BAY 41-2272 Attenuates CTGF Expression via sGC/cGMP-Independent Pathway in TGFβ1-Activated Hepatic Stellate Cells

Activation of hepatic stellate cells (HSCs) is a critical pathogenic feature of liver fibrosis and cirrhosis. BAY 41-2272 is a canonical non-nitric oxide (NO)-based soluble guanylyl cyclase (sGC) stimulator that triggers cyclic guanosine monophosphate (cGMP) signaling for attenuation of fibrotic disorders; however, the impact of BAY 41-2272 on HSC activation remains ill-defined. Transforming growth factor (TGF)β and its downstream connective tissue growth factor (CTGF or cellular communication network factor 2, CCN2) are critical fibrogenic cytokines for accelerating HSC activation. Here, we identified that BAY 41-2272 significantly inhibited the TGFβ1-induced mRNA and protein expression of CTGF in mouse primary HSCs. Indeed, BAY 41-2272 increased the sGC activity and cGMP levels that were potentiated by two NO donors and inhibited by a specific sGC inhibitor, ODQ. Surprisingly, the inhibitory effects of BAY 41-2272 on CTGF expression were independent of the sGC/cGMP pathway in TGFβ1-activated primary HSCs. BAY 41-2272 selectively restricted the TGFβ1-induced phosphorylation of Akt but not canonical Smad2/3 in primary HSCs. Together, we illustrate a unique framework of BAY 41-2272 for inhibiting TGFβ1-induced CTGF upregulation and HSC activation via a noncanonical Akt-dependent but sGC/cGMP-independent pathway.


Introduction
The overwhelming activation and proliferation of hepatic stellate cells (HSCs) is an important factor in hepatic fibrosis, an advanced pathogenic condition in liver cirrhosis, portal hypertension, and hepatocellular carcinoma. During acute or chronic injuries, star-like shaped HSCs shift from quiescent non-proliferative status to active state with contractile, proliferative, and fibrogenic properties, purchased from Cell Signaling Technology (Beverly, MA, USA). Antibodies against GAPDH and CTGF were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Other chemicals were purchased from Sigma (St. Louis, MO, USA).

Isolation of Mouse Primary HSCs
The study was approved by and followed the guidelines of the Institutional Animal Care and Use Committee of Chang Gung University, Taiwan (IACUC approval No.: CGU15-085 and date: 11 November 2015). We purchased 8-to 10-week-old C57BL/6 male mice from BioLASCO (Taiwan). Five mice were kept in a cage under a 12-12-h light-dark cycle and were provided with water and food. HSCs were isolated from livers of 8-to 10-week-old C57BL/6 male mice using a modified, previously described procedure [30,31]. Briefly, mice were euthanized by CO 2 and perfused with normal saline via the inferior vena cava followed by injection with 0.1% (w/v) collagenase IV in HBSS. Livers were taken out and agitated in 0.1% (w/v) collagenase IV at 37 • C for 10 min. Cells were filtered through a nylon mesh and purified via Percoll gradient centrifugation. The isolated HSCs were cultured in Roswell Park Memorial Institute (RPMI) 1640 media supplemented with 10% fetal bovine serum and 10% horse serum at 37 • C in a humidified incubator with 5% CO 2 . The purity of HSCs was above 95%, as confirmed by their astrocytes, intracellular lipid droplets, and vitamin A autofluorescence [32]. HSCs were cultured for 7 days and then spread out as the next generation. The fifth to seventh generation of HSCs were seeded overnight for the following experiments.

Cell Viability
HSCs were cultured and starved in RPMI 1640 media for 24 h. Cells were treated with indicated compounds and then activated with TGFβ1 (5 ng/mL) for 0, 24, or 48 h. Cells were subsequently incubated with the WST-1 reagent at 37 • C for 2 h. The cell viability was monitored spectrophotometrically at 450 nm (Thermo Fisher Scientific; Waltham, MA, USA).

Immunofluorescence Staining
Cells placed on cover slides were fixed with 4% formaldehyde for 10 min and then incubated with 5% goat serum for 60 min. Protein levels were determined using primary antibodies against CTGF (catalog no. sc-25440, Santa Cruz Biotechnology) in 5% bovine serum albumin (BSA) at room temperature for 1 h and fluorescein isothiocyanate (FITC)-conjugated secondary antibody for another 1 h. Nuclei were counterstained with Hoechst (1 µg/mL). Images were obtained by fluorescent microscopy (OLYMPUS IX 81; Olympus, Tokyo, Japan).

RNA Isolation and Quantitative Real-Time PCR
Total RNA was extracted from HSCs using TRIzol reagent and 1 µg RNA was used as a template for cDNA synthesis by iScript cDNA Synthesis Kit according to the manufacturer's protocol. mRNA levels were determined using Power SYBR Green PCR Master Mix and quantitative PCR with CFX Connect Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). Primers for mouse CTGF (5 -GGAATTGTGACCTGAGTGACT-3 and 5 -TGAGCCAGCCATTTCTTAATAAAG-3 ) and mouse GAPDH (5 -AAGGAGTAAGAAACCCTGGAC-3 and 5 -GATGGAAATTGTGAGGGAGATG-3 ) were used. The real-time PCR was conducted at 95 • C for 10 min, followed by 40 cycles of denaturation at 95 • C for 15 s, and annealing/extension at 60 • C for 1 min. PCR conditions were optimized to achieve a single peak by melting curve analysis on the CFX Connect system.

Statistical Analysis
Data were expressed as the mean ± standard error of mean (SD). Statistical comparisons were made between two groups using Student's t-test. p < 0.05 was considered statistically significant.

BAY 41-2272 sGC Stimulator Inhibited TGFβ1-Induced CTGF Expression and Cell Proliferation in Primary HSCs
To evaluate the biological significance of sGC/cGMP signaling in primary HSCs, we first evaluated the effect of a typical sGC stimulator BAY 41-2272 on TGFβ1-induced CTGF expression that is a critical pro-fibrotic cytokine for HSC activation. BAY 41-2272 dose-dependently (1-10 µM) inhibited the CTGF expression in primary TGFβ1-activated HSCs, as determined by Western blot and immunofluorescent staining ( Figure 1A,B). The TGFβ1-induced proliferation of primary HSCs was also significantly restricted by BAY 41-2272 (10 µM; Figure 1C), suggesting that the compound may serve as an agent to ameliorate HSC activation.

The BAY 41-2272-Inhibited CTGF Expression and Cell Proliferation Was not via sGC/cGMP Pathway in TGFβ1-Activated Primary HSCs
To examine whether the inhibitory effects of BAY 41-2272 on HSC activation is through the activated sGC/cGMP signaling, we checked the CTGF expression in TGFβ1-activated primary HSCs in the presence or absence of inhibitors, ODQ (sGC inhibitor), and KT5823 (cGMP-dependent protein kinase (PKG) inhibitor). Surprisingly, the BAY 41-2272-repressed mRNA and protein expressions of CTGF were not affected by ODQ (1 µM) or KT5823 (3 µM) in TGFβ1-activated HSCs ( Figure 3). Moreover, the NO donors SNAP (0.1 µM) and SNP (0.1 µM) did not alter the inhibitory effects of BAY 41-2272 on the TGFβ1-induced mRNA and protein expression of CTGF in primary HSCs ( Figure 4A-C). The BAY 41-2272-repressed cell proliferation was not changed in the presence of SNAP or SNP in TGFβ1-activated primary HSCs ( Figure 4D), suggesting that the effects of BAY 41-2272 against HSC activation may be independent on its canonical sGC/cGMP signaling.

The BAY 41-2272-Inhibited CTGF Expression and Cell Proliferation Was not via sGC/cGMP Pathway in TGFβ1-Activated Primary HSCs
To examine whether the inhibitory effects of BAY 41-2272 on HSC activation is through the activated sGC/cGMP signaling, we checked the CTGF expression in TGFβ1-activated primary HSCs in the presence or absence of inhibitors, ODQ (sGC inhibitor), and KT5823 (cGMP-dependent protein kinase (PKG) inhibitor). Surprisingly, the BAY 41-2272-repressed mRNA and protein expressions of CTGF were not affected by ODQ (1 μM) or KT5823 (3 μM) in TGFβ1-activated HSCs ( Figure 3). Moreover, the NO donors SNAP (0.1 μM) and SNP (0.1 μM) did not alter the inhibitory effects of BAY 41-2272 on the TGFβ1-induced mRNA and protein expression of CTGF in primary HSCs ( Figure  4A-C). The BAY 41-2272-repressed cell proliferation was not changed in the presence of SNAP or SNP in TGFβ1-activated primary HSCs ( Figure 4D), suggesting that the effects of BAY 41-2272 against HSC activation may be independent on its canonical sGC/cGMP signaling.   pathway in TGFβ1-activated HSCs. HSCs were pretreated with or without ODQ (1 μM) or KT5823 (cGMP-dependent protein kinase (PKG) inhibitor) (3 μM) PKG inhibitors for 15 min. HSCs were sequentially incubated with BAY 41-2272 (10 μM) for 30 min before activation by TGFβ1 (5 ng/mL) for another 6 h. (A) mRNA levels of CTGF or GAPDH were determined by quantitative RT-PCR. (B,C) Expressed CTGF and GAPDH proteins were determined by Western blot using the corresponding antibodies. All data are expressed as mean ± SD (n = 3). *** p < 0.001 compared with the TGFβ1 alone.  cGMP-dependent phosphodiesterases (PDEs; PDE1, 2, 3, 5, 6, 9, 10, and 11) are responsible for the conversion from active cGMP to inactive 5 GMP [33]. To further understand the effect of the sGC/cGMP pathway on HSC activation, we used PDE inhibitors to examine the cGMP production and CTGF expression in primary HSCs. First, the non-specific PDE inhibitor 3-isobutyl-1-methylxanthine (IBMX) apparently increased the BAY 41-2272-induced intracellular cGMP levels in the presence or absence of SNAP or SNP. The cGMP levels were also inhibited by ODQ ( Figure 5A). We also screened the isozyme-specific inhibitors of PDEs (vinpocetine for PDE1, erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA) for PDE2, zaprinast and tadalafil for PDE5, BAY 73-6691 for PDE9) to check the BAY 41-2272-induced cGMP production in HSCs. The PDE9 inhibitor BAY 73-6691 showed the strongest effect on increasing the BAY 41-2272-induced cGMP levels ( Figure 5B), suggesting that PDE9 is important for the sGC/cGMP pathway in HSCs. However, the PDE9 inhibitor did not change the inhibitory effect of BAY 41-2272 on CTGF expression in TGFβ1-activated HSCs ( Figure 5C). This demonstrates further that the inhibitory effect of BAY 41-2272 on CTGF expression is via the sGC/cGMP-independent pathway. PDE9) to check the BAY 41-2272-induced cGMP production in HSCs. The PDE9 inhibitor BAY 73-6691 showed the strongest effect on increasing the BAY 41-2272-induced cGMP levels ( Figure 5B), suggesting that PDE9 is important for the sGC/cGMP pathway in HSCs. However, the PDE9 inhibitor did not change the inhibitory effect of BAY 41-2272 on CTGF expression in TGFβ1-activated HSCs ( Figure 5C). This demonstrates further that the inhibitory effect of BAY 41-2272 on CTGF expression is via the sGC/cGMP-independent pathway. PDE9 inhibitor for 15 min. HSCs were sequentially incubated with BAY 41-2272 (10 μM) for 30 min before activation by TGFβ1 (5 ng/mL) for another 6 h. Expressed CTGF and GAPDH were measured by Western blot using the corresponding antibodies. All data are expressed as mean ± SD (n = 4). * p < 0.05, ** p < 0.01, *** p < 0.001 compared with the basal (A,B) or TGFβ1 alone (C). HSCs were sequentially incubated with BAY 41-2272 (10 µM) for 30 min before activation by TGFβ1 (5 ng/mL) for another 6 h. Expressed CTGF and GAPDH were measured by Western blot using the corresponding antibodies. All data are expressed as mean ± SD (n = 4). * p < 0.05, ** p < 0.01, *** p < 0.001 compared with the basal (A,B) or TGFβ1 alone (C).

The TGFβ1-Induced CTGF Expression is Independent of cGMP Formation in Primary HSCs
The sGC stimulator BAY 41-2272 has the potential of inhibiting HSC activation by prohibiting CTGF in TGFβ1-activated primary HSCs; however, this is separate from its ability to elicit sGC activity and cGMP formation. To further characterize the correlation between sGC/cGMP signaling and pro-fibrotic CTGF expression, we used another heme-dependent sGC stimulator, BAY 63-2521, and a heme-independent sGC activator, BAY 58-2667, to analyze the cGMP formation and TGFβ1-induced CTGF expression in primary HSCs. BAY 63-2521 (10 µM) alone increased the intracellular cGMP generation in HSCs that was further increased by synergetic incubation with NO donors SNP (0.1 µM) or SNAP (0.1 µM). Moreover, the PDE inhibitor IBMX increased the BAY 63-2521-induced cGMP levels in HSCs with or without SNAP or SNP. All the increased cGMP levels were inhibited by ODQ (1 µM) ( Figure 6A). On the other hand, BAY 58-2667 had no effect on cGMP formation in HSCs, even in the presence of SNAP, SNP, IBMX, and/or ODQ ( Figure 6B,C). These results indicate that heme-independent sGC is the major type of sGC in HSCs and may have unique functions in these cells.
BAY 63-2521 and BAY 58-2667 did not affect CTGF expression in TGFβ1-activated HSCs ( Figure 7A,B). An analog of cGMP, 8-Br-cGMP (bromo-cGMP), also did not alter the TGFβ1-induced CTGF expression in HSCs ( Figure 7C), suggesting that cGMP signaling may not be involved in CTGF production in TGFβ1-activated HSCs. Together, the activated sGC/cGMP signaling exhibits no advantage for blocking HSC activation via preventing CTGF expression. activity and cGMP formation. To further characterize the correlation between sGC/cGMP signaling and pro-fibrotic CTGF expression, we used another heme-dependent sGC stimulator, BAY 63-2521, and a heme-independent sGC activator, BAY 58-2667, to analyze the cGMP formation and TGFβ1induced CTGF expression in primary HSCs. BAY 63-2521 (10 μM) alone increased the intracellular cGMP generation in HSCs that was further increased by synergetic incubation with NO donors SNP (0.1 μM) or SNAP (0.1 μM). Moreover, the PDE inhibitor IBMX increased the BAY 63-2521-induced cGMP levels in HSCs with or without SNAP or SNP. All the increased cGMP levels were inhibited by ODQ (1 μM) ( Figure 6A). On the other hand, BAY 58-2667 had no effect on cGMP formation in HSCs, even in the presence of SNAP, SNP, IBMX, and/or ODQ ( Figure 6B,C). These results indicate that heme-independent sGC is the major type of sGC in HSCs and may have unique functions in these cells. BAY 63-2521 and BAY 58-2667 did not affect CTGF expression in TGFβ1-activated HSCs ( Figure  7A,B). An analog of cGMP, 8-Br-cGMP (bromo-cGMP), also did not alter the TGFβ1-induced CTGF expression in HSCs ( Figure 7C), suggesting that cGMP signaling may not be involved in CTGF production in TGFβ1-activated HSCs. Together, the activated sGC/cGMP signaling exhibits no advantage for blocking HSC activation via preventing CTGF expression. Expressed CTGF and GAPDH were determined by Western blot using the corresponding antibodies. All data are expressed as mean ± SD (n = 3).

Discussion
The fibrogenic cytokines TGFβ and CTGF trigger HSC activation as a pathogenic factor during liver fibrosis. Emerging evidence shows that stopping TGFβ-CTGF upregulation and sGC modulators are effective ways to control HSC activation [11,16,20,23]; however, the full understanding of their correlation is still elusive. BAY 41-2272 is a well-described sGC stimulator that is used to ameliorate fibrotic lesion [24][25][26][27][28][29], but its application in HSC remains undefined. In the present study, we manifested the pharmacological effects and mechanisms of BAY 41-2272 and sGC/cGMP signaling in TGFβ-activated primary mouse HSCs. BAY 41-2272 remarkably restricted the TGFβ1-induced CTGF upregulation and cell proliferation in mouse primary HSCs, along with increased cGMP levels and sGC activity. However, the inhibitory effects of BAY 41-2272 on TGFβ1- levels were determined by Western blot using antibodies against p-Smad2, p-Smad3, Smad2/3, p-Akt (S473 or T308), Akt, CTGF, or GAPDH. (E) Cell viability was measured using the WST-1 assay and monitored spectrophotometrically at 450 nm. All data are expressed as mean ± SD (n = 3). * p < 0.05, ** p < 0.01, *** p < 0.001 compared with the basal (B) or TGFβ1 alone (C-E). # p < 0.05, ## p < 0.01 compared with the corresponding DMSO group.

Discussion
The fibrogenic cytokines TGFβ and CTGF trigger HSC activation as a pathogenic factor during liver fibrosis. Emerging evidence shows that stopping TGFβ-CTGF upregulation and sGC modulators are effective ways to control HSC activation [11,16,20,23]; however, the full understanding of their correlation is still elusive. BAY 41-2272 is a well-described sGC stimulator that is used to ameliorate fibrotic lesion [24][25][26][27][28][29], but its application in HSC remains undefined. In the present study, we manifested the pharmacological effects and mechanisms of BAY 41-2272 and sGC/cGMP signaling in TGFβ-activated primary mouse HSCs. BAY 41-2272 remarkably restricted the TGFβ1-induced CTGF upregulation and cell proliferation in mouse primary HSCs, along with increased cGMP levels and sGC activity. However, the inhibitory effects of BAY 41-2272 on TGFβ1-induced HSC activation and induction of sGC/cGMP signaling are astonishingly unconnected. BAY 41-2272 restricted the Akt activation instead of canonical Smad2/3 pathway in TGFβ1-activated HSCs, providing a plausible cellular basis for the mechanistic actions of BAY 41-2272 and Akt for HSC activation.
A growing body of evidence indicates that reversing to quiescent status or preventing proliferation is an effective way to treat activated HSCs [4,36,37]. TGFβ has been well-established for inducing HSCs into an activated status and becoming, in turn, myofibroblast-like cells with proliferative, contractile, and fibrogenic properties [16][17][18]. TGFβ signaling upregulates another fibrogenic factor, CTGF, which contributes to HSC activation [20,38,39]; inhibition of CTGF is an option for attenuating HSC activation [40][41][42]. In this study, we used isolated primary HSCs to address the possibility of targeting TGFβ1-induced CTGF expression. Both RNA and protein levels of CTGF are upregulated in TGFβ1-activated HSCs and hence the inhibitory effects of BAY 41-2272 on CTGF upregulation and HSC proliferation should be relevant to its anti-fibrotic effect on HSC activation (Figures 1 and 3). Therefore, targeting CTGF upregulation may be an effective way of attenuating TGFβ-induced HSC activation.
The NO/sGC/cGMP pathway maintains various physiological homeostasis, and induction of sGC/cGMP signaling has been proposed to inhibit fibrogenesis [11,43]. However, the role of NO in regulating HSCs remains controversial. For example, the nitrovasodilator-mediated contraction and proliferation of HSCs are both NO/cGMP-dependent and -independent [14,44,45]. Lipopolysaccharide (LPS) triggers HSCs to release NO, but this is not related to HSC-induced hepatocyte proliferation [46,47]. Moreover, release of NO inhibits HSC activation [48]. Therefore, NO signaling may show diverse functions during HSC activation. SNP and SNAP act as NO donors via enzymatic oxidation and chemical reaction, respectively [49]. Both SNP and SNAP synergistically potentiated the BAY 41-2272-induced sGC activities and cGMP levels in HSCs ( Figure 2); however, SNP and SNAP did not affect the BAY 41-2272-inhibited CTGF upregulation and proliferation of HSCs ( Figure 4). We suggest that NO production may not participate in TGFβ1-induced HSC activation, although it augments sGC/cGMP signaling. sGCs are expressed in HSCs and not in hepatocytes [50] and are divided into reduced and oxidized forms that possess heme-dependent and heme-independent properties, respectively [11,13]. Many sGC modulators have been developed: (1) sGC stimulators to sensitize the reduced and heme-containing sGC to NO, and (2) sGC activators to activate oxidized and heme-free sGC [51][52][53] to generate cGMP. Exposure of 8-Br-cGMP cGMP analog leads to an inhibition of TGFβ-induced fibrogenesis in renal, cardiac, and dermal fibroblasts [26,54,55], suggesting that the sGC/cGMP pathway may be preferential signaling for TGFβ-induced HSC activation. Here, the BAY 41-2272 sGC stimulator showed potent inducible and inhibitory effects on cGMP levels and TGFβ1-increased CTGF expression in HSCs (Figures 1 and 2). However, only increased cGMP levels and sGC activity but not decreased CTGF expression was reversed by the ODQ sGC inhibitor in BAY 41-2272-treated HSCs (Figures 2 and 3). Importantly, the elevation of cGMP signaling by KT5823 (PKG inhibitor), BAY 73-6691 (PDE9 inhibitor), and 8-Br-cGMP (cGMP analog) failed to alter the TGFβ1-induced CTGF expression in HSCs ( Figure 3C, Figure 5C, and Figure 7C). Together, our results may support the existence of a unique framework making TGFβ-based HSC activation independent of sGC/cGMP signaling.
In general, TGFβ binding with the TGFβ receptor leads to phosphorylation of the receptor-activated Smad2 and Smad3. Additionally, non-canonical Smad-independent signaling pathways such as PI3K/Akt are activated by TGFβ and provide a broad TGFβ-induced intracellular crosstalk [18,19]. Targeting PI3K/Akt signaling also restrains the progression of HSC activation, including cell proliferation and CTGF expression [34,35,[60][61][62], supporting the biological significance of the PI3K/Akt pathway for TGFβ-dependent HSC activation. In this study, BAY 41-2272 reduced the TGFβ1-induced phosphorylation of Akt but not canonical Smad2/3. Pharmacological inhibition of PI3K/Akt also apparently blocked the TGFβ1-induced CTGF expression and proliferation of HSCs ( Figure 8). Because the effect of BAY 41-2272 on Akt phosphorylation was transient, we cannot exclude the possibility that the inhibitory effects of BAY 41-2272 in HSCs are mediated by other targets. This is the first time that BAY 41-2272 was explored to treat TGFβ-mediated HSC activation via targeting the PI3K/Akt that plays an important role in this process.

Conclusions
We illustrate that the BAY 41-2272 sGC stimulator attenuates the TGFβ1-induced CTGF expression and cell proliferation through the Akt signaling and not the sGC/cGMP pathway in mouse primary HSCs. Our results provide a novel insight into the TGFβ1-and BAY 41-2272-based regulatory networks and molecular profiles for HSC activation.