artery ligation was responsible for development of cardiac hypertrophy in rats.
Isolated cardiac fibroblasts treated with IL-6 or AngII was used to generate hypertrophy in vitro. Cultured cardiac fibroblasts were treated with 10 Ϫ8 mol/liter (Sar 1 )-Angiotensin II with or without neutralizing IL-6 antibody, and secreted IL-6 levels in culture supernatant were estimated by ELISA. A significant increase in secreted IL-6 levels was observed in supernatants of AngII-treated fibroblasts (283.5 Ϯ 38.21 pg/ml) compared with untreated control cells (127.25 Ϯ 11.59 pg/ml), which was reduced significantly when IL-6 was preincubated with neutralizing IL-6 antibody before treatment (data not shown).
Up-regulation of Collagen Expression during Hypertrophy-A significant increase in collagen expression was observed in the ventricular tissue of renal artery-ligated rats (Fig. 1C). The role of IL-6 in the collagen induction process in our renal arteryligated rats was confirmed by treating the ligated rats with neutralizing IL-6 antibody, which showed a significant decrease (29.72%) in total ventricular collagen content (Fig. 1C). Similarly, IL-6 neutralization in vitro also resulted in significant decrease (48.24%) in IL-6-induced collagen expression in cultured cardiac fibroblast cells (Fig. 1D).
Phosphorylation of STAT and MAPK Proteins during Hypertrophy-Time-dependent phosphorylation of STAT1 (Tyr-701 and Ser-727), STAT3 (Tyr-705 and Ser-727), p38 MAPK, and ERK1/2 was observed after IL-6 treatment of cultured cardiac fibroblasts. Maximum phosphorylation of all these signaling proteins was observed within 15-30 min of IL-6 treatment (supplemental Figs. 1A and 2A). Earlier results were corroborated by an immunofluorescence study where IL-6treated (20 min) cardiac fibroblasts showed induction and nuclear translocation of p-STAT1-Tyr-701, p-STAT3-Tyr-705, p-p38 MAPK, and p-ERK1/2 proteins compared with untreated cells (supplemental Figs. 1, B and C, and 2, B and C). Cells were counterstained with anti-vimentin antibody (green fluorescence) for fibroblasts specificity. A similar time-dependent phosphorylation of these proteins was observed in AngIItreated cardiac fibroblasts (data not shown).
To block the phosphorylation of these proteins, cultured cardiac fibroblasts were pretreated with specific inhibitors to these No change was observed in total p38 MAPK and ERK1/2 levels after inhibitor treatment (Fig. 2B).

Inhibition of STAT3 Activation in Vivo in Renal Artery-ligated Rats Resulted in Decreased Cardiac Collagen Synthesis-
Concurrent to in vitro results, significant phosphorylation of STAT3 at Tyr-705 (4.69 Ϯ 0.21-fold; p Ͻ 0.001) as well as Ser-727 (3.25 Ϯ 0.18-fold; p Ͻ 0.001) was observed in hypertrophied rat hearts compared to sham-operated animals (Fig. 3, A  and B), with no change in total STAT3 levels. Phosphorylation of STAT1 (Tyr-701 and Ser-727), p38 MAPK, and ERK1/2 was also significantly pronounced in renal artery-ligated rat hearts compared to sham-operated controls (data not shown). To block phosphorylation of STAT3 in vivo, renal artery-ligated animals were treated with either S31-201 or STAT3 siRNA, as described under "Experimental Procedures." Western blotting with whole cell extracts prepared from these ventricular tissues showed significant down-regulation in p-STAT3-Tyr-705 levels in STAT3 siRNA as well as S31-201-treated samples (2.9 Ϯ 0.15-fold and 2.6 Ϯ 0.14-fold; p Ͻ 0.01 respectively) compared to ligated animals treated either with DMSO or nonspecific siRNA (Fig. 3, A and B). Although, S31-201 treatment did not change p-STAT3-Ser-727 or total STAT3 levels, STAT3 siRNA-treated renal artery-ligated rats showed a decrease in p-STAT3-Ser-727 and total STAT3 levels compared to DMSO or nonspecific siRNA-treated ligated rats (Fig. 3, A and B). Thus, the efficacy of STAT3 siRNA in down-regulation of STAT3 protein in vivo was confirmed.
A significant increase in expression of both collagen 1 (3.6 Ϯ 0.18-fold; p Ͻ 0.01) and collagen 3 (2.9 Ϯ 0.2-fold; p Ͻ 0.01) genes was also observed in vivo in hypertrophied rat hearts compared to sham-operated controls (Fig. 3, C and   D). A significant increase in total collagen content was also observed in ligated heart tissues compared to sham-operated animals (383.79 Ϯ 40.9 g/g wet tissue versus 195.64 Ϯ 28.232 g/g wet tissue, p Ͻ 0.01) as revealed by hydroxyproline assay (Fig. 3E). Masson's Trichrome staining of ventricular tissue sections revealed substantial infiltration and deposition of collagen in renal artery-ligated rats compared with control tissue sections. Quantitative analysis of the Masson's Trichrome-stained tissue sections revealed that collagen volume fraction (%) in ligated tissue was significantly higher than that of sham-operated animals (20.38 Ϯ 0.99 versus 9.42 Ϯ 0.62; p Ͻ 0.001; data not shown).
Because STAT3 inhibitor treatment in vitro showed maximum down-regulation of collagen, in vivo experiments were done to confirm the role of STAT3 in collagen biosynthesis process. S31-201 and STAT3 siRNA-treated renal artery-ligated animals showed a significant decrease in both collagen 1 (2.3 Ϯ 0.2-fold; p Ͻ 0.01) and collagen 3 (2.2 Ϯ 0.14-fold; p Ͻ 0.01) gene expression compared to renal artery-ligated rats with DMSO or nonspecific siRNA treatment (Fig. 3, C and D). In addition, significant lowering of total ventricular collagen content was also observed in both STAT3 siRNA as well as S31-201-treated animals compared to ligated ones (213.6 Ϯ 29.87 and 222.8 Ϯ 26.1 g/g, respectively versus 383.79 Ϯ 40.9 g/g wet tissue; p Ͻ 0.01; Fig. 3E). In vivo treatment with p38 MAPK inhibitor (SB203580) had no significant effect on collagen gene expression (data not shown).
Inhibition of STAT3 Activation Resulted in Regression of Cardiac Hypertrophy in Vivo-Previous results confirmed that STAT3 inhibition resulted in down-regulation of collagen biosynthesis. However, whether this had a downstream effect on cardiac hypertrophy process, HW/BW ratio and expression of hypertrophy marker genes were measured. S31-201 and STAT3 siRNA treatment of ligated rats resulted in a significant decrease in HW to BW ratio in these animals (3.15 Ϯ 0.077 and 3.07 Ϯ 0.21, respectively, versus 4.03 Ϯ 0.098, p Ͻ 0.001; Fig.  respectively, compared to controls treated with either DMSO or nonspecific siRNA (Fig. 5, B and C). Inhibition of STAT3 also resulted in significantly improved cardiac function in these renal artery-ligated animals as evident by decreased left ventricular diastolic dimensions and increased fractional shortening % as observed by M-mode echocardiography analysis (Table 1).

DISCUSSION
Cardiac fibrosis leading to ventricular stiffness is responsible for the development of diverse pathological states including cardiac dysfunction during various heart diseases, and accumulation of major extracellular matrix protein collagen 1 is one of the main events in cardiac remodeling process in diseased hearts (31). Increased deposition of collagen proteins was observed in hypertensive heart disease and in patients with dilated cardiomyopathy and end stage heart failure (31). Significant up-regulation of collagen 1 and collagen 3 expression as well as excess collagen deposition was evident in our in vivo pressure overload hypertrophy model (Figs. 1C and 3, C-E). Many growth factors and cytokines have been shown to modulate cardiac hypertrophy and collagen synthesis involving var- NSsiRNA. GAPDH was used as the internal loading control. D, shown is a graphic representation of the above RT-PCR results. **, p Ͻ 0.01 with respect to Sham-treated; † †, p Ͻ 0.01 with respect to ligated (L) or LϩNSsiRNA treatment; n ϭ 10 in each group. E, a significant increase in total ventricular collagen was observed in renal artery-ligated rats as revealed by hydroxyproline assay. Total ventricular collagen was decreased significantly in STAT3 siRNA (213.6 Ϯ 29.87 g/g wet tissue) as well as S31-201 (222.8 Ϯ 26.1 g/g wet tissue)-treated ligated rats compared with only ligated animals (383.79 Ϯ 40.9 g/g wet tissue). Results are expressed as ϮS.E. of three independent experiments. **, p Ͻ 0.01 with respect to Sham-treated; † †, p Ͻ 0.01 with respect to L or LϩNSsiRNA treatment; n ϭ 10 in each group).
ious signaling pathways in isolated systems (3). Increased activity of IL-6, a proinflammatory cytokine during hypertrophy and heart failure has been reported by several authors (32). Meléndez et al. (11) showed that IL-6 infusion causes left ventricular hypertrophy and myocardial fibrosis in rats. AngII, known to induce cardiac hypertrophy involving renin-angiotensin system, was shown to induce IL-6 production both in cultured cardiac fibroblasts and during maladaptive pressure overload hypertrophy via activation of gp130 (9,10,13). IL-6 has been shown to increase collagen synthesis both in in vitro cultured fibroblasts and in in vivo rat hearts (8,9,11). Consistent with these previous studies, we also observed increased levels of IL-6 after AngII treatment of cardiac fibroblasts (data not shown) as well as up-regulation of collagen expression by IL-6 (Figs. 1D and 2, C and D). Despite its significant role, precise mechanism of IL-6-induced hypertrophy has not yet been clearly understood. In this study we have tried to look into the involvement of various signaling molecules associated with IL-6-mediated fibrosis and eventual compromised cardiac function. To confirm the role of IL-6 during hypertrophy, neutralizing IL-6 antibody was administered to the renal artery-ligated animals to neutralize the serum IL-6. IL-6 neutralization partially reduced cardiac hypertrophy as well as collagen synthesis in rat heart (Fig. 1, B and C). Total regression of hypertrophy after IL-6  neutralization did not occur, as several other factors also play a role during cardiac hypertrophy in vivo (33).
Downstream signaling by IL-6-gp130 axis involves activation of two major signaling pathways, namely JAK-STAT and MAPK-ERK; thus, significance of both pathways during hypertrophy was analyzed. Previous studies have shown increased tyrosine phosphorylation of STAT1 and STAT3 proteins during cardiac hypertrophy. Activation of STAT1 and STAT3 in the heart plays a very significant role in cardiac pathophysiology during various diseases, and their activity is increased during mechanical overload (34). During this study too, time-dependent phosphorylation of both STAT1 and STAT3 at tyrosine as well as serine residues in IL-6-stimulated cultured cardiac fibroblasts was observed (supplemental Fig. 1, A-C). Serine phosphorylation of both STATs achieved a maximum level before Tyr phosphorylation, which is suggestive of their role in regulation of full transcriptional activity of STAT proteins.
Although, total STAT3 deletion leads to embryonic lethality, cardiac-specific STAT3 deletion causes increased infarct size after ischemia/reperfusion injury (35,36). STAT3 overexpression results in decreased infarct size induced by ischemia/reperfusion injury and gives protection to cardiac myocytes during hypoxia/reoxygenation induced cell loss (37,38). Heart failure patients with end stage-dilated cardiomyopathy shows increased tyrosine phosphorylation of STAT3, but it is also established that STAT3 activation is instrumental in inhibition of apoptosis and necessary to promote compensatory hypertrophy and thus helps in prevention of heart failure (39). During ischemia, STAT3 protects the myocardium against STAT1-induced apoptosis (40,41). Although various roles of both these STAT proteins have been illustrated in heart, their involvement in collagen up-regulation and fibrosis process was not described. Primarily it was hypothesized that the calcium channel blocker Mibefradil inhibits AngII or aldosterone-induced cardiac fibrosis, suggesting the role of calcium signaling (42). Tranilast, used to treat allergic manifestations, also attenuated renovascular hypertension and myocardial fibrosis via inhibition of TGF-␤ signaling (43). AngII-induced cardiac fibrosis in mice was also attenuated through modulation of TGF-␤/Smad signaling cascade with allopurinol (44). In addition, angiotensin-converting enzyme inhibitor, ramipril and enalapril, treatment also reverses renal failureinduced left ventricular hypertrophy and fibrosis in rats (45,46). In our study inhibition of STAT3 phosphorylation with both S31-201 as well as STAT3 siRNA resulted in decreased collagen synthesis by IL-6 in cultured cardiac fibroblasts and also during pressure overload-induced cardiac hypertrophy in vivo (Figs. 2, C and D,  and 3, C-E). Although STAT1 was activated during this process, it did not have any significant role in collagen induction process (Fig.   2, C and D). Studies have shown involvement of STAT1 in induction of proapoptotic genes and also development of hypoxic eccentric hypertrophy together with STAT3 (47). STAT1 also has been shown to inhibit endothelial cell migration and tube formation in heart and thus opposes the effect of STAT3 in this respect (48,49). Further investigation is needed to delineate the role of STAT1 induction during IL-6-induced cardiac hypertrophy.
Similar to JAK-STAT pathways, MAP kinases phosphorylate their target proteins at specific serine/threonine residues to regulate their activity in myocardium (50). Increased activity of p38 MAPK has been observed during pressure overload-induced hypertrophy and in human heart failure (51)(52)(53). Inhibition of MAPK signaling has been shown to block hypertrophy both in vitro as well as in vivo and end organ damage during hypertensive heart disease, suggesting their role in hypertrophic signaling (52,54). IL-6 has been shown to activate serine/threonine kinases, which includes activation of p38 MAPK and ERK1/2 during cardiac remodeling (32,33). In addition to activation of p38 MAPK during hypertrophy, its role in the collagen induction process in isolated fibroblasts stimulated by IL-6 was observed in this study. Blocking p38 MAPK phosphorylation using its inhibitor (SB203580), collagen up-regulation was particularly regressed. The effect of p38 MAPK inhibitor was lower (ϳ19%) compared to STAT3 inhibitor (ϳ44% regression; Fig. 2, C and D) in vitro. Although, in vivo p38 MAPK inhibition could not regress the collagen induction process in this study, there are reports indicating the role of p38 MAPK signaling in serine phosphorylation of STAT proteins, which could be another possible mechanism by which MAPK affects the collagen induction process during hypertrophy (14,15,18). Again, several studies have established the role of ERK1/2 signaling in development of cardiac hypertrophy (55,56), corroborating our result that has shown prominent ERK1/2 activation in vitro (supplemental Fig. 2, A and C) as well as the in vivo (data not shown) model of hypertrophy, but at the same time, ERK1/2 inhibitor had no significant effect on collagen up-regulation process during hypertrophy (Fig. 2, C and D).
The extracellular collagen degradation is regulated by the ratio of MMPs and TIMPs (6), and their increased expression in failing myocardium has been reported (5). In renal artery-ligated hearts, a significant increase in expression of MMP-2, MMP-9, TIMP-1, and TIMP-2 was observed (Fig. 4, A and B). In contrast, only TIMP-1 expression was induced after in vitro treatment of cardiac fibroblasts by IL-6. This further confirms that IL-6 alone may result in induction of hypertrophy markers in vitro, but other cytokines in addition to IL-6 might play significant role in collagen modulation in renal artery-ligated hypertrophy in vivo (33). Compared with MMPs, a pronounced increase in expression of TIMPs was observed, suggesting lesser degradation and thus increased accumulation of collagen in the myocardium during hypertrophy. In addition, increased LOX expression (Fig. 4, A and B) further confirms increased collagen turnover during hypertrophy. Moreover, decreased expression of all these genes in vivo after treatment with the STAT3 inhibitor and siRNA suggest that activation of STAT3 during hypertrophy directly induces collagen synthesis as well as indirectly regulates collagen modulators to alter their expression to indirectly up-regulate collagen accumulation.
In vivo treatment with STAT3 inhibitor (S31-201) and STAT3 siRNA to the renal artery-ligated rats also resulted in regression of hypertrophy, as evident by a significant decrease in HW to BW ratio along with significant down-regulation in expression of hypertrophy marker genes and proto-oncogenes (Fig. 5, A-C). Reversal of hypertrophy after STAT3 inhibition by parthenolide has also recently been shown by Skoumal et al. (57), but the mechanism by which STAT3 could modulate hypertrophy process was not defined. Furthermore, it was shown that parthenolide treatment attenuated AngII-induced IL-6 mRNA levels (57), but in our in vivo rat model of hypertrophy, STAT3 inhibitors did not show any decrease in circulating IL-6 level (data not shown), suggesting that STAT3 is the downstream signaling molecule in IL-6-mediated collagen synthesis during hypertrophy. Results also suggest that reversal of hypertrophy process by STAT3 inhibition is associated with down-regulation of collagen synthesis although the hypertrophy regression due to STAT3 inhibition was much more pronounced compared with IL-6 neutralization in hypertrophy model in vivo. It is possible that during renal artery ligation other than IL-6, other cytokines and growth factors are induced that may also modulate cardiac fibrosis and resultant hypertrophy (33,58). Regression of cardiac hypertrophy was directly correlated with improved cardiac function in these animals, as shown by M-mode echocardiography data marked by decreased left ventricular diastolic dimensions and significantly increased % fractional shortening (Table 1). This is the first report on positive role of STAT3 during IL-6mediated collagen gene up-regulation during hypertrophy. Specific STAT3 inhibition not only down-regulated IL-6-induced collagen synthesis significantly but also effectively regressed the pressure overload-induced cardiac hypertrophy by reducing the fibrosis process. Although, both STAT3 and STAT1 have been implicated to play a role in cardiac hypertrophy, our results suggest that STAT3 and not STAT1 is the major transcription factor that up-regulates myocardial collagen synthesis during hypertrophy, and p38 MAPK activation may have an indirect role by enhancing STAT3 signaling. Thus, STAT3 could be a potential therapeutic target for regulating cardiac fibrosis during myocardial pressure-overload hypertrophy and improvement of cardiac function.