Protective Effect of High-Intensity Interval Training (HIIT) and Moderate-Intensity Continuous Training (MICT) against Vascular Dysfunction in Hyperglycemic Rats

Background Hyperglycemia is a major risk factor for endothelial dysfunction. Endothelial dysfunction is associated with the inability of endothelial cells to maintain homeostasis of the cardiovascular system. Regular exercise may be considered as an effective and low-cost nonpharmacological tool for improving vascular function, though there is no agreement on the best type of exercise. Objectives To determine how high-intensity interval training (HIIT) and moderate-intensity continuous training (MICT) may prevent endothelial dysfunction under hyperglycemic conditions, and to compare these two interventions. Method Twenty-four eight-week-old male Wistar rats were randomly assigned into four groups: healthy nonexercising control (C), hyperglycemic control (HG-C), hyperglycemic + HIIT (HG-IT), and hyperglycemic + MICT (HG-CT). Hyperglycemia was induced by a single injection of streptozotocin. Hyperglycemic animals were subjected to HIIT or MICT protocols six days a week for six weeks. Decapitation was performed the day after the exercise protocols were completed. The ascending aorta (until the abdominal artery) was examined. An enzyme-linked immunosorbent assay (ELISA) was used to measure the glucagon-likepeptide-1 (GLP-1), endothelial nitric oxide synthase (eNOS), and tumor necrosis factor-alpha (TNFα) levels. A colorimetric assay was used to measure superoxide dismutase (SOD) activity and malondialdehyde (MDA) levels. Quantitative real-time polymerase chain reaction (PCR) was used to measure the expression of the receptor for advanced glycation end-products (RAGE) and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB). Hematoxylin and eosin (H&E) staining was used to histologically analyze the aortas. Results There was a significantly higher level of GLP-1 and lower expression of RAGE, NF-κB, and TNFα in the HG-IT and HG-CT group compared to the HG-C group. Microscopic examination of aortic tissue showed a better tissue arrangement in both treatment groups than in the HG-C group. Except for the MDA level, there were no significant differences in any of the measured parameters between the HG-IT and HG-CT groups. Conclusion Under hyperglycemic conditions, both HIIT and MICT have a protective role against endothelial dysfunction.


Introduction
Diabetes mellitus (DM) is characterized by high blood sugar levels that are accompanied by impaired carbohydrate, lipid, and protein metabolism as a result of insulin function insufciency [1]. Data from the International Diabetes Federation (IDF) Diabetic Atlas showed that in 2019, 463 million people worldwide had DM, and the number is expected to increase to 700 million by 2045 [2].
DM poses a signifcant problem in the health sector and is associated with an increased risk of cardiovascular disease. Type 1 and type 2 DM are both signifcant risk factors for stroke, atherosclerosis, coronary artery disease (CAD), and peripheral arterial disease (PAD), which are responsible for over 80% of all deaths of people with DM [3]. Cardiovascular complications are the main cause of mortality in patients with type 2 DM (T2DM), and the development of cardiovascular complications caused by DM is triggered by endothelial dysfunction [4].
Vascular endothelial cells make a considerable contribution to the maintenance of cardiovascular system homeostasis through the secretion of various vasoactive agents. Tese include prostaglandin I 2 , endothelium-derived hyperpolarizing factor, nitric oxide (NO), angiotensin II, thromboxane A2, and endothelin-1 (ET-1). Disturbances in endothelial cells could lead to endothelial dysfunction, a condition marked by reduced NO activation or decreased NO production by endothelial nitric oxide synthase (eNOS) in endothelial cells [4].
Chronic hyperglycemia is now recognized as a key contributor to the development of vascular dysfunction in patients with DM. Hyperglycemia causes several changes in vascular tissue at the cellular level, all of which could accelerate the development of diabetic complications. Te three known main pathways in the pathogenesis of organ disorders under hyperglycemic conditions are the polyol pathway, the formation of advanced glycation end-products (AGEs) followed by overexpression of the receptor for AGEs (RAGE) gene, and activation of the protein kinase C (PKC) isoform via diacylglycerol (DAG) synthesis [5]. Tese pathways contribute to creating oxidative stress through increased production of reactive oxygen species (ROS), such as lipid peroxidase products, and decreased production of antioxidants, such as superoxide dismutase (SOD). Tis results in infammation that leads to organ dysfunction [5,6].
Te interaction between AGEs and RAGE can activate many signaling pathways, one of which is the release of proinfammatory mediators that trigger the accumulation of intracellular ROS. Te accumulation of ROS also activates the proinfammatory agent tumor necrosis factor-alpha (TNFα), which further induces nuclear factor kappa-light-chainenhancer of activated B cells (NF-κB) [7,8]. NF-κB activation is essential for early events of atherosclerotic lesion formation [9]. NF-κB acts as a transcription factor that regulates the expression of hundreds of genes that regulate critical physiological processes such as infammation, immunity, proliferation, and cell death. NF-κB controls both innate and adaptive immune response that plays a signifcant role in the initiation and development of atherosclerosis [10,11]. T cells, and macrophages, for instance, initiate activation of interferon (IFN)-c, a key trigger for the infammation process that leads to atherosclerosis formation [12].
Tere is great interest within the health industry in developing ways to manage endothelial dysfunction under hyperglycemic conditions to prevent cardiovascular complications. Te potential of incretin hormones (intestine secretion insulin), such as glucagon-likepeptide-1 (GLP-1), is one of the recent fndings. GLP-1 can increase the synthesis and phosphorylation activity of eNOS and thus increase NO production. GLP-1 also inhibits NF-κB activation through suppression of RAGE expression by increasing intracellular cAMP. Tis entails reducing binding between AGEs and RAGE, thereby reducing infammation and promoting endothelial regeneration. Te endothelial repair process prevents changes in blood vessel structure, such as the thickening of the tunica intima-media [13][14][15][16]. However, where and how GLP-1 afects the structure and function of vessels in hyperglycemic conditions have yet to be investigated.
Several studies have shown that exercise is a crucial intervention to prevent cardiovascular system complications associated with DM. In diabetic patients with insulin defciency, regular physical exercise has been shown to facilitate the entry of glucose into muscle cells, thereby increasing insulin sensitivity. Sports activities can increase glucosecarrying protein levels, thereby reducing insulin resistance [17].
According to a study by Gibala et al. [18], exercise intensity is the key factor infuencing peroxisome proliferatoractivated receptor gamma coactivator 1-alpha (PGC-1α) activation in human skeletal muscle. Te positive efects of the PGC-1α activation pathway include increased oxygen uptake, formation of antioxidants, increased oxidative capacity, and activation of anti-infammatory pathways. Physical exercise also increases angiogenesis and accelerates blood fow, leading to reduced vasoconstriction and the occurrence of hypoxia [19]. Physical exercise causes a decrease in the recruitment of adhesion molecules, thus reducing the infammatory process in the vascular wall [20]. Speretta [21] found that physical exercise positively afected the infammation parameters, body weight, adipocyte area, and lipid profle in obese rats, but the infuence of the diferent types of exercise gave diferent results on the analyzed parameters.
Aerobic exercise training is a well-established means for improving the vascular health of individuals with endothelial dysfunction. Most studies have shown that highintensity interval training (HIIT) improves endothelial function more than moderate-intensity continuous training (MICT). Consistent with these fndings, HIIT is reportedly more efective than MICT in increasing antioxidant status and NO availability. HIIT generates better blood fow than MICTdue to the large shear stress it induces, and shear stress induces eNOS to produce NO to protect against oxidative stress [22]. How HIIT and MICT difer in afecting endothelial function in the hyperglycemic condition through the increase of GLP-1 release remains elusive.
Te main purpose of this study was to compare and determine how HIIT and MICT may prevent endothelial dysfunction under hyperglycemic conditions. We compared the efects of both types of exercise training on GLP-1 level, RAGE gene expression, eNOS level, malondialdehyde (MDA) level, SOD activity, TNFα level, NF-κβ gene expression, and the histological appearance of aortic tissue from hyperglycemic rats.

Materials and Methods
Tis is an experimental study without blinding. Analysis was not performed by investigators blinded to the allocation of animals. Tis research was approved by the Health Research Ethics Committee of FKUI (number 1018/UN2.F1/ETIK/2017).

Animals.
Twenty-four male Wistar rats (Rattus norvegicus), aged eight weeks and weighing 180-300 g, were used in this study. Te rats were given standard food and drink ad libitum. Te cages were kept clean and set to 12 h of light and 12 h of darkness. Te ambient temperature was maintained at 23.1°C. Before the experiment, the rats were adjusted to the research environment for seven days. Te rats were randomly divided into four groups: the control group with no intervention (C), the hyperglycemic group without intervention (HG-C), the hyperglycemic group that underwent the HIIT program (HG-IT), and the hyperglycemic group that underwent the MICT program (HG-CT). Te rats in all the groups, except the C group, were injected intraperitoneal with a single dose of 40 mg/kg streptozotocin (STZ) (Sigma-Aldrich, Saint Louis, MO) dissolved in 0.1 M citrate bufer (pH 4.5) to induce hyperglycemia. Blood glucose was measured 72 h after injection, and blood was taken from the tail vein. Rats with a blood glucose level >200 mg/dL were considered hyperglycemic. All rats developed hyperglycemia 72 h after injection.

Exercise Procedures.
Physical exercise was performed six days a week, on a treadmill, with increasing speed, duration, and intensity each day. Te MICT started at a speed of 27 m/ min for 25 min, and the duration was increased every day (Table 1). Te HIIT regimen was divided into odd and even days. On odd days, the training started at a speed of 40 m/ min, with two repetitions (reps) of 3 min interspersed with 1 min of active rest (total duration � 7 min). Te speed, reps, and duration were increased every subsequent odd day. On even days, the training started at a speed of 54 m/min, with three reps of 30 s interspersed with 1 min of active rest (total duration � 3.5 min). Te speed, reps, and duration were increased every subsequent even day ( Table 2). In both the MICT and HIIT, the training regimen started with a warmup session (5 min at 16 m/min) and ended with a cool-down session (5 min at 16 m/min) (Tables 1 and 2). Te exercise prescriptions were adapted from Afzalpour et al. [23]. Exercise intensity was confrmed using a treadmill speed reference for eight-week-old untrained Wistar rats from Qin et al. [24]. Blood sugar was evaluated each week during the provision of physical exercise ( Figure 1).

Tissue Collection.
Te day after the exercise regimen was completed, decapitation was performed. Te rats underwent glucose loading (2.5 g glucose/kg) for 30 min before surgery to induce GLP-1 release. Total anesthesia was conducted by intraperitoneal injection of combined xylazine hydrochloride (0.01 ml/kg) and ketamine (0.05 ml/kg). Te ascending aorta until the abdominal aorta was surgically removed. Aortic tissue was rinsed in phosphate bufer saline (PBS) (pH 7.4) to remove excess blood and stored at −80°C.
On the day of measurement, the tissue was thawed at 2-8°C and was chopped in PBS with a glass homogenizer on ice. Te homogenate was centrifuged at 2000-3000 RPM for 20 minutes and the supernatant was taken for measurement.

Enzyme-Linked Immunosorbent Assay.
Te concentrations of GLP-1 and eNOS were measured using ELISA kits from Bioenzy (BZ-08189170-EB and BZ-08185640-EB) and the concentration of TNFα was measured using ELISA kits from Cusabio (CSB-E11987 r) according to the manufacturer's instructions. Total protein was measured using the Bradford assay for normalization to GLP-1 and TNFα (Bio-Rad Laboratories).

Colorimetric
Assay. SOD activity was measured using a colorimetric assay kit from Elabscience (E-BC-K020-M). Te measurement was carried out using the water-soluble tetrazolium salts-1(WST-1) principal method. MDA levels were assessed using a colorimetric assay kit from Elabscience (E-BC-K025-S) and a modifed spectrophotometric thiobarbituric acid (TBA) test method.
2.6. Quantitative Real-Time PCR. Total RNA was extracted from aortic umbilical tissue using the Quick-RNA MiniPrep Plus kit (Zymo Research) and was reverse transcribed using the ReverTra ACE ® qPCR RT Master Mix with gDNA remover (Toyobo) cDNA synthesis kit, according to the manufacturer's instructions. A real-time polymerase chain reaction (RT-PCR) was performed using the SensiFAST SYBR Hi-Rox Kit (Bioline). Te thermal cycling program was 95°C for 15 min followed by 40 cycles at 95°C for 20 s, 58°C for 30 s, and then 72°C for 30 s. Te RAGE gene was detected, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as the internal control or for normalization. Te gene primer sequences used were as follows: RAGE: forward 5′-CAGAAACCGGTGATGAAGGAC-3′ and reverse 5′-TCTGGGTTGTCGTTTCGC-3′; NF-κB: forward 5′-CGACGTATTGCTGTGCCTTC-3′ and reverse 5′-TTGAGATCTGCCCAGGTGGTA-3′; and GAPDH: forward 5′-GCATCTTCTTGTGCAGTGCC-3′ and reverse 5'-TACGGCCAAATCCGTTCACA-3′. Te 2ΔΔct method was used to evaluate the relative expression in the control and treatment groups.

Hematoxylin and Eosin (H&E) Staining.
Te tissue samples from the ascending aorta until the abdominal aorta of the rats in all study groups were fxed with optimal cutting temperature (OCT) medium. Tey were then cut using a cryostat (Leica Model CM1950) with 4.5 um-5 um thick metal grids and brushes. Tey were stained using hematoxylin-eosin solution (0.1% Mayers hematoxylin) (Sigma; MHS-16).
Te image of the stained preparation was taken using a Leica microscope and the Leica Application Suite software. Te images were analyzed using the image J image processing program. Te thickness of the aorta was measured starting from the tunica intima to the media. It was measured from four felds of view in a clockwise direction (12, 3, 6, and 9) and viewed with x400 magnifcation. Te diameter of the aorta was viewed by x40 magnifcation and measured in the largest and smallest part; the results were added and then divided by two.

Statistical
Analysis. Data are presented as mean-± standard deviation (SD) for normally distributed data and as mean (interquartile range) if the data are not normally  (   Journal of Nutrition and Metabolism

Blood Glucose.
All rats had normal blood glucose levels before the start of the study. Except for the control group that did not get streptozotocin injection, blood glucose levels were always maintained above 200 mg/dL during the study. Although still considered hyperglycemia, the mean blood glucose level in the exercised group (both MICT and HIIT) tends to decline each week. On the contrary, the hyperglycemia group that did not receive exercise showed similar  6 Journal of Nutrition and Metabolism blood glucose levels in each week ( Figure 1). Statistical analysis using two-way ANOVA showed a signifcant difference in mean blood glucose levels between the control group and all other groups all week, except Week 0 (p < 0.01 ). Tere was no signifcant diference in blood glucose levels between the HG-C, HG-IT, and HG-CT groups each week (p > 0.01). Figure 2, GLP-1 levels in the aortic tissues of the HG-C group (674.54 ± 134.97 ng/mg protein) were signifcantly lower than those in the aortic tissues of the HG-CT group (1,129.78 ± 224 ng/mg protein; p � 0.002) and the HG-IT group (91,099.03 ± 375.9 ng/mg protein; p � 0.009). Tere was no signifcant diference between the C group and both intervention groups (HG-CT and HG-IT). Tere was no signifcant diference in GLP-1 levels between the HG-CT and HG-IT groups. Figure 3 illustrates that there was no statistical diference in eNOS levels within each group. Te mean eNOS level in the HG-IT group was 47.915 ± 12.5 ng/mg protein and 37.653 ± 13.259 ng/mg protein in the HG-CT group. Figure 4 shows that the TNFα concentration in the HG-C group (444 ± 124.56 ng/mg protein) was signifcantly higher (p � 0.001) than in the HG-CT group (218 ± 44.22 ng/mg protein) and the HG-IT group (215 ± 89.121 ng/mg protein). Tere was no signifcant difference in TNFα levels between the HG-CT and HG-IT groups. Figure 5, there was no signifcant diference in the SOD activity within all groups.

Colorimetric Assay. As shown in
In terms of MDA levels, the C group had a signifcantly lower MDA level (3.65 ± 0.92 nmol/mgprotein) when compared to the HG-C group (p < 0.05) (Figure 6). Te MDA level in the HG-IT group (6.5 ± 1.67 nmol/mgprotein) was signifcantly (p = 0.006) lower than that in the HG-C group (10 ± 2.4 nmol/mgprotein). However, there was no signifcant diference in MDA levels between the HG-C and HG-CT groups. Figure 7 shows the level of RAGE gene expression in the abdominal aorta tissue samples. RAGE gene expression was signifcantly higher in the HG-C group (2.74 ± 0.3) compared to the C group (1.004 ± 0.28; p � 0.002), the HG-CT group (1.2 ± 1.34; p � 0.041), and the HG-IT group (1.125 ± 1.09; p � 0.002). Tere was no signifcant diference in RAGE gene expression between the HG-CT and HG-IT groups.

Quantitative Real-Time PCR.
As shown in Figure 8, the HG-C group (2.1 ± 0.4) had the highest NF-κB gene expression and was signifcantly higher than that in the C group (1.03 ± 0.31; p � 0.004), the HG-CT group (1.2 ± 0.71; p � 0.05), and the HG-IT group (1.4 ± 0.57; p � 0.016). Tere was no signifcant diference in NF-κB gene expression between the HG-CT and HG-IT groups.
It was found that there were no signifcant diferences in the aorta diameter and wall thickness between the groups (Figures 9 and 10). However, when the tissues were examined using light microscopy, tunica intima (endothelial) detachment was observed in the HG-C group samples. An irregularity in the arrangement of collagen fbers and smooth muscle cells (SMCs) in the tunica media was also seen (Figures 11 and 12).

Discussion
Our study examined the direct efect of 6 weeks of exercise on the aorta of rats under hyperglycemic conditions. To the best of our knowledge, it is the frst to explore the efect of exercise on the GLP-1 levels and how it then inhibits AGE- RAGE, hinders infammation, reduces stress oxidative, and protects vascular structure and function. Te main fnding of this study is a higher GLP-1 concentration, lower RAGE gene expression, lower TNFα concentration, and lower NFκβ gene expression in hyperglycemic rats that performed both HIIT and MICT compared to hyperglycemic control rats. It is worth noting that lower MDA concentrations were found only in HIIT rats, while no diferences in SOD activity or eNOS levels were found across all groups.
GLP-1 is one of the incretin hormones that have multifaceted and broad pharmacological potential. It is activated by binding to its receptor, GLP-1R. GLP-1R is expressed in various organs in the body, including the cardiovascular system, and specifcally in SMCs, the endothelium, and cardiomyocytes [13,25]. According to the fndings of Bhats et al. [26], GLP-1 level in rats with T2DM is lower compared to healthy rats. Tis is presumably due to several factors. Te frst factor is chronic hyperglycemia (>72 h). After oral consumption of glucose, the glucose level in the intestine increases, which induces L cells to increase GLP-1 secretion. However, in the case of chronic hyperglycemia, diabetic products are triggered, such as glycated serum (GS) and glucolipoxity [26,27]. GS plays a role in reducing the expression of prohormone convertase 1/3 (PC 1/3), an enzyme that functions in the posttranslation process of converting proglucagon into GLP-1 in intestinal L cells [28]. Te second factor is the presence of systemic infammation caused by chronic   [29] In T2DM, adipocytes and adipose tissue macrophages are the major sources of increased circulating IL-6. Tese infammatory cytokines may be circulated throughout the body, afecting organs and tissues further away, such as the kidneys, skeletal and heart muscle, and circulating leukocytes [30]. IL-6 that was synthesized and secreted during infammatory conditions lead to sustained high ROS levels [31]. ROS interferes with insulin signaling, which leads to inhibition of glucose uptake and disruption of insulin secretion. Insulin is known to stimulate proglucagon gene expression and GLP-1 synthesis. Hence, disruption of insulin secretion results in GLP-1 defciency. Te fnal factor is the presence of dipeptidyl peptidase 4 (DPP-4). Tis enzyme can degrade the active form of GLP-1 (7-36) to a noninsulin tropic form of GLP-1 (9-36) that causes the active form of GLP-1 to last for only one to two minutes [25,32]. In contrast to Bhat et al. who measured GLP-1 in animal serum immediately after the intervention (up to 1 hour), GLP-1 measurements in our study were performed on aortic tissue 24 hours after the intervention. We suspect it contributes to the lack of diference in GLP-1 levels between healthy and diabetic rats in our study. Here, we have shown that rats performing HIIT and MICT had higher levels of GLP-1 than hyperglycemic control rats. Tis could be due to sympathetic activation provoked by exercise that induces modulation of GLP-1 release via an increase in epinephrine [33][34][35]. Epinephrine is known to stimulate the release of GLP-1 in the mouse ileum [36]. Ellingsgaard et al. also showed that physical exercise could increase GLP-1 levels through skeletal muscle contraction-induced interleukin 6 (IL-6) secretion. Muscle contraction by both type I and type II muscle fbers increases cytosolic calcium (Ca 2+ ) and activates p38 mitogen-activated protein kinase (MAPK) and/or calcineurin, which can further activate transcription factors of genes that encode IL-6 in the nucleus of myocytes [37][38][39]. IL-6 is then released into the bloodstream and binds to receptors located on intestinal L cells [40,41]. Tis binding increases the phosphorylation of signal transducer and transcription activator 3 (STAT3) via Janus kinase 2 (JAK 2), which triggers an increase in Ca 2+ levels, thereby inducing GLP-exocytosis [41]. IL-6 can also induce GLP-1 synthesis by increasing proglucagon transcription and PC1/3 expression in intestinal L cells [41]. Te result of our study shows that under hyperglycemic condition, RAGE levels are lower in rats that performed both HIIT and MICT compared to hyperglycemic control rats. Physical exercise indirectly inhibits AGEs-RAGE activation via the GLP-1 signaling pathway. Tang et al. showed that GLP-1 could inhibit RAGE expression [13]. When GLP-1 binds to GLP-1R, it is activated and this increases the cAMP level, which activates the protein kinase A (PKA) signaling pathway. Te cAMP pathway suppresses RAGE gene expression and reduces AGEs-RAGE activation [13]. In addition, Lim et al. showed that under hyperglycemic conditions, GLP-1 could increase eNOS mRNA expression via the cAMP signaling pathway that blocks the AGEs-RAGE pathway [42]. Terefore, it is speculated that physical exercise, which stimulates an increase in GLP-1 secretion, increases eNOS mRNA expression [42]. According to a study by Farinha et al., highintensity exercise training decreased sRAGE in insulindependent patients, which may be related to greater plasma total antioxidant capacity levels and enhanced antioxidant enzyme activity [43]. Exercise is thought to encourage a more efective utilization of reactive intermediates of glycolytic and polyol pathways, thus reducing their availability for reaction with amino groups and the formation of AGEs [44].
Under the hyperglycemic conditions associated with T2DM, AGEs increase rapidly and reduce the half-life of eNOS mRNA, which causes eNOS defciency [45]. In addition, the AGEs-RAGE interaction in the formation of an extracellular matrix causes structural changes in the blood   vessel walls, thereby worsening vascular conditions [45]. Changes in the structure of blood vessel walls disrupt the supply of nutrients and materials to various body organs because the blood vessels constrict, and NO is not released. Te efects of these morphological changes can increase the risk of stroke, blindness, ischemia, retinopathy, atherosclerosis, heart attack, kidney failure, and gangrene of the lower extremities. It is known that physical exercise has a positive impact on preventing vascular complications in DM through the inhibition of chronic infammation. You   how exercise training reduces chronic infammation. Every bout of exercise causes contracting muscle tissue to produce muscle-derived anti-infammatory "myokine." Exercise also has efect on adipose tissue to reduce hypoxia and local adipose tissue infammation. In the immune system, exercise lowers the number of proinfammation production. Specifcally, on endothelial cells, exercise training reduces leukocyte adhesion and cytokine formation [20,46]. A study by Pramaningtyas showed that physical exercise triggers an increase in the NO released by eNOS through shear stress [47]. Te NO then difuses into the vascular smooth muscle and activates soluble guanylate cyclase-cyclic guanosine monophosphate (sGC-cGMP) signaling to inhibit L-type Ca 2+ channel located on vascular smooth muscle cell membranes and to induce a decrease in cytosolic Ca 2+ so that relaxation can occur [48]. During physical exercise, blood fow increases. Blood pressure exerts radial compression and circular-attraction forces on various types of cells in the vessel wall and on extracellular matrix proteins. Blood fow also applies a longitudinal frictional force on the tunica intima of blood vessels and produces shear stress (velocity per unit area) [49]. In turn, shear stress promotes optimal regulation of eNOS. Chengji et al. showed that six weeks of physical exercise of diferent intensities could increase eNOS mRNA, eNOS activity, and levels of NO product (nitrite and nitrate). Increasing the intensity of physical exercise further increases eNOS levels. Tis is thought to be infuenced by the speed of the shear stress induced by the intensity of the physical exercise [48]. A study by Grijalva et al. found signifcantly higher eNOS protein levels and eNOS dimerization in the myocardium of a diabetic rat model given exercise training compared to sedentary rats [50]. In this study, we failed to demonstrate diferences in eNOS activity in the aorta tissue across the groups. We can only hypothesize that decapitation of the rat's 24-hour postexercise played a role in afecting the levels of eNOS measured in the aorta.
Our study results show no diferences in eNOS levels between groups given HIIT and MICT. Te activity of eNOS is afected by 2 pathways. Te frst pathway is the initiation of eNOS by shear stress. Te speed of shear stress is higher with increased intensity of physical exercise and thus further increases eNOS levels [48]. Trough this mechanism, higher intensity in HIIT compared to MICT results in greater shear stress. Te second pathway is eNOS release through AGE-RAGE inhibition. Physical exercise triggers the release of IL-6 that causes an increase of GLP-1. Further increased level of GLP-1 inhibits AGE-RAGE. Te increase in plasma IL-6 is more infuenced by duration so that the intensity is adjusted to the duration [37].
In the hyperglycemic state, very high levels of lipid peroxidation and decreased antioxidant defense mechanisms can simultaneously cause damage to cellular organelles and create oxidative stress. In patients with T2DM, the increase in lipid peroxidation products (MDA) afects the decreasing antioxidant enzymes, such as glutathione (GSH) and SOD [51]. Our results are consistent with previous studies. Te mean MDA level in the control group was signifcantly lower than that in the hyperglycemic control group. However, compared to the hyperglycemic control group, in this study, we found that the MDA level was only lower in HIIT and not MICT group. Tis is in line with the fndings of Mitranun et al. Tey also compared MICT with HIIT and found that MDA levels in subjects that performed HIIT were signifcantly lower than the levels in subjects that performed MICT [52]. We assumed that the processes through which interval exercise reduces oxidative damage were more efective compared to continuous training. Interval aerobic training appears to have a stronger efect on blood fow and shear stress than continuous training, which in turn improved NO bioavailability and endotheliumdependent vasodilation. An earlier study by Liu et al. found that human mesenchymal stem cells respond better to intermittent fuid shear stress than continuous fuid shear stress to promote osteogenic diferentiation [53]. Mitranun et al. showed increased plasma nitrite and nitrate concentrations in the interval group but not in the continuous group. Terefore, reduced oxidative stress may be associated with improved endothelial function following interval training [52]. It is also plausible that interval training exercise sessions caused higher shear stress and thus lead to greater cellular and molecular reactions [54]. According to a study by Lee et al., moderate-intensity physical exercise can increase SOD activity under hyperglycemic conditions due to decreased NADPH oxidase activity, thus inhibiting endothelial dysfunction [55]. NADPH oxidase is a membrane-bound protein with the main function to transfer electrons across the plasma membrane to molecular oxygen, which results in the generation of the superoxide anion and subsequently reactive oxygen species (ROS), including hydrogen peroxide (H 2 O 2 ) and hydroxyl radicals (OH • ). [56].
Sarasvati et al. demonstrated a signifcant increase in SOD activity in subjects who performed HIIT compared to subjects who performed MICT. It is assumed that MICT does not produce an adaptive response to oxidative stress via activation of the nuclear erythroid 2-related factor (Nrf2) signal. Nrf2 is a redox transcription factor and a major regulator of antioxidants [57]. Te higher SOD activity in subjects who performed HIIT was due to the activation of the redox protein sensitivity pathway via higher ROS production [58]. HIIT could provoke greater blood fow in the large blood vessels during exercise in which shear stress increases and induces NO bioavailability [57]. We did not fnd any signifcant diferences in SOD activity among the groups, which is in contrast to the fndings of Sarasvati et al. However, our result is in line with the study result of Kesavulu et al. which showed an increased erythrocyte catalase (CAT) activity and decreased glutathione peroxidase (GPx) activity in diabetic patients compared to controls, but there was no signifcant change in superoxide dismutase (SOD) activity [59].
Nuclear transcription factor κB (NF-κB) is activated by AGEs and in turn promotes the transcription of infammatory markers, which produces vasoconstrictor molecules and oxidative damage [60]. In our study, NF-κB gene expression in the hyperglycemic control group was signifcantly higher compared to the expression in the control group. Tis result is in line with previous studies that found hyperglycemia can produce oxidative stress in the endothelium that activates NF-κB [61,62]. Recent studies show that NF-κB is the main suspect and a key player in the development of insulin resistance and T2DM [61]. TNFα, a regulated product of NF-κB and a strong activator of NF-κB, induces insulin resistance mainly through serine phosphorylation of insulin receptor substrate-1 (IRS1) [61]. Research by Patel et al. has shown that long-standing hyperglycemia will generate AGEs and directly activate NF-κB in vascular smooth muscle cells (VSMCs) [61]. Under hyperglycemic conditions, VSMCs showed a higher basal NF-κB activity than under normoglycemic conditions [61].
Previous studies have stated that undertaking physical activity can reduce NF-κB expression through the IL-10 pathway. Anti-infammatory IL-10 inhibits TNFα by suppressing NF-κB P65 expression in macrophages [63]. Tis is in accordance with the result of our study which showed signifcantly lower NF-κB expression in animals that performed both HIIT and MICT compared to animals in the hyperglycemic control group. Kahkha et al. reported that serum NF-κB levels were higher in subjects who performed HIIT than in those who performed aerobic exercise [64]. However, there was no signifcant diference between the two training methods found in our study result.
In this study, we examined the efect of HIIT and MICT on infammation by analyzing TNFα levels. TNFα is a potent proinfammatory agent that regulates many aspects of macrophage function. It is rapidly released after trauma or bacterial infection, and it is one of the most abundant early mediators in infammation [65]. Our results showed that the expression of TNFα was signifcantly higher in the hyperglycemic control group than in the exercising group. Tis is in accordance with previous studies that have shown that exercising at a certain intensity can provide positive benefts in terms of reducing infammatory factors [66]. Over the past two decades, it has become increasingly clear that chronic infammation is an important contributing factor in the development of chronic noncommunicable diseases, such as T2DM. Te infammatory process partly mediates vascular complications in patients with diabetes, and there is evidence that IL-1β and TNFα are the major proinfammatory mediators in cell damage and insulin resistance [66]. During exercise, skeletal muscle cells contract and thus activate c-JunN-terminal kinase (JNK/AP-1) and MAPK, which stimulate IL-1ra and IL-10, thereby inhibiting the production of TNFα [67,68]. Our results indicate that both HIIT and MICT have a positive efect in that they reduce TNFα under hyperglycemic conditions.
In chronic hyperglycemia, the aorta is at risk of endothelial dysfunction. Endothelial dysfunction is a condition that can be initiated through the activation of AGEs-RAGE [13]. It can afect the histology of the aorta, especially the thickness of the intima-media and the lumen diameter. However, in this study, there was no diference in the aorta diameter and the thickness of tunica intima-media across all the groups. It is assumed that six weeks is not a sufcient length of time for signifcant hyperglycemia-related histological changes to occur in the aorta. Nevertheless, we found that the tunica intima in the hyperglycemic control group was detached (endothelial detachment). Tis means that the aorta was not evenly enveloped, as shown in Figure 11. In addition, there was an irregularity in the arrangement of the collagen fbers and SMCs of the tunica media, a characteristic of endothelial dysfunction. Tis is thought to be related to apoptosis induced by hyperglycemia. Furthermore, the increase in AGEs can cause changes in endothelial cell structure through AGE-RAGE activation. Changes in the structure of endothelial cells increase vascular permeability, which is followed by loss of adherence junctions, contraction of SMCs, and focal adhesion redistribution. Increased permeability results in protein leakage into the tunica media. Moreover, SMC contraction has a vital role in the regulation of endothelial permeability. Te actin-myosin interaction is regulated by myosin light chain (MLC) phosphorylation, which causes changes in the distribution of F-actin and barrier failure, resulting in irregularities in the tunica media. Optimal control of MLC is mediated by a balance between myosin light chain phosphatase (MLCP) and myosin light chain kinase (MLCK) [13].

Conclusions
In this study, hyperglycemic rats that performed HIIT and MICT had signifcantly higher GLP-1 levels, lower RAGE gene expression, lower TNFα levels, and lower NF-κB gene expression than hyperglycemic rats that had no intervention. Microscopic examination of aortic tissue revealed that hyperglycemic rats that underwent the exercise treatment had better tissue arrangement than hyperglycemic rats that did not undergo the exercise treatment. Except for the MDA levels, it can be concluded that, in the presence of hyperglycemia, HIIT and MICT have similar protective efects against endothelial dysfunction. HIIT has the advantage of being less time-consuming than MICT.

Data Availability
Te datasets generated and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Conflicts of Interest
Te authors declare that they have no conficts of interest.