International Journal of Diabetes Mellitus, December 2010

International Journal of Diabetes Mellitus 2 (2010) 139–140

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Editorial

Raising the priority accorded to diabetes in global health and development: A promising response. . . On 13 May 2010, the United Nations General Assembly (UNGA) passed a resolution A/RES/64/265 on non-communicable diseases (NCDs). This step is of historic significance in global health and development, as the resolution recognizes the enormous human suffering, premature death and the seriously negative socioeconomic impact caused by NCDs [1]. These diseases, mainly diabetes, cardiovascular diseases, cancers and chronic lung diseases are emerging as a major threat to global development. Their magnitude is rapidly increasing, because of population ageing, unplanned urbanization and globalization of trade and marketing. These preventable problems, largely caused by unhealthy diet, physical inactivity, being overweight and obese, tobacco use and the harmful use of alcohol, are now causing an estimated 36 million deaths every year, including 9 million people dying prematurely before the age of 60 years [2]. Their major impact is on developing populations. Around 90% of deaths before the age of 60 occur in developing countries and economies in transition, in particular among the poorest and the most vulnerable people. NCDs are currently the second leading cause of death for women in low-income countries, and the leading causes of death for women in middle-income countries. Twice as many women die (per 1000 adults) from non-communicable diseases in Africa as in high-income countries [3]. Diabetes and other NCDs, which share the same risk factors, are a development issue because of the loss of household income from unhealthy behavior, from loss of productivity due to disease, disability and premature death, and from the high cost of health care which drives families below the poverty line. Additionally, the level of exposure of people in developing countries to unhealthy diets, physical inactivity, tobacco use and the harmful use of alcohol is higher than in high-income countries where a higher proportion of the population tends to be protected by comprehensive interventions aiming to promote healthier behavior. Also, affordable and accessible primary health care services for early detection, effective treatment and prevention of complications are often inadequate in developing countries [4]. NCDs and their risk factors are also closely related to poverty, and contribute to poverty at the household level. Studies in developing countries demonstrate how health care for a family member with diabetes can consume a considerable proportion of household income, and how treatment of heart disease and other cardiovascular complications greatly increases the likelihood of falling into poverty in developing countries, and due to ‘‘catastrophic” out of pocket expenditure and loss of income from ill-health [4]. NCDs are reported by the World Economic Forum to be a leading macroeconomic risk at a global level [5].

There is also evidence that diabetes and other NCDs are undermining the attainment of the Millennium Development Goals (MDGs). The links between smoking, diabetes and tuberculosis, the rising prevalence of high blood pressure and diabetes, and the increased exposure to NCD risk factors among women of child-bearing age in developing counties have direct consequences in terms of maternal health complications, pregnancy outcomes and child survival [6,7]. As a result, the 63rd World Health Assembly urged Member States, international development partners and WHO, in a resolution on health-related Millennium Development Goals, to recognize the growing burden of NCDs [8]. Demographic and epidemiological transitions, together with unplanned urbanization and globalization, have brought about a deadly interplay between infectious diseases and NCDs like diabetes. Both are major challenges in many developing countries, and both need to be effectively addressed. This interplay must no longer remain nameless and faceless: it represents the health risks and disease profile of the twenty-first century, and it is emerging as a major socioeconomic risk. Addressing this phenomenon, from a health point view requires a more integrated and effective approach to prevention and treatment – one that is based upon strengthening health systems, rather than only peering through the keyhole of one specific disease or another. From the broader development perspective, preventing and minimizing the adverse socioeconomic impact requires the active engagement of all government sectors, the industry and civil society in reducing the level of diabetes risk factors and determinants. Many more gains can be achieved by influencing the policies of non-health sectors like finance, industry, trade, urban development and education than by changes in health policies alone. The challenge can be effectively addressed [9]. There is a sound global vision and a clear road map for global and national action. The vision is represented by the Global Strategy for the Prevention and Control of NCDs, which was endorsed by the World Health Assembly in 2000 [10]. The road map is guided by its six-year Action plan, which was developed with Member States and endorsed by the Assembly in 2008. The Action Plan has six objectives, with clear sets of actions under each objective for countries, the WHO and international partners. The World Health Assembly also endorsed specific global strategies and tools to reduce the exposure to the four key risk factors, such as the WHO Framework Convention on Tobacco Control, the Global Strategy on Diet, Physical Activity and Health, and the Global Strategy to Reduce the Harmful Use of Alcohol. Objective one of the Action Plan for Global Strategy specifically focuses on integrating the prevention of diabetes and other NCDs into the global development agendas and related national

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. doi:10.1016/j.ijdm.2010.10.001

140

Editorial / International Journal of Diabetes Mellitus 2 (2010) 139–140

initiatives. To support the implementation of this objective, evidence linking NCDs to socioeconomic development has been examined and discussed in several key events organized over the last two years, in close collaboration with Member States, the United Nations Department of Economic and Social Affairs (UNDESA) and relevant United Nations Regional Commissions. Proposals for addressing the NCDs burden and its developmental challenges were made in Ministerial consultations and the Meeting of the High-level Segment of the UN Economic and Social Council in July 2009. These have made a significant contribution to the consultations of countries that have led to the recent UNGA resolution. The UNGA resolution is a major political event in the struggle against diabetes and other NCDs, and places them at the center of socioeconomic development. The resolution calls for a Highlevel Meeting of the United Nations General Assembly in September 2011 with the participation of Heads of State and Government to address the prevention and control of NCDs. Establishing a global forum on the prevention of diabetes and other NCDs emphasizes the underlying social and environmental drivers of these health problems and their implications for poverty. It is also a clear recognition of the threat that NCDs pose to the economies of countries, leading to increasing inequalities between countries and populations, thereby threatening the achievement of internationally agreed development goals, including the Millennium Development Goals. There is no doubt that the UNGA resolution has already made a major contribution to increasing the awareness of the need for global coordinated action to prevent diabetes and other NCDs, and there is now great hope that the High-level Meeting will focus on galvanizing action at global and national levels, so as to halt and address the health and socio-economic impact of diabetes and other NCDs through multi-sectoral approaches. However, the success of the Meeting will depend on the contribution that countries, the diabetes and NCDs community and other stakeholders will make in supporting the UN discussions over the coming months, and in providing technical guidance on challenges like

integrating the surveillance of NCDs into national information systems, successful mechanisms and approaches for engaging nonhealth sectors in prevention initiatives and in strengthening health systems to deliver more effective care. References [1] United Nations General Assembly. Resolution 64/265. Prevention and control of noncommunicable diseases; 2010. [2] World Health Organization. Global burden of disease 2004 update. <www.who.int/healthinfo/global_burden_disease/2004_report_update>. [3] World Health Organization. Women and health: today’s evidence, tomorrow’s evidence. . [4] World Health Organization. Discussion paper: noncommunicable diseases, poverty and the development agenda (July 2009) ECOSOC high-level segment; 2009. . [5] World Economic Forum. Global risks 2010. A global risk network report; 2010. . [6] Gajalakshmi V, et al. Smoking and mortality from tuberculosis and other diseases in India. Retrospective study of 43,000 adult male deaths and 35,000 controls. Lancet 2003. [7] World Health Organization. Equity, social determinants and public health programmes. WHO; 2010. [8] World Health Organization. Resolution WHA63.15 monitoring of the achievement of the health-related Millennium Development Goals. WHO; 2010. [9] Gaziano TA, Galea G, Reddy KS. Scaling up interventions for chronic disease prevention: the evidence. Lancet 2007;370:1939–46. [10] World Health Organization. Resolution WHA53.17. Prevention and control of noncommunicable diseases. WHO; 2000.

Ala Alwan Assistant Director General, World Health Organization, Geneva, Switzerland E-mail address: [email protected]

International Journal of Diabetes Mellitus 2 (2010) 141–143

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

Free radical activity in hypertensive type 2 diabetic patients Suvarna Prasad ⇑, Ajay Kumar Sinha Department of Biochemistry, M. M. Institute of Medical Sciences & Reasearch, Mullana, Ambala, Haryana, India

a r t i c l e

i n f o

Article history: Received 24 July 2010 Accepted 4 October 2010 Keywords: Superoxide dismutase Nitric oxide Malondialdehyde Type 2 diabetes mellitus Hypertension

a b s t r a c t Background: Free radical activity is an important cause of vascular complications in type 1 diabetes mellitus. But data regarding vascular complications in type 2 diabetes mellitus are scant. Objectives: The aim of this study was to examine free radical activity in type 2 diabetic patients with hypertension compared to those without hypertension. Materials and Methods: The serum levels of lipid peroxidation product, MDA malondialdehyde), the free radical scavenger, SOD (superoxide dismutase) and NO (nitric oxide) were studied in 50 type 2 diabetic outpatients. Controls were regarded as those diabetic outpatients who did not have hypertension. Result: Among 50 patients thus studied 19 were hypertensive. The concentration (median (range)) of both SOD (21.31(5.33–26.64) vs. 16.65(6.66–22.64) U/dl; p < 0.05) and NO (18.54 (11.40–37.07) vs. 21.39(15.69–35.65) U/dl; p < 0.05) were reduced in the hypertensive group. Similarly, concentration (median (range) of MDA (359(231–718) vs. 385(256–666) lmoles/dl; p < 0.01 were increased in the hypertensive group. Conclusion: The reduction in serum levels of SOD and NO with a concomitant increase in serum MDA levels is consistent with an increase in free radical activity in hypertensive type 2 diabetics. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction Free radical activity has been implicated in the development of diabetic vascular complications in type 1 diabetes mellitus. It plays an important role in both microvascular and macrovascular complications in diabetes mellitus. However, data regarding vascular complications in type 2 diabetes mellitus are scant. Cardiovascular diseases (CVD) are the major causes of mortality in persons with diabetes, and many factors including hypertension contribute to this high prevalence of CVD. Hypertension is twice as frequent in patients with diabetes compared with patients without the disease. Furthermore, up to 75% of CVD in diabetes may be attributable to hypertension, leading to recommendations for more aggressive treatment for those having hypertension in this disease [1]. Long-term complications of diabetes are supposed to be, at least in part, mediated by increased free radical generation and subsequent oxidative stress. In this study, we have attempted to summarize the experimental evidence in this field, and to emphasize the possible importance of oxidative stress in the development of diabetic vascular complications. Free radicals may be defined as any chemical species that contains unpaired electrons. Unpaired electrons increase the chemical ⇑ Corresponding author. Address: House No. E-54, GH-94, SEC-20, Panchkula, Haryana 134112, India. Tel.: +91 9872174466, +91 9257206509. E-mail address: [email protected] (S. Prasad).

reactivity of an atom or a molecule. Common examples of free radicals include the hydroxyl radical (0H), super oxide anion (02 ), transition metals, such as iron (Fe), copper (Cu), nitric oxide (NO) and peroxynitrite (OONO) [2]. Free radicals and reactive nonradical species derived from radicals exist in biological cells and tissues at very low concentrations [3,4]. Halliwell and Gutteridge [3] have defined antioxidants as substances that are able, at relatively low concentrations, to compete with other oxidizable substrates, and thus, to significantly delay or inhibit the oxidation of these substrates. This definition includes the enzymes SOD, glutathione peroxidase (GPx) and catalase, as well as nonenzymatic compounds such as tocopherol (vitamin E), b-carotene, ascorbic acid (vitamin C), and glutathione, which scavenge the reactive oxygen species. 2. Materials and methods The aim of this study was to investigate whether the serum levels of lipid peroxidation product, malondialdehyde (MDA), serum superoxide dismutase (SOD) and serum nitric oxide (NO) were altered in normotensive type 2 diabetic patients and type 2 diabetic patients who subsequently developed hypertension. We selected 50 type 2 diabetic patients. 19 of these type 2 diabetic patients had subsequently developed hypertension. The criteria for hypertension were a mean arterial pressure of greater than the upper range of accepted normal pressure, and a mean arterial pressure of greater than 110 mm of Hg (normal is 90 mm of Hg) that is considered to be hypertensive.

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.10.002

142

S. Prasad, A.K. Sinha / International Journal of Diabetes Mellitus 2 (2010) 141–143

Type 2 diabetic patients were subjected to an evaluation of their hypertensive state by measuring their blood pressure, three times a day for a period of one week, and the average was taken for the evaluation of blood (mean arterial blood pressure = 1/3 of pulse pressure + diastolic pressure). Blood pressure was measured in the supine position, manually in both arms, by a calibrated sphygmomanometer. Subjects underwent a medical history screening, a physical examination and laboratory analysis, which included CBC, serum electrolytes, blood urea, creatinine, fasting blood glucose and HbAc 1, ECG, and echocardiography. Exclusion criteria included tobacco, caffeine use, cardiac and pulmonary disease and evidence of left ventricular hypertrophy. The hypertensive patients were on calcium channel blockers. All medications were stopped 12 h before blood sample collection. In the morning fasting blood sample was taken. Venous blood was collected from the anterior aspect of the forearm with the help of disposable syringes. Serum was separated within 1 h after refrigeration. The straw colored supernatant serum was centrifuged and separated. All three tests were carried out within 2 h, after obtaining venous blood. The method of testing for MDA (malondialdehyde) as a marker of lipid peroxidation product was that of determined through the method of Okhawa et al. (1974) who measured MDA as thiobarbituric acid reactive substances (TBARS) [5]. Superoxide dismutase was assessed by the method of Kakkar et al. [6]. Nitric oxide was assessed through the method of Green et al. [7]. In this method, nitric oxide in serum was estimated indirectly by measuring the amount of nitrates formed from nitric oxide. 3. Observation and results These results are consistent with a significant increase in free radical activity in type 2 diabetic patients with coexistent hypertension. The SOD and NO levels were significantly decreased and MDA levels were significantly increased in those type 2 diabetic patients who had coexistent hypertension (Table 1). 4. Discussion Increased concentrations of oxygen-derived free radicals are implicated in the pathogenesis of vascular complications in diabetes. Superoxide anion appears to block endothelium derived nitric oxide mediated relaxation by inactivating the eNOS. In a hyperglycemic state, the production of superoxide is stimulated, and the en-

zyme superoxide dismutase is inhibited by non-enzymatic glycosylation known as Maillard reaction [8]. Glycation was shown to affect the C-terminal end of the enzyme, reducing its heparin binding affinity. Thus, protection against extra cellular radicals by cell surface attached SOD may be impaired in diabetes leaving the endothelial cell susceptible to damage by super oxide anion. The addition of exogenous SOD restores normal or unmasks even greater acetylcholine induced relaxation in diabetic aorta. Thus, in diabetic conditions, normal levels of antioxidant enzymes may be insufficient or may be functionally impaired, so as to preserve a physiological contractile response [9]. Nitric oxide and superoxide anion readily react to form peroxynitrite (OONO) at nearly diffusion limited rate. During physiologic conditions O2 scavengers and formation of OONO are minimal. During pathologic conditions such as in the presence of increased concentrations of O2 or after O2 scavengers are exhausted, significant concentrations of OONO may be produced. Peroxynitrite directly causes oxidation, peroxidation, and nitration of biologically important molecules (e.g. lipids protein and DNA). It is more cytotoxic than NO in a variety of experimental conditions [10]. An important example of a reaction caused by OONO is the nitration of tyrosine. Tyrosine nitration inhibits tyrosine phosphorylation, alters the dynamics of assembly and disassembly of cytoskeletal proteins, and inhibits tyrosine hydroxylase, thereby inhibiting the cycloskeletal movements of endothelial cells [10]. Nitric oxide has contrasting effects on lipids, particularly on oxidation of LDL lipoproteins in the pathogenesis of atherosclerotic lesions. NO inhibits lipid peroxidation by inhibiting radical chain propagation, reactions via radical reaction with lipid peroxyl and alkoxyl group. As a ligand to iron (and other transition metals), NO modulates the peroxidant effects of iron and thereby limits the formation of hydroxyl radicals and iron dependent electron transfer reactions. NO inhibits all and OONO mediated lipoprotein oxidation in macrophage and endothelial cell systems. However, NO, induced OONO formation can oxidize low density lipoproteins to potentially atherogenic species. In summary, OONO is more cytotoxic than NO in a variety of experimental systems and the balance of NO, O2 , and OONO , scavenging systems determine whether biologically relevant OONO concentrations will occur in tissues. Thus, the endothelium appears to modulate vascular functions by releasing relaxant substances like NO and constrictor substances like superoxide. Superoxide may play a key role in the relationship between cardiovascular diseases and metabolic disorders like diabetes mellitus, and will almost certainly prove to be a focus for future therapies [10].

Table 1 Clinical and Laboratory data of diabetic patients without hypertension as control/diabetic patients with hypertension. Parameters

Controls

Hypertensives

Number Sex (M/F) Age (In years) Wt. (kg) Duration of diabetes (years) Fasting plasma glucose (mg/dl) Glycosylated Haemoglobin (%) Mean arterial pressure (mmHg) MDA (moles/dl) SOD UB/dl NO (U/dl)

31 24/7 54 (35–65) 54 (40–70) 4 (0A–14) 175 ± 67 9.6 ± 1.7 97 ± 2 359 (231–718) p < 0.01, SD ± 55.85 21.31 (5.33–26.64) SD ± 4.64 18.54 (11.40–37.07) SD + 4.34

19 15/4 51 (40–71) 52 (40–71) 3 (0–20) 223± 84 p < 0.001 11 ± 2.8 p < 0.001 116 ± 6 p < 0.05 385 (256–666), SD ± 63.34 16.65 (6.66–22.64) SD ± 4.08 p < 0.05 21.39 (15.69–35.65) SD ± 4.79 p < 0.05,

Median (Range), SOD, superoxide dismutase, NO, nitric oxide; MDA, malondialdehdye. A, newly diagnosed type 2 diabetic patients. B, one unit of enzyme is defined as enzyme concentration required to inhibit the optical density of chromogen production by 50% in 1 min. Table 1 shows the clinical and biochemical details of the study groups. There were no significant differences in the age, weight, sex, and duration of diabetes between the two study groups. Patients with hypertension had higher plasma fasting glucose levels (p < 0.001) and glycosylated hemoglobin levels (p < 0.001) compared to those without hypertension.

S. Prasad, A.K. Sinha / International Journal of Diabetes Mellitus 2 (2010) 141–143

5. Conclusion Much evidence suggests that free radical over generation may be considered the key in the generation of insulin resistance, diabetes and cardiovascular disease. Many new specific causal antioxidants are being developed [10,11], and may become important tools in opposing the increasing epidemic of diabetes a real emergency in our future. It has been demonstrated that insulin resistance is associated in humans with reduced intracellular antioxidant defense [12], and that diabetic subjects prone to complications may have a defective intracellular antioxidant response [13,14] where what we call genetic predisposition to diabetes, as well as liability to its late complications, might be based on a deficient ROS-scavenging ability in b-cells and/or in target tissues such as endothelium. Oxidative stress is involved in various cardiovascular diseases, including atherosclerosis, hypertension and the aging process; therefore, therapeutic strategies to modulate this maladaptive response should become a target for future extensive investigation, and could have a broad application [15]. References [1] Sowers James R, Murray E, Edward DF. Diabetes, hypertension, and cardiovascular disease: an update. Hypertension 2001;37:1053–9. [2] Betteridge DJ. What is oxidative stress? Metabolism 2000;49:3–8.

143

[3] Tirzitis G, Bartosz G. Determination of antiradical and antioxidant activity: basic principles and new insights. Acta Biochim Pol 2010;57:139–42. [4] Sies H. Strategies of antioxidant defence. Eur J Biochem 2005;215:213–9. [5] Ohkawa H, Ohishi N, Yogi K. Assay of lipid peroxides in animal tissues by thiobarbituric acid reaction. Annal Biochem 1979;95:351–8. [6] Kakkar P, Awasthi S, Vishwanathan PN. Oxidative changes in brain of aniline exposed rats. Arch Environ Contam Toxicol 1992;23:307–9. [7] Green LC, Wagner DA, Glogowski J, Skipper PL, Wishnok JS, Tannebaum SR. Analysis of nitrate, nitrite, and [N 15] nitrate in biological fluids. Anal Biochem 1982;126:131–8. [8] Taboshashi K, Saito Y. The role of superoxide anion in relationship between the cardiovascular diseases and the metabolic disorders associated with obesity. Nippon Rinsho 2000;58:1592–7. [9] Szaleczky E, Prechl J, Feher J, Somogyi A. Alterations in enzymatic defence in diabetes mellitus – a rational approach. Postgrad Med J 1999;75:13–7. [10] Ceriello A. New Insights on oxidative stress and diabetic complications may lead to a causal antioxidant therapy. Diabetes Care 2003;26:1589–96. [11] Cuzzocrea S, Riley DP, Caputi AP, Salvemini D. Antioxidant therapy: a new pharmacological approach in shock, inflammation, and ischemia/reperfusion injury. Pharmacol Rev 2001;53:1159. [12] Ceriello A, Morocutti A, Mercuri F, Quagliaro L, Moro M, Damante G, et al. Defective intracellular antioxidant enzyme production in type 1 diabetic patients with nephropathy. Diabetes 2000;49:2170–7. [13] Hodgkinson AD, Bartlett T, Oates PJ, Millward BA, Demaine AG. The response of antioxidant genes to hyperglycaemia is abnormal in patients with type 1 diabetes and diabetic nephropathy. Diabetes 2003;52:846–51. [14] Bruce CR, Carey AL, Hawley JA, Febbraio MA. Intramuscular heat shock protein 72 and heme oxygenase-1 mRNA is reduced in patients with type 2 diabetes: evidence that insulin resistance is associated with a disturbed antioxidant defence mechanism. Diabetes 2003;52:2338–45. [15] Tsutsui Hiroyuki, Kinugawa Shintaro, Matsushima Shouji. Mitochondrial oxidative stress and dysfunction in myocardial remodelling. Cardiovasc Res 2009;81:449–56.

International Journal of Diabetes Mellitus 2 (2010) 144–147

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

Association of serum free IGF-1 and IGFBP-1 with insulin sensitivity in impaired glucose tolerance (IGT) Golam Kabir a,b,⇑, Mosaraf Hossain a,b, M. Omar Faruque a, Naimul Hassan a,b, Zahid Hassan a, Quamrun Nahar a, Sultana Marufa Shefin c, Mohammad Alauddin b, Liaquat Ali a a b c

Biomedical Research Group (BMRG), BIRDEM, Dhaka, Bangladesh Department of Biochemistry & Molecular Biology, University of Chittagong, Chittagong-4331, Bangladesh Dept. of Endocrinology and Diabetology, BIRDEM, Dhaka, Bangladesh

a r t i c l e

i n f o

Article history: Received 31 May 2010 Accepted 4 September 2010

Keywords: Free IGF-1 IGFBP-1 BMI IGT

a b s t r a c t Aim and Background: Free IGF-1 and IGFBP-1 are associated with obesity which is one of the major features of insulin resistance. But very few studies exist on free IGF-1 and IGFBP-1 in IGT subjects. The present study was undertaken to investigate the association of free IGF-1 and IGFBP-1 with insulin sensitivity in IGT subjects. Subjects and Methods: Ninety-one subjects with impaired glucose tolerance (IGT) were studied along with age- sex- and BMI-matched sixty-one healthy Controls without family history of diabetes or prediabetes. Insulin, free IGF-1 and IGFBP-1 were measured by standard ELISA method. Insulin secretory capacity (HOMA B) and insulin sensitivity (HOMA S) were calculated using fasting glucose and fasting insulin by HOMA-CIGMA software. Results: Fasting free IGF-1 and IGFBP-1 levels were not significantly different among the study groups. In stepwise multiple regression analysis, when free IGF-1 was considered as a dependent variable with other independent variables, model 1 (b = 0.352, p = 0.03), model 2 (b = 0.355, p = 0.033) and model 5 (b = 0.378, p = 0.026) have shown significant association of fasting glucose with free IGF-1. Similarly when IGFBP-1 was considered as a dependent variable, model 4 (b = 0.865, p = 0.03) and model 5 (b = 0.1.07, p = 0.004) have shown negative association of fasting glucose with IGFBP-1. In this analysis model 5 have also shown negative association of HOMA S with IGFBP-1 (b = 1.015, p = 0.017). Conclusion: IGF1 and IGFBP-1 seems to be negatively associated with fasting glucose in IGT subjects and insulin sensitivity (HOMA S) may also be negatively associated with IGFBP-1 in IGT subjects. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction Insulin, like growth factor-1 (IGF-1), is a multipotent growth factor with important action on normal tissue growth and metabolism. In addition, IGF-1 has been suggested to have beneficial effects on glucose homeostasis, due to its glucose lowering and insulin sensitizing actions. Epidemiological studies suggest that IGF-1 is also involved in the development of common cancer, atherosclerosis and type 2 diabetes [1–4]. In several pathological states, an impairment of IGF-1 action on glucose metabolism has been recorded, along with insulin resistance [5–7]; however, it is not known whether IGF-1 and insulin resistance are always associated and it is not clear whether resistance, when it does occur, affects only glucose uptake and metabolism or protein metabolism as well. Frystyk et al. (1999) have found that circulating fasting free IGF-I is increasingly elevated with increasing obesity, whereas ⇑ Corresponding author at: Department of Biochemistry & Molecular Biology, University of Chittagong, Chittagong-4331, Bangladesh. Tel.: +8801711943681. E-mail address: [email protected] (G. Kabir).

serum total IGF-I is normal. They have proposed that elevated serum free IGF-I may be caused by insulin resistance inducing hyperinsulinemia which suppress IGFBP-1 [8]. Although IGF-1 is structurally related to insulin, unlike insulin, it circulates bound to specific proteins called IGF binding proteins (IGFBPs) with variable affinity [9]. IGFBP-1 levels have been shown to be elevated in type 1 diabetes and in patients with insulin resistance syndromes. Type 2 diabetes tends to have low serum IGFBP-1 levels. Patients with growth hormone deficiency tend to have elevated IGFBP-1 levels [10]. Insulin inhibits the hepatic synthesis and secretion of IGFBP-1 [11,12] and increases the portal concentrations of insulin decrease serum levels of IGFBP-1 in obese subjects [8]. Frystyk et al., 1999 [8] have shown that simple obesity was associated with reduced levels of IGFBP-1 when compared to lean control and obese type 2 diabetes. Free IGF-1 and IGFBP-1 have been well studied in type 1 diabetic subjects and also in type 2 diabetic subjects with higher BMI (BMI > 30). In developing countries like Bangladesh, type 2 diabetic patients mostly possess lower to normal BMI, and no reports of free IGF-1 and IGFBP-1 exist in this physiological

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.09.003

145

G. Kabir et al. / International Journal of Diabetes Mellitus 2 (2010) 144–147

condition. But this is very important because BMI is a major risk factor for insulin resistance. The variation in serum concentrations of IGF-1 and IGFBP-1 occurs due to racial variation. Moreover Impaired glucose tolerance (IGT), which is also known as the defect of insulin resistance, is not well explored in regard to free IGF-1 and IGFBP-1 issues, which may help to know the early mechanisms of the onset of insulin resistance and the development of diabetes. The present study has been undertaken to explore the association of insulin resistance with free IGF-1 and IGFBP-1 in IGT subjects. 2. Materials and methods This cross-sectional observational study was conducted in Bangladesh Institute of Research and Rehabilitation in Diabetes, Endocrine and Metabolic Disorders (BIRDEM), Dhaka. A group of 91 impaired glucose tolerant (IGT) subjects were selected purposively from the Out-Patient Department (OPD) of BIRDEM, along with a group of 61 age-, sex- and BMI-matched healthy subjects without family history of diabetes as Controls from the friend circle of the IGT subjects considering the same socio-economic status. Written consent was taken from all the volunteers; clinical examinations Table 1 Clinical characteristics of the study subjects. Variable

Control (n = 61)

IGT (n = 91)

Age (yrs) BMI (kg/m2) WHR MUAC (mm) Triceps (mm) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting glucose (mmol/l) Postprandial glucose (mmol/l)

36 ± 6 25.4 ± 3.8 0.91 ± 0.05 298 ± 27 14.6 ± 5.0 115 ± 9 77 ± 9 5.2 ± 0.5 5.9 ± 1.7

43 ± 9 26.3 ± 4.0 0.92 ± 0.05 303 ± 30 15.3 ± 5.4 122 ± 17 82 ± 9* 5.7 ± 0.6* 9.2 ± 1.3*

Results are expressed as M ± SD. * p < 0.05, significantly different compared to controls when using Student’s ‘t’ test. Table 2 Serum insulinemic, lipid profile, IGF-1 and IGFBP-1 status of the study subjects. Variables

Control (n = 61)

Fasting Insulin (pmol/l) HOMA B HOMA S Triglyceride (mg/dl) Cholesterol (mg/dl) HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl) Free IGF-1 (pg/ml) IGFBP-1 (ng/ml)

were undertaken by a registered physician using a predesigned questionnaire. Anthropometric measurements were taken using standard methods. Subjects were requested to come on a prescheduled morning, after overnight fasting for the fasting blood sample; subjects were then given 75 gm anhydrous glucose dissolved in 250 ml water. Blood was taken at fasting conditions and 2 h after glucose loading. Serum glucose, cholesterol, triglyceride and HDL were determined by the enzymatic colorimetric method, using commercial kits (Randox Laboratories Ltd., UK). The LDL cholesterol in serum was calculated by using the formula: LDLcholesterol = Total cholesterol (TG/5 + HDL-cholesterol). Serum insulin levels were determined by enzyme linked immunosorbent assay (ELISA) method (Linco Research Inc., USA). Serum free IGF-1 and IGFBP-1 concentrations were measured by enzyme linked immunosorbent assay (ELISA) method (Ray Biotech, USA). Insulin secretory capacity (HOMA B) and insulin sensitivity (HOMA S) were calculated from fasting glucose and fasting insulin using HOMA-CIGMA software [13]. 2.1. Statistical analysis Data were expressed as mean ± SD (standard deviation), median (range) and/or percentage (%) as appropriate using SPSS (Statistical Package for Social Science) software for Windows version 10 (SPSS Inc., Chicago, Illinois, USA). The statistical significance of the differences between the values was assessed by Student’s ‘t’ test or Mann–Whitney U test (as appropriate). A two-tailed p value of <0.05 was considered to be statistically significant. 3. Results and observations 3.1. Clinical characteristics of the study subjects Anthropometric measurements (BMI, WHR, MUAC, Triceps) showed no difference I in controls and IGT subjects. Diastolic blood pressure (mmHg) was significantly (p = 0.007) higher in IGT subjects compared to that of Controls (Table 1). 3.2. Insulinimic status of the study subjects

IGT (n = 91) *

51.7 (7.8–155.9) 99 (21–187) 86 (29–554) 136 (52–408) 192 (90–261) 30 (13–59) 125 (46–203) 118.2 (39.4–486.1) 11.5 (1.1–83.97)

67.7 (6.9–237.6) 95(26–278) 66 (20–661)* 150 (50–491) 194 (105–298) 30 (18–54) 127 (63–239) 118.2 (19.9–465.9) 13.8 (1.6–68.1)

HOMA%B = B cell function assessed by homeostasis model assessment; HOMA%S = insulin sensitivity assessed by homeostasis model assessment; Free IGF-1 = free insulin like growth factor-1 and IGFBP-1 = insulin like growth factor binding protein-1. * p < 0.05, significantly different compared to controls when using Student’s ‘t’ test.

Fasting serum insulin level was significantly higher in IGT (p = 0.004) compared to that of Controls. Insulin sensitivity (HOMA S) was significantly lower in IGT subjects (p = 0.001) compared to that of Controls. Fasting serum free IGF-1 level and IGFBP-1 level of IGT subjects showed no significant difference compared to that of Controls (Table 2). 3.3. Bivariate correlation Pearson’s correlation analysis has shown a significant association of fasting serum glucose with free IGF-1 (r = 0.337, p = 0.025) but not with IGFBP-1 (Table 3).

Table 3 Correlation of serum free IGF-1 and IGFBP-1 with different variables among the study groups. Group Control

BMI Free IGF-1 IGFBP-1

IGT

Free IGF-1 IGFBP-1

r p r p r p r p

0.078 0.709 0.192 0.197 0.240 0.142 0.056 0.685

F_G 0.116 0.582 0.16 0.275 0.337 0.025 0.121 0.380

F_INS 0.099 0.639 0.186 0.205 0.053 0.749 0.210 0.124

Data are expressed as correlation coefficient (Paerson’s rho) r values and p = level of significance.

TG 0.295 0.152 0.227 0.120 0.097 0.555 0.202 0.138

CHOL 0.086 0.683 0.043 0.773 0.168 0.306 0.005 0.974

HOMA-B% -0.056 0.789 0.062 0.676 0.180 0.275 0.157 0.253

HOMA-S% -0.106 0.614 0.29 0.153 0.044 0.791 0.218 0.111

146

G. Kabir et al. / International Journal of Diabetes Mellitus 2 (2010) 144–147

Table 4 Stepwise Multiple regression analysis of free IGF-1 (dependent variable) in IGT subjects. Variables

Model 1 p

b F_GLU TG T_CHOL F_INS HOMA S% HOMA B% BMI R2

Model 2

0.352

p

b

0.030

0.355 0.019

0.100

Model 3 b

0.033 0.904

0.074

0.321 0.071 0.167

Model 4

Model 5

p

b

p

0.058 0.677 0.339

0.318 0.085 0.163 0122

0.062 0.621 0.353 0.455

0.073

p

b

0.061

Model 6

0.378 0.014 0.276 0.637 0.814

0.026 0.936 0.128 0.141 0.062

0.133

Model 7 p

b 0.051 0.007 0.291 1.110 0.664 0.773

p

b

0.892 0.968 0.107 0.056 0.136 0.212

0.149

0.131 0.002 0.306 0.967 0.519 0.802 0.173 0.149

0.734 0.991 0.092 0.104 0.265 0.197 0.325

b stands for standardized regression coefficients. R2 for adjusted R square (Multiple coefficient of determination).

Table 5 Stepwise multiple regression analysis of IGFBP-1 (dependent variable) in IGT subjects. Variables

HOMA S% HOMA B% F_INS F_GLU BMI R2

Model 1

Model 2

Model 3

b

p

b

p

0.145

0.392

0.200 0.077

0.414 0.751

0.007

0.033

Model 4 p

b 0.188 0.267 0.577

0.643 0.359 0.237

0.020

Model 5 p

b 0.587 1.014 0.065 0.865 0.094

0.170 0.115 0.904 0.030

p

b 1.015 1.084 0.363 1.070 0.477 0.270

0.017 0.063 0.473 0.004 0.006

b stands for standardized regression coefficients. R2 for adjusted R square (Multiple coefficient of determination).

3.4. Stepwise multiple regression analysis In stepwise multiple regression analysis, when free IGF-1 was considered as a dependent variable with other independent variables, model 1 (b = 0.352, p = 0.03), model 2 (b = 0.355, p = 0.033) and model 5 (b = 0.378, p = 0.026) have shown a significant association of fasting glucose with free IGF-1 (Table 4). Similarly when IGFBP-1 was considered as a dependent variable, model 4 (b = 0.865, p = 0.03) and model 5 (b = 0.1.07, p = 0.004) have shown a negative association of fasting glucose with IGFBP-1 (Table 5). In this analysis model 5 has also shown a negative association of HOMA S with IGFBP-1 (b = 1.015, p = 0.017).

4. Discussion It has been documented that free IGF-1 and IGFBP-1 are associated with type1 diabetes, as well as with obesity [14]. Studies on obese type 2 diabetic subjects have shown an increasing tendency of free IGF-1 and a decreasing tendency of IGFBP-1. Studies on obese IGT subjects have also claimed similar results. Unfortunately, however, no studies exist on IGT with lower to normal BMI, which mostly dominates in the developing countries like Bangladesh. In this study the median (range) value of serum free IGF-1 (pg/ ml) in healthy subjects was 118 (39–486). A study of Danish population has shown that free IGF-1 is significantly higher in obese healthy subjects compared to lean healthy subjects [14]. The present data show that Bangladeshi subjects have much lower levels of free IGF-1 than that of the European population. A study of Bangladeshi children aged 5–6 yrs, irrespective of gender, has shown that total IGF-1 concentration level is much lower than that of European children aged 3–6 yrs [15]. Another study in United States based on age-adjusted Asian, African–American and Caucasian population have shown that Asian population have significantly lower IGF-1 than Caucasians and African–Americans, which indicates that IGF-1 has considerable racial variations [16].

In the present study, the serum level of free IGF-1 was not significantly higher in IGT subjects compared to healthy Controls. In a study by Frystyk et al. (1999), it has been shown that the levels of free IGF-1 were increased in obese controls (BMI, 31.6 ± 0.7) compared to lean controls (BMI, 22.8 ± 0.2), but in obese type 2 diabetes (BMI, 32.3 ± 0.8) the levels of free IGF-1 did not differ significantly from either lean or obese controls [8]. A study on the Korean population has also shown that free IGF-1 concentrations were significantly elevated in obese subjects (free IGF-1. 1.46 ± 1.1 lg/l; BMI 30 ± 2.5) when compared to controls (free IGF-1. 0.91 ± 0.9 lg/l; BMI. 21.3 ± 1.4). There is an increasing tendency for free IGF-1 in IGT (140 pg/ml) subjects than those of controls (96 pg/ml) [9]. No studies have so far looked at free IGF-1 in IGT or any other prediabetic subject. In this study the values of serum IGFBP-1 had no significant difference in IGT subjects, compared to controls. Similar values of IGFBP-1 were found in control subjects with a normal BMI in the Korean and Danish population [9,14]. In these studies they have shown that obese people have significantly lower values of IGFBP-1 compared to lean controls and obese type 2 diabetic subjects. Another study done in the USA has shown that IGFBP-1 in type 1 diabetes was significantly higher compared to healthy controls and type 2 diabetic subjects [17]. Free IGF-1 was significantly (r = 0.337, p = 0.025) associated with fasting serum glucose in simple Pearson’s correlation which is also reflected in stepwise multiple regression where both free IGF-1 (model 1, 2 and 5 in Table 4) and IGFBP-1 (model 4 and 5 in Table 5) showed themselves to be negatively associated with fasting serum glucose. HOMA S in stepwise multiple regressions have also been shown to be significantly associated with IGFBP-1 (model 5 in Table 5). Previous studies [9,14] documented that obesity tends to lower the level of IGFBP-1, and in general, it is accepted that obesity is associated with hyperglycemia, so hyperglycemia may lower IGFBP-1 or vice versa. In this study although the studied subjects were not highly obese, this idea strongly follows, and BMI in stepwise multiple regression analysis has shown significantly to be associated with IGFBP-1 (model 5 in Table 5).

G. Kabir et al. / International Journal of Diabetes Mellitus 2 (2010) 144–147

5. Conclusion IGF1 and IGFBP-1 seem to be negatively associated with fasting glucose in IGT subjects and insulin sensitivity (HOMA S) may also be negatively associated with IGFBP-1 in IGT subjects.

Acknowledgement Authors greatly acknowledge the Diabetic Association of Bangladesh and International Program in the Chemical Sciences (IPICS), Uppsala University, Sweden for the financial support of this study.

References [1] Jan Frystyk. Free insulin-like growth factors – measurements and relationships to growth hormone secretion and glucose homeostasis. Growth Horm IGF Res 2004;14(5):337–75. [2] Khandwala HM, McCutcheon IE, Flyvbjerg A, Friend KE. The effects on insulin like growth factors on tumorigenesis, neoplastic growth. Endocr. Rev 2000;21:215–44. [3] Vaessen N, Heutink P, Janseen JA, Witteman JC, Testers L, Hofman A, et al. A polymorphism in the gene for IGF-1; functional properties and risk for type 2 diabetes and myocardial infraction. Diabetes 2001;50:637–42. [4] Moschos SJ, Mantzoros CS. The role of IGF system in cancer: from basic to clinical studies and clinical applications. Oncology 2002;63:317–32. [5] Sowell MO, Robinson KA, Buse MG. Phenylarsine oxide and denervation effects on hormone-stimulated glucose transport. Am J Physiol 1988;255: E159–65.

147

[6] Dohm GL, Elton CW, Raju MS, Mooney ND, Dimarchi R, Pories WJ, et al. IGF-I stimulated glucose transport in human skeletal muscle and IGF-I resistance in obesity and NIDDM. Diabetes 1990;39:1028–32. [7] Sherwin RS, Greenawalt K, Shulman GI. Simultaneous in-sulin-like growth factor I and insulin resistance in obese zucker rats. Diabetes 1992;41:691–7. [8] Frystyk J, Vestbo E, SkjÒrbÒk C, Mogensen CE, Èrskov H. Free insulin-like growth factors in human obesity. Metabolism 1995;44(Suppl 4):37–44. [9] Nam SY, Lee EJ, Kim KR. Effect of obesity on total and free insulin-like growth factor (IGF)-1, and their relationship to IGF-binding protein (BP)-1, IGFBP-2, IGFBP-3, insulin, and growth hormone. Int J Obes Relat Metab Disord 1997;21:355–9. [10] Hills FA, Gunn LK, Hardiman. ‘‘IGFBP-1 in the placenta, membranes and fetal circulation: levels at term and preterm delivery”. Early Hum Dev 1996;44(1):71–6. [11] Brismar K, Fernqvist Forbes E, Wahren J, Hall K. Effect of insulin on the hepatic production of insulin-like growth factor-binding protein-1 (IGFBP-1), IGFBP-3, and IGF-I in insulin-dependent diabetes. J Clin Endocrinol Metab 1994;79:872–8. [12] Lee PD, Conover CA, Powell DR. Regulation and function of insulin-like growth factor-binding protein-1. Proc Soc Exp Biol Med 1993;204(1):4–29. [13] Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998;21:2191–2. [14] Frystyk J, Skjñrbñk C, Vestbo E, Fisker S, érskov H. Circulating levels of free insulin-like growth factors in obese subjects: the impact of type 2 diabetes. Diabetes Metab Res Rev 1999;15:314–22. [15] Burtis CA, Ashwood ER, Bruns DE Tietz. Textbook of Clinical chemistry and Molecular diagnostics. Fourth edition. (Elsevier) 2006, p. 1971–75. [16] Elizabeth A. Platz,2 Michael N. Pollak, Eric B. Rimm, Noreen Majeed, Yuzhen Tao, Walter C. Willett, et al. Racial variation in insulin-like growth factor-1 and binding protein-3concentrations in middle-aged men. Cancer, Epidemiol Biomarkers Prevention 1999;8:1107–1110. . [17] Jehle PM, Jehle DR, Mohan S, Bohm BO. Serum levels of insulin-like growth factor system components and relationship to bone metabolism in type 1 and type 2 diabetes mellitus patients. J Endocrinol 1998;159(2):297–306.

International Journal of Diabetes Mellitus 2 (2010) 148–153

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

The prevalence and incidence of and risk factors for, micro-albuminuria among urban Africans with type 1 diabetes in South Africa: An inter-ethnic study W.J. Kalk ⇑, F.J. Raal, B.I. Joffe Division of Endocrinology and Metabolism, University of the Witwatersrand, Johannesburg, South Africa

a r t i c l e

i n f o

Article history: Received 10 August 2010 Accepted 13 October 2010

Keywords: Sub-Saharan Africans Type 1 diabetes Micro-albuminuria Nephropathy

a b s t r a c t Background: Type 1 diabetes (T1DM) in sub-Saharan Africans is rare and is associated with high mortality from nephropathy. We studied the prevalence and potential risk factors for microalbuminuria (MA) in African and in age-of-onset matched white patients with T1DM. Risk factors for MA were evaluated prospectively in an African cohort. Materials and Methods: 68 African and 134 white patients, age at diagnosis 10–40 years, duration of diabetes > 2 years, were evaluated for MA; 48 Africans were followed prospectively. Results: Africans had shorter duration of diabetes (median, 8 years vs 11 years), higher HbA1c (10.62(SD 2.52)%, vs 9.02(2.44)%, lower cholesterol (4.45(1.04) vs 5.45(1.16)mmol/l), and fewer (23.5% vs 54.5%) had adolescent diabetes onset (p 0.0030 for each); the prevalence of MA was 39.7% and 24.6% respectively (p = 0.0155). In multiple regression analysis MA was associated with mean HbA1c (p < 0.0001), younger age at diagnosis (p = 0.0060), SBP (p = 0.0012) and African race (p = 0.0287). Prospectively, Africans developing MA (45%) had higher mean HbA1c levels (p = 0.0001), were more likely to have had adolescent onset of DM (33.3% vs 8.0%, p = 0.0310) and lower BMI (p = 0.0340); logistic regression revealed that higher HbA1c and SBP, and lower BMI predicted MA. Nine of 16 African subjects progressed to macroalbuminuria; they were characterised only by extremely poor glycaemic control (mean HbA1c, 13.49(2.00)%). Conclusions: Microalbuminuria, and severe hyperglycaemia, are common in diabetic Africans with short duration TIDM; MA may rapidly progress to macroalbumiuria. African race may be associated with increased susceptibility to diabetic nephropathy. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction Type 1diabetes (T1DM) is relatively rare in sub-Saharan Africans, especially in young children – the peak age of onset is about a decade older than in white Europeans [1–3]. Although data are limited, available information indicates that the prognosis in T1DM is poor in Africa, as a result of both acute and long-term complications [2,4]. Diabetic nephropathy appears to be particularly frequent in diabetic Africans and is a major cause of morbidity and mortality [4–6], perhaps more so than in comparable populations of European extraction; no comparative data have been published. Micro-albuminuria (MA) is a marker of early diabetic renal disease and is a precursor of overt diabetic nephropathy, although it may regress in a substantial proportion of patients [7]. Influences on and risk factors for the development of early diabetic nephropathy or its progression in T1DM include gender, age of onset of diabetes, ⇑ Corresponding author. Present address: Endocrinology Unit, Musgrove Park Hospital, Parkfield Drive, Taunton, TA1 5DA, Somerset, UK. Tel.: + 44 0 1823 344536; fax: + 44 0 1823 344542. E-mail addresses: [email protected], [email protected] (W.J. Kalk), Fredrick. [email protected] (F.J. Raal), [email protected] (B.I. Joffe).

duration of disease, poor glycaemic control, blood pressure, lipids, central obesity and psychosocial and genetic factors [7–20]. The great majority of patients studied have been of European extraction and publications on renal involvement in T1DM from Africa are few and cross-sectional [2,4,22]. Moreover, uncertainty remains about some potential risk factors, notably the roles of blood pressure and dyslipidaemia in the genesis of MA. We have, therefore, investigated the prevalence of early diabetic nephropathy and some associations in a group of African patients with TIDM in urban South Africa, a population in epidemiological transition and characterised by relatively low serum lipid concentrations; an age matched white patient group from the same institution was studied by way of comparison. In a subgroup of Africans, the incidence of and potential risk factors for MA and for progression to macro-albuminuria were evaluated in a prospectively studied cohort. 2. Subjects and methods Patients attending the Diabetes Service at the Johannesburg Academic Hospital, South Africa, between 1994 and 2008, with age at diagnosis of diabetes 640 years, were studied. African

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.10.003

149

W.J. Kalk et al. / International Journal of Diabetes Mellitus 2 (2010) 148–153

Table 1 The clinical and laboratory characteristics of all patients with type 1 diabetes and in the African and White subjects separately, excluding those with overt nephropathy. Data are expressed as mean (SD) or median (IQR).

% male Age (years) Age diagnosis Adolescent onset (%) Duration DM BMI Hypertension (%) SBP DBP Smokers (%) (n = 165) HbA1c (%) Cholesterol (mmol/l) LDL-C (mmol/l) (n = 128) HDL-C (mmol/l) (n = 79) Trigycerides(mmol/l) (n = 133) Creatinine (umol/l)

All subjects (n = 202)

African (n = 68)

White (n = 134)

p

60.4 34.7 (10.2) 22.5 (8.1) 44.1 9.0 (5.0,15.3) 24.3 (4.3) 29.2 121.1 (14.1) 76.0 (8.8) 39.0 9.56 (2.57) 5.09 (1.21) 3.03 (1.03) 1.30 (0.42) 1.00 (0.80,1.58) 88.0 (81.0, 98.0)

55.9 34.9 (8.6) 26.2 (7.3) 23.5 8.0 (4.0,11.0) 24.6 (3.9) 33.8 119.3 (14.4) 76.9 (9.5) 31.3 10.62 (2.52) 4.45 (1.04) 2.50 (0.82) 1.31 (0.41) 0.90 (0.70,1.20) 87.0 (79.0,96.0)

62.7 34.6 (11.0) 20.6 (8.0) 54.5 11.0 (6.0,18.0) 24.0 (4.7) 26.9 121.3 (15.3) 75.5 (8.4) 43.1 9.02 (2.44) 5.42 (1.16) 3.34 (1.03) 1.43 (0.42) 1.10 (0.90,1.80) 89.5 (81.8,99.0)

0.459 0.844 <0.0001 <0.0001 0.0003 0.408 0.304 0.367 0.290 0.115 <0.0001 <0.0001 <0.0001 0.114 0.0026 0.092

patients were defined as those with African parentage with indigenous African names; several ‘tribal’ affiliations and first languages were included; all were born in southern Africa. White patients were those of northern and southern European extraction. None of the patients had known genetic admixtures. Type 1 diabetes was diagnosed according to ADA criteria [23]. All subjects had symptomatic hyperglycaemia at their diagnosis; many presented with keto-acidosis. All were considered to be ‘ketosis-prone’, and all required insulin therapy from diagnosis. Because the precise classification of diabetes in adult Africans can be difficult [23] as type 2 diabetes may commonly present with keto-acidosis [24]; among the African subjects aged 31–40 years at diagnosis, only those who presented acutely, required insulin continuously from diagnosis and/or were positive for antibodies against glutamic acid decarboxylase (GAD-65 antibodies) [25] were accepted as having T1DM. As persistent abnormal urinary albumin excretion seldom occurs before two to three years of diabetes, only patients with duration of diabetes of more than two years were included in the analyses. Blood pressure (BP) was assessed with the subjects seated after at least five minutes rest; large cuffs were used for obese patients. Pre-existing hypertension was diagnosed in the absence of prior abnormal urinary albumin excretion if the systolic BP (SBP) was consistently P140 mmHg systolic and/or diastolic BP (DBP) P90 mmHg or in individuals treated for hypertension. Hypertensive patients were treated with angiotensin converting enzyme inhibitors and diuretics, usually both in the African subjects, with the addition of calcium channel blockers as third line agents, in accordance with local protocols. HMG-Co A reductase inhibitors (statins) were used only in the later part of the study; cholesterol data included statin treated patients. Obesity was evaluated by body mass index (BMI; mass kg/height m2). Subjects were classified as smokers if they were current or past users of tobacco. Urinary albumin excretion was assessed from urine albumin:creatinine ratio measurements (ACR, mg/mmol); MA was defined as an ACR >2.5 mg/mmol in males or >3.5 mg/mmol in females in at least two of three urine collections, and macro-albuminuria – overt nephropathy – was defined as an ACR P30 mg/mmol. Since the day-to-day variability of urinary albumin excretion is some 50%, for patients who provided only a single urine specimen urinary ACRs P5.0 mg/mmol and P7.0 mg/mmol for men and women, respectively (i.e., double the conventional cut-points) were used to define MA. The methods for HbA1c (DCCT linked), serum lipids and creatinine, urinary albumin and creatinine measurements have been described previously [26]. Mean values for blood pressure, HbA1c and total cholesterol prior to the onset of MA, or until the end of follow-up in those with normal ACR, were used in the analyses of putative risk factors.

Data are expressed as a mean (SD) or median (inter-quartile range – IQR) for variables with a skewed distribution. Differences between groups were evaluated by the unpaired T-test or the Mann Whitney U-test and the Chi squared test. Multiple and logistic regression analyses were used to assess the independent associations of putative risk factors with MA. The study was approved by the Committee for Research on Human Subjects of the University of Witwatersrand. 3. Results Ninety-one African patients with type 1 diabetes attended at least once during the study period. Of these, 78(85.7%) had both a duration of diabetes greater than 2 years and at least one urinary ACR measurement; 10(12.8%) had macro-albuminuria when first evaluated and were, therefore, excluded. Thus, 68 patients with duration >2 years were available for analyses for the prevalence of MA. The youngest age at diagnosis of diabetes among the Africans was 10 years; therefore, only white patients aged 10–40 years at diagnosis of T1DM attending during the same period were included in the comparative analyses (n = 134). Micro-albuminuria was significantly more prevalent in African than in white patients – 39.7% and 24.6%, respectively (p = 0.0155), despite a shorter median duration of diabetes (8.0 vs 11.0 years respectively) and similar blood pressures and prevalence of pre-existing hypertension in each group. Blood pressure in hypertensive subjects was higher than in those with normotension, but hypertension was reasonably well controlled (hypertensive patients, 131.4/80.0 vs 116.2/74.3 mmHg in normotensive subjects; p < 0.001 for both). Other group differences include proportionately fewer African patients with diagnosis during adolescence (age 10–20 years), higher mean levels for HbA1c but lower total and LDL cholesterol and triglyceride concentrations (Table 1). Data on African and white patients without and with MA are shown in Table 2. Among the African subjects, those with MA were characterised by a significantly higher BP, mean HbA1c and cholesterol levels and lower BMI; more patients with MA (32% vs 17%, p = 0.161) had adolescent onset of diabetes. Among white patients, the presence of MA was associated with female gender, younger age at diagnosis and more frequent adolescent onset, a higher prevalence of hypertension and higher SBP and higher HbA1c. Multiple regression analysis of all patients, after incorporating all measurements significantly different in the univariate analyses as the independent variables revealed significant independent associations between MA and younger age at the onset of diabetes (p = 0.0060), higher HbA1c (p < 0.0001) and SBP (p = 0.0012), the presence of pre-existing hypertension (p = 0.0068), lower BMI

150

W.J. Kalk et al. / International Journal of Diabetes Mellitus 2 (2010) 148–153

Table 2 Comparison of variables in African and White patients without (MA ( )) and with micro-albuminuria (MA (+)). Data are expressed as mean (SD) or median (IQR). Categories marked with an asterisk (*) indicate significant differences between African and White patients with mico-albuminuria. Urine ACR

Males (%) Age (years) Age at diagnosis (years) Adolescent onset (%) Duration (years) BMI Hypertension (%) Systolic BP Diastolic BP Smokers (%) HbA1c (%) Cholesterol (mmol/) LDL-C (mmol/l) HDL-C (mmol/l) Triglyceride (mmol/l) Creatinine (umol/l) *

African

White

MA ( ) (n = 40)

MA (±) (n = 28)

p

MA ( ) (n = 101)

MA (±)(n = 33)

p

52.5 35.4 (7.9) 27.0 (6.5) 17.5 8.5 (4.0, 11.8) 25.6 (4.1) 27.5 115.6 (13.3) 74.9 (8.3) 30.8 9.55 (1.98) 4.23 (0.99) 2.40 (0.76) 1.26* (0.37) 0.90 (0.7,1.3) 86.0 (78, 95)

60.7 34.1 (9.6) 25.0* (8.3) 32.1* 7.5* (4.0, 9.8) 23.1 (3.0) 42.9 124.6 (14.6) 79.9 (10.5) 32.0 12.14* (2.46) 4.76* (1.03) 2.64 (0.91) 1.39 (0.45) 0.90 (0.7,1.2) 90.5 (79,98)

0.203 0.527 0.286 0.161 0.760 0.0126 0.301 0.0109 0.0317 0.601 <0.0001 0.0392 0.341 0.251 0.704 0.248

70.3 35.1 (10.5) 22.1 (7.5) 47.5 11.0 (6, 17) 24.1 (4.0) 20.8 119.1 (12.3) 75.2 (7.7) 45.7 8.83 (2.22) 5.26 (1.09) 3.30 (1.03) 1.42 (0.44) 1.10 (0.9, 1.8) 90.0 (81,99)

39.4 33.1 (12.3) 16.3* (7.2) 75.7* 12.0* (7, 23) 23.9 (6.0) 45.5 128.3 (20.7) 76.4 (10.4) 34.5 9.64* (2.96) 5.66* (1.36) 3.51* (1.03) 1.50 (0.38) 1.25 (0.9,2.0) 88.0 (82,104)

0.0014 0.368 0.0002 0.0047 0.190 0.914 0.0055 0.0194 0.5499 0.284 0.154 0.179 0.450 0.518 0.916 0.939

Significant differences between the African and white patients with MA (p < 0.05).

(p = 0.0184) and African race (p = 0.0287) (R2 = 0.3145); weighting for the duration of diabetes increased the significance of the associations (R2 = 0.3717). Regression analysis was carried out separately for each group. In the African group, significant independent associations were found between the presence of MA and higher HbA1c and (p < 0.0001) SBP (p = 0.0195) and prior hypertension (p = 0.0348) and lower BMI (p = 0.0022), (R2 = 0.4617). In the white patients, there were significant independent associations between MA and female gender (p = 0.0002), younger age at diagnosis (p < 0.0001), higher systolic BP (p = 0.0144) and hypertension (p = 0.0477) (R2 = 0.2925). Comparisons between subjects with MA revealed group differences in gender, age at diagnosis, duration of diabetes at the onset of MA (medians, African 6.0 years, white subjects 12.0 years) and HbA1c and total cholesterol concentrations. In order to evaluate possible differences in susceptibility to renal damage between the races from very severe long term hyperglycaemia, we compared the prevalence of MA in the total African group and in the subset of white patients whose HbA1c was greater than the group median (>8.73%; n = 68; mean HbA1c, 10.89(1.89)%, vs 10.58(2.55)% in the African group, p = 0.4288). As in the total white group, in comparison with the Africans, this subgroup was characterised by significantly younger age at diagnosis and longer median duration of diabetes – 11.0 vs 8.0 years in the Africans (p = 0.0009). The prevalence of MA was 30.9%, vs 40.3% in the Africans (p = 0.253). Among patients with MA, the duration of diabetes was greater in the white subgroup – 12.0 years, vs 7.0 years in the Africans (p = 0.0007). All other variables were similar in these MA-positive groups, except, again, the younger age at the onset in the white subjects. Multiple regression analysis, with the presence of MA as the dependent variable in the combined white subgroup and Africans revealed significant associations between the development of MA and HbA1c (p < 0.0001), SBP (p = 0.0003) young age at diagnosis of diabetes (p = 0.0010) and African race (p = 0.0020). Since the development of diabetic microvascular complications is a function of the duration of diabetes and average long-term glycaemic control (8, 9), the total ‘glycaemic exposure’ prior to the onset of MA was calculated as [mean HbA1c x duration of diabetes] prior to the onset of MA: in the white subgroup, 140(95, 197) HbA1c.years and in the Africans 82(51, 132) HbA1c.years (p = 0.0126). Forty-eight African subjects had sufficient longitudinal data suitable for the analysis of potential risk factors for MA (duration

8.0(4.0, 11.3) years). Their characteristics were similar to those of the larger African group and 70% had documented acute onset of diabetes. Anti-GAD-65 antibodies were detected, some years after diagnosis, in 11 of the 21(59%) who were tested; the insulin dose at the end of follow-up was 0.72(0.63, 1.00) U/kg. In this cohort 21 subjects (45%) developed persistent micro-albuminuria (excluding two subjects who had clearly documented transient micro-albuminuria): prior to the onset of MA, they were characterised in comparison to those who remained free of persistent MA, by higher mean levels of HbA1c (12.41(2.06)% vs 9.62(2.25)%, p = 0.0001) and mean total cholesterol (4.77(0.97) vs 3.86(0.61)mmol/l, p = 0.0008) and lower BMI (22.6(2.8) vs 26.4(3.9), p = 0.0340) and they were more likely to have developed diabetes during adolescence (33.3% vs 8.0%, p = 0.0310); all other variables measured, including blood pressures, duration of DM and renal function were similar. Logistic regression analysis, with the development of MA as the dependent variable, and all parameters significantly different in the univariate analyses entered as independent variables, demonstrated that the independent predictors of MA in these African patients were higher levels of mean HbA1c (p = 0.007, odds ratio (OR, 95% CI) 2.98 (1.35–6.60)), higher mean SBP (p = 0.038; odds ratio 1.16(1.01–1.33), and lower BMI (p = 0.012; OR 0.59(0.39– 0.89)). Further follow-up data on urinary ACR were available in 16 patients who had developed persistent MA. Micro-albuminuria progressed to macro-albuminuria (ACR > 30 mg/mmol) in nine subjects with median duration of MA of 3.0 years (range 1–5 years) and total duration of diabetes of 8.0 years (range 4–16 years); in the remaining seven patients MA persisted for a median of 3.0 years (range 1–6 years) with total duration of diabetes of 10.0 years (range 4–24 years). The mean HbA1c level in those who progressed to macro-albuminuria was 13.49(2.00)%, and 11.38(1.62)% in subjects with persistent MA during follow-up (p = 0.040); other parameters were similar.

4. Discussion The clinical variables in the African and white patient groups were generally well matched – for age, BMI, prevalence of hypertension, BP and renal function, with an unexpected male predominance in both groups. A substantial proportion of the white patients attending our clinic who had an age of the onset of diabetes <10 years were not included in the analyses so as to reduce the possible bias of the longer duration of pre-adolescent diabetes

W.J. Kalk et al. / International Journal of Diabetes Mellitus 2 (2010) 148–153

[27,28]. Despite these omissions, the age at diabetes diagnosis was much younger in the white group, reflecting the low rate of adolescent onset in the Africans [3]. The major additional group differences were the strikingly worse glycaemic control in the African patients and their lower lipid levels, anticipated from the general African population [29]. In the white group, the mean HbA1c was similar to those reported from several European and North American studies [7,11,15,18], while in the Africans, it was much higher, although comparable to levels in some European centres [30]. The prevalence of MA was significantly greater in the African (40%) than in the white patients (25%), despite a 50% shorter duration of diabetes prior to the onset of MA in the former, raising the possibility of a greater predisposition among this population. Indeed, African race was independently associated with the development of MA. Other risk factors, such as poor glycaemic control, blood pressure, female gender and younger age at diagnosis are common to previous reports. Micro-albuminuria was also more prevalent than recently reported from Tanzania but the duration of diabetes was shorter in that study [21]. When the two race groups were evaluated separately, some similarities and differences in risk factors were observed. In both groups, MA was associated with higher blood pressures and worse glycaemic control and in the Africans also with lower BMI and higher total cholesterol concentrations, which were, however, lower than in the MA-positive white patients. In the white group MA was also associated with younger age at diagnosis and female gender [15,16] but not with cholesterol or triglyceride concentrations. In neither group was duration of diabetes associated with MA, probably because of the short duration of follow-up. So as to offset the strikingly worse long-term glycaemic control in the Africans, we repeated the group comparisons after excluding the best controlled white patients – those with mean HbA1c less than the median (68.73%). Mean HbA1c levels in this very poorly controlled white subgroup were close to those in the Africans. The prevalence of MA remained higher in the African group (40% vs 31% in the white subgroup); however, the duration of diabetes was substantially shorter in the African subjects. Thus, the estimated total exposure to similar severe hyperglycaemia was significantly lower in the African patients for a comparable prevalence of MA, suggesting greater renal susceptibility to hyperglycaemia. Regression analysis supported this possibility, in that among these patients with similar degrees of hyperglycaemia, the African race remained a (more) significant independent predictor of MA; other significant associations are well known. The incidence of MA in the smaller longitudinally studied African cohort was similarly high, at 45%, (median duration 8 years), comparable to a 6-year development of any proteinuria recently reported in African Americans with TIDM [19]. In keeping with the cross-sectional analysis, independent predictors for the development of MA had a higher mean HbA1c and mean SPB and lower BMI. Racial differences in the apparent susceptibility to diabetic nephropathy have been noted before. African-Americans with type 1 diabetes exhibited earlier onset and a greater age-specific incidence of end stage renal disease than white patients [31]; similar patterns were noted in patients with type 2 diabetes. Recent analyses, mainly in type 2 diabetes, have identified genes which are closely associated with diabetic nephropathy, some of which may predispose to, and others which may protect against this complication [32,33]. While some genomic regions underlying nephropathy susceptibility are common to the several populations studied, others appear to be more race-specific, notably between African and European Americans [32]. Moreover, genes which are ‘protective’ in European Americans may not exert this effect in African Americans [33]. Studies in white patients, however, suggest that the genetic basis for nephropathy may be different in type1 and type 2

151

diabetes [34]; inter-ethnic genetic studies are awaited. Apart from African-derived populations, others with type 1 diabetes may also be at increased risk for nephropathy in comparison to patients of European extraction [35]. In type 1 diabetes, poor glycaemic control has been observed to be a consistent indicator of risk for MA, proportional to the average degree of hyperglycaemia [8,30]. Among our African subjects mean HbA1c was also the strongest independent risk factor for progression to MA; indeed, 73% had a mean HbA1c above 9.0%, a level strongly associated with the development of early diabetic renal disease in young white patients [36]. This high incidence of MA was comparable to the 10 year incidence among the worst controlled (HbA1c > 8.6%) Danish patients after some 10 years of diabetes [11]. Although mean blood pressures in the African patients were mainly normal, a higher systolic BP predicted the onset of MA [9,17]. It has been postulated that among white diabetic subjects, slightly elevated blood pressures may interact with hyperglycaemia [37] and perhaps a genetic predisposition [20], to cause glomerular injury. Furthermore, there may be patient subgroups with abnormal renal vulnerability to small increments in blood pressure [38]. In urban African populations, in comparison with those of European extraction, hypertension occurs more frequently, starts earlier and appears to have greater harmful effects on the kidneys [39], possibly mediated via differences in renal sodium handling and lower renin concentrations [40]. Thus, it is plausible that there may be racial differences in renal susceptibility to small elevations in blood pressure in the presence of hyperglycaemia or to hyperglycaemia per se. In type 1 diabetic African Americans, systolic BP was also a risk factor for progression to proteinuria [19]. In some European cohorts, however, a rise in blood pressure has been found only some years after the onset of MA [41], a pattern observed in an early analysis from our unit in young white patients [42]. Dyslipidaemias have been cited as potential risk factors for MA in Type 1 diabetes [10,14,17,18]. In our African patients, total and LDL-cholesterol and triglyceride concentrations were relatively low, similar to those in the non diabetic African population and lower than in the local white population [29]. Although MA-positive African patients had higher levels, cholesterol was not independently associated with MA – concentrations that correlated significantly with HbA1c (r = 0.4595, p = 0.0010). Furthermore, neither triglycerides nor HDL cholesterol concentrations were associated with MA in either group. Of note is the observation that the relatively low lipid levels in the Africans did not appear to protect against MA. Increased BMI or waist circumference and elevated triglycerides are features of metabolic syndrome, which are said to be risk factors for nephropathy [18]. Paradoxically, in our African group, lower BMI appeared to be an independent predictor of MA. While it is unlikely that relative leanness is itself a risk factor, it may correlate with some other unmeasured factors, possibly social or economic, in this deprived population [43] or with low adherence to prescribed insulin doses. In our African patients, neither the intensity of insulin therapy nor total prescribed daily insulin doses influenced glycaemic control or the development of MA. Significant hyperglycaemia appears to promote the progression of MA to macro-albuminuria [7,11]. In the African subjects followed after the onset of persistent MA, macro-albuminuria developed after 1–5 years in more than half, despite documented therapy with angiotensin converting enzyme inhibitors. These patients were characterised by extremely poor average glycemic control (mean HbA1c, 13.49%). Rapid progression from normo- to macro-albuminuria has also been documented in type 1 diabetic African Americans [19]. A limitation of this study is the relatively small number of African patients studied longitudinally and some gaps in the data col-

152

W.J. Kalk et al. / International Journal of Diabetes Mellitus 2 (2010) 148–153

lection. However, type 1 diabetes remains a relatively rare condition in sub-Saharan Africa and prospective data collection is difficult in this mobile population in South Africa [5,23], often characterised by erratic clinic attendance. Gaps in data collection might have lead to misclassification in some subjects, especially those with transient MA, which appeared to be infrequent, and over-estimation of the real prevalence and incidence of MA. The use of a higher ACR cut point to define MA should have reduced this problem. It is also possible that our data may have been biased as a consequence of early mortality from acute metabolic complications [2,5,23,24]. Lastly, our patients may not be representative of all type 1 diabetic patients in the country, as our hospital’s diabetes service was relatively well endowed in comparison with smaller centres [44,45]. On the other hand, the potential strengths of our analyses lie in the use of multiple measurements obtained at patients’ annual reviews, prior to the onset of MA. Furthermore, the associations we observed with the development of MA were statistically strong and were mainly similar to previous reports. Notably, severe chronic hyperglycaemia was a feature of our African patients, despite routine care which included skilled diabetes nurse specialists/educators fluent in the indigenous African languages. Comparable levels of hyperglycaemia and a high prevalence of early diabetic complications have been noted recently in other transitional peoples, such as Maori and Pacific Islanders [46] and in young adult African Americans [47]. Detailed information on the type of insulin therapy was available in 37 patients: 22 received twice daily biphasic injections and 15 intensive basal bolus treatments. Intensity of therapy did not influence the risk of developing MA (intensive therapy in eight of 17 with MA; seven of 20 without MA); the daily insulin doses prescribed were also similar in each group (MA-positive, 0.72 U/kg; MA-negative, 73 U/kg). Influences on glycaemic control, such as poverty, depression and psychosocial issues [19,46,48,49] may be more prevalent in transitional communities. Thus for Africans with type 1 diabetes, there is undoubtedly a special need for the provision of holistic diabetes services aimed at reducing their high rates hyperglycaemia, complications and premature death [2,5]. There is accumulating evidence that improved routine services can lead to better outcomes in type 1 diabetic population [50]. In conclusion, we have found a high incidence of micro-albuminuria among Africans with type 1 diabetes predicted primarily by their extremely poor glycaemic control but possibly also related to an increased susceptibility to diabetic glomerular injury. Recognition of their high incidence of diabetic renal disease, its early onset and potential for rapid progression should improve their prognosis. Duality of interest The authors declare that there is no duality of interest associated with this manuscript. Acknowledgements The study was supported by a grant from the South African Medical Research Council. We thank sisters Alice Tshabalala and Sheila Nzelu for their assistance. References [1] Swai ABM, Lutale JL, McLarty DG. Prospective study of incidence of juvenile diabetes mellitus over 10 years in Dar es Salaam, Tanzania. Br Med J 1993;306:1570–2. [2] 0sei K, Schuster DP, Amoah AG, Owusu SK. Diabetes in Africa. Pathogenesis of type 1 and type 2 diabetes mellitus in Sub-Saharan Africa: implications for transitional populations. J Cardiovasc Risk 2003;10:85–96.

[3] Kalk WJ, Huddle KRL, Raal FJ. The age of onset and sex distribution of insulindependent diabetes mellitus in Africans in South Africa. Postgrad Med J 1993;69:552–6. [4] Gill GV, Huddle KR, Rolfe M. Mortality and outcome of insulin-dependent diabetes in Soweto, South Africa. Diabet Med 1995;12:546–50. [5] Gill GV, Huddle KRL, Monkoe G. Long-term (20 years) outcome and mortality of type 1 diabetic patients in Soweto, South Africa. Diabet Med 2005;22:1642–6. [6] Majaliwa ES, Sanyiwa A, Munubhi E, Mohn A, Ramaiya K, Charelli F, et al. Survey on acute and chronic complications in children and adolescents with type 1 diabetes at Muhimbili National Hospital, Tanzania. Diabetes Care 2007;30:2187–92. [7] Giorgino F, Laviola L, Cavallo Perin P, Solnica B, Fuller J, Chaturvedi N. Factors associated with progression to macroalbuminuria in microalbuminuric type 1 diabetic patients: the EURODIAB Prospective Complications Study. Diabetologia 2004;47:1020–8. [8] The Diabetes C. Ontrol and Complications Study Group: the effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86. [9] Warram JH, Gearin G, Laffel L, Krowelski AS. Effect of duration of type 1 diabetes on the prevalence of stages of diabetic nephropathy defined by urinary albumin/creatinine ratio. J Amer Soc Nephrol 1996;7:930–7. [10] Chaturvedi N, Bandinelli S, Mangili R, Penno G, Rottiers RE, Fuller JH, et al. Microalbuminuria in type 1 diabetes: rates, risk factors and glycemic threshold. Kidney Int 2001;60:219–27. [11] Rossing P, Hougaard P, Parving H-H. Risk factors for development of incipient and overt diabetic nephropathy in type 1 diabetic patients. Diabetes Care 2002;25:859–64. [12] Thorn LM, Forsblom C, Fagerudd J, Thomas MC, Pettersson-Fernholm K, Saraheimo M, et al. Groop P-H on behalf of the FinnDiane study Group: Metabolic syndrome in type 1 diabetes. Association with diabetic nephropathy and glycaemic control (the FinnDiane Study). Diabetes Care 2005;28:2019–24. [13] Thomas MC, Forsblom C, Rosengard-Barlund M, Cooper ME, Mills V, Taskinen M-R, et al. Serum lipids and the progression of nephropathy in type 1 diabetes. Diabetes Care 2006;29:317–22. [14] Stone ML, Lee JW, Craig ME, Verge CF, Chan AK, Donaghue KC. Natural history and risk factors for microalbuminuria in adolescents with type 1 diabetes. Diabetes Care 2006;29:2072–7. [15] Gallego PH, Bulsara MK, Frazer F, Lafferty AR, Davis EA, Jones TW. Prevalence and risk factors for microalbuminuria in a population-based sample of children and adolescents with T1DM in Western Australia. Paediatr Diabetes 2006;7:165–72. [16] Sibley SD, Boer IH, Steffes M, Brunzell JD, The Diabetes control and Complications Trial/Epidemiology of Diabetes Intervention and Complications (DCCT/EDIC) research Study Group. Intra-abdominal fat and elevated urine albumin excretion in men with type 1 diabetes. Diabetes Care 2007;30:1898–900. [17] Raile K, Galler A, Hofer S, Herbst A, Dunstheimer D, Busch P, et al. Diabetic nephropathy in 27, 805 children, adolescents, and adults with type 1 diabetes. Diabetes Care 2007;30:2523–8. [18] Kilpatrick ES, Rigby AS, Atkin SL. Insulin resistance, the metabolic syndrome, and complication risk in type 1 diabetes. Diabetes Care 2007;30:707–12. [19] Roy MS, Affouf M, Roy A. Six-year incidence of proteinuria in type 1 diabetic African Americans. Diabetes Care 2007;30:1807–12. [20] Al-Kateb H, Boright AP, Mirea L, Xie X, Sutradhar R, Mowjoodi A, et al. Multiple superoxide dismutase 1/splicing factor serine alanine 15 variants are associated with the development and progression of diabetic nephropathy. The diabetes control and complications trial/epidemiology of diabetes interventions and complications genetic study. Diabetes 2008;57: 218–28. [21] Lutale JJK, Thordonson H, Abbas Gulam, Vetvik K. Microalbuminuria among type 1 and type 2 diabetic patients of African origin in Dar es Salaam, Tanzania. BMC Nephrol 2007;8:1–11 (accessed 18 July 2009). [22] Gill GV, Mbanya J-C, Ramaiya KL, Tesfaye S. A sub-Saharan African perspective of diabetes. Diabetologia 2009;52:8–16. [23] The Expert Committee on the Diagnosis and Classification of Diabetes mellitus. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 1997;20:1183–97. [24] Zouvannis M, Piertese AC, Seftel HC, Joffe BI. Clinical characteristics and outcome of hyperglycaemic emergencies in Johannesburg Africans. Diabet Med 1997;14:603–6. [25] Panz V, Kalk WJ, Zouvanis M, Joffe BI. Distribution of autoantibodies to glutamic acis decarboxylase across the spectrum of diabetes mellitus seen in South Africa. Diabet Med 2000;17:524–7. [26] Kalk WJ, Joffe BI. The metabolic syndrome, insulin resistance and its surrogates in African and white subjects with type 2 diabetes in South Africa. Metab Syndr Relat Disord 2008;6:247–55. [27] Kalk WJ, Osler C, Taylor D, Panz V. Prior long-term glycaemic control and insulin therapy in insulin-dependent diabetes in adolescents with microalbuminuria. Diab Res Clin Pract 1990;9:83–8. [28] Schultz CJ, Knonpelska-Bahu T, Dalton RN, Carroll TA, Sratton I, Gale EAM, et al. Microalbuminuria prevalence varies with age, sex, and puberty in children with type 1 daibetes followed from diagnosis in a longitudinal study. Diabetes Care 1999;22:495–502.

W.J. Kalk et al. / International Journal of Diabetes Mellitus 2 (2010) 148–153 [29] McQueen MJ, Hawken S, Wang X, Ounpuu S, Snidermab J, Steyn K, et al. For the INTERHEART study investigators: Lipids, lipoproteins, and apolipoproteins as risk markers of myocardial infarction in 52 countries (the INTERHEART study): a case-control study. Lancet 2008;372:224–33. [30] Stephenson L, Fuller JH. On behalf of the EURODIAB IDDM Complications Study Group: Microvascular and acute complications in IDDM patients: the EURODIAB IDDM Complications Study. Diabetologia 1994;37:278–85. [31] Cowie CC, Port FK, Wolfe RA, Savage PJ, Moll PP, Hawthorne VM. Disparities in incidence of diabetic ens-stage renal disease according to race and type of diabetes. N Engl J Med 1989;321:1074–9. [32] Iyengar SK, Abboud HE, Goddard KAB, Saad MF, Adler SG, Arar NH, et al. Genome-wide scans for diabetic nephropathy and albuminuria in multiethnic populations. Diabetes 2007;56:1577–85. [33] McDonough CW, Hicks PJ, Lu L, Langefeld CD, Freedman BI, Bowden DW. The influence of carnosinase gene polymorphism on diabetic nephropathy risk in African Americans. Hum Genet 2009;126:265–75. [34] Pezzolesi MG, Poznik GD, Mychaleckyj LC, Paterson AD, Barati MT, Klein JB, et al. Genome-wide association scan for diabetic nephropathy susceptibility in type 1 diabetes. Diabetes 2009;58:1403–10. [35] Joshy G, Dunn P, Fisher M, Lawrenson R. Ethnic differences in the natural history among diabetic patients in New Zealand: hospital admission rate for renal complications and incidence of end-stage renal disease and renal death. Diabetologia 2009;52:1474–8. [36] Dunger DB, Schwarze CP, Cooper JD, Widmer B, Neil HAW, Shield J, et al. Can we identify adolescents at high risk for nephropathy before the development of microalbuminuria? Diabet Med 2007;24:131–6. [37] The Microalbuminuria Collaborative Study Group. Predictors of the development of microalbuminuria in patients with Type 1 diabetes mellitus: a seven year prospective study. Diabet Med 1999;16:918–25. [38] Stephenson JM, Fuller JH, Viberti G-C, Sjolie A-K, Navelesi R. The EURODIAB IDDM Complications Study Group: Blood pressure, retinopathy and urinary albumin excretion in IDDM: the EURODIAB IDDM Complications Study. Diabetologia 1995;38:599–603. [39] Seedat YK. Hypertension in black South Africans. J Hum Hypertens 1995;13:96–111.

153

[40] Opie LH, Seedat YK. Hypertension in sub-Saharan African populations. Circulation 2005;112:3562–8. [41] Schultz CJ, Neil HAW, Dalton RN, Konopelska Bahu T. Dunger DB on behalf of the Oxford Regional Prospective Study Group: blood pressure does not rise before the onset of microalbuminuria in children followed from diagnosis of type 1 diabetes. Diabetes Care 2001;24:555–60. [42] Raal FJ, Kalk WJ, Taylor D, Osler CE, Panz VR. The relationship between the development and progression of microalbuminuria and arterial blood pressure in type 1 (insulin-dependent) diabetes mellitus. Diabetes Res Clin Pract 1992;16:221–7. [43] Carter PJ, Cutfield WS, Hofman PL, Gunn AJ, Wison DA, Reed PW, et al. Ethnicity and social deprivation independently influence metabolic control in children with type 1 diabetes. Diabetologia 2008;51:1835–42. [44] Beattie A, Kalk WJ, Price M, Rispel L, Broomberg J, Cabral J. The management of diabetes at primary level in South Africa: the results of a facility-based assessment. J Royal Soc Health 1998;118:338–45. [45] Kalk WJ, Veriawa Y, Osler C. A survey of hospital outpatient services for chronic diseases in Gauteng. S Afr Med J 2000;90:57–61. [46] Scott A, Toomath R, Bouchier D, Bruce R, Crook N, Carroll D, et al. First national audit of the outcome of care in young people with diabetes in New Zealand: high prevalence of nephropathy in Maori and Pacific Islanders. N Zealand Med J 2006;199:1–12. [47] Bosnyak Z, Nishimura R, Hagan Hughes M, Tajima N, Becker TJ. Excess mortality in black compares with white patients with type 1 diabetes: an examination of underlying causes. Diabet Med 2005;22:1636–41. [48] Bryden KS, Peveler RC, Stein A, Neil A, Mayou RA, Dunger DB. Clinical and psychological course of diabetes from adolescence to young adulthood. Diabetes Care 2001;24:1536–40. [49] Chida Y, Hamer M. An association of adverse psychosocial factors with diabetes mellitus: a meta-analytic review of longitudinal cohort studies. Diabetologia 2008;51:2168–78. [50] Hovind P, Tarnow L, Rossing K, Rossing P, Larsen N, Binder C, et al. Decreasing incidence of diabetic microangiopathy in type 1 diabetes. Diabetes Care 2003;26:1258–64.

International Journal of Diabetes Mellitus 2 (2010) 154–157

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

Association between socio-demographic factors and diabetes mellitus in the north of Iran: A population-based study Gholamreza Veghari ⇑, Mehdi Sedaghat, Hamidreza Joshaghani, Sed Ahmad Hoseini, Farhad Niknezad, Abdolhamid Angizeh, Ebrahim Tazik, Pooneh Moharloei Golestan University of Medical Sciences, Iran

a r t i c l e

i n f o

Article history: Received 11 May 2010 Accepted 1 September 2010

Keywords: FBS Socio-demographic BMI Iran

a b s t r a c t Objective: This study considers the prevalence of DM and some related factors among adults in the Golestan province (north of Iran) in 2006. Methods: This is a Crossectional–Descriptive and population-based study, carried out among 1999 cases (1000 men and 999 women) between 25 and 65 years old. Participants were chosen by cluster and stratified sampling in urban and rural areas. Data on socio-demographic factors were collected using questionnaire, and anthropometric and biochemical indexes were measured. Fasting Blood Sugar (FBS) equal to or over 126 mg/dl was classified as type 2 DM. Results: Mean of age was 39.2 years and mean ± SD of FBS among men and women was 94.51 ± 32.91 and 98.2 ± 40.1 mg/dl, respectively. Prevalence of DM was 8.3% [(men = 6.8% and women = 9.7%), (urban = 10.5% and villages = 6.4%)]. Twenty-five percent of patients were undiagnosed as whole, 43% of patients were unaware of their problem, in men more than women (48.5% versus 39.2%) and in rural area more than in urban area (35.1% versus 54.4%). We showed a positive and significant correlation between FBS and age, waist circumference and BMI (P = 0.01). Conclusion: DM was the one of the biggest health problems in the north of Iran, and half of them were unaware of their morbidity. DM was influenced by socio-demographic factors. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction The number of people suffering from DM is increasing due to population growth, aging, urbanization, low physical activity and the high prevalence of obesity [1,2]. Quantifying the prevalence of DM and the number of people affected by diabetes, now and in the future, is important in permitting national planning and allocation of resources. DM and its complications are a major cause of morbidity and mortality in developing countries. Successful management of DM requires that we understand the beliefs, lifestyles, attitudes, family and social networks of the patients being treated [3]. Veghari [4] announced that diabetes is one of the health problem in north of Iran and that patients do not have an effective knowledge about their diet and blood glucose controlling methods. Hadaegh [5] in Iran [5] reported that DM is a health problem, and most of patients were unaware of their problem. The studies of Azimi-Nezhad [6] and Maddah [7] in Iran showed that DM was related to socioeconomic factors. Janghorbani [8]

⇑ Corresponding author. Address: Golestan Cardiovascular Research Center and Dept. of Biochemistry and Nutrition, School of Medicine, Gorgan, Iran. E-mail address: [email protected] (G. Veghari).

reported that metabolic syndrome was common among 65% of DM in Iran. Of the 1,600,000 population in the north of Iran, 66.39% are 15– 64 years old, whereas 43.9% and 56.1% are living in urban and rural areas, respectively [9]. Agriculture is the main job in rural areas. Different ethnic groups, such as Fars(native), Turkman and Sistani, are living in this region. Due to the restriction in executing epidemiological projects, there has been no study of the DM in this area up till now; therefore it was necessary to design a research project to address this. The aims of this study are to determine the prevalence of DM and some socio-demographic factors such as sex, age, BMI, central obesity, residential area, physical activity, economic status and level of education in the north of Iran in 2006. 2. Material and methods This population based cross sectional descriptive study was carried out in 1999 adults (1000 males and 999 females) between 25 and 65 years old. Participants were chosen by cluster and stratified sampling in urban and rural areas. According to American Diabetes Association (ADA) criteria, Fasting Blood Sugar (FBS) equal to or more than 126 mg/dl was diagnosed as type 2 DM [10].

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.09.001

155

G. Veghari et al. / International Journal of Diabetes Mellitus 2 (2010) 154–157

FBS were determined using laboratory kits (enzymatic methods) and spectrophotometry technique. Data on socio-demographic factors and physical activity were collected using questionnaires. Anthropometric and biochemical indexes were examined. Body Mass Index (BMI) was calculated by dividing weight (kg) to height (m2). Those with a BMI of 25.0–29.9 kg/m2 were classified as overweight, while those with a BMI P 30.0 kg/ m2 were classified as obese and BMI P 40 classified as pathologically obese [11]. Central obesity was defined based on waist circumference (men P 102 and women P 88 cm) [12]. Weight measurement without shoes and clothing was carried out using a balance, and recorded to the nearest 0.5 kg. Height and waist were measured to the nearest 0.5 cm, while the participants were standing on their feet. Waist circumference was measured using a tape measure over the iliac and lower border of the ribs. Economic status, with regard to Iranian social-economic was categorized as based on home ownership, number of the rooms in the house, owning of a private car, structure of the house and the number of the family members. According to this list, the scoring of economic status of the sample population in this study was as follows: low P 1, moderate = 2–3, and good P 4. Physical activity was categorized as based on activity during daily work, in five categories: (1) no physical activities (without moving from one place to another place); (2) low activity (physical activity involving the extension of muscular-skeletal and moving from one place to another place); (3) moderate activity (physical activity sometimes involving an increase in respiratory rate like cleanliness, gardening, building painter,. . .); (4) high activity (physical activity involving a highly increased reparatory rate such as manual labor, building labor, porter,. . .) and (5) all of the above activities during the day. Educational level categories were based on three levels: illiterate, 0–12 years schooling and college. Data were analyzed using SPSS version 16.0 and statistical significance was defined as a p-value of <0.05. Analysis of variance (ANOVA) and chi-2 tests were used to compare group means and frequencies, respectively.

3. Results Mean and standard deviation of FBS are present in Table 1, and prevalence of DM was shown in Table 2. Mean and standard deviation of FBS was 96.40 ± 36.38 mg/dl and the prevalence of DM was 8.3%. Nearly 25% of total cases of diabetes were undiagnosed. The portion of type 2 DM is 95.4%. There was a positive and significant correlation between age and blood glucose (P < 0.05). The mean of blood glucose was shown to be more significant in women more than men (P < 0.05). The prevalence of DM in women was 3.1% more than in men, and in 55–65 years olds, was five times more than the 25–35 years old group. Statistically significant differences were shown among four age groups, based on the mean of FBS (P < 0.001). The mean of blood glucose among central obese people was 10.1 mg/dl more than in normal people, and the prevalence of DM was also 7.3% more. The prevalence of DM in urban areas was more than rural (10.4% versus 6.4%), and statistical differences were significant (P = 0.003). Meanwhile, there was a direct relationship between FBS level and economic status, but statistical differences were not significant between the three economic groups. Physical activity has a marked effect on FBS level, and the prevalence of DM in the low physical activity group was two times as much as in the high physical activity group, and the statistical differences were significant (P = 0.04). There was a positive correlation between BMI and FBS level, and it was elevated to 1.77 mg/

Table 1 FBS and socio-demographic factors among adult people in the north of Iran. Characteristics

N

Mean(SD) mg/dl

ANOVA test P-value

Sex Male Female

1000 999

94.56(32.2) 98.25(40.10)

0.023

Age group(y)* 25–35 35–45 45–55 55–65

548 538 489 419

87.74(27.01) 94.36(33.59) 103.33(44.18) 102.39(38.04)

0.001

Central obesity No Yes

1113 850

92.05(29.67) 102.65(44.84)

0.001

Residential area Urban Village

931 1067

99.79(40.18) 93.45(32.45)

0.001

Economic status** Low Moderate Good

211 1717 70

93.83(38.30) 96.68(36.29) 97.30(32.70)

0.549

Physical activity*** Low Moderate High Whole

508 883 83 96

99.34(40.67) 93.34(30.90) 88.39(18.17) 96.75(36.48)

0.004

BMI* <18.5 18.5–24.9 25–29.9 30–34.9 35–39.9

58 665 663 538 42

87.69(16.62) 91.12(31.67) 95.94(33.72) 103.35(44.53) 113.67(52.19)

0.001

Educated level Illiterate 0–12 year schooling College

731 1147 120

98.90(39.21) 95.34(35.41) 91.41(25.24)

0.04

Overall

1999

94.40(36.38)

SD: standard deviation. FBS equal to or more than 126 mg/dL defines hyperglycemia and diabetes mellitus. * The mean of FBS has a positive and significant correlation with age and BMI. ** Although a positive correlation has shown between economic status and FBS, statistical differences is not significant. *** There is a negative and significant correlation between FBS and physical activity.

dl per 1 kg/m2 of BMI increasing, as whole. The prevalence of DM among pathologic obese people was five times more than in normal people. Statistical differences were significant (P = 0.001). There was a statistical significant difference among the three educational levels and DM was significantly observed in illiterate people, more so than in other educated groups (P = 0.004). The prevalence of DM among central obese people was significantly more than in normal people (P = 0.001). 4. Discussion The results of this study were discussed from two aspects, prevalence and certain factors related to DM. In the present study, the prevalence of DM was 8.3%, and 25% of them were undiagnosed. The prevalence of diabetes was estimated to be 10% and 8.1% in women and men in Theran (center of Iran), respectively; based on this study, 40% of patients were undiagnosed [5]. Another study [6] reported that the prevalence of DM in Iran was 5.5%. The proportion of undiagnosed diabetes in China population was 70.5% and 58% in rural and urban areas, respectively [13]. In comparison with other studies, the undiagnosed DM rate in the north of Iran is appropriate. King and et al. [2] were estimated the prevalence of DM in Iran up to 5.5% in 1995, 6.8% in 2000 and 6.8% in 2025. The prevalence

156

G. Veghari et al. / International Journal of Diabetes Mellitus 2 (2010) 154–157

Table 2 Relationship between hyperglycemia and socio-demographic factors among adult people in the north of Iran. Characteristics

N

Hyperglycemia N (%)

Chi 2 test P-value

Sex Male Female

1000 999

68(6.8) 97(9.7)

0.029

Age group(y)* 25–35 35–45 45–55 55–65

548 538 489 419

14(2.8) 26(5.2) 53(10.6) 72(14.5)

0.001

1113 850

58(5.2) 106(12.5)

0.001

Residential area Urban Village

931 1067

97(10.4) 68(6.4)

0.003

Economic status Low Moderate Good

211 1717 70

14(6.6) 146(8.5) 5(7.1)

0.665

508 883 83 96

51(10.0) 55(6.2) 4(4.8) 7(7.3)

0.038

BMI* <18.5 18.5–24.9 25–29.9 30–34.9 35–39.9

58 665 663 538 42

2(3.4) 32(4.8) 51(7.7) 69(12.8) 8(19.0)

0.001

Educated leve*l Illiterate 0–12 year schooling College

731 1147 120

72(9.8) 89(7.8) 4(3.3)

0.056

Overall

1999

16.5(8.3)

Central obesity No Yes

Acknowledgements

*

Physical activity Low Moderate High Whole

*

quantity and quality of diet, duration of diabetes morbidity and ethnic differences in this area. These are the limitations of this study. Briefly, our study has shown that DM was a health problem in the north of Iran, and one to four of diabetic patients remained undiagnosed. Socio-economic status, obesity and low physical activity are predisposing factors for DM morbidity. A screening and intervention program for preventing of DM in the north of Iran is necessary.

This paper was created from a provincial incommunicable study, and is based on 258888 official document was justified fro publication. The authors wish to thank the medical and administrative staff in the Primary Health Care Centers of Golestan University of Medical Sciences for their valuable assistance during the field work. References

*

Statistical differences are significant.

of DM in our study was approximately equal to other studies. We showed the prevalence of DM in women 3.1% more than men. Another study [2] have similar result. In religions countries like Iran, because of family priorities women can not exercise in public places. Like the other study [2], we have DM prevalence in urban more than rural residents. The prevalence of DM increased with age, low physical activity, central obesity and BMI. Some studies [14–16] showed a higher DM rate in older people. Patrick [17] and Wannamethee [18] showed that the prevalence of DM in low active people is more than in highly active people. Harris in US [15], Simmons [19] O’Rahilly [20], Regzedmaa [21], Brancati [22] and Thompson [23] announced that obesity and central obesity are two risk factors for the incidence of DM. Like other studies [24,25] there was no positive correlation between economic status and educational level. Meanwhile, Weissman [26] showed that low economic status was associated with low access to health care. Limited access to health care not only influences the use of preventive services [27], but also elevates the risk of a decline in health [28]. In the present study, there was an association between illiteracy and DM which was consistent with other studies [29–31]. Moreover, limited literacy is associated with a decreased knowledge of medical conditions [32,33]. These persons have no access to knowledge about self care for the prevention or treatment of DM and other diseases [34,35]. We had no information about medical supervision status, and we did not determine all of the factors related to DM, such as

[1] Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004;27:1047–53. [2] King H, Aubert RE, Herman WH. Global burden of diabetes, 1995–2025. Prevalence, numerical estimates, and projections. Diabetes Care 1998;21:1414–31. [3] Bradley C, Gamsu DS. For the Psychological Well-being Working Group of the WHO/IDF St Vincent Declaration Action Programme for Diabetes. Measures of psychological well-being and treatment satisfaction developed from the responses of people with tablet-treated diabetes. Diabet Med 1994;7:510–6. [4] Veghari G, Marjani A, Joshaghani H. The study of diabetes mellitus in Gorgan, Iran. Saudi Med J 2007;28(8):1300–1. [5] Hadaegh F, Bozorgmanesh M, Harati H, Saadat N, Azizi F. High prevalence of diabetes and abnormal glucose tolerance in urban Iranians aged over 20 years: determining an effective screening strategy for un-diagnosed diabetes. Iranian J Endocrinol Metab 2008;9(4):384–91. [6] Azimi-Nezhad M, Ghayour-Mobarhan M, Parizadeh MR, Safarian M, Esmaeili H, Parizadeh SM, et al. Prevalence of type 2 diabetes mellitus in Iran and its relationship with gender, urbanization, education, marital status and occupation. Singapore Med J 2008;49(7):571–6. [7] Maddah M. Association of diabetes with living area in Iranian women. Int J Cardiol 2008; 20 [Epub ahead of print]. [8] Janghorbani M, Amini M. Metabolic syndrome in type 2 diabetes mellitus in Isfahan, Iran: prevalence and risk factors. Metab Syndr Relat Disord 2007;5(3):243–54. [9] Statistical Center of Iran. Population and Housing Census. Available from: http://www.sci.org.ir; 2006. [10] American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2004;27:S5–10. [11] WHO. Obesity: Preventing and managing the global epidemic. WHO/NUT/98. Geneva, Switzerland: World Health Organization; 1998. [12] Molarius A, Seidell JC, Visscher TL, Hofman A. Misclassification of high-risk older subjects using waist action levels established for young and middle-aged adults – results from the Rotterdam Study. J Am Geriatr Soc 2000;48(12):1638–45. [13] Dong Y, Gao W, Nan H, Yu H, Li F, Duan W, et al. Prevalence of type 2 diabetes in urban and rural Chinese populations in Qingdao, China. Diabet Med 2005;22(10):1427–33. [14] Harris MI, Hadden WC, Knowler WC, Bennett PH. Prevalence of diabetes and impaired glucose tolerance and plasma glucose levels in US population aged 20–74 year. Diabetes 1987;36(4):523–34. [15] Porapakkham Y, Pattaraarchachai J, Aekplakorn W. Prevalence, awareness, treatment and control of hypertension and diabetes mellitus among the elderly: the 2004 National Health Examination Survey III, Thailand. Singapore Med J 2008;49(11):868–73. [16] Bruce A. Reeder, Aubie Angel T, Marielle Ledoux, Simon W. Rabkin, Kue Young T, et al. Obesity and its relation to cardiovascular disease risk factors in Canadian adults. Can Med Assoc J 1992;146(11):2009–19. [17] Patrick W. Sullivan, Elaine H. Morrato, Vahram Ghushchyan, Holly R. Wyatt, James O. Hill. Obesity, inactivity, and the prevalence of diabetes and diabetesrelated cardiovascular comorbidities in the US, 2000–2002. Diabetes Care 2005;28(7):1599–603. [18] Wannamethee SG, Shaper AG. Weight change and duration of overweight and obesity in the incidence of type 2 diabetes. Diabetes Care 1999;22(8):1266–72. [19] Simmons D, Rush E, Crook N. Prevalence of undiagnosed diabetes, impaired glucose tolerance, and impaired fasting glucose among Maori in Te Wai o Rona: diabetes prevention strategy. NZ Med J 2009;122(1288):30–8. [20] O’Rahilly S. Non-insulin dependent diabetes mellitus: the gathering storm. BMJ 1997;314:955–9.

G. Veghari et al. / International Journal of Diabetes Mellitus 2 (2010) 154–157 [21] Regzedmaa Nyamdorj, Qing Qiao, Stefan Söderberg, Pitkäniemi Janne M, Zimmet Paul Z, Shaw Jonathan E, et al. BMI compared with central obesity indicators as a predictor of diabetes incidence in Mauritius. Obesity 2009;17(2):342–8. [22] Brancati FL, Wang NY, Mead LA, Liang KY, Klag MJ. Body weight patterns from 20 to 49 years of age and subsequent risk for diabetes mellitus: the Johns Hopkins Precursors Study. Arch Int Med 1999;159(9):957–63. [23] Thompson D, Edelsberg J, Colditz GA, Bird AP, Oster G. Lifetime health and economic consequences of obesity. Arch Int Med 1999;159(18):2177–83. [24] Bourdel-Marchasson I, Helmer C, Barberger-Gateau P, Peuchant E, Fevrier B, Ritchie K, et al. Characteristics of undiagnosed diabetes in community-dwelling French elderly: the 3C study. Diabetes Res Clin Pract 2006;76:257–64. [25] Wilder RP, Majumdar SR, Klarenbach SW, Jacobs P. Socio-economic status and undiagnosed diabetes. Diabetes Res Clin Pract 2005;70:26–30. [26] Weissman JS, Stern R, Fielding SL, Epstein AM. Delayed access to health care: risk factors, reasons, and consequences. Ann Int Med 1991;114:325–31. [27] DeVoe JE, Fryer GE, Phillips R, Green L. Receipt of preventive care among adults: insurance status and usual source of care. Am J Public Health 2003;93:786–91. [28] Baker DW, Sudano JJ, Albert JM, Borawski EA, Dor A. Lack of health insurance and decline in overall health in late middle age. N Engl J Med 2001;345: 1106–12.

157

[29] Al-Moosa S, Allin S, Jemiai N, Al-Lawati J, Mossialos E. Diabetes and urbanization in the Omani population: an analysis of national survey data. Popul Health Metr 2006;4:5. [30] Gupta A, Gupta R, Sarna M, Rastogi S, Gupta VP, Kothari K. Prevalence of diabetes, impaired fasting glucose and insulin resistance syndrome in an urban Indian population. Diabetes Res Clin Pract 2003;61(1):69–76. [31] Laramee AS, Morris N, Littenberg B. Relationship of literacy and heart failure in adults with diabetes. BMC Health Serv Res 2007;7:98. [32] Lindau ST, Tomori C, Lyons T, Langseth L, Bennett CL, Garcia P. The association of health literacy with cervical cancer prevention knowledge and health behaviors in a multiethnic cohort of women. Am J Obstet Gynecol 2002;186(5):938–43. [33] Schillinger D, Grumbach K, Piette J, Wang F, Osmond D, Daher C, et al. Association of health literacy with diabetes outcomes. JAMA 2002;288(4):475–82. [34] Kalichman SC, Benotsch E, Suarez T, Catz S, Miller J, Rompa D. Health literacy and health-related knowledge among persons living with HIV/AIDS. Am J Prev Med 2000;18(4):325–31. [35] Scott TL, Gazmararian JA, Williams MV, Baker DW. Health literacy and preventive health care use among medicare enrollees in a managed care organization. Med Care 2002;40:395–404.

International Journal of Diabetes Mellitus 2 (2010) 158–164

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

A new paradigm between mechanical scaling and root planing combined with adjunctive chemotherapy for glycated hemoglobin improvement in diabetics Sultan Al Mubarak a,⇑, Marwan Abou Rass b, Abdulaziz Alsuwyed c, Khalid Al-Zoman a, Abdulaziz Al Sohail b, Samia Sobki d, Mohammed Tariq e, Asirvatham Alwin Robert f, Sebastian Ciancio g, Paresh Dandona h a

King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia Prince Abdulrahman Bin Abdulaziz Institute for Higher Dental Studies, Riyadh, Saudi Arabia Dental Department, King Abdulaziz Medical City, Riyadh, Saudi Arabia d Department of Pathology, Armed Forces Hospital, Riyadh, Saudi Arabia e Research Center, Armed Forces Hospital, Riyadh, Saudi Arabia f Research Center, Sultan Bin Abdulaziz Humanitarian City, Saudi Arabia g Department of Periodontics and Endodontics, School of Dental Medicine, State University of New York at Buffalo, NY, USA h Department of Endocrinology State University of New York at Buffalo, NY, USA b c

a r t i c l e

i n f o

Article history: Received 29 May 2010 Accepted 31 August 2010

Keywords: Diabetics Gingival health Periodontal disease

a b s t r a c t Aim: The objective of the study was to evaluate the effectiveness of scaling and root planing (SRP) and adjunctive chemotherapy (doxycycline hyclate, 20 mg) on gingival health, specific cytokines and glycemic control in diabetic subjects. Methods: Three hundred and forty-six type 1 and 2 diabetic subjects were randomized into four test groups: (1) one session of SRP at the baseline visit and placebo tablets twice/day, started at the baseline visit, for 3 months, (2) one session of SRP at the baseline visit, and doxycycline hyclate (20 mg, twice/day) started at the baseline visit for 3 months, (3) two sessions of SRP, first at the baseline visit and second at the 6 months, with placebo tablets twice/day started at the baseline visit and 6-month visit, for 3 months at each visit, and (4) two sessions of SRP, first at the baseline visit and the second at the 6-month visit, and doxycycline hyclate 20 mg twice/day, started at the baseline visit and the 6-month visit, for 3 months at each visit. Venous blood samples were obtained to evaluate TNF-a, IL-1a and glycated hemoglobin (HbA1c); dental measurements were also included. Results: HbA1c showed significant improvement (P < 0.05) only for subjects with glycated hemoglobin 68.8% within each group, as well as when subjects were combined together. All groups achieved statistically significant improvements for most of the dental parameters at follow-up visits (P < 0.05) compared to the baseline. Conclusions: Eliminating periodontal inflammation may significantly reduce glycated hemoglobin levels for subjects with HbA1c 68.8%; furthermore, SRP and adjunctive therapy improved periodontal inflammation in diabetics. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction The World Health Organization (WHO) and International Diabetes Federation (IDF) have predicted that the number of diabetics will increase significantly by the year 2030 to approximately 366 million, an increase of 214% compared to the percentage in 2006 [1]. Diabetes is associated with several complications, and some types have been linked to a chronic hyperglycemic state. Diabetes is also frequently associated with pathological changes in the blood ⇑ Corresponding author. Address: King Faisal Specialist Hospital and Research Center, P.O. Box 3354, Riyadh 11211, Saudi Arabia. Tel.: +966 1 4424238; fax: +966 1 4427894. E-mail address: [email protected] (S. Al Mubarak).

vessel walls [2]. The IDF and American College of Endocrinology (ACE) recommend HbA1c values of below 6.5%, while the American Diabetes Association (ADA) recommends that the HbA1c be below 7.0% for most patients [3]. Loe [4] reported that diabetes is a risk factor for periodontitis, and periodontal disease is the sixth-leading complication of diabetes [4]. Hyperglycemia appears to trigger a series of events leading to a higher risk of infection. The association between diabetes and an increased susceptibility to oral infection, including periodontal disease, is significant [5]. Chronic periodontitis is a slowly progressing disease that is primarily the result of an inflammatory response to plaque and calculus accumulation [6]. A more rapidly progressing clinical presentation of chronic periodontitis has been described in diabetic subjects [3,5]. Periodontal disease may also be an independent predictor of incident type 2

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.08.006

159

S. Al Mubarak et al. / International Journal of Diabetes Mellitus 2 (2010) 158–164

diabetes, according to a study published in the July issue of Diabetes Care [7]. Studies have shown that diabetic patients with periodontal infection have a greater risk of worsening glycemic control over time, when compared to diabetic subjects without periodontitis [8]. The level of specific inflammatory markers (TNF-alpha) in the gingival crevice fluid has also been related to the level of glycemic control in diabetic patients [9]. The diabetic state has also been shown to have an upregulated monocytic TNF-alpha secretion phenotype which, in the presence of Gramnegative bacterial challenge, is associated with the expression of more severe periodontal disease [9]. Several studies addressed the effect of periodontal treatment on the glycemic control of diabetic subjects [9–14]. However, the role of periodontal treatment on metabolic control within diabetics is still controversial [11,12,14–16]. Some studies show that periodontal treatment improves periodontal status in diabetic subjects [16]. However, other studies report that a further improvement in metabolic control is achieved when mechanical periodontal treatments and systemic antibiotics are used together [10,17]. Studies have also demonstrated that the subgingival delivery of doxycycline improves dental parameters, i.e., PPD, CAL, BOP, GI and PI in subjects with chronic periodontitis, when used in conjunction with supragingival scaling and dental prophylaxis. Mechanical therapy is the standard treatment in arresting disease progression and inflammation, and non-surgical care with adjunctive pharmacotherapies (such as antimicrobials and/or antibiotics aimed at modifying the destructive host response) has proved to be of additional benefit [11,16,18,19]. The aim of this study is to evaluate the effectiveness of scaling and root planing (SRP) therapy with or without chemotherapy (doxycycline hyclate, 20 mg) on gingival health, specific cytokines, and glycemic control in diabetic subjects. 2. Subject and methods 2.1. Study design This study is a double-blinded, randomized, placebo-controlled, 12-month multi-center trial. Subject selection was conducted using eligibility screening following an assessment of periodontal status during baseline visits, and eligible subjects were given individual patient numbers; this was to validate the blinding and randomization analysis. The subjects at each hospital were allocated numbers which were used to randomize the different groups. This was achieved by distributing the subject number among the four test groups by sequentially allocating them to one of four alphabetical codes relating to a test group (i.e., A for group 1, B for group 2, C for group 3, and D for group 4). The four test groups were: (1) one session of SRP at the baseline visit only and placebo tablets twice/ day, starting at baseline visit and continuing for 3 months only; (2) one session of SRP at the baseline visit only and doxycycline hyclate (20 mg, twice/day) starting at the baseline visit and continuing for 3 months only; (3) two sessions of SRP, the first at the baseline visit and the second at the 6-month visit and placebo tablets twice/day starting at the baseline visit and the 6-month visit,

continuing for 3 months after each visit; and (4) two sessions of SRP, first at the baseline visit and the second at the 6-month visit, with doxycycline hyclate 20 mg, twice/day starting at the baseline visit and 6-month visit, continuing for 3 months after each visit (Table 1). Clinical measurements included the following periodontal parameters: probing pocket depth (PPD), clinical attachment level (CAL), gingival index (GI), plaque index (PI) and bleeding on probing (BOP). Fasting venous blood samples (20 mL) were obtained from the antecubital vein by venipuncture using a 27-G butterfly needle in the morning between 8:00 a.m. to 10:30 a.m. for the laboratory analysis of cytokines [Tumor Necrosis Factor alpha (TNF-a), Interleukin-1 alpha (IL-1a)] and to evaluate glycated hemoglobin (HbA1c) in all subjects. It should be mentioned here that the examiners as well as the dental hygienist in the different hospital were all blinded to the assigned treatment. Subjects were also blinded to the prescribed medication (doxycycline hyclate, 20 mg or placebo). 2.2. Study population This study was conducted on 369 diabetic subjects registered at Riyadh Armed Forces Hospital, King Faisal Specialist Hospital and Research Center, King Abdul Aziz Medical City, Naval Base Hospital and Sultan Bin Adulaziz Humanitarian City, Riyadh, Saudi Arabia. The subjects were recruited from the above hospitals during their routine dental follow-up appointments. All subjects who were willing to participate in this research were asked to sign an informed consent agreement to participate in this study. The study was approved by the Research and Ethics Committee of the five involved hospitals and the study registered with International Standard Randomized Controlled Trial Number (ISRCTN-11742127). Twenty-three subjects did not continue at several intervals during the 12 months of the study, for reasons that violated the inclusion criteria – i.e., taking an antibiotic during the study period, extraction of tooth/teeth, minor or major surgical intervention during the study period, or due to loss of contact information. It should be mentioned here that those who failed to show up at any time point after the baseline visit or failed to show at the last visit (12 months) were also completely excluded from the study. A total of 346 subjects continued until the end of the study. Inclusion criteria were as follows: age range between 18 and 65 years old; diabetes identified as type 1 or 2; diabetes diagnosed P1 year; diabetes under control by oral hypoglycemic agent or insulin or both; constant type and dose of diabetic medication administered for the previous 6 months; and good physical condition with no serious medical conditions or transmittable diseases – i.e., malignant disease; active hepatitis; freedom from any cardiac condition that would require antibiotic prophylaxis prior to SRP; a minimum of 18 remaining natural and non-capped teeth; a minimum of six sites in a minimum of two different quadrants with PPD P5 mm but 68 mm; no treatment with SRP in the 6 months prior to the baseline visit; visible supragingival calculus in a minimum of four teeth in two different quadrants; the absence of orthodontic bands and brackets and/or dental appliances that would compromise the scored index; no use of antibiotics within the 3 months prior to the

Table 1 Distribution of different groups based on the treatment. Months

Baseline

3 months

6 months

9 months

12 months

Groups Group-1 Group-2 Group-3 Group-4

SRP + placebo (for 3 months) SRP + doxycycline hyclate (for 3 months) SRP + placebo (for 3 months) SRP + doxycycline hyclate (for 3 months)

– – – –

– – SRP + placebo (for 3 months) SRP + doxycycline hyclate (for 3 months)

– – – –

– – – –

SRP: scaling and root planing.

160

S. Al Mubarak et al. / International Journal of Diabetes Mellitus 2 (2010) 158–164

baseline appointment; and female subjects not pregnant or nursing.

performed at baseline and during the 3-, 6-, 9- and 12-month visits. 2.4. Cytokines analysis

2.3. Assessment procedures Three trained, precalibrated examiners (periodontists) were allocated among the four centers. They were assigned to perform clinical assessments for all subjects throughout the study. Interand intra-examiner variability in the dental examination criteria were tested by performing duplicate examinations on 16 randomly selected subjects among the four centers at the baseline visit. The percentages of agreement were 93% for PPD, 87% for CAL, 94% for BOP, 91% for PI, and 88% for GI. An assessment for the each subject was conducted at baseline and at 3, 6, 9 and 12 months. Clinical dental measurements included PPD, CAL, BOP, GI, and PI; they were collected at the baseline, and at 3, 6, 9 and 12 months. Each subject received full mouth supragingival and subgingival debridement using ultrasonic and hand instrumentation. This treatment was undertaken at one or two different visits, with a maximum of seven days between the first and second visits, by four trained dental hygienists, who were recruited solely for this purpose, at the four centers. Dental hygiene aids were provided for the subjects – i.e., toothbrush and toothpaste at the baseline visit. Written oral hygiene instructions were given to all subjects within the different treatment groups at each session, including appropriate teeth brushing technique (Bass method) [20] and demonstration of the proper use of inter dental brushes and dental floss. The Bass method calls for up and down strokes on the sides of the teeth with back and forth strokes on the tops of teeth. PPD, CAL, and BOP were measured on all existing teeth based on the above inclusion criteria at the six sites (mesio-buccal, medbuccal, disto-buccal, mesio-lingual, med-lingual and disto-lingual) using pressure-sensitive periodontal probes (Ò Florida Probe Corporation, 3700 NW 91st Street, C-100, Gainesville, FL 32606, USA). Four teeth within a minimum of two quadrants were selected, based on the criteria mentioned earlier for the follow-up examination at the 3-, 6-, 9- and 12-month visits. CAL was measured for teeth using the cemento-enamel junction (CEJ) as a reference point. GI [21], PI [22] were measured for all teeth on the facial, lingual, mesial and distal surfaces excluding the third molars. GI scores were as follows: 0 = normal gingiva, no inflammation, discoloration or bleeding; 1 = mild inflammation, slight color change, mild alteration of gingival surface, no bleeding; 2 = moderate inflammation, erythema and swelling, bleeding on probing or when pressure was applied; and 3 = severe inflammation, erythema and swelling, tendency to spontaneous bleeding, perhaps ulceration. The PI scores were as follows: 0 = no plaque; 1 = thin film of plaque at the gingival margin, visible only when scraped with an explorer; 2 = moderate amount of plaque along the gingival margin, which can be seen by the naked eye; 3 = heavy plaque accumulation at the gingival margin; interdental space filled with plaque. BOP was measured for all teeth at the six sites for each tooth, and rated as follows: 0 = no bleeding within 15 s after probing, or 1 = bleeding within 15 s after probing. The HbA1c test was

A commercially available human interleukin enzyme-linked immunosorbent assay kit (Duo Set, ELISA Development System, UK) was used to determine the effect of the periodontal treatment on TNF-alpha, IL-1 alpha. The standards and samples were incubated in a 96-well polystyrene microplate coated with Capture Antibodies TNF-alpha, and IL-1 alpha, respectively. The interleukins in the samples were bound to the wells, and the other components of the samples were removed by washing and aspiration. The interleukins were detected by biotinylated goat anti-human antibodies. The amount of peroxidase bound to each well was measured by adding a tetramethylbenzidine (TMB) substrate. The reaction was quenched by the addition of 2 N sulphuric acid. The plate was read at 450 nm. The concentration of the interleukins in the serum samples was calculated by interpolation from a standard curve. 2.5. Statistical analysis Statistical analyzes were performed on the data obtained from all subjects who completed the study and had no substantial protocol violations. All the available data from these subjects were analyzed, and no imputations were carried out for missing data. The results of dental parameters (PPD, CAL, GI, PI, and BOP), cytokines (TNF-a and IL-1a) and HbA1c measurements were analyzed using one-way analysis of variance (ANOVA). The Tukey–Kramer multiple comparisons test was used for comparisons among test groups. P-values <0.05 were assumed to be statistically significant. 3. Results The age, gender and distribution of patients to the different study groups are illustrated in Table 2. A total of 346 subjects continued until the end of the study. 33 subjects were defined as type 1 diabetics out of the total sample; the remaining were defined as type 2 diabetics. 3.1. HbA1c (%) Group 1 showed a marked but insignificant reduction at 3 and 6 months (8.89 ± 0.34% and 8.97 ± 0.48%, respectively). However, it rebound to a higher (but statistically insignificant) level than that recorded at the baseline visit (9.87 ± 0.33% and 9.90 ± 0.52%, respectively). Group 2 showed the same trend as group 1, and the changes were not significant. Group 3 did not show significant changes at the 3-, 6-, 9-, and 12-month visit (9.05 ± 0.39%, 8.58 ± 0.54%, 8.82 ± 0.42%, and 8.90 ± 0.53%, respectively) as compared to the baseline value (8.87 ± 0.29%), with no statistical difference between the different follow-up visits. Compared to the baseline value the group 4 showed a slight increase, with no

Table 2 Distribution of subjects in respective study groups. Group

1 2 3 4 Total

Subjects distribution

98 93 75 80 346

Gender

Mean age (years)

Male

Female

41 44 34 42 161

57 49 41 38 185

51 ± 6 48 ± 5 47 ± 4 43 ± 6 47 ± 6

Types of diabetes Type-1

Type-2

9 9 7 8 33

89 84 68 72 313

S. Al Mubarak et al. / International Journal of Diabetes Mellitus 2 (2010) 158–164

significant differences at the 3-, 6- and 12-month visits (9.14 ± 0.30%, 9.20 ± 0.28%, and 9.78 ± 0.38%, respectively) except 9-month visit (9.03 ± 0.35%). However, no significant differences were observed between the different groups (Fig. 1a). A scaled-down statistical analysis of HbA1c for those subjects with baseline readings 68.8% within each individual treatment group showed a steady but continuous (and significant) numeric reduction associated with an improvement in periodontal health (Fig. 1b). Interestingly, this significant reduction was observed when those subjects with HbA1c 68.8% (n = 132) within the four groups combined together (Fig. 1c).

161

(1, 2, 3, and 4) or between the different groups (Table 3) except group 3 at 6-month and group 4 at 9-month visit compared to the baseline readings. 3.3. IL-1a (pg/ml) The results showing the effect of different treatments on TNFalpha are described in Table 3. There were no significant changes at the different time points within the four examination groups (1, 2, 3, and 4) or between the different groups (Table 3) except group 3 at 12-month and group 4 at 12-month visit compared to the baseline readings.

3.2. TNF-a (pg/ml) The results showing the effect of different treatments on TNFalpha are described in Table 3. There were no significant changes at the different time points within the four examination groups

3.4. Clinical periodontal parameters 3.4.1. Dental indices Changes within the periodontal parameters are illustrated in Table 4. It should be noted here that there were significant changes in PPD, PI, GI, and BOP for all groups, as compared to the baseline in the follow-up visits; such changes were not observed for CAL (Table 4). 4. Discussion

Fig. 1a. HbAlc (%) level within different treatment groups. Values are mean ± SEM, *P-values versus base line, Tukey–Kramer multiple comparisons test. Groups compared by Tukey–Kramer multiple comparisons test: $ 1 vs. 3 (P < 0.05), f 2 vs. 4 (P < 0.05).

Fig. 1b. Periodontal treatment on HbAlc (%) (=<8.8). Values are mean ± SEM, *P-values versus base line, Tukey–Kramer multiple comparisons test, *P < 0.05, **P < 0.01. Groups compared by Tukey–Kramer multiple comparisons test:   1 vs. 4 (P < 0.05).

Fig. 1c. Periodontal treatment on HbA1c (%) (=<8.8). For all groups.

Adult periodontitis is a chronic inflammatory condition, characterized by acute episodes of periodontal destruction occurring in a susceptible host. The successful long-term management of periodontitis may require an integrated and tailored treatment that ensures that mechanical debridement will help in protecting the susceptible host and eliminating the causes of periodontitis. Previous studies have demonstrated that the subgingival delivery of doxycycline improves dental parameters, i.e., PPD, CAL, BOP, GI and PI in subjects with chronic periodontitis, when used in conjunction with supragingival scaling and dental prophylaxis [14,19,23]. Extending the longevity of teeth and reducing pathological, microbial anaerobic bacteria are the important reasons for performing periodontal therapy, to stabilize periodontal health within susceptible individuals and to prevent further loss of periodontal support [24]. Many studies have addressed the effect of periodontal treatment on the glycemic control of diabetic subjects [10–14,16]. The outcome of these studies is controversial. Some studies have shown that periodontal treatment has made little – if any-clinical improvement to glycemic control within diabetic subjects [10,16]. However, other studies show that periodontal treatment improves glycemic control in diabetic subjects [16], and report that there are further significant numeric as well as clinical improvements achieved in metabolic control when mechanical periodontal treatments and systemic antibiotics were combined together [10,17]. A reduction in glycated hemoglobin was noted in type 1 diabetes subjects following periodontal therapy, combined with systemic antibiotic treatment [17]. Similar results were found in type 2 diabetes subjects using systemic doxycycline [10]. It should be mentioned that in the present study, there were no significant changes to be observed in HbA1c; in addition, such changes were not equal for all groups within the study population or for subjects within each group. This is in agreement with previous studies, which show that mechanical periodontal treatment demonstrates an improvement in periodontal status without changes in glycemic control [10,14,19,25]. However, other studies have reported improvements in periodontal status and glycemic control when mechanical treatment and systemic antibiotics are included [10,19]. Williams and Mahan [17] also demonstrate that periodontal therapy improves metabolic control, as indicated by reduced insulin requirements and reductions in the blood glucose level.

162

S. Al Mubarak et al. / International Journal of Diabetes Mellitus 2 (2010) 158–164

Table 3 Effect of different treatments on IL-1 alpha (pg/ml) and TNF-alpha (pg/ml). Periodontal parameter

Group

Baseline

3 months

6 months

9 months

IL-1 alpha (pg/ml)

1 2 3 4

3.16 ± 0.43 3.02 ± 0.37 3.08 ± 0.27 3.04 ± 0.43

3.42 ± 0.42 3.16 ± 0.37 3.1 ± 0.43 3.16 ± 0.38

2.85 ± 0.39 2.97 ± 0.37 2.63 ± 0.39 2.57 ± 0.41

2.74 ± 0.37 2.73 ± 0.32 3.05 ± 0.37 2.89 ± 0.38

TNF-alpha (pg/ml)

1 2 3 4

12.06 ± 2.43 11.87 ± 2.12 11.06 ± 1.73 11.4 ± 2.45

12.86 ± 2.54 12.42 ± 2.67 12.96 ± 2.12 12.06 ± 2.03

10.28 ± 2.16 11.96 ± 2.42 9.07 ± 2.02*à 10.13 ± 2.97

12.1 ± 2.14 10.73 ± 2.22 11.06 ± 2.87 8.26 ± 2.83* 

12 months 2.83 ± 0.41 2.72 ± 0.42 2.46 ± 0.36* 2.35 ± 0.36*  11.02 ± 2.34 9.87 ± 2.72 11 ± 2.65 10.24 ± 2.24

Values are mean ± SEM. P-values versus base line, Tukey–Kramer multiple comparisons test, *P < 0.05. Groups compared by Tukey–Kramer multiple comparisons test:   1 vs. 4 (P < 0.05), à 2 vs. 3 (P < 0.05).

*

Table 4 Effect of different treatments on periodontal parameter. Periodontal parameter

Group

3 months

6 months

9 months

12 Months

PI (mean) all teeth

1 2 3 4

2.51 ± 0.08 2.3 ± 0.1 2.15 ± 0.1 2.11 ± 0.1

1.71 ± 0.14*** 1.53 ± 0.13*** 1.5 ± 0.13*** 1.66 ± 0.13**

1.87 ± 0.13*** 1.82 ± 0.12** 1.59 ± 0.16** 1.55 ± 0.13** 

1.95 ± 0.15*** 1.46 ± 0.16*** 1.4 ± 0.18***$ 1.8 ± 0.13

1.46 ± 0.21*** 1.69 ± 0.16** 1.51 ± 0.22* 1.64 ± 0.18*

GI (mean) all teeth

1 2 3 4

2.38 ± 0.07 2.15 ± 0.11 2.06 ± 0.11 2.06 ± 0.1

1.49 ± 0.12*** 1.23 ± 0.12*** 1.27 ± 0.12*** 1.47 ± 0.13***

1.5 ± 0.12*** 1.55 ± 0.1*** 1.39 ± 0.13*** 1.53 ± 0.12***

1.78 ± 0.13*** 1.34 ± 0.16*** # 1.4 ± 0.17***$ 1.64 ± 0.13***

1.64 ± 0.2*** 1.35 ± 0.21*** 1.34 ± 0.25*** 1.27 ± 0.23*** 

PPD (mm) experimental teeth only

1 2 3 4

4.06 ± 0.13 3.9 ± 0.14 3.86 ± 0.18 3.92 ± 0.13

3.2 ± 0.13*** 2.88 ± 0.15*** 2.7 ± 0.14***$ 2.86 ± 0.15***

2.94 ± 0.16*** 3.05 ± 0.17*** 2.74 ± 0.16*** 2.94 ± 0.15***

3.16 ± 0.26*** 2.82 ± 0.22*** 2.63 ± 0.22*** 2.99 ± 0.16***

2.69 ± 0.25*** 2.4 ± 0.2*** 2.46 ± 0.35*** 2.29 ± 0.27***

CAL (mm) experimental teeth only

1 2 3 4

5.58 ± 0.23 5.01 ± 0.2 5.32 ± 0.27 4.96 ± 0.18

4.92 ± 0.22 4.43 ± 0.23 4.42 ± 0.24* 4.45 ± 0.19

4.87 ± 0.25 4.6 ± 0.25 4.53 ± 0.25 4.64 ± 0.2

4.8 ± 0.31 4.47 ± 0.3 4.48 ± 4.41 4.65 ± 0.17

5.25 ± 0.29 4.37 ± 0.25 4.89 ± 0.29 3.83 ± 0.18

35.2 ± 4.5*** 24.1 ± 4.2*** 21.9 ± 3.9*** 32.5 ± 5.3***

31.2 ± 5.3*** 31.2 ± 4.5*** 21.4 ± 4.2*** 24.6 ± 4.2***

34.8 ± 5.8*** 28 ± 5.4*** 24.3 ± 6.4*** 28.4 ± 5.9***

5.25 ± 6*** 15.9 ± 6.4*** 20.9 ± 8.4*** 15.1 ± 6.1***

BOP (%) all teeth

1 2 3 4

Baseline

75.8 ± 3.3 75.01 ± 3.4 61.7 ± 4.8 65.2 ± 4.7

 

PPD – probing pocket depth, CAL – clinical attachment level, GI – gingival index, PI – plaque index, BOP – bleeding on probing. Values are mean ± SEM, *P-values versus base line, Tukey–Kramer multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001. Groups compared by Tukey–Kramer multiple comparisons test: # 1 vs. 2 (P < 0.05), $ 1 vs. 3 (P < 0.05),   1 vs. 4 (P < 0.05).

A scaled-down statistical analysis of HbA1c for those subjects with baseline readings 68.8% within each individual treatment group showed a steady but continuous (and significant) numeric reduction associated with an improvement in periodontal health (Fig. 1b). Interestingly, this significant reduction was observed when those subjects with HbA1c 68.8% (n = 132) within the four groups are combined together (Fig. 1c). In addition, there was a statistically significant improvement at each time point when compared to the baseline reading; also, when the HbA1c value became higher within that range (68.8%), the reduction was more noticeable. This observation may be explained by the impact of the local oral infection and periodontal inflammation, i.e., swelling, bleeding and calculus accumulation on the underlying systemic conditions may be systemically effective, and add numerical impact and value, which could further provoke an underlying inflammatory response when HbA1c is higher within that range (68.8%) [4,15] and enhance the glycemic disturbance. In addition, when combined with high HbA1c levels within the range 68.8%, this may synergistically raise the HbA1c level within this group of diabetic patients. Therefore, the improvement may be numerically noticeable, due to the improvement in previously mentioned local factors, which may contribute to the systemic improvement maintained by hypoglycemic medications. On the other hand, when the glycemic level is uncontrolled (i.e.,

>9%), which may occur because of several major systemic factors (i.e., inappropriate dose of hypoglycemic medications, uncontrolled diabetes), the influence of periodontal therapy is defined hereby as local factors, i.e., improvement in periodontal infection will be diminished, and may not enhance improvement compared to the underlying compromised systemic condition (poor glycemic control). Also, it was noticeable that when the HbA1c level approached near to the normal range (6.5%), such associations became weak or/and insignificant. This complex phenomenon may be explained hypothetically by the fact that when the glycated hemoglobin level is relatively controlled (6.5%), any improvement within the local oral environment (measured by periodontal indices) may have minimal impact, if any, on the systemic improvement measured by HbA1c. Thus, periodontal therapy may enhance improvement near the normal range of glycated hemoglobin for those subjects under good diabetic control. The evidence and reasoning described herein may improve the understanding of the significant conflict and controversy that resulted from previous studies, as most of those studies did not establish a clear and appropriate link to the trends and levels of glycated hemoglobin. Further studies may be needed to test these findings. Studies have shown that TNF has a broad range of biological effects, including the stimulation of bone resorption [26]. The present study shows that there is a general trend toward a slight

S. Al Mubarak et al. / International Journal of Diabetes Mellitus 2 (2010) 158–164

mean reduction among the four groups when compared to the baseline measurements. However, this change is not significant when comparing different groups, or when comparing any of the follow-up time points to the baseline reading (Table 3). This may be explained by the fact that the improvement of certain inflammatory mediators (i.e., TNF-a) is sensitive to the given treatment and that this effect may not last long as such levels will rebound within 24 h [27]. Therefore, because periodontal inflammation is a chronic disease that recurs in most of the affected subjects at different levels after the given treatment, the level of TNF-a may have reduced further after the prescribed treatment but rebound before the next evaluation visit. The potential role of IL-1 in periodontal tissue destruction involves negative effects on periodontal ligament cells [28]. IL-1 has been shown to have strong stimulatory effects on increased bone resorption and inhibitory effects on bone formation [28]. Other studies have clearly demonstrated that interleukin-1 is a potent stimulator of bone resorption and that this effect is mediated via prostaglandin E2 (PGE2) [29]. The present study shows that all groups achieved a mean improvement in IL-1a by the end of the study, compared to baseline levels (Table 3). However, none of the groups achieved any clinical or statistical significance or added advantage when compared to each other. In comparing the results of this study to those of previous studies [30,31], it is suggested that SRP reduces TNF-a and IL-1a levels in the studied population. However, the reductions found were not statistically significant. In the present study, the severity of periodontal disease measured by dental indices (PPD, CAL, GI, PI, and BOP) was approximately within the same range at baseline, with no significant difference among the four groups (Table 4). Clinical improvement was statistically significant in most of the clinical periodontal parameters (PPD, GI, PI, and BOP) within the four treatment groups at the 3-, 6-, 9-, and 12-month visits (Table 4). This is in accordance with the results of several previous studies [32,33]. The findings indicate that (1) periodontal maintenance therapy, including scaling and root planing and tooth debridement, when given periodically to diabetics, and (2) oral hygiene instruction, which was given at each recall visit, resulted in markedly improved oral hygiene conditions for all treatment groups (Table 4), a result that supports the findings of earlier studies [32,33]. Therefore, it is confirmed that scaling and root planing (SRP) has a significant and beneficial effect on periodontal health in these diabetic subjects, and reduced tissue breakdown. However, adjunctive therapy (groups 2 and 4) had numeric but not statistically significant clinical advantages. The present study shows that although there is a general trend toward improvement (as measured by dental indices at each follow-up visit), group 4 subjects had a statistically significant improvement at the 12-month visit, compared to the baseline level. This is in agreement with existing research [34] and may be due to the efficacy of the repeated sessions of SRP, as well as the repeated adjunctive chemotherapy treatment (doxycycline hyclate, 20 mg) that might help in reducing gingival inflammation. It is consistent with earlier reports describing the efficacy of regular and low-dose tetracyclines, such as minocycline and low-dose doxycyline [35]. The beneficial effects of adjunctive chemotherapy (doxycycline hyclate, 20 mg), along with SRP (group 2), may help in reducing gingival inflammation. Accordingly, our observation needs further validation to provide better evidence in this regard. Whether there is a cut-off point in the HbA1c level that can be improved by treating local factors, i.e., periodontal disease is an interesting point to discuss in future studies. Further studies are needed to enhance our understanding of the role of periodontal treatment in diabetes mellitus. In conclusion to this study, the present results suggest that the improvement in periodontal inflammation measured by periodontal indices may lead

163

to a significant improvement in the glycated hemoglobin level for subjects with HbA1c 68.8%. Contribution All authors contributed equally to the conception, design, and interpretation of data and the final manuscript. Conflict-of-interest disclosure The authors declare no competing financial interests. Acknowledgements This project was funded by King Abdulaziz City for Science and Technology (KACST), Riyadh, Saudi Arabia (Research Grant #3/ 21/T1). References [1] Agarwal S, Raman R, Paul PG, Rani PK, Uthra S, Gayathree R, et al. Sankara Nethralaya-Diabetic Retinopathy Epidemiology and Molecular Genetic Study (SN-DREAMS 1): study design and research methodology. Ophthalmic Epidemiol 2005;12(2):143–53. [2] Colca JR. Discontinued drugs in 2008: endocrine and metabolic. Expert Opin Investig Drugs 2009;18(9):1243–55. [3] Executive summary: standards of medical care in diabetes. Diabetes Care 2009;32(1):S6–12. [4] Loe H. Periodontal disease. The sixth complication of diabetes mellitus. Diabetes Care 1993;16(1):329–34. [5] Cullinan MP, Ford PJ, Seymour GJ. Periodontal disease and systemic health: current status. Aust Dent J 2009;54(Suppl 1):S62–9. [6] Calabrese N, Galgut P, Mordan N. Identification of Actinobacillus actinomycetemcomitans, Treponema denticola and Porphyromonas gingivalis within human dental calculus: a pilot investigation. J Int Acad Periodontol 2007;9(4):118–28. [7] Demmer RT, Jacobs Jr DR, Desvarieux M. Periodontal disease and incident type 2 diabetes: results from the First National Health and Nutrition Examination Survey and its epidemiologic follow-up study. Diabetes Care 2008;31(7):1373–9. [8] Taylor GW, Burt BA, Becker MP, Genco RJ, Shlossman M, Knowler WC, et al. Severe periodontitis and risk for poor glycemic control in patients with noninsulin-dependent diabetes mellitus. J Periodontol 1996;67(Suppl. 10):1085–93. [9] Salvi GE, Collins JG, Yalda B, Arnold RR, Lang NP, Offenbacher S. Monocytic TNF alpha secretion patterns in IDDM patients with periodontal diseases. J Clin Periodontol 1997;24(1):8–16. [10] Grossi SG, Skrepcinski FB, DeCaro T, Robertson DC, Ho AW, Dunford RG, et al. Treatment of periodontal disease in diabetics reduces glycated hemoglobin. J Periodontol 1997;68(8):713–9. [11] Navarro-Sanchez AB, Faria-Almeida R, Bascones-Martinez A. Effect of nonsurgical periodontal therapy on clinical and immunological response and glycaemic control in type 2 diabetic patients with moderate periodontitis. J Clin Periodontol 2007;34(10):835–43. [12] Santos VR, Lima JA, De Mendonca AC, Braz Maximo MB, Faveri M, Duarte PM. Effectiveness of full-mouth and partial-mouth scaling and root planing in treating chronic periodontitis in subjects with type 2 diabetes. J Periodontol 2009;80(8):1237–45. [13] Taylor GW. Periodontal treatment and its effects on glycemic control: a review of the evidence. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1999;87(3):311–6. [14] Tervonen T, Lamminsalo S, Hiltunen L, Raunio T, Knuuttila M. Resolution of periodontal inflammation does not guarantee improved glycemic control in type 1 diabetic subjects. J Clin Periodontol 2009;36(1):51–7. [15] Kasaj A, Gortan-Kasaj A, Willerhausen B, Hoffmann O, Angelov N, Zafiropoulos GG. The relationship of periodontitis and diabetes mellitus. Acta Med Croatica 2007;61(4):369–74. [16] Montoya-Carralero JM, Saura-Perez M, Canteras-Jordana M, Morata-Murcia IM. Reduction of HbA1c levels following nonsurgical treatment of periodontal disease in type 2 diabetics. Med Oral Patol Oral Cir Bucal 2010. [17] Williams Jr RC, Mahan CJ. Periodontal disease and diabetes in young adults. J Am Med Assoc 1960;20(172):776–8. [18] Garofalo GS. Relationships between diabetes mellitus and periodontal disease: current knowledges and therapeutic prospects. Clin Ter 2008;159(2):97–104. [19] Promsudthi A, Pimapansri S, Deerochanawong C, Kanchanavasita W. The effect of periodontal therapy on uncontrolled type 2 diabetes mellitus in older subjects. Oral Dis 2005;11(5):293–8. [20] Crews K, O’Hara J, Gordy F, Penton N. The Bass technique: Charles Cassidy Bass’ legacy. Miss Dent Assoc J 1995;51(2):18–20.

164

S. Al Mubarak et al. / International Journal of Diabetes Mellitus 2 (2010) 158–164

[21] Loe H, Silness J. Periodontal disease in pregnancy. I. Prevalence and severity. Acta Odontol Scand 1963;21:533–51. [22] Silness J, Loe H. Periodontal disease in pregnancy. Ii. Correlation between oral hygiene and periodontal condition. Acta Odontol Scand 1964;22: 121–35. [23] Ciancio S, Ashley R. Safety and efficacy of sub-antimicrobial-dose doxycycline therapy in patients with adult periodontitis. Adv Dent Res 1998;12(2):27–31. [24] Darby IB, Hodge PJ, Riggio MP, Kinane DF. Clinical and microbiological effect of scaling and root planing in smoker and non-smoker chronic and aggressive periodontitis patients. J Clin Periodontol 2005;32(2):200–6. [25] Christgau M, Palitzsch KD, Schmalz G, Kreiner U, Frenzel S. Healing response to non-surgical periodontal therapy in patients with diabetes mellitus: clinical, microbiological, and immunologic results. J Clin Periodontol 1998;25(2): 112–24. [26] Hounoki H, Sugiyama E, Mohamed SG, Shinoda K, Taki H, Abdel-Aziz HO, et al. Activation of peroxisome proliferator-activated receptor gamma inhibits TNFalpha-mediated osteoclast differentiation in human peripheral monocytes in part via suppression of monocyte chemoattractant protein-1 expression. Bone 2008;42(4):765–74. [27] Al-Mohanna F, Saleh S, Parhar RS, Collison K. IL-12-dependent nuclear factorkappaB activation leads to de novo synthesis and release of IL-8 and TNF-alpha in human neutrophils. J Leukoc Biol 2002;72(5):995–1002. [28] Tatakis DN. Interleukin-1 and bone metabolism: a review. J Periodontol 1993;64(5):416–31.

[29] Rausch-Fan X, Ulm C, Jensen-Jarolim E, Schedle A, Boltz-Nitulescu G, Rausch WD, et al. Interleukin-1beta-induced prostaglandin E2 production by human gingival fibroblasts is upregulated by glycine. J Periodontol 2005;76(7): 1182–8. [30] Al Mubarak S, Ciancio S, Aljada A, Mohanty P, Ross C, Dandona P. Comparative evaluation of adjunctive oral irrigation in diabetics. J Clin Periodontol 2002;29(4):295–300. [31] Chaudhary LR, Spelsberg TC, Riggs BL. Production of various cytokines by normal human osteoblast-like cells in response to interleukin-1 beta and tumor necrosis factor-alpha: lack of regulation by 17 beta-estradiol. Endocrinology 1992;130(5):2528–34. [32] Knofler GU, Purschwitz RE, Jentsch HF. Clinical evaluation of partial- and fullmouth scaling in the treatment of chronic periodontitis. J Periodontol 2007;78(11):2135–42. [33] Paolantonio M, D’Angelo M, Grassi RF, Perinetti G, Piccolomini R, Pizzo G, et al. Clinical and microbiologic effects of subgingival controlled-release delivery of chlorhexidine chip in the treatment of periodontitis: a multicenter study. J Periodontol 2008;79(2):271–82. [34] Caton JG, Ciancio SG, Blieden TM, Bradshaw M, Crout RJ, Hefti AF, et al. Treatment with subantimicrobial dose doxycycline improves the efficacy of scaling and root planing in patients with adult periodontitis. J Periodontol 2000;71(4):521–32. [35] Golub LM, Ciancio S, Ramamamurthy NS, Leung M, McNamara TF. Low-dose doxycycline therapy: effect on gingival and crevicular fluid collagenase activity in humans. J Periodontal Res 1990;25(6):321–30.

International Journal of Diabetes Mellitus 2 (2010) 165–168

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

Association of serum TNF-a and IL-6 with insulin secretion and insulin resistance in IFG and IGT subjects in a Bangladeshi population q Mosaraf Hossain a,b,⇑, M Omar Faruque a, Golam Kabir a,b, Naimul Hassan a,b, Dwaipayan Sikdar b, Quamrun Nahar a, Liaquat Ali a a b

Biomedical Research Group (BMRG), BIRDEM, Dhaka-1000, Bangladesh Department of Biochemistry & Molecular Biology, University of Chittagong, Chittagong-4331, Bangladesh

a r t i c l e

i n f o

Article history: Received 16 March 2010 Accepted 28 August 2010

Keywords: Cytokines IGT IFG HOMA

a b s t r a c t Background: TNF-a and IL-6 have been shown to be associated with insulin resistance in T2 DM subjects. However, their causal role in the development of diabetes is still unsettled. Subjects and methods: In the present study 106 prediabetic subjects (17 IFG, 60 IGT, 29 IFG-IGT) were studied along with age- and BMI-matched 56 healthy controls. Insulin was measured by ELISA; TNF-a and IL-6 were measured by chemiluminescence-based EIA. Results: Fasting serum TNF-a and fasting serum IL-6 levels were not significantly different among the groups. However, a significant association of TNF-a (p < 0.013) with prediabetic (IGR) state on binary logistic regression analysis was shown when the effects of sex, BMI, WHR, HOMA B and HOMA S were adjusted. On multinomial logistic regression analysis a significant positive association of TNF-a was observed with IGT and IFG-IGT subjects (p = 0.008, p = 0.008) when the effects of sex, BMI, WHR, HOMA B and HOMA S were adjusted. On multiple linear regression analysis TNF-a showed a significant positive association with insulin secretory capacity when adjusting the effects of the confounding factors. Conclusions: TNF-a is positively associated with IGT and IFG-IGT state and may have a causal relation with insulin secretory defect in IGR or prediabetic subjects. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction Inflammatory mechanisms play a key role in the pathogenesis of diabetes mellitus. Individuals who progress to type 2 diabetes display features of low-grade inflammation before the onset of diabetes. This low-grade inflammation has been suggested as being involved in the pathogenetic processes that cause type 2 diabetes [1]. Several humoral markers of inflammation are elevated in humans with type 2 diabetes [2]. Epidemiological evidence suggests that inflammatory markers predict the development of diabetes and glucose disorders [3]. Tumor necrosis factor-a (TNF-a) and Interleukin-6 (IL-6) are two major proinflammatory markers or cytokines, acting primarily as autocrine and/or paracrine factors. TNF-a is secreted by several types of cells such as macrophages, monocytes, neutrophils and T-cells. Increased TNF-a expression has been observed in adipose tissue derived from obese rodents or human subjects and has been q Sources of Financial support: Diabetic Association of Bangladesh; International Program in the Chemical Sciences (IPICS), Uppsala University, Sweden. ⇑ Corresponding author. Address: Biomedical Research Group (BMRG), Room no336A, BIRDEM, 122 Kazi Nazrul Islam Avenue, Dhaka-1000, Bangladesh. Tel.: +880 2 861664150x2578; +880 1716735474 (M); fax: +880 2 8611138. E-mail address: [email protected] (M. Hossain).

implicated as an important factor in obesity-associated insulin resistance and pathogenesis of type 2 diabetes [4]. Multiple mechanisms have been suggested to account for the metabolic effects of TNF-a. These include the downregulation of genes required for normal insulin action, the negative regulation of PPARc an important insulin-sensitizing nuclear receptor, the direct effects on insulin signaling and the induction of elevated free fatty acids via the stimulation of lipolysis [5]. On the other hand, interleukin-6 (IL-6) is an immune protein of the hematopoietins family. It is a monomer of 184 amino acids produced by T-cells, macrophages and endothelial cells found on a single gene located at 7p21. It is secreted from adipose tissue during resting conditions and from muscle during strenuous exercise [6]. The role of IL-6 in insulin resistance is controversial [7]. TNF-a and IL-6 are increased in subjects with IGT (female Turkish subject and Italian Caucasians) [8,9] but another study contradicts this result [10]. Although TNF-a and IL-6 are claimed to be associated with insulin resistance in type 2 diabetes [11,12], very few studies exist where all the three IGR (Impaired Glucose Regulation) (IFG, IGT, IFG-IGT) groups were studied regarding TNF-a and IL-6, which are important to know whether these adipocytokine hormones are the causal factors in the development of type 2 diabetes. Hence, the present study aims to find out the relation

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.08.004

166

M. Hossain et al. / International Journal of Diabetes Mellitus 2 (2010) 165–168

Table 1 Clinical and biochemical characteristics of the study subjects (IFG, IGT and IFG-IGT). Parameters

Controls (n = 56)

Age (years) Body mass index (kg/m2) Waist-to-hip ratio Neck circumference (cm) MUAC (mm) Triceps (mm) Body fat mass (%) Systolic blood pressure (mm Hg) Diastolic blood Pressure (mm Hg) Fasting glucose (mmol/l) 2 h Glucose (mmol/l) Triglyceride (mg/dl) Cholesterol (mg/dl) HDL cholesterol (mg/dl) LDL cholesterol (mg/dl) Serum GPT (U/L) Serum creatinine (mg/dl) Fasting insulin (pmol/l) HOMA B HOMA S TNF-a (pg/ml) IL-6 (pg/ml)

38 ± 6 25.4 ± 3.9 0.90 ± 0.05 35.4 ± 3.0 296 ± 28 14.4 ± 5.2 29.2 ± 6.2 114 ± 8 76 ± 8 5.1 ± 0.4 5.8 ± 1.1 149 (52–376) 188 (90–261) 31 (18–59) 127 (45–194) 27 (12–122) 1.1 (0.8–1.6) 50 (7–155) 99 (21–187) 88 (29–554) 9.9 (4.8–28.9) 2.8 (1.1–10.6)

Prediabetic subjects (n = 106) IFG (n = 17)

IGT (n = 60)

IFG-IGT (n = 29)

43 ± 6 26.6 ± 4.6 0.92 ± 0.04 36.1 ± 2.8 302 ± 33 15.5 ± 7.1 29.3 ± 7.3 120 ± 21 82 ± 9 6.2 ± 0.3* 6.7 ± 0.9* 184 (61–415) 194 (150–259) 28 (15–40) 124 (74–153) 16 (10–74) 1.1(0.9–1.3) 64 (23–121) 76 (39–113)* 68 (36–180) 11.1 (5.5–24.3) 3.2 (1.3–10.5)

41 ± 7 26.5 ± 3.5 0.91 ± 0.05 35.3 ± 3.6 306 ± 24 15.4 ± 4.9 29.7 ± 6.0 121 ± 16 82 ± 9* 5.4 ± 0.2* 9.6 ± 0.8* 165 (50–491) 196(105–298) 30 (18–54) 132 (62–239) 23 (10–162) 1.1 (0.8–1.4) 66 (6–237) 99 (26–278) 66 (19–660) 14.6 (4.2–47.5) 3.1 (1.0–17.8)

43 ± 8 26.1 ± 4.2 0.93 ± 0.05* 37.0 ± 3.7 300 ± 32 15.5 ± 5.7 29.2 ± 0.5 124 ± 19* 81 ± 9* 6.4 ± 0.3* 8.9 ± 1.0* 191 (50–401) 200 (124–261) 29 (21–43) 132 (77–192) 32 (16–92) 1.1 (0.8–1.6) 77 (22–181)* 79 (37–157)* 55 (24–193)* 10.3 (5.1–36.7) 3.3 (1.3–10)

HOMA %B = B cell function and HOMA %S = insulin sensitivity, assessed by homeostasis model assessment; TNF-a = tumor necrosis factor-alpha; IL-6 = interleukin-6. * p < 0.05, significantly different compared to controls when using Student’s ‘t’ test.

of TNF-a and IL-6 with insulin secretory capacity or insulin sensitivity in prediabetic subjects (IFG, IGT and combined IFG-IGT).

2. Materials and methods This cross-sectional observational study was conducted in the Research Division, Bangladesh Institute of Research and Rehabilitation in Diabetes, Endocrine and Metabolic Disorders (BIRDEM), Dhaka. A group of 17 impaired fasting glucose (IFG), 60 impaired glucose tolerance (IGT) and 29 combined IFG-IGT subjects were recruited purposively from the Out-Patient Department (OPD) of BIRDEM, along with a group of 56 age-, sex- and BMI-matched healthy subjects without a family history of diabetes, as controls from the friend circle of the Impaired Glucose Regulation (IGR) subjects considering the same socio-economic status. Subjects were considered as IFG, IGT and IFG-IGT using the WHO guidelines, 1999 [13]. Written consent was taken from all the volunteers; clinical examination was done by a registered physician using a predesigned questionnaire. Anthropometric measurements were taken using standard methods. Subjects were requested to come on a prescheduled morning after overnight fasting for the fasting blood sample; subjects were then given 75 gm of anhydrous glucose, dissolved in 250 ml water. Blood was taken by venepuncture at fasting condition, and 2 h after glucose loading. Fasting and postprandial serum glucose was measured using the glucose–oxidase method, and the fasting serum lipid profile (cholesterol, triglyceride and HDL) was determined by enzymatic colorimetric method, fasting serum creatinine by alkaline picrate method, and serum SGPT by UV-spectrophotometric method using commercial kits (Randox Laboratories Ltd., UK). Serum LDL was calculated using the formula of Friedewald [14]. Fasting serum insulin levels were determined through the enzyme-linked immunosorbent assay (ELISA) method (Linco Research Inc., USA). Fasting serum TNF-a and IL-6 concentrations were measured by solid-phase, enzyme-labeled, chemiluminescent immunometric assay (IMMULITE, DPC, USA). Insulin secretory capacity (HOMA B%) and insulin sensitivity (HOMA S%) were calculated from fasting glucose and fasting insulin using HOMA-CIGMA software [15].

3. Statistical analysis Statistical analysis was performed using SPSS (Statistical Package for Social Science) software for Windows version 10 (SPSS Inc., Chicago, Illinois, USA). Fasting levels of biomarkers (for glucose both fasting & postprandial) were used for statistical analysis. All the data were expressed as mean ± SD (standard deviation), median (range) and/or percentage (%) as appropriate. The statistical significance of differences between the values was assessed by ANOVA or Mann–Whitney U test (as appropriate). Logistic and Multiple regressions were performed among on the parameters. A two-tailed p value of <0.05 was considered as statistically significant. 4. Results Subjects with IFG-IGT had significantly higher waist–hip ratio (WHR) (p = 0.04), systolic blood pressure (p = 0.03), and diastolic blood pressure (p = 0.04) compared to controls. Diastolic blood pressure was also significantly higher in IGT subjects (p = 0.02) compared to controls. Fasting serum insulin was significantly higher in IFG-IGT (p = 0.004) group compared to controls. Insulin secretory capacity (HOMA B) was significantly lower in IFG (p = 0.004) and IFG-IGT (p = 0.008) subjects compared to controls. Insulin sensitivity (HOMA S) was significantly lower in IFG-IGT (p = 0.004) compared to controls. Fasting serum TNF-a and IL-6 levels were not significantly different in any of the prediabetic group compared to controls (Table 1). 4.1. Logistic regression analysis In binary logistic regression analysis, IFG, IGT and IFG-IGT subjects were considered together as a single IGR group (Table 2) and it was found that TNF-a was positively associated (p = 0.013) with IGR subjects adjusted by age, sex, BMI, HOMA B and HOMAS. In the same analysis HOMA B and HOMA S were negatively associated (p = 0.003 and p = 0.001 respectively) with IGR subjects (Table 2). Considering IFG, IGT and IFG-IGT as single groups, we performed

M. Hossain et al. / International Journal of Diabetes Mellitus 2 (2010) 165–168 Table 2 Logistic regression analysis of TNF-a and IL-6 adjusted with confounding factors (Taking control as reference). Group

Variables

Binary logistic regression IGR TNF-a IL-6 HOMA B HOMA S BMI Age Sex Constant Multinomial logistic regression IFG Intercept TNF-a IL-6 HOMA B HOMA S BMI Age Sex IGT Intercept TNF-a IL-6 HOMA B HOMA S BMI Age Sex IFG + IGT Intercept TNF-a IL-6 HOMA B HOMA S BMI Age Sex

Beta

S.E.

p value

Exp(B)

0.111 0.033 0.025 0.035 0.015 0.177 0.462 1.992

0.045 0.113 0.008 0.011 0.067 0.046 0.534 2.896

0.013 0.771 0.003 0.001 0.825 0.000 0.387 0.491

1.118 0.968 0.975 0.966 0.985 1.194 1.587 0.136

7.575 0.079 0.095 0.118 0.066 0.064 0.130 0.380 5.613 0.125 0.022 0.009 0.023 0.043 0.181 0.987 13.390 0.141 0.100 0.130 0.094 0.052 0.127 0.010

5.291 0.067 0.171 0.030 0.020 0.106 0.059 0.852 3.311 0.047 0.123 0.009 0.012 0.072 0.047 0.587 5.247 0.053 0.183 0.028 0.021 0.108 0.059 0.817

0.152 0.235 0.581 0.001 0.001 0.543 0.028 0.656 0.090 0.008 0.861 0.342 0.050 0.548 0.000 0.092 0.011 0.008 0.585 0.001 0.001 0.628 0.030 0.990

1.082 1.099 0.888 0.936 1.066 1.139 0.684 1.133 0.979 0.991 0.978 0.958 1.198 2.684 1.151 0.905 0.878 0.911 0.949 1.136 1.010

Table 3 Multiple linear regression analysis of different factors affecting HOMA B in IGR subjects. HOMA B vs

Controls Beta

TNF-a IL-6 BMI Age Sex

1.395 0.002 0.427 0.028 0.052

IGR p value 0.174 0.992 0.026 0.880 0.804

Beta 0.238 0.044 0.244 0.084 0.005

p value 0.023 0.668 0.020 0.407 0.961

multinomial regression analysis, and TNF-a showed a positive significant association (p = 0.008) with IGT and IFG-IGT subjects when age, sex, BMI, HOMA B, HOMA S and IL-6 were adjusted. In the same analysis, it was shown that HOMA B had negative association (p = 0.001) with IFG and IFG-IGT subjects, and HOMA S also had negative association with IFG (p = 0.001), IGT (p = 0.05) and IFGIGT (p = 0.001) subjects (Table 2).

4.2. Multiple linear regression analysis In multiple linear regression analysis, TNF-a has shown a significant positive association with insulin secretory capacity (HOMA B) when adjusting the effects of the confounding factors - age, sex and BMI in IGR subjects (Table 3). Such an association of IL-6 with prediabetic state and its underlying defects was not evident. Neither TNF-a nor IL-6 in IGR subjects has found any association with insulin resistance according to the data.

167

5. Discussion It has been suggested that type 2 diabetes mellitus is a disease of the innate immune system responsible for an ongoing cytokinemediated acute phase response and low-grade chronic inflammation, which may be involved in the atherosclerosis of diabetes mellitus [16]. Therefore it seems important to determine whether signs of an activated innate immune system are present before the onset of type 2 diabetes mellitus. Epidemiological evidence suggests that inflammatory markers such as TNF-a and IL-6 predict the development of diabetes and glucose disorders [3]. TNF-a contributes to the pathogenesis of insulin resistance, type 2 diabetes, and abnormal adiposity or lipid disorders [17]. Some authors have found increased serum TNF-a and IL-6 concentrations in type 2 DM and IGT subjects [9,18] but Choi et al. (2004) [10] have not found any association of TNF-a and IL-6 with IGT. However, TNF-a and IL-6 have not been extensively studied in all the three groups of IGR (IFG, IGT and IFG-IGT) subjects. It is important to know whether it has any relation with insulin resistance and insulin secretory defects which appear at the prediabetic stage. In this study, we have not found any increment of serum TNF-a in IGR subjects as compared to controls. TNF-a has been shown to be secreted in the adipocytes of both mice and human [19]. In the adipose tissue of rodents with genetic obesity and insulin resistance, increased levels of both RNA and protein of TNF-a were described, supporting a link between obesity, diabetes and TNF-a. Body Mass Index (BMI) of the subjects in the present study was of normal range, and there was no significant difference between the control subjects and the IGR groups (IFG, IGT and IFG-IGT), which may be the reason for the similar serum TNF-a concentrations among the groups in the studied subjects. A study on Korean subjects showed that serum TNF-a concentration was not elevated in prediabetic patients, as compared to controls [10], which supports the present study and gives a similar reason for the subjects not being obese. However, when the effects of HOMA B, HOMA S, age, sex and BMI were excluded using binary logistic regression, serum TNF-a levels in the present study showed a significant association with IGR. With regards to multinomial regression analysis, to observe the association in individual groups of IGR, TNF-a showed a positive significant association (p = 0.008) with IGT and IFG-IGT subjects when age, sex, BMI, HOMA B, HOMA S and IL-6 were adjusted for. After this, we considered whether TNF-a is associated with the principal features of type 2 diabetes, i.e., insulin resistance or insulin secretory defect. On multiple linear regression analysis with HOMA B and HOMA S as dependent variables, TNF-a showed a significant positive association with insulin secretory capacity when adjusting the effects of the confounding factors - age, sex and BMI in IGR subjects. Thus, there may be a causal relationship between TNF-a with insulin secretory defect in prediabetic and IGR subjects. Circulating IL-6 levels have been reported to be elevated in subjects with type 2 diabetes [24]. Moreover, IL-6 independently predicts the risk of type 2 diabetes [25]. We have not found any association of serum IL-6 level with IGR, or with any of the two principal features of diabetes. Serum IL-6 levels in IFG, IGT, IFGIGT and even in control subjects are controversial [8]. Earlier studies have documented that serum IL-6 concentrations are higher in IGT than in NGT (Normal Glucose Tolerance) [11,20,21], while other studies have found no elevation in serum IL-6 concentrations in prediabetic patients, as compared to controls [10]. A study undertaken with Italian Caucasians has shown that IGT and type 2 DM, but not IFG, are associated with elevated IL-6 levels [9]. This may happen due to variations in racial, environmental or geographical conditions or nutritional status.

168

M. Hossain et al. / International Journal of Diabetes Mellitus 2 (2010) 165–168

Although binary and multinomial regression analyses have shown significant association of TNF-a with IGR, no difference of circulating concentrations of TNF-a and IL-6 was found between controls and IGR subjects. Studies suggest that in type 2 diabetes mellitus, both TNF-a and IL-6 are increased [9,22,23]. Therefore the present study suggests that insulin resistance in type 2 diabetes mellitus may lead to, rather than be the consequence of, raised concentrations of inflammatory mediators. In conclusion, it may be stated that TNF-a is positively associated with IGT and IFG-IGT and may have a causal relation with an insulin secretory defect in IGR or prediabetic subjects. Acknowledgements Authors greatly acknowledge the Diabetic Association of Bangladesh and the International Program in the Chemical Sciences (IPICS), Uppsala University, Sweden, for the financial support of this study. References [1] Kristiansen Ole P. Thomas mandrup-poulsen interleukin-6 and diabetes. Diabetes 2005;54:S114–24. [2] Caballero AE. Endothelial dysfunction, inflammation, and insulin resistance. A focus on subjects at risk for type 2 diabetes. Curr Diab 2004;4:237–46. [3] Schmidt MI, Duncan BB, Sharrett AR, Lindberg G, Savage PJ, Offenbacher S, et al. Markers of inflammation and prediction of diabetes mellitus in adults (Atherosclerosis Risk in Communities study): a cohort study. Lancet 1999;353:1649–52. [4] Hotamisligil GS. Tumor necrosis factor (TNF-a) alpha inhibits signaling from the insulin receptor. Proc Natl Acad Sci USA 1994;91:4854–8. [5] Chen QI, Pekala Philip H. Tumor necrosis factor – alpha induced insulin resistance in adiposites (44471). PSEBM 2000;223:128–35. [6] Kristina Wallenius, Jansson JO, Wallenius V. The therapeutic potential of interleukin- 6 in treating obesity. Expert Opin Biol Ther 2003;3:1061–70. [7] Carey AL, Febbraio MA. Interleukin-6 and insulin sensitivity: friend or foe? Diabetologia 2004;47:1135–42. [8] Konukoglu D, Hatemi H, Bayer H, Bagriacik N. Relation between serum concentrations of interleukin-6 and tumor necrosis factor-a in female Turkish subjects with normal and impaired glucose tolerance. Horm Metab Res 2006;38:34–7. [9] Cardellini M, Andreozzi F, Laratta E, Marini MA, Lauro R, Hribal ML, et al. Plasma IL-6 levels are increased in subjects with impaired glucose tolerance but not in those with impaired fasting glucose in a cohort of Italian caucasians. Diabetes Metab Res 2007;23(2):141–5.

[10] Choi KM, Lee J, Lee KW, Seo JA, Oh JH, Kim SG, et al. Comparison of serum concentrations of C-reactive protein, TNF-a and IL-6 between elderly Korean women with normal and Impaired Glucose Tolerance. Diabetes Res Clin Pract 2004;64:99–106. [11] Miyazaki Y, Pipek R, Mandarino LJ, De Fronzo RA. Tumor necrosis factor alpha and insulin resistance in obese type 2 diabetic patients. Int J Obes Relat Metab Disord 2003;27:88–94. [12] Andreozzi Francesco, Laratta Emanuela, Cardellini Marina, Marini Maria A, Lauro Renato, Hribal Marta L, et al. Plasma interleukin-6 levels are independently associated with insulin secretion in a cohort of Italiancaucasian nondiabetic subjects. Diabetes 2006;55:2021–4. [13] World Health Organization Consultation. Definition, diagnosis and classification of diabetes mellitus and its complications, part 1: diagnosis and classification of diabetes mellitus, Report of a WHO Consultation, Geneva: World Health Organization; 1999. [14] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499–502. [15] Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998;21:2191–2. [16] Fernandez-Real JM, Ricart W. Insulin resistance and chronic cardiovascular inflammatory syndrome. Endrocrine Rev 2002;24:278–301. [17] Moller DE. Potential role TNF-a in the pathogengsis of insulin resistance and type 2 diabetes trends. Endocrinal Metab 2000;11:212–7. [18] Gupta A, Ten S, Anhalt H. Serum levels of soluble TNF-a receptor 2 are linked to insulin resistance and glucose tolerance in children. Pediatric Endocrinol Metab 2005;18:75–82. [19] Saghizadeh M, Ong JM, Garvey TW, Henry R, Kern PA. The expression of TNF-a by human muscle. J Clin Invest 1996;97:1111–6. [20] Esposito K, Nappo F, Marfella R, Giugliano G, Giugliano F, Ciotola M, et al. Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans; Role of oxidative stress. Circulation 2002;106:2067–72. [21] Muller S, Martin S, Koenig W, Hanifi-Moghaddam P, Rathmann W, Haastert B, et al. Impaired glucose tolerance is associated with increased serum concentrations of interleukin-6 and co-regulated acute-phase proteins but not TNF-a or its receptors. Diabetologia 2002;45:805–12. [22] Spranger J, Kroke A, Möhlig M, Ristow M, Boeing H, Pfeiffer AFH. Role of TNF alpha in the development of type 2 diabetes: preliminary results of a population-based, prospective study. Experimental and Clinical Endocrinology & Diabetes Thema: Poster Diabetes, Metabolism and Gastrointestinal Hormones II 2002. p. 194. [23] Joshi SV, Tambwekar SR, Khadalia K, Dhar HL. Role of inflammatory marker interleukin 6 (il-6) and insulin in diabetes and diabetic neuropathy. Bombay Hosp J 2008;50:466–71. [24] Pickup JC, Mattock MB, Chusney GD, Burt D. NIDDM as a disease of the innate immune system: association of acute-phase reactants and interleukin-6 with metabolic syndrome X. Diabetologia 1997;40:1286–92. [25] Joachim Spranger, Anja Kroke, Matthias Mohlig, Kurt Hoffmann, Bergmann Manuela M, Michael Ristow, et al. Inflammatory cytokines and the risk to develop type 2 diabetes. Diabetes 2003;52:812–7.

International Journal of Diabetes Mellitus 2 (2010) 169–174

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

High glucose-induced DNA-binding activities of nuclear factor of activated T cells 5 and carbohydrate response element binding protein to the myo-inositol oxygenase gene are inhibited by sorbinil in peripheral blood mononuclear cells from patients with type 1 diabetes mellitus and nephropathy Bingmei Yang ⇑, Andrea Hodgkinson, Beverley A. Millward, Andrew G. Demaine Molecular Medicine Research Group, Institute of Biomedical and Clinical Science, Peninsula Medical School, University of Plymouth, Plymouth PL6 8BU, United Kingdom

a r t i c l e

i n f o

Article history: Received 12 May 2010 Accepted 28 August 2010

Keywords: Myo-inositol Myo-inositol oxygenase Type 1 diabetes mellitus Diabetic nephropathy Aldose reductase Aldose reductase inhibitors

a b s t r a c t Aims: To investigate whether high glucose induces myo-inositol oxygenase (MIOX) expression in peripheral blood mononuclear cells through transcription factors, nuclear factor of activated T cells 5 (NFAT5) and carbohydrate response element binding protein (ChREBP), which may contribute to the pathogenesis of diabetic nephropathy. Methods: 34 patients with type 1 diabetes mellitus (20 with nephropathy, 14 without complications) and 9 healthy controls were recruited in this study. Peripheral blood mononuclear cells were exposed to normal, high glucose conditions with/without an aldose reductase inhibitor (ARI), using electrophoretic mobility shift assays the DNA-binding activities of NFAT5 and ChREBP to corresponding sites in the promoter region of MIOX gene were analysed. The protein levels and the enzyme activity of MIOX were measured. Results: DNA-binding activities of NFAT5 and ChREBP were increased under high glucose conditions and decreased in the presence of the ARI in all groups. In the presence of ARI, the DNA-binding activities of NFAT5 and ChREBP were significantly decreased by 41% (NFAT5: 0.91 ± 0.06 vs. 1.54 ± 0.12; p = 0.01) and 49% (ChREBP: 0.92 ± 0.08 vs. 1.81 ± 0.22; p = 0.001) compared with high glucose in patients with nephropathy. ARI treatment decreased the protein levels of MIOX under high glucose conditions in patients with nephropathy (0.81 + 0.19 vs. 1.3 + 0.04; p = 0.049). Summary/conclusions: There was a trend for increased binding activities of NFAT5 and ChREBP accompanied with increased protein levels under high glucose, particularly in patients with nephropathy. ARI treatment prevented these increases and this effect was more obvious in the patients with nephropathy compared to the uncomplicated subjects. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction It has been widely accepted that long-term exposure to high blood glucose plays an important role in the development of diabetic nephropathy [1]. High glucose, together with increased flux through the polyol pathway causes myo-inositol (MI) depletion which may have a role to play in the development of diabetic nephropathy. However, the mechanism of MI depletion is still unclear. It has been proposed that myo-inositol oxygenase (MIOX) plays a key role in causing the depletion of MI. There are two enzymes in the polyol pathway, aldose reductase (AKR1B1) and sorbitol dehydrogenase. Together, they convert glu⇑ Corresponding author. Address: Molecular Medicine Research Group, The John Bull Building, Research Way, Peninsula Medical School, University of Plymouth, Plymouth PL6 8BU, United Kingdom. Tel.: +44 1752 437415; fax: +44 1752 517846. E-mail addresses: [email protected], [email protected] (B. Yang).

cose to fructose via sorbitol. Under physiological conditions, most of the cellular glucose is phosphorylated into glucose 6-phosphate, and only a minor portion is metabolised through the polyol pathway. However, high glucose conditions may saturate the enzymes involved in the phosphorylation of glucose, and as a result, onethird of intracellular glucose enters the polyol pathway, which causes an accumulation of sorbitol within cells. Accumulation of sorbitol is accompanied by a depletion of MI, and this alters Na+/K+ ATPase activity and impairs phosphatidylinositol syntheses, which is a common precursor of important secondary signalling molecules. Several studies have shown that MI depletion is associated with diabetic nephropathy, retinopathy, neuropathy and diabetic cataracts [2–7]. MIOX is the first and rate-limiting enzyme in the MI metabolism pathway, which is the glucuronate– xylulose pathway and is the only pathway for MI catabolism. It has been shown that an increase in MIOX enzyme activity is in proportion to serum glucose concentrations [8]. Therefore, MIOX

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.08.005

170

B. Yang et al. / International Journal of Diabetes Mellitus 2 (2010) 169–174

may be responsible for the MI depletion found in diabetic complications. MIOX is predominantly expressed in the kidney, nerves and liver [9–12]. The promoter region of both the human and murine MIOX genes contain osmotic response element(s) (OREs) and carbohydrate response element(s) (ChREs). The binding activities of the nuclear factor of activated T cells 5 (NFAT5) and carbohydrate response element binding protein (ChREBP) to OREs and ChREs, respectively, significantly increase under high glucose conditions [8,13]. High glucose and osmolytes significantly increase the transcriptional activity of MIOX by in vitro luciferase assays. Furthermore, increased MIOX expression has been shown in the kidneys of diabetic mice [8]. In preliminary experiments, we have shown that human peripheral blood mononuclear cells (PBMCs) express the MIOX gene by a direct sequencing of reverse transcriptase polymerase chain reaction (RT-PCR) products and Western blotting. In related studies, we have shown that NFAT5 is involved in the regulation of AKR1B1 under high glucose conditions [14] and in the presence of an aldose reductase inhibitor (ARI), the binding activities of nuclear factor kappa B (NFjB) [15] and NFAT5 (unpublished data) to OREs in the promoter of the AKR1B1 gene were suppressed. Therefore, our hypotheses are that high glucose increases MIOX protein levels and activity which might be regulated through increased binding activities of NFAT5 to OREs and ChREBP to ChREs in the promoter region of the MIOX gene. These binding activities may be different in the PBMCs from patients with nephropathy, compared to those without complications. Inhibition of AKR1B1 may suppress MIOX protein levels and activity through reducing the production of sorbitol through the polyol pathway. To the best of our knowledge, there has been no study that has investigated the factors involved in the regulation of the MIOX gene and enzyme activity in patients with type 1 diabetes mellitus (T1D) and nephropathy. Therefore, the aims of this study were to investigate the regulation of MIOX expression under high glucose conditions in the PBMCs from patients with T1D, with or without nephropathy. 2. Materials and methods 2.1. Subjects The following Caucasoid subjects were included in this study: 34 patients with T1D and 9 ethnically matched healthy controls. All patients with T1D as defined by The Expert Committee On The Diagnosis And Classification Of Diabetes Mellitus [16] had attended the Diabetes Clinic, Derriford Hospital, Plymouth. The study was approved by the Local Research Ethical Committee, and informed consent was obtained from all subjects. The criteria for diabetic microvascular complications have been published previously [17]. 2.1.1. Uncomplicated Patients (n = 14) have been diagnosed with T1D for at least 20 years but remain free of retinopathy (fewer than five dots or blots per fundus), proteinuria (urine Albustix negative on at least three consecutive occasions over 12 months) and neuropathy (overt neuropathy was defined if there was any clinical evidence of peripheral or autonomic neuropathy). 2.1.2. Diabetic nephropaths Patients (n = 20) have had T1D for at least 8 years with persistent proteinuria (urine Albustix positive on at least three consecutive occasions over 12 months or three consecutive total urinary protein excretion rates >0.5 g/24 h) in the absence of hematuria or infection on midstream urine samples. Diabetic nephropathy

was always associated with retinopathy. Retinopathy was defined as more than five dots or blots per eye; hard or soft exudates, new vessels, or fluorescein angiographic evidence of maculopathy or previous laser treatment for pre-proliferative or proliferative retinopathy; and maculopathy or vitreous haemorrhage. Fundoscope was performed by both a diabetologist and an ophthalmologist.

2.2. Cell isolation and cultures Peripheral venous blood samples (20 ml) were collected into 5% EDTA Vacutainers (Becton Dickinson, UK). The peripheral blood mononuclear cells (PBMCs) were separated by using Histopaque (Sigma, Dorset, UK) and grown in RPMI 1640 supplemented with D-glucose at a concentration of 5.5 mmol/l, 10% calf serum and 2 mmol/l L-glutamine, 100 units/ml penicillin G sodium and 100 mg/ml streptomycin sulfate with PHA-P at a concentration of 5 lg/ml in a 37 °C incubator with a controlled, humidified atmosphere of 95% air/5% CO2. The cells were divided into three groups in 200 ml-flasks. Group 1 (normal conditions-NG): cells were cultured in the above medium (D-glucose at a final concentration of 5.5 mmol/l). Group 2 (high glucose conditions-HG): 19.5 mmol/l extra D-glucose was added into above-mentioned media (D-glucose at a final concentration of 25 mmol/l). Group 3 (aldose reductase inhibitor conditions (ARI) with HG): sorbinil (at a final concentration of 10 lmol/l), was added 3 h later after 19.5 mmol/l extra Dglucose was added to the media. All cells were incubated for 5 days. At the end of the incubation time, cells were harvested and nuclear and cytoplasmic proteins were extracted as below.

2.3. Extraction of nuclear protein and cytoplasmic proteins Cells were collected and re-suspended in 100 ll of buffer A (10 lmol/l HEPES, pH 7.9, 1.5 mmol/l MgCl2, 0.5 mmol/l dithiothreitol (DTT), 0.2% NP-40, 100 mmol/l 4-(2-aminoethyl)-bezenesulfonyl fluoride (AEBSF), 18.4 mg/ml sodium orthovanadate, 42 mg/ml sodium fluoride and 2.2 mg/ml aprotonin) and held on ice for 15 min. The resulting cell lysate was then centrifuged at 13,000 rpm for 10 min. The supernatant containing cytoplasmic proteins was transferred into a fresh tube and stored at 80 °C for Western blotting and MIOX activity assays. The nuclear pellets were re-suspended in 50 ll of buffer C (20 mmol/l HEPES pH 7.9, 25% glycerol, 0.42 mol/l NaCl, 1.5 mmol/l MgCl2, 0.5 mmol/l DTT, 0.2 mmol/l EDTA, 100 mmol/l AEBSF, 18.4 mg/ml sodium orthovanadate, 42 mg/ml sodium fluoride and 2.2 mg/ml aprotonin), and incubated on ice for 10 min. After centrifugation at 13,000 rpm for 10 min the supernatant containing the nuclear protein was transferred into a fresh tube and stored at 80 °C until use in the electrophoretic mobility shift assay (EMSA). The concentrations of both nuclear and cytoplasmic proteins were determined using a CoomassieÒ Plus Protein Assay kit (Peribo Science Ltd., Chest, UK).

2.4. Electrophoretic mobility shift assay NFAT5 and ChREBP probes with consensus sequence to the OER and ChRE motifs CCTCCTCCAGGAAAGCCTTTACCCTCC and GAGCACGTGACCTACCCGTGTTG GGACACGTGAGG [8,13] of the MIOX gene were labeled with [a-32P] deoxy-ATP by T4 polynucleotide kinase (Amersham Pharmacia Biotech, Buckinghamshire, UK). The labeled probes along with the gel-binding buffer were incubated with 25 lg of nuclear proteins at room temperature for 20 min. The binding mixtures were resolved by electrophoresis on a 4% non-denaturing polyacrylamide gel at 100 V for 3–4 h. The gel was exposed to X-Omax photographic paper.

171

B. Yang et al. / International Journal of Diabetes Mellitus 2 (2010) 169–174

2.5. Western blotting

3. Results

All of the cytoplasmic proteins from each individual within each group was pooled together and 50 lg of this was loaded onto a 7.5 or 10% precast SDS–PAGE (BIO-RAD Laboratory Ltd., Hempstead, UK), electrophoresed for 2–3 h at 100 V and then transferred to nitrocellular membrane (Amersham, USA) overnight. Next, the membrane was blocked with 5% non-fat milk and 0.05% Tween 20-PBS for 1 h at room temperature. Immunoblotting was performed with a primary goat antibody against human MIOX (Insight Biotechnology Ltd., Wembley, UK) in a 1:500 dilution. Secondary antibody against goat IgG of horse-radish peroxidase-conjugated was used in 1:5000 dilution (Sigma, UK). A chemiluminescence kit (Pierce, UK) and Kodak X-Omax film (Amersham, UK) were used to detect the amount of protein. In order to have equal amounts of proteins loaded in wells, a-tubulin levels were measured using a mouse antibody against human a-tubulin in 1:5000 dilution (Sigma, UK) and a secondary antibody against mouse IgG of horse-radish peroxidase-conjugated in 1:5000 dilution (Sigma, UK). DNA-binding activities of NFAT5 and ChREBP and protein levels of MIOX were analysed and quantified using a phosphoimager (BIO-RAD, UK) with multi-analyst software. All results were expressed as means of fold increase, due to high glucose treatment, calculated by dividing the density in high glucose-treated cells by the density in untreated cells, or as a mean fold decrease due to ARI treatment, calculated by dividing the density in high glucose with ARI-treated cells by density in high glucose-treated cells.

Clinical characteristics of patients with T1D with or without diabetic nephropathy are shown in Table 1. There were no differences in age, gender, age at onset of diabetes, duration of diabetes, Haemoglobin A1c (Hb1Ac) and plasma glucose levels between the two groups. Estimated glomerular filtration rate (eGFR) was significantly lower in patients with nephropathy compared with uncomplicated subjects (59.4 ± 4.5 vs. 74.5 ± 3.2; p = 0.021). The DNA-binding activities of NFAT5 to the ORE and ChREBP to the ChRE motifs in PBMCs exposed to either normal glucose, high glucose or, ARI with high glucose conditions, from nephropaths, uncomplicated and normal control subjects are shown in Fig. 1a and b. DNA-binding activities of NFAT5 and ChREBP to their motifs increased under high glucose conditions in all study groups; however, these increases did not reach statistical significance. ARI treatment significantly decreased the DNA-binding activities of

2.6. MIOX activity assay Pooled cytoplasmic proteins from each subject group were used to perform MIOX enzyme activity by following the methods used by Prabhu’s group [13]. Two-hundred and fifty microgram of cytoplasmic proteins was mixed with 50 mM sodium acetate buffer (pH 6.0), 2 mM L-cysteine, 1 mM ferrous ammonium sulfate and with 60 mM of MI or without MI as a negative control and paralleled tubes with D-glucuronic acid as positive controls. The mixture then was incubated at 30 °C for 1 hour and the reaction was terminated by the addition of 100 ll of 25% trichloroacetic acid. The products, D-glucuronic acid in the supernatant was determined by orcinol reagent [13,18,19]. Briefly, 200 mg of orcinol were dissolved in 81.4 ml of concentrated HCl and then 2 ml of 2% ferric chloride was added in. The final solution was made up to 100 ml with H2O. This solution must be made fresh. Two millilitre of the orcinol reagent was then added into the reaction tubes, which then were incubated at 100 °C in a water bath for 30 min. After they were cooled down to room temperature, 200 ll was transferred into 96-well-microplates in triplicates and D-glucuronic acid was measured on a spectrophotometer (GENios, Tecan, UK) at a wavelength of 650 nm. The MIOX enzyme activity was normalized by the activity without MI in the incubation reaction tubes. MIOX enzyme activities were expressed as mean of fold increase due to high glucose treatment, calculated by dividing the optical density (OD) value in high glucose-treated cells by the OD value in untreated cells or as mean of fold decrease due to ARI treatment, calculated by dividing the OD value in high glucose with ARI-treated cells by the OD value in high glucose-treated cells.

2.7. Statistical analysis Data were analysed through SPSS. t-test, or one-way analysis of variance (ANOVA) was used to test DNA-binding activities, MIOX protein level, and MIOX enzyme activity between different groups. A value of p < 0.05 was considered to be statistically significant.

Table 1 Clinical characteristics of patients with type 1 diabetes mellitus and normal controls.

Male:Female Age (years) Age at onset of diabetes (years) Duration of diabetes (years) Plasma glucose (mmol/L) HbA1c (%) eGFR (mls/min/ 1.73 m2)

Nephropaths (n = 20)

Uncomplicated (n = 14)

Normal controls (n = 9)

6:14 49.6 ± 13.9 (30–82) 12.3 ± 8.8 (1– 35) 34.9 ± 8.1 (21–49) 12.7 ± 0.5

4:10 48.1 ± 14.9 (19– 74) 15.2 ± 9.4) (2– 38) 34.3 ± 9.4) (20– 53) 11.0 ± 0.4

4:5 31.5 ± 5.14 (26– 41) –

–

7.9 ± 0.1 59.4 ± 4.5*

8.4 ± 0.1 74.5 ± 3.2

– –

–

Data are means ± SE (ranges) in years. Levels of plasma, Haemoglobin A1c (HbA1c) and estimated glomerular filtration rate (eGFR) are expressed as means + SE in mmol/l, % and mls/min/1.73 m2, respectively. Diabetic nephropaths, uncomplicated and normal controls were defined in Section 2.1. * Vs. the uncomplicated, p = 0.02.

a

Nephropath

Uncomplicated

Normal control

NFAT5

NG HG ARI+HG

b

Nephropath

NG

HG ARI+HG

Uncomplicated

NG

HG ARI+HG

Normal control

ChREBP

NG HG ARI+HG

NG

HG ARI+HG

NG

HG ARI+HG

Fig. 1. NFAT5: nuclear factor of activated T cell 5; ChREBP: carbohydrate response element binding protein; NG: cells were cultured under normal conditions (D-glucose at a final concentration of 5.5 mmol/l). HG: cells were cultured under high glucose conditions (D-glucose at a final concentration of 25 mmol/l). ARI + HG: cells were cultured under high glucose conditions with aldose reductase inhibitor, sorbinil (at a final concentration of 10 lmol/l).

172

B. Yang et al. / International Journal of Diabetes Mellitus 2 (2010) 169–174

Table 2 DNA-binding activities of NFAT5 and ChREBP to the ORE and ChRE in the promoter of the MIOX gene; MIOX protein levels and MIOX enzyme activity in patients with T1D with diabetic nephropathy or uncomplicated and normal controls.

DNA binding ChREBP to ChRE NFAT5 to ORE MIOX proteins MIOX activity

Stress

Nephropaths (n = 20)

Uncomplicated (n = 14)

Normal controls (n = 9)

HG/NG ARI + HG/HG HG/NG ARI+HG/HG HG/NG ARI + HG/HG HG/NG ARI + HG/HG

1.81 ± 0.22 (n = 11)* 0.92 ± 0.08 (n = 11) 1.54 ± 0.12 (n = 13)# 0.91 ± 0.06 (n = 13) 1.3 ± 0.04 [n = 6]§ 0.81 ± 0.19 [n = 6] 1.1 ± 0.4 [n = 3] 0.8 ± 0.1 [n = 3]

1.30 ± 0.15 (n = 8) 0.92 ± 0.15 (n = 8) 1.20 ± 0.10 (n = 8) 1.13 ± 0.24 (n = 8) 1.03 ± 0.04 [n = 4] 0.85 ± 0.05 [n = 4] 1.1 ± 0.1 [n = 3] 1.0 ± 0.4 [n = 3]

1.41 ± 0.18 (n = 8) 0.95 ± 0.21 (n=8) 1.24 ± 0.10 (n = 9) 0.90 ± 0.06 (n = 9) 1.16 ± 0.56 [n = 3] 0.73 ± 0.12 [n = 3] 0.9 ± 0.1 [n = 3] 1.2 ± 0.3 [n = 3]

Abbreviations: NFAT5, nuclear factor of activated T cell 5; ChREBP, carbohydrate response element binding protein; MIOX, myo-inositol oxygenase; ORE, osmotic response element; ChRE, carbohydrate response element. NG: cells were cultured under normal conditions (D-glucose at a final concentration of 5.5 mmol/l). HG: cells were cultured under high glucose conditions (D-glucose at a final concentration of 25 mmol/l). ARI + HG: cells were cultured under high glucose conditions with aldose reductase inhibitor, sorbinil (at a final concentration of 10 lmol/l). The density value was defined as 1 under NG conditions. Data are expressed as means of fold changes ± SE under HG compared to NG conditions (HG/NG) or under the ARI with HG conditions compared to HG conditions without ARI (ARI + HG/HG) in peripheral blood mononuclear cells from subjects with diabetic nephropathy or uncomplicated and normal controls. Sample sizes in the gel shift assay are shown within parentheses. The subjects in the electrophoretic mobility shift assays were overlapped. Pooled cytoplasmic proteins from each subject group were used to perform Western blotting for measuring levels of MIOX protein and an assay for determining MIOX enzyme activity. Results are shown herein from 3 to 6 independent measurements (square brackets). * Vs. nephropaths under ARI + HG conditions (p = 0.001). # Vs. nephropaths under ARI + HG conditions (p = 0.01). § Vs. nephropaths under ARI + HG conditions (p = 0.049).

Uncomplicated

Nephropath

Normal control

MIOX α-tubulin NG

HG

ARI+HG

NG

HG

ARI+HG

NG

HG

ARI+HG

Fig. 2. MIOX: myo-inositol oxygenase; NG: cells were cultured under normal conditions (D-glucose at a final concentration of 5.5 mmol/l). HG: cells were cultured under high glucose conditions (D-glucose at a final concentration of 25 mmol/l). ARI + HG: cells were cultured under high glucose conditions with aldose reductase inhibitor, sorbinil (at a final concentration of 10 lmol/l).

NFAT5 to the ORE motif by 41% (1.54 ± 0.12 vs. 0.91 ± 0.06, p = 0.01) and ChREBP to the ChRE motif by 49% (1.81 ± 0.22 vs. 0.92 ± 0.08, p = 0.001) in the nephropaths (Table 2). However, the fold decrease due to ARI in the binding activities of NFAT5 and ChREBP were not significantly different in DN compared to the uncomplicated group (NFAT5: 0.91 ± 0.06 vs. 1.13 ± 0.24, p > 0.05; ChREBP: 0.92 ± 0.08 vs. 0.92 ± 0.15; p > 0.05) (Table 2). Furthermore, in the presence of ARI, the protein levels of MIOX were decreased in patients with nephropathy and this decrease just reached statistical significance (1.3 + 0.04 vs. 0.81 + 0.19; p = 0.049). Again, the fold decrease in the MIOX protein levels was not significant between patients with nephropathy and the uncomplicated group (0.81 + 0.19 vs. 0.85 + 0.05; p > 0.05) (Table 2, Fig. 2). Furthermore, the DNA-binding activities of NFAT5 and ChREBP are correlated within individuals (data not shown), when there is an increase in DNA binding of NFAT5 there is also increased DNA binding of ChREBP. MIOX enzyme activity was assessed by an orcinol reagent. There was no significant difference of MIOX enzymatic activity under different conditions and between all study groups (Table 2).

4. Discussion There is considerable evidence that altered inositol metabolism (MI depletion) due to hyperglycaemia is associated with microvascular complications in both T1D and T2D [2–7]. A decreased uptake rate of MI under high glucose conditions has been suggested to contribute to MI depletion, seen in diabetes and its complications

through both competitive (its stereochemical similarity between and MI) and non-competitive mechanisms (sorbitol accumulation as a result of increased flux of polyol pathway) [20]. Evidence supporting the non-competitive mechanism is the fact that the ARI significantly ameliorated the decrease in MI uptake. This suggests a close link between MI depletion and the polyol pathway. Recent studies on the MIOX gene structure provided another possible mechanism by which MI levels are regulated in the cells. MIOX catalysis is the first committed step in the glucuronate–xylulose pathway, which is the only pathway for MI metabolism. As the expression of the MIOX gene is ORE and ChREdependent, its expression levels may be regulated by NFAT5 and ChREBP under high glucose conditions [8]. The glucuronate–xylulose pathway can also provide xylulose 5-phosphate, a cellular sensor that activates ChREBP during hyperglycaemia [21]. Therefore, high glucose may not only directly cause increased DNA-binding activity of ChREBP to ChRE, but also the end product, xylulose 5phosphate, provides a positive feedback loop on MIOX expression. In this study, we have demonstrated that high glucose induced the binding activities of NFAT5 and ChREBP to ORE and ChRE in the MIOX gene, respectively, although, these increases were not statistically significant. This trend may suggest that high glucose is involved in the regulation of MIOX gene expression. However, these increases were much higher in patients with nephropathy, and were also accompanied by an increase in MIOX protein levels. These results are similar to our previous studies, which have shown that the DNA-binding activity of NFAT5/NFjB to ORE/jB motif in the promoter region of AKR1B1 [14,15] is increased in PBMCs from patients with T1D and nephropathy, compared to D-glucose

B. Yang et al. / International Journal of Diabetes Mellitus 2 (2010) 169–174

the uncomplicated group under high glucose conditions. Furthermore, our previous studies have shown that mRNA and protein levels of AKR1B1 and sorbitol dehydrogenase were increased in PBMCs from patients with T1D and nephropathy [14,22]. These results clearly indicated that the response to uptake and metabolism of D-glucose is different for patients with T1D and with or without nephropathy. The exact mechanisms have still to be elucidated and explored, but genetic associations such as AKR1B1 or glucose transporter 1 (GLUT1) gene polymorphisms, may contribute to the differences in gene expression [23,24]. Interestingly, there are single nucleotide polymorphisms close to the OREs in the MIOX gene. This could be another possible factor affecting MIOX expression levels through altered DNA-binding activity. A recent study conducted in a renal tubule cell line [25] showed an increase in MIOX mRNA and protein levels in the kidney of rats with diabetic nephropathy, whilst also demonstrating that in vitro high glucose significantly increased MIOX secretion in rat renal tubular epithelial cells, suggesting that hyperglycaemia may be a direct cause of the MIOX increase in the kidney. In diabetic complications, increased glucose levels are associated with high sorbitol accumulation in the kidney, nerve, retina, and lens followed by a depletion of MI [2,3,5]. The MI metabolic pathway is a source of xylitol, which would contribute to the increase in AKR1B1 expression during hyperglycaemia. This could establish another positive feedback loop on MIOX expression. Consequently, the application of an ARI could decrease MIOX expression through the possible involvement of sorbitol in the process of the DNA-binding of NFAT5 and ChREBP to the MIOX promoter, as we showed in our study. In the presence of the ARI, sorbinil, the DNA-binding activities of NFAT5 and ChREBP to ORE and ChRE in the MIOX gene as well as MIOX protein levels were decreased and these decreases were especially noticeable in patients with nephropathy. These results support the evidence that an increased flux through the polyol pathway is involved in diabetic complications, and the accumulation of sorbitol produced by the increased activity of AKR1B1 is linked to the expression of MIOX. ARIs have been shown to improve MI uptake under high glucose conditions. This suggests that the depletion of MI results from a glucose-induced decrease in MI uptake, which is associated with increased AKR1B1 enzyme activity [26,27]. Our findings, together with those reported in cell lines and human neutrophils [28] strongly suggest that the increased expression and activity of AKR1B1 is intricately linked with the activation of MIOX. It has been suggested that the inhibitors of MIOX could be of therapeutic value. Several studies have demonstrated the therapeutic potential of inositol administration [29–31], which suggested that regulation of MI levels may represent a useful strategy. Inhibition of MIOX, as the first enzyme in the only known pathway for inositol catabolism, should raise inositol levels in diabetes, and may thus help counteract hyperglycaemia. The absence of any significant detectable changes in MIOX enzyme activity in our subjects is disappointing, and may be due to a number of reasons including the assay itself. Most assays done by other groups have been on homogeneous MIOX enzyme obtained from a centrifuged homogenate of rat or hog kidneys. There is no publication to our knowledge that describes the MIOX assay in PBMCs. Although, our RT-PCR (data not shown) and Western blotting demonstrated that PBMCs expressed high levels of MIOX, MIOX enzyme activity may not be very high in these cases. On the other hand, the assay may not be sensitive and specific enough to detect MIOX activity in the PBMCs extracted from our subjects prepared in our way. In summary, we have shown a trend for high glucose-increased binding activities of NFAT5 and ChREBP accompanied with increased protein levels, particularly in patients with nephropathy. ARI treatment prevented these increases, and this effect was more

173

obvious in the patients with nephropathy, compared to the uncomplicated subjects. Our results suggested that NFAT5 and ChREBP may be involved in the regulation of MIOX expression under high glucose conditions in patients with T1D with nephropathy and that sorbitol may be involved in the process of DNA binding of NFAT5 and ChREBP in the MIOX promoter. Using ARIs may slow down, or prevent the development of diabetic nephropathy by inhibiting both the polyol and glucuronate–xylulose pathways to decrease the production of sorbitol and to restore the MI levels within cells, respectively. Acknowledgements Grant support: This work was supported by Diabetes UK. We would like to thank all staff that work at Diabetes Research Network in Plymouth Campus, UK for organising and collecting patient samples. We would like to thank Dr. Peter Oates (Department of Cardiovascular and Metabolic Diseases, Pfizer Global Research and Development, Groton, CT 06340 USA) for providing aldose reductase inhibitor: sorbinil. References [1] Brownlee M. The pathobiology of diabetic complications: a unifying mechanism. Diabetes 2005;54:1615–25. [2] Greene DA, Chakrabarti S, Lattimer SA, Sima AA. Role of sorbitol accumulation and myo-inositol depletion in paranodal swelling of large myelinated nerve fibers in the insulin-deficient spontaneously diabetic bio-breeding rat. Reversal by insulin replacement, an aldose reductase inhibitor, and myoinositol. J Clin Invest 1987;79:1479–85. [3] Raccah D, Coste T, Cameron NE, Dufayet D, Vague P, Hohman TC. Effect of the aldose reductase inhibitor tolrestat on nerve conduction velocity, Na/K ATPase activity, and polyols in red blood cells, sciatic nerve, kidney cotex, and kidney medulla of diabetic rats. J Diabetes Complications 1998;12:154–62. [4] Cohen AM, Wald H, Popovtzer M, Rosenmann E. Effect of myo-inositol supplementation on the development of renal pathological changes in the Cohen diabetic (type 2) rat. Diabetologia 1995;38:899–905. [5] Del Monte MA, Rabbani R, Diaz TC, Lattimer SA, Nakamura J, Brennan MC, et al. Sorbitol, myo-inositol, and rod outer segment phagocytosis in cultured hRPE cells exposed to glucose. In vitro model of myo-inositol depletion hypothesis of diabetic complications. Diabetes 1991;40:1335–45. [6] Lin LR, Reddy VN, Giblin FJ, Kador PF, Kinoshita H. Polyol accumulation in cultured human lens epithelial cells. Exp Eye Res 1991;52:93–100. [7] Arner RJ, Prabhu KS, Krishnan V, Johnson MC, Reddy CC. Expression of myoinositol oxygenase in tissue susceptible to diabetic complications. Biochem Biophys Res Commun 2006;339:816–20. [8] Nayak B, Xie P, Akagi S, Yang Q, Sun L, Wada J, et al. Modulation of renalspecific oxidoreductase/myo-inositol oxygenase by high glucose ambience. Proc Natl Acad Sci USA 2005;102:17952–7. [9] Arner RJ, Prabhu KS, Thompson JT, Hildenbrandt GR, Liken AD, Reddy CC. Myoinositol oxygenase: molecular cloning and expression of a unique enzyme that oxidizes myo-inositol and D-chiro-inositol. Biochem J 2001;360:313–20. [10] Arner RJ, Prabhu KS, Reddy CC. Molecular cloning, expression, and characterization of myo-inositol oxygenase from mouse, rat, and human kidney. Biochem Biophys Res Commun 2004;324:1386–92. [11] Yang Q, Dixit B, Wada J, Tian Y, Wallner EI, et al. Identification of a renalspecific oxido-reductase in newborn diabetic mice. Proc Natl Acad Sci USA 2000;97:9896–901. [12] Danesh FR, Wada J, Wallner EI, Sahai A, Srivastva SK, Kanwar YS. Gene regulation of aldose-aldehyde- and a renal specific oxido reductase (RSOR) in the pathobiology of diabetes mellitus. Curr Med Chem 2003;10:1399–406. [13] Prabhu KS, Arner RJ, Vunta H, Reddy CC. Up-regulation of human myo-inositol oxygenase by hyperosmotic stress in renal proximal tubular epithelial cells. J Biol Chem 2005;280:19895–901. [14] Yang B, Hodgkinson AD, Oates P, Kwon HM, Millward BA, Demaine AG. Elevated activity of transcription factor nuclear factor of activated T-cells 5 (NFAT5) and diabetic nephropathy. Diabetes 2006;55:1450–5. [15] Yang B, Hodgkinson AD, Oates P, Millward BA, Demaine AG. High glucose induction of DNA binding activity of the transcription factor NFjB in patients with diabetic nephropathy. Biochim Biophys Acta 2008;1782:295–302. [16] The Expert Committee on the diagnosis and classification of diabetes mellitus: Report of the Expert Committee on the diagnosis and classification of diabetes mellitus, Diabetes Care 2003;26:S5–20. [17] Heesom AE, Hibberd ML, Millward BA, Demaine AG. Polymorphism in the 5’end of the aldose reductase gene is strongly associated with the development of diabetic nephropathy in type 1 diabetes. Diabetes 1997;46:287–91. [18] Reddy CC, Swan JS, Hamilton GA. Myo-inositol oxygenase from hog kidney. I. purification and characterization of the oxygenase and of an enzyme complex

174

[19]

[20]

[21]

[22]

[23]

[24]

[25]

B. Yang et al. / International Journal of Diabetes Mellitus 2 (2010) 169–174 containing the oxygenase and D-glucuronate reductase. J Biol Chem 1981;256:8510–8. Reddy CC, Pierzchala PA, Hamilton GA. Myo-inositol oxygenase from hog kidney. II. Catalytic properties of the homogeneous enzyme. J Biol Chem 1981;256:8519–24. Haneda M, Kikkawa R, Arimura T, Ebata K, Togawa M, Maeda S, et al. Glucose inhibits myo-inositol uptake and reduces myo-inositol content in cultured rat glomerular mesangial cells. Metabolism 1990;39:40–5. Kabashima T, Kawaguchi T, Wadzinski BE, Uyeda K. Xylulose 5-phosphate mediates glucose-induced lipogenesis by xylulose 5-phosphate-activated protein phosphatase in rat liver. Proc Natl Acad Sci USA 2003;100:5107–12. Hodgkinson AD, Sondergaard KL, Yang B, Cross DF, Millward BA, Demaine AG. Aldose reductase expression is induced by hyperglycaemia in diabetic nephropathy. Kidney Int 2001;60:211–8. Yang B, Millward A, Demaine A. Functional differences between the susceptibility Z-2/C-106 and protective Z+2/T-106 promoter region polymorphisms of the aldose reductase gene may account for the association with diabetic microvascular complications. Biochim Biophys Acta 2003;1639:1–7. Hodgkinson AD, Page T, Millward BA, Demaine AG. A novel polymorphism in the 5’ flanking region of the glucose transporter (GLUT1) gene is strongly associated with diabetic nephropathy in patients with type 1 diabetes mellitus. J Diabetes Complications 2005;19:65–9. Lu Y, Liu C, Miao X, Xu K, Wu X, Liu C. Increased expression of myo-inositol oxygenase is involved in the tubulointerstitial injury of diabetic nephropathy. Exp Clin Endocrinol Diabetes 2009;117:257–65.

[26] Beyer-Mears A, Diecke FP, Mistry K, Cruz E. Comparison of the effects of Zopolrestat and sorbinil on lens myo-inositol influx. Pharmacology 1997;54:76–83. [27] Whiting PH, Palmano KP, Hawthorne JN. Enzymes of my-inositol and inositol lipid metabolism in rats with streptozotocin-induced diabetes. Biochem J 1979;179:549–53. [28] Suzuki K, Kawamura T, Sakakibara F, Sasaki H, Sano T, Sakamoto N, et al. Effect of aldose reductase inhibitors on glucose-induced changes in sorbitol and myo-inositol metabolism in human neutrophils. Diabetes Med 1999;16:67–73. [29] Nascimento NR, Lessa LM, Kerntopf MR, Sousa CM, Alves RS, Queiroz MG, et al. Inositols prevent and reverse endothelial dysfunction in diabetic rat and rabbit vasculature metabolically and by scavenging superoxide. Proc Natl Acad Sci USA 2006;103:218–23. [30] Ortmeyer HK, Huang LC, Zhang L, Hansen BC, Larner J. Chiroinositol deficiency and insulin resistance. II. Acute effects of D-chiroinositol administration in streptozotocin-diabetic rats, normal rats given a glucose load, and spontaneously insulin-resistant rhesus monkeys. Endocrinology 1993;132:646–51. [31] Brautigan DL, Brown M, Grindrod S, Chinigo G, Kruszewski A, Lukasik SM, et al. Allosteric activation of protein phosphatase 2C by D-chiro-inositolgalactosamine, a putative mediator mimetic of insulin action. Biochemistry 2005;44:11067–73.

International Journal of Diabetes Mellitus 2 (2010) 175–178

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

Early metabolic imprinting as a determinant of childhood obesity C. Scerri a, C. Savona-Ventura b,⇑ a b

School Medical Services, Department of Health, Malta Department of Obstetrics and Gynaecology, University of Malta Medical School, Malta

a r t i c l e

i n f o

Article history: Received 24 July 2010 Accepted 28 August 2010

Keywords: Birth weight Intrauterine nutrition Infant feeding Childhood obesity

a b s t r a c t Aims: Childhood obesity has seen an alarming increase in recent decades. This study was designed to assess the role of family history and perinatal programming in the aetiology of childhood obesity in a population known to have a high risk of developing metabolic syndrome. Methodology: The study was carried out among two study populations of children. The first was a population of 206 mixed-gender 5-year-old children; the second of 230 mixed-gender 9-year-old children. The children underwent standard anthropomorphic measurements that were correlated to family history of metabolic syndrome-related illness, the child’s birth weight and a history of breastfeeding in early infant life. Results: No statistically significant correlation was noted with a family history of metabolic syndrome; but a definite (P = 0.04) negative correlation was noted with breastfeeding in the 5-year-old children. Children of low birth weight appeared to retain a lower body weight at five years of age than their higher birth weight counterparts (P = 0.002). The pattern changed to suggest a U-shaped distribution of obesity among the various birth weight groups of children, though statistical significance was noted only for the macrosomic group (P = 0.002). Conclusions: The study confirms the importance of intrauterine and early infant nutrition towards the development of childhood and later obesity. Children of low or high birth weight should be considered at risk and parents are advised actively regarding health lifestyle and nutrition options. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction A number of population studies have suggested that the developed world is currently seeing an increase in childhood obesity which has reached alarming levels. The reasons for this changing epidemiology are multifactorial. While genetic defects have been linked to syndromatic and monogenetic obesity syndromes, altered nutritional and social habits are strong contributors to the increase in weight gain during childhood [1]. A further contributor to childhood obesity appears to be perinatal metabolic programming or imprinting, whereby intrauterine and early postnatal nutrition modulates the risk for obesity and the metabolic syndrome later on in life [2,3]. The Maltese population is a relatively insular, small central Mediterranean island population that has been identified as having a high prevalence of Type 2 diabetes mellitus and associated metabolic syndrome co-morbidities. An association between this high metabolic syndrome prevalence to ‘‘endogenous teratogenesis” in this population has been demonstrated in a number of studies

⇑ Corresponding author. Address: Department of Obstetrics and Gynaecology, University of Malta Medical School, Tal-Qroqq, Msida, Malta. Tel.: +356 21435396. E-mail address: [email protected] (C. Savona-Ventura).

[4]. Childhood obesity in this population, in conformity to what is occurring in the developed world, has also shown an increase over the last decades. The 2001–02 Health Behaviour in Schoolchildren Study (HBSC) has shown that the overweight-obesity prevalence among Maltese children stood at 33.3% (overweight 25.4%; obese 7.9%) [5]. The present study attempts to investigate the role of family history and perinatal feeding in the aetiology of childhood obesity in this high risk population. 2. Methods The study was conducted among two groups of children aged five and nine years, recruited from six selected schools covering the different regions in Malta. The 5-year-old study group included 206 individuals of different genders; while the 9-year-old study group included 230 individuals. The children were assessed for obesity with standard anthropomorphic measurements including body weight, standing height and waist circumference. BMI definitions for overweight-obesity and waist circumference centiles expected at the two chosen age groups according to gender were adapted from the literature. The BMI cutoff values for overweight 5-year-old boys and girls were considered to be 17.42 and 17.15 kg/m2, respectively; for 9-year olds, the figures were 19.10

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.08.003

176

C. Scerri, C. Savona-Ventura / International Journal of Diabetes Mellitus 2 (2010) 175–178

Table 1 Family history and early infant feeding determinants by BMI. Body mass index

5-year-old group

9-year-old group

Males and females

Lean

Overweight – Obese

Lean

Overweight – Obese

Family history of metabolic disease No Yes

116 80.6% 28 19.4%

46 74.2% 16 25.8% P = 0.40 ns

68 70.8% 28 29.2%

89 66.4% 45 33.6% P = 0.57 ns

Breastfeeding No Yes

51 36.7% 88 63.3%

33 53.2% 29 46.8% P = 0.04 sig

29 30.2% 67 69.8%

29 33.0% 59 67.0% P = 0.81 ns

*Chi-square statistical 2  2 test.

and 19.07 kg/m2, respectively. The 75th percentile waist circumference measurements for 5-year-old boys and girls were 56.5 cm for both genders; for 9-year-old, the figures were 67.0 and 65.7 cm, respectively [6,7]. The anthropomorphic examination was followed by a parent-administered standard questionnaire that included specific questions related to a family history of metabolic syndromerelated illness (hypertension and/or diabetes mellitus), the child’s birth weight and a history of breastfeeding in early infant life. These parameters were statistically correlated to present obesity status, using the chi-square test and student t-test as appropriate. Ethical approval for the study was obtained from the University of Malta Medical School Research Ethics Committee.

Table 2 Mean anthropomorphic values at 5-years of age by birth weight. Birth weight (g) Males and females

<2500

2500–3999

P4000

BMI (kg/m2)

16.1 ± 2.0 [21] P = 0.24 ns

16.7 ± 2.2 [161]

17.5 ± 2.7 [13] P = 0.22 ns

11 68.7% 5 31.3% *P = 0.44 ns 18.1 ± 3.5 [21] P = 0.002 sig 52.7 ± 3.7 [20] P = 0.21 ns

104 80.6% 25 19.4%

2 8.0% 23 92.0% *P < 0.0001sig 22.7 ± 4.8 [13] P = 0.08 ns 55.5 ± 5.4 [13] P = 0.69 ns

BMI (kg/m2) Lean Overweight/obese Body weight (kg)

Waist circumference (cm)

3. Results Waist circumference (cm) <75th centile >75th centile Height (m)

15 78.9% 4 21.1% *P = 0.72 ns 1.1 ± 0.1 [21] P = 1.0 ns

54.7 ± 7.0 [161]

116 72.0% 45 28.0% 1.1 ± 0.05 [161]

10 76.9% 3 23.1% *P = 0.96 ns 1.1 ± 0.05 [13] P = 1.0 ns

*Chi-square statistical 2  2 tests comparing to BW 2500–3999 group. Student statistical tests compared to group with birth weight 2500–3999 g.

At nine years of age, there appeared to be an increase in the anthropomorphic measurements in infants born with a weight of 4000 g or more, when compared to infants born with a weight of 2500–3999 g. Statistical significance in the mean body weight was shown for the higher birth weight group (Fig. 1). The mean measurement of BMI and waist circumference in children born with a birth weight of P4000 g was also statistically higher.

50 <2500 g

Mean Body Weight [kg]

The prevalence of childhood overweight-obesity in Maltese 5-year-old children based on the cut-off points defined in the literature was 28.8% for boys and 32.7% for girls. There was no statistically significant difference in the BMI distribution between the genders (P = 0.70). These proportions increased markedly with increasing age, so that 48.9% of Maltese 9-year-old males and 45.1% of girls were found to be overweight-obese. Again, there was no statistically significant difference between the genders (P = 0.39). A family history of metabolic disease, as defined by a history of hypertension or diabetes mellitus in either or both of the parents, did not appear to correlate with an increased risk of childhood obesity at both age groups, though the rate of metabolic abnormalities in the parents did show a non-statistically significant increase in the overweight-obese groups of students. The rate of breastfeeding in the overweight-obese 5-year-old group of children showed a statistically significant low rate when compared to the lean 5-yearold children (P = 0.04). While the observation persisted in the 9year-old children, the differences were not statistically significant in this age group (P = 0.81) (Table 1). It would appear that there is a progressive increase in anthropomorphic mean measures determined by weight at birth when assessed in children now aged 5 years. However the large majority of these differences did not exhibit a statistical significance. The mean body weight of children born with a birth weight under 2500 g, however, was statistically (P = 0.002) lower than that registered for infants born with a weight of 2500–3999 g. The proportion of overweight/obese children in those born with a birth weight P4000 g was statistically higher than in those born with a weight of 2500–3999 g (P < 0.0001). There were no statistically significant differences between birth weight and anthropomorphic characteristics when BMI and waist circumference for the combined male and female population were classified according to the defined parameters (Table 2).

20.8 ± 3.7 [162]

40

p=0.38

p=0.004

2500-3999 g >=4000 g

30 p=0.002

p=0.08

20 10 0 5-yrs-old

9-yrs-old

Fig. 1. Mean body weight by birth weight in the two age groups.

C. Scerri, C. Savona-Ventura / International Journal of Diabetes Mellitus 2 (2010) 175–178 Table 3 Mean anthropomorphic values at 9-years of age by birth weight. Birth weight (g) Males and females

<2500

2500–3999

P4000

BMI [kg/m2)

21.3 ± 5.3 [9] P = 0.14 ns

19.2 ± 4.0 [135]

22.5 ± 4.7 [18] P = 0.002 sig

2 25.0% 6 75.0% *P = 0.03 sig 38.1 ± 13.3 [9] P = 0.38 ns 67.2 ± 11.1 [9] P = 0.49 ns

75 69.4% 33 30.6%

4 22.2% 14 77.8% *P < 0.0001 sig 42.0 ± 9.0 [18] P = 0.004 sig 71.6 ± 8.1 [18] P = 0.01 sig

BMI (kg/m2) Lean Overweight/obese Body weight (kg)

Waist circumference (cm)

Waist circumference (cm) <75th centile 75th centile Height (m)

3 33.3% 6 66.7% P = 0.15 ns 1.3 ± 0.1 [9] P = 1.0 ns

35.2 ± 9.2 [135] 64.6 ± 10.9 [134]

86 64.2% 48 35.8% 1.3 ± 0.1 [135]

13 41.9% 18 58.1% *P = 0.04 sig 1.4 ± 0.1 [18] P = 0.0001 sig

*Chi-square statistical 2  2 tests comparing to BW 2500–3999 group. Student statistical tests compared to group with birth weight 2500–3999 g.

Similar statistically significant observations can be made when the data are grouped according to the literature-defined parameters for BMI and waist circumference centiles. Children with high birth weight also appear to be marginally taller. The proportion of overweight/obese children in those born with a birth weight <2500 g was also statistically higher than in those born with a weight of 2500–3999 g (P = 0.03). A higher BMI, body weight and waist circumference are also noted in infants with a low (<2500 g) but the differences did not reach statistical significance (Table 3). 4. Conclusions Childhood obesity and the consequences of a life-long exposure to the obese state have become a global concern, particularly in the developed world. A steady increase in body weight has been noted in the last decades in many developed countries. This observation has also been made for the Maltese population, and the 2001–02 Health Behaviour in Schoolchildren Study (HBSC) has shown that the overweight-obesity prevalence among Maltese children was 33.3% (overweight 25.4%; obese 7.9%) [5]. The present study suggests that the prevalence of overweight-obesity in 9-year-old Maltese children has now reached a rate of 45.1–48.9% depending on gender. Childhood obesity has been shown to confer long-term effects on mortality and morbitity. Children who have a predisposition to develop obesity have been shown to have pre-existing ‘‘nature” or ‘‘nurture” contributors. The identification of these factors, and their relative importance in a particular population, would serve to identify children at risk, and hence promote targeted lifestyle interventions early in life, to prevent the persistence of the metabolic state into adult life with its attendant morbidities. In the present study, when a family history defined by a history of hypertension or diabetes mellitus in either or both parents was correlated against childhood obesity in both the age groups studied, an increased tendency for a positive association was found, however without showing statistical significance. This inter-relationship confirms previous studies [8,9]. Glowinska et al. reported that in a series of patients from referral clinics, approximately one third of obese children were found to have a positive family history of cardiovascular disease (CVD) (defined as CVD, myocardial infarction, stroke, or recognized CVD risk factors, including obesity,

177

hypertension, and diabetes). Similarly, Robinson et al reported that a family history of hypertension was associated with higher child body mass index. In this present study, it can be argued that since the parents of the children were still of a relatively young age group, full-blown clinically identified metabolic disease may not yet have set in. It is also possible that not all the parents interviewed in this study were properly aware of their health status and a further percentage of these were actually suffering from elements of insulin resistance, and were still not aware of it. Followup interviews 10–15 years later may show a higher prevalence of metabolic disorders in the parents. The study protocol could have been better designed to include formal examination and investigations of the parents to assess for features of metabolic disease, rather than rely on self-filled questionnaires. Early infant nutrition has also been suggested as playing a role in the development of adult-onset metabolic disease. The extent and duration of breastfeeding have been previously reported to be inversely associated with the risk of obesity in later childhood; the relationship being possibly mediated by physiologic factors in human milk as well as by the feeding and parenting patterns associated with nursing [10,11]. A similar association was found in this current study. In the 5-year-old population studied, a significant negative relationship was found between these children, and a history of breastfeeding. Thus, breastfed children showed a lower prevalence of overweight-obesity, as compared to those who were bottle-fed. This relationship was also evident in the 9-year-old population, but here a statistical significance was not reached. The loss of statistical strength may be possibly attributed by the introduction of other influencing environmental factors, such as diet and lack of physical exercise, with increasing age of the child. These factors by 9 years of age would cancel out and modify the metabolic imprinting of breast feeding. The present study has shown that breastfed children have a reduced risk of becoming overweight-obese. This confirms previous studies [10,11]. Breastfeeding intrinsically allows a baby to set its appetite regulatory pathways in such a way as to limit the propensity to overeat [12]. A number of epidemiological studies have identified that intrauterine nutrition can determine the development of metabolic syndrome in adult life – foetal origins of adult-onset disease theory. These observations have been emulated in the Maltese population [4]. The present study has shown an inter-relationship between birth weight and the risk of developing overweight-obesity in childhood. The macrosomic infant born with a birth weight of 4000 g or more has been shown to have higher mean anthropomorphic measurements both at 5 years and 9 years of age when compared to the corresponding children born with a birth weight of 2500–3999 g. Statistical significance was only shown at 9 years of age; though the proportion of overweight/obese children in the high birth rate group was statistically higher than in those children born with a birth weight of 2500–3999 g. The low birth weight (<2500 g) individuals showed lower mean anthropomorphic measurements at 5 years of age with statistical significance being shown for mean body weight. At 9 years of age, there was a non-statistical significant rise in mean anthropomorphic measurement. The proportion of overweight/obese individuals in those born with a low birth weight was statistically higher than in those born with a birth weight of 2500–3999 g. It would thus appear that from birth to 5 years, these low birth weight children are still passing through their ‘catch up growth period’, correcting for the effects of their in utero period of limited nutritional resources. By 9 years of age, these children have caught up and passed through their ‘catch up growth period’, so that their anthropomorphic parameters now lie in the overweight-obese range. During and following the ‘catch up growth’ the children are exposed to a practically limitless supply of calories. The present findings observed in the 9-year-old population support the

178

C. Scerri, C. Savona-Ventura / International Journal of Diabetes Mellitus 2 (2010) 175–178

previously reported ‘U’ odds-risk pattern described in the thrifty phenotype hypothesis of obesity. The thrifty phenotype hypothesis postulates that poor nutrition in foetal life is detrimental to the development and functioning of b-cells and insulin-sensitive tissues, resulting in the emergence of insulin resistance or metabolic syndrome later in life [12]. A similar U-shaped inter-relationship has been described in the Maltese population, in relation to the risk of developing gestational diabetes [13]. The findings engendered by the present study indicate that environmental factors, including intrauterine and postnatal nutrition, strongly influence the risk of eventual obesity development. The family physician must use all the opportunities presented to him during the antenatal period and in the postpartum period to identify those infants particularly at risk of becoming overweight-obese in childhood. Early educational interventions with the parents must be made to encourage breast feeding, and to introduce healthy nutrition and lifestyle practices. Prevention and early identification are the key to decreasing the prevalence of obesity in childhood. References [1] Koletzko B, Girardet JP, Klish W, Tabacco O. Obesity in children and adolescents worldwide: current views on future directions – working group report of the first congress of pediatric gastroenterology, hepatology, and nutrition. J Pediatr Gastroenterol Nutr 2002;35(Suppl.2):S205–12. [2] Barker DJ. In utero programming of chronic disease. Clin Sci (Lond) 1998;95(2):115–28.

[3] Waterland RA, Garza C. Potential mechanisms of metabolic imprinting that leads to chronic disease. Am J Clin Nutr 1999;69(2):179–97. [4] Savona-Ventura C. The thrifty diet phenotype – a case for endogenous physiological teratogenesis. In: Engels JV, editor. Birth Defects: New Research. USA: Nova Science Publishers; 2006. p. 183–200. [5] Janssen I, Katzmarzyk PT, Boyce WF, Vereecken C, Mulvihill C, Roberts C, et al. Health behaviour in school-aged children obesity working group. Comparison of overweight and obesity prevalence in school-aged youth from 34 countries and their relationships with physical activity and dietary patterns. Obes Rev 2005;6(2):126–32. [6] Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000;320(7244):1240–3. [7] Fernández JR, Redden DT, Pietrobelli A, Allison DB. Waist circumference percentiles in nationally representative samples of African-American, European-American, and Mexican-American children and adolescents. J Pediatr 2004;145(4):439–44. [8] Glowinska B, Urban M, Koput A. Cardiovascular risk factors in children with obesity, hypertension and diabetes: lipoprotein(a) levels and body mass index correlate with family history of cardiovascular disease. Eur J Pediatr 2002;161(10):511–8. [9] Robinson RF, Batisky DL, Hayes JR, Nahata MC, Mahan JD. Body mass index in primary and secondary pediatric hypertension. Pediatr Nephrol 2004;19(12):1379–84. [10] Buttigieg A. A study of childhood obesity and associated factors. Thesis – Master of Science in Public Health Medicine, Malta: University of Malta; 1997. [11] von Kries R, Koletzko B, Sauerwald T, von Mutius E, Barnert D, Grunert V, et al. Breast feeding and obesity: cross sectional study. BMJ 1999;319(7203): 147–50. [12] Miller J, Rosenbloom A, Silverstein J. Childhood obesity. J Clin Endocrinol Metab 2004;89(9):4211–8. [13] Savona-Ventura C, Chircop M. Birth weight influence on the subsequent development of gestational diabetes mellitus. Acta Diabetol 2003;40(2): 101–4.

International Journal of Diabetes Mellitus 2 (2010) 196–198

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Case Report

Spontaneous gas gangrene of the scrotum in patient with severe diabetic ketoacidosis Chu Zhang, Lizhen Ma, Fengying Peng, Yin Wu, Yu Chen, Yuhong Zhan, Xianfeng Zhang ⇑ Department of Endocrinology and Metabolism, Hangzhou Hospital, Nanjing Medical University, China

a r t i c l e

i n f o

Article history: Received 26 September 2010 Accepted 30 September 2010

Keywords: Diabetic ketoacidosis Ketosis-prone T2DM Fournier’s gangrene Clostridium perfringens

a b s t r a c t A 52-year-old Chinese man presented with severe diabetic ketoacidosis and markedly inflated scrotum. Computed tomographic scans of the lower pelvis show extensive gas accumulation and inflammatory changes in both sides of the scrotum. Gas gangrene of the scrotum was diagnosed and radical debridement along with other proactive anti-ketoacidosis therapy was performed immediately. Clostridium perfringens was found in cultures of necrotic tissue that verified the diagnosis and the patient was cured through multiple proactive treatments. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction Spontaneous gas gangrene of the scrotum, also known as Fournier’s gangrene (FG) is an extremely rare but life-threatening skin and soft tissue infective disease in the perineal region [1]. The infection commonly starts as cellulitis adjacent to the portal of entry, and rapidly progresses to extensive tissue necrosis. Without aggressive treatment, the patient will die from sepsis and multiple organ failure [2]. Multiple microbial infections by aerobes and anaerobes are always found in cultures from the wounds, most of which are normal commensals in the perineum and genitalia. Because of the impaired host cellular immunity, these conditional pathogens become virulent, and act synergistically to invade tissue and cause extensive damage [3]. Some comorbid systemic disorders with cellular immunity impairment, including AIDS, diabetes, alcohol abuse, leukemia, chemotherapy and chronic corticosteroid use [4] are identified in patients with FG. Diabetic ketoacidosis (DKA) is an acute complication of diabetes, characterized by circulation failure and acidosis. In developing countries, even with improved healthcare systems and reliable insulin supply, mortality and morbidity from DKA remain high

Abbreviations: DKA, diabetic ketoacidosis; WBCs, white blood cells; DM, diabetes mellitus; LADA, late onset autoimmune diabetes in adult; MRI, magnetic resonance imaging; CT, computed tomography. ⇑ Corresponding author. Address: Department of Endocrinology and Metabolism, Hangzhou Hospital, Nanjing Medical University, Xueshi Road 4[#], Hangzhou City, Zhejiang Province 310006, China. Tel.: +86 0571 87065701. E-mail address: [email protected] (X. Zhang).

[5]. Although impaired tissue perfusion and defective immune response presented in DKA could be predisposing conditions for FG, there has been no previous report of specifically pinpointing the DKA with FG. In the present case, we show how a patient with severe DKA and FG recovered through prompt, accurate diagnosis and emergent intervention.

2. Case presentation A 52-year-old previously healthy man was hospitalized on account of poor health. He mainly complained about polydipsia, fever and shortness of breath for 9 days, and a swollen scrotum was found 4 days later. Fever (always) occurred after rigors, and his maximum temperature reached 38.8 °C. The swollen scrotum was accompanied by urodynia, hematuria and dysuria. He had not previously been diagnosed as diabetic, but his father had had type 2 diabetes (T2DM) for 34 years. Upon admission, he was in confusion, with a body temperature of 38.6 °C, systolic blood pressure at 145 mm Hg, pulse at 96/min and breath rate at 24/min. His height was 174 cm, his body weight 82 kg and BMI 27. His skin was dry, and his extremities were cold. The scrotum was found to be swollen almost as big as a football, and was of a dark red colour; there was no tenderness but crepitation could be heard on touching. Laboratory examination revealed DKA and leucocytosis; plasma sugar was 16.95 mmol/L, blood WBCs was 38.2  109/L, plasma beta-hydroxybutyrate was 1898 mmol/L, urine acetone was 3+, urine lencocyte count was 20 per high power lens, pH was 7.06, HCO3 was 4 mmol/L, BE was 28 mmol/L, GAD-Ab, ICA and IAA

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.09.004

C. Zhang et al. / International Journal of Diabetes Mellitus 2 (2010) 196–198

were negative. Ultrasonography showed a filled cyst with hydronephrosis in both sides. A Lower Pelvis CT scan revealed extensive gas accumulation within the scrotum and inflammatory changes to the scrotum wall (Fig. 1). DKA and spontaneous gas gangrene of the scrotum was diagnosed, conventional DKA treatment with fluid replacement and continual insulin infusion were given, 6000 ml isotonic saline water was given in first 24 h, thorough debridement and drainage were performed immediately, the wound surface was thoroughly rinsed with hydrogen peroxide, Tazocin and clindamycin were given as empirical antibiotic therapy. Cultures of necrotic tissue and urine from a urethral catheter showed Clostridium perfringens, and an antibiotic sensitivity test justified the empirical antibiotic choice. The antibiotic sensitivity test also showed that the patient was piperacillin, cefoperazone, clindamycin, Doxycycline, ofloxacin, and metronidazole sensitive. Patient recovered consciousness in first day of admission, temperature dropped back to normal range on the 4th day. He was given tazocin and clindamycin for 4 weeks. When he was discharged from the hospital 4 weeks later, his blood glucose was well controlled through insulin, and the infected scrotum was completely healed. When the patient went back clinic for a routine visit 6 months later, we discontinued his insulin therapy and solely used metformin; his glucose was well controlled thereafter. 3. Discussion FG is a rare, fulminant form of infective necrotizing fasciitis of the perineal, genital, or perianal regions, characterized by rapid progression of necrotising inflammation and by the violent life-threatening course of the disease. Without appropriate treatment, the patient soon dies after septic shock and multiorgan failure [2]. DM is reported to be present in 20–70% of patients with Fournier’s gangrene [6], and the mortality rate of Fournier’s gangrene is higher in patients with DM [7], as increased tissue glucose, poor tissue perfusion, and weak immune response in DM lead to a faster progress of infection and rapid development of systemic sepsis. DKA is an acute complication of DM, characterized by haemodynamic instability and acidosis, pathophysiological states facilitating FG to occur, and also being promoted by FG. Although such a severe infection like FG could be one factor that causes the initiation and development of DKA, these are often seen in patients with type 1 DM. The patient in the present case was in his fifties, with

197

negative GAD-Ab, ICA and IAA, and his BMI was 27. All these are inconsistent with the diagnosis of type 1 DM or LADA. As severe DKA developed in such a short time, combined with other characteristics of patient, such as the age, BMI and negative autoantibody, we proposed that the patient be diagnosed as ketosis-prone T2DM. Another characteristic of ketosis-prone T2DM is that insulin secretion is markedly impaired at presentation, but intensified diabetic management results in significant improvement in beta-cell function and insulin sensitivity, sufficient to allow discontinuation of insulin therapy within a few months of follow-up [8]. Actually, the patient in the present case switched his therapy from insulin to oral anti-diabetic glucose 6 months later, and this also supported the diagnosis of ketosis-prone type 2 diabetes. FG commonly starts as cellulites adjacent to the portal of entry. For those idiopathic cases, the source of infection can be ascribed to the gastrointestinal and genitourinary tract [9]. In the present case, without evidence of cutaneous injury or perianal abscess, most possibly, it was the infection of the urinary tract that initiated gas gangrene in the scrotum, as urine leucocyte counts was 20 per high power lens and culture of urine from urethral catheter showed Clostridium perfringens, the same pathogen in necrotic tissue. Clostridium perfringens, the organism responsible for classical myonecrotic gas gangrene, is frequently identified along with other bacteria in FG [10]. Under normal conditions, Clostridium species colonize in gastrointestinal and genitourinary tract, and become pathogens when predisposing factors are present, such as compromised immunity, diabetes, or invasive procedures within the urinary tract, gastrointestinal tract, or skin. In the local wounds of patients with FG, they produce various exotoxin and enzymes such as a toxin, collagenase, heparinase, and hyaluronidase, which lead to tissue necrosis, gas accumulation and the spread of infection[11]. Once FG is diagnosed, aggressive debridement of local wounds is warranted, along with other medical intervention like fluid replacement for haemodynamic stabilization, and multi broad spectrum antibiotics, to cover all possible organisms. For patients of DKA and FG, the significance of fluid replacement is even more vital than it is for patient with DKA or FG alone, since both DKA and FG are in favor of circulation failure. The usual antibiotic combination includes penicillin or clindamycin for the streptococcal species, third generation cephalosporin or carbapenems for the gram negative organisms, plus metronidazole for the anaerobes [12]. Hyperbaric oxygen could be used as an adjunctive therapy

Fig. 1. (A) sagittal, (B) coronal, and (C) axial CT scan shows an accumulation of gas in the scrotum (white arrow) and extensive inflammation of scrotum wall.

198

C. Zhang et al. / International Journal of Diabetes Mellitus 2 (2010) 196–198

in the treatment of FG, even though there is no conclusive evidence regarding its effectiveness. In conclusion, DKA with FG is a rare clinical emergency, but mortality on account of the disease is high, and early diagnosis and proactive medical intervention remain critical for patients’ survival. Clinical presentation in many patients during early stage may not be prominent. The detection of gas in the tissue by X-ray, CT, MRI or Ultrasonography should be a warning sign pointing to spontaneous gas gangrene. On the other hand, the proactive management of perineal injury or infection of patients with DKA or diabetes is also of extreme importance, in case FG may develop in these situations. Conflict of interest None. Reference [1] Smith GL, Bunker CB, Dinneen MD. Fournier’s gangrene. Br J Urol 1998;81:347–55.

[2] Pawlowski W, Wronski M, Krasnodebski IW. Fournier’s gangrene. Pol Merkuriusz Lek 2004;17:85–7. [3] Rotstein OD, Pruett TL, Simmons RL. Mechanisms of microbial synergy in polymicrobial surgical infections. Rev Infect Dis 1985;7:151–70. [4] Anaya DA, Dellinger EP. Necrotizing soft-tissue infection: diagnosis and management. Clin Infect Dis 2007;44:705–10. [5] Otieno CF, Kayima JK, Omonge EO, Oyoo GO. Diabetic ketoacidosis: risk factors, mechanisms and management strategies in sub-Saharan Africa: a review. East Afr Med J 2005;82:S197–203. [6] Morpurgo E, Galandiuk S. Fournier’s gangrene. Surg Clin North Am 2002;82:1213–24. [7] Korkut M, Icoz G, Dayangac M, Akgun E, Yeniay L, Erdogan O, et al. Outcome analysis in patients with Fournier’s gangrene: report of 45 cases. Dis Colon Rectum 2003;46:649–52. [8] Umpierrez GE, Smiley D, Kitabchi AE. Narrative review: ketosis-prone type 2 diabetes mellitus. Ann Intern Med 2006;144:350–7. [9] Asci R, Sarikaya S, Buyukalpelli R, Yilmaz AF, Yildiz S. Fournier’s gangrene: risk assessment and enzymatic debridement with lyophilized collagenase application. Eur Urol 1998;34:411–8. [10] Stevens DL. The pathogenesis of clostridial myonecrosis. Int J Med Microbiol 2000;290:497–502. [11] Rood JI. Virulence genes of Clostridium perfringens. Annu Rev Microbiol 1998;52:333–60. [12] Kobayashi S. Fournier’s gangrene. Am J Surg 2008;195:257–8.

International Journal of Diabetes Mellitus 2 (2010) 179–183

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

Association of CD36 gene variants rs1761667 (G > A) and rs1527483 (C > T) with Type 2 diabetes in North Indian population Monisha Banerjee a,⇑,1, Sunaina Gautam a,1, Madhukar Saxena a, Hemant Kumar Bid b,2, C.G. Agrawal c a

Molecular and Human Genetics Laboratory, Department of Zoology, University of Lucknow, Lucknow 226 007, India Department of Endocrinology, Central Drug Research Centre, Lucknow, India c Department of Medicine, Chhatrapati Sahuji Maharaj Medical University, Lucknow, India b

a r t i c l e

i n f o

Article history: Received 28 June 2010 Accepted 28 August 2010

Keywords: Single nucleotide polymorphism Diabetes CD36 Indians Oxidized low density lipoproteins

a b s t r a c t Introduction: Type 2 diabetes mellitus (T2DM) affects huge populations in India and abroad. Genetic polymorphisms (SNPs) in scavenger receptors such as CD36 have been implicated in the pathogenesis of diabetic atherosclerosis and cardiovascular diseases. Since CD36 gene expression contributes to T2DM, we proposed to study the association of two of its polymorphisms. Methods: A population of 400 subjects from North India was analyzed according to clinical parameters. Out of them 150 controls and 250 T2DM patients were genotyped for two SNPs namely rs1761667 (G > A) and rs1527483 (C > T) in the CD36 gene using polymerase chain reaction and restriction fragment length polymorphism (PCR–RFLP) followed by statistical analysis. Results and discussion: No association of rs1527483 (C > T) SNP was observed in T2DM patients. The GA genotype was prevalent in 76.0% diabetic population and a highly significant genotypic association of rs1761667 (G > A) SNP in CD36 gene was observed in them (P = <0.001; OR 3.173; CI 1.098–9.174). Sample characteristics showed a highly significant association with lipid profile (P = <0.001). The ‘GA’ genotype in combination with CC genotype showed a significant association with TC (P = 0.020), LDL (P = 0.005) and VLDL (P = 0.029). In addition, the haplotype analysis CC/GA (/+) and CT/GA (/+) showed a strong association with TC and LDL (P < 0.05). The presence of 31118 A* allele in haplotypes showed strong association with T2DM (P = <0.005). Conclusion: Out of the two CD36 gene polymorphisms tested, rs1761667 SNP is significantly associated with T2DM with the GA heterozygous genotype showing highest frequency among the T2DM patients. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction Type 2 diabetes mellitus (T2DM) is a part of metabolic syndrome (MetS) which is affected both by environmental, as well as genetic factors. Many studies suggest that net lipid accumulation, caused by an imbalance between fatty acid delivery/synthesis and fatty acid oxidation, results in the activation of a serine kinase cascade [1,2]. This in turn, inhibits insulin signaling, resulting in

Abbreviations: CD36, cluster of differentiation 36; oxLDL, oxidized low density lipoproteins; SNPs, single nucleotide polymorphisms; FAT, fatty acid transporter; GP88, glycoprotein 88; EDTA, ethyl diamine tetra acetic acid; NEB, New England Biolabs. ⇑ Corresponding author. E-mail addresses: [email protected], banerjee_m@lkouniv. ac.in (M. Banerjee), [email protected] (S. Gautam), madhukarbio@gmail. com (M. Saxena), [email protected] (H.K. Bid), [email protected] (C.G. Agrawal). 1 These authors contributed equally to this work. 2 Present address: Research Associate, University of Southern California, Norris Comprehensive Cancer Center, Los Angeles, CA 90033, USA. Tel.:+1 323 865 0535.

insulin resistance in liver and skeletal muscle, the organs responsible for majority of glucose disposal. CD36 (FAT, SCARB3, GP88, glycoprotein IV (gpIV) and glycoprotein IIIb (gpIIIb)) is a broadly expressed 88 kDa membrane transporter glycoprotein that acts as a facilitator of fatty acid (FA) uptake, a signaling molecule and a receptor for a wide range of ligands [3]. In addition to FAs, CD36 binds to native lipoproteins and functions in the uptake of cholesteryl esters, facilitates the uptake of oxidized low/high-density lipoproteins and cholesterol [4–8]. As a result of its many ligands and functions, CD36 could impact a variety of conditions linked with the metabolic syndrome, including diabetes, insulin resistance, inflammation and atherosclerosis [9,10]. The CD36 gene spans 36 Kb (7q11.2–7q21.11) and is comprised of 15 alternatively spliced exons that are differentially regulated by several upstream promoters [11,12]; CD36 plays an important role in lipid metabolism and its gene polymorphisms are related to hypertension [13], MetS and high-density lipoprotein cholesterol (HDL-C) [14]. Since there have been very few studies on the association of genetic variants in the CD36 gene with T2DM [15–17], we proposed to investigate two probable CD36 gene

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.08.002

180

M. Banerjee et al. / International Journal of Diabetes Mellitus 2 (2010) 179–183

Table 1 Primer sequences, PCR conditions, amplicon sizes, restriction enzymes with product sizes of SNPs in CD36 gene. SNP SNP (C > T) rs1527483

Primer sequence 0

0

0

0

F: 5 -CGCTACAACAATTTTATAGATTTTGAC-3

25,444 intron 11

R: 5 -TGAAATAAAAATAATCTTGTCGATGA-3

SNP (G > A) rs1761667

F: 50 -CAAAATCACAATCTATTCAAGACCA-30

31118 Promoter 50 region of exon 1A

R: 50 -TTTTGGGAGAAATTCTGAAGAG-30

Annealing temp. (°C)

Product size (bp)

Restriction enzyme/allele sizes

55

252

56

190

Taq I CC 160,70,22 CT 230,160,70,22 TT 230,22 Hha I GG 138, 52 GA 190, 138, 52 AA 190

2. Material and methods

DNA (100–150 ng), 10 pmol of each primer, 200 lM dNTPs, and 05 U of Taq DNA polymerase (MBI-Fermentas, USA) in a gradient Master Cycler (Eppendorf, USA). The PCR products were digested with the respective restriction enzymes, resolved on 2% agarose and 12% polyacrylamide gels.

2.1. Patients and clinical evaluation

2.3. Statistical analysis

Type 2 diabetic patients, 22–77 years of age (n = 250) were enrolled from the outpatient Diabetes Clinic of Chhatrapati Shahuji Maharaj Medical University (CSMMU), Lucknow under the supervision of expert clinicians. Age/sex-matched normal healthy controls (n = 150) were screened from healthy staff members of both Universities. The study was approved by the Medical Ethical Committee of CSMMU and a written informed consent was taken from all subjects enrolled for the study. Controls showing a normal oral glucose tolerance test were included in the study, whereas those having a history of coronary artery disease or other metabolic disorders were excluded. Subjects with fasting glucose concentrations P126 mg/dl or 2-h glucose concentrations P200 mg/dl after a 75g oral glucose tolerance test were categorized in the diabetes group. Medical records of these patients were reviewed to ascertain diabetes-associated complications. A self-administered questionnaire was used to record the clinical history of diabetes, associated complications such as hypertension as well as family history. All patients were on oral hypoglycemic agents to maintain a normal glucose level in their blood. Estimations of plasma glucose (mg/dl), serum insulin (mg/dl) and lipid profile (total serum cholesterol (TC), high-density lipoprotein cholesterol (HDL-C) and serum triglycerides) were done using commercially available Ecoline kits from Merck by double beam spectrophotometer at 550 nm (TGL-C), 510 nm (serum creatinine), 500 nm (total cholesterol) and 560 nm (HDL-C) [18]. Height, weight and waist circumference were measured to calculate body mass index (BMI) and waist hip ratio (WHR). Systolic and diastolic blood pressures were measured in the sitting position with an appropriately sized cuff after a 5 min rest. Clinical details of patients and controls were recorded.

Allele frequencies, genotype frequencies and carriage rates of the alleles in all the groups were compared using 2  2 contingency table by Fisher’s exact test. The Hardy–Weinberg equilibrium at individual loci was assessed by chi-square (v2) statistics using SPSS (v 15.0). Allele frequency was calculated as the number of occurrences of the test allele in the population divided by the total number of alleles. Carriage rate was calculated as the number of individuals carrying at least one copy of test allele divided by the total number of individuals. Association of various clinical parameters with different CD36 (SNPs C/T and G/A) genotypes in T2DM patients and controls was calculated by 2  2 paired t-test. Differences were considered statistically significant for P < 0.05. The strength of association of SNP-SNP combinations was determined by Odds ratio (OR) at 95% confidence interval (CI) using Logistic Regression Analysis (SPSS). Multivariate Logistic regression analysis was used to examine the association of biochemical parameters with different haplotypes of the two SNPs. Data were presented as mean ± SD.

polymorphisms that might have an important role to play in T2DM and related complications. This is perhaps the first report of CD36 gene polymorphism study in T2DM patients from Northern India.

2.2. DNA extraction and CD36 gene polymorphisms Five milliliter of blood sample was taken in EDTA vials from both the groups. Genomic DNA was extracted from peripheral blood leucocytes using the standard salting out method [19]. Two CD36 single nucleotide polymorphisms (SNPs) viz. G > A (rs1761667) in the 31118 promoter region of exon 1A and C > T (rs1527483) in intron 11 were genotyped in 150 controls and 250 T2DM patients using polymerase chain reaction and restriction fragment length polymorphism (PCR–RFLP) method. Primers for the SNPs were designed and restriction enzymes (REs) were identified using the Primer 3 and NEB cutter softwares respectively. Details including the location of SNPs in the CD36 gene, primer sequences and REs with product sizes are presented in Table 1. PCR was performed in a 25 ll reaction mixture containing genomic

3. Results 3.1. Characteristics of the sample population Out of the total number of subjects (n = 400) included in the study, 150 were healthy controls with an average age of 47.07 ± 6.01 years, their fasting and post prandial glucose levels were 109.69 ± 51.96 and 143.69 ± 21.46 mg/dl, respectively. The average age of the patients (n = 250) was 48.39 ± 9.91 years, their fasting and postprandial glucose levels were 166.14 ± 65.76 and 266.40 ± 88.45 mg/dl, respectively. The patients average systolic BP was slightly elevated (135.22 ± 19.35 mm Hg) and the mean diastolic pressure was almost normal (84.27 ± 10.77 mm Hg). Total cholesterol (225.83 ± 34.83 mm/dl) and LDL-C (158.00 ± 37.04 mm Hg) levels were slightly raised and HDL-C was low (43.47 ± 5.26 mm Hg) in patients. Therefore, the lipid profile in patients showed a significant difference when compared to control subjects (P < 0.001). However, no significant difference was observed in BMI, WHR and serum creatinine levels between the T2DM and control groups (P > 0.05) (Table 2). 3.2. Genetic analysis PCR–RFLP results showing the pattern of genotypes for the two SNPs (rs1527483 and rs1761667) in the CD36 gene have been represented in Fig. 1a and b. The allele and genotype frequency distribution and carriage rate of CD36 gene polymorphisms among 150 controls and 250 patients

M. Banerjee et al. / International Journal of Diabetes Mellitus 2 (2010) 179–183

181

Table 2 Sample characteristics for 400 subjects from North Indian population. Sample characteristics

Controls (n = 150)

Patients (n = 250)

P-value

Age (years) Basal metabolic index, BMI (kg/m2) Waist hip ratio, WHR Fasting plasma glucose, F (mg/dl) Post prandial plasma glucose, PP (mg/dl) Blood pressure systolic, SBP (mmHg) Blood pressure diastolic, DBP (mmHg) Total-cholesterol, TC (mg/dl) Triglyceride, TGL (mg/dl) HDL-cholesterol, HDL (mg/dl) LDL-cholesterol, LDL (mg/dl) VLDL-cholesterol, VLDL (mg/dl) Serum creatinine, S. Cret. (mg/dl)

47.07 ± 6.01 23.43 ± 3.83 0.97 ± 0.06 109.69 ± 51.96 143.69 ± 21.46 117.40 ± 5.48 76.80 ± 6.48 171.65 ± 41.02 151.87 ± 64.23 59.07 ± 13.93 82.21 ± 48.94 30.37 ± 12.85 1.08 ± 0.16

48.39 ± 9.91 24.49 ± 4.60 0.92 ± 0.08 166.14 ± 65.76 266.04 ± 88.45 135.22 ± 19.35 84.27 ± 10.77 225.83 ± 34.83 112.04 ± 16.88 43.47 ± 5.26 158.00 ± 37.04 22.44 ± 3.56 1.04 ± 0.09

0.064 0.701 1.000 0.216 0.313 0.454 0.846 <0.001 <0.001 <0.001 <0.001 <0.001 0.912

bp

M1

CT

CC

TT

M2

AA

230 160 70

(a) bp

GA

GG

190 138

52

(b) Fig. 1. (a) Ethidium bromide stained agarose gel of SNP (C > T) rs1527483 showing different genotypes; CT: 230, 160, 70 bp; CC: 160, 70; TT: 230; M1: 100 bp ladder (b) silver stain polyacrylamide gel of SNP (G > A) rs1761667; GA: 190, 138, 52; GG: 138, 52; AA: 190; M2: 50 bp ladder.

are shown in Table 3. In case of both SNPs, the allele and genotype frequencies in control and patient groups were in Hardy–Weinberg Equilibrium (HWE). The minor allele frequencies (MAF) for both SNPs in our analysis were P0.1 (Table 3). For C/T polymorphism, no allelic and genotypic associations were observed in the present study (P > 0.05). Most subjects were homozygous wild type (CC), while the TT genotype was rare in the study population (TT genotype frequency 6 0.01). Allele frequencies of C and T were almost same in controls and T2DM patients (Table 3). In case of the G/A promoter polymorphism, although no allelic association was observed (P = 0.244), the frequency of the GA genotype was significantly higher in T2DM (76.0%) when compared to controls (49.0%, P = <0.001). SNP (G > A) genotypes showed a highly significant association (P < 0.001; OR 3.173; CI 1.098–9.174) with T2DM when compared to controls (Table 3). Haplotype analysis using logistic regression analysis showed the possible effect of two genotypes, i.e. double combinations of SNPs (C > T) and (G > A) on the risk of developing T2DM. Out of

the nine possible haplotypes, we obtained only six with frequencies ranging from 6% to 37% in this population (Table 4). The frequency of recessive genotypes (AA and TT) was very low in the population, so their combination with other genotypes was not considered. Results showed significant association in case of CC/GG and CC/ GA haplotypes (P = <0.05). The risk of CC/AA (+/+, +/) also showed significant association (P = <0.05) while the risk of CT genotype with GG/GA was highly significant (P = <0.05) (Table 4). The risk of GA and GG without CT genotype was 0.3–2.6 times higher (P = <0.001), (OR; 0.290 and 2.597; CI 95%; 0.178–0.474 and 1.567–4.303). The risk of diabetes in the combination CT/GA (/ ) also showed significant association (P = 0.025). Analysis of the association of haplotypes (double combinations of genotypes) with biochemical parameters using multivariate logistic regression brought forth some interesting results. We found that the subjects without CC/GG genotypes (/) showed significant association with TGL and VLDL (P = 0.006). In case of CC/GG (/+) subjects significant association was observed with HDL, LDL and S. Cret. (P = <0.05). Subjects with CC/GA genotypes (+/+) showed significant association with TC (P = 0.020), LDL (P = 0.005) and VLDL (P = 0.029) and subjects without GA showed significant association with HDL (P = 0.003) and LDL (P = 0.005). Subjects with haplotype CC/GA (/+) showed significant association with TC (P = 0.005) and LDL (P = <0.001) while CC/GA (/) with TC (P = 0.007), LDL (P = 0.001) and S. Cret. (P = 0.009). Subjects with CC genotype in combination with AA (+/+) also showed significant association with TC and LDL (P = 0.003). Subjects having CC without AA genotype (+/) and CT with GG (+/+) showed significant association with HDL (P = 0.034, 0.036). In case of CT/GG (+/ +) subjects significant association was observed with LDL (P = 0.001), but CT/GG (+/) and CT/GA (+/) showed significant association with TGL and VLDL (P = 0.006) and TC, LDL, S. Cret. (P = 0.007, 0.001, 0.009) respectively. In case of CT/GA (/+) and (/) genotypes, TC, LDL, VLDL (P = <0.05) and S. Cret. (P = <0.001) showed significant association, respectively (Table 5). Other haplotypes such as CT/AA, TT/GG etc. were not found in our population. 4. Discussion The role of CD36 in lipid metabolism, metabolic syndrome and T2DM prompted us to investigate its SNP association with T2DM in our North Indian population. In the present study, we demonstrated that the GA heterozygous genotype of a promoter polymorphism (rs1761667) in the CD36 gene was more prevalent in the T2DM patients and showed association with diabetes in the North Indian population. A similar association of a promoter polymorphism (rs1527479) in the CD36 gene with insulin resistance and

182

M. Banerjee et al. / International Journal of Diabetes Mellitus 2 (2010) 179–183

Table 3 Allele, genotype and carriage rate frequencies (F) of SNPs and their association status with T2DM. SNPs

Allele frequency

Genotype frequency

Carriage rate

Association

Controls F (*n)

T2DM patients F (*n)

Controls F (**n)

T2DM patients F (**n)

Controls F(***n)

Patients F(***n)

Allele (df = 1)

Genotype (df = 2)

Carrier rate (df = 1)

SNP (C > T) rs1527483

C = 0.89 (268)

C = 0.88 (439)

CC = 0.76 (190) CT = 0.24 (59)

C = 0.99 (148)

C = 0.99 (249)

v2 = 0.429

v 2 = 2.331

v 2 = 0.493

Intron 11

T = 0.11 (32)

T = 0.12 (61)

CC = 0.80 (120) CT = 0.19 (28) TT = 0.01 (02)

TT = 0.00 (01)

T = 0.20 (30)

T = 0.24 (60)

P = 0.512

P = 0.312

P = 0.483

SNP (G > A) rs1761667

G = 0.64 (193)

G = 0.60 (301)

GG = 0.22 (56) GA = 0.76 (189)

G = 0.89 (133)

G = 0.98 (245)

v 2 = 1.356

v2 = 35.24

v 2 = 0.887

Promoter

A = 0.36 (107)

A = 0.40 (199)

GG = 0.40 (60) GA = 0.49 (73) AA = 0.11 (17)

AA = 0.02 (05)

A = 0.60 (90)

A = 0.78 (194)

P = 0.244

P = < 0.001

P = 0.346

*

Number of respective alleles. Genotypes. *** Carriers of alleles in the study population. **

Table 4 Distribution of double combinations of the SNPs (C > T) and (G > A) in CD36 gene in T2DM patients and controls. Genotypes

Controls (n = 150)

Patients (n = 216)

CC&GG (+&+) (+&) (&+) (&)

45 75 11 19

(30.0%) (50.0%) (7.38%) (12.67%)

38 (17.59%) 142 (65.74%) 9 (4.16%) 27 (1.25%)

0.011 0.950 0.307 0.075

1.0 (Ref.) 1.032 (0.387–2.753) 1.737 (0.603–5.006) 2.314 (0.918–5.831)

CC&GA (+&+) (+&) (&+) (&)

56 64 19 11

(37.34%) (42.67%) (12.66%) (7.33%)

136 (62.96%) 44 (20.37%) 27 (12.5%) 9 (4.17%)

<0.001 0.114 0.307 0.042

1.0 (Ref.) 1.709 (0.880–3.321) 0.576 (0.200–1.659) 0.484 (0.240–0.975)

CC&AA (+&+) (+&)

15 (10%) 105 (70.0%)

5 (2.32%) 175 (81.02%)

0.022 0.047

(&+)

4 (2.66%)

0 (0.0%)

0.557

(&)

26 (17.34%)

36 (16.67%)

0.546

CT&GG (+&+) (+&) (&+) (&)

9 (6.0%) 19 (12.67%) 51 (34.0%) 71 (47.34%)

9 (4.17%) 27 (12.50%) 39 (18.05%) 141 (65.28%)

0.002 0.604 <0.001 0.092

1.0 (Ref.) 1.308 (0.475–3.604) 2.597 (1.567–4.303) 1.858 (0.905–3.817)

CT&GA (+&+) (+&) (&+) (&)

17 11 57 65

26 (12.03%) 9 (4.17%) 136 (62.96%) 45 (20.83%)

<0.001 0.203 <0.001 0.025

1.0 (Ref.) 0.641 (0.323–1.272) 0.290 (0.178–0.474) 0.343 (0.135–0.872)

(11.34%) (7.34%) (38.0%) (43.34%)

Pvalue

OR (95% CI)

1.0 (Ref.) 164.198 (0.000– 4.9  1011) 682.054 (0.000– 2.0  1012) 820.991 (0.000– 2.4  1012)

T2DM was reported in the Caucasian population at risk for MetS [17]. However, the allele and genotype frequencies of C/T (rs1527483) polymorphism in intron 11 were similar in controls and T2DM patients and did not show any association with T2DM. This suggested that the regulatory region of the gene is responsible for the varied expression of the CD36 gene in normal and diseased conditions since a variant located in the CD36 upstream promoter determines the binding site for transcription factors [13]. A study in Italian men has shown that a haplotype of five SNPs spanning the CD36 gene was associated with higher free fatty acid and triglyceride levels thereby modulating lipid metabolism and cardiovascular risk [13]. It was also suggested that since CD36 modulates the uptake of modified lipoproteins and is related to insulin resistance through its contribution to fatty acid metabo-

lism, the variants of the CD36 gene may contribute to the pathogenesis of diabetes in African Americans [14]. A recent study on the variants of the CD36 gene in Boston and Puerto Rican adults also revealed its association with metabolic syndrome which can increase the risk of cardiovascular disease and T2DM [20]. However, the minor allele frequency of the SNP G > A (rs1761667) was slightly lower in our population (0.36) compared to 0.46 in Puerto Ricans [20]. A genetic study in the French diabetic population revealed a rare non-sense mutation in the CD36 gene while a promoter variant 178A/C was significantly associated with adiponectin levels and represented a putative marker for insulin resistance [15,16]. Another common polymorphism (478 C/T) associated with CD36 deficiency was reported in Japanese patients with heart disease [21]. Evidence from the literature has shown that CD36 expression in monocytes is up regulated by oxidized low density lipoprotein and its level increases in T2DM, hyperglycemia and related atherosclerosis probably through its contribution to disturbed fatty acid metabolism, [22–24] suggesting a possible connection between atherosclerosis and insulin resistant states through CD36 [25]. The increased hepatic CD36 protein expression in response to diets rich in fatty acids and/or to obesity contributes to aberrant liver fatty acid uptake and subsequent dyslipidemia [26] and the increased foam cell formation in T2DM is caused by decreased macrophage cholesterol efflux to HDL. Reports have suggested that it was associated with abnormal glucose metabolism, and altered serum lipids and CD36 deficiency is a risk factor for MetS [27]. Some studies have reported alterations in FFAs and TGL with a common haplotype in CD36, especially the SNPs rs1761667 and rs1049673 have been associated with elevated plasma FFAs in white men without diabetes [13]. In our study, rs1527483 (C > T) showed highly significant association with lipid levels (P = < 0.001) when considered alone. However, on haplotype analyses significant associations between CD36 SNPs and biochemical parameters were evident (Table 5). A significant difference was observed in total cholesterol (TC) with CC genotype in combination with GA and AA, but CT without GA was also associated with it. Subjects without CC/GA genotype (/) were also associated with TC. TC also showed significant association with GA genotype minus CC and CT. Significant differences in TGL levels were also observed in subjects without CC and GG genotypes but subjects with CT/GG (+/) were also associated with it. HDL is only associated with CC genotype except CT/GG (+/+) in combination without GA and AA. LDL showed its association with CC genotype in combination with GA, without GA and with AA, but CT genotype showed its associa-

M. Banerjee et al. / International Journal of Diabetes Mellitus 2 (2010) 179–183 Table 5 Association comparison of haplotypes of rs1527483 and rs1761667 with biochemical parameters. Genotypes

Biochemical parameters P-value TC

TGL

HDL

LDL

VLDL

S. Cret.

CC&GG

+/+ +/ /+ /

0.745 0.636 0.402 0.447

0.572 0.720 0.772 0.006

0.536 0.577 0.029 0.889

0.564 0.851 <0.001 0.549

0.636 0.955 0.772 0.006

0.348 0.804 0.023 0.462

CC&GA

+/+ +/ /+ /

0.020 0.240 0.005 0.007

0.258 0.392 0.070 0.382

0.990 0.003 0.611 0.091

0.005 0.045 <0.001 0.001

0.029 0.151 0.070 0.382

0.620 0.275 0.807 0.009

CC&AA

+/+ +/ /

0.003 0.449 0.993

0.065 0.380 0.446

0.276 0.034 0.943

0.003 0.224 0.027

0.065 0.394 0.568

0.287 0.953 0.867

CT&GG

+/+ +/ /+ /

0.058 0.447 0.745 0.636

0.701 0.006 0.572 0.720

0.036 0.889 0.536 0.577

0.001 0.549 0.564 0.851

0.701 0.006 0.636 0.955

0.232 0.890 0.348 0.804

CT&GA

+/+ +/ /+ /

0.690 0.007 0.020 0.487

0.700 0.382 0.258 0.226

0.669 0.091 0.111 0.704

0.306 0.001 0.005 0.092

0.715 0.382 0.029 0.199

0.413 0.009 0.620 <0.001

tion with GG and without GA genotypes. Subjects showed its significant association with VLDL only in case of CC/GG (/), CC/ GA (+/+), CT/GG (+/) and CT/GA (/+). In case of S. Cret. subjects without CC and GA, subjects with CT and without GA, subjects without CT and GA, subjects with GG and GG without CC showed a highly significant association. The association of these haplotypes with biochemical parameters indicates that somehow they are responsible for the manifestation of the disease, for example, although CT genotype was not prevalent among the North Indian diabetic patients but in combination with the high-risk genotype GA showed significant associations with TC, LDL, VLDL etc. suggesting that the selected SNPs have a synergistic effect on T2DM. In addition, the presence of the minor allele A* of rs1761667 in the -31118 promoter region of the CD36 gene seemed to contribute greatly in elevating the risk of developing this disorder. 5. Conclusions In conclusion, this is perhaps the first study to examine the association of CD36 polymorphisms in North Indian adults, a high-risk urban population. Considering that CD36 plays a crucial role in T2DM, we showed that the 5’ promoter SNP rs1761667 has a significant association with T2DM patients in this population. Our results suggest that individuals having a ‘GA’ genotype might be susceptible to T2DM and may be at risk of developing related complications. Moreover, it has provided a lead for future studies to examine the role of other CD36 variants in the development of T2DM in India and other ethnic populations. 6. Competing interests The authors declare that they have no competing interests. Acknowledgements The authors would like to thank the Department of Biotechnology (DBT), Ministry of Science and Technology, Government of India, New Delhi, India for funding the work. Authors SG and MS would like to acknowledge the Rajiv Gandhi National Fellowship, University Grants Commission, New Delhi, India and the DBT, New Delhi for their respective Junior Research Fellowships.

183

References [1] Shulman GI. Cellular mechanisms of insulin resistance. J Clin Invest 2000;106: 171–6. [2] Boden G, Shulman GI. Free fatty acids in obesity and type 2 diabetes: defining their role in the development of insulin resistance and b-cell dysfunction. Eur J Clin Invest 2001;32:14–23. [3] Febbraio M, Silverstein RL. CD36: implications in cardiovascular disease. Int J Biochem Cell Biol 2007;39:2012–30. [4] Calvo D, Gomez-Coronado D, Suarez Y, Lasuncion MA, Vega MA. Human CD36 is a high affinity receptor for the native lipoproteins HDL, LDL, and VLDL. J Lipid Res 1998;39:777–88. [5] Connelly MA, Klein SM, Azhar S, Abumrad NA, Williams DL. Comparison of class B scavenger receptors, CD36 and scavenger receptor BI (SR-BI), shows that both receptors mediate high density lipoprotein-cholesteryl ester selective uptake but SR-BI exhibits a unique enhancement of cholesteryl ester uptake. J Biol Chem 1999;274:41–7. [6] Nicholson AC, Hajjar DP. CD36, oxidized LDL and PPAR gamma: pathological interactions in macrophages and atherosclerosis. Vascul Pharmacol 2004;41:139–46. [7] Nassir F, Wilson B, Han X, Gross RW, Abumrad NA. CD36 is important for fatty acid and cholesterol uptake by the proximal but not distal intestine. J Biol Chem 2007;282:19493–501. [8] Thorne RF, Mhaidat NM, Ralston KJ, Burns GF. CD36 is a receptor for oxidized high density lipoprotein: implications for the development of atherosclerosis. FEBS Lett 2007;581:1227–32. [9] Hirano K, Kuwasako T, Nakagawa-Toyama Y, Janabi M, Yamashita S, Matsuzawa Y. Pathophysiology of human genetic CD36 deficiency. Trends Cardiovasc Med 2003;13:136–41. [10] Kontrova K, Zidkova J, Bartos B, Skop V, Sajdok J, Kazdova L, et al. CD36 regulates fatty acid composition and sensitivity to insulin in 3T3-L1 adipocytes. Physiol Res 2007;56:493–6. [11] Armesilla AL, Vega MA. Structural organization of the gene for human CD36 glycoprotein. J Biol Chem 1994;269:18985–91. [12] Rac ME, Safranow K, Poncyljusz W. Molecular basis of human CD36 gene mutations. Mol Med 2007;13:288–96. [13] Ma X, Bacci S, Mlynarski W, Gottardo L, Soccio T, Menzaghi C, et al. A common haplotype at the CD36 locus is associated with high free fatty acid levels and increased cardiovascular risk in Caucasians. Hum Mol Genet 2004;13: 2197–205. [14] Love-Gregory L, Sherva R, Sun L, Wasson J, Schappe T, Doria A, et al. Variants in the CD36 gene associate with the metabolic syndrome and high-density lipoprotein cholesterol. Hum Mol Genet 2008;17:1695–704. [15] Lepretre F, Linton KJ, Lacquemant C, Vatin V, Samson C, Dina C, et al. Genetic study of the CD36 gene in a French diabetic population. Diab Metab 2004;30: 459–63. [16] Lepretre F, Vasseur F, Vaxillaire M, Scherer PE, Ali S, Linton K, et al. A CD36 nonsense mutation associated with insulin resistance and familial type 2 diabetes. Hum Mut 2004;24:104–9. [17] Corpeleijn E, van der Kallen CJH, Kruijshoop M, Magagnin MGP, de Bruin TWA, Feskens EJM, et al. Direct association of a promoter polymorphism in the CD36/FAT fatty acid transporter gene with Type 2 diabetes mellitus and insulin resistance. Diab Med 2006;23:907–11. [18] Bachorik PS, Albers JJ. Precipitation methods for quantification of lipoproteins. Meth Enzymol 1986;129:78–100. [19] Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988;16:1215. [20] Noel SE, Lai CQ, Mattei J, Parnell LD, Ordovas JM, Tuker KL. Variants of the CD36 gene and metabolic syndrome in Boston Puerto Rican adults. Atherosclerosis 2010;211:210–5. [21] Kashiwagi H, Tomiyama Y, Nozaki S, Kiyoi T, Tadokoro S, Matsumoto K, et al. Analysis of genetic abnormalities in types I CD36 deficiency in Japan: identification and cell biological characterization of two novel mutations that cause CD36 deficiency in man. Hum Genet 2001;108: 459–66. [22] Griffin ERA, Hamel N, Fu C, Bush H, McCaffrey T, Asch AS. A link between diabetes and atherosclerosis: glucose regulates expression of CD36 at the level of translation. Nat Med 2001;7:840–6. [23] Handberg A, Levin K, Hojlund K, Nielsen HB. Identification of the oxidized lowdensity lipoprotein scavenger receptor CD36 in plasma: a novel marker of insulin resistance. Circulation 2006;114:1169–76. [24] Pravenec M, Kurtz TW. Molecular genetics of experimental hypertension and the metabolic syndrome: from gene pathways to new therapies. Hypertension 2007;49:941–52. [25] Kashyap SR, Ioachimescu AG, Gornik HL, Gopan T, Davidson M, Makdissi A, et al. Lipid-induced insulin resitance is associated with increased monocyte expression of scavenger receptor CD36 and internalization of oxidized LDL. Obesity (Silver Spring) 2009;17:2142–8. [26] Koonen DPY, Jacobs RL, Febbraio M, Young ME, Soltys CLM, Ong H, et al. Increased hepatic CD36 expression contributes to dyslipidemia associated with diet-induced obesity. Diabetes 2007;56:2863–71. [27] Yamashita S, Hirano K, Kuwasako T, Janabi M, Toyama Y, Ishigami M, et al. Physiological and pathological roles of a multi-ligand receptor CD36 in atherogenesis; insights from CD36-deficient patients. Mol Cell Biochem 2007;299:19–22.

International Journal of Diabetes Mellitus 2 (2010) 184–188

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Original Article

Current clinical status and complications among type 2 diabetic patients in Universiti Sains Malaysia hospital Salwa Selim Ibrahim Abougalambou a,⇑, Mafauzy Mohamed b, Syed Azhar Syed Sulaiman a, Ayman S. Abougalambou c, Mohamed Azmi Hassali d a

Discipline of Clinical Pharmacy, School of Pharmaceutical Sciences, Universiti Sains Malaysia (USM), Malaysia Department of Medicine, School of Medical Sciences, Health Campus, Universiti Sains Malaysia (USM), Malaysia National Heart Institute (IJN), Kuala Lumpur, Malaysia d Discipline of Social and Administrative Pharmacy, School of Pharmaceutical Sciences, Universiti Sains Malaysia (USM), Malaysia b c

a r t i c l e

i n f o

Article history: Received 1 July 2010 Accepted 25 August 2010

Keywords: Type 2 Diabetes Mellitus Microvascular complication Macrovascular complication Dyslipidaemia Hypertension Obesity

a b s t r a c t Objective: To identify current clinical status of type 2 diabetic outpatients and to determine the prevalence of diabetic complications and risk factors. Material and method: Prospective cross-sectional study design was used in the data collection process. The study sample consists of 1077 type 2 Diabetes Mellitus outpatients who fit the inclusion criteria. All the patients were recruited from the diabetic outpatient clinics from Hospital Universiti Sains Malaysia (HUSM). The study period was from January till December 2007. Demographic data, clinical status of diabetes and its complications were collected and analyzed for the prevalence of complications and risk factors. Results: One thousand and seventy seven type 2 diabetes outpatients were included in the present study. Mean age was 58.3 years and duration of diabetes was 11 years. Only 23.4% of the subjects achieved HbA1c of 67%, 53.5% of patients had achieved target FPG 66.7 mmol/l, and 60.4% of the patients had achieved optimal postprandial plasma glucose level <10 mmol/l. The overall prevalence of dyslipidaemia was 93.7%, hypertension was 92.7% and obesity was 81.5%. Nephropathy was the most common complication accounting for 91.0% followed by neuropathy 54.4%, retinopathy 39.3%, and macrovascular complications 17.5%. The vascular complications were significantly associated with the age (P < 0.001), BMI (P < 0.001), and triglyceride (P < 0.001). Conclusion: The prevalence of dyslipidaemia, hypertension and obesity were high. The high prevalence of vascular complications was associated with age, BMI and triglyceride of diabetic patients. Effort to treat triglyceride appropriately among elderly diabetic patients could be considered as a prime target. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction Type 2 Diabetes Mellitus (DM) is rapidly rising as a global health care problem that is threatening to reach pandemic levels by 2030. In 2003, an estimated 194 million adults had diabetes worldwide (5.1%) [1]. This prevalence increased to 6.0% in 2007, and is predicted to increase to 7.3% by 2025 [2]. People (380 million) are expected to have diabetes in 2025 [2]. In Malaysia, the Third National Health and Morbidity Survey [3] showed that the prevalence of type 2 DM for adults aged 30 years and above was found to be 14.9% in 2006. The presence of hypertension in diabetic patients has dramatically increased the rate of complication [4]. When happening together, the two disease entities appear to aggravate one another, ⇑ Corresponding author. E-mail address: [email protected] (S.S.I. Abougalambou).

worsening both diabetes and cardiovascular end points [5]. Dyslipidaemia is a major risk factor for macrovascular disease. The prevalence of dyslipidaemia is increased by at least twofold in the presence of type 2 DM, and involves all classes of lipoprotein [6]. Obesity is a great public health concern, because it is directly related to the development of diabetes, hypertension, and ultimately, congestive heart failure. Overweight adults are more likely to experience problems, including higher morbidity and mortality [7]. The present study aimed to identify the current clinical status of type 2 diabetic outpatients in tertiary center and to estimate the prevalence of vascular complications and other selected risk factors. The rationale of this study will be to provide suitable baseline data regarding the current status of type 2 diabetic patients at HUSM, and the rate of vascular complications among diabetic patients. Knowing the factors that affect the development of vascular complications may allow clinicians to draw appropriate plans for preventing or slowing down the progress of diabetic complications.

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.08.001

185

S.S.I. Abougalambou et al. / International Journal of Diabetes Mellitus 2 (2010) 184–188

2. Methodology The medical records were studied either directly from the diabetes clinic after the patients consulted the doctors or from the patient medical record center. The patients selected were type 2 diabetic outpatients, aged over 18 years, with active follow-up at the diabetic clinic. The exclusion criteria for this study included patients who were suffering from juvenile diabetes, gestational diabetes, thyroid problems, obstructive liver disease, advanced renal failure, and tuberculosis. A prospective study was conducted for a study period of one year (1st January 2007 till 31st December 2007) in order to identify the characteristics of type 2 diabetic outpatients in a tertiary center, and to determine the prevalence of diabetic complications associated with outpatient diabetic care at HUSM, which is located in the state of Kelantan, Malaysia. The study design is an observational, prospective cross-sectional study. Non-probability sampling method (convenience sample technique) was applied. The research’s protocol was approved by the Human Research and Ethics Committee of the School of Medicine in the Universiti Sains Malaysia. Signed informed consent was obtained from all patients. 2.1. Data collection The outpatient diabetic clinic recording lists of patients who attended the diabetic clinic in HUSM were captured from the diabetic clinic registration book. Based on glycaemic control tests (HbA1c, FPG, PPG), the medical records were then retrieved from the record office using the patient’s name. The medical records review was undertaken by a single researcher, and the required information including demographic, co-morbidity characteristics, detailed physical and biochemical information and therapy to be reviewed and recorded in a data collection form. Socio-demographic characteristics included age, sex and race, alcohol, smoking history, physical activity and level of education. Physical examination included: pulse rate, height, weight and waist circumference. Blood pressure was measured twice and average reading was taken. Hypertension was defined as systolic blood pressure of >130 mmHg or diastolic blood pressure of >80 mmHg or current use of antihypertensive drugs also has been diagnosed as hypertension [8]. Laboratory results included fasting plasma glucose (FPG), postprandial plasma glucose (PPG), HbA1c level, and lipid profile. Dyslipidaemia was defined as a fasting cholesterol of greater than 4.5 mmol/l, LDL-C greater than 2.6 mmol/l, Triglyceride greater than 1.7 mmol/l, HDL-C less than 1.0 mmol/l in males and less than 1.3 mmol/l in females [9]. Diabetic retinopathy (DR) was diagnosed with the presence of retinal hemorrhages, exudates and macular edema [10]. Neuropathy was diagnosed in the presence of persistent numbness, paresthesia, loss of hearing of the tuning fork and sense of vibration [10]. Diabetic nephropathy (DN) was considered by positive persistent proteinuria for at least three consecutive readings per year, and/ or serum creatinine (SCr) >130 lmol/L and/or GFR <60 ml/min [10]. Coronary artery disease was diagnosed by documented angina symptoms and confirmed by ECG, or from the results of percutaneous transluminal coronary angiography (PTCA) in patient’s records [11]. Cerebrovascular disease was defined by the presence of transient ischemic attack or stroke in the past medical history [11]. 2.2. Statistical analysis Statistical analyses were performed using statistical packages for social sciences (SPSS) version 12 (SPSS Inc., 2003). Demographic data were expressed as mean ± SD. Distributions and frequencies

of the independent variable were examined. Data exploration was undertaken to include descriptive statistics for each variable. Frequencies and percentages for independent variable were calculated. In simple logistic analysis, each independent variable was analyzed to look at any significant association with dependent variable (vascular complications) and preceded to by multiple logistic regressions, to confirm the association after excluding confounders. The results of simple logistic regression analysis were recorded as beta, P-value, crude odds ratio and 95% confidence interval. Multivariate analysis was performed on numerical and categorical analysis variable by using binary logistic regression to eliminate confounding effect as there are more than one independent variable. The findings of the final model were presented with adjusted odds ratios (OR), its 95% confidence interval (CI) and corresponding P-value. The level of significance was set at 0.05. 3. Results All type 2 diabetic outpatients who attended diabetic clinic during study periods and fulfilled the inclusion criteria were selected. Demographic, health characteristics and clinical variables of all type 2 diabetic patients included in this study are shown in Tables 1–3. One thousand and seventy seven type 2 diabetic outpatients participated in the present study. The age of patients was from 18 to 88, the mean (±SD) age was 58.3 (±9.80) years, and (55.8%) were female. By assessing the levels of education in patients, those who had education less than secondary school level constituted the majority 53.9% of the patients (Table 1). The mean duration of diabetes was 11 ± 6.81 years, ranging from less than 1 year to 40 years, and 74.6% had diabetes for more than 5 years (Table 2). The mean HbA1c was 8.7% ± 2.3, while mean fasting plasma glucose levels were 7.8 ± 3.7 mmol/l, and the mean postprandial plasma glucose levels were 10.0 ± 4.3 mmol/l. The majority of patients 998 (92.7%) had hypertension, the mean (±SD) systolic blood pressure (SBP) was 135.9 (±19.78) mmHg and the mean

Table 1 Demographic characteristics of type 2 diabetic patients. Variable

n (%)

Gender Male Female

476 (44.2) 601 (55.8)

Age (years) 635 >35–50 >50–65 >65

15 (1.4) 194 (18) 626 (58.1) 242 (22.5)

Race Malay Chinese Indian

916 (85.1) 150 (13.9) 11 (1.0)

Smoking history Current smoker Pervious smoker Never smoked

66 (6.1) 81 (7.5) 930 (86.4)

Physical activity Active P150 min/week Non-active <150 min/week

471 (43.7) 606 (56.3)

Level of education <Secondary school PSecondary school

580 (53.9) 497 (46.1)

Family history Yes No

141 (13.1%) 936 (86.9%)

186

S.S.I. Abougalambou et al. / International Journal of Diabetes Mellitus 2 (2010) 184–188 Table 2 Health characteristics of type 2 diabetic patients. Variable

Table 4 Types of vascular complications among type 2 DM patients.

n (%) 2

BMI (kg/m ) Asia pacific Target 623 kg/m2 Non-target >23 kg/m2

199 (18.5) 878 (81.5)

Waist circumference category AP (cm) Target (male) 690 cm 100 (9.3) Non-target (male) >90 cm 376 (34.9) Target (female) <80 cm 50 (4.6) Non-target (female) P80 cm 551 (51.2) Diabetes duration (years) 65 years >5–10 years >10–15 years >15–20 years >20 years

273 294 256 136 118

HPT duration category (years) Free from HPT 63 years >3–6 years >6–9 years >9 years

56 (5.2) 204 (18.9) 288 (26.7) 160 (14.9) 369 (34.3)

(25.4) (27.3) (23.7) (12.6) (11)

Type of complications

n (%)

No complications Microvascular complications Retinopathy Neuropathy Nephropathy Retinopathy, neuropathy Retinopathy, nephropathy Neuropathy, nephropathy Retinopathy, neuropathy and nephropathy Microvascular and macrovascular complications Total

48 (4.5) 841 (78.0) 15 (1.4%) 19 (1.8%) 273 (25.3%) 14 (1.3%) 87 (8.1%) 221 (20.5%) 212 (19.6%) 188 (17.5%) 1077 (100%)

1.40 ± 0.54 mmol/L and mean TG was1.74 ± 0.85 mmol/l. Non-achievement of the ADA guideline for LDL-C, total cholesterol, triglycerides, and HDL-C were 54.3%, 36.8%, 42%, and 32%, respectively (Table 3). In the aspect of the Metabolic Syndrome, the overall prevalence of dyslipidaemia was 93.7%, hypertension was 92.7%, and obesity (BMI >23 kg/m2) was 81.5%. 3.1. Type of vascular complications among type 2 DM patients

Table 3 Characteristics of clinical variables of type 2 DM patients. Variables

n (%)

Mean (±SD)

HbA1c (%) Optimal <7% Fair 7–8% Poor >8%

252 (23.4) 258 (24) 567 (52.6)

8.72 (±2.34)

Fasting plasma glucose (mmol/l) Optimal <6.7 mmol/l Fair 6.7–7.8 mmol/l Poor >7.8 mmol/l

498 (46.3) 163 (15.1) 416 (38.6)

7.89 (±3.72)

Most of the patients, 841 (78%) had microvascular complications alone, 188 (17.5%) had a combination of microvascular and macrovascular complications, and of these, 137 (12.8%) had coronary heart disease, only 51 (4.7%) had cerebrovascular disease and a minority 48 (4.5%) had no complications (Table 4). Diabetic nephropathy was the most common complication, accounting for 91.0%, followed by neuropathy 54.4%, retinopathy 39.3%, and macrovascular complications (17.5%).

10.03 (±4.38)

3.2. Univariate analysis of risk factors affecting the development of complications

PPG (mmol/l) Control <10.0 mmol/l Uncontrolled P10.0 mmol/l Hypertension control Systolic blood pressure (mmHg) 6120 mmHg >120–139 mmHg 140–159 mmHg P160 mmHg Diastolic blood pressure (mmHg) <80 mmHg 80–89 mmHg 90–99 mmHg >100 mmHg

634 (58.9) 443 (41.1)

332 (30.8) 289 (26.8) 296 (27.5) 160 (14.9) 753 (69.9) 33 (3.1) 213 (19.8) 78 (7.2)

LDL cholesterol (mmol/l) ADA Normal <2.6 mmol/l Border high 2.6–3.3 mmol/l High 3.4–4.1 mmol/l Very high >4.1 mmol/l

493 (45.7) 285 (26.5) 186 (17.3) 113 (10.5)

Total cholesterol (mmol/l) ADA Target <5.2 mmol/l Non-target P5.2 mmol/l

681 (63.2) 396 (36.8)

Triglycerides (mmol/l) ADA Normal <1.7 mmol/l Border high 1.7–2.3 mmol/l High 2.4–5.7 mmol/l

625 (58) 223 (20.7) 229 (21.3)

HDL cholesterol (mmol/l) ADA Target (male) >1.0 mmol/l (40 mg/dl) Non-target (male) 61.0 mmol/l (640 mg/dl) Target (female) >1.3 mmol/l (>50 mg/dl) Non-target (female) 61.3 mmol/l (650 mg/dl)

384 (35.7) 92 (8.5) 348 (32.3) 253 (23.5)

135.98 (±19.78)

80.62 (±9.83)

Table 5 shows the simple logistic regression of risk factors affecting the development of vascular complications. There were significant associations between the complications and age, BMI, WC, PPG, triglyceride, duration of diabetes and systolic blood pressure. 3.3. Final model of multivariate analysis on complications Using a backward stepwise logistic regression, all factors found to be significant at P-value <0.05 during the previous analysis, were introduced together in one multivariate analysis. Statistically, vari-

2.82 (±1.08)

4.98 (±1.17)

Table 5 Univariate analysis of risk factors affecting the development of complications. Variables Age BMI WC Duration of diabetes HbA1c FPG PPG Triglyceride Total cholesterol HDL LDL Systolic blood pressure Diastolic blood pressure

1.74 (±0.85)

1.40 (±0.54)

(±SD) diastolic blood pressure (BPD) was 80.6 (±9.83) mmHg. For lipid profile, the mean LDL-C was 2.82 ± 1.08 mmol/l, the mean total cholesterol was 4.98 ± 1.17 mmol/l, while mean HDL-C was

a

ba 0.14 0.14 0.05 0.12 0.02 0.04 0.09 0.77 0.16 0.06 0.06 0.02 0.01

Crude OR (95%CI)

P-value

1.15 0.86 0.94 1.12 0.97 1.04 1.10 2.18 1.18 0.93 1.07 1.02 1.01

<0.001 <0.001 <0.001 <0.001 0.65 0.317 0.027 0.001 0.214 0.793 0.630 0.003 0.452

(1.11–1.19) (0.82–0.90) (0.92–0.96) (1.06–1.190 (0.86–1.09) (0.95–1.14) (1.01–1.19) (1.35–3.50) (0.90–1.54) (0.57–1.53) (0.81–1.41) (1.01–1.04) (0.98–1.04)

Simple logistic regression (outcome as complications).

S.S.I. Abougalambou et al. / International Journal of Diabetes Mellitus 2 (2010) 184–188 Table 6 Factors significantly associated with development of complications. Independent variables Age BMI Triglyceride

ba 0.13 0.14 1.11

OR (95.0% CI)

P-value

1.14 (1.10–1.18) 0.86 (0.81–0.91) 3.06 (1.82–5.13)

<0.001 <0.001 <0.001

Overall correctly classified percentage = 95.5%. Area under curve = 91.7%. a Multiple logistic regression.

ables at P-value <0.05 were accepted. Three variables remained in the final model. These were age, BMI and triglyceride as shown in Table 6. 4. Discussion This study undertaken with 1077 diabetic type 2 outpatients is large enough to evaluate the current status of diabetic patients and the burden of diabetic complications among Malaysian people at a tertiary center. Body mass index and waist circumference were two measurements of obesity in the present study. Looking at current results, 199 (18.5%) of patients had BMI at Asian Pacific target 6 23 kg/ m2 and 878 (81.5%) had non-target BMI. According to Asian Pacific type 2 DM policy group [9] waist circumference target, a total of 100 (9.3%) male patients were within targets, while 376 (34.9%) male patients did not achieve targets. A total of 551 (51.2%) female patients did not achieve the target, while only 50 (4.6%) female patients achieved the target. Prospective epidemiological studies showed that waist circumference has been an independent predictor of type 2 DM risk [12,13]. The majority of diabetic patients do not perform physical activity regularly, and do not adhere to dietary advice. Also patients taking oral hypoglycaemic commonly gain weight due to medication [14]. In the current study 76.6% of the patients did not achieve the ADA guidelines of HbA1c, 46.5% and 39.6% of patients recruited in this study had not achieved a target of FPG and PPG levels respectively. Dyslipidaemia was found in 93.7% of patients, of which total cholesterol was more than normal in 36.8%, LDL in 54.3% and triglycerides in 42% but HDL was lower than normal in 32%. A study by Akbar et al. [15] suggested that poor control was associated with poor diet compliance and use of multiple medications. Proper management and control of this disease are needed among elderly patients. To achieve glycaemic control, hypertension and dyslipidaemia the result from this study found that HbA1c achieved the target in only 24.3% of the patients, 47.2% of patients achieved blood pressure targets and LDL-C of being less than 2.6 mmol/l was achieved in only 45.8% of all patients, regardless of any treatment strategy; all of these accounted for an increase in diabetic complications. The explanation of the result from the present study for the current state of diabetes may be due to various reasons, like advancing age, obesity and genetic factors, disease unawareness, socioeconomic status, sedentary life style, long duration of diabetes, intake of unhealthy food and poor compliance with treatments. This study has shown a prevalence of macrovascular disease of 17.5% among diabetics and a percentage of macrovascular disease lower than that in a study in UAE by Al-Maskari et al. [16]. The latter study found a prevalence of macrovascular disease in 29.5% of diabetics. The differences in our rates of macrovascular complications among type 2 DM patients, as compared with others, may be attributed to the differences in study design and the population characteristics of various studies. The present study shows that the prevalence rate of retinopathy was 39.3%. The prevalence of retinopathy demonstrates wide vari-

187

ations between countries; in type 2 DM it ranges from 17% in Switzerland to 52% in the United Kingdom [17]. Prevalence of neuropathy was 54.7%, this percentage of neuropathy is higher than that in a study by Tesfaye et al. [18], who recruited 3250 diabetic patients and reported the prevalence of neuropathy in 28% of them, but in other studies it was between 25% and 60% [19,20]. The prevalence of nephropathy was 91%. This is considered as a high percentage in comparison with other studies on diabetic nephropathy which occurs in 40% of diabetic patients [21]. In the present study a high percentage of patients have microvascular rather than macrovascular complications. In comparisons regarding the prevalence of diabetic microvascular complications, the study shows that the prevalence of diabetic nephropathy is more than neuropathy and retinopathy. This may be because the patients in the current study have hypertension and dyslipidaemia, which are related to renal complications. In addition, the population in this study is Asian, where the prevalence of nephropathy is more than the other people. This may be due to genetic factors. Another possible explanation for the high percentage of complication in the present study is that most of the patients are elderly. 5. Conclusion The prevalence of dyslipidaemia, hypertension and obesity were high in this population. The unsatisfactory control of metabolic status may be due to age, long duration, obesity and level of education of diabetic patients. The rate of vascular complications among type 2 diabetic patients was high. Identifying factors associated with the development of complications would be able to prevent the complications. From this study, the findings indicated that age, BMI and triglyceride concentrations are associated with vascular complications. More attention must be paid to elderly diabetic patients with appropriate treatment for high triglyceride. Diabetic patients need more efforts to be spent on them. Screening and intervention programs should be implemented early at the diagnosis stage, and risk factors should be treated aggressively. Public health strategies are required in order to improve the current status of diabetic patients and to decrease the rate of prevalence of vascular complications. 6. Limitation This analysis was based on type 2 DM clinic at HUSM (tertiary center); thus data from other centers are required to determine whether the finding in this study can be generalised to diabetes care setting in general. Furthermore the population of this study was that of diabetic outpatients. Acknowledgment We would like to acknowledge the Institute of Postgraduate Studies (IPS) Universiti Sains Malaysia for its support in this research. References [1] Sicree R, Shaw J, Zimmet P. Executive summary. In: Gan D, editor. Diabetes atlas. 2nd ed. Brussels: International Diabetes Federation and World Diabetes Foundation; 2003. [2] Sicree R, Shaw J, Zimmet P. Prevalence and projections. In: Gan D, editor. Diabetes atlas. 3rd ed. Brussels: International Diabetes Federation; 2006. p. 16–104. [3] The Third National Health Morbidity Survey (NHMS III) Diabetes Group. Ministry of Health Malaysia; 2006. [4] Canadian Diabetes A. Clinical practice guidelines for prevention and management of diabetetic. Can Diabet J Diab 2003;27(1);S113116.

188

S.S.I. Abougalambou et al. / International Journal of Diabetes Mellitus 2 (2010) 184–188

[5] Stamler J, Vaccaro O, Neaton J. Diabetes, other risk factors, and 12 years cardiovascular mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes Care 1993;16:434–44. [6] Jean P, John R. Diabetes in old age: establishing the diagnosis. 2nd ed, Chapter 3. Second Edited by Alan J & Paul finucane, John Wiley & Sons Ltd.; 2001. [7] Liu S, Manson J. Dietary carbohydrates, physical inactivity, obesity and the metabolic syndrome as predictors of coronary heart disease. Curr Opin Lipidol 2001;12:395–404. [8] JNC II (2003). Complete report. The seventh report of the joint national committee on prevention, detection, evaluation and treatment of high blood pressure. Hypertension 2003;42:1206–52. [9] Asian-Pacific Type 2 Diabetes Policy Group. Type 2 DM practical targets and treatments: screening, diagnosis, management, treatment, monitoring, education and prevention. 4th ed.; 2005. [10] Alwakeel S, Al-Suwaida A, Isnani A, Al-Harbi A, Alam A. Concomitant macro and microvascular complications in diabetic nephropathy. Saudi J Kidney Dis Transplant 2009;20(3):402–9. [11] Al-Maskari F, El-Sadig M, Norman J. The prevalence of macrovascular complications among diabetic patients in the United Arab Emirates. Cardiovas Diabetol 2007;6:24. [12] Koh-Banerjee P, Wang Y, Hu F, Spiegelman D, Willett W, Rimm E. Changes in body weight and body fat distribution as risk factors for clinical diabetes in US men. Am J Epidemiol 2004;159:1150–9.

[13] Wang Y, Rimm E, Stampfer M, Willett W, Hu F. Comparison of abdominal adiposity and overall obesity in predicting risk of type 2 DM among men. Am J Clin Nutr 2005;81:555–63. [14] Pi-Sunyer FX. Weight loss in type 2 diabetic patients. Diabetes Care 2005;28:1526–7. [15] Akbar D, Al-Gamdi A, Hejazi N. Poor lipid control in type 2 diabetics with and without ischemic heart disease. Endocrine 2003;1(3):217–20. [16] Al-Maskari F, El-Sadig M. Prevalence of diabetic retinopathy in the United Arab Emirates: a cross-sectional survey. BMC Ophthalmol 2007;16:7–11. [17] Amos A, Mccarty D, Zimmet P. The rising global burden of diabetes and its complications: estimates and projections to the year 2010. Diabet Med 1997;14(5):S1–85. [18] Tesfaye S, Stevens L, Stephenson J, Fuller JH, Plater M, Ionescu-Tirgoviste C, et al. Prevalence of diabetic peripheral neuropathy and its relation to glycaemic control and potential risk factors. The EURODIAB IDDM complications study. Diabetologia 1996;39:1377–84. [19] Boru U, Alp R, Sargin H. Prevalence of peripheral neuropathy in type 2 diabetic patients attending a diabetes center in Turkey. Endocr J 2004;51:563–7. [20] Tres G, Lisboa H, Syllos R, Canani L, Gross J. Prevalence and characteristics of diabetic polyneuropathy in Passo Fundo, South of Brazil. Arq Bras Endocrinol Metabol 2007;51:987–92. [21] Parving H. Benefits of and cost of antihypertensive treatment in incipient and overt diabetic nephropathy. J Hyperten 1998;16:99–101.

International Journal of Diabetes Mellitus 2 (2010) 189–195

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus journal homepage: www.elsevier.com/locate/ijdm

Review

Endothelial cell dysfunction in hyperglycemia: Phenotypic change, intracellular signaling modification, ultrastructural alteration, and potential clinical outcomes Doina Popov ⇑ Laboratory of Vascular Dysfunction in Diabetes and Obesity, Institute of Cellular Biology and Pathology ‘‘N. Simionescu”, Romania

a r t i c l e

i n f o

Article history: Received 29 June 2010 Accepted 6 September 2010

Keywords: Quiescence Inflammation Proliferation Apoptosis Transcription factors Signaling pathways

a b s t r a c t Hyperglycemia, the hallmark of Diabetes mellitus, is a major risk factor for endothelial dysfunction and vascular complication. In recent years, significant achievements have been made in understanding endothelial cell dysfunction triggered by high glucose concentration. The purpose of this review is to discuss the results of these recent developments. First, the remarkable plasticity of vascular endothelial cell in response to the high glucose insult is emphasized. This is evident through the switch in the cell’s normal quiescent profile into new phenotypes, endowed with biosynthetic, inflammatory, adhesive, proliferative, migratory, pro-atherogenic, and pro-coagulant properties, frequently overlapping each other. Then, we underline the imbalanced expression and activity of transcription and signaling pathways, and the intense metabolic activity that accompanies the change in endothelial cell phenotype. As an adaptation to the high glucose-induced biochemical modification, a severe alteration of cell structure is produced. The review concludes with the clinical outcomes of the subject, emphasizing the high glucose-associated endothelial cell dysfunctional molecules of potential for targeting, and for reducing the impact of hyperglycemia on vascular endothelium. Such interventions may lead to a more efficient therapy for the benefit of those diabetic patients who are at increased cardiovascular risk. Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved.

1. Introduction Endothelial cells (ECs) are flat epithelial cells that form a monolayer that lines the internal lumen of the blood vessels. In normal, physiological condition, ECs are exposed to circulating blood glucose levels in the range of 3.6–5.8 mmol/L, which are tightly regulated as part of metabolic homeostasis. The cells are metabolically active, and produce mediators that affect vascular tone, cell adhesion, and the homeostasis of clotting, and fibrinolysis. Through its contribution to hemostasis, ECs ensure fluidity of the blood. In normal conditions, the phenotype of EC is characterized as quiescent, with turnover rates of the order of months to years. Latest reports have identified several molecules that control EC quiescence. These are the circulating form of human Bone Morphogenetic Protein-9 (BMP-9) [1], the cytosolic phospholipase A2-a when sequestrated within Golgi apparatus [2], the transcription factor E2-2 (member of the basic helix–loop-helix family) [3], the very low-density lipoprotein receptor [4], and Angiopoietin 1 (Ang1), the ligand for the receptor tyrosine kinase Tie2 [5–7]. Shear stress is also a potent physiological regulator of EC quiescence. As a biomarker of ⇑ Address: Laboratory of Vascular Dysfunction in Diabetes and Obesity, Institute of Cellular Biology and Pathology ‘‘N. Simionescu” of the Romanian Academy, 8, B.P. Hasdeu Street, Bucharest 050568, Romania. Tel: +40 213194518; fax: +40 213194519. E-mail address: [email protected]

vascular quiescence stands the anti-angiogenic R-ras gene expression [8]. 2. High glucose concentration induces phenotypic switch and modifies the intracellular signaling in vascular endothelial cells Exposure of vascular ECs to glucose levels over than 10 mmol/L (in vitro or in vivo, as in Diabetes mellitus) is regarded as a high glucose (HG) condition. The over normal glucose concentration perturbs cells homeostasis and biochemistry, triggering modifications both in large vessels (macrovasculature) and in small blood vessels, such as arterioles, venules, and capillaries (microvasculature). As a consequence of HG concentration, EC quiescence is lost, cells acquire new phenotypes, their normal function is impaired, and ‘‘endothelial cell dysfunction” is installed. EC dysfunction is characterized by one or more of the following features: deficiency in bioavailable nitric oxide (NO), reduced endothelium-mediated vasorelaxation, hemodynamic deregulation, impaired fibrinolytic ability, enhanced turnover, overproduction of growth factors, increased expression of adhesion molecules and inflammatory genes, excessive generation of reactive oxygen species (ROS), increased oxidant stress, and enhanced permeability of the cell layer [9–12]. In examination of HG effects on vascular EC, one should also take into account that over physiological glucose concentrations lead to the accelerated formation of multiple biochemical

1877-5934/$ - see front matter Ó 2010 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdm.2010.09.002

190

D. Popov / International Journal of Diabetes Mellitus 2 (2010) 189–195

species unusual in physiological conditions. Among these are the nonenzymatic reactive Amadori products, 3-deoxyglucosone, diacylglycerol, methylglyoxal, advanced glycation end products (AGEs), ROS, and nitrosylated species, which further amplify the imbalance that portrays HG-associated EC dysfunction. ROS production also triggers the peroxidation of plasma membrane polyunsaturated fatty acids like linoleic acid and arachidonic acid, generating endogenous 4-hydroxy nonenal, a highly reactive carbonyl compound. The increased oxidative stress seems to be a common alteration, triggered by a Type 2 diabetes milieu, in which hyperglycemia is adjoined by insulin resistance, hyperinsulinemia, and dyslipidemia [13]. There is a common agreement that endothelial dysfunction precedes the development of micro- and macrovascular complications associated with Type 2 diabetes, such as nephropathy, retinopathy, atherosclerosis, and coronary artery disease; the underlying mechanism includes the accelerated formation of AGEs, activation of protein kinase C, increased pro-inflammatory signaling, and impaired sensitivity of the PI 3-kinase signaling pathways [14]. Recent data show that etiopathogenesis of EC dysfunction differs in Types 1 and 2 diabetes [15]; it is present at the earliest stages of metabolic syndrome and insulin resistance, and may precede the clinical diagnosis of Type 2 diabetes by several years [16]. A first issue examined in this overview is the remarkable plasticity of EC in HG conditions, allowing the transition of the normal quiescent profile into a spectrum of new biosynthetic, pro-inflammatory, pro-adhesive, migratory, pro-atherogenic, pro-coagulant, proliferative, pro-apoptotic, and/or senescent phenotypes; these frequently overlap, e.g. the biosynthetic phenotype is also adhesive, pro-inflammatory, and pro-atherogenic, while the proapoptotic phenotype is a senescent one. As a function of the duration of HG exposure (in vitro) and of circulating glucose level (in vivo) vascular ECs gradually turn into biosynthetic cells, endowed with an over developed rough endoplasmic reticulum (rER); however, in this condition, the protein folding process within rER might be affected, and the endoplasmic reticulum stress is installed [17]. In time, and also as a function of glucose concentration, ECs enlarged and thickened their basal lamina by mechanisms that involve complex biochemical changes [18]. To this modification contribute TGF-b and its receptor ALK1 [19], fibronectin over expression [20], AGEs [21], as well as AGEs cross-linking to collagen molecules; the latter products are less sensitive to degradation, and promote extracellular matrix accumulation [22]. HG concentration also induces pro-inflammatory and pro-adhesive phenotypic changes of vascular ECs [23,24]; in these circumstances, cell surfaces express adhesion molecules (intracellular adhesion molecule-1, vascular cell adhesion molecule-1, and endothelial selectin), interleukin (IL)-1b expression becomes up-regulated [24], and secretion of VEGF, IL-8, IL-6, and TNF-a attains significantly increased levels [25]. Along with the typical cytokines (TNF-a and IL-1b), and chemokines (such as the monocyte chemoattractant protein-1), the latest reports emphasize that the proinflammatory phenotype of EC is associated with an increased expression of inflammatory genes (e.g. the Neuronatin gene) [26], with the presence of ROS and phosphorylation of ERK1/2, c-Jun NH2 – terminal kinase (JNK), NF-kB [23,27,28], and of NF-jB transcription factor inhibitor IkBa [29]. Interestingly, the effect of HG on this phenotypic change is dependent on the anatomic position of the vessel, as well as on the duration of diabetes. Thus, regions of arteries exposed to low shear stress are susceptible to inflammation, whereas regions exposed to high shear stress are protected; in the latter areas, the transcription factor NF-E2-related factor 2 inhibits p38 phosphorylation, and suppresses EC dysfunction [30]. Moreover, the duration of diabetes selectively up-regulates the inflammatory genes expression, i.e. in short-term diabetes, the

mRNA transcripts for chemokine ligands CCL2 and CCL5 were upregulated in the aortic ECs, while at later stages of diabetes these genes were up-regulated in both the aortic and venous ECs [31]. HG significantly enhanced the migration of ECs (within the retina) concomitant with the sustained activation of the downstream prosurvival and promigratory signaling pathways, including Src kinase, phosphatidylinositol 3-kinase/Akt1/endothelial NO synthase, and ERKs [32]. Recent reports demonstrate that EC migration is NO-induced in a process, which implies the inactivation of the transcription factor FOXO3a and subsequent down-regulation of peroxisome proliferator-activated receptor ccoactivator 1a (PGC-1a) [33]. Much of the current literature deals with the migratory properties of ECs as manifest in the angiogenesis process, and with endothelial progenitor cells (EPCs) migration during vascular repairs. Although these issues are beyond the subject of this review, an interesting recent report underlines that Type 2 Diabetes mellitus impairs EPC migration, in a process linked to the stimulation of CXC receptor-4 (CXCR4) and activation of PI3K/Akt/eNOS signaling pathway [34]. Interestingly, although hyperglycemia is acknowledged to be an independent risk factor for developing diabetes-associated atherosclerosis, the pro-inflammatory environment in diabetes is the critical factor conditioning the early pro-atherosclerotic actions of HG [35]. Reportedly, the presence of hyperglycemia stimulates expression of thioredoxin-interacting protein (TXNIP) [36] with a possible role in the EC pro-atherosclerotic response to extracellular diabetic-like environment [37]. Another imbalance associated with the effect of HG concentration is the promotion of a pro-coagulant condition of endothelium (by production of pro-coagulant mediators such as plasminogen activator inhibitor-1, fibrinogen, and P-selectin) associated with a reduced fibrinolitic activity [15,38]. Reportedly, Type 2 diabetes patients may be especially vulnerable to prothrombotic events when hyperglycemia is concurrent with systemic inflammation [39]. Not only is the endothelial cell affected by HG, but also the circulating cells. Thus, a recent report emphasizes that a glucoseregulated protein (GRP78) is involved in platelet deposition during interaction with the vascular wall [40]. High glucose concentration may also trigger two opposed phenotypic changes of vascular EC: in some circumstances, EC turns into proliferative, while in others, into apoptotic. An important player in the induction of EC proliferation is nicotinamide phosphoribosyltransferase (Nampt); which enables cells to resist HG concentration, and to use excess glucose to support replicative longevity and angiogenic activity [41]. For the retinal ECs exposed to HG, it was demonstrated that proliferation was accelerated as the number of pericytes gradually decreased [42]. The altered hemodynamic forces generated by changes in blood flow, influence also EC proliferation [43]. Although there is a common agreement on HG concentration as an inductor for EC apoptosis [44,45], the circumstances that favor this process are diabetes duration (via selective up-regulated caspase-1mRNA) [31], the sequential activation of ROS, JNK, and caspase-3 [46,47], the down-regulation of connexins Cx43 [48], Cx37, and Cx40 [49], and the presence of auto-antibodies in patients with macular edema or progression to albuminuria [50]. The intracellular consequences of EC apoptotic changes consist in DNA fragmentation, and mitochondrial dysfunction, manifest by alteration of membrane potential, release of cytochrome-c, and mitochondrial fragmentation, and these changes play a potential pathogenic role in mediating the risk of Type 2 Diabetes mellitus. Thus, defective or insufficient mitochondrial function may lead to the chronic accumulation of lipid oxidative metabolites that can mediate insulin resistance and secretory dysfunction [48,51,52]. Moreover, in HG condition, a senescent phenotype of EC occurs [53]; reportedly, cell turnover and oxidative stress are engaged in

D. Popov / International Journal of Diabetes Mellitus 2 (2010) 189–195

this process by mechanisms involving telomere shortening [54]. Other contributors to the senescent phenotype of EC are the decreased expression of NAD+-dependent deacetylase sirtuin1, and the activation of p53 acetylation [55]. Taken together, the multitude of new phenotypes acquired by EC under the insult of HG concentration reveals remarkable cell plasticity, linked to the alteration of specific molecules and intracellular pathways. Comprehensive recent reports advanced deeper into EC biochemistry in HG conditions, unveiling the mechanisms that underlie its intense metabolic activity.

3. High glucose concentration intensifies the metabolic activity of vascular endothelial cells The intensification of metabolic activity of ECs exposed to HG is the result of an imbalance between up-regulation of several transcription and signaling factors, and down-regulation of other intracellular molecules. Examples of activated transcription factors are the aryl hydrocarbon receptor transcription factor [56], p300 (a transcriptional co-activator with histone acetyl transferase activity) [57], and COUP-TFII (the chicken ovalbumin upstream promoter-transcription factor II) [58]. In human umbilical vein ECs, HG has time-dependent effects on COUP-TFII, i.e. the shortterm (60–240 min) HG stimulation increases its expression, while long term stimulation (48 h) down-regulates its expression [58]. Other laboratories reported that HG treatment and diabetes produced a deregulated activation of ETS (E-twenty six), one of the largest families of transcription factors; this activation blocks the functional activity of progenitor cells and their commitment toward the EC lineage [59]. Another activated transcription factor is FOXO1; the mechanism is triggered by HG, acting through oxidative stress pathway. Subsequently, activated FOXO1 promotes inducible nitric oxide synthase (iNOS)-dependent NO-peroxynitrite generation, which leads in turn to LDL oxidation and eNOS dysfunction [60]. Moreover, in diabetic rats, the activation of transcription factor FOXO1 was linked to retinal microvascular cell loss [61]. In HG conditions, as well as in diabetes, activation of EC intracellular signaling pathways occurs. The recent data emphasize the activation of JAK2/STAT3 pathway and of Vascular Endothelial Growth Factor (VEGF) [62], of NAD(P)H oxidase followed by ROS generation [63], and of protein kinase C [13]. Among the last family of enzymes, the activation of PKC-a, -b1/2, and PKC-d isoforms is linked to the development of diabetes pathologies affecting both large vessels (in atherosclerosis and cardiomyopathy) and small vessels (in retinopathy, nephropathy, and neuropathy) [64]. Another modification induced by HG is the augmented expression of lipid peroxidation products [65], methylglyoxal (an AGE precursor molecule) [66], angiotensin II receptor 1 [67], neutrophins p75 receptor [68], Transient Receptor Potential ion channel protein 1 [69], poly(adenosine diphosphate-ribose) polymerase, [70], and fibronectin [20]. The intracellular calcium [Ca2+]i concentration is also enhanced by HG condition [71]. A further characteristic of EC dysfunction triggered by HG consists in the increased production of vasoconstrictor prostanoids (endoperoxides and prostacyclin) that activate Thromboxane A2 (TP) receptors on the underlying SMCs, contributing to amplified vascular wall contractility [72]. In contradistinction to the above-mentioned activating effects, HG concentration down-regulates a variety of EC molecules. Examples are the AGER1 (an AGE-receptor that counteracts Receptor for Advanced Glycation End products (RAGE)) [73], UCP2 [63], and eNOS. As shown by the latest reports, the mechanisms that may account for the decline in eNOS activity consist in enhanced ROS formation by NADPH oxidase and uncoupled eNOS [74], diminished

191

Hsp-90–eNOS interaction, due to the reaction between heat shock protein-90 and the inhibitor jB kinase (IKK)b [75], and inhibition of enzyme activity of dimethylarginine dimethylaminohydrolase, resulting in an accumulation of asymmetric dimethylarginine; the latter competes with the eNOS substrate, L-arginine, and inhibits NO formation [76]. Such a mechanism explains the other characteristic of EC dysfunction in hyperglycemia and Diabetes mellitus, i.e. the impeded endothelium-dependent relaxation of the vascular wall. The mitochondria are also vulnerable in HGexposed ECs; dysregulation in fuel-sensing molecules, AMP-activated protein kinase (AMPK), and the histone/protein deacetylase SIRT1 predisposes to Type 2 diabetes and atherosclerotic cardiovascular disease [77]. The cellular fueling system (the tricarboxylic acid oxidation cycle and the fatty acid b-oxidation pathway) is a target for various noxious stimuli, some generated within mitochondria themselves, such as the ROS [78]. Overproduction of the latter by mitochondrial electron transport chain serves as a causal link between elevated glucose and three major pathways responsible for hyperglycemic damage, i.e. the activation of the hexosamine pathway, the increased formation of AGE, and the activation of PKC isoforms [17]. Collectively, the imbalanced biochemical pathways (described above) portray the dysfunctional condition of EC facing HG concentration either in vitro or in vivo, as in diabetic vasculature. As an adaptation to the modified biochemistry and dysfunction, a modification to cellular structure is produced. In time, the structural modifications aggravate damaging the EC and the vascular wall.

4. High glucose concentration modifies vascular endothelial cell ultrastructure To evaluate HG-induced alterations of EC structure, the basic features of cells’ normal ultrastructure are briefly mentioned. EC plasmalemma expose differentiated macro- and micro-domains and membrane-associated receptors; the cells contain coated vesicles and caveolae (formerly known as plasmalemmal vesicles) endowed with specific receptors, and transendothelial channels [79]. The EC is quiescent in physiologic conditions (see Section 1), display rather rare organelles involved in biosynthetic activities (rER and Golgi apparatus) or in degradations (such as multivesicular bodies, a lysosomal like compartment, and lysosomes), and produce a unique, thin basal lamina. The ultrastructural alterations induced by HG in vascular ECs are documented both by in vitro (human aortic ECs cultured for 1–2 weeks in a medium supplemented with 25 mM D-glucose) and in vivo studies (the vasculature of mice and golden Syrian hamsters at 6 weeks and 6 months, respectively after streptozotocin injection) [80,81]. The common morphologic feature is the gradual and significant enrichment of biosynthetic organelles. As examples, the Golgi complex is evident in the aortic and capillary ECs (Fig. 1a and b), and the abundance in rER is obvious in ECs of the athero-susceptible aortic arch (Fig. 1c), of retinal venules (Fig. 1d), and of the femoral artery (Fig. 1e). Reportedly, diabetes imparts diffuse endothelial perturbation in both arterial and venous endothelium [31]. Morphologic evidence for the cells’ active metabolism is also provided through the presence of multivesicular bodies on electron-micrographs. One to three copies of these organelles are evident per EC in capillaries of retinal inner layer of diabetic animals (Fig. 1f). As a result of prolonged and direct contact with the hyperglycemic milieu, mitochondrial fragmentation occurs [51], and the cellular cytoskeleton is reorganized; the latter occupies almost the entire cytoplasm in myocardial capillary ECs (Fig. 1g); others have reported recently the reorganization of actin filaments in human ECs exposed to nonenzymatically glycated bovine serum albumin [12]. In addition, the intact microtubule and actin cytoskeleton is

192

D. Popov / International Journal of Diabetes Mellitus 2 (2010) 189–195

Fig. 1. Electron microscopic images of endothelial cells morphology in various vascular beds of streptozotocin-injected mice and hamsters (at 4 weeks since induction of diabetes). (a) Aortic endothelium; (b, g, and h) myocardial capillary endothelium; (c) aortic arch endothelium; (d) retinal venular endothelium; (e) femoral artery endothelium; (f and i) capillary endothelium of the retina inner layer. EC, endothelial cell; l, vascular lumen; mvb, multivesicular body; Go, Golgi apparatus; rER, rough endoplasmic reticulum; WPb, Weibel Palade body; f, cytoskeletal filaments; m, mitochondria; bl, basal lamina; c, collagen; ecm, extracellular matrix; CM, cardiomyocyte; SMC, smooth muscle cell. Bars: a, c, e, and g: 0.12 lm; b and h: 0.27 lm; d: 0.183 lm; f and i: 0.081 lm.

required for heparanase secretion by ECs exposed to hyperglycemia [23]. It is generally recognized that diabetes is associated with the thickening of the capillary ECs basal lamina; we show here that basal lamina turns into a folded structure (Fig. 1h), and generates reduplications with a multilayer appearance (Fig. 1i); this phenomenon was also observed at the interstitial capillaries of diabetic golden Syrian hamsters [82] and at the endoneurial capillaries of diabetic cats [83]. Functionally, the thickening of EC basal lamina impairs the amount and selectivity of transport of metabolic products and nutrients between circulation and the tissues [84]. Together, the above ultrastructural modifications are associated with the biosynthetic phenotype of vascular ECs. On electron-micrographs, the proliferative phenotype of endothelium is assessed by the presence of two centriols (paired organelles involved in mitosis) in close proximity to each other (Fig. 2). Since the ultrastructural changes appear as secondary to the biochemical alterations and EC dysfunction, in order to alleviate them, it is essential to find the optimal moment for therapeutic intervention, when the molecular alterations are still reversible. Examples for uncovering such issues are illustrated by ongoing research.

5. Potential clinical outcomes: toward the attenuation of ECs disturbances induced by hyperglycemia The main goal of translational medicine is to target the disturbed molecules/pathways in therapies aimed at attenuating HG-

triggered EC dysfunction. The majority of such strategies intend to reduce the oxidative stress generated in hyperglycemia/diabetes by using a plethora of antioxidant molecules. Mitochondria-derived ROS generation can be inhibited by Pigment Epithelium-Derived Factor (PEDF) that also decreases lipid peroxidation, and downregulates VEGF; thus it appears that PEDF treatment may be beneficial in diabetic retinopathy [62,65]. Moreover, azaserine functions as antioxidant and protects against hyperglycemic endothelial damage [23]. ROS generation in EC can be regulated by overexpression of certain molecules, such as transcription factor NFE2-Related Factor-2, AGER1, peroxisome proliferator-activated receptor-cco-activator 1-a, and AMP-activated protein kinase (AMPK) [73,85–87]. Activation of the latter suppresses 26S proteasome-mediated degradation of GTP-cyclohydrolase and up-regulates mitochondrial UCP2, resulting in the inhibition of superoxide anion production and prostacyclin synthase nitration; the final result is the prevention of the oxidative stress induced by hyperglycemia and the normalization of EC function [63,88]. In line with this, the mammalian homolog of the fish calcium regulatory hormone stanniocalcin-1 attenuates endothelial superoxide anions generation and activation of inflammatory pathways, and maintains tight junction proteins expression, preserving the EC monolayer seal [28]. As for the attenuation of inflammatory pathways activation, a potential therapeutic method that improves vascular barrier function consists in targeting key signaling molecules that mediate endothelial-junction-cytoskeleton dissociation [89]. A further approach is to inhibit the activity of certain molecules in specific

D. Popov / International Journal of Diabetes Mellitus 2 (2010) 189–195

193

Fig. 2. Proliferative endothelial cells in a myocardial capillary (a) and a brain capillary (b) of streptozotocin-injected hamster (at 4 weeks since induction of diabetes). EC, endothelial cell; l, vascular lumen; Go, Golgi apparatus; c, centriol; ecm, extracellular matrix; CM, cardiomyocyte. Bars: a: 0.12 lm; b: 0.183 lm.

vascular beds. Thus, in brain microvascular ECs inhibition of glycogen synthase kinase 3b reduced adhesion molecules expression, and decreased endothelial leukocyte adhesion under inflammatory conditions [90]; in diabetic retinopathy, inhibition of a4 integrin/ CD49d signaling pathway attenuated the diabetes-induced upregulation of NF-kB activation, reduced leukocyte adhesion to EC surface, and thus may hold promise for the clinical activity [91]. Another useful molecule is sphingosine-1-phosphate that inducts the expression of MAP kinase phosphatase-3, inhibiting the highglucose-mediated ERK1/2 phosphorylation and exerting an antiinflammatory effect on diabetic ECs [92]. The accelerated proliferation of retinal ECs and the decrease in pericytes number can be prevented through the addition of bioactive TGF-b and by aldose reductase inhibition [42]. In addition, enhancing endothelial Nampt activity may be beneficial in scenarios requiring ECs-based vascular repair and regeneration during HG, such as diabetes-related vascular disease [41]. Inhibition of ECs migration can be achieved by activating Activin Receptor-Like Kinase 1 [93]. Moreover, enhanced production of thrombospondin (TSP2) is described as an inhibitor of EC migration and capillary morphogenesis during neovascularization [94]. The latest reports also indicate possible strategies for reducing ECs apoptosis and senescence. Reportedly, the HG-induced apoptosis is reduced by fenofibrate [44] and is inhibited by quercetin sulfate/glucuronide, the metabolite of quercetin in blood [47]. Moreover, enhanced IGF-1 signaling inhibits glucose-induced apoptosis in human umbilical vein ECs by reducing mitochondrial dysfunction, and maintaining the mitochondrial retention of cytochrome-c [95]. The prevention of HG-induced EC senescence is exerted by the activation of Sirt1 (a NAD-dependent deacetylase) or disruption of p53 pathway [55]. L-arginine also exerts an antisenescence effect on human umbilical vein ECs exposed to 33 mmol/L glucose (via the PI3K/Akt pathway), and the authors claim it might be a therapeutic agent for diabetic vascular complications [53]. In addition, statins and peroxynitrite scavengers possess the ability to reduce senescence in laboratory models of disease (reviewed in Ref. [54]). In summary, three systems may at present alleviate HGinduced vascular EC dysfunctional profiles: exposure to inhibitors of certain molecules/pathways or down-regulation of specific molecules, activation or over-expression of particular molecules and intracellular pathways, and correction of biochemical alterations (if still reversible) towards rehabilitation of the quiescent condition. Although translational medicine may potentially target the intracellular molecules/pathways identified thus far (as shown above), the avenue is far from reaching an end. Still not well understood are the time-dependence and the vascular bed specificity of HG-induced EC biochemical changes, and substantiation of mechanisms which operate in continuous, fenestrated or discontinuous

endothelia. Less clear still are the intracellular pathways which lead to basal lamina reduplication in HG conditions. In addition, despite the recent progress made in reducing the HG-associated deleterious effects on EC, the existence of a causal relationship between telomere dysfunction and endothelial senescence is yet to be demonstrated [54]. 6. Concluding remarks This literature reviewed outlines the most recent mechanisms underlining EC dysfunction, triggered by over-physiological glucose concentration. Modified biochemistry is linked to the phenotypic switch of EC that acquires new properties (biosynthetic, inflammatory, adhesive, migratory, pro-atherogenic, pro-coagulant, proliferative, apoptotic or even senescent), rather than overlapping each other. Such changes are associated with the diabetic vasculopathy of large vessels and microvasculature. The results of the ongoing studies emphasize substantial progress in identifying ECdisturbed molecules/pathways under the insult of HG; and these may hold therapeutic potential, and be of benefit to the diabetic patient. Finally, it is important to point out that the reported effects of HG concentration are not the outcome of a unique insult, but of a multitude of deleterious molecules formed in HG conditions. These include the nonenzymatic glycation products, the AGE-proteins, the oxidant free radical species, among other components of the diabetic environment. Moreover, under insulin resistant conditions, as in Type 2 Diabetes mellitus, increased insulin concentrations and/or impaired insulin-signaling pathways may also contribute to endothelial dysfunction [13]. Therefore, the effects of HG emphasized in the largest part of the literature reviewed here should be viewed within the more complex perspective of the diabetic milieu. Acknowledgement This work was supported by funds from the Romanian Academy. References [1] David L, Mallet C, Keramidas M, Lamande N, Gasc JM, Dupuis-Girod S, et al. Bone morphogenetic protein-9 is a vascular quiescence factor. Circ Res 2008;102:914–22. [2] Herbert SP, Odell AF, Ponnambalam S, Walker JH. Activation of cytosolic phospholipase A2-a as a novel mechanism regulating endothelial cell cycle progression and angiogenesis. J Biol Chem 2009;284:5784–96. [3] Tanaka A, Itoh F, Nishiyama K, Takezawa T, Kurihara H, Itoh S, et al. Inhibition of endothelial cell activation by bHLH protein E2-2 and its impairment of angiogenesis. Blood 2010;115:4138–47.

194

D. Popov / International Journal of Diabetes Mellitus 2 (2010) 189–195

[4] Jiang A, Hu W, Meng H, Gao H, Qiao X. Loss of VLDL receptor activates retinal vascular endothelial cells and promotes angiogenesis. Invest Ophthalmol Vis Sci 2009;50:844–50. [5] Sako K, Fukuhara S, Minami T, Hamakubo T, Song H, Kodama T, et al. Angiopoietin-1 induces Krüppel-like Factor 2 expression through a Phosphoinositide 3-Kinase/AKT-dependent activation of myocyte enhancer factor. J Biol Chem 2009;284:5592–601. [6] Singh H, Brindle NP, Zammit VA. High glucose and elevated fatty acids suppress signaling by the endothelium protective ligand angiopoietin-1. Microvasc Res 2010;79:121–7. [7] Fukuhara S, Sako K, Noda K, Zhang J, Minami M, Mochizuki N. Angiopoietin-1/ Tie2 receptor signaling in vascular quiescence and angiogenesis. Histol Histopathol 2010;25:387–96. [8] Xu L, Komatsu M. Promoter cloning and characterization of the anti-vascular proliferation gene R-ras. Role of ETS- and SP-binding motifs. J Biol Chem 2009;284:2706–18. [9] Laughlin MH, Newcomer SC, Bender SB. Importance of hemodynamic forces as signals for exercise-induced changes in endothelial cell phenotype. J Appl Physiol 2008;104:588–600. [10] Cade WT. Diabetes-related microvascular and macrovascular diseases in the physical therapy setting. Phys Ther 2008;88:1322–35. [11] Addabbo F, Montagnani M, Goligorsky MS. Mitochondria and reactive oxygen species. Hypertension 2009;53:885–92. [12] Hirose A, Tanikawa T, Mori H, Okada Y, Tanaka Y. Advanced glycation end products increase endothelial permeability through the RAGE/Rho signaling pathway. FEBS Lett 2010;584:61–6. [13] Potenza MA, Gagliardi S, Nacci C, Carratu MR, Montagnani M. Endothelial dysfunction in diabetes: from mechanisms to therapeutic targets. Curr Med Chem 2009;16:94–112. [14] Nacci C, Tarquinio M, Montagnani M. Molecular and clinical aspects of endothelial dysfunction in diabetes. Intern Emerg Med 2009;4:107–16. [15] Singh DK, Winocour P, Farrington K. Endothelial cell dysfunction, medial arterial calcification and osteoprotegerin in diabetes. Br J Diabetes Vasc Dis 2010;10:71–7. [16] Highlander P, Shaw GP. Current pharmacotherapeutic concepts for the treatment of cardiovascular disease in diabetics. Therap Adv Cardiovasc Dis 2010;4:43–54. [17] Bakker W, Eringa EC, Sipkema P, van Hinsbergh VW. Endothelial dysfunction and diabetes: roles of hyperglycemia, impaired insulin signaling and obesity. Cell Tissue Res 2009;335:165–89. [18] Roy S, Trudeau K, Roy S, Behl Y, Dhar S, Chronopoulos A. New insights into hyperglycemia-induced molecular changes in microvascular cells. J Dental Res 2010;89:116–27. [19] Van Geest RJ, Klaassen I, Vogels IM, Van Noorden CJ, Schlingemann RO. Differential TGF-b signaling in retinal vascular cells: a role in diabetic retinopathy? Invest Ophthalmol Vis Sci 2010;51:1857–65. [20] Cherian S, Roy S, Pinheiro A, Roy S. Tight glycemic control regulates fibronectin expression and basement membrane thickening in retinal and glomerular capillaries of diabetic rats. Invest Ophthalmol Vis Sci 2009;50:943–9. [21] McDonald DM, Coleman G, Bhatwadekar A, Gardiner TA, Stitt AW. Advanced glycation of the Arg–Gly–Asp (RGD) tripeptide motif modulates retinal microvascular endothelial cell dysfunction. Mol Vis 2009;15:1509–20. [22] Martinez-Lemus LA, Hill MA, Meininger Gerald A. The plastic nature of the vascular wall: a continuum of remodeling events contributing to control of arteriolar diameter and structure. Physiology 2009;4:45–57. [23] Rajapakse AG, Ming X-F, Carvas JM, Yang Z. The hexosamine biosynthesis inhibitor azaserine prevents endothelial inflammation and dysfunction under hyperglycemic condition through antioxidant effects. Am J Physiol Heart Circ Physiol 2009;296:H815–22. [24] Liu Y, Biarnes M, Gerhardinger C. Effect of diabetes duration and high glucose on retinal Interleukin-1b expression. Invest Ophthalmol Vis Sci 2009;50:EAbstract 4984. [25] Bhatwadekar AD, Glenn JV, Curtis TM, Grant MB, Stitt AW, Gardiner TA. Retinal endothelial cell apoptosis stimulates recruitment of endothelial progenitor cells. Invest Ophthalmol Vis Sci 2009;50:4967–73. [26] Mzhavia N, Yu S, Ikeda S, Chu TT, Goldberg I, Dansky HM. Neuronatin: a new inflammation gene expressed on the aortic endothelium of diabetic mice. Diabetes 2008;57:2774–83. [27] Sarai K, Shikata K, Shikata Y, Omori K, Watanabe N, Sasaki M, et al. Endothelial barrier protection by FTY720 under hyperglycemic condition: involvement of focal adhesion kinase, small GTPases, and adherens junction proteins. Am J Physiol Cell Physiol 2009;297:C945–54. [28] Sheikh-Hamad D. Mammalian stanniocalcin-1 activates mitochondrial antioxidant pathways: new paradigms for regulation of macrophages and endothelium. Am J Physiol Renal Physiol 2010;298:F248–54. [29] Sweet IR, Gilbert M, Maloney E, Hockenbery DM, Schwartz MW, Kim F. Endothelial inflammation induced by excess glucose is associated with cytosolic glucose 6-phosphate but not increased mitochondrial respiration. Diabetologia 2009;52:921–31. [30] Zakkar M, Van der Heiden K, Luong LA, Chaudhury H, Cuhlmann S, Hamdulay SS, et al. Activation of Nrf2 in endothelial cells protects arteries from exhibiting a proinflammatory state. Arterioscler Thromb Vasc Biol 2009;29:1851–7. [31] Bucciarelli LG, Pollreisz A, Kebschull M, Ganda A, Kalea AZ, Hudson BI, et al. Inflammatory stress in primary venous and aortic endothelial cells of type 1 diabetic mice. Diab Vasc Dis Res 2009;6:249–61.

[32] Huang Q, Sheibani N. High glucose promotes retinal endothelial cell migration through activation of Src, PI3K/Akt1/eNOS, and ERKs. Am J Physiol Cell Physiol 2008;295:C1647–57. [33] Borniquel S, García-Quintáns N, Valle I, Olmos Y, Wild B, Martínez-Granero F, et al. Inactivation of Foxo3a and subsequent downregulation of PGC-1a mediate Nitric Oxide-induced endothelial cell migration. Mol Cell Biol 2010;30:4035–44. [34] Hamed S, Brenner B, Abassi Z, Aharon A, Daoud D, Roguin A. Hyperglycemia and oxidized-LDL exert a deleterious effect on endothelial progenitor cell migration in Type 2 Diabetes Mellitus. Thromb Res 2010;126:166–74. [35] Azcutia V, Abu-Taha M, Romacho T, Vazquez-Bella M, Matesanz N, Luscinskas FW, et al. Inflammation determines the pro-adhesive properties of high extracellular D-glucose in human endothelial cells in vitro and rat microvessels in vivo. PLoS One 2010;5:e10091. [36] Perrone L, Devi TS, Hosoya K-I, Terasaki T, Singh LP. Thioredoxin Interacting Protein (TXNIP) induces inflammation through chromatin modification in retinal capillary endothelial cells under diabetic conditions. J Cell Physiol 2009;221:262–72. [37] Zitman-Gal T, Green J, Pasmanik-Chor M, Oron-Karni V, Bernheim J. Endothelial pro-atherosclerotic response to extracellular diabetic-like environment: possible role of thioredoxin-interacting protein. Nephrol Dial Transplant 2010;25:2141–9. [38] Goldberg RB. Cytokine and cytokine-like inflammation markers, endothelial dysfunction, and imbalanced coagulation in development of diabetes and its complications. J Clin Endocrinol Metab 2009;94:3171–82. [39] Stegenga ME, van der Crabben SN, Blümer RME, Levi M, Meijers JCM, Serlie MJ, et al. Hyperglycemia enhances coagulation and reduces neutrophil degranulation, whereas hyperinsulinemia inhibits fibrinolysis during human endotoxemia. Blood 2008;112:82–9. [40] Molins B, Peña E, Padro T, Casani L, Mendieta C, Badimon L. Glucose-Regulated Protein 78 and platelet deposition: effect of Rosuvastatin. Arterioscler Thromb Vasc Biol 2010;30:1246–52. [41] Borradaile NM, Pickering JG. Nicotinamide phosphoribosyltransferase imparts human endothelial cells with extended replicative lifespan and enhanced angiogenic capacity in a high glucose environment. Aging Cell 2009;8:100–12. [42] Takamura Y, Tomomatsu T, Kubo E, Tsuzuki S, Akagi Y. Role of the polyol pathway in high glucose-induced apoptosis of retinal pericytes and proliferation of endothelial cells. Invest Ophthalmol Vis Sci 2008;49:3216–23. [43] Resnick N, Yahav H, Shay-Salit A, Shushy M, Schubert S, Chen L, et al. Fluid shear stress and the vascular endothelium: for better and for worse. Prog Biophys Mol Biol 2003;81:177–99. [44] Zanetti M, Stocca A, Dapas B, Farra R, Uxa L, Bosutti A, et al. Inhibitory effects of fenofibrate on apoptosis and cell proliferation in human endothelial cells in high glucose. J Mol Med 2008;86:185–95. [45] Chen G, Shen X, Yao J, Chen F, Lin X, Qiao Y, et al. Ablation of NF-kappaB expression by small interference RNA prevents the dysfunction of human umbilical vein endothelial cells induced by high glucose. Endocrine 2009;35:63–74. [46] Wang S, Hirschberg R. Diabetes-relevant regulation of cultured blood outgrowth endothelial cells. Microvasc Res 2009;78:174–9. [47] Chao CL, Hou YC, Chao PD, Weng CS, Ho FM. The antioxidant effects of quercetin metabolites on the prevention of high glucose-induced apoptosis of human umbilical vein endothelial cells. Br J Nutr 2009;101:1165–70. [48] Li A-F, Roy S. High glucose-induced downregulation of connexin 43 expression promotes apoptosis in microvascular endothelial cells. Invest Ophthalmol Vis Sci 2009;50:1400–7. [49] Hou CJ-Y, Tsai C-H, Su C-H, Wu Y-J, Chen S-J, Chiu J-J, et al. Diabetes reduces aortic endothelial gap junctions in ApoE-deficient mice: simvastatin exacerbates the reduction. J Histochem Cytochem 2008;56:745–52. [50] Zimering MB, Pan Z. Autoantibodies in Type 2 diabetes induce stress fiber formation and apoptosis in endothelial cells. J Clin Endocrinol Metab 2009;94:2171–7. [51] Trudeau K, Molina A, Roy S. High glucose alters mitochondrial morphology and membrane potential heterogeneity in retinal endothelial cells. Invest Ophthalmol Vis Sci 2009;50. E-Abstract 33/A20. [52] Patti M-E, Corvera S. The role of mitochondria in the pathogenesis of Type 2 diabetes. Endocr Rev 2010;31:364–95. [53] Zhong W, Zou G, Gu J, Zhang J. L-arginine attenuates high glucose-accelerated senescence in human umbilical vein endothelial cells. Diabetes Res Clin Pract 2010;89:38–45. [54] Erusalimsky JD. Vascular endothelial senescence: from mechanisms to pathophysiology. J Appl Physiol 2009;106:326–32. [55] Orimo M, Minamino T, Miyauchi H, Tateno K, Okada S, Moriya J, et al. Protective role of SIRT1 in diabetic vascular dysfunction. Arterioscler Thromb Vasc Biol 2009;29:889–94. [56] Dabir P, Marinic TE, Krukovets I, Stenina OI. Aryl Hydrocarbon Receptor is activated by glucose and regulates the thrombospondin-1 gene promoter in endothelial cells. Circ Res 2008;102:1558–65. [57] Chen S, Feng B, George B, Chakrabarti R, Chen M, Chakrabarti S. Transcriptional coactivator p300 regulates glucose-induced gene expression in endothelial cells. Am J Physiol Endocrinol Metab 2010;298:E127–37. [58] Brunssen C, Korten S, Brux M, Seifert S, Roesler J, Bornstein SR, et al. COUP-TFII is regulated by high glucose in endothelial cells. Horm Metab Res 2010;42:81–7.

D. Popov / International Journal of Diabetes Mellitus 2 (2010) 189–195 [59] Seeger FH, Chen L, Spyridopoulos I, Altschmied J, Aicher A, Haendeler J. Downregulation of ETS rescues diabetes-induced reduction of endothelial progenitor cells. PLoS One 2009;4(2):e4529. [60] Tanaka J, Qiang L, Banks AS, Welch CL, Matsumoto M, Kitamura T, et al. Foxo1 links hyperglycemia to LDL oxidation and endothelial nitric oxide synthase dysfunction in vascular endothelial cells. Diabetes 2009;58:2344–54. [61] Behl Y, Krothapalli P, Tesfahun D, Roy S, Graves DT. FOXO1 plays an important role in enhanced microvascular cell apoptosis and microvascular cell loss in Type 1 and Type 2 diabetic rats. Diabetes 2009;58:917–25. [62] Zheng Z, Chen H, Zhao H, Liu K, Luo D, Chen Y, et al. Inhibition of JAK2/STAT3mediated VEGF upregulation under high glucose conditions by PEDF through a mitochondrial ROS pathway in vitro. Invest Ophthalmol Vis Sci 2010;51:64–71. [63] Xie Z, Zhang J, Wu J, Viollet B, Zhou M-H. Upregulation of mitochondrial Uncoupling Protein-2 by the AMP-activated protein kinase in endothelial cells attenuates oxidative stress in diabetes. Diabetes 2008;57:3222–30. [64] Geraldes P, King GL. Activation of Protein Kinase C isoforms and its impact on diabetic complications. Circ Res 2010;106:1319–31. [65] Banumathi E, Sheikpranbabu S, Haribalaganesh R, Gurunathan S. PEDF prevents reactive oxygen species generation and retinal endothelial cell damage at high glucose levels. Exp Eye Res 2010;90:89–96. [66] Brouwers O, Niessen PM, Haenen G, Miyata T, Brownlee M, Stehouwer CD, et al. Hyperglycemia-induced impairment of endothelium-dependent vasorelaxation in rat mesenteric arteries is mediated by intracellular methylglyoxal levels in a pathway dependent on oxidative stress. Diabetologia 2010;53:989–1000. [67] Min Q, Bai YT, Jia G, Wu J, Xiang JZ. High glucose enhances angiotensin-IImediated peroxisome proliferation-activated receptor-gamma inactivation in human coronary artery endothelial cells. Exp Mol Pathol 2010;88: 133–7. [68] Caporali A, Pani E, Horrevoets AJG, Kraenkel N, Oikawa A, Sala-Newby GB, et al. Neurotrophin p75 Receptor (p75NTR) promotes endothelial cell apoptosis and inhibits angiogenesis. Implications for diabetes-induced impaired neovascularization in ischemic limb muscles. Circ Res 2008;103:e15–26. [69] Bishara NB, Ding H. Glucose enhances expression of TRPC1 and calcium entry in endothelial cells. Am J Physiol Heart Circ Physiol 2010;298:H171–8. [70] Mahrouf-Yorgov M, Marie N, Borderie D, Djelidi R, Bonnefont-Rousselot D, Legrand A, et al. Metformin suppresses high glucose-induced poly(adenosine diphosphate-ribose) polymerase overactivation in aortic endothelial cells. Metabolism 2009;58:525–33. [71] Wang S, Peng Q, Zhang J, Liu L. Na+/H+ exchanger is required for hyperglycaemia-induced endothelial dysfunction via calcium-dependent calpain. Cardiovasc Res 2008;80:255–62. [72] Vanhoutte PM, Shimokawa H, Tang EH, Feletou M. Endothelial dysfunction and vascular disease. Acta Physiol (Oxf) 2009;196:193–222. [73] Cai W, He JC, Zhu L, Chen X, Striker GE, Vlassara H. AGE-receptor-1 counteracts cellular oxidant stress induced by AGEs via negative regulation of p66shcdependent FKHRL1 phosphorylation. Am J Physiol Cell Physiol 2008;294:C145–52. [74] Jaimes EA, Hua P, Tian R-X, Raij L. Human glomerular endothelium: interplay among glucose, free fatty acids, angiotensin II, and oxidative stress. Am J Physiol Renal Physiol 2010;298:F125–32. [75] Mohan S, Konopinski R, Yan B, Centonze VE, Natarajan M. High glucoseinduced IKK–Hsp-90 interaction contributes to endothelial dysfunction. Am J Physiol Cell Physiol 2009;296:C182–92. [76] Lu CW, Guo Z, Feng M, Wu ZZ, He ZM, Xiong Y. Ex vivo gene transferring of human dimethylarginine dimethylaminohydrolase-2 improved endothelial dysfunction in diabetic rat aortas and high glucose-treated endothelial cells. Atherosclerosis 2010;209:66–73. [77] Ruderman Neil B, Julia Xu X, Nelson Lauren, Cacicedo José M, Saha Asish K, Lan Fan, et al. AMPK and SIRTI: a long-standing partnership? Am J Physiol Endocrinol Metab 2010;298:E751–60.

195

[78] Addabbo F, Ratliff B, Park H-C, Kuo M-C, Ungvari Z, Ciszar A, et al. The Krebs cycle and mitochondrial mass are early victims of endothelial dysfunction. Proteomic approach. Am J Pathol 2009;174:34–43. [79] Simionescu M, Popov D, Sima A. Endothelial dysfunction in diabetes. In: Cheta DM, editor. Vascular involvement in diabetes: clinical, experimental and beyond. Bucharest: Karger and the Publishing House of the Romanian Academy; 2004. p. 15–34. [80] Simionescu M, Popov D, Sima A, Hasu M, Costache G, Faitar S, et al. Pathobiochemistry of combined diabetes and atherosclerosis studied on a novel animal model: the hyperlipemic hyperglycemic hamster. Am J Pathol 1996;148:997–1014. [81] Popov D, Simionescu M. Cellular mechanisms and signalling pathways activated by high glucose and AGE-albumin in the aortic endothelium. Arch Physiol Biochem 2006;112:265–73. [82] Popov D. Towards understanding the mechanisms of impeded vascular function in the diabetic kidney. In: Simionescu M, Sima A, Popov D, editors. Cellular dysfunction in atherosclerosis and diabetes – reports from bench to bedside. Bucharest: The Publishing House of the Romanian Academy; 2004. p. 244–59. [83] Estrella JS, Nelson RN, Sturges BK, Vernau KM, Williams DC, LeCouteur RA, et al. Endoneurial microvascular pathology in feline diabetic neuropathy. Microvasc Res 2008;75:403–10. [84] Dokken BB. The pathophysiology of cardiovascular disease and diabetes: beyond blood pressure and lipids. Diabetes Spectrum 2008;21:160–5. [85] Xue M, Qian Q, Adaikalakoteswari A, Rabbani N, Babaei-Jadidi R, Thornalley PJ. Activation of NF-E2-Related Factor-2 reverses biochemical dysfunction of endothelial cells induced by hyperglycemia linked to vascular disease. Diabetes 2008;57:2809–17. [86] Torreggiani M, Liu H, Wu J, Zheng F, Cai W, Striker G, et al. Advanced Glycation End Product Receptor-1 transgenic mice are resistant to inflammation, oxidative stress, and post-injury intimal hyperplasia. Am J Pathol 2009;175:1722–32. [87] Won JC, Park J-Y, Kim YM, Koh EH, Seol S, Jeon BH, et al. Peroxisome Proliferator-Activated Receptor-c Coactivator 1-a overexpression prevents endothelial apoptosis by increasing ATP/ADP translocase activity. Arterioscler Thromb Vasc Biol 2010;30:290–7. [88] Wang S, Xu J, Song P, Viollet B, Zou M-H. In vivo activation of AMP-activated protein kinase attenuates diabetes-enhanced degradation of GTP Cyclohydrolase I. Diabetes 2009;58:1893–901. [89] Kumar P, Shen Q, Pivetti CD, Lee ES, Wu MH, Yuan SY. Molecular mechanisms of endothelial hyperpermeability: implications in inflammation. Expert Rev Mol Med 2009;11:e19. [90] Ramirez SH, Fan S, Zhang M, Papugani A, Reichenbach N, Dykstra H, et al. Inhibition of Glycogen Synthase Kinase 3b (GSK3b) decreases inflammatory responses in brain endothelial cells. Am J Pathol 2010;176:881–92. [91] Iliaki E, Poulaki V, Mitsiades N, Mitsiades CS, Miller JW, Gragoudas ES. Role of 4 Integrin (CD49d) in the pathogenesis of diabetic retinopathy. Invest Ophthalmol Vis Sci 2009;50:4898–904. [92] Whetzel AM, Bolick DT, Hedrick CC. Sphingosine-1-phosphate inhibits high glucose-mediated ERK1/2 action in endothelium through induction of MAP kinase phosphatase-3. Am J Physiol Cell Physiol 2009;296:C339–45. [93] Ray BN, Lee NY, How T, Blobe GC. ALK5 phosphorylation of the endoglin cytoplasmic domain regulates Smad1/5/8 signaling and endothelial cell migration. Carcinogenesis 2010;31:435–41. [94] Tang Y, Scheef EA, Wang S, Sorenson CM, Marcus CB, Jefcoate CR, et al. CYP1B1 expression promotes the proangiogenic phenotype of endothelium through decreased intracellular oxidative stress and thrombospondin-2 expression. Blood 2009;113:744–54. [95] Li Y, Wu H, Khardori R, Song YH, Lu YW, Geng YJ. Insulin-like growth factor-1 receptor activation prevents high glucose-induced mitochondrial dysfunction, cytochrome-c release and apoptosis. Biochem Biophys Res Commun 2009;384:259–64.