Inhibition of PHLPP1 ameliorates cardiac dysfunction via activation of the PI3K/Akt/mTOR signalling pathway in diabetic cardiomyopathy.

Abstract Background Pleckstrin homology (PH) domain leucine‐rich repeat protein phosphatase 1 (PHLPP1) is a kind of serine/threonine phosphatase, whose dysregulation is accompanied with numerous human diseases. However, its role in diabetic cardiomyopathy remains unclear. We explored the underlying function and mechanism of PHLPP1 in diabetic cardiomyopathy (DCM). Method In vivo, Type 1 diabetic rats were induced by intraperitoneal injection of 60 mg/kg streptozotocin (STZ). Lentivirus‐mediated short hairpin RNA (shRNA) was used to knock down the expression of PHLPP1. In vitro, primary neonatal rat cardiomyocytes and H9C2 cells were incubated in 5.5 mmol/L glucose (normal glucose, NG) or 33.3 mmol/L glucose (high glucose, HG). PHLPP1 expression was inhibited by PHLPP1‐siRNA to probe into the function of PHLPP1 in high glucose‐induced apoptosis in H9c2 cells. Results Diabetic rats showed up‐regulated PHLPP1 expression, left ventricular dysfunction, increased myocardial apoptosis and fibrosis. PHLPP1 inhibition alleviated cardiac dysfunction. Additionally, PHLPP1 inhibition significantly reduced HG‐induced apoptosis and restored PI3K/AKT/mTOR pathway activity in H9c2 cells. Furthermore, pretreatment with LY294002, an inhibitor of PI3K/Akt/mTOR pathway, abolished the anti‐apoptotic effect of PHLPP1 inhibition. Conclusion Our study indicated that PHLPP1 inhibition alleviated cardiac dysfunction via activating the PI3K/Akt/mTOR signalling pathway in DCM. Therefore, PHLPP1 may be a novel therapeutic target for human DCM.


| INTRODUC TI ON
Diabetes mellitus is one of the most important events to cause mortality and morbidity all over the world, 1 representing a global burden on human health and economics. Over half of diabetics die from cardiovascular disease, including diabetic cardiomyopathy (DCM). 2 DCM, unique to other complications such as hypertension as well as coronary artery disease, 3,4 is distinguished by left ventricular (LV) hypertrophy, LV systolic and diastolic dysfunction. [5][6][7] The pathogenesis of diabetic cardiomyopathy is complicated, including oxidative stress, 8,9 interstitial fibrosis 10 and myocardial apoptosis. 2 Above all, chronic hyperglycaemia is the most prominent pathophysiological stimuli involved in the development of DCM. 11,12 However, the molecular mechanisms of chronic hyperglycaemia-induced cardiomyocyte apoptosis remain to be elucidated.
Pleckstrin homology (PH) domain leucine-rich repeat protein phosphatase 1 (PHLPP1), a newfound serine/threonine phosphatases, is a member of the type 2C phosphatase (PP2C) family. 13 It is well known that PHLPP1 plays a significant role in suppressing cell survival. [14][15][16][17] And mounting evidence support the fact that PHLPP1 could directly dephosphorylate the hydrophobic motif of Akt (Ser473) which will lead to the inhibition of kinase activity and suppress the occurrence and progression in several cancers, such as colon cancer, 14,18 prostate cancer, 15 pancreatic cancer 19 and lymphoma. 20 Moreover, a previous study stated that impaired insulin action could increase PHLPP1 level in the adipose and muscle tissues of obese patients. 21 While many studies have studied the function of PHLPP1, the role of PHLPP1 in DCM has not been reported yet.
The phosphoinositide 3-kinase (PI3K)/protein kinase B (Akt)/ mammalian target of rapamycin (mTOR) signalling pathway is central to regulating cell transcription, metabolism, survival and inflammation. 22,23 It is well established that during diabetes, defective insulin signalling down-regulates the activity of PI3K, 24 leading to Akt and mTOR inactivation. Previous studies have demonstrated that Akt is closely related to PHLPP1 in cell survival and inflammation. 25 However, the role of the PHLPP1 in PI3K/Akt/mTOR signal pathway in the process of DCM has not been investigated.
Here, we hypothesized that PHLPP1 inhibition is likely to have a protective effect in DCM. To make clear the role of PHLPP1 in the development of DCM, we investigated in primary rat cardiomyocytes as well as H9c2 cells induced by high glucose in vitro and in a DCM rat model in vivo.

| Animals
Sixty male 4-week-old Sprague Dawley (SD) rats purchased from Beijing Weitong Lihua Experimental Animal Technology (Beijing, China) were housed at 22 ± 2°C under an alternating 12-hour light/ dark cycle with free access to water and rat chow. After adaptive feeding with standard experimental rat chow for 1 week, all rats were randomly grouped into the following four (n = 15 for each group): control, diabetes mellitus (DM), DM + shRNA-negative control (shN.C) and DM + shRNA-PHLPP1. All diabetic rats were induced by a single intraperitoneally injection of high-dose streptozotocin (60 mg/kg, STZ, Solarbio) dissolved in citrate buffer (0.5 mL, pH 4.5), and the control group was administered with the same volume of citrate buffer. One week after STZ administration, glucometer (Roche) was used to measure tail vein glucose levels. Only rats with blood glucose ≥16.7 mmol/L were considered as type 1 diabetic rats. Twelve weeks later, an amount of 1 × 10 8 UT/50 μL of lentivector with PHLPP1 shRNA (GenePharma) or 50 μL lentivehicle (GenePharma) was injected into the jugular vein. Four weeks after PHLPP1 shRNA injection, rats were anaesthetized with 3% pentobarbital sodium intraperitoneally (35 mg/kg) and then killed. Hearts were weighted and immersion fixed in formalin for the following histology analysis. All experiments conformed to animal protocols approved by the Shandong University Animal Care Committee.

| Cardiac function measurement
Rats were anaesthetized by inhalation of isoflurane gas (2.5%) until immobile in the imaging laboratory and kept in anaesthesia with 1.5% isoflurane by a nose cone connected to the anaesthesia machine.
Cardiac function was evaluated with the Vevo770 imaging system (VisualSonic) before lentivirus infection and before killing. The left ventricular ejection fraction (LVEF), fractional shortening (FS), peak E, peak A, early (E′), late (A′) and left ventricular end-diastolic dimension (LVEDd) were measured. The ratio of early-to-late mitral flow velocity (E/A) and diastolic velocity ratio (E′/A′) were calculated.

| Histology and immunohistochemistry
The formalin-fixed and paraffin-embedded rat heart tissues were cut into 5 μm sections for the following analyses. Heart sections were stained with Haematoxylin and eosin (HE) to measure cardiomyocyte width. Masson's trichrome and Sirius red staining was performed to observe interstitial collagen deposition. The cardiomyocyte cell diameter and fibrotic area were quantified by ImageJ software, and 500 cells per rat were measured for analysis. The slides were incubated overnight at 4°C with the specific primary antibodies against PHLPP1 (dilution, 1:50; Proteintech, 22789-1-AP), collagen I (dilution, 1:200; Abcam) and collagen III (dilution, 1:500; Abcam) for immunohistochemistry. The next day, the samples were incubated with secondary antibodies for 30 minutes at 37°C to detect the expression of PHLPP1, collagen I and collagen III.

| TUNEL assay
Apoptotic cells in myocardium were detected with a in situ cell death detection kit (Roche) following the manufacturer's instructions.
H9c2 cells were fixed with 4% paraformaldehyde for 20 minutes at room temperature, incubated in TUNEL reaction mixture for 60 minutes at 37°C and then sealed in Prolong Gold Anti-Fade Reagent with DAPI (Invitrogen). Images were acquired via a fluorescence microscopy (Nikon).

| Cell culture
Primary cardiomyocytes were obtained from neonate SD rats with 8% horse serum and 5% calf serum for 2 hours at 37°C in a humidified atmosphere containing 5% CO 2 . After the incubation, the unattached cells were collected and considered as primary cardiomyocytes, while the attached cells were considered to be myocardial fibroblasts and discarded. After the following 3-4 days, the primary cardiomyocytes were incubated in serum-free DMEM for 8 hours before treatment with individual procedure. H9c2 cardiomyoblasts were passaged every 2 days and seeded in 6-well culture plates. When cell populations reached 50% confluence, the medium was substituted with FBS-free DMEM for 4 hours serum starvation.
After that, cells were incubated in normal glucose (NG; 5.5 mmol/L glucose) or high glucose (HG; 33.3 mmol/L glucose) and harvested at 6, 12, 24 and 48 hours of HG treatment. To explore the potential role of PHLPP1 on PI3K, Akt and mTOR phosphorylation as well as on cardiomyocytes apoptosis, PHLPP1-siRNA plasmids were transfected into H9c2 cardiomyoblasts 24 hours prior to HG stimulation.
Moreover, in order to explore the potential role of PHLPP1 in PI3K/ Akt/mTOR pathway under HG stimulation, a specific PI3K inhibitor LY294002 (MCE, HY-10108, 25 μmol/L) was added to the H9c2 cardiomyoblasts medium 2 hours prior to the HG (33.3 mmol/L) treatment.
After HG groups were incubated with HG for 6 hours, H9c2 cardiomyoblasts were exposed to DCFH-DA (10 μmol/L) for 30 minutes at 37°C. Fluorescence intensity was observed via a fluorescence microscope (Nikon).

| PHLPP1 siRNA production and infection in H9c2 cardiomyoblasts
The expression of PHLPP1 in H9c2 cardiomyoblasts were inhib-

| Western blot analysis
Equal amounts of protein were obtained from freshly dissected rat hearts, primary neonatal cardiomyocytes and H9c2 cardiomyoblasts following the procedure of BCA Protein Assay Kit (Solarbio). And then, the protein was separated through 8% or 10% sodium dode-

| Real time RT-PCR
Total RNA samples were extracted from rat hearts by Trizol reagent (Ambion) following the manufacturer's protocol.
β-actin gene was selected as the endogenous reference using the primes 5′-CTCTGTGTGGATTGGTGGCT-3′ (forward) and 5′-CGCAGCTCAGTAACAGTCCG-3′ (reverse). All experiments were repeated at least three times. The 2-ΔΔCT method was used to calculate the fold changes of β-MHC and BNP.

| Statistical analysis
Each experiment was performed at least 3 times. Data are reported as the means ± standard deviation. Differences between two groups were performed by unpaired t test, and multiple groups involved one-way ANOVA followed by Scheffe's test or Bonferroni's post hoc test or Dunnet's multiple-to-one comparison test. P < .05 was considered as statistically significant. All statistical analyses were carried out using Prism 6.0 (Graphpad) and SPSS 20.0.

| Diabetes increased myocardial PHLPP1 expression and PHLPP1 down-regulation prevented diabetes-induced myocardial remodelling
PHLPP1 protein level in the diabetic rat heart was much higher than that in controls, and the PHLPP1 expression was reduced in shPHLPP1-treated diabetic rat hearts compared with vehicle-treated diabetic rats as demonstrated by Western blotting (P < .05; Figure 1A). HE stain showed that PHLPP1 down-regulation restored the increment of cardiomyocyte cell diameter (P < .05; Figure 1B,E). Moreover, qRT-PCR indicated that diabetes-induced up-regulation of β-MHC and BNP was reduced after shPHLPP1 treatment. (P < .05; Figure 1C,D). Finally, the ratio of heart weight to bodyweight was increased in diabetic rats than controls (P < .05; Table 1). And the ratio of heart weight to bodyweight of shPHLPP1-treated diabetic rats appeared to be lower than that of the untreated diabetic rats, but this difference did not achieve statistical significance (Table 1).

| PHLPP1 down-regulation attenuated cardiac dysfunction in diabetes
Sixteen weeks after diabetes induction, echocardiography showed that LVEF, FS, the E/A ratio and the E′/A′ ratio in DM group was significantly decreased than control group and PHLPP1 knockdown reversed this reduction compared with vehicle group (P < .05) (Figure 2A-E). Moreover, LVEDd was moderate higher in diabetic rats than that in control rats, and PHLPP1 knock-down attenuated wall thickening compared with vehicle group (P < .05) ( Figure 2A,F).
Meanwhile, the level of cleaved caspase-3 as well as Bax/Bcl-2 ratio was also expressively elevated in diabetic group (P < .05) ( Figure 3C,D).

| HG increased the expression of PHLPP1 in primary neonatal rat cardiomyocytes and H9c2 cardiomyoblasts
The PHLPP1 protein level showed a significant increase after 24 hours HG treatment in both primary neonatal cardiomyocytes (P < .05) ( Figure 4A) and H9c2 cardiomyoblasts (P < .05) ( Figure 4B).
Immunofluorescence microscopy revealed that PHLPP1 is located in the nucleus, the cytoplasm and membrane of both primary cardiomyocytes and H9c2 cardiomyoblasts. Furthermore, culturing in HG environment significantly increased PHLPP1 expression ( Figure 4C).

| ROS mediated HG-induced PHLPP1 overexpression
HG treatment significantly induced ROS overproduction (P < .05) ( Figure 5A). To explore the role of ROS in the HG-induced overexpression of PHLPP1 in H9c2 cells, we used a ROS inhibitor, N-acetyl cysteine (NAC, 3 mmol/L) to inhibit the production of ROS.

| Inhibition of PHLPP1 reverted HG-induced p-PI3K, p-AKT and p-mTOR inactivation in the H9c2 cardiomyoblasts
PHLPP1-siRNA plasmids were transfected 24 hours prior to HG stimulation in H9c2 cardiomyoblasts. Western blot analysis suggested that HG treatment significantly reduced PI3K, AKT and mTOR phosphorylation compared with the control group. However, these changes were reversed after inhibition of PHLPP1 (P < .05) ( Figure 7).

| Inhibition of PHLPP1 prevented HG-induced apoptosis via the PI3K/AKT/mTOR pathway in the H9c2 cardiomyoblasts
Twenty-four hours after down-regulating PHLPP1 expression with PHLPP1-siRNA, H9c2 cardiomyoblasts were treated with a PI3K inhibitor, LY294002 (MCE, 25 μmol/L) 2 hours prior to the 24 hours of HG treatment. TUNEL assay showed that the proportion of apoptotic cells was increased in the HG + siN.C + LY294002 group compared with the HG + siN.C group, decreased in the HG + siPHLPP1 group compared with the HG + siN.C group and significantly increased in the HG + siPHLPP1 + LY294002 group compared with the HG + siPHLPP1 group (P < .05) ( Figure 8A). Consistent with these findings, the expression of cleaved caspase-3 and the ratio of Bax to Bcl-2 increased upon LY294002 treatment (P < .05) ( Figure 8B,C).

| D ISCUSS I ON
Diabetes mellitus has been recognized as a significant health problem worldwide. 26  other heart diseases. [27][28][29] The potential pathophysiological mechanisms of diabetic cardiomyopathy including oxidative stress, 8,9 interstitial fibrosis 10 as well as myocardial apoptosis, 2 remain poorly understood.
Pleckstrin homology (PH) domain leucine-rich repeat protein phosphatase 1 (PHLPP1) is a kind of serine/threonine phosphatase, 13 whose dysregulation is associated with numerous human diseases. PHLPP1 is known to directly dephosphorylate the hydrophobic motif of Akt (Ser473) to keep the balance between cell survival and apoptosis in several cancers. 30 Recently, Moc C and colleagues have shown that the deletion of PHLPP1 In our present study, diabetic rat hearts showed elevated PHLPP1 expression. We found the novel result that the down-regulating PHLPP1 expression can improve cardiac function and remodelling, alleviate collagen deposition, attenuate HG-induced cardiomyocytes apoptosis, restore PI3k/Akt/mTOR pathway activity. Thus, PHLPP1 plays a pivotal role in the progression of DCM.
Long-term hyperglycaemia and hyperglycaemia-induced oxidative stress are chief reasons for the development of DCM. 8,35 Additionally, cardiac oxidative stress is associated with cell death and myocardial fibrosis, leading to potential fatal cardiac events. 36 It occurs under the condition of overproduction of ROS. In our current study, high glucose induced overexpression of PHLPP1 in both primary neonatal rat cardiomyocytes and H9c2 cardiomyoblasts.
And high-glucose treatment results in excessive synthesis of ROS In addition to cardiomyocyte apoptosis, intensive cardiac fibrosis is another crucial characteristic of DCM. 41,42 Excess production and deposition of ECM, principally collagen I and III, can alter the structure of the heart and lead to severe cardiac dysfunction. 10,42 F I G U R E 8 TUNEL analyses and Western blotting demonstrated that LY294002 effectively reversed the anti-apoptotic effect of PHLPP1 inhibition. A, The cell apoptosis rates detected by TUNEL were reduced after PHLPP1 inhibition; however, this reduction was reversed after pretreated with LY294002. B, The expression of cleaved caspase3 was reduced after PHLPP1 inhibition, while LY294002 reversed this reduction. C, The ratio of Bax to Bcl2 was reduced after PHLPP1 inhibition but it was enhanced when treatment with LY294002. NG: 5.5 mmol/L glucose; HG: 33.3 mmol/L glucose. siN.C: negative control siRNA. siPHLPP1: PHLPP1 siRNA. All experiments were performed at least 3 times. Data are expressed as the means ± SD. Statistical analysis was performed using one-way ANOVA followed by Bonferroni's post hoc test. *P < .05 compared with HG + siNG. #P < .05 compared with HG + siPHLPP1 And HG stimulates MMP-2 and MMP-9 expression in diabetic hearts which accelerates tissue remodelling and cardiac fibrosis.
Foetal gene re-expression and natriuretic peptides up-regulation indicated reduction of cardiac function decreased. In our study, shPHLPP1 treatment reduced β-MHC re-expression and BNP up-regulation in diabetic ret hearts. And diabetic-induced aberrant deposition of collagen I and III as well as increased expression of MMPs were reduced after shPHLPP1 treatment. However, we have not explored the mechanisms of PHLPP1 inhibition reduced cardiac fibrosis.
The data presented here demonstrate that PHLPP1 gene silencing improves myocardial function and protects DCM from myocardial apoptosis and fibrosis. Moreover, PHLPP1 down-regulation plays a crucial role in hyperglycaemia-induced cardiomyocyte apoptosis via activating the PI3K/AKT/mTOR pathways.
However, the effects and molecular mechanism of PHLPP1 inhibition in myocardial fibroblasts have not been fully elucidated.
Further investigations using myocardial fibroblasts and spontaneous diabetes model, such as db/db mice or gene knockout mice should be needed to uncover these mechanisms. Given the cardio-protective effects of PHLPP1 inhibition, PHLPP1 may be a novel therapeutic target in the treatment of DCM.