Effect Of Heme Oxygenase-1 On Diabetic Angiopathies And The Study Of Its Preliminary Mechanisms

Diabetic vascular complications, which affect the quality of life of diabetes patients, are also responsible for substantial morbidity and mortality in the diabetic population. Recently, researchers begin to pay close attention to the theory of oxidative stress, indicating that oxidative stress plays a crucial role in the pathogenesis of late diabetic vascular complications. Although our understanding of how diabetes-induced oxidative stress ultimately leads to vascular damage has advanced considerably in recent years, effective strategies about how to enhance the anti-oxidative activity of vascular wall remain limited. Recently, increasing attention has been paid to the beneficial effects of heme oxygenase-1(HO-1) in the cardiovascular system. HO is a microsomal enzyme that catalyzes the degradation of heme into biliverdin, which is subsequently reduced to bilirubin, one atom of iron and one molecule of carbon monoxide (CO). HO-1 is proposed as an inducible protein known as heat shock protein 32 and its breakdown products play many vital physiological and pathological roles in anti-oxidant processes and regulation of vascular tension. These phenomena imply that the induction of HO-1 might function as an alternative defense mechanism against the vascular injury induced by diabetes mellitus(DM). The present study was designed to observe the effect of HO-1 on vascular function and morphology of diabetic rats and demonstrate the probable mechanisms involved at different levels including in vivo and in vitro. In this study, a retroviral vector containing the human HO-1 gene was constructed and transfected into vascular endothelial cells to investigate the cellular protection of HO-1 against the injury induced by high glucose and free fatty acids(FFA) and the effects of overexpression of HO-1 on cell proliferation. These would provide experimental evidences for clinical use of gene therapy in diabetic vascular complications. 1 Effect of HO-1 on vascular function and morphology in diabetic rats The purpose of the present study was to investigate the effect of HO-1 inducer (hemin) and inhibitor (zinc protoporphyrin-Ⅸ,ZnPP) on vascular function and morphology of diabetic rats. Male Sprague-Dawley(SD) rats were randomly divided into four experimental groups, ①vehicle-treated control group;②DM group SD rats were rendered diabetic by a single intraperitoneal injection of 50 mg/kg streptozotocin (STZ) and were killed at 5-, 10-and 15-week after the successful inducement of diabetes; ③hemin group Rats after induction of diabetes were treated daily with abdominal injections of hemin (30μmol/kg) for 5 weeks; ④ZnPP group Diabetic rats were treated daily with abdominal injections of ZnPP (10μmol/kg) for 5weeks. The body weight was determined every week and the blood glucose every other week in the whole experiments. The systolic blood pressure (SBP) and heart beats was recorded weekly during the experiments by the tail-cuff plethysmographic method. Contents of serum FFA were detected using test kits. Contraction-response to phenylephrine(PE), endothelium-dependent relaxation response to ACh or endothelium-independent relaxation response to sodium nitroprusside(SNP) were studied in thoracic aortic rings (TARs) with the isolated artery ring technique. HO-1 mRNA and protein expression of vascular tissue in 5-week each group were examined by RT-PCR and Western blot. The specific RT-PCR band relative expression quantity was represented with the ratio of band peak area integrated optical density(IOD) of HO-1 to band peak area IOD of β-actin. The ultrastructure of thoracic aorta was observed under scanning electron microscope (SEM). The experimental results were as follows: (1) Changes of Blood glucose, body weight, SBP and heart beats Blood glucose concentration increased significantly from the control values after induction of diabetes and remained high(>16mmol/L) throughout the experimental period, all P<0.01. The body weights of 5-, 10-and 15-weekdiabetic rats were significantly lower than the corresponding values obtained in control rats at the same time points, P<0.01. Diabetic rats progressively developed hypertension as evident from the elevated SBP with no change in the heart rate. Although, 5-week of treatment with hemin had no significant effect on the concentration of blood glucose and body weights, it significantly decreased SBP level and there was no difference when compared with that in control group. Treatment of diabetic rats with ZnPP elevated blood glucose and SBP, while decreased body weights compared to DM group. (2)Serum FFA contents The serum FFA contents of 5-week and 10-week DM group were not changed significantly compared to that of control group at the same time points, P>0.05. Elevated FFA contents were observed only in 15-week DM group compared to the corresponding control as well as 5-and 10-week DM group. Hemin treatment had no effect on FFA content compared to 5-week DM group and control group, while ZnPP increased FFA level significantly. (3) Vascular reactivity The endothelium-dependent relaxation responses to ACh(10–6mol/L) in each time-point diabetic rats were diminished compared with that of control group, P<0.05. Administration with hemin in diabetic rats could improve vasorelaxation disorder and the curve of cumulative dose responses to ACh (10-8~10–5mol/L) of hemin group shifted upward compared with that of DM group; while treatment with ZnPP in diabetic rats could aggravated the vasorelaxation disturbance. Contractive responses to PE (10–6mol/L) were significantly higher in TARs from diabetic rats than those from control group, while the contractile responses to PE were significantly reduced in TARs from hemin group. ZnPP treatment did not significantly alter the contractions to PE in aortas of diabetic rats. No changes were detected in endothelium-independent relaxation responses to SNP (10–6mol/L) among the four groups. . (4) HO-1 mRNA and protein assay HO-1 mRNA and protein expression were not significantly changed in aorta of diabetic rats relative to control rats, P>0.05. However, more significant HO-1 mRNA and protein expression wereobserved in hemin group, and the IOD values were 2.01 times and 1.75 times of those of control group. (5) Ultrastucture changes Scanning electron micrographs of rat thoracic aortic endothelium from control group showed intact endothelial cell layer; while DM group exhibited structural damages of endothelial cell and cell gap; compared with DM group, ZnPP induced serious structural damages and hemin attenuated such morphologic changes of endothelial cells. The results suggested that upregulation of HO-1, by an HO-1 inducer, might provide a novel means to improve vascular dysfunction and ameliorated endothelial injury in diabetes. 2 The protective role of HO-1 in diabetic vascular complications and its properties as an antioxidant Hyperglycemia and high level of FFA were known to increase the generation of reactive oxygenase species(ROS), ultimately leading to increased oxidative stress in a variety of tissues. The objective of the following study was to explore whether the effect of HO-1 in diabetic vascular complications was due to its anti-oxidant capacity. The same animal model was used as before. Serum total antioxidant capacities (TAOC) and malondialdehyde (MDA ) contents were detected by using kits. Expression and distribution of tumor necrosis factor α(TNF-α), an important pro-inflammation cytokine, were examined and observed by RT-PCR and immunohsitochemistry in aorta. We noticed that there was no change in HO-1 mRNA and protein expression in the aorta of 5-week diabetic rats relative to vehicle-treated control rats. To further explore the relation between HO-1 gene regulation and oxidative stress in diabetic rats, we examined HO-1 expression in lung and heart by RT-PCR and immunochemistry and compared the changes to that in aorta. The results were as follows: (1) Changes of serum TAOC and MDA contents Serum TAOC of diabetic rats was decreased in a time-dependent manner (6.24±0.87 U/ml after 5 weeks;5.32±0.91 U/ml after 10weeks;4.02±0.91 U/ml after15 weeks;vs. control 8.84±0.81 U/ml ), and the content of MDA was progressivelyincreased (5.63±0.52 nmol/L after 5 weeks;6.82±0.47 nmol/L after 10 weeks;8.13±0.89 nmol/L after 15 weeks;vs. control 4.56±0.28 nmol/L). After pretreatment with hemin for 5 weeks in diabetic rats,there was no markedly change in serum MDA but TAOC was higher than that in 5 week DM group. Alternatively,pretreatment with ZnPP in diabetic rats,a lower content of TAOC (4.33±0.44 U/ml)and a higher level of MDA (7.51±0.49 nmol/L ) in serum were measured. (2) TNF-αexpression assay The ratio of TNF-αto β-actin in aorta from control group was (10.08±3.21)%, while that in 5-week DM group increased to (20.66±4.47)%, P<0.05. The ratio reduced to (13.53±2.97)% after the treatment with hemin and there was no significant difference between hemin group and control. ZnPP treatment resulted no change of TNF-αexpression in diabetic rats, P>0.05. The immunohistochemical results demonstrated that TNF-αpositive cells in aorta decreased in hemin group compared with DM group. (3) HO-1 expression in lung and heart Elevated expression of HO-1 mRNA and protein were observed in lung and heart of diabetic rats. These findings suggested that upregulation of HO-1 may increase the anti-oxidant capacities, decrease lipid peroxidation powerfully and inhibit the increase of pro-inflammatory cytokines in diabetic rats. The protective effect of HO-1 on diabetic vascular injury was due to its property as an antioxidants. The expression of HO-1 was upregulated in response to oxidative stimuli in diabetic lung and heart. Since HO plays an important role in defense mechanisms, the sustained low level of HO-1 in aorta may weaken vascular defense against the increased oxidative stress in diabetes. It is essential to enhance HO-1 expression against diabetic angiopathies. 3 The protective role of HO/CO system in diabetic vascular complications and its relation with NO As a gaseous messenger similar to nitric oxide(NO), CO has been demonstrated to share many properties with NO, including signal transduction and vasodilatation. We studied the regulation of NO pathway to elucidate thepossible interactive relationship between NO and CO in diabetic rats. Blood sample was collected by cardiac puncture and the level of CO was detected by blood gas analysis. NO production was assessed by measuring the concentration of nitrite,a stable degradation product of NO,meanwhile the activity of inducible nitric oxide synthase (iNOS) and endothelial nitric oxide synthase(eNOS) was measured through spectrophotometric analysis according to the procedures described by the technical manual. iNOS and eNOS were detected using RT-PCR and the distribution of iNOS was observed by immunohistochemistry too. The experimental results were as follows: (1) Changes of blood CO contents and serum NO contents The CO content in the blood was decreased in 5-week(1.50±0.21% vs. control 2.50±0.61%),and restored to normal in 10-week (2.25±0.54%) and 15-week (2.26±0.67%) in diabetic rats. Furthermore the NO level in serum of 5-week diabetic rats markedly increased to 53.30±10.98mol/L (vs. control 18.60±3.74?mol/L)and then decreased to 34.65±7.22 mol/L and 30.02±9.28 mol/L separately in 10-and 15-week diabetic rats, but was still higher than that of control. Pretreatment with hemin in diabetic rats,a high level of CO (3.20±0.73%) and decreased production of NO(48.67±9.70?mol/L) was measured. Alternatively,pretreatment with ZnPP in diabetic rats could inhibit the production of CO(0.93±0.35%) and meantime a higher level of NO(92.25±18.76m?ol/L) in serum was measured. (2) Changes of iNOS and eNOS activity and expression iNOS activity in serum of diabetic rats significantly advanced (13.90±2.05 U/ml after 5 weeks;15.06±1.89 U/ml after 10 weeks;14.98±2.37 U/ml after 15 weeks;vs. control 3.91±0.84 U/ml ) and upregulated iNOS expression in aorta from 5-week diabetic rats was also observed compared with that of control detected by RT-PCR; While eNOS activity significantly inhibited (11.13±7.29U/ml after 5 weeks;13.11±5.05 U/ml after 10 weeks;11.35±5.33 U/ml after 15 weeks;vs. control 19.32±3.95U/ml ) and its mRNA was down-regulated. Pretreatment with hemin for 5 weeks in diabetic rats could inhibit the mRNA expression and activity of iNOS(5.85±1.53U/ml) and promote the expressionand activity of eNOS(19.49±4.44mol/L) compared with those of 5-week diabetic rats. ZnPP treatment in diabetic rats could enhance iNOS activity (18.08±1.99U/ml) but had no effect on eNOS activity and mRNA expression. (3) Immunohistochemistry assay The immunohistochemical results demonstrated that iNOS positive cells in aorta decreased in hemin group compared with DM group. Our results indicated that increased iNOS/NO production at early diabetic stage was one of the reasons that lead to vascular dysfunction. We also observed that administration of hemin, an inducer of HO-1, could promote CO contents in blood and eNOS expression,inhibit iNOS activity and decrease NO production,thus improve the vasodilatation dysfunction of artery. 4 Effects of HO-1 gene transfer on endothelial cells injury induced by high D-glucose and free fatty acid 4.1 Human umbilical vein endothelial cells were transfected with retroviral vector XM6/hHO-1 We established an in vitro transfection of human HO-1(hHO-1) gene into human umbilical vein endothelial cell line ECV304 mediated by a retroviral vector. The procession of transfection was as follows: Construction of the retroviral vector XM6/hHO-1: A 987bp HindIII-cleaved fragment of hHO-1 cDNA was released from plasmid pGEM7zf/hHO-1, and inserted into HindIII site of XM6 vector. The orientation of recombinant construct was determined by restriction digestion with ApaI and the correct plamsid construct, designated as XM6/hHO-1, was solely chosen for the subsequent viral infection. Packaging of retroviral recombinant and transfection of endothelial cell XM6/hHO-1 DNA was transfected into PA317 cells with Lipofectin according to the manufacturer's manuals. The cells were selected for neomycin resistance (NeoR) in medium containing G418 (400 μg/ml). After selected with G418 for 2 weeks, the individual cell-cluster of G418-resistant was formed. The well-isolated clusters were trypsinized and harvested for continuous enrichment. For each isolated clone, the viral titer was determined by infectionof NIH/3T3 fibroblasts and only cell clone with titers beyond 1×106 colony-forming units (CFU) per ml was exclusively adopted in the following experiments. NIH/3T3 cell was also used to detection of helper virus in retrovirus stocks. Using supernatants of the PA317/ hHO-1 cells, ECV304 cells were infected with above retrovieuses. After selection with 500μg/ml of G418, individual G418-resistant clones were obtained. The ECV304 cells transfected with XM6/hHO-1 were nominated ECV304/hHO-1. Analysis of hHO-1 gene expression in transduced endothelial cells. RT-PCR detection, Western blot and immunocytochemistry techniques were used to analyses HO-1 mRNA and protein expression of ECV304 and ECV304/hHO-1 cells. The results obtained were as follows: (1) Viral production and titer The abundant virus titers as 4~10×105CFU/ml were obtained, indicating the sufficient infection rate of hHO-1 recombinant and there was no helper virus in the retrovirus stocks. (2)HO-1 mRNA detection and HO-1 protein expression We found that HO-1 gene-transduced endothelial cells exhibited a strong signal for HO-1 RT-PCR products compared to the non-transfected ones. The quantitative evaluation showed that HO-1 protein was increased 1.7-fold in the transfected cells. In addition, HO-1 protein could be obviously detected using immunocytochemical staining in ECV304/HO-1 cells, while the positive signal was weaker in ECV304 cells. Transfection of mammal cells with retroviral vector possesses considerable advantages including relatively high infective efficiency and stable expression. The above results indicated that the construction strategy of the recombinant vector was correctly performed and the retrovirus containing HO-1 could be stably and efficiently expressed in transfected ECV304 cells. Moreover, the introduction of exogenous HO-1 gene mediated by retrovirus allowed the functional expression of this foreign gene in the target endothelial cells. 4.2 Effects of HO-1 gene transfer on endothelial cells injury induced byhigh D-glucose and free fatty acid The present study was to investigate the effect of overexpression of HO-1 gene on endothelial cells injury induced by high D-glucose and FFA. Cells were divided into three groups:①ECV304 cells group;②ECV304/hHO-1 cells group;③ECV304/hHO-1+ZnPP(20μmol/L) group. Cells in each group were cultured in media containing D-glucose (5.5mmol/L, 20mmol/L) and/or palmitic acid (125μmol/L, 250μmol/L, 375μmol/L). After incubated for 48h, cell viability in each group was determined by MTT assay. Cellular MDA content was tested as a marker of lipid peroxidation. The leakage rate of lactate dehydrogenase (LDH) was measured by colorimetric assay. Flow cytometric determination of DNA content was used as an index of cell apoptosis. The results were as follows: (1)After incubation of ECV304 cells with D-glucose(5mmol/L, 20mmol/L) only or palmitic acid(125μmol/L) for 48h, no change of cell viability was seen as compared with that of the same treatment in ECV304/hHO-1 group . Howerver, exposure of ECV304 cells to high D-glucose(20mmol/L) plus a series of concentration of palmitic acid resulted in a significant decrease in cell viability, P<0.05, and compared with ECV304 cells with the same treatment, cell viability in ECV304/hHO-1 group was higher, P<0.05. ZnPP at a pharmacological dose of 20μmol/L caused decreased cell viability when compared with ECV304 cell incubated with the same media. (2)The exposure to D-glucose(5.5mmol/L) for 48h did not result in any significant difference in LDH leakage rate or cellular MDA contents among the three groups, P>0.05. However, after incubation with D-glucose (20mmol/L) and palmitic acid(250μmol/L) for 48h, both LDH leakage rate and MDA contents in the three groups were increased compared with those of ECV304 cells incubated with D-glucose (5.5mmol/L), P<0.05. But these cell damages indexes in ECV304/hHO-1 were significantly relieved as compared with that in ECV304 group with the same treatment. Compared with ECV304/HO-1 group, when ECV304/hHO-1 cells were treated with ZnPP, increased LDH leakage rate and MDA contents were observed as c, P<0.05.(3)There was no significant difference of apoptotic rates among the three groups after exposure to D-glucose(5.5mmol/L) for 48h, P>0.05. After incubation with D-glucose(20mmol/L) and palmitic acid(250μmol/L) for 48h, apoptosis in ECV304/hHO-1 treated with ZnPP group was remarkably increased compared with that in ECV304 and ECV304/hHO-1 group with the same treatment, P<0.05. These findings suggest that the upregulation of HO-1 gene expression was able to enhance endothelial cell resistance against injury induced by high D-glucose and FFA, which might be related to its effectivenese as an antioxidants. CONCLUSIONS 1. Upregulation of HO-1 gene expression in diabetic rats could partly reverse vascular dysfunction and hypertension; while downregulation of HO-1 could worsen vascular dysfunction, aggravate hyperglycemia and accelerate the rise of FFA. 2. It was further demonstrated that upregulation of HO-1 might increase the anti-oxidant capacities, decrease lipid peroxidation powerfully and inhibit the generation of pro-inflammatory cytokines in diabetic rats. The protective effect of HO-1 on diabetic vascular injury was due to its property as an antioxidants. The sustained low level of HO-1 in aorta might weaken vascular defense against the increased oxidative stress in diabetes. It was essential to enhance HO-1 expression in vascular wall against diabetic angiopathies. 3. Blood CO content decreased, then increased in the diabetic rats, while the change of NO was contrary to that of CO. Upregulation of HO-1 could promote CO content and eNOS expression,while inhibit iNOS activity and decrease NO production,thus improve the vasodilatation dysfunction of artery. 4. Upregulation of HO-1 gene expression in endothelial cells by transfer...