Zinc Prevents the Development of Diabetic Cardiomyopathy in db/db Mice

Diabetic cardiomyopathy (DCM) is highly prevalent in type 2 diabetes (T2DM) patients. Zinc is an important essential trace metal, whose deficiency is associated with various chronic ailments, including vascular diseases. We assessed T2DM B6.BKS(D)-Leprdb/J (db/db) mice fed for six months on a normal diet containing three zinc levels (deficient, adequate, and supplemented), to explore the role of zinc in DCM development and progression. Cardiac function, reflected by ejection fraction, was significantly decreased, along with increased left ventricle mass and heart weight to tibial length ratio, in db/db mice. As a molecular cardiac hypertrophy marker, atrial natriuretic peptide levels were also significantly increased. Cardiac dysfunction and hypertrophy were accompanied by significantly increased fibrotic (elevated collagen accumulation as well as transforming growth factor β and connective tissue growth factor levels) and inflammatory (enhanced expression of tumor necrosis factor alpha, interleukin-1β, caspase recruitment domain family member 9, and B-cell lymphoma/leukemia 10, and activated p38 mitogen-activated protein kinase) responses in the heart. All these diabetic effects were exacerbated by zinc deficiency, and not affected by zinc supplementation, respectively. Mechanistically, oxidative stress and damage, mirrored by the accumulation of 3-nitrotyrosine and 4-hydroxy-2-nonenal, was significantly increased along with significantly decreased expression of Nrf2 and its downstream antioxidants (NQO-1 and catalase). This was also exacerbated by zinc deficiency in the db/db mouse heart. These results suggested that zinc deficiency promotes the development and progression of DCM in T2DM db/db mice. The exacerbated effects by zinc deficiency on the heart of db/db mice may be related to further suppression of Nrf2 expression and function.


Introduction
Diabetic cardiomyopathy (DCM) is highly prevalent in asymptomatic type 2 diabetes (T2DM) patients. Diabetes is associated with an increased risk of developing heart failure; indeed,

General Features of db/db Mice after Treatment with Different Zn Amounts
Mice were fed normal diet with different amounts of Zn, including Zn deficient (ZD), Zn adequate (ZN), Zn supplemented (ZS), and ZS for the first three months and switched to ZN (ZS-N) groups, starting at the age of 17 weeks. Non-fasting blood glucose levels ( Figure 1A) and blood GHbA1c ( Figure 1B) were increased in db/db mice compared with WT animals. Insulin resistance was assessed by intraperitoneal glucose tolerance test (IPGTT) with injection of 2 g/kg body weight, and blood glucose was assessed at the six month time point ( Figure 1C), followed by AUC determination ( Figure 1D). db/db mice showed significantly increased IPGTT, as reflected by elevated AUCs at six months ( Figure 1C,D). Interestingly, ZD worsened the db/db mouse condition, further increasing non-fasting blood glucose levels, blood GHbA1c, and IPGTT, while ZS and ZS-N showed no significant changes compared with the ZN group. Calorie intake was not significantly changed based on daily food consumption. groups, starting at the age of 17 weeks. Non-fasting blood glucose levels ( Figure 1A) and blood GHbA1c ( Figure 1B) were increased in db/db mice compared with WT animals. Insulin resistance was assessed by intraperitoneal glucose tolerance test (IPGTT) with injection of 2 g/kg body weight, and blood glucose was assessed at the six month time point ( Figure 1C), followed by AUC determination ( Figure 1D). db/db mice showed significantly increased IPGTT, as reflected by elevated AUCs at six months ( Figure 1C,D). Interestingly, ZD worsened the db/db mouse condition, further increasing non-fasting blood glucose levels, blood GHbA1c, and IPGTT, while ZS and ZS-N showed no significant changes compared with the ZN group. Calorie intake was not significantly changed based on daily food consumption. ; seven C57BL/6J mice were fed ZN as the control group (WT). All mice were fed from the age of 17 weeks, and sacrificed six months later (as 6M). Nonfasting glucose levels (A) and blood GHbA1c (g/L) (B) were evaluated. IPGTT (C) at six months and related area under curve (AUC) values (D) were assessed. Zn levels in the liver were measured in WT and diabetic mouse groups (E). Data were presented as mean ± SD (n = 7 for WT control and db/db-ZD (db/ZD) groups; n = 6 for db/db-ZN (db/ZN) group; n = 4 for db/db-ZS (db/ZS) and db/db-ZS-N (db/ZS-N) groups). Differences were assessed by ANOVA with Tukey-Kramer post hoc analysis. *, p < 0.05 vs. WT group; #, p < 0.05 vs. db/ZN group.

Figure 1.
Systemic effects of different Zn amounts in db/db mice. Twenty-one B6.BKS(D)-Leprdb/J mice were randomly divided into four groups and fed normal chow (10% Cal from fat) with different amounts of Zn (ZD, ZN, ZS and ZS-N, respectively); seven C57BL/6J mice were fed ZN as the control group (WT). All mice were fed from the age of 17 weeks, and sacrificed six months later (as 6M). Non-fasting glucose levels (A) and blood GHbA1c (g/L) (B) were evaluated. IPGTT (C) at six months and related area under curve (AUC) values (D) were assessed. Zn levels in the liver were measured in WT and diabetic mouse groups (E). Data were presented as mean ± SD (n = 7 for WT control and db/db-ZD (db/ZD) groups; n = 6 for db/db-ZN (db/ZN) group; n = 4 for db/db-ZS (db/ZS) and db/db-ZS-N (db/ZS-N) groups). Differences were assessed by ANOVA with Tukey-Kramer post hoc analysis. *, p < 0.05 vs. WT group; #, p < 0.05 vs. db/ZN group.

Zn Levels in Cardiac and Liver Tissues
Because cardiac tissue and serum specimens were limited, we measured hepatic Zn levels. As shown in Figure 1E, hepatic Zn levels in db/db mice were decreased significantly compared with the amounts of WT mice. Compared with the ZN group, the ZD group showed further decreased levels of hepatic Zn, while no significant change was found in the ZS and ZS-N groups.

Cardiac Hypertrophy and Function
Echocardiographic analysis showed that EF% (Figure 2A) in db/db mice decreased significantly, with markedly increased LV mass ( Figure 2B) and heart weight to tibial length ratio ( Figure 2C). ANP protein expression levels were significantly higher in db/db mice than in WT animals ( Figure 2D). All these parameters were significantly worsened by ZD, compared with the ZN group, but not significantly changed in the ZS and ZS-N groups.

Zn Levels in Cardiac and Liver Tissues
Because cardiac tissue and serum specimens were limited, we measured hepatic Zn levels. As shown in Figure 1E, hepatic Zn levels in db/db mice were decreased significantly compared with the amounts of WT mice. Compared with the ZN group, the ZD group showed further decreased levels of hepatic Zn, while no significant change was found in the ZS and ZS-N groups.

Cardiac Hypertrophy and Function
Echocardiographic analysis showed that EF% (Figure 2A) in db/db mice decreased significantly, with markedly increased LV mass ( Figure 2B) and heart weight to tibial length ratio ( Figure 2C). ANP protein expression levels were significantly higher in db/db mice than in WT animals ( Figure 2D). All these parameters were significantly worsened by ZD, compared with the ZN group, but not significantly changed in the ZS and ZS-N groups. Zn deficiency exacerbates diabetic-induced heart hypertrophy and function. Animals were treated as described in Figure 1. EF% (A) and corrected Left ventricular (LV) mass (mg/g) (B) were examined by echocardiography. Heart-weight to tibial length ratio (C) and atrial natriuretic peptide (ANP) protein levels obtained by Western Blot (D) were assessed as indicators of heart hypertrophy and function. Data were presented as mean ± SD (n = 4-7, details in Figure 1). *, p < 0.05 vs. WT group; #, p < 0.05 vs. db/ZN group. Zn deficiency exacerbates diabetic-induced heart hypertrophy and function. Animals were treated as described in Figure 1. EF% (A) and corrected Left ventricular (LV) mass (mg/g) (B) were examined by echocardiography. Heart-weight to tibial length ratio (C) and atrial natriuretic peptide (ANP) protein levels obtained by Western Blot (D) were assessed as indicators of heart hypertrophy and function. Data were presented as mean ± SD (n = 4-7, details in Figure 1). *, p < 0.05 vs. WT group; #, p < 0.05 vs. db/ZN group.

Zn Prevents Cardiac Fibrosis in db/db Mice
Sirius Red Staining followed by semi-quantitative analysis ( Figure 3A) indicated increased fibrosis in all four db/db groups compared with WT controls. Among the db/db mice, the db/db-ZD group had worse outcome compared with db/db-ZN animals, while the ZS and ZS-N groups showed no significant change in terms of cardiac fibrosis severity compared to the ZN group. Western blot demonstrated that TGF-β and CTGF amounts in the diabetic groups were significantly higher compared with WT group levels ( Figure 3B). Meanwhile, db/db-ZD animals had significantly higher expression levels of TGF-β and CTGF compared with the db/db-ZN group. However, the ZS, ZN and ZS-N groups showed similar TGF-β and CTGF levels.

Zn Prevents Cardiac Fibrosis in db/db Mice
Sirius Red Staining followed by semi-quantitative analysis ( Figure 3A) indicated increased fibrosis in all four db/db groups compared with WT controls. Among the db/db mice, the db/db-ZD group had worse outcome compared with db/db-ZN animals, while the ZS and ZS-N groups showed no significant change in terms of cardiac fibrosis severity compared to the ZN group. Western blot demonstrated that TGF-β and CTGF amounts in the diabetic groups were significantly higher compared with WT group levels ( Figure 3B). Meanwhile, db/db-ZD animals had significantly higher expression levels of TGF-β and CTGF compared with the db/db-ZN group. However, the ZS, ZN and ZS-N groups showed similar TGF-β and CTGF levels. . Zn deficiency exacerbates diabetes-induced cardiac fibrosis. Sirius Red Staining (A) was used to assess fibrosis in the heart, Scale bar = 25 µM. Western blot was used to evaluate fibrosis related factors (B), including transforming growth factor β (TGF-β) and connective tissue growth factor (CTGF). Data were presented as mean ± SD (n = 4-7, details in Figure 1). *, p < 0.05 vs. WT group; #, p < 0.05 vs. db/ZN group. Zn deficiency exacerbates diabetes-induced cardiac fibrosis. Sirius Red Staining (A) was used to assess fibrosis in the heart, Scale bar = 25 µM. Western blot was used to evaluate fibrosis related factors (B), including transforming growth factor β (TGF-β) and connective tissue growth factor (CTGF). Data were presented as mean ± SD (n = 4-7, details in Figure 1). *, p < 0.05 vs. WT group; #, p < 0.05 vs. db/ZN group.

Zn Prevents Cardiac Inflammation in db/db Mice
Given that hyperglycemia induces chronic inflammation as a causative factor of cardiac remodeling, the expression levels of inflammatory markers, including p-P38, p38, TNF-α, IL-1β ( Figure 4A), CARD9, and BCL10 ( Figure 4B), were examined by Western blot. Interestingly, diabetic mice showed significantly increased expression levels of these proteins in cardiomyopathy at the 6M time-point; these changes were even more notable in the ZD group. However, Zn supplementation (ZS and ZS-N) alleviated these effects, and the amounts of the above inflammatory factors were similar to ZN group values.

Zn Prevents Cardiac Inflammation in db/db Mice
Given that hyperglycemia induces chronic inflammation as a causative factor of cardiac remodeling, the expression levels of inflammatory markers, including p-P38, p38, TNF-α, IL-1β ( Figure 4A), CARD9, and BCL10 ( Figure 4B), were examined by Western blot. Interestingly, diabetic mice showed significantly increased expression levels of these proteins in cardiomyopathy at the 6M time-point; these changes were even more notable in the ZD group. However, Zn supplementation (ZS and ZS-N) alleviated these effects, and the amounts of the above inflammatory factors were similar to ZN group values.

Zn Attenuation of Cardiac Oxidative Stress Is Probably Associated with Nrf2 Activation to Upregulate Downstream Antioxidants
Considering that inflammation is often accompanied by oxidative stress, such damage was evaluated by Western blot for 3-NT ( Figure 5A) and 4-HNE ( Figure 5B); significantly increased amounts of these proteins were found in diabetic mice at the 6M time-point. ZD exacerbated cardiac oxidative stress in diabetic mice, while the ZS and ZS-N groups showed 3-NT and 4-HNE levels similar to ZN group values.

Zn Attenuation of Cardiac Oxidative Stress Is Probably Associated with Nrf2 Activation to Upregulate Downstream Antioxidants
Considering that inflammation is often accompanied by oxidative stress, such damage was evaluated by Western blot for 3-NT ( Figure 5A) and 4-HNE ( Figure 5B); significantly increased amounts of these proteins were found in diabetic mice at the 6M time-point. ZD exacerbated cardiac oxidative stress in diabetic mice, while the ZS and ZS-N groups showed 3-NT and 4-HNE levels similar to ZN group values. To explore the mechanism behind the protective effects of Zn toward diabetic-induced pathological changes in the mouse heart, nuclear transcription factor (Nrf2) expression was analyzed at the protein and gene levels, respectively, by Western blot and RT-PCR. Western blot showed that db/db mice displayed significantly decreased amounts of Nrf2 and the downstream catalase (CAT) and NQO1 proteins in cardiac tissues. Zn deficiency aggravated these changes, while the ZS and ZS-N groups showed similar protein levels of Nrf2, CAT, and NQO1 to the ZN group ( Figure 6A). At Data were presented as mean ± SD (n = 4-7, details in Figure 1). *, p < 0.05 vs. WT group; #, p < 0.05 vs. db/ZN group.
To explore the mechanism behind the protective effects of Zn toward diabetic-induced pathological changes in the mouse heart, nuclear transcription factor (Nrf2) expression was analyzed at the protein and gene levels, respectively, by Western blot and RT-PCR. Western blot showed that db/db mice displayed significantly decreased amounts of Nrf2 and the downstream catalase (CAT) and NQO1 proteins in cardiac tissues. Zn deficiency aggravated these changes, while the ZS and ZS-N groups showed similar protein levels of Nrf2, CAT, and NQO1 to the ZN group ( Figure 6A). At the transcriptional level, mRNA amounts of Nrf2, catalase and NQO-1 were significantly decreased in diabetic mice of the ZD group, while the ZS and ZS-N groups showed increased mRNA levels of these genes ( Figure 6B). the transcriptional level, mRNA amounts of Nrf2, catalase and NQO-1 were significantly decreased in diabetic mice of the ZD group, while the ZS and ZS-N groups showed increased mRNA levels of these genes ( Figure 6B).  Figure 1). *, p < 0.05 vs. WT group; #, p < 0.05 vs. db/ZN group.

Discussion
Both diabetes and Zn deficiency are global health problems [1,11,20]. Diabetic patients often suffer from Zn deficiency at the late disease stage, particularly those whose glucose is poorly controlled [21][22][23][24]. Zn supplementation has beneficial effects on glucose and lipid control [25]. In the present study, an animal model of T2DM with leptin receptor defect was used to demonstrate that  Figure 1). *, p < 0.05 vs. WT group; #, p < 0.05 vs. db/ZN group.

Discussion
Both diabetes and Zn deficiency are global health problems [1,11,20]. Diabetic patients often suffer from Zn deficiency at the late disease stage, particularly those whose glucose is poorly controlled [21][22][23][24]. Zn supplementation has beneficial effects on glucose and lipid control [25]. In the present study, an animal model of T2DM with leptin receptor defect was used to demonstrate that Zn deficiency significantly exacerbates obesity-induced cardiac oxidative damage, inflammation, and fibrosis in diabetes, effects associated with significantly decreased expression of Nrf2 and the downstream antioxidants NQO-1 and CAT. These changes were exacerbated by Zn deficiency, but not affected by Zn supplementation in db/db mice. As an adaptive mechanism, Nrf2 is quickly upregulated in cells and tissues in response to various oxidative stresses, but downregulated at late stage after exposure to chronic oxidative stress [17,[26][27][28]. In the current study, cardiac Nrf2 expression levels were decreased in diabetic mice after six months on HFD, probably because the diabetes duration was long enough. Moreover, diabetic mice with Zn deficiency showed further downregulation of Nrf2 and the downstream effectors NQO-1 and catalase. More importantly, Zn deficiency-exacerbated Nrf2 downregulation resulted in severely increased oxidative damage, cardiac inflammation, and fibrosis. Meanwhile, an important finding was that Zn deficiency aggravated diabetes-induced pathogenic changes, in association with further Nrf2 downregulation in the T2DM model.
However, other studies demonstrated that CARD9 mediates the activation of p38 MAPK, which is pivotal in multiple immune responses and inflammation activation [34][35][36]. CARD9 signaling allows Toll like receptor (nucleotide-binding oligomerization domain) pathways to induce p38 MAPK activation [34,37]. Previous studies reported that CARD9 knockout attenuates cardiac dysfunction by abrogating increased p38 MAPK phosphorylation in obese mice, as well as cardiac inflammation and fibrosis after angiotensin II treatment [31,38]. The present study demonstrated that Zn deficiency exacerbated BCL10 and CARD9 expression changes in db/db mice, although there were no significant differences among the ZN, ZS, and ZS-N groups.
Our team and others have reported that overexpression of metallothionein, catalase, and manganese superoxide dismutase in the heart reverses diabetic cardiomyopathy in animal models of both T1DM and T2DM [39][40][41][42]. Thus, strategies that either reduce ROS or augment myocardial antioxidant defense mechanisms might have therapeutic efficacy in improving myocardial function in diabetes. As shown above, db/db mice displayed significantly decreased levels of Nrf2 and downstream antioxidants, including catalase, compared with WT control mice. Among the db/db mice, the expression levels of Nrf2 and downstream antioxidants were lowest in the ZD group, which may be the main reason why Zn deficiency exacerbates obesity/T2DM-induced cardiac remodeling and dysfunction. Indeed, low antioxidant capacity may lead to high oxidative stress that induces BCL10/CARD9-mediated p38 MAPK activation, resulting in cardiac inflammation and remodeling, and even dysfunction. In addition, since the ZD group also showed worst outcomes of blood glucose and GHbA1c levels as well as insulin resistance, the worsening effects of ZD on these systemic alterations may also be in part responsible for the aggravated development of DCM.
It should be mentioned that, in the present study, Zn supplementation to db/db mice did not exert any beneficial effects on T2DM-induced-cardiac pathogenesis compared to the ZN group. In contrast, we recently demonstrated that HFD induces cardiac hypertrophy along with inflammation, which were exacerbated and attenuated by Zn deficiency and supplementation, respectively, compared with ZN groups, in HFD-induced obese WT mice [15]. The discrepancy between the two studies suggests that Zn supplementation-mediated cardiac protection from obesity and/or T2DM is likely dependent on leptin-mediated signaling, as shown by the impact of Zn on leptin levels [18,19]; this will be further explored in future studies.

Animals
B6.BKS(D)-Leprdb/J (db/db) and C57BL/6J mice were obtained from the Jackson Laboratory (Bar Harbor, ME, USA), and housed in the University of Louisville Research Resources Center at 22 • C under a 12:12 h light/dark cycle, with tap water and rodent diet ad libitum. All experimental procedures were approved by the Institutional Animal Care and Use Committee of the University of Louisville (3 April 2015), and carried out in accordance with the Guide for the Care and Use of Laboratory Animals, Eighth Edition (Library of Congress Control Number: 2010940400, revised 2011). Twenty-one B6.BKS (D)-Leprdb/J mice were randomly divided into four groups. The first 17 mice were fed a normal diet (10% calories from fat) with different amounts of Zn, including 10 mg (deficient, ZD; n = 7), 30 mg (adequate, ZN; n = 6), and 90 mg (supplemented, ZS; n = 4) per 4057 Kcal, respectively, for six months. The fourth group of four db/db mice were fed ZS for the first three months and switched to ZN for the subsequent three months, the same as the db/db/ZS-N group. Meanwhile, seven C57BL/6J mice were fed a ZN diet for six months as a control group. After six months, all mice were sacrificed at 6M.

Zn Concentration Assessment in the Liver Tissue
Each liver sample (30 mg wet-weight) was digested with 1 mL 70% concentrated nitric acid, in an 85 • C water bath for 3 h. After digestion, each sample was diluted 35 times. The digested samples were then filtered using a PTFE 0.2 mm filter. Liver Zn levels were assessed by Atomic Absorption Spectroscopy (AAS) on an iCE-3000 AAS instrument from Thermo Fisher Scientific (Waltham, MA, USA). Zn levels were calculated based on a standard curve and presented as ng/mg wet tissue.

Sirius Red Staining
Heart fibrosis was assessed by Sirius red staining for collagen, with a mixture of 0.1% Sirius red F3BA and 0.25% Fast Green FCF. Collagen amounts in the myocardium were evaluated using Pro Plus 6.0 software (Media Cybernetics Inc., Bethesda, MD, USA).

Quantitative Real-Time PCR
Total RNA was extracted from heart tissues with TRIzol-reagent (RNA STAT60 Tel-Test; Austin, Texas, USA). RNA amounts and purity were determined on a Nanodrop ND-1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA). Random-primed strand complimentary DNA (cDNA) was synthesized from total RNA using GoScript Reverse Transcription System (Promega, Madison, WI, USA) following the manufacturer's instructions. Primers for NQO-1 (Mm01253561_m1), catalase (Mm00437992_m1), Nfe2/2 (Mm00477784_m1), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Mm99999915_g1) were purchased from Thermo Fisher Scientific. Quantitative RT-PCR (qRT-PCR) was performed in duplicate in a 20 µL reaction system comprising 10 µL of TaqMan Universal PCR Master Mix, 1 µL of each primer, and 3 µL of cDNA, on an ABI 7500 RT-PCR system (Applied Biosystems, Foster City, CA, USA). C t values were obtained, and relative gene expression was assessed by the 2 −∆∆Ct method, with GAPDH used for normalization.

Statistical Analysis
Data are mean ± standard deviation (SD) with 4-7 animals per group. Group comparisons were performed by one-way analysis of variance (ANOVA), followed by pairwise repetitive comparisons with Tukey test, using the Origin 8.0 Lab data analysis and graphing software. p < 0.05 was considered statistically significant.

Conclusions
In conclusion, this is the first study to our knowledge providing evidence regarding the relationship between DCM and long-term Zn treatment in T2DM animal models. We confirmed that Zn deficiency contributes to the pathological progression of cardiac disorders induced by T2DM, and revealed the partial protective effect of long-term Zn supplementation on DCM. These findings confirm the significant role of Zn in preventing diabetic related cardiac disorders in T2DM mouse models. Optimal Zn supplementation dose for T2DM mice could be identified to prevent DCM. Overall, Zn supplements might be highly potent in the prevention and/or treatment of T2DM related DCM.

Conflicts of Interest:
The authors declare no conflict of interest.