Wogonin was purchased from Spring & Autumn Biotec Co., Ltd. (Nanjing, China). Isoprenaline was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). All-trans-retinoic acid (RA), DBcAMP, and phorbol 12, 13-dibutyrate (PDBU) were purchased from Sigma-Aldrich LLC. (Shanghai, China). MG132 was obtained from Selleck Chemical (Houston, TX, United States). The vectors pUSEamp(+)/myc-tagged Akt (constitutively active, CA) and pUSEamp(+)-Pik3ca were kindly provided by Liangyou Rui from the University of Michigan. The vectors pcDNA3HA and pGL3-Basic were provided by Dongping Wei from Nanjing First Hospital. The empty vector pAdeno-MCMV-MCS-3Flag and pDONR223 vector carrying a human Nedd4l gene were purchased from Obio Technology Corp., Ltd. (Shanghai, China) and Public Protein/Plasmid Library (Nanjing, China), respectively. The primers 5′-TCGAGCTCAAGCTTCGAATTCATGGAGCGACCCTATACATTT-3′ and 5′-GTCATCCTTGTAGTCGGATCCATCCACCCCTTCAAATCCTT-3′ were used to subclone human Nedd4l cDNA from pDONR223-Nedd4l by PCR. The PCR products were recombined with pAdeno-MCMV-MCS-3Flag vector cut by EcoR1/BamH1 to obtain the expression vector pAdeno-MCMV-MCS-3Flag-Nedd4l. Pik3ca cDNA was subcloned from pUSEamp(+)-Pik3ca using the primers 5′-GATCCCCCGGGCTGCAGGAATTCATGGGGAGCAGCAAGAGCAAG-3′ and 5′-ATAGAATAGGGCCCCCCCTCGAGTCAGTTCAAAGCATGCTG-3′. The PCR products were recombined with pcDNA3HA vector cut by EcoR1/Xho1 to obtain the expression vector pcDNA3HA-Pik3ca.

Male ICR mice were purchased from Model Animal Research Center of Nanjing University. They were housed in a pathogen-free barrier facility with a 12-h light/dark cycle and given free access to food and water. Eight-week-old mice (n = 29) were divided into five groups as indicated (Figure 2). Under inhalation anesthesia by isoflurane, an osmotic minipump (model 2004D; Alzet, Durect Corp, Cupertino, CA, United States) was implanted subcutaneously on the back of the neck (Krenek et al., 2009; Heap et al., 2015) and the delivery of isoprenaline (5 mg /kg/day) was started. Immediately after implantation, wogonin was administered by intraperitoneal injection for 2 weeks at a dose of 1 mg/kg or 10 mg/kg once daily. All animal experiments complied with the guidelines of the Nanjing Medical University Regulations of Animal Experiments and were approved by the Animal Experiment Committee of the Nanjing Medical University.

Rat fetal cardiomyocytes (H9c2) were obtained from American Type Collection (ATCC, Manassas, VA, United States). H9c2 cells were cultured in Dulbecco’s modified Eagle’s high-glucose medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (Hyclone) in a humidified incubator in an atmosphere of 5% CO₂ at 37°C. When the cell confluence reached to 70–80%, cell culture medium was switched to DMEM containing 1% FBS and 1 μM RA for further 5-day culture before experimental treatments.

After the 2-week medication, mice were anesthetized with isoflurane for echocardiographic examination. The images were obtained using a Vevo 2100 system with a 45 MHz probe (Visualsonics, Toronto, ON, Canada) to evaluate the cardiac function and chamber size. M-mode tracings were used to measure interventricular septum diameter (IVSd), left ventricular posterior wall diameter (LVPWd), left ventricular end-diastolic diameter (LVIDd), and left ventricular end-systolic diameter (LVIDs). Ejection fraction (EF, %) and left ventricular (LV) mass were calculated as left ventricular end-systolic volume (LVESV) = 7.0/(2.4 + LVIDs)^∗LVIDs³; left ventricular end-diastolic volume (LVEDV) = 7.0/(2.4 + LVIDd)^∗ LVIDd³. EF = 100^∗((LVEDV-LVESV)/LVEDV). LV Mass = 1.053^∗((LVIDd+LVPWd+IVSd)³-LVIDd³).

Mouse hearts were fixed in 4% paraformaldehyde solution for 48 h, dehydrated in ascending grades of ethanol, and embedded in wax. Heart sections were prepared for routine haematoxylin and eosin (H&E) staining. H9c2 cells were grown on the glass coverslips for immunofluorescence staining. Cells were washed twice with PBS, fixed for 10 min in 4% paraformaldehyde solution, and permeabilized with 0.1% Triton X-100 plus 1% bovine serum albumin (BSA) for 1 h. Then the cells were incubated with alpha-smooth muscle actin (α-SMA) antibody (1:100 dilution) for 4 h at 37°C followed by staining with a 1:100 dilution of 4′,6-diamidino-2-phenylindole (DAPI) for 10 min. A model IX72 fluorescence inverted microscope (Olympus, Tokyo, Japan) was used to observe the cells. α-SMA antibody was purchased from Abcam (Cambridge, United Kingdom). Antibody against DAPI was purchased from Sigma-Aldrich (St. Louis, MO, United States).

Total RNA extracted from mouse heart tissues and H9c2 cells were reverse-transcribed as previously described (Chen et al., 2016). qPCR was performed using SYBR Premix Master Mix (Thermo Fisher Scientific Inc., Shanghai, China). Relative mRNA levels of target genes were quantified using comparative threshold (CT) normalized to house-keeping genes [ribosomal protein, large, P0 (36B4) for mouse heart or β-actin for H9c2 cells]. The primer sequences used for qPCR are demonstrated in Table 1.

As previously described (Chen et al., 2016), extracts from mouse hearts or cells were immunoblotted using primary antibodies against Nedd4l, Pik3ca, p-Akt, Akt, cAMP response elements binding (CREB), phospho (p)-CREB, c-jun N-terminal kinase (JNK), p-JNK, p-p38, p38, extracellular signal-regulated kinase 1/2 (ERK1/2), p-ERK1/2, forkhead box protein O1 (FoxO1), p-FoxO1, FoxO3a, p-FoxO3a, ubiquitin, G-protein-coupled receptor kinase 2 (GRK2), and β-actin (all from Cell Signaling Technologies, Beverly, MA, United States). Blots were incubated with secondary antibody conjugated with horseradish peroxidase (Cell Signaling Technologies) and visualized using an ECL kit (Millipore, Billerica, MA, United States). The chemiluminiscence signals were detected by a model 4200SF device (Tanon Shanghai, China).

The -2000 to +1 promoter region of human Nedd4l was amplified by PCR with genomic DNA from HepG2 cells using primers 5′-GGTACCGAGCTCTTACGCGTAGGTAGAGAAGCACTGACTCC-3′ and 5′-CTTAGATCGCAGATCTCGAGCGGCCGGGCTTTCC-3′. The PCR products were recombined with pGL3-Basic vector cut by Mlu1/Xho1 to obtain the reporter construct pGL3–2000/+1LUC. H9c2 cells were transfected with luciferase reporter plasmids plus the internal control vector pRL-TK-Renilla for luciferase assays in 24-well plates. After an overnight culture, cells were incubated for 24 h in complete medium supplemented with wogonin (10 μM). Luciferase activity was measured using the dual-luciferase reporter assay system (Promega, Madison, WI, United States) and normalized to Renilla luciferase.

The data and statistical analyses in this study comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2015). Data are presented as means ± SE. Data between groups were analyzed by Student’s t-test or one-way ANOVA followed by LSD–Dunn multiple comparisons. p < 0.05 was considered statistically significant.

To evaluate the anti-hypertrophic effect of wogonin, myocardial hypertrophy in mice was replicated by subcutaneous implantation of mini-pump delivering isoprenaline for 14 days. Wogonin failed to prevent the isoprenaline-mediated reduction in body weight (BW) but did reverse the abnormal increase in heart weight (HW) (Table 2 and Figure 2A). Wogonin alone had no effects on BW or HW. The detailed anatomical changes in the heart were detected by echocardiography and reflected via hypertrophic indexes including HW, HW/BW, IVSd, LVPWd, LV Mass, LVIDd, and LVIDs. EF is to evaluate the left ventricular systolic function. Isoprenaline treatment increased HW, HW/BW, IVSd, LVPWd, and LV Mass but decreased LVIDd and LVIDs. Those changes, showing typical for concentric hypertrophy, were all reversed by wogonin. EF was neither worsened by isoprenaline nor affected by wogonin. Histologic analysis of heart section slices also revealed that wogonin attenuated the enlargement in heart size after isoprenaline treatment (Figures 2A,B). Atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) are two well-known natriuretic peptides secreted by cardiac muscle cells when hypertrophy occurs (Vasan, 2006). Thus, the expressions of ANP and BNP in heart were determined as the important biomarkers for myocardial hypertrophy. As shown in Figure 2C, wogonin relieved isoprenaline-induced increase in mRNA levels of ANP and BNP in a dose-dependent manner but wogonin alone did not affect their expression.

The H9c2 cell line was originally derived from embryonic rat ventricular tissue. These cells are widely used as an in vitro tool to investigate pathomechanism of cardiomyocytes, including cardiac hypertrophy (Watkins et al., 2011). H9c2 cells are undifferentiated myoblasts. The addition of RA drives cell differentiation toward a cardiac phenotype (Branco et al., 2015). The increased mRNA levels of Tnnc1 and Tnnt2, the cardiac-specific genes, indicated the success in cell differentiation (Figure 3A). Myocardial hypertrophy was replicated by isoprenaline treatment in differentiated H9c2 cells. The mRNA levels of ANP and BNP were also investigated as hypertrophic markers. Compared to the dose of 1 μM, wogonin (10 μM) more effectively reduced the cellular mRNA levels of ANP and BNP which had been enhanced by isoprenaline treatment (Figure 3B). In cardiomyocytes, α-SMA is a major constituent of the cytoskeleton, which determines the cell shape. Thus, staining of α-SMA was performed to observe the morphology of H9c2 cells. The average area of H9c2 cells was significantly expanded by isoprenaline treatment, while wogonin relieved this effect (Figures 3C,D).

Isoprenaline induces cardiomyocyte hypertrophy via multiple pathways, including PI3K/Akt, cAMP/PKA/CREB, MAPKs (JNK, p38, and ERK), and PKC (Heineke and Molkentin, 2006). The phosphorylation levels of Akt, CREB, JNK, p38, and ERK were significantly increased by isoprenaline. Wogonin reversed the phosphorylation levels of Akt and CREB (Figures 4A,B). CREB phosphorylation was mediated by two pathways, including cAMP/PKA and PI3K/Akt, during isoprenaline treatment (Bullock and Habener, 1998; Khalilimeybodi et al., 2017). Wogonin did not affect CREB phosphorylation induced by dibutyl cyclic adenosine acid (DB-cAMP), the analog of cAMP (Figures 4C,D), implying that wogonin reduces CREB phosphorylation via the PI3K/Akt pathway instead of the cAMP/PKA pathway. PKCs were activated by its selective agonist, PDBU, which increased the mRNA levels of ANP and BNP. However, wogonin did not affect PKC-induced transcription of ANP and BNP, indicating that the anti-hypertrophy effect of wogonin is not involved in the PKC pathway (Figure 4E).

FoxO1, FoxO3a, and CREB mediate PI3K/Akt pathway-induced gene expression involved in myocardial hypertrophy (Ronnebaum and Patterson, 2010; Khalilimeybodi et al., 2017). We observed that isoprenaline increased the phosphorylation levels of Akt, and the downstream FoxO1, FoxO3a, and CREB, which were all inhibited by wogonin (Figures 4A,B, 5A,B). Thus, we reasoned that wogonin specifically acts on the PI3K/Akt signaling pathway. The fact that constitutively active Akt [Akt (CA)] transfected into H9c2 cells significantly enhanced the phosphorylation of FoxO1 and FoxO3a, which were not affected by wogonin, implies that the target of wogonin is located in the upstream of Akt (Figures 5C,D). As class IA PI3K, one of three types of PI3Ks, plays an important role in cardiac growth and function (Shioi et al., 2000), we overexpressed Pik3ca, the catalytic subunit of class IA PI3K, into H9c2 cells to initiate the phosphorylation of Akt. Interestingly, wogonin reversed the Akt phosphorylation by reducing the level of Pik3ca protein (Figures 5E,F). Consistently, α-SMA staining showed that Pik3ca overexpression amplified the size of H9c2 cells and wogonin reversed this pathological change (Figures 5G,H). When catecholamine binding to β-adrenoceptors, the G protein–coupled receptor kinase-2 (GRK2) mediates the translocation of PI3K to β-adrenoceptors and then enhances the recruitment of β-arrestin and AP-2, which finally results in the internalization and downregulation of β-adrenoceptors (Naga Prasad et al., 2002). It has reported that disrupts the interaction between PI3K and GRK2 by displacing class I PI3K isoforms blocks agonist-stimulated β-adrenoceptors internalization (Nienaber et al., 2003). To test whether wogonin potentially affects β-adrenoceptor function by modulating the interaction between PI3K and GRK2, precipitation was performed. As shown in Supplementary Figure S1, wogonin treatment reduced the binding of GRK2 to Pik3ca in H9c2 cells, mostly due to wogonin-induced reduction in Pik3ca expression.

Pik3ca was significantly increased at the protein and mRNA levels in H9c2 cells with isoprenaline treatment (Figures 6A,B). Wogonin reduced the protein level of Pik3ca but failed to decrease its mRNA level (Figures 6A,B). We reasoned that wogonin accelerates the degradation of Pik3ca protein. The classic protein degradation process depends on the ubiquitination-proteasome system, which includes protein-ubiquitination by ubiquitin ligase and subsequent protein-recognization by proteasomes for retrogradation. Thus, we utilized the proteasome inhibitor MG132 to prevent the ubiquitinated proteins from proteasome-mediated degradation. Wogonin induced an enhancement in the ubiquitination of plenty of proteins, which should include Pik3ca, since MG132 reversed the wogonin-induced downregulation of Pik3ca (Figure 6A). To confirm the specificity of the wogonin-mediated ubiquitination of Pik3ca, hemaglutinin (HA)-tagged Pik3ca was transfected into H9c2 cells. The cells were then treated with wogonin and MG132. The original protein amounts of HA-tagged Pik3ca were equal in each condition and immunoprecipitation was easier to operate with HA-tagged beads. As shown in Figure 6C, the precipitated Pik3ca displayed higher ubiquitination levels after wogonin treatment.

Nedd4l is the specific ubiquitin ligase of Pik3ca, belonging to the Nedd4 (neural precursor cell-expressed developmentally down-regulated gene 4) family (Yang and Kumar, 2010; Wang Z. et al., 2016). Isoprenaline treatment significantly reduced the mRNA and protein levels of Nedd4l in H9c2 cells (Figures 7A–C), which led to the upregulation in protein level of Pik3ca (Figure 6A). Besides, isoprenaline also increased the mRNA level of Pik3ca (Figure 6B). Wogonin reversed the inhibitory effect of isoprenaline on Nedd4l expression at the transcriptional and post-transcriptional levels (Figures 7A–C). We expressed Flag-tagged Nedd4l in H9c2 cells by transient transfection and found that wogonin treatment did not affect the protein levels of exogenous Nedd4l (Figures 7D,Eindicating that wogonin regulates Nedd4l expression only at the transcriptional level. This was confirmed in the luciferase assay, as wogonin significantly enhanced the Nedd4l promoter-driven luciferase activity (Figure 7F). In line with the data from cells, wogonin treatment significantly reversed the downregulation by isoprenaline on protein level of Nedd4l in mouse heart (Figures 7G,H).