Estrogen deprivation aggravates cardiac hypertrophy in nonobese Type 2 diabetic Goto–Kakizaki (GK) rats

Both Type 2 diabetes mellitus (T2DM) and estrogen deprivation have been shown to be associated with the development of cardiovascular disease and adverse cardiac remodeling. However, the role of estrogen deprivation on adverse cardiac remodeling in nonobese T2DM rats has not been clearly elucidated. We hypothesized that estrogen-deprivation aggravates adverse cardiac remodeling in Goto–Kakizaki (GK) rats. Wild-type (WT) and GK rats at the age of 9 months old were divided into two subgroups to have either a sham operation (WTS, GKS) or a bilateral ovariectomy (WTO, GKO) (n = 6/subgroup). Four months after the operation, the rats were killed, and the heart was excised rapidly. Metabolic parameters, cardiomyocytes hypertrophy, cardiac fibrosis, and biochemical parameters were determined. GK rats had hyperglycemia with hypoinsulinemia, and estrogen deprivation did not increase the severity of T2DM. Cardiac hypertrophy, cardiac oxidative stress, and phosphor-antinuclear factor κB were higher in WTO and GKS rats than WTS rats, and they markedly increased in GKO rats compared with GKS rats. Furthermore, cardiac fibrosis, transforming growth factor-β, Bax, phosphor-p38, and peroxisome proliferator- activated receptor γ coactivator-1α expression were increased in GKS and GKO rats compared with the lean rats. However, mitochondrial dynamics proteins including dynamin-related protein 1 and mitofusin-2 were not altered by T2DM and estrogen deprivation. Although estrogen deprivation did not aggravate T2DM in GK rats, it increased the severity of cardiac hypertrophy by provoking cardiac inflammation and oxidative stress in nonobese GK rats.


Introduction
Type 2 diabetes mellitus (T2DM) has been proposed as a global health problem [1]. In advanced state of uncontrolled T2DM patients, there is impaired pancreatic β-cell function, and diminished insulin sensitivity in the insulin responsive tissues including the heart [2], leading to hypoinsulinemia and hyperglycemia. Several experimental models have been use to replicate human T2DM such as db/db mice, Zucker diabetic rats, and Goto-Kakizaki (GK) rats. GK rats are genetically engineered nonobese rats with hyperglycemia and hypoinsulinemia, and have been used as a nonobese T2DM model [3]. T2DM is associated with cardiovascular disease. Several pieces of evidence showed that GK rats have developed pathological conditions in the heart including cardiac dysfunction, hypertrophy, and fibrosis [4][5][6].
Estrogen deprivation is another important factor that has been shown to increase the risk of T2DM [7], and is also associated with cardiac dysfunction [8]. Our recent study reported that estrogen deprivation aggravated cardiac dysfunction and cardiac mitochondrial dysfunction in obese-insulin resistant rats [9]. Furthermore, in various models such as ovariectomized rats [10], ovariectomized hypertensive rats [11,12], db/db mice [13], and GK rats [4-6], both estrogen deprivation and T2DM have been shown to be associated with cardiac remodeling including cardiac fibrosis and hypertrophy. Although several studies reported the deleterious effects of either estrogen deprivation or T2DM on the heart [4-6,10-12], research regarding the effects of estrogen deprivation together with T2DM on the heart is limited. There is some supporting evidence from an in vitro study showing that ventricular cardiomyocytes from ovariectomized rats had a higher degree of contractile dysfunction when exposed to high-glucose solution, compared with cardiomyocytes from sham operated rats [14]. However, the role of estrogen deprivation on metabolic parameters in the case of T2DM is still controversial [15][16][17][18]. Although estrogen deprivation has been shown to exacerbate insulin resistance and inflammation [16][17][18], it did not cause overt hyperglycemia [15].
Mitochondria play an important role in all organs in our body especially the heart. In the state of energy depletion, peroxisome proliferator-activated receptor γ coactivator-1α (PGC-1α) is activated to increase mitochondrial biogenesis and energy production [19]. Regarding the process of mitochondrial energy production, long chain free fatty acids would be taken up into the mitochondria via carnitine palmitoyltranferase 1 (CPT-1) to increase mitochondrial fatty acid oxidation [20]. However, the effects of estrogen deprivation, along with T2DM on cardiac mitochondrial metabolism, are still unclear. Recently, growing evidence showed the adverse effects of estrogen deprivation in T2DM subjects such as increased bone loss [21] and pancreatic β-cell dysfunction in T2DM animal models [17]. Although there are studies that demonstrated the deleterious effects of estrogen deprivation in T2DM subjects, there is still a paucity of study regarding the effect of estrogen deprivation on cardiac remodeling and function in nonobese T2DM.
In the present study, we investigated the effects of estrogen deprivation on metabolic parameters, cardiac remodeling, and cardiac mitochondrial biogenesis and dynamics in GK rats. We tested the hypothesis that estrogen deprivation aggravates T2DM, cardiac remodeling, and cardiac mitochondrial biogenesis and dynamics in GK rats.

Materials and methods
The present study was approved by the Laboratory Animal Ethics Committee of Faculty of Science, and National Laboratory Animal Center, Mahidol University, Thailand, in compliance with NIH guidelines, and in accordance with the ARRIVE guidelines for reporting experiments involving animals [22]. Nine-month-old GK and Wistar Hannover (wild-type; WT) were used in the present study, and rats were housed in a 12:12 h dark:light cycle in a controlled temperature (23-25 • C) room. Rats in each strain were randomly divided into two subgroups to have either a sham operation or a bilateral ovariectomy (OVX). All rats were killed at the age of 12.5 months old by deeply anesthesia with sodium pentobarbital (150 mg/kg, Ceva animal health, Bangkok, Thailand), and the hearts were excised rapidly.

Bilateral OVX
A bilateral OVX was performed under sterile surgical techniques as previously described [23]. Briefly, rats were anesthetized using sodium pentobarbital, and the 1.5 cm incision was made in the paralumbar region, then the distal part of the fallopian tubes was identified, ligated, and then removed. Successful surgery was confirmed by uterine atrophy and vaginal smear.

Metabolic profiles
Body weight, fasting blood glucose, and plasma insulin were determined at post-OVX. Blood glucose level was determined using an Accu-Chek active blood glucose meter (Roche Diagnostics, Germany). Plasma insulin level was determined using a commercialELISA kit (Millipore, MO, U.S.A.). Plasma estradiol level was determined using a commercial ELISA kit (Cayman chemical, MI, U.S.A.). Plasma cholesterol and LDL levels were determined using chemiluminescent assays (Architect Plus; Abbott Diagnostics, Abbott Park, IL, U.S.A.).

Histological study
At the end of the experiment, the hearts were excised rapidly. The midsection of the left ventricle (LV) was used to determine cardiac hypertrophy and interstitial fibrosis. In brief, the mid LV section was embedded in paraffin and sliced at 10 μm thickness intervals. The slices were stained with Picrosirius Red to quantify the amount of interstitial collagen in the heart tissue. An increase in the interstitial collagen deposit in the heart indicated the severity of cardiac fibrosis. Another slice from the same sample was stained with hematoxylin and eosin (H&E) to determine the cardiomyocyte size. Fifty LV cardiomyocytes with equal-sized nuclei were randomly selected for analysis of the cross-sectional area from prescanned images. Both cardiac fibrosis and hypertrophy were analyzed using Aperio ScanScope (Aperio, Aperio Technologies Inc., California, U.S.A.) [24].

Protein expression analysis
Protein expression was determined using Western blot analysis. The myocardial protein was extracted from snap-frozen heart tissue. The heart was homogenized in an extraction buffer containing 20 mM Tris/HCl (pH 6.8), 1 mM sodium orthovanadate, 5 mM sodium fluoride, and a protease inhibitor (Calbiochem, Darmstadt, Germany). Then, the heart tissue was centrifuged at 16000g for 10 min, the supernatant was retained, and the protein concentration was determined using a Bio-Rad protein assay kit (Bio-Rad Laboratories, CA, U.S.A.). Sixty micrograms of total protein was mixed with a loading buffer containing 5% β-mercaptoethanol, 0.05% Bromophenol Blue, 75 mM Tris/HCl (pH 6.8), 2% SDS, and 10% glycerol. The protein was loaded onto 10% SDS/acrylamide gels, and transferred to a 0.45 μm pore size nitrocellulose membrane (GE Healthcare Bio-Sciences, MA, U.S.A.) in a glycine/methanol-transferred buffer using a Wet/ Tank

Cardiac oxidative stress level
Cardiac malondialdehyde (MDA) level was used to determine cardiac oxidative stress level. Myocardial tissue was homogenized with a homogenization buffer. The mixture was mixed with 10% trichloroacetic acid containing 50 ppm BHT, and then centrifuged at 3300g for 10 min. The supernatant was mixed with 0.44 M H 3 PO 4 and 0.6% thiobartiburic acid, and heated at 90 • C to produce the pink-colored product called thiobartiburic acid reactive substances (TBARS). The TBARS was fractionated on the adsorption column with the high-performance liquid chromatography (HPLC) technique (Thermo Fisher Scientific, MA, U.S.A.). Cardiac TBARS concentration was determined directly from a standard curve and reported as an MDA equivalent concentration. Additionally, a small amount of supernatant from the myocardial tissue was used to determine protein concentration using a Bio-Rad protein assay kit (Bio-Rad Laboratories, CA, U.S.A.). Cardiac MDA level was represented as μM/mg protein [9].

Statistical analysis
Data were expressed as mean + − SE. A one-way analysis of variance (ANOVA) followed by Tukey's post-hoc test was used to test the difference between groups. P<0.05 was considered as being statistically significant.

Effects of estrogen deprivation on metabolic profiles in GK and WT rats
At the age of 12.5 months, wild-type ovariectomized rats (WTO) had a higher body weight than wild-type rats with sham operation (WTS). In GK rats, both GK rats with sham operation (GKS) and GK ovariectomized rats (GKO) exhibited T2DM as indicated by increased plasma glucose and decreased plasma insulin compared with WTS rats (Table 1). Moreover, the body weight was markedly increased in GKO rats compared with GKS rats. However, there was no significant difference in plasma glucose and insulin levels between GKS and GKO rats. In OVX group, plasma estradiol levels were significantly decreased in both WTO and GKO rats, compared with their sham operation ( Table  1). For plasma lipid profiles, plasma cholesterol and LDL levels were not different between WTS and GKS rats. In OVX group, plasma cholesterol and LDL levels were increased in both WTO and GKO rats, compared with their sham operation ( Table 1).

Effects of estrogen deprivation on cardiac oxidative stress and inflammation in GK and WT rats
The cardiac MDA level was higher in WTO rats than in WTS rats. In the GK groups, the cardiac MDA level was higher in GKS rats compared with WTS rats. Moreover, GKO rats had a higher cardiac MDA level than WTS, WTO, and GKS rats ( Figure 1A).  In the present study, NF-κB phosphorylation was used as a determinant inflammatory marker. We found that NF-κB phosphorylation was increased in WTO rats, compared with WTS rats. In GK groups, NF-κB phosphorylation was higher in GKS rats than WTS rats, and NF-κB phosphorylation was greater in GKO rats than WTS and GKS rats ( Figure 1B).

Effects of estrogen deprivation on cardiac remodeling in GK and WT rats
In the present study, heart weight/body weight ratio was used to represent the severity of cardiac hypertrophy. In sham operation group, our data showed that GKS rats had higher heart weight/body weight than WTS rats. In OVX group, heart weight/body weight ratio was increased in WTO rats, compared with WTS rats. Moreover, heart weight/body weight ratio was markedly increased in GKO rats, compared with WTS, WTO, and GKS rats ( Table 1).
In addition, cardiomyocyte hypertrophy and cardiac fibrosis were determined as markers for cardiac remodeling. Regarding cardiomyocyte hypertrophy, data from a histological study showed that WTO rats had a larger cardiomyocyte cross-sectional area (CSA) than WTS rats (Figure 2A). In GK rats, CSA was increased in GKS rats, compared with WTS and WTO rats. Moreover, GKO rats had a larger CSA than WTS, WTO, and GKS rats (Figure 2A).
Regarding cardiac fibrosis, our results demonstrated that cardiac fibrosis, as indicated by the intensity of Picrosirius Red, was significantly increased in GKS and GKO rats compared with WTS and WTO rats. We also found that cardiac fibrosis showed no difference between GKS and GKO rats ( Figure 2B). Our results also showed that Bax expression was increased in GKS and GKO rats, compared with WTS rats ( Figure 3A). However, Bcl-2 expression was not different between groups ( Figure 3B). Furthermore, the p-p38 was increased in GKS compared with WTS rats ( Figure 3C). In GKO rats, we found that p-p38 was higher in GKO rats, compared with WTO and GKS rats ( Figure  3C). The TGF-β expression was also increased in GKS and GKO rats, compared with WTS and WTO rats ( Figure  3D). However, there was no difference in both p-p38 MAPK and TGF-β expression between GKS and GKO rats ( Figure 3C and D).

Effects of estrogen deprivation on cardiac mitochondrial metabolism in GK and WT rats
Our data demonstrated that CPT-1 expression was not different between groups ( Figure 4A). However, a mitochondrial biogenesis marker, PGC-1α expression was increased in GKS and GKO rats, compared with WTS rats. Moreover, there was no difference in PGC-1α expression between GKS and GKO rats ( Figure 4B). The expression of mitochondrial fission and fusion proteins, DRP-1 and Mfn-2, were not different between groups ( Figure 4C and D).

Discussion
The main focus in the present study was investigating the effects of estrogen deprivation on the heart of a T2DM rat model. The major findings are that estrogen deprivation could lead to: (1) increased body weight in both the WT and GK rats, however, it did not aggravate the severity of T2DM in the GK rats; (2) increased cardiac oxidative stress and inflammation were associated with cardiac hypertrophy in both the WT and GK rats; (3) no exacerbation of cardiac fibrosis, apoptosis, and cardiac mitochondrial biogenesis in the WT and GK rats, and (4) no alteration of mitochondrial fatty acid oxidation and mitochondrial fission and fusion protein expression in both the WT and GK rats.
In WTO rats, estrogen deprivation augmented lipogenesis, adipocyte hypertrophy, reduced fatty acid oxidation, and increased lipid accumulation in adipose tissue, all of which caused an increased body weight [25]. Moreover, adipocyte hypertrophy and lipid accumulation could increase oxidative stress and inflammation, and NF-κB activation. Since NF-κB regulates hypertrophic genes expression, its activation could lead to cardiac hypertrophy as observed in WTO rats. However, the levels of oxidative stress and inflammation could be insufficient to induce collagen deposition, cardiac fibrosis, as well as cardiac apoptosis [26].  A GK rats is a genetically engineered-nonobese-T2DM model, as indicated by hypoinsulinemia with hyperglycemia [3]. Our data were consistent with previous studies which showed that GK rats had lower body weight than WT rats [27,28]. Furthermore, GK rats had insulin resistance, impaired pancreatic β-cells function, decreased β-cells volume, and reduced insulin secretion [29,30] since they were at the age of 2 months old [30], leading to hypoinsulinemia with hyperglycemia in their advanced age. After OVX, our results demonstrated that at the age of 12.5 months, estrogen-deprived GK rats had increased body weight, but this did not increase the severity of T2DM in these GK rats. However, we found that our data were consistent with ZDF rats, which is a genetically engineered obese-T2DM rat [31,32]. These data suggested that bilateral OVX did not alter blood glucose levels or plasma insulin levels in T2DM rats. Unlike a diet induced prediabetic animal model, bilateral OVX increased fasting plasma glucose and insulin levels [33][34][35]. Although estrogen deprivation increased cardiac inflammation and oxidative stress, it could be insufficient to aggravate hyperglycemia in nonobese GK rats. In addition, several studies reported that insulin sensitivity [36] and pancreatic β-cells function [37] were not affected by estrogen deprivation. Therefore, estrogen deprivation did not aggravate hyperglycemia in nonobese GK rats. In contrary to the prediabetic and diabetic obese animals, several studies reported that estrogen deprivation reduced insulin sensitivity and increased plasma glucose levels [35,38], led to increased severity of T2DM. Moreover, the future research is needed to clarify the different role of estrogen deprivation on metabolic function in nonobese T2DM versus obese T2DM subjects.
In the present study, cardiac hypertrophy developed in the estrogen-deprived WT and estrogen-deprived GK rats. However, our results showed that estrogen deprivation aggravated the severity of cardiac hypertrophy in the GK rats.
Oxidative stress and inflammation have been demonstrated as key contributing factors in the development of cardiac hypertrophy [26,39]. Oxidative stress activates a variety of cardiac hypertrophy signaling kinase and transcription factors including NF-κB [26]. NF-κB is also a downstream signaling molecule of TNF-α (a proinflammatory cytokine). Thus, oxidative stress could also initiate an inflammatory response [26]. Our data showed that both estrogen deprivation and hyperglycemia increased oxidative stress and NF-κB activation, and associated with the development of cardiac hypertrophy. Furthermore, our results demonstrated that a combination of estrogen deprivation and hyperglycemia as seen in OVX-GK rats had a higher level of oxidative stress and inflammation than either estrogen deprivation or hyperglycemia alone. This synergistic adverse effect could be responsible for the increased severity of cardiac hypertrophy observed in our OVX-GK rats.
Cardiac fibrosis is caused by the accumulation of fibroblast and extracellular matrix proteins including collagen in the heart, leading to abnormal LV architecture and LV dysfunction [40]. Our data indicated that only T2DM condition as seen in GK rats was associated with cardiac fibrosis, but not with estrogen deprivation in WT rats, suggesting that although estrogen deprivation increased oxidative stress, it did not trigger the cardiac fibrosis pathway in this model. Several studies have demonstrated that T2DM causes cardiac fibrosis though various mechanisms [7,41]. It has been shown that TGF-β and p38 play important roles during cardiac remodeling and fibrosis [40]. In the present study, their expression was increased only in the T2DM rats, whereas estrogen deprivation in the WT did not affect the expression of these proteins in the T2DM rats. Since TGF-β is responsible for the production and deposition of collagen in the heart tissue [42], increased TGF-β expression could cause increased collagen deposition in the T2DM heart [43], and also cardiac fibrosis in the GK rats that was observed in the present study. Previous studies demonstrated that TGF-β activation also increased Bax expression, leading to cellular apoptosis [43,44], thus indicating the association between a higher level of TGF-β expression and increased cardiac apoptosis. Our data supported this notion as indicated by the findings that TGF-β activation is associated with increased interstitial collagen deposition and cardiac apoptosis in T2DM rats ( Figure 5).
In the present study, we determined cardiac mitochondrial metabolism in estrogen-deprived T2DM rats. Our data indicate that T2DM, but not estrogen deprivation in the WT rat has an increased PGC-1α expression. However, neither T2DM nor estrogen deprivation altered other mitochondrial metabolism protein expression, including CPT-1, DRP-1, and Mfn-2. PGC-1α has been shown to increase after p38 activation [45]. PGC-1α plays a role in mitochondrial biogenesis by increasing the mitochondrial content [46]. During cellular stress, ATP depletion is observed, and PGC-1α is activated in order to increase ATP levels [46]. In a PGC-1α gene deletion model, the lack of PGC-1α reduces cardiac ATP production [45,46]. PGC-1α also increases oxidative capacity by increasing mitochondrial fatty acid oxidation or increasing respiratory chain activity [46]. Since we did not find any changes in mitochondrial fatty acid oxidation as indicated by unaltered CPT-1 expression, PGC-1α may increase oxidative capacity by increasing respiratory chain activity through the activation of nuclear respiratory factor-1 (NRF-1) [46]. Although we did not determine NRF-1 levels in the heart, several studies reported NRF-1 was activated by PGC-1α to increase mitochondrial biogenesis and increased oxidative capacity in the pathologic heart such as heart failure and cardiomyopathy [47][48][49][50]. This could be the mechanism responsible for the increase in ATP production during stress in T2DM. Since Estrogen deprivation together with T2DM aggravates cardiac hypertrophy through oxidative stress and an inflammation pathway. However, T2DM increases cardiac fibrosis through TGF-β-p38MAPK and apoptosis pathway, and T2DM also augmented PGC-1α to increase ATP production as a compensatory mechanism.
estrogen deprivation did not alter cardiac mitochondrial metabolism in both WT and GK rats, our findings suggest that the ATP production system was affected negatively only in the T2DM model, and that PGC-1α served as a compensatory mechanism to enhance ATP levels in the T2DM rats.
Our previous study has shown that estrogen deprivation aggravated cardiac mitochondrial dysfunction in obese-insulin resistant rats [9]. In the present study, we found that T2DM did not affect cardiac mitochondrial dynamics. Data from T1DM also showed that it did not alter cardiac mitochondrial dynamics [51]. On the other hand, mitochondrial fission was increased in subjects with obesity and hyperinsulinemia [52,53]. Therefore, these accumulative data suggested that mitochondrial dynamics dysregulation might be affected by obesity and hyperinsulinemia, but was not affected by hyperglycemia and hypoinsulinemia.
Estrogen deprivation aggravated cardiac hypertrophy, and related to an increasing oxidative stress levels and inflammation in nonobese GK rats. However, estrogen deprivation did not aggravate the severity of T2DM and cardiac fibrosis in both normal and nonobese GK rats.