Involvement Of Heme Oxygenase-1 (HO-1) In Experimental Diabetes And Association Of Susceptibility To The Development Of Type 2 Diabetes With HO-1 Gene Promoter Polymorphism

Diabetes mellitus (DM) is one of the most common chronic diseases with endocrine and metabolic disorders characterized by the leading death caused by late macro- and micro-vascular complications. Type 2 diabetes mellitus (T2DM), the most frequent form of DM, is a complex disease (multiple gene-multiple factor disease) and attributed to both the pancreaticβcells dysfunction and insulin resistance (IR). The natural progression of T2DM experiences a prolonged pre-diabetic period of impaired glucose regulation (IGR), consisting of two distinct disorders: impaired fasting glucose (IFG) or impaired glucose tolerance (IGT). Lines of evidence from epidemiological researches have demonstrated IFG and IGT are both the risk factor of DM and cardiovascular disease. Owing to the sedentary life style, unhealthy diet habits as well as increased stress states, the prevalence of DM has increased strikingly and the suffering people become more youthful. Furthermore, an insidious epidemic of IFG/IGT also occurs.Although the majority of currently available therapies for diabetes could afford effective glycemic control, they are always accompanied by serious adverse effects for long-term use. So, it seems to be an urgent necessity to explore the new therapeutic paradigm based on the etiology of DM, which, however, still remains to be determined. Notably, manifold animal and human studies have disclosed oxidative stress contributes, at least in part, to the progression of T2DM and the following long-term complications. Antioxidants used as adjuvant in the treatment of DM can alleviate the oxidative stress and lipid peroxidation, but not all the researches reached consistency, suggesting the existence of large inter-individual responses dependent on genetic background involved in determining a person's susceptibility to oxidative stress. Thereafter, deeply understanding the inherited factors that influence a patient's susceptibility to oxidative stress may lead to the development of better and more comprehensive therapies for T2DM. Among the panel of potential candidate genes, heme oxygenase-1 (HO-1) seems to draw much attention with its potent anti-inflammatory, anti-oxidant, and anti-proliferative effects, close correlation with oxidative stress-related disease as well as its inducibility by various dietary antioxidants.The role of HO-1 implicated in diabetes has been reported by most studies so far in vitro assay exposed to glucose, whereas less and arguable information are available for diabetic models. Certainly, it is still not clear whether HO-1 exerts the similar biological effect in human diabetes, the relevant clinical studies are just being undertaken and deserved deeper investigations. Additionally, genetic variability is proposed to modulate the HO-1 response to exogenous stimuli, i.e., human present apparent differences in their ability to cope with a HO-1 response. To date, two potentially functional polymorphisms in the promoter region of human HO-1 gene have been identified: a (GT)n repeat microsatellite polymorphism and a single nucleotide polymorphism (SNP), T(-413)A, and hypothesized to modulate individual HO-1 response to a given stimulus. Thus, association studies between individual characteristics of HO-1 promoter genotype and susceptibility to T2DM may provide substantial theoretical guidance for its feasibility of clinical application to treat T2DM.To summarize, the study herein was carried out based on both the animal models and epidemiology angle respectively, to explore the effect of a natural antioxidant on oxidative stress pathways in renal mitochondria of diabetic mice mediated by HO-1, and to describe the association of susceptibility to the development of T2DM with HO-1 gene promoter polymorphism.PartⅠAnimal experiment (Effect of a natural antioxidant on oxidative stress pathways in renal mitochondria of diabetic mice mediated by HO-1)Objectives: This study described the dynamic changes of renal lipid peroxidation, anti-oxidative defense system and HO-1 of diabetic mice in different stages of DM (4 w and 8 w) to elucidate the potential role of HO-1 in the progression of diabetic nephropathy (DN). Meanwhile, the protective effect of a natural antioxidant (MG) and the underlying mechanism was explored to evaluate the possibility of HO-1 to screen for antioxidants.Methods: Male Balb/c mice were rendered diabetic by a single intra-peritoneal injection of alloxan. Control and diabetic mice were further subdivided according to their treatments: Control (saline), low dose MG (150 mg/kg) and high dose MG (300 mg/kg), which were administered immediately after confirmation of hyperglycemia by gavage daily. Mice were killed by cervical dislocation at 4th and 8th week respectively. Serum glucose, fructosamine, lipid profile and renal function were evaluated; renal homogenate were subjected to determination of malondialdehyde (MDA) and glutathione (GSH) concentration, manganese superoxide dismutase (Mn-SOD), catalase (CAT), glutathione peroxidase (GSH-Px) and HO-1 activities, together with HO-1 and Mn-SOD mRNA expression; paraffin-embedded renal tissues was used for routine histopathological examination.Results:(1) Clinical monitoring of animals: Mice demonstrated the typical symptoms of DM, including polydipsia, polyphagia, polyuria and loss of body weight after alloxan induction. While treatment with low dose of MG effectively attenuated the polydipsia and polyuria symptom of diabetic mice from the 3th week, and partially but significantly increased body weight of diabetic mice compared to diabetic control at the end of the experiment.(2) Effects of MG on serum glucose, fructosamine and lipid profile: Diabetic mice suffered permanent hyperglycemia and demonstrated higher fructosamine levels than control group both at 4th week and 8th week after alloxan injection. Besides, there was a significant positive correlation between serum glucose and fructosamine. Higher glucose levels were partially restored to basal values both after MG short-term (4 w) and long-term (8 w) administration, although a complete restoration to normalcy was not achieved. Alloxan didn't induce abnormal lipid profile until 8th week, confirmed by significantly increased total cholesterol and triglyceride concentration as compared with non-diabetic mice. The increased total cholesterol level was restored to normalcy after MG administration.(3) Effect of MG on renal histomorphology and function: Diabetic mice didn't demonstrate early pathological features of DN and renal dysfunction until 8th week, and the latter could be seen from the significantly higher urea nitrogen and creatinine levels than that of non-diabetic mice. The increase in creatinine level was partially prevented by treatment with low dose of MG.(4) Effect of MG on lipid peroxidation and antioxidant status in renal mitochondria: Renal mitochondrial MDA levels were both increased in diabetic mice compared with control mice at 4th and 8th week. Treatment with MG after 8 w completely abolished the accumulation of MDA. On the contrary, the level of GSH in diabetic group after short-term and long-term trial was lower than that in control group. With respect to the anti-oxidative enzymes, a significant increase in activities of renal mitochondrial Mn-SOD and GSH-Px was detected in diabetic mice 4 w after alloxan injection compared to non-diabetic mice, which was prevented by treatment with low dose of MG. However, the activity of renal mitochondrial Mn-SOD was reduced significantly in diabetic mice at 8th week, which could not be significantly reversed by MG but a nonsignificant increase in Mn-SOD activity was found after low dose of MG administration, showing no significant difference with non-diabetic mice. As far as CAT was concerned, decreased activity was observed when compared to non-diabetic mice both at 4th week and 8th week after alloxan injection.(5) Effect of MG on HO-1 activity in renal microsome: Diabetic mice demonstrated a significant higher and lower HO-1 activity 4 w and 8 w after alloxan injection respectively with respect to the corresponding non-diabetic mice. Following 4 weeks of treatment, both dose of MG resulted in a significant decrease in HO-1 activity with respect to non-treated diabetic mice, whereas after 8 weeks of treatment, only low dose of MG increased HO-1 activity as compared to non-treated diabetic mice.(6) Effect of MG on HO-1, Mn-SOD mRNA expression in the kidney: The mRNA expressions of HO-1 and Mn-SOD in the kidney of diabetic mice were significantly enhanced at 4th week as compared with control mice, which were significantly ameliorated after treatment with low dose of MG, although a complete restoration to basal levels was not achieved for the latter. Whereas alloxan markedly decreased mRNA expression of HO-1 and Mn-SOD as compared to non-diabetic control at 8th week, which could be both returned to normalcy by low dose of MG. Moreover, treatment with high dose of MG also induced a higher mRNA expression level of HO-1 than that in non-treated diabetic mice.Conclusions:(1) Oxidative stress was implicated in the progress of DM and DN, mediated by an important oxidative stress-responsive protein, HO-1.(2) HO-1 exhibited either pro- or anti-oxidant activities, and such a dual role may depend on particular condition in the development of DM, contributing to the regulation of some anti-oxidative enzymes.(3) Treatment with MG could eliminate renal oxidative stress in DM, and prevent the development of DN through its anti-oxidative action relevant to inhibition/activation of HO-1 due to its dual roles. PartⅡMolecular epidemiological research (Association of susceptibility to the development of T2DM with HO-1 gene promoter polymorphism)Section 1 Role of oxidative stress in the progression of T2DMObjectives: A large-scale case-control study was conducted among populations with different glucose metabolism state (normal glucose tolerance [NGT] people, IGR and newly-diagnosed T2DM patients), to describe the epidemiological features of IGR and newly-diagnosed T2DM patients, explore the inter-relationships among oxidative stress status, IR and pancreaticβcell secretary function, and illuminate the role of oxidative stress in the etiology of T2DM. This seems promising for the primary prevention of T2DM in the pre-diabetic IGR patients.Methods: The study population consisted of 371 patients of IGR (defined as IFG or IGT) and 1103 newly-diagnosed T2DM patients consecutively recruited from those attending the outpatient clinics of Department of Endocrinology, Tongji Medical College Hospital and 1615 age- and sex-matched healthy controls with NGT were recruited from an unselected population undergoing a routine health check-up in the same hospital from December, 2004 to November, 2007. IGR and T2DM patients met the well-established diagnostic criteria recommended by American diabetes association (ADA) and World Health Organization (WHO) incorporating both fasting plasma glucose (FPG) and 2-h oral glucose tolerance test (OGTT). A standard questionnaire was used to collect information about demography, history of diseases, family history of DM, diet habits, tobacco and alcohol consumption, and physical activity. IR index andβcell function were calculated by homeostasis model assessment (HOMA). Plasma MDA and total anti-oxidative capacity (TAC) status, and erythrocyte GSH content and SOD activity were determined. Comet assay were employed to evaluate DNA damage. Also, inter-relationships between the aforementioned parameters were analyzed.Results:(1) Clinical characteristics of the study population: Patients with IGR and T2DM had more conventional risk factors for cardiovascular events than did the NGT people, such as higher BMI value and percentage of hypertension, current drinker and family history of DM than NGT subjects, except for the information about tobacco consumption, dietary habits and physical activity, which contributed equally to different glucose metabolism groups.(2) Biochemical characteristics of the study population with different glucose metabolism state: With regard to biochemical indices, patients with IGR showed a slight but significant hyperglycemia and apparent hyperinsulinemia, thus leading to a higher HOMA-IR and lower HOMA-βcell than NGT people. Diabetic patients suffered from a more pronounced hyperglycemia than IGR group, and displayed strikingly increased HbA1c level and HOMA-IR, accompanied with decreased HOMA-βcell value when compared to NGT and IGR groups. Furthermore, diabetic subjects demonstrated a slightly abnormal lipid profile, with a higher triglyceride and total cholesterol level concentration than both NGT and IGR groups.(3) Oxidative stress and anti-oxidative status in different glucose metabolism population: Diabetic patients exhibited a higher plasma MDA concentration but a lower plasma TAC level than NGT and IGR group. Compared to NGT people, IGR and diabetic patients showed reduced erythrocyte SOD activity, while GSH concentration remained in normalcy. Correlation analysis revealed a strong positive association between IR with MDA concentration, whereas negative correlations with TAC status and SOD activity. With respect toβcells function, a positive association with TAC status and an inverse correlation with GSH were observed respectively.(4) DNA oxidative damage in the study population: Comet assay was performed in a small-scale people (27 NGT, 16 IGR and 34 T2DM) to assess DNA damage of PBMCs, and displayed NGT people exhibited lower levels of both % Tail DNA and olive tail moment (OTM) value than IGR and diabetic patients, and the latter appeared more severe than the former. Linear correlation analysis of % Tail DNA showed significant positive correlations with BMI, FPG, HOMA-IR, and a negative association with HOMA-βcell. Similarly, OTM value directly correlated with FPG and HOMA-IR, and inversely correlated with HOMA-βcell.Conclusions: The present study could depict the outline of the possible underlying mechanism of the pathogenesis of T2DM via investigating oxidative stress pathways and DNA damage in IGR and newly-diagnosed T2DM patients. Even at the time of IGR state, the mildly increased blood glucose at fasting or 2-h after glucose load had resulted in intensified oxidative stress or compromised anti-oxidative defense system or both, leading to oxidative DNA damage, which possibly contributed to pancreaticβcells dysfunction, IR and the following more pronounced hyperglycemia. This vicious cycle finally induced the deterioration into diabetes.Section 2 Association of susceptibility to the development of T2DM with HO-1 gene promoter polymorphismObjectives: To examine the potential association between individual HO-1 expression level and different glucose metabolism states (subjects recruited in Section 1), and to clarify the role of two polymorphisms in the promoter region of HO-1 ([GT]n microsatellite polymorphism and T[-413]A SNP) in susceptibility to T2DM as well as their functional relevance..Methods: The intracellular HO-1 expression in PBMCs was detected by flow cytometry. Genomic DNA was extracted from the venous blood samples and subjected to determine the genotypes of (GT)n microsatellite polymorphism by fragment analysis and T(-413)A SNP by TaqMan probes respectively. The linkage disequilibrium (LD) between these two polymorphisms was evaluated by the linkage disequilibrium analysis program (LDA), and haplotypes were constructed using PHASE 2.0 program. The association of IGR and T2DM events with the specific polymorphism or haplotypes in HO-1 gene promoter was assessed with adjustments for other known risk factors of T2DM. Moreover, to evaluate the functional relevance of these two polymorphisms, the differences of HO-1 expression levels across genotypes adjusted by those aforementioned contributing factors was compared. Results:(1) Investigation of association between individual HO-1 expression levels with different glucose metabolism states: There was a significantly decreased HO-1 expression level in both the IGR and diabetic patients when compared with NGT subjects.(2) Association study of (GT)n microsatellite polymorphism with T2DM: The numbers of (GT)n repeats in HO-1 gene promoter region in our population ranged from 12 to 40. The distribution of allelic subgroups was comparable in groups with different glucose metabolism. All were bimodal with the two most common alleles being 23 and 30 repeats. According to this distribution and the previously reported results of HO-1 association and functional studies, we divided the allelic repeats into two subclasses: <25 (GT)n repeats was designated as class S (short), and repeats with≥25 (GT)n as class L (long). The proportion of allelic frequency in class L and genotypic frequency in class L/L were significantly higher in patients with IGR and T2DM than that in those with NGT. After adjusting for traditional risk factors of diabetes, L/L genotype was associated with an increased risk of IGR as compared to S/S genotype (OR, 1.739; 95% CI, 1.026-2.949; P=0.040); the presence of L allele (S/L and L/L genotypes) was an independent risk factor for T2DM, and the adjusted ORs were 1.446 (95% CI, 1.051-1.990; P=0.023) and 1.621 (95% CI, 1.113-2.362; P=0.012) for genotype S/L and L/L versus genotype S/S, respectively. When IGR and T2DM groups were combined, similar results were obtained and the adjusted ORs were 1.352 (95% CI, 0.998-1.831; P=0.048) and 1.651 (95% CI, 1.156-2.358; P=0.006) for genotype S/L and L/L.(3) Association study of T(-413)A SNP with T2DM: Our analysis of the association study between T(-413)A SNP in HO-1 gene promoter and IGR or T2DM did not yield any significant difference among the subjects with different glucose metabolism states, either in allelic (T- or A- allele) or genotypic (genotype T/T, T/A and A/A) frequencies.(4) Analysis of LD and haplotypes of (GT)n microsatellite polymorphism and T(-413)A SNP: Considering their close location of these two polymorphisms, LD analysis was performed and a moderately strong degree of LD between these two polymorphism was observed (D'=0.72). Among the four estimated haplotypes constructed in our study population, the two most prevalent haplotypes were TS and AL. Compared to the most frequent hyplotype TS, haplotype TL was associated with an increased risk of IGR (OR, 1.816; 95% CI, 1.179-2.796; P=0.007) and T2DM (OR, 1.487; 95% CI, 1.091-2.027; P=0.012), as was also observed for the risk of development into T2DM (IGR+T2DM) (OR, 1.555; 95% CI, 1.158-2.089; P= 0.003).(5) Functional relevance of these two polymorphisms: HO-1 expression level in PBMCs decreased with the increasing number of (GT)n-L allele among all the three groups, with L/L homozygous individuals exhibiting the lowest HO-1 expression level. This trend reached significance only in IGR and T2DM group, but not for NGT group. However, no significant differences were observed in HO-1 expression level between individuals with different genotypes of T(-413)A SNP across each group. On the basis of the hyplotype analysis, the association between HO-1 expression level and various haplotype pairs were subsequently explored. Within the class of A/A genotype, S-allele carriers with a haplotype pairs of AS/AS had a significantly higher HO-1 expression level than the L-allele carriers (haplotype pairs: AL/AL) in both the NGT group and T2DM group. When the T/T genotype was considered, similar effects were observed in diabetic patients, i.e., HO-1 expression level in patients carrying TS/TS haplotype pairs was significantly higher than those carrying TL/TL haplotype pairs.Conclusions:(1) (GT)n microsatellite polymorphism was associated with increased susceptibility to IGR and T2DM, ascribed to its regulatory effect on HO-1 protein expression. Long (GT)n repeats (L,≥25 GT repeats) carriers showed a decreased HO-1 protein expression and an increased risk of developing T2DM.(2) (GT)n microsatellite polymorphism and T(-413)A SNP were in LD with each other in our population, and haplotype TL was associated with an increased risk of IGR and T2DM.(3) (GT)n microsatellite polymorphism appeared to play an important role in the functional relevance.(4) (GT)n microsatellite polymorphism can serve as a new genetic marker for T2DM in our population, which may provide important clinical implications for the prevention and treatment of IGR and T2DM.