Iron Reduces the Anti-In ammatory Effect of Omega-3 Polyunsaturated Fatty Acids on the Heart of STZ- and HFD-Induced Diabetic Rats

Maryam Khajvand-Abedini Hamadan University of Medical Sciences Medical School Nasrin Ziamajidi Hamadan University of Medical Sciences Medical School Alireza Nourian Bu-Ali Sina University: Bu Ali Sina University Mahdi Bahmani Hamadan University of Medical Sciences Medical School Roghayeh Abbasalipourkabir (  abbasalipourkabir@umsha.ac.ir ) Hamadan University of Medical Sciences Medical School https://orcid.org/0000-0002-6345-1410


Introduction
Diabetes mellitus, as one of the biggest health problems worldwide, is characterized by hyperglycemia resulted from abnormalities in insulin secretion or action (1). One of the harmful effects of hyperglycemia is causing cardiovascular complications associated with metabolic changes such as endothelial dysfunction as well as mild changes in myocardial structure (2). Long-term hyperglycemia causes a nonenzymatic reaction in glucose along with high-half-life protein molecules and lipid molecules, which is called the glycation process. Accordingly, Glycation in the pathogenesis of long-term complications of diabetes plays an important role through the recipient of the advanced glycation products (RAGEs) (3). Oxidative stress can drive glycation, while during a faulty cycle AGE (Advanced Glycation End Products), its level also increases by accelerating the oxidation process (4).
There are lots of reports on that coronary artery disease and cardiovascular complications along with endothelial dysfunction, in ammation, and heart brosis are common in the diabetic population (5). Cardiac in ammation is associated with an increment in the expressions of intercellular adhesive molecules (ICAM-1) and vascular adhesive molecules (VCAM-1) (6). Thus, the presence and accumulation of AGEs in diabetes through the formation of cross-linkages in collagen molecules could potentially accelerate the procedures of endothelial dysfunction, myocardial thickness and brosis, and oxidative stress, which consequently lead to damage to internal myocardial cells (7,8). Therefore, increasing the level and activity of AGE increases RAGE, reactive oxygen species (ROS), expressions of ICAM-1 and VCAM-1, and in ammatory cytokines like tumor necrosis factor-α (TNF-α), which eventually result in cardiac complications in diabetes (9).
Since the long-term usage of anti-diabetic medications has several diverse side effects, during the discovery of anti-diabetic drugs, researchers have turned their attention to those dietary supplements with minimal side effects (10). Recent studies have recommended the administrations of omega-3 polyunsaturated fatty acids (ω-3 PUFAs), docosahexaenoic acid (DHA, 22:6n-3), and eicosapentaenoic acid (EPA, 20:5n-3) supplements, as effective anti-in ammatory agents on controlling and improving metabolic disorders, for patients with type 2 diabetes mellitus (T2DM) (11,12).
The high sensitivity of T2DM patients to systemic in ammation, drugs, and the altered iron metabolism, which are known as risk factors for anemia in these patients, increases the risk of developing anemia in them. Additionally, due to poor diet and nutritional de ciencies of cyanocobalamin, folate, and iron, T2DM patients are more prone to nutritional anemia (13). This complication occurs simultaneously with the development of diabetes, partly due to the impaired iron absorption, which lead to renal and cardiovascular abnormalities in diabetic patients (14,15). Therefore, common ferrous sulfate supplementation to manage iron de ciency anemia in diabetes causes molecular damages, which are due to the catalytic role of this supplement in the reaction of Fenton and the resulting free radicals (16).
However, the results of some in vitro studies have shown that the concomitant use of bivalent metal and ω-3 PUFAs supplements can exacerbate the level of in ammation in T2DM patients, because in the presence of glucose, proteins, and divalent metals, especially for iron, ω-3 PUFAs increase the productions of AGEs and malondialdehyde (17,18). Several studies have shown that the productions of AGEs, oxidative stress, and other body peroxidants can be effective on increasing in ammation and dysfunction of heart tissue. Due to the lack of adequate information on the effects of ω-3 PUFAs and iron supplements on the development of cardiac complications in diabetic patients or animal models of T2DM, in this study, we aimed to examine the effect of the concomitant use of iron and ω-3 PUFA supplements on oxidative stress and in ammation as well as the expressions of VCAM-1 and ICAM-1 factors in the heart tissue of the animal model T2DM.

Experimental design
Thirty Wistar male rats aged between 8 and 10 weeks and weighing between 180 and 220 g were prepared from the animal house of Hamadan University of Medical Sciences. The animals were then wellmaintained within the animal house under standard condition (at 22-24 °C with 12:12 hours light and dark cycle) with free access to food and water. Subsequently, the animals were randomly divided into ve groups (n = 6 in each group) as follows: Groups 2 to 5 received high-fat diet with 60% fat (HFD) for 8 weeks and a single-dose of 40 mg/kg streptozotocin (STZ) in citrate buffer with pH = 4.5 intraperitoneally after ve to six hours fasting (19). Group 1 has also received a normal diet for 8 weeks. Fasting blood sugar (FBS) obtained from tail blood sampling was then measured at 12 and 72 hours after the STZ administration using a strip-operated blood sugar sensor (Accuchek; Roche, Germany). Animals with FBS > 250 mg/dL were considered as T2DM model. The treatment regimen used in this study was based on a 40-day period. Accordingly, this regimen included daily gavage of moderate doses of iron sulfate (35 mg/kg/day (16, 20) and ω-3 PUFAs supplements (45 mg/kg EPA and 30 mg/kg DHA) (21), in the designed groups and groups 1 and 2 have received normal saline solution. Thereafter, the rats were sacri ced and their hearts were dissected, washed rapidly with cold phosphate-buffered saline (PBS, pH 7.4), frozen in liquid nitrogen, and nally kept at -70 °C (16, 21). The composition of high-fat diet is presented in Table 1 (22).

Preparation of cardiac homogenates
About 100 mg of heart tissue was weighed, washed three times with ice-cold PBS, and then homogenized with liquid nitrogen. The obtained mixture was re-suspended in ice-cold lysis buffer (10 mM (4-(2hydroxyethyl)-1-piperazineethanesulfonic acid), 10 mM KCl, 1.5 mM MgCl2, 1 mM EDTA, 0.1% Triton X100, and protease inhibitor cocktail, pH = 7.9) and then incubated on ice for 20 min. Thereafter, homogenate was centrifuged at 10000 g for 15 min at 4 °C to remove tissue remnants. The supernatant was nally restored at -70 °C until measuring the TAC & TOS parameters (23).

TOS assay
The determination of the TOS level was conducted in the obtained tissue samples using the ferrous ion oxidation xylenol orange (FOX1) method. Notably, the basis of this method is the oxidation of ferrous ions to ferric ions in the presence of oxidizing agents and under acidic conditions. Finally, the ferric ions formed by the xylene orange produced a color complex, which was detected at a wavelength of 560 nm (24).

TAC assay
The TAC level was determined in the obtained tissue samples using the ferric reducing antioxidant potential (FRAP) method. Notably, the basis of this method is the reduction of ferric tripyridyl triazine (FeIIITPTZ) to ferropyridyl triazine (FeIITPTZ) in the presence of biological antioxidant agents. Finally, ferrous pyridyl triazine (FeIITPTZ) formed a blue complex whose absorption was compared to that of a known standard (FeSO 4 .7H 2 O) at 593 nm (25).

Lipid peroxidation assay
The determination of aldehydes produced during lipid peroxidation was conducted in the obtained homogenate in terms of the manufacturer's instructions by TBA (thiobarbituric acid) reaction (26).
Notably, the basis of this method is the reaction of MDA, which is the end product of lipid peroxidation, along with TBA to produce a color complex. Finally, the TBA assay kit (Kiazist, Iran) was used in terms of the manufacturer's instructions, and the MDA value was then reported as nanomoles per mg protein.

Total protein assay
Protein content was determined in the homogenized tissue used for each assay using the Bradford method. Correspondingly, in this method, 750 µL of Kumasi Blue was mixed with 50 µL of the samples and then incubated for 10 minutes at room temperature. At last, the absorbance of the samples was measured at 595 nm. A standard curve for bovine serum albumin (BSA) was also drawn up (27).

ELISA analysis of TNF-α
After preparing of heart homogenate, TNF-α protein concentration was determined in terms of the manufacturer's instructions using Rat TNF-α ELISA MAX™ Deluxe set kit (BioLegend, San Diego, USA).

Measurement of heart total cholesterol and triglyceride content
The heart tissue lipid of the animals was measured by the method described by Folch et al. (1957) (28). After the homogenization of the tissue (≈ 50 mg) in liquid nitrogen and radio-immunoprecipitation assay (RIPA) buffer, the tissue sample was re-suspended in chloroform-methanol solution (1: 2) with a nal volume of 20 folds of the weight. Afterward, the homogenate was centrifuged to recover the liquid phase, and the solvent was then washed with 0.9% NaCl (4 mL for a mixture of 20 mL volume). To separate the 2 phases, the mixture was centrifuged at 1000 g (10 min), the upper phase was removed, and the lower chloroform phase containing tissue lipids was evaporated in a vacuum containing N 2 gas. Triglyceride and cholesterol assay kits (Pars Azmun, Iran) were used to measure total cholesterol and triglyceride contents of heart tissue.

RAGE, TNF-α, ICAM-1, and VCAM-1 genes' expressions
Cardiac tissue RNA extraction was conducted using Kiazol reagent in terms of the manufacturer's protocol (Kiazist, Iran). Thereafter, the nano drops and 1% agarose gel electrophoresis were applied to ensure its quality. Some of the total RNA extracted (1-5 µg) were reversely transcribed into cDNA using a cDNA synthesis kit (CinaClon, Iran). Next, a combination of 2 µL of cDNA made with 10 µL of SYBR Green, 1 µL of the leading primer, and 1 µL ladder primer, was used for RT-PCR (The sequence of primers is presented in Table 2). Subsequently, real time-PCR (RT-PCR) was done using the Light Cycler®96 system (Roche, USA). The Ampli cation Protocol with 40-45 cycles is shown in Table 3. Table 2 The sequences of the primers utilized in the RT-PCR in this research.

Gene
Forward primer Reverse primer Product size  At the end of each cycle, the uorescence signal was recognized. During this process, the β-actin gene was utilized as an internal control. The relative expression of the gene was achieved by counting the number of PCR cycles (CT) using a comparative method CT (ΔCT, ΔCT = CT β−actin -CT target gene ). Method 2 − ΔΔCT was also used to evaluate the relative expression of each gene in the control and treatment groups.

Histological evaluation
The formalin-xed heart tissue samples were dehydrated in ascending ethanol concentrations, cleared in xylene, in ltrated, and then embedded in para n. They were then sectioned at 5 µm thick using a rotary microtome (Leica RM2255, Germany). Afterward, these sections were stained with hematoxylin and eosin (H&E), and studied independently by a pathologist using a light microscope (Olympus CX41, Japan) equipped with a digital camera (Olympus DP25, Germany).

Statistical analysis of the data
The experimental results are expressed as the means ± SEM. All the statistics were performed using the SPSS software version 16.0 (SPSS for Windows; http://www.spss.com) and GraphPad Prism version 8.0 (http://www.graphpad.com). To nd out the difference, a one-way analysis of variance (ANOVA) was used, followed by the Tukey test. The level of p < 0.05 was considered as statistically signi cant.

Results
Fasting blood sugar and body weight Before and after the treatment, body weight (BW) and FBS change of all rats were recorded. As shown in Tables 4 and 5, BW gain was not signi cantly different among the groups. Before the treatment, FBS has signi cantly increased in diabetic rats (p < 0.001) compared to the normal control group. After the treatment, both FBS and BW also revealed a signi cant (p < 0.001) increase in diabetics groups compared to the normal group. Co-treatment of ω-3 PUFAs and Fe 2+ supplements in the groups D.ω, D.F, and D.ω.F led to a signi cant (p < 0.01 and p < 0.001, respectively) reduction in BW compared to the normal control rats.

Evaluation of oxidative stress markers in cardiac homogenates
The results of TAC and TOS, as oxidative stress markers, and OSI (oxidative stress index) and MDA ( nal product of lipid peroxidation) are shown in Fig. 1 (A-D). As shown in Fig. 1-A, there is a signi cant decrease in the level of TAC tissue in the groups D and DF (p < 0.001) and the group D.ω.F (p < 0.05) compared to the group CN. Additionally, groups D and D.F exhibited a signi cant decrease in TAC compared to the group D.ω (p < 0.001 and p < 0.05, respectively).

Gene expression of cardiac in ammation markers
The mRNA levels of RAGE, TNF-α, ICAM-1, and VCAM-1 genes were determined in the hearts of all the studied groups' rats using QRT-PCR analysis. The groups D, D.F, and D.ω.F showed signi cantly increased gene expressions of RAGE (p < 0.001) and ICAM-1 (p < 0.05) in comparison with the group CN (Figs. <link rid=" g4">4</link>-A and 4-B). However, the groups D and DF had a signi cant increase in the expression of the TNF-α gene compared to the group CN (P < 0.001, Fig. 4-C). Besides, the expression of VCAM-1 gene has signi cantly increased in the groups D and D.F compared with the CN group (p < 0.001 Fig. 4-D).
Comparison of the group D.ω.F with the CN group also showed an increase in mRNA expressions of TNFα and VCAM-1 genes (p < 0.001 4-C and p < 0.01 4-D, respectively). In the expressions of RAGE, ICAM-1, TNF-α, and VCAm-1 genes, the group D had a signi cant increase compared to the group D.ω (p < 0.05).
In addition, the comparison of the D.ω.F group with the D.ω group showed a signi cant increase in TNF-α gene expression level (p < 0.01 Fig. 4-C) as well as a an increase in VCAM-1 gene expression level; however, it was not signi cant (p > 0.05; p = 0.096 Fig. 4-D). RAGE gene expression increase in the D.F group was not signi cant compared to the group D.ω (p > 0.05; p = 0.075 Fig. 4-A).

Discussion
According to the results of the present research, there is a notable increase in BW following HFD utilization. Moreover, it was found that some strains (Wistar and Sprague-Dawley rats) are more vulnerable to obesity when receiving HFD, which may be due to the hyperphagic nature of these animals that are resistant to the insulin function and the reduced hypothalamic peptides. Furthermore, the alterations in BW gain mostly depend on a large extent fat composition of the diet (29-31), the deposition of the saturated fats resulted from the consumption of a diet rich in energy in different layers of body fat due to the reduced energy consumption compared to animals fed with a normal diet (32,33). A non-signi cant BW loss was also shown in our results after the injection of STZ that can increase levels of leptin and adiponectin and a concurrency between adipogenesis and lipolysis in the diabetic rats (34,35). The reduced BW in diabetic rats in the present study is consistent with the results of the Gundala's study in 2019 (36). It was suggested that ω-3 PUFA enhances lipid oxidation through the activation of peroxisome proliferator-activated receptor α (PPAR α) and suppression of lipogenic enzymes, which make obesity resistance (37,38 (39) in their study reported that iron can cause loss of BW in HFD/STZ-treated rats compared to the control group. However, its precise mechanism has not been determined yet (39). Sahin et al. 2007 (40) suggested that the HFD/STZ model can be considered as a novel animal model for the induction of T2DM and the study of antidiabetic and anti-in ammatory compounds (40).  reported a signi cant increase in FBS in diabetic rats after the STZ injection compared to the normal control group (34). This is likely due to compensatory hyperinsulinemia to control glucose homeostasis (41). After the treatment period, FBS levels were signi cantly higher in diabetic groups compared to the normal control group. Besides, a non-signi cant decrease in FBS levels were observed in the D.ω group. The positive effect of ω-3 PUFA on increasing glycolysis and glycogenesis and its negative effect on lowering FBS levels via increasing insulin resistance and exacerbating glucose intolerance have also been reported (42) (43-46).
The mechanisms of STZ include elective devastation of pancreatic beta cells and a decreased activity of insulin secretion (47,48). It was shown that chronic hyperglycemia leads to the formation of AGE through the glycation process of long half-life proteins and lipids. Accordingly, the accumulation of these compounds and their binding to RAGE cause cross-linking in collagen molecules, which consequently increase myocardial thickness, oxidative stress, and brosis, and eventually damage myocardial cells (7,8). Furthermore, it has been reported that oxidative stress can lead to the development of the glycation process. In contrast, it was found that AGE increases its level by accelerating the oxidation process (4).
Thus, the oxidative stress caused by the binding of AGE to its receptor, in turn, leads to numerous pathological changes in the gene expression and impairs the structure and function of the heart in diabetic patients (49,50). According to the ndings of the present study, the increased OSI was observed in the heart tissues of the groups D, D.F, and D.ω.F compared to the CN group. Moreover, there was a signi cant increase in the OSI level in the groups D, D.F, and D.ω.F compared to the D.ω group. Furthermore, the administration of ω-3 PUFA as a supplement, signi cantly reduced OSI compared to the group D.
In line with the ndings pf a study by Flavia et al (2014), the effective role of iron supplements in the high production of ROS was one of the major mechanisms proposed to explain cardiac changes, myocardial cell death, and cardiac brosis in diabetic rats (51). Ginty and Conklin (2012) suggested that the ω-3 PUFA diet may improve the antioxidant capacity of diabetic rats (52). In the present study, it was expected that iron supplementation would have a negative outcome on the antioxidant effect of ω-3 PUFA supplementation. After comparing the group D.ω.F with the group D.F, it was found that the level of OSI has more signi cantly increased. Therefore, concurrent use of both supplements in diabetics can reduce the oxidative capacity, may be due to the negative effect of free radicals, resulted from the reactions of the divalent metal iron, on the antioxidant properties of ω-3 PUFA. In contrast, some studies have shown that diets containing sh oil, besides increasing cellular oxidative damage, produce antioxidant effects (30).
Based on the evidence, ferrous sulfate causes damage to cell membranes and their lipoproteins. This damage occurs through high production of toxic hydroxyl radicals resulted from the Fenton/Haber-Weiss reaction, followed by the free radical chain reaction, called lipid peroxidation, and the increased production of malondialdehyde (53). It is noteworthy that some factors like AGEs cause a substantial rise in PUFA peroxidation and MDA production in the presence of Fe + 2 and Fe + 3 (54). In the present study, it was found that the groups D, D.F, and D.ω.F had a signi cant increase in MDA level, as a lipid peroxidation index, in comparison with the group CN, probably due to the effect of divalent metals on raising the amount of lipid peroxidation. However, there was a non-signi cant decrease in lipid peroxidation in the group D.ω compared to the group D. Additionally, a study by Wei Bao et al. (2012) showed that there is no association between dietary or supplementation of iron and the risk of T2DM (55).
According to the results, gene expression of RAGE signi cantly increased in the heart tissues of the group D compared to the group CN. In addition, compared to CN group, the induction of T2DM in group D more increased gene expression and tissue level of TNF-α, as an in ammatory cytokine, as well as gene Nevertheless, prescribing iron and vitamin supplements as well as the supplements containing DHA and EPA, as anti-in ammatory agents, to T2DM patients are recommended to control and improve metabolic disorders effectively (11,12). Omega-3 lowers obstruction, in ammation, brosis, and hypertrophy caused by heart disease through disrupting JNK signaling, reducing TNF-α expression, and producing molecules involved in cardiac in ammation such as ICAM-1 and VCAM-1 (59-61). Besides, the expression of genes and tissue level of the TNF-α have signi cantly induced in the D.ω.F group compared to the D.ω group. In the current study, it was found that only the D.ω group signi cantly decreased gene expressions of RAGE and TNF-α and tissue level of TNF-α protein. It was also found that the gene expressions of RAGE and TNF-α, and tissue level of TNF-α protein signi cantly increased in the groups D.F, D.ω.F, CN, and D.F compared to the group D.ω. Although the administration of omega-3 alone decreased genes expression of ICAM-1 and VCAM-1 compared to the group D, co-treatment with Fe and omega-3 in diabetic rats led to a signi cant increase and a non-signi cant increase in the levels of genes ICAM-1 and VCAM-1, respectively. It was also found that in hyperglycemia, iron supplementation has a negative effect on the anti-in ammatory properties of ω-3 PUFAs by catalyzing their oxidation (62). In contrast, a study by Kosacka (2016) reported that a lack of iron in the diet caused a mild in ammation in diabetic rats (63).
A diabetic animal model with STZ-HFD induced mild hyperglycemia, hypertriglyceridemia, hypercholesterolemia, and conditions similar to insulin resistance in pre-diabetic humans. In this regard, it seems that the induction of diabetes mainly occurs through the glucose-fatty acid cycle or Randle cycle circulating FFAs levels can lead to a sudden and signi cant increase in insulin resistance through the accumulation of high rate of TG in the heart (68). The reduced insulin's role, as a regulator of lipid homeostasis in insulin resistance, is accompanied with the increased lipogenesis in hepatocyte and lipolysis in adipocytes leading to the increased blood FFA and TG as well as the impaired handling lipids levels in adipose tissue, liver, and muscle. Thus, insulin resistance leads to some metabolic and functional disorders of the heart and the increased insulin levels stimulate the transfer of FFAs to cardiomyocytes (69, 70). Subsequently, hyperlipidemia and compensatory hyperinsulinemia enhance the transfer and utilization of FFAs in myocytes. Accordingly, if myocytes receive FFAs neyond their oxidative capacity, they would be accumulated leading to lipotoxicity. This phenomenon increases cardiac dysfunction through various mechanisms such as ROS production and the increased insulin signaling disorder (68). Accordingly, the ndings of the lipid pro le in the heart tissues of the studied rats showed that the tissue levels of total cholesterol and triglyceride had a signi cant increase in the groups D and D.F compared to the group CN. Due to the wide application of iron supplements and multivitamins, and on the other hand, the extensive tissue damage in diabetes due to iron storage through high production of ROS, low iron intake in the diet is recommended to have a bene cial effect on lowering serum and tissue cholesterol and triglycerides levels as well as heart pathology.
Additionally, it was indicated that the increased liver iron levels increase cholesterol by producing glutathione, which increases the activity of HMG-CoA reductase, a rate-limiting enzyme in cholesterol biosynthesis. Therefore, the increased liver iron resulted from supplements is likely to cause lipemia, cardiac TG accumulation, and severe pathological changes (51,71,72). The ndings of the present study con rmed the possible effect of ω-3 PUFAs supplementation on lowering triglycerides and total cholesterol levels in type 2 diabetics, which is line with the ndings reported by Friday et al. (1989) and Rudkowska (2010) (44,73). Considering the relationship between blood FFAs levels and cardiac TGs accumulation (74), it is expected that ω-3 PUFAs are effective on reducing cardiac TGs through reducing plasma TG levels, improving lipemia by stimulating the intestine to reduce chylomicron secretion, increasing LPL activity and tissue blood ow, and improving the stimulatory effect of insulin and PPARs receptors on increasing chylomicron clearance (75,76). The present study revealed the reduced tissue levels of total cholesterol and TG in the group D.ω compared to the other diabetic groups; however, this reduction was only signi cant in comparison with the group D.F.
Histological ndings revealed some pathological changes caused by diabetes in the heart of animals.
Following the defects in the proteins that make up the hallmark of the heart muscle, the intercalated discs (IDs) as an intercellular communication surrounding, has emerged cardiomyopathy or other heart disorders (77,78). Besides, changes in the organization of the layered structure of heart tissue along with changes in the deformation and location of the myocytes nucleus, which is usually spindle-shaped and located along with the longitudinal axis or in the center of the cell, may indicate a change in the morphology of the heart muscle cell in diabetes (79,80). In conclusion, the ndings of this study con rmed the devastating effects of iron on the anti-in ammatory effect of omega-3 polyunsaturated fatty acids on the heart of diabetic rats

Conclusion
In people with T2DM, the in ammatory conditions caused by hyperglycemia can be reduced by ω-3 PUFAs as anti-in ammatory agents. It is likely that the increased peroxidation of ω-3 PUFAs in the presence of iron as well as high glucose concentrations may reduce their antioxidant properties.
Accordingly, it can be suggested that modifying the way of taking these supplements may minimize the prevalence of diabetes' complications, especially cardiovascular disease. However, more research is required to support this study.
Declarations Figure 1 The effect of treatments on the oxidative stress markers in cardiac tissue of studied animals (n= 6).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.