articles
nature publishing group
Integrative Physiology
Oxidant Stress in Healthy Normal-weight, Overweight, and Obese Individuals Louise A. Brown1,2, Catherine J. Kerr1, Paul Whiting3, Nicholas Finer2,4, Jane McEneny5 and Tony Ashton1 This study was undertaken to investigate the association among BMI and lipid hydroperoxide (LH), total antioxidant status (TAS), superoxide dismutase (SOD), and reduced glutathione (GSH). Ninety (n = 90) healthy males and females (n = 23/67) (29 normal weight (BMI: 22.74 ± 0.25 kg/m2), 36 overweight (BMI: 27.18 ± 0.23 kg/m2), and 25 obese (33.78 ± 0.48 kg/m2)) participated in the study. Data collected included anthropometric measures, fasting blood glucose, lipid profile, LH, TAS, and enzymatic antioxidants (SOD, and reduced GSH). The results of the study showed that obese individuals had significantly increased LH levels compared to normal-weight individuals (obese vs. normal weight (0.88 ± 0.05 vs. 0.67 ± 0.03 µmol/l, P < 0.01)) but the increased levels were not significantly different when compared to the overweight group (obese vs. overweight (0.88 ± 0.05 vs. 0.79 ± 0.05 µmol/l)). No other consistent significant differences in TAS, SOD, and GSH were identified between groups. This study concluded that only obesity and not moderate overweight elevates LH levels. Furthermore, the levels of TAS, SOD, and GSH in obesity do not explain the increased LH levels observed in obesity. Obesity (2009) 17, 460–466. doi:10.1038/oby.2008.590
Introduction
Oxidant stress in obesity may be an important pathogenic mechanism in the obesity-associated metabolic syndrome (1), which includes the coexistence of several risk factors for atherosclerosis, including hyperglycemia, dyslipidemia, and hypertension. Oxidant stress has also been shown to play a critical role in the pathogenesis of various diseases such as cancer, cardiovascular disease, and diabetes mellitus (2). Oxidant stress results when free-radical formation is greatly increased or protective antioxidant mechanisms are compromised (3). Free-radical formation or reactive oxygen species includes the oxygen-derived free radicals and oxidants, superoxide, hydrogen peroxide, nitric oxide, and hydroxyl radicals (4), which are capable of damaging DNA, lipids, and proteins (5). Reactive oxygen species levels are controlled though an intricate network of enzymes and antioxidant molecules that are responsible for consumption of reactive oxygen species (6). Several research studies have suggested that obesity is associated with increased oxidant stress (7–13) i.e., increased freeradical production and/or depleted cellular antioxidant defence systems (3). Possible mechanisms contributing to the obesityassociated oxidant stress include increased oxygen consumption and subsequent radical production via mitochondrial respiration, diminished antioxidant capacity, increased fat
deposition, and cell injury causing increased rates of radical formation such as O2− and OH− (14). In addition, hyperglycemia, hypertension, and hyperleptinemia are also possible sources of increased oxidant stress in the obese state (15). It is not known whether obesity-associated oxidant stress is related to excess adipose tissue accumulation or is a consequence of obesity-related diseases i.e., hypertension, hyperlipidemia, hyperleptinemia, and hyperglycemia (16). This study aimed to identify the relationship among BMI (normal weight, overweight, and obese) and lipid hydroperoxide (LH), total antioxidant status (TAS), superoxide dismutase (SOD), and reduced glutathione (GSH) levels. Methods And Procedures Subject characteristics The study group consisted of 90 (n = 90) healthy male and female volunteers (closely sex-matched by ratio in each subject group) who were divided into three categories of BMI: 18.5–24.99, 25–29.99, and ≥30 kg/m2 according to the World Health Organization classifications of normal weight, overweight, and obese, respectively (17) (29 normal weight (BMI: 22.74 ± 0.25 kg/m2), 36 overweight (BMI: 27.18 ± 0.23 kg/ m2), and 25 obese (BMI: 33.78 ± 0.48 kg/m2) healthy male and females (n = 23/67)) (see Table 1 for subject characteristics). Volunteers were invited to take part in the study by local advertisement. An inclusion criterion was age between 18 and 50 years old and BMI between 18.5 and 45 kg/m2. Subjects with diabetes, cardiovascular or cerebrovascular
1 School of Physical Education and Sport Sciences, University of Bedfordshire, Bedford, UK; 2Centre for Obesity Research, Luton and Dunstable Hospital NHS Trust, Luton, UK; 3Faculty of Health and Life Sciences, Hawthorn Building, Leicester, UK; 4Wellcome Clinical Research Facility, Addenbrooke’s Hospital, Cambridge, UK; 5 Department of Medicine, Nutrition and Metabolism Group, The Queen’s University of Belfast, Centre for Clinical and Population Sciences, Belfast, UK. Correspondence: Louise A. Brown (
[email protected])
Received 2 July 2007; accepted 26 May 2008; published online 8 January 2009. doi:10.1038/oby.2008.590
460
VOLUME 17 NUMBER 3 | MARCH 2009 | www.obesityjournal.org
articles Integrative Physiology disease, hepatic or renal disease, tobacco abuse, or those on hormone replacement therapy were excluded. In addition subjects were excluded if they were hypertensive (with or without treatment), taking treatment for dyslipidemia, taking any antioxidant supplementations, or a smoker. The clinical definition for diabetes included fasting plasma glucose ≥7.0 mg/dl (World Health Organization criteria), oral hypoglycemic use, or insulin use (18). High blood pressure (BP) was defined as mean systolic BP ≥140 mm Hg, diastolic BP ≥90 mm Hg (Joint National Committee VI) (19). Written informed consent was obtained from all the subjects after they had been given a full explanation of the study. The research was given ethical approval by Bedfordshire Local Research Ethics Committee. Study design Subjects visited the Centre for Obesity Research on one occasion between 9 and 12 am and were instructed to fast for 10–12 h and refrain from exercise, caffeine, and alcohol intake for 48 h before the study visit. Subjects were also asked to maintain their usual dietary pattern. Anthropometric measurements Height was measured to the nearest 0.1 cm without shoes using a freestanding stadiometer. Weight and body composition were assessed using whole body air displacement plethysmography, BOD POD (Life Measurement Instruments, Concord, CA). The intra- and interassay coefficients of variation for body fat using the BOD POD were 4.7 and 5.8%, respectively. Waist circumference was measured using a tape measure to the nearest 1 mm at a point midway between the costal margin and the iliac crest and in line with the mid-axillary line (17). Cardiovascular measurements Heart rate was measured at the radial artery and BP was measured using a mercury sphygmomanometer. Hematological sampling and analysis Venous blood was collected following a 12-h overnight fast from an antecubital forearm vein, using the Vacutainer method (Becton– Dickinson, Oxford, UK). Blood samples were distributed between EDTA and plain serum tubes. Samples were centrifuged (3,000 rpm for 10 min), and then stored as separate aliquots at −70 and −20 °C. For assessment of SOD, the red cells in EDTA were washed twice with saline, then hemolyzed by adding 1.5 ml of water, and the subsequent 2 ml solution was stored at −70 °C. For assessment of GSH, 200 µl EDTA whole blood was added to 2.0 ml of distilled water. Lysate (200 µl) was then removed and stored at −20 °C for estimation of hemoglobin. Precipitating solution (3 ml; 100 ml containing 1.67 g of glacial metaphosphoric acid, 0.2 g of disodium EDTA, and 30 g of sodium chloride) was added to the remaining 2 ml of hemolysate. After standing for 5 min the mixture was filtered through medium grade filter paper and stored in Eppendorf tubes at −20 °C. In addition to the assays described below, other biochemical measurements including fasting plasma glucose, plasma cholesterol, plasma low-density lipoprotein, plasma high-density lipoprotein, and plasma triglycerides were performed. Reagent kits were supplied by Roche Diagnostics, UK.
Lipid peroxidation. Lipid peroxidation was estimated by measuring LH concentrations in serum using an assay developed by Wolff (20), which is a ferrous iron/xylenol orange assay that quantifies the susceptibility to iron-induced LH formation in blood. In general, in dilute acids, hydroperoxides oxidize ferrous ions to ferric ions, and the resultant ferric ions are used as an indirect measure of hydroperoxide content, which can be detected by ferric-sensitive dyes. A bluepurple colored complex is produced with the selective binding of xylenol orange to the ferric ions produced. The absorption was then measured at 560 nm. obesity | VOLUME 17 NUMBER 3 | MARCH 2009
TAS. The ferric reducing ability of plasma (FRAP) assay was used to assess TAS in plasma by the method of Benzie and Strain (21). In brief, the FRAP assay is a simple and reliable test measuring the total reactive (“antioxidant”) capacity of biological fluids. Ferric to ferrous ion reduction at low pH causes a blue-colored ferrous–tripyridyltriazine complex to form. FRAP values are obtained by comparing the absorbance change at 593 nm in test reaction mixtures with those containing ferrous ions of known concentration. Erythrocyte-GSH. GSH was measured using the method of Beutler (22). Virtually all of the nonprotein sulfhydryl compounds of red cells are in the form of GSH. 5,5′ Dithiobis (2-nitro benzoic acid) is a disulfide compound readily reduced by sulfhydryl compounds that forms a highly colored yellow anion. The optical density of this yellow substance is measured at 412 nm and is directly proportional to the GSH concentration present. Erythrocyte SOD. SOD in erythrocytes was measured according to the method of Winterbourn et al. (23). Red blood cells were hemolyzed with cold distilled water, and extraction was performed using an ethanol/chloroform mixture (1:1). SOD activity (units per gram of hemoglobin) was then measured in the supernatant. Statistics Values are reported as the mean ± s.e.m. or median and range. Data were analyzed by nonparametric methods to avoid assumptions about the distribution of the measured variables. The Kruskal–Wallis analysis of variance test was used to compare groups and the Bonferroni-corrected Mann–Whitney U-test was used as a more conservative measure of significance for multiple comparisons. Associations between parameters were assessed using the Spearman rank correlation test. All tests were considered significant at P < 0.05 or, in case of k comparisons, when P < 0.05/k. Statistical analysis was performed using a computer software package (SPSS for Windows, version 13.0). Results
Subject characteristics for the normal-weight, overweight, and obese groups are shown in Table 1. Body weight, body fat (kg), and waist circumference all progressively increased significantly (P < 0.001) with higher BMI values. Percentage total body fat also increased with higher BMI values (normal weight vs. obese, P < 0.01 and overweight vs. obese, P < 0.01), but the overweight group only showed an increased trend for a higher percentage total body fat compared to the normal-weight group (P = 0.018). Other significant differences between groups with increasing BMI values include increased diastolic BP (normal weight vs. overweight, P < 0.01 and normal weight vs. obese, P < 0.01), increased systolic BP (normal weight vs. overweight, P < 0.01 and normal weight vs. obese, P < 0.01), increased cholesterol (normal weight vs. overweight, P < 0.01 and normal weight vs. obese, P < 0.01), decreased high-density lipoprotein concentration (normal weight vs. obese, P < 0.01), increased low-density lipoprotein concentration (normal weight vs. overweight, P < 0.01 and normal weight vs. obese, P < 0.01). Fasting blood glucose and triglycerides were similar in all subjects. In addition, the Bonferroni-corrected Mann–Whitney U-test did not detect any significant differences in age among the normal-weight, overweight, and obese groups. The mean values for LH, TAS, SOD, and GSH for the normalweight, overweight, and obese groups are given in Table 2. LH progressively increased with higher BMI values (Kruskal– Wallis test, P = 0.021) with the obese group demonstrating 461
articles Integrative Physiology Table 1 Subject characteristics Characteristics Age, years Male:female ratio
Normal weight (NW) (n = 29) 31.03 ± 1.70 7:22
Overweight (OW) (n = 36)
P value
Obese (O) (n = 25)
35.14 ± 1.51
NW vs. OW
NW vs. O
OW vs. O
0.12
0.02
0.13
