Exogenous adiponectin with full Peroxisome proliferator-activated receptor agonists (pioglitazone) abrogates streptozotocin induced oxidative stress and vascular abnormalities in Spontaneously hypertensive rats.

Background Oxidative stress, associates with metabolic and anthropometric perturbations, leads to reactive oxygen species production and decrease in plasma adiponectin concentration. We investigated pharmacodynamically the pathophysiological role and potential implication of exogenously administered adiponectin with full and partial peroxisome proliferator-activated receptor-gamma (PPAR-γ) agonists on modulation of oxidative stress, metabolic dysregulation and antioxidant potential in streptozotocin induced Spontaneously hypertensive rats (SHR). Methods Group I (WKY) serve as normotensive control, whereas 42 male SHRs were randomized equally into 7 groups (n = 6), group II: SHR control, groups III: SHR diabetic control, group IV, V and VI treated with irbesartan (30 mg/kg), pioglitazone (10 mg/kg) and adiponectin (2.5 µg/kg), whereas, groups VII and VIII received co-treatments as irbesartan + adiponectin), (pioglitazone + adiponectin) respectively. Diabetes was induced using an intra-peritoneal injection of Streptozotocin (40 mg/kg). Plasma adiponectin, lipid contents, arterial stiffness with oxidative stress bio-markers were measured using an in-vitro and in-vivo analysis. Results Diabetic SHRs exhibited hyperglycaemia, hypertriglyceridemia, hypercholesterolemia, increased arterial stiffness with reduced plasma adiponectin and antioxidant enzymatic levels (P < 0.05). Diabetic SHRs pre-treated with pioglitazone and adiponectin separately exerted improvements in antioxidant enzyme activities, abrogated arterial stiffness, offset the increased production of reactive oxygen species and dyslipidaemic effects of STZ, except for blood pressure values which were more pronounced with irbesartan treated groups (all P < 0.05).

We prepared a model of Type 1 diabetic SHRs using a single intra-peritoneal injection (I/P) of (STZ) (Nova Laboratories, Sdn, Bhd, Malaysia), 40 mg/kg body weight, dissolved in citrate buffer (10 mM, pH 4.5) (28), whereas, all the STZ induced SHRs were given glucose (10%) in for the rst 48 hours after injection to offset the early hypoglycaemic shock. Blood glucose levels were evaluated using a standard Glucometer (Free Style, Abbott, Malaysia) and rats with glucose levels > 300 mg/dL at the 7th day were selected for the experiment. Physiological and metabolic perturbations including body weight, 24 hrs water intake and urine collection and were performed on day 0, to establish the basal variables, following on day 08, 21 and 28 of the experiment. Systemic hemodynamic parameters including systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP) and heart rate (HR) were measured non-invasively (NIBP) using the CODA equipment (Kent Scienti c Corporation, Torrington, CT) on similar day pattern as for metabolic and physiological indices. Pulse wave velocity (PWV) was measured on the acute study day i.e. day 28. Blood samples (2ml) were obtained on each day of urine collection, and plasma samples were obtained following centrifugation and stored at -30•C for further biochemical analysis including oxidative and antioxidant bio-markers, plasma levels of cholesterol, triglycerides, and low & high density lipoproteins measured values. The estimation of plasma adiponectin concentration was carried out using a quantitative assay max rat adiponectin Elisa kit (Chemtron, Biotechnology Sdn, Malaysia).

Measurement of in-vivo oxidative stress and antioxidant markers
The collected blood plasma samples before the termination of acute experiment, were subjected to a variety of biochemical analysis in order to access the oxidative and anti-oxidative status of experimental diabetic SHRs. The levels of plasma oxidative stress bio-markers including malondialdehyde (MDA), antioxidant enzymes activities i.e. total superoxide dismutase (TSOD, nitric oxide (NO), total antioxidative activity (TAOC) and glutathione peroxidase (GSH) in collected plasma samples were measured using the spectrophotometric detection kits following the instruction manual provided by Institute of Biological Engineering of Nanjing Jianchen, Nanjing, China.
Plasma malondialdehyde In the biological system oxygen free radicals can be generated by enzymatic and non-enzymatic reactions. Oxygen free radicals upon generation react with polyunsaturated fatty acids resulting in lipid peroxidation and generate lipid peroxide such as aldehyde group (malondialdehyde MDA), ketone and hydroxyl group with some oxygen free radicals. MDA, a product of lipid peroxidation reactions, generated as a result of the reaction between free radicals and polyunsaturated fatty acids in the cell membrane (30). Therefore, evaluation of the MDA concentration in the biological samples could re ect the extent of lipid peroxidation and indirectly signify the extent of cell oxidative state.
Total superoxide dismutase SOD plays an important role in cellular environments in the prevention of diseases linked to oxidative stress. SOD scavenges the superoxide anion free radicals and protects the cells from being injured from oxidative stress in a biological system. We investigated SOD measurement in blood plasma samples using the method as described by Oyanagui (31).

Nitric oxide
The universal inter and intracellular molecule, nitric oxide (NO), is involved in regulating pathophysiology of CVS. Its biological activity is recognized as EDRF responsible for vasodilatation. It is a gaseous free biological molecule with a half-life of few seconds or less in vivo, whereas, it's altered levels are associated with several pathological conditions like hypertension, hypoxia and diabetes mellitus. The NO detection kit utilizes the nitrate reductase method and provides an accurate and convenient method for the measurement of total nitrate/nitrite concentration in the biological sample.
Total anti-oxidant capacity The antioxidant defence consists of enzymatic and non-enzymatic components. Defence system protects the biological system from oxidation through three pathways. Firstly, it eliminates activated oxygen and free radicals, secondly decomposes superoxide to block the oxidation chain and lastly, gets rid of catalytic metal ions (32). All different antioxidants yields greater protection against attack by nitrogen radicals and reactive oxygen. Hence, total anti-oxidant capacity (T-AOC) provides more concise biological information about anti-oxidant status of an organism compared to that obtained by the measurement of individual components.

Plasma glutathione
Glutathione is naturally occurring tri-peptide and is a signi cant component of antioxidant system and offers protection against oxidative damage and in the detoxi cation processes in the cell. Glutathione is mostly present in its reduced form (GSH) than in the oxidized form (GSSG coupled to computerized data acquisition system (PowerlabÒ, ADInstruments, Australia). A midline abdominal incision was carried out to expose the left kidney and the whole dissection process was done using electrical cautery knife and the abdominal contents were moved with great care to the right to get the clear exposure of the left kidney. The left kidney was exposed via a ventral mid-line incision and a laser Doppler probe (OxyFlo® Probe, Oxford Ltd., UK) attached to the Powerlab® system was placed on the dorsal surface of kidney for the direct observation of renal cortical blood perfusion (RCBP) values throughout the experiment. Additionally, the left iliac artery was catheterized with a PP50 cannula and was advanced through abdominal aorta lay at the entrance of renal artery, whereas PP50 cannula was kept patent via infusing saline @ 6 ml/hr. A time period of 60 minutes were allowed to stabilize the animals after the completion of surgical protocol. Blood pressure waves from the two pressure transducers were simultaneously imported and displayed on a data acquisition system at a sampling rate of 400/s for 30 min. The measurement of PWV was done as per our lab technique methodology, described by Swarup et al., (35), and was calculated by dividing the propagation distance (d) by propagation time (t) and expressed as meters per second.

Propagation distance and time
After euthenisation of animal, the full length of aorta was exposed and tip of the two cannulae from the carotid and iliac arteries were identi ed and marked. The distance between these two points was determined by using a cotton thread. After that the thread was laid straight for the measurement of distance and used to calculate PWV. At the end of recording, the animal was euthanized with sodium pentobarbitone (200 mg/kg) and the full length of aorta was exposed. A damp silk thread was placed along contour of aorta and marked at the tips of the two cannulae. The thread was then removed, laid straight and the distance between the two marks was measured. This pulse wave propagation distance was used to calculate the PWV. The propagation time was determined using a manual 'foot to foot' technique. The time consumed by the pulse wave (t) to move from the aortic arch to abdominal aorta was measured manually by the time delay between the upstrokes (foot) of each pressure wave front. The manual foot to foot method has been widely used and is considered a reliable method for determining PWV (35) (26). The time for the propagation of the pulse wave from the aortic arch to the abdominal aorta was measured by the time delay between the upstrokes (foot) of each pressure wave front. The average of 10 normal consecutive cardiac cycles was used to calculate the propagation time. Any abnormal waveform within the cycles measured was rejected and next viable waveform was measured. All the animals used in the experiment were sacri ced with an overdose of sodium pentobarbitone, Nembutal®, CEVA, France), at the termination of the study and were disposed off in accordance with guidelines of Animal Ethics Committee of Universiti Sains Malaysia, Malaysia.

Statistical analysis
The statistical analysis was performed using GraphPad Prisim® version 5.00 for Windows (GraphPad Software, San Diego California U.S.A). Metabolic parameters including body weight, blood glucose level and plasma adiponectin concentration, the haemodynamics parameters during the treatment period were analyzed using repeated measures one-way ANOVA followed by Bonferroni post hoc test. Data expressed as mean±S.E.M and differences between the means were considered signi cant at the 5% level.

Biochemical and metabolic indices
The mean values for metabolic indices including body weight, uid intake, urine output and blood glucose concentration of all eight experimental groups were measured on four occasions during the study period, i.e. on day 0, day 8, day 21, and day 28 and are given in  (Table 1).

Systemic haemodynamics
As per study protocol, baseline values and the changes in the systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP) and heart rate (HR) of eight groups of experimental rats were measured by the tail cuff method on day 0, day 8 after induction of diabetes, day 21 and nally on day 28 of the study, as shown in Hear rate of all groups were observed on the same pattern of days i.e., day 0, 8, 21 and 28. The heart rate of SHR+CNT group remained signi cantly higher as compared to WKY+CNT on all four points of observation, whereas, heart rate of SHR+STZ+CNT group was higher as compared to SHR+CNT on day 21 and 28 (P<0.05). However, treating diabetic SHRs signi cantly reduced the heart rate in SHR+STZ+Irb, SHR+STZ+Pio and SHR+STZ+Adp as compared to SHR diabetic control group on day 28 only, (P<0.05), whereas, the values obtained in SHR+STZ+Adp were of greater extent as compared to SHR+STZ+Irb and SHR+STZ+Pio groups. No signi cant effect was observed in case of combined treatment of adiponectin with either irbesartan or pioglitazone, (P>0.05), (Table 2).

Plasma adiponectin and Lipid pro le determination
Plasma adiponectin concentration and lipid pro le were measured on day 28 only in SHR and SHR diabetic pre-treated groups. A signi cant increases in plasma adiponectin concentration was observed in WKY and SHR control groups as compared to SHR+STZ group (P<0.05). The diabetic SHRs treated with irbesartan (30mg/kg/day), pioglitazone (10mg/kg/day) and adiponectin (2.5µg/kg/day), expressed signi cant increase in plasma adiponectin concentration as compared to SHR+STZ+CNT group (P<0.05).
Moreover, the combined treatment of adiponectin in SHR+STZ+Irb+Adp and SHR+STZ+Pio+Adp signi cantly increased plasma concentration of adiponectin as compared to their separate treatments, (P<0.05), however, the extent of increase in plasma concentration of adiponectin in SHR+STZ+Pio+Adp group was greater as compared to the combination of adiponectin with irbesartan in SHR+STZ+Irb+Adp group (P<0.05), but didn't reach to the level of WKY control group ( Figure 1).
As far as lipid pro le of SHR diabetic treated groups are concerned, SHR+STZ showed a signi cant increase in triglycerides, low density lipoproteins, total serum cholesterol and decrease in high density lipoproteins as compared to SHR+CNT group as shown in table 3 (P<0.05). Interestingly, adiponectin treatment caused a signi cant improvement in all these parameters (P<0.05), whereas, combination of adiponectin with pioglitazone caused more signi cant decrease in triglycerides, low density lipoproteins, total serum cholesterol, and increases in high density lipoproteins as compared to separate treatment of adiponectin and combination of adiponectin with irbesartan (P<0.05), thus improved the lipid contents of diabetic treated SHRs (Table 3).
Moreover, it was observed that the pulse wave velocity (PWV) of SHR control group was signi cantly higher as compared to WKY control group, whereas after induction of diabetes the velocity of SHR diabetic group was signi cantly higher compared to SHR group. This increase in PWV was blunted in SHR+STZ+Irb (6.17±0.17)m/s, SHR+STZ+Pio+Adp (6.14±0.21)m/s, and SHR+STZ+Adp (5.49±0.22)m/s, however, the tendency to decrease PWV in adiponectin group was more as compared to separate irbesartan and pioglitazone groups. Adiponectin with pioglitazone in SHR+STZ+Pio+Adp group further reduce PWV and reach to the level of WKY group (5.27±0.31) m/s, (P<0.05), (Figure 3).

Antioxidant Biomarkers
Plasma total superoxide dismutase and malondialdehyde The plasma total superoxide dismutase (T-SOD) of eight experimental groups including diabetic control SHRs and diabetic treated SHRs was measured as shown in gure 4. We observed that plasma T-SOD level of SHR groups of rats was signi cantly lower as compared to WKY control i.  (Figure 8).

Discussion
To the best of our knowledge, this study is amongst few investigating the pathophysiological role and impact of exogenously administered adiponectin with PPAR-γ agonists in Streptozotocin (STZ) induced Spontaneously hypertensive rats (SHRs) by measuring in-vivo and in-vitro antioxidant potential, plasma lipid contents, glycaemic and endogenous adiponectin levels with systemic and renal blood pressure measurements. The present study also assessed the relationship between pulse wave velocity (PWV) and oxidative stress makers. Our study indicates that STZ administration leads to a complex mechanism of diabetes and hypertension development, possibly due to the enhanced oxidative stress, indicated by increase MDA and decreased plasma SOD, NOx and T-AOC levels. The 28 days study period, pharmacodynamically, revealed that adiponectin, as a biomarker, in combination with full PPAR-γ agonist, pioglitazone abrogates oxidative stress including PWV, ameliorates lipid pro le, systemic and renal blood pressure without effecting glycemic levels, signifying the synergistic antioxidant potential and vasodilator action in pretreated diabetic SHRs.
Spontaneously hypertensive rats (SHRs) are more susceptible to diabetogenic effect of Streptozotocin, most frequently used for the induction of diabetes (36) and causes increased production of reactive oxygen species (ROS) with activation of poly adenosine diphosphate ribosylation and nitric oxide release in SHRs (37). Consequently, we attempted to develop a well-known rat model of combined state of essential hypertension with diabetes.
Physiological and metabolic indices were kept into consideration to assess the experimental diabetes in SHRs, including body weight, which was signi cantly reduced as one of the pronounced effect of STZ on β -cells (38). Polyuria and polydipsia were also observed as signi cant metabolic perturbations of diabetic SHRs and found to be in accordance with the observations of Khan (39).
Hyperglycemia with elevated B.P cause damage to the vascular endothelial cells with increased oxidative stress & vascular reactivity (40) and are considered as vital phenomenon of metabolic syndrome (2). In our study, glycemic level was not in uenced after either irbesartan, adiponectin or pioglitazone either singly or combination treatment protocol. This probably corresponds with the type of the diabetic model using STZ similar to human type 1 diabetic model. It is well known that endothelial dysfunction, occurs in diabetic complications (6), associates with atherosclerotic progression (4)  Moreover, we also observed that hyperglycemic SHRs exhibited decreased RCBP which supports previous observations of our laboratory ndings in diabetic model of rats (41), (39), which probably due to stimulation of local ANG-II & intra-renal RAAS (45). In our ndings, three weeks irbesartan (partial PPARγ agonist) in combination with adiponectin signi cantly reduced RCBP, SBP and MAP values to a larger extent as compared to adiponectin and pioglitazone (full PPAR-γ agonist) either singly or combination pre-treatments, which could be possibly due to up regulation of PPAR-γ receptors besides an increase in production of nitric oxide (NO) (46), (47). The signi cance of NO in the kidney vasculature cannot be ruled out which performs various pivotal role including renal hemodynamics regulation, modulation of renal sympathetic neural activity and inhibition of tubular sodium reabsorptive mechanism (48). We observed that irbesartan @ (30 mg/kg/day) caused maximal dose for blockade of RAAS whilst its partial PPAR-γ agonistic activity also contributed in its B.P reduction and renoprotective characteristics in nondiabetic SHRs as observed previously in our ndings (49).
Nonetheless, regulation of MAP and vascular tone depends upon NO, which acts as endothelium-derived molecule (50), whereas, plasma adiponectin stimulates production of NO with reduction in sensitivity to Ang-II (51). Adiponectin receptors (Adipo R1 and Adipo R2) in endothelial cells mediate adiponectininduced phosphorylation of AMPK and eNOS which together lead to an increase in NO production (52). In our ndings activation of PPAR-γ with partial and full agonists (irbesartan and pioglitazone) respectively up-regulates plasma adiponectin levels probably by stimulating the expression of proteins involved in adiponectin assembly, for instance endoplasmic reticulum oxidoreductin-1 protein (Erol-Lα) and adiponectin secretion such as disul de-bond A oxidoreductase-like protein (DsbA-L), (53), however, we did not measure these proteins in our experimental protocol.
Moreover, in our ndings, the heart rate of STZ induced SHRs remained higher which could be due to the SNS over-activation (54), whereas, hypertensive state correlate with SNS activity, which, therefore, intricately involved with the initiation and progression of hypertension causing an increases in the heart rate (55) and supports our values obtained in diabetic SHRs. Previous ndings con rms the adiponectin existence in the cerebrospinal uid (56), thus controls and reduces the sympathetic nerve activity and heart rate (57), indicating adiponectin merely responsible for the reduction in heart rate of diabetic SHRs treated groups.

Adiponectin concentration in plasma and lipid pro le
Diabetes induced by high-dose STZ is similar to human type 1 diabetic model (58), thus reduction of plasma adiponectin concentration with STZ administration would contribute to the diabetic condition of SHRs, and is in agreement with ndings of Thule (59). Interestingly, in our experimental protocol STZ induced SHRs treated with pioglitazone for 3 weeks in combination with exogenous adiponectin signi cantly increased adiponectin levels as compared to the other sets of treatment. It is also evident that pioglitazone while acting as an agonist for PPAR-γ, improve endothelial function (60), with B.P reduction and lipid metabolism (61) via stimulating the production of plasma adiponectin (22) and reduction in vascular sensitivity in diabetic SHRs (21).
In addition, we also measured lipid contents of experimental diabetic and genetic model of hypertensive pre-treated rats. Plasma triglyceride concentrations were higher in control SHRs as compared to control WKY during treatment period, whereas, STZ treatment aggravate the condition in genetic model of hypertensive rats, leading to a further signi cant increase in plasma triglyceride, LDL, total serum cholesterol, with a decrease in adiponectin and HDL plasma concentrations, indicating anthropometric and physiological disorders. Previous studies reveals that full-length adiponectin activates AMP-activated protein kinase (AMPK) phosphorylation, (62) stimulates fatty-acid oxidation and glucose utilization by activating AMP-activated protein kinase, thus suppressing gluconeogenesis in liver (15). However, phosphorylation of AMPK regulates enzymes responsible for the synthesis of triglycerides and fatty acids with their transcription factors, thus constrains basal and oxidized low-density lipoproteins through NADPH oxidase inhibition in endothelial cells (63), eventually leading to decrease in adipose tissue mass through activation of adiponectin receptors present mainly in lateral hypothalamic nuclei (64). Therefore, PPAR-γ agonists used in our study probably in uenced the gene expression responsible for lipid and carbohydrate metabolism without affecting glycaemic levels in diabetic SHRs. To support our observations previous observation proved that pioglitazone attenuated dyslipidemia in cyclosporine induced hypertensive rats (65), whereas, in another study, Hussain proved a much greater bene cial effect of a combination of rosiglitazone and telmisartan offered more improvement in serum TGs and adiponectin (66). Interestingly, treating diabetic rats with exogenous adiponectin and pioglitazone as full PPAR-γ agonist, produced signi cant attenuation of metabolic dysfunctions, as evidenced by the signi cant decrease in TC, TGs, LDL, but an increase in HDL and adiponectin plasma concentrations as similar conclusion were drawn for plasma adiponectin concentration.

Pulse wave velocity and antioxidant changes
Oxidative stress de nes an imbalance between production of free radicals, its reactive metabolites and so-called oxidants or reactive oxygen species (ROS), whereas, their elimination by protective mechanisms, referred to as antioxidants. This imbalance leads to damage of important biomolecules and cells, with potential impact on the whole organism (67), however, oxidative stress and reactive oxygen derivatives further aggravates in diabetes and hypertension (7). In SHR, oxidative stress appears to be the cause of hypertension development on a larger scale and major effect of PPAR-γ activation is the reduction of oxidative stress levels (68). Recent epidemiological studies together with human diabetic models have suggested an association between adiponectin concentration and oxidative stress, thus, decreased circulating adiponectin levels predominates in increased oxidative stress, which are closely linked with metabolic syndrome (69), (12), and a the key feature of increased production of ROS and proin ammatory pathways (11). Production of ROS reduces bioavailability of NO due to uncoupling of eNOS, with enhanced levels of superoxide anions leading to formation of peroxynitrite, thus aggravates the impairment of eNOS activity and reduces NO production (70). In our experiment, an imbalance between antioxidants and oxidative stress was observed in diabetic SHRs, which can be con rmed from the increased plasma levels of free radical mediated products of lipid peroxidation (MDA), decreased plasma concentration of enzymatic antioxidant SOD and non-enzymatic antioxidant GSH. A decrease in TAC further con rms this imbalance indicates free radicals production with weak antioxidant defence system in diabetic SHRs, signifying the importance of OS as common denominator in all these pathways.
Arterial stiffness is linked with endothelial dysfunction, whereas, the pulse wave velocity (PWV) is considered as surrogate marker (a well-established index for arterial stiffness) (26) and vascular diseases. The stiffer artery would lead to increase in the PWV due to the persistent hyperglycaemia leads to depletion of antioxidant defence mechanism, generating free radicals (71) resulting in endothelial dysfunction and reduced vascular elasticity. Therefore, pulse wave velocity of diabetic SHRs was signi cantly higher as compared to control rats indicating the marked decrease in the extensibility of blood vessels in diabetic condition leading to increased arterial stiffness.
However, we observed that exogenously administered adiponectin attenuated the arterial stiffness (PWV) of diabetic SHRs along decrease in SBP and MAP, which could be at least partially mediated through its potent antioxidant characteristics and was attenuated by blocking endothelial derived nitric oxide synthase activity, suggesting that relaxant effect was possibly mediated by nitric oxide. However, the combination with pioglitazone resulted in signi cantly greater decrease in PWV as compared with combined treatment of adiponectin with irbesartan and separate treatments. Previous clinical studies have demonstrated that partial PPAR-γ agonist (ARBs), protects vascular endothelium via an increase of endothelial NO synthesis (72) and plasma adiponectin concentration (47) thus prevents endothelial dysfunction more effectively as compared to non PPARγ-agonists ARBs (73). Likewise, full PPAR-γ agonist, pioglitazone, stimulates the production of NO and moderates oxidative stress through activation of signaling cascades, such as cAMP-PKA and AMPK-eNOS component (60), and by increasing glutathione levels, thus supports the fact that AMPK serves as a major downstream molecule for adiponectin production (55). In this study, combined treatment of exogenous adiponectin with full PPAR-γ agonist (pioglitazone) signi cantly attenuates the oxidative status to a larger extent as compared to cotreatment of adiponectin with irbesartan in experimentally induced diabetic SHRs. There was a marked increase in NO, SOD and T-AOC plasma levels that indicates improvement in arterial stiffness with decreased oxidative stress in diabetic SHRs. Similarly the reduced lipid peroxidation (MDA) values denotes decrease in free radical production, thus substantiates our ndings and supports our hypothesis tested. Increased antioxidant levels (SOD and GSH) imply better defense against ROS. These antioxidants protect the cells from oxidative damage, thereby decreasing the oxidative stress mediated vascular complications through antioxidant-mediated pathways.

Conclusion
In a nut shell, exogenous adiponectin administration attenuated the vascular abnormalities, uctuating from endothelial dysfunction to ROS production, through nitric oxide and antioxidant enzymatic properties with abrogation of arterial stiffness. Nonetheless, owing to the full PPAR-γ agonist activity of pioglitazone, co-treatment with adiponectin signi cantly augmented to a larger extent with improvement in oxidative status, serum triglycerides, restoration of atrial stiffness (in-vivo biomarker) with antioxidant enzymatic potential indicating a degree of synergism existence between adiponectin and pioglitazone. All procedures and animal handling were carried out in accordance with the guidelines research centre