RIPK3-Mediated Necroptosis in Diabetic Cardiomyopathy Requires CaMKII Activation

Activation of Ca2+/calmodulin-dependent protein kinase (CaMKII) has been proved to play a vital role in cardiovascular diseases. Receptor-interaction protein kinase 3- (RIPK3-) mediated necroptosis has crucially participated in cardiac dysfunction. The study is aimed at investigating the effect as well as the mechanism of CaMKII activation and necroptosis on diabetic cardiomyopathy (DCM). Wild-type (WT) and the RIPK3 gene knockout (RIPK3−/−) mice were intraperitoneally injected with 60 mg/kg/d streptozotocin (STZ) for 5 consecutive days. After 12 w of feeding, 100 μL recombinant adenovirus solution carrying inhibitor 1 of protein phosphatase 1 (I1PP1) gene was injected into the caudal vein of mice. Echocardiography, myocardial injury, CaMKII activity, necroptosis, RIPK1 expression, mixed lineage kinase domain-like protein (MLKL) phosphorylation, and mitochondrial ultrastructure were measured. The results showed that cardiac dysfunction, CaMKII activation, and necroptosis were aggravated in streptozotocin- (STZ-) stimulated mice, as well as in (Lepr) KO/KO (db/db) mice. RIPK3 deficiency alleviated cardiac dysfunction, CaMKII activation, and necroptosis in DCM. Furthermore, I1PP1 overexpression reversed cardiac dysfunction, myocardial injury and necroptosis augment, and CaMKII activity enhancement in WT mice with DCM but not in RIPK3−/− mice with DCM. The present study demonstrated that CaMKII activation and necroptosis augment in DCM via a RIPK3-dependent manner, which may provide therapeutic strategies for DCM.


Introduction
Diabetes mellitus (DM) is an independent risk factor of diabetic cardiomyopathy (DCM), in the deficiency of coronary artery diseases, hypertension, and other cardiovascular risk factors. DCM is characterized by a series of structural and functional abnormalities, including myocardial stiffness, contractility impairment, myocardial fibrosis, and hypertrophy [1,2]. Special attention must be paid to DCM due to its concealed onset, rapid evolution but poor treatment efficacy. Possible pathophysiological factors of DCM involve hyperglycemia, hyperlipidemia, insulin resistance, oxidative stress, endoplasmic reticulum stress, cardiomyocyte death, and mitochondrial dysfunction [3,4]. However, the exact mechanism is still obscure. Investigation to find out potential preventative and therapeutic strategies is an urgent problem to be solved.
Recent advances have revealed a new necrotic type of regulated cell death, necroptosis, which is morphologically distinct from apoptosis and necrosis [5]. Necroptosis is characterized by cell enlargement, organelle swelling, and plasma membrane rupture, followed by cell disintegration and intracellular component release without obvious changes in nuclear chromatin [6]. Receptor-interaction protein kinase 3 (RIPK3) plays the core role in the necroptotic signaling pathway. RIPK3, along with receptor-interacting protein 1 (RIP1) and mixed lineage kinase domain-like protein (MLKL), executes necroptosis [7,8]. Besides, Zhang et al. identified Ca 2+ /calmodulin-dependent protein kinase (CaM-KII) as a novel substrate of RIPK3 mediating ischemia-and oxidative stress-induced myocardial necroptosis [9]. Emerging evidences have corroborated the vital role of necroptosis in cardiovascular diseases. However, contribution of necroptosis in DCM remains insufficient.
CaMKII has four isoforms: α, β, γ, and δ, among which CaMKIIδ is prominent in the heart. CaMKIIδ expresses three variants CaMKIIδA, CaMKIIδB, and CaMKIIδC after alternative splicing of its exon 14, 15, or 16 by splicing factors. Alternative splicing of CaMKIIδ is strictly regulated. Once it is disordered, the expression of three variants imbalances to lead to cardiomyocyte dysfunction and ultimately heart diseases [22].
Protein phosphatase 1 (PP1) is a serine/threonine phosphatase mainly present in the heart and plays a vital role in regulating the phosphorylation of splicing factors to eventually mediate CaMKIIδ alternative splicing [14]. PP1 also regulates CaMKII dephosphorylation. It has been shown that PP1 expression was elevated in the heart of heart failure patients [23]. However, the clinical application value of PP1 has not been found. Inhibitor 1 of protein phosphatase 1 (I1PP1) binds to PP1 and subsequently inhibits its activity. I1PP1 transgenic mice demonstrated minor hypertrophy after transverse aortic constriction (TAC). Our previous studies have demonstrated that CaMKII regulation by I1PP1 alleviates necroptosis in high glucose-induced cardiomyocyte injury [24].
In the present study, we investigated the role of necroptosis in the development of DCM and the molecular mechanism underlying necroptosis. We also try to demonstrate the contribution of I1PP1 on the regulation of CaMKIIδ alternative splicing and CaMKII activity to ameliorate necroptosis in the myocardium of streptozotocin-(STZ-) induced wild-type and RIPK3 -/diabetic mice and db/db mice. 2.2. Establishment of Mouse Diabetic Models. After adaptive feeding for a week, mice were injected with 60 mg/kg/d STZ (Sigma, USA), dissolved in 0.1 mol/L citrate buffer for 5 consecutive days after a 12-hour overnight fast [25]. Mice in the control group were injected with the same amount of citrate buffer. Mice with fasting blood glucose (FBG) level above 16.7 mmol/L were considered as the diabetic mice. Mice were randomly divided into different groups: WT group (WT), WT mice with DCM group (WT-DCM), RIPK3 KO group (RIPK3 -/-), and RIPK3 KO mice with DCM group (RIPK3 -/--DCM).

Measurement of the Glycosylated Hemoglobin (HbA1c).
12 weeks after STZ injection, blood samples were collected and heparinized. The level of HbA1c was measured using the Glycosylated Hemoglobin A1c Assay Kit according to the manufacturer's instructions (Jiancheng Bioengineering Institute, Nanjing, China).

Measurement of Triglyceride (TG)
. Blood samples were obtained and centrifuged at 1000 g for 10 min. The supernatant was collected, and serum TG content was measured with the Triglyceride Assay Kit (Jiancheng Bioengineering Institute, Nanjing, China). Absorbance was recorded at the wavelength of 510 nm.
2.5. Echocardiography. 12 weeks after STZ injection, the mice were anesthetized with 1.5% isoflurane and examined by 2-D guided M-mode echocardiography (Visual Sonic Vevo 2100, Toronto, ON, Canada) to assess cardiac configuration and function. Ejection fraction (EF) was calculated with the percentage of the difference between end-diastolic volume (EDV) and end-systolic volume (ESV) to EDV, whereas fractional shortening (FS) was calculated with the percentage of the difference between end-diastolic diameter (EDD) and end-systolic diameter (ESD) to EDD. In addition, the ratio of early (E) to late (A) diastolic velocity ratio (E/A) was calculated. All indexes were obtained from three consecutive beats.  2.11. Real-Time Polymerase Chain Reaction. The total RNA of the myocardium was extracted from the myocardium using a Trizol separation reagent. The cDNA was then synthesized by the Prime Script™ RT Master Mix Kit (Takara, Kyoto, Japan) before quantitative real-time PCR were performed using the SYBR Green Fast qPCR mix (Takara) with the ABI Ste-pOne PCR System (ABI, Carlsbad, CA, USA). All primers used were the following: atrial natriuretic peptide-(ANP-) F 5 ′ -GAGAAGATGCCGGTAGAAGA-3 ′ , ANP-R 5 ′ -AAGCAC TGCCGTCTCTCAGA-3′; brain natriuretic peptide (BNP-F), Experimental cycle threshold values were normalized to 18S, a housekeeping gene, and relative mRNA expression was calculated versus a control sample. 2.13. Statistical Analysis. All data were presented as standard error of the mean (S.E.M.). Statistical analysis was performed by the unpaired Student t test on comparisons between two groups and by the one-way ANOVA test and followed by the Bonferroni post hoc test on comparisons among multiple groups. P < 0:05 was considered statistically significant.

Cardiac Dysfunction, CaMKIIδ Activity, and Necroptosis
Are Augmented in DCM. FPG, HbA1c, serum TG, and myocardial hypertrophic gene expression were increased significantly in mice with DCM, which suggested that STZ injection produced diabetes mellitus ( Figure S1). Mice with FBG more than 16.7 mmol/L were confirmed as diabetic mice and used for further experiments. Compared to the mice in the control group, mice with DCM suffered diastolic and systolic dysfunctions, elucidated by decreased EF, FS, and E/A (Figures 1(a)-1(c)). It suggested that DCM was successfully established in the present study. HE staining displayed hypertrophy, distortion, and irregular arrangement of myocardial cells in the hearts of mice with DCM ( Figure 1(d)). The cTnI level was detected to evaluate myocardial injury. The cTnI level was significantly higher in the DCM group than that in the control group (Figure 1(e)). Disorder of CaMKIIδ alternative splicing is prone to promote cardiomyocyte dysfunction and ultimately heart diseases [24,26]. Since no specific antibodies of CaMKIIδ variants were available, CaMKIIδA, CaMKIIδB, and CaMKIIδC mRNA expressions were measured by quantitative real-time PCR. There was a significant decrease of CaMKIIδA and CaMKIIδB but an increase of CaMKIIδC in diabetic mice, indicating disorder of CaMKIIδ alternative splicing in mice with DCM ( Figure 1(f)). Moreover, oxidation and phosphorylation of CaMKII also increased in the DCM group (Figures 1(g) and 1(h)). Various studies have confirmed that RIPK3 was involved in necroptosis. Our study verified that RIPK3 was significantly increased in the hearts of DCM mice (Figure 1(i)). As apoptosis is a critical manifestation of necroptosis, TUNEL staining and cleaved-caspase 3  5 Oxidative Medicine and Cellular Longevity expression indicated that the necroptotic cardiomyocytes in mice with DCM were significantly more than that in the control mice (Figures 1(j)-1(l)). All these data suggested that cardiac dysfunction, CaMKIIδ activity, and necroptosis were augmented in DCM.

Oxidative Medicine and Cellular Longevity
In accordance, both systolic and diastolic cardiac functions of db/db mice were significantly impaired as compared to WT mice (Figures 2(a)-2(c)), CaMKIIδ variant expression disordered and RIPK3 expression was increased in the myocardium of db/db mice (Figures 2(d) and 2(e)).

RIPK3 Deficiency Alleviates Cardiac Dysfunction,
CaMKIIδ Alternative Splicing Disorder, and Necroptosis in DCM. To further investigate the contribution of RIPK3 in the development of DCM, RIPK3-KO (RIPK3 -/-) mice were used in our study. In diabetic mice, RIPK3 deficiency significantly enhanced EF, FS, and E/A, suggesting both systolic and diastolic functions were improved (Figures 3(a)-3(c)). Notably, RIPK3 -/hearts were resistant to myocardial injury in DCM, as evidenced by improved structure of myocardium and reduced cTnI level (Figures 3(d) and 3(e)). Disorder of CaMKIIδ alternative splicing was alleviated in RIPK3 deficiency mice with DCM (Figure 3(f)). TUNEL staining and western blot assay revealed that necroptosis was attenuated in RIPK3 deficiency mice with DCM (Figures 3(g)-3(k)). These data suggested that the impairment of cardiac dysfunction and the augment of necroptosis were attenuated in mice with DCM if RIPK3 was deficient.
3.3. RIPK3 Deficiency Decreases RIPK1 Expression, MLKL Phosphorylation, and CaMKII Activity and Improves Myocardial Mitochondrial Ultrastructure in Mice with DCM. Besides RIPK3, previous studies have demonstrated that RIPK1 also participated in the process of necroptosis, and phosphorylation of MLKL was an essential effector molecule of necroptosis [27]. In the present study, RIPK1 expression and MLKL phosphorylation were markedly elevated in the myocardium of mice with DCM, which were significantly decreased in RIPK3 -/mice with DCM (Figures 4(a) and 4(b)). A recent study identified CaMKII as one of the substrates of RIPK3 in cardiac ischemic disease [8]. Our results revealed that RIPK3 deficiency attenuated CaMKII oxidation and phosphorylation in the myocardium of mice with DCM (Figures 4(c) and 4(d)). TEM analysis of myocardial mitochondrial ultrastructure showed irregular and swelled mitochondria and fractured mitochondrial cristae in the left ventricular myocardium of mice with DCM. However, the abnormalities of mitochondria with DCM were improved in RIPK3 -/mice (Figure 4(e)).         Figure S2). EF, FS, and E/A were improved after I1PP1 overexpression in WT mice with DCM but not in RIPK3 -/mice with DCM. Moreover, cardiac function of RIPK3 -/mice with DCM was further improved as compared to WT mice with DCM after I1PP1 overexpression (Figures 5(a)-5(e)). Similarly, I1PP1 overexpression reduced cTnI level, improved myocardial structure, and alleviated necroptosis augment in WT mice with DCM but not in RIPK3 -/mice with DCM. In addition, lower cTnI level, better myocardial structure, and weaker necroptosis augment were detected in RIPK3 -/mice with DCM as compared to WT mice with DCM after I1PP1 overexpression (Figures 5(f)-5(j)). All the above data suggested CaMKII was possible downstream of RIPK3 during necroptosis in DCM. The effects of I1PP1 overexpression were also detected in db/db mice. Results showed that EF, FS, and E/A were significantly elevated in db/db mice after I1PP1 overexpression (Figure 6), indicating CaMKII regulation was beneficial to improve cardiac function of db/db mice.

I1PP1 Overexpression Inhibits Oxidation and Phosphorylation of CaMKII and Improves Myocardial
Mitochondrial Ultrastructure in Mice with DCM. Further studies were done to investigate the mechanism of I1PP1 overexpression on necroptosis in DCM mice. Oxidation and phosphorylation of CaMKII were inhibited after I1PP1 overexpression in WT mice with DCM but not in RIPK3 -/mice with DCM. Lower CaMKII oxidation and phosphorylation were measured in RIPK3 -/mice with DCM as compared to WT mice with DCM after I1PP1 overexpression (Figures 7(a) and 7(b)). I1PP overexpression improved mitochondrial ultrastructure in WT mice with DCM but not in RIPK3 -/mice with DCM. Better mitochondrial ultrastructure was observed in RIPK3 -/mice with DCM as compared to WT mice with DCM after I1PP1 overexpression (Figure 7(c)).

Discussion
Accumulated evidence has shown that diabetes results in structural and functional impairment of the hearts. DCM features glucotoxicity, lipotoxicity, and hypertrophy. The results showed increased FBG, HbA1c, serum TG, ANP, and BNP expression in mice with DCM. However, there are still no specific strategies for prevention and treatment of DCM [28]. Necroptosis, a form of regulated necrosis, has been contributable to cardiovascular diseases, such as myocardial infarction, atherosclerosis, and abdominal aortic aneurysm [29]. Our study verified that necroptosis was augmented in mice with DCM. Given RIPK3 is a critical   molecule for necroptosis in large amount types of cells, the differences between WT mice and RIPK3 -/mice during DCM were further investigated. Furthermore, RIPK3 -/alleviated cardiac dysfunction, myocardial injury, and necroptosis in diabetic mice. RIPK1, one common partner of RIPK3, interacts with RIPK3 via a shared RIP homotypic interaction motif domain to form a complex called "necrosome" [30,31]. Necrostatin-1 (Nec-1), an inhibitor of RIPK1, exerts a protective effect against necroptosis during multiple diseases [32]. MLKL, a recently acknowledged substrate of RIPK3, oligomerizes and translocates to the plasma membrane after being phosphorylated by RIPK3, finally leading to necroptotic cell death. A previous study demonstrated that MLKL deficiency protected against necroptosis in cells including tumor cells, macrophages, and fibroblasts [33][34][35]. Our results revealed that both RIPK1 expression and MLKL phosphorylation were significantly upregulated in the myocardium of mice with DCM. Moreover, RIPK3 -/mice with DCM showed decreased RIPK1 expression and MLKL phosphorylation. Anyhow, the detailed pathway of RIPK3-evoked myocardial necroptosis was unknown.
CaMKII is a pleiotropic signal that regulates gene expression, contractility, metabolism, Ca 2+ cycling, and cell survival of cardiomyocytes [36,37]. CaMKII is inactivated under basal conditions. Sustained activation of CaMKII promotes cardiomyocyte death under conditions of oxidative stress, hyperglycemia, and ischemic and hypoxic injury [38][39][40]. Zhang et al. reported that RIPK3 contributed to myocardial necroptosis by oxidizing and phosphorylating CaMKII after ischemia-reperfusion injury or doxorubicin challenge [9]. Our previous study demonstrated CaMKII an alternative substrate of RIPK3 in high glucose-induced cardiomyocyte necroptosis [24]. Our present study found significant enhancement of phosphorylation and oxidation of CaMKII both in DCM mice and in AGE-stimulated cardiomyocytes, which were markedly reversed in RIPK3 -/mice. Moreover, no matter I1PP1 is overexpressed or not, cardiac function was improved and necroptosis was suppressed in RIPK3 -/mice. Taken together, the CaMKII signal pathway was possible downstream or substrate of RIPK3 in DCM.
Various studies have confirmed mitochondrial dysfunction as a key underlying mechanism of necroptosis. Mitochondria are a major source of energy and ROS generation, and mitochondrial dysfunction leads to ROS overproduction as well as cell death [41]. RIPK3 increased mitochondrial localization of NADPH oxidase-4 (NOX4) but inhibited mitochondrial complex I and -III and finally promoted kidney tubular injury [42]. Besides, knockdown of RIPK3 reversed the reduced adenosine triphosphate (ATP) production and attenuated the mitochondrial permeability transi-tion pore (mPTP) opening of cardiomyocytes induced by hypoxia-reoxygenation (HR) injury [43]. Our present study found that RIPK3 deficiency improved the ultrastructure of mitochondria in the myocardium of diabetic mice. These results highlighted that mitochondrial injury is possibly responsible for RIPK3-induced necroptosis in DCM.
PP1 is a serine/threonine protein phosphatase that dephosphorylates diverse cellular substrates, acting as a key regulator of some cellular processes [44]. I1PP1 is the first recognized endogenous inhibitor of PP1 and greatly low expressed in cardiomyocytes under basal conditions. Early studies have been reported overexpression of I1PP1 in engineered heart tissue, and rat cardiac myocytes improved

16
Oxidative Medicine and Cellular Longevity contractile function [45]. Our previous studies also uncovered the role of I1PP1 overexpression in protecting against myocardial ischemia-reperfusion injury through CaMKII regulation and necroptosis alleviating in high glucoseinduced cardiomyocyte injury [24,46]. Our present study corroborated the important role of overexpression of I1PP1 in inhibiting necroptosis to alleviate diabetic cardiomyopathy by regulating CaMKIIδ alternative splicing and CaMKII activity. However, the protective effect of I1PP1 overexpression on DCM was less stronger than that of RIPK3 deficiency. Moreover, I1PP1 overexpression showed no cardioprotective effects in diabetic mice without the RIPK3 gene, indicating that CaMKII might be served as the substrate to mediate necroptosis in the development of DCM.
In conclusion, our study proved there was serious necroptosis in DCM. RIPK3 deficiency alleviated myocardial injury, improved cardiac function, suppressed CaMKII activation, and attenuated necroptosis in mice with DCM mice.
Taken together, CaMKII activation and necroptosis augment in diabetic cardiomyopathy via a RIPK3-dependent manner ( Figure 8). These findings will be a benefit to the understanding of the pathology of diabetic cardiomyopathy and provide novel therapeutic strategies for diabetic cardiomyopathy.

Data Availability
The data used to support the findings of this study are available from the corresponding authors upon request.

Disclosure
The manuscript has been presented as preprint according to the following link: https://authorea.com/doi/full/10.22541/au .158465453.30390557.