LCZ696 ameliorates doxorubicin-induced cardiomyocyte toxicity in rats

Doxorubicin (DOX)-based chemotherapy induces cardiotoxicity, which is considered the main bottleneck for its clinical application. In this study, we investigated the potential benefit of LCZ696, an angiotensin receptor–neprilysin inhibitor against DOX-induced cardiotoxicity in rats and H9c2 cells and determined whether the mechanism underlying any such effects involves its antioxidant activity. Male Sprague–Dawley rats were randomly separated into four groups, each consisting of 15 rats (DOX (1.5 mg/kg/day intraperitoneally for 10 days followed by non-treatment for 8 days); DOX + valsartan (31 mg/kg/day by gavage from day 1 to day 18); DOX + LCZ696 (68 mg/kg/day by gavage from day 1 to day 18); and control (saline intraperitoneally for 10 days). DOX-induced elevation of cardiac troponin T levels on day 18 was significantly reduced by LCZ696, but not valsartan. The DOX-induced increase in myocardial reactive oxygen species (ROS) levels determined using dihydroethidium was significantly ameliorated by LCZ696, but not valsartan, and was accompanied by the suppression of DOX-induced increase in p47phox. LCZ696 recovered the DOX-induced decrease in phosphorylation of adenosine monophosphate-activated protein kinase and increased the ratio of Bax and Bcl-2. In H9c2 cardiomyocytes, LCZ696 reduced DOX-induced mitochondrial ROS generation and improved cell viability more than valsartan. Our findings indicated that LCZ696 ameliorated DOX-induced cardiotoxicity in rat hearts in vivo and in vitro, possibly by mediating a decrease in oxidative stress.


Results
Effects of LCZ696 on DOX-induced changes in hemodynamic parameters and heart weight. Table 1 shows the change in heart rate, blood pressure, body weight, heart weight, heart weight to body weight ratio, fractional shortening, and ejection fraction in treated rats. During the experimental period, the change in heart rate over time did not differ among the four groups. Systolic blood pressure on day 11 in the DOX (rats that received DOX), DOX + VAL (rats that received DOX and valsartan), and DOX + LCZ (rats that received DOX and LCZ696) groups was significantly lower than that in the control (rats that received saline) group. On day 18, systolic blood pressure in the DOX + LCZ group was significantly lower than that in the control and DOX groups.
The heart weight to body weight ratio in the DOX group was significantly higher than that in the control group. However, the heart weight to body weight ratio in the DOX + VAL and DOX + LCZ groups was significantly lower than that in the control and DOX groups.
Left-ventricular (LV) function was assessed by echocardiography on days 0, 11, and 18. The fractional shortening and ejection fraction on day 11 in the DOX, DOX + VAL, and DOX + LCZ groups were significantly lower than those in the control group. On day 18, the same parameters remained significantly lower than those in the control group. DOX-induced LV dysfunction was not prevented by valsartan and LCZ696.

Effects of LCZ696 on cardiac troponin T in DOX-treated rats.
To assess the change in myocardial injury, the serum level of cardiac troponin T was measured. On day 11, it was increased in the DOX, DOX + VAL, and DOX + LCZ groups compared to that in the control group (Fig. 1A). On day 18, the same parameter was increased in the DOX group and significantly reduced in the DOX + LCZ group, but not in the DOX + VAL group (Fig. 1B).
Effects of LCZ696 on DOX-induced oxidative stress in rat. We measured the MDA level to test whether pre-treatment with LCZ696 ameliorated DOX-induced oxidative stress in rats. On day 11, the serum MDA level was significantly reduced in the DOX + LCZ group compared to that in the DOX group ( Fig. 2A). On day 18, these levels did not differ among the DOX, DOX + VAL, and DOX + LCZ groups (Fig. 2B).
To evaluate ROS production in cardiac tissue, dihydroethidium (DHE) staining of the frozen LV section of rats was performed. The fluorescence intensity of DHE, which was increased in the DOX group, was significantly reduced in the DOX + LCZ group, but not in the DOX + VAL group (Fig. 2C,D). Table 1. Change in the heart rate, blood pressure, body weight, heart weight, heart weight to body weight ratio, fractional shortening, and ejection fraction in the treated rats. Data are expressed as the mean ± SEM. DOX, doxorubicin; VAL, valsartan; LCZ, LCZ696. n = 15 in each group. *p < 0.05 versus control and # p < 0.05 versus DOX. We investigated the expression of p47phox, a key subunit of nicotinamide adenine dinucleotide phosphate oxidase that plays a pivotal role in inducing oxidative stress 19 . On day 18, real-time polymerase chain reaction (PCR) and western blotting showed that the mRNA and protein expression of p47phox was increased in the DOX group (Fig. 2E,F,G). These increases in expression were significantly reduced by LCZ696, but not valsartan.
Effects of LCZ696 on AMPK and Bcl-2 family in the cardiac tissue. Previous studies have demonstrated that DOX inhibits AMPK activation 20 . The phosphorylated AMPK level was decreased in the DOX group compared to that in the control group (Fig. 3A,B). This alteration of phosphorylated AMPK induced by DOX was recovered significantly by the LCZ696 group but not by valsartan.
Proteins of the Bcl-2 family play critical roles in regulating apoptosis. Bax and Bcl-2 are key players in this family. The expression of Bax did not differ among the control, DOX, DOX + VAL, and DOX + LCZ696 groups (Fig. 3C). However, the expression of Bcl-2 decreased in the DOX group and was recovered by LCZ696. The DOX group's Bax/Bcl-2 ratio increase was also significantly reduced by LCZ696, but not valsartan (Fig. 3D). In addition, we evaluated the effect of LCZ696 on the activation of caspase-3 (Fig. 3E,F). The level of cleaved caspase-3 protein in the DOX group was significantly higher than that in the control group. This alteration of cleaved caspase-3 protein expression induced by DOX was reversed by LCZ696, but not valsartan.
Effects of LCZ696 on DOX-induced morphological changes in rat myocardium. To microscopically investigate the effect of LCZ696 on DOX-induced change in rat myocardium, we used hematoxylin and eosin (HE) and Masson's trichrome staining. HE staining showed that the cross-sectional area of cardiomyocytes was significantly lesser in the DOX and DOX + VAL groups, but not in the DOX + LCZ group, when compared to that in the control group (Fig. 4A,B). Furthermore, with respect to the effect of LCZ696 on cardiac fibrosis in rats, Masson's trichrome staining showed increased intestinal fibrosis in the DOX group, and significantly decreased intestinal fibrosis in the DOX + LCZ group, but not in the DOX + VAL group (Fig. 4C,D). The increase in perivascular fibrosis in the DOX group was significantly reduced in the DOX + LCZ group, but not in the DOX + LCZ group ( Fig. 4E and 4F).
Expression of collagen type I alpha 1 (COL1A1), tumor necrosis factor-alpha (TNFa), and atrial natriuretic peptide (ANP) mRNA in the cardiac tissue. Real-time PCR was performed to evaluate the mRNA expression of collagen. On day 18, the mRNA levels of COL1A1 in the DOX and DOX + VAL groups were significantly increased compared to those in the control group (Fig. 5A). The mRNA level of COL1A1 in the DOX + LCZ, but not in the DOX + VAL, was significantly lower than that in the DOX group.
Because TNFa plays an important role in DOX-induced cardiotoxicity 21,22 , the expression of TNFa mRNA was evaluated using real-time PCR (Fig. 5B). On day 18, it was significantly higher in the DOX group than that in the control; however, the increased expression of TNFa in response to DOX treatment was not significantly reduced upon treatment with valsartan or LCZ696.
We also analyzed the expression of ANP mRNA using real-time PCR (Fig. 5C). On day 18, it was significantly higher in the DOX group than that in the control. The increased expression of ANP in response to DOX was significantly reduced upon treatment with valsartan or LCZ696.
Measurements of serum N-terminal pro-brain natriuretic peptide (NT-proBNP) levels.

Effects of LCZ696 on DOX-induced mitochondrial oxidative stress in H9c2 cells.
We investigated the effect of valsartan and LCZ696 on DOX-induced mitochondrial ROS level in H9c2 cells. Mitochondrial superoxide production was measured via fluorescence-activated cell sorting-based quantification of fluorescent MitoSOX Red. As shown in Fig. 6A, H9c2 cells displayed marked mitochondrial ROS production after 6 h of   www.nature.com/scientificreports/ treatment with 5 μM DOX, whereas mitochondrial ROS production was clearly inhibited in the cells pre-treated with LCZ696. The mean fluorescence of intensity increased above that of the control and DOX-treated cells. However, pre-treatment with valsartan attenuated the increased fluorescence, and pre-treatment with LCZ696 further inhibited the increased fluorescence (Fig. 6B).

Effects of LCZ696 on DOX-induced cytotoxicity in H9c2 cells.
To analyze the effect of LCZ696 on DOX-induced cytotoxicity, cell viability was measured using MTS assay after exposing H9c2 cells to DOX. As shown in Fig. 6C, cell viability was markedly decreased after 24 h of treatment with 5 μM DOX. Pre-treatment with 10 μM and 20 μM LCZ696 for 12 h attenuated the decreased cell viability, whereas pre-treatment with 10 μM and 20 μM valsartan did not inhibit DOX-induced cytotoxicity.

Discussion
Herein, we investigated the prophylactic effects of LCZ696 against DOX-induced cardiotoxicity and the underlying potential mechanisms. Our results showed that LCZ696 treatment ameliorated DOX-induced oxidative stress and cardiac fibrosis in rats. However, DOX-induced cardiac dysfunction was not prevented by LCZ696 in this model. The study of hemodynamic parameters in the DOX group showed a significant decrease in fractional shortening and ejection fraction which were not prevented by valsartan and LCZ696. Heart weight and heart weight to body weight ratio were reduced by DOX and a significant decrease in systolic blood pressure during treatment with DOX was observed, consistent with previous reports 5, 23 . These experiments were performed to investigate the early pathophysiological mechanisms of DOX-induced cardiotoxicity. Boutagy et al. recently reported that LCZ696 attenuated the DOX-induced reduction of LV systolic function in rats receiving DOX over 3 weeks, which contradicts our findings 24 . However, we obtained insufficient evidence for the efficacy of LCZ696, possibly because of the short treatment duration. With respect to the change in systolic blood pressure, its reduction on day 11 was similar among the DOX, DOX + VAL, and DOX + LCZ groups, probably due to DOX-induced LV dysfunction. However, the reduction in systolic blood pressure on day 18 in the DOX + LCZ group, but not in the DOX VAL groups, remained significant compared to that in the control and DOX groups. An explanation is that sacubitril in LCZ696 elevates the levels of circulating natriuretic peptides such as ANP, BNP, and C-type natriuretic peptide, which induce vasodilation 25 .
This study demonstrated that treatment with LCZ696 reduced DOX-induced oxidative stress both in rats and H9c2 cells, and that this effect of LCZ696 was stronger than that of valsartan. The inhibition of ARB suppresses oxidative stress in spontaneously hypertensive rats 26 . In addition, in cardiomyocytes, natriuretic peptides play a role in the development of the anti-oxidative stress response that is induced upon the inhibition of NADPH oxidases 27 . Consistent with previous data, this study showed that the DOX-induced upregulation of 47phox was reduced by LCZ696. Thus, the additive antioxidant effects of angiotensin blockade and increased natriuretic peptides may contribute to the greater effect of LCZ696. Mitochondria are a major source of ROS during HF 28 . Xia et al. reported that LCZ696 alleviated the increase in dynamin-related protein 1, a fission protein of mitochondria, in DOX-induced cardiotoxicity 29 . This study suggests that a decrease in ROS production may indicate mitochondrial dysfunction in the DOX-induced cardiotoxicity model.
Mitochondrial impairment, increased apoptosis, dysregulated autophagy, and increased fibrosis have also been shown to play crucial roles in doxorubicin cardiotoxicity. AMPK30 links these cellular processes. Several studies have shown that cardiac AMPK is inhibited by DOX 20,30 . In this study, LCZ696 significantly increased the expression of phosphorylated AMPK. The effect of LCZ696 on AMPK activation may be involved in ameliorating DOX-induced cardiotoxicity in rat hearts.
Proteins of the Bcl-2 family play critical roles in the regulation of apoptosis 31 . Bax and Bcl-2 are the key players in this family, and the ratio of Bax/Bcl-2 is critical for the induction of apoptosis. The increased ratio of BAX/ Bcl-2 leads to the activation of caspase-3 32 . We found that LCZ696 decreased the increased ratio of Bax/Bcl-2 by DOX and an increased in cleaved caspase -3. These data suggest that the anti-apoptotic effect contributed to the alleviation of DOX-induced cardiotoxicity.
Our study also demonstrated that LCZ696 ameliorated cardiac fibrosis. An antifibrotic effect of LCZ696 has been reported in rodent models of HF characterized by pressure-overload cardiac hypertrophy 33 , myocardial infarction 34 , chronic kidney disease 35 , isoproterenol-induced cardiotoxicity 14 , diabetic cardiomyopathy 36 , DOXinduced dilated cardiomyopathy 29 , and obesity-related diastolic dysfunction 15 . Although the precise mechanisms underlying the protective effect of LCZ696 in preventing DOX-induced cardiac fibrosis remain unclear, we previously reported that the mRNA level of transforming growth factor = beta in isoproterenol-treated left ventricle was suppressed in the LCZ696 group compared with that of the control groups 14 . Neprilysin inhibition increases circulating levels of natriuretic peptides, which would increase the generation of cyclic guanosine monophosphate 11 . The increased cyclic guanosine monophosphate is a potent inhibitor of collagen production through transforming growth factor-beta in fibroblasts in vitro 25 . Consistent with our findings, in the sub-analysis of the PARADIGM-HF trial, the effects of sacubitril/valsartan on biomarkers of extracellular matrix in heart failure with reduced ejection fraction were evaluated, and sacubitril/valsartan was found to reduce profibrotic signaling 37 .
This study had a few limitations. First, it evaluated the effect of LCZ696 on the early pathophysiological change in DOX-induced cardiomyopathy. Further study is needed to clarify the benefit of LCZ696 on chronic DOX-induced cardiotoxicity. Second, although the differential effects between LCZ696 and valsartan may be due to neprilysin blockade, the neprilysin substrates responsible cannot be determined. Neprilysin has several peptide substrates including bradykinin, substance P, adrenomedullin, angiotensin I and II, and endothelin-1 38 . In this study, we did not evaluate the specific effect of sacubitril on cardiotoxicity after treatment with LCZ696. Therefore, it is necessary to determine the exact mechanisms contributing to the beneficial effects of LCZ696.
In summary, our data demonstrated that LCZ696 protected the rat hearts from DOX-induced cardiotoxicity in vivo and in vitro. This effect may be mediated via decreased oxidative stress. Our results suggest that LCZ696 can be potentially used to treat DOX-induced cardiotoxicity. Potential clinical benefits should be investigated in large prospective clinical studies in the future. www.nature.com/scientificreports/ randomly allocated to four groups (n = 15) as follows: (1) Control: rats received 2 mL/kg/day saline intraperitoneally from day 1 to day 10 followed by non-treatment for 8 days.

Methods
(2) DOX: rats received 1.5 mg/kg/day DOX (Wako, Tokyo, Japan) intraperitoneally from day 1 to day 10 followed by non-treatment for 8 days, as previously described 5 . A total dose of 15 mg/kg DOX was administered over the 10-day period. (3) DOX + VAL: rats received 1.5 mg/kg/day DOX intraperitoneally from day 1 to day 10 followed by non-treatment for 8 days and 31 mg/kg/day valsartan from day 1 to day 18 orally. (4) DOX + LCZ: rats received 1.5 mg/kg/day DOX intraperitoneally from day 1 to day 10 followed by non-treatment for 8 days and 68 mg/kg/day LCZ696 from day 1 to day 18 orally. On day 18, the rats were anesthetized by administering 60 mg/kg sodium thiopental intraperitoneally and euthanized. Valsartan was purchased from Sigma-Aldrich (St. Louis, MO, USA). LCZ696 for animal experiments was supplied by Novartis Pharma K.K. (Basel, Switzerland) and included molecular moieties of valsartan and sacubitril at a 1:1 ratio. In this study, doses of 68 mg/kg and 31 mg/kg, respectively were selected for LCZ696 and valsartan because several experimental studies using rat models have employed these doses 14,35,36,39,40 .
Blood pressure and pulse rate measurement. Systolic blood pressure and pulse rate were measured at 0, 11 and 18 days using tail-cuff plethysmography (MK-2000, Muromachi, Tokyo, Japan) 41 . An average of three measurements was used.
Echocardiography. Transthoracic echocardiography was performed using a 10-MHz phased array transducer (Aplio ver. 6.0; Toshiba, Tokyo, Japan) under 2% isoflurane. The papillary muscles were visualized with M-mode echocardiography using a parasternal short-axis view. The LV internal diameter at end-diastole and end-systole are measured during diastole and systole. Fractional shortening was then calculated according to the formula FS = [(LV internal diameter at end-diastole) − (LV internal diameter at end-systole)/LV internal diameter at end-diastole] × 100. Ejection fraction was calculated using end-systolic and end-diastolic volumes according to Teichholz's formula 37 .
Measurements of cardiac troponin T. Serum cardiac troponin T was measured using an enzyme-linked immunosorbent assay kit (RK03996, ABclonal, Woburn, MA, USA) according to the protocol. Briefly, the samples and standard substances were added into 96-well plate and incubated for 60 min at 37 °C, cultured with biotin-conjugated antibodies for 60 min, and later seeded with streptavidin-peroxidase complex for 30 min before adding tetramethylbenzidine substrate. Finally, the optical density values at 450 nm were detected using a microplate reader (Bio-Rad Laboratories, Hercules, CA) for calculating the quantity. Histology. The LV cardiac tissues taken for histological analysis were fixed with 4% paraformaldehyde prepared in phosphate-buffered saline, embedded in paraffin, and cut into 5-µm-thick sections. Subsequently, these sections were stained with HE for evaluating the cross-sectional areas of the cardiomyocytes and Masson's trichrome for evaluating the percentage of fibrosis compared to the total area. The cardiomyocyte cross-sectional areas and fibrotic area were examined using a BZ-700 microscope equipped with a BZ analyzer (Keyence, Osaka, Japan). Intestinal fibrosis area (%) and perivascular fibrosis area (%) were calculated using the following formulas: intestinal fibrosis = collagen area/total area × 100 and perivascular fibrosis area (%) = area occupied by the collagen/total area of the vessel section × 100. The percentage of the fibrotic area was analyzed using WinROOF Version 5.7 (MITANI Corporation, Fukui, Japan) 42 . Real-time PCR. Total RNA was extracted using the TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol. RNA was quantified using Implen's Nanophotometer (Munich, Germany). cDNA was synthesized using the Prime Script RT reagent kit (Takara). Real-time qPCR was performed using the Ste-pOnePlus Real-Time PCR System (Applied Biosystems) 43 . All mRNA-specific labeled primers were purchased from Applied Biosystems and the expression of the following genes was investigated: Col1a1 (Rn01523309_m), Tnfa (Rn01525859_g1), and Anp (Rn00664637_g1). Gapdh (Rn01775763_g1) was used as an internal control, and the data were analyzed using the 2 −ΔΔCt method 43 .
H9c2 cell culture. H9c2 cells derived from the rat myocardium were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). H9c2 cells were cultured in high glucose Dulbecco's modified Eagle's medium, supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, and 2 mM L-glutamine. Dulbecco's modified Eagle's medium, fetal bovine serum, and penicillin/streptomycin were purchased from Gibco (Grand Island, NY, USA). Cells were maintained in a humidified incubator (37 °C, 5% CO 2 ) and were always used at less than 80% confluence. For cell culture experiments, LCZ696 was purchased from Cayman Chemical (Ann Arbor, MI, USA). Statistical analysis. Data are expressed as the mean ± standard error. Comparisons between multiple groups were done by the one-way analysis of variance followed by Bonferroni post hoc test for data with a normal distribution or Kruskal-Wallis test with Dunn's post hoc test for non-normal distribution data. A p value < 0.05 was considered statistically significant. Data were analyzed using SPSS for Windows (ver. 25; SPSS Inc., Chicago, IL, USA).

Data availability
The datasets generated during the present study are available from the corresponding author on reasonable request.