Despite evidences of association between basic redox biology and metabolic syndrome (MetS), few studies have evaluated indices that account for multiple oxidative effectors for MetS. Oxidative balance score (OBS) has indicated the role of oxidative stress in chronic disease pathophysiology. In this study, we evaluated OBS as an oxidative balance indicator for estimating risk of MetS with 6414 study participants. OBS is a multiple exogenous factor score for development of disease; therefore, we investigated interplay between oxidative balance and genetic variation for development of MetS focusing on biological pathways by using gene-set-enrichment analysis. As a result, participants in the highest quartile of OBS were less likely to be at risk for MetS than those in the lowest quartile. In addition, persons in the highest quartile of OBS had the lowest level of inflammatory markers including C-reactive protein and WBC. With GWAS-based pathway analysis, we found that VEGF signaling pathway, glutathione metabolism, and Rac-1 pathway were significantly enriched biological pathways involved with OBS on MetS. These findings suggested that mechanism of angiogenesis, oxidative stress, and inflammation can be involved in interaction between OBS and genetic variation on risk of MetS.
Oxidative stress is a complex and multifactorial process that results from an imbalance between antioxidant protection and reactive oxygen species produced by prooxidants [
Studies of diet and health have demonstrated that combination of antioxidant factors can be more strongly associated with disease risk than any single nutrient [
The number of individuals with metabolic syndrome (MetS) is increasing worldwide. It was reported that MetS affects 20% to 30% of the population in developed countries [
It is well accepted that not only exogenous factors including diet and smoking but also endogenous factors like genetic variation contribute to the development of MetS. Genome-wide association studies (GWAS) have identified important susceptible loci [
Genetic variation also modulates reactive oxidative stress, which may influence several diseases through their interaction with OBS components [
The purpose of this study is to investigate the relationship between OBS and risk of MetS. First, we used OBS as an oxidative balance indicator for estimating risk of MetS. OBS is multiple exogenous factor score for development of disease; therefore, we investigated interplay between oxidative balance and genetic variation for development of MetS. We focused on biological pathways by using gene-set-enrichment analysis (GSEA) for investigating how OBS affects development of MetS with genetic factors.
This study was based on the Korea Association Resource (KARE) data included in the Korean Genome Epidemiology Study (KoGES). The details of the study design and procedures used in the KoGES have been described previously [
MetS is characterized by the clustering of several components including abdominal obesity, hypertension, dyslipidemia, insulin resistance, and glucose intolerance that are important precursor of cardiovascular disease and type 2 diabetes. MetS was defined by the presence of three or more of the following five components according to the NCEP-ATP III criteria [
Diabetes was diagnosed in subjects who had any one of the following: (1) fasting plasma glucose ≥ 126 mg/dL; (2) postprandial 2 h plasma glucose ≥ 200 mg/dL after a 75 g OGTT; and (3) HbA1c ≥ 6.5% [
The OBS was calculated by combining information from a total of 7 a priori selected pro- and antioxidant factors, including three prooxidant (smoking, alcohol consumption, and total iron intake) and four antioxidant exposures (
Baseline characteristics of this study by OBS quantile
Characteristics | Q1 ( | Q2 ( | Q3 ( | Q4 ( | |
---|---|---|---|---|---|
Age (years) | | | | | <0.001 |
Sex | <0.001 | ||||
Male | 1432 (69) | 569 (49) | 817 (37) | 199 (22) | |
Female | 637 (31) | 600 (51) | 1378 (63) | 711 (78) | |
Region | <0.001 | ||||
Rural (Ansung) | 1194 (58) | 618 (53) | 989 (45) | 419 (46) | |
Urban (Ansan) | 875 (42) | 551 (47) | 1206 (55) | 491 (54) | |
BMI | | | | | 0.105 |
Education (years) | <0.001 | ||||
Elementary school or less | 261 (14) | 300 (12) | 98 (9) | 88 ( | |
Middle school graduate | 814 (44) | 1122 (46) | 475 (44) | 420 (44) | |
High school or higher | 770 (42) | 1036 (42) | 503 (47) | 440 (47) | |
Monthly income | <0.001 | ||||
<1000 | 724 (40) | 876 (36) | 318 (30) | 259 (27) | |
1000–2000 | 422 (23) | 550 (23) | 238 (22) | 200 (21) | |
≥2000 | 682 (37) | 450 (41) | 516 (48) | 486 (52) | |
Total energy intake | | | | | <0.001 |
Depending on contribution of OBS components on oxidative balance, we scored OBS with 2 additional weighting methods besides equal weighting method. For beta coefficient weighting, the coefficient estimates for each of the components obtained from the regression model were used to calculate weights for OBS-beta coefficient. Coefficients were multiplied by −1 so that components inversely associated with metabolic disease risk had positive weight and vice versa. Principal component analysis (PCA) was a weight method using variation reduction [
The KARE dataset consists of the individual SNP chip genotypes and the epidemiological/clinical phenotypes for studying the genetic components of Korean public health. DNA samples were isolated from the peripheral blood of all participants and were genotyped with Affymetrix Genome Wide Human SNP array 5.0 (Affymetrix Inc, Rockville, MD, USA). The obtained KARE dataset passed the quality control criteria and was reported in previous GWAS reports [
It has now been recognized that a majority of biological behaviors are manifested from a complex interaction of biological pathway. Pathway-based approaches using GWAS data are now used routinely to study complex disease like MetS [
In our study, 344,396 SNPs were mapped to genes with 100 kb boundaries and pathways with <20 genes or >200 genes were excluded from further analysis to reduce the multiple-testing issue and to avoid testing overly narrow or broad functional categories [
OBS components except for physical activity were not normally distributed, so they were log-transformed when assigning OBS. Tests for linear trend were performed using a sum of scores with values from Q1 to Q4, consistent with the quartile grouping. Each OBS weighting method was divided into quartiles, with the lowest quartile (predominance of prooxidants) used as reference. Even though obesity is a main factor of oxidative stress, BMI is a strong risk factor for MetS; therefore, we remove it from OBS components but controlled for it in the statistical models.
Logistic regression analyses were used to examine the relationship between OBS and incidence of MetS adjusting for age, geographic area, sex, and BMI. For comparing the effect of different weighting methods, receiver operating characteristics (ROC) curves and the respective areas under the curves (AUCs) were calculated.
For investigating relationship between OBS and inflammation, we examined the association of OBS with CRP (C-reactive protein) and WBC (white blood cell count). CRP was not normally distributed, and so it was log transformed. The results of the linear regression models were expressed as regression coefficients and their corresponding 95% confidence intervals adjusted for same potential confounding factors as the previous ones.
For elucidating biological process through interaction of gene and oxidative stress by pathway analysis, we used to test gene-environment interaction by performing a 1df test of
A total of 6414 participants (mean age, 56.07 years, SD, 8.29 years) were included in this study. The distribution of characteristics according to quartile of OBS-equal weight is shown in Table
Oxidative balance score assignment scheme.
OBS components | Score assignment scheme |
---|---|
Iron | 0 = high (3rd tertile), 1 = medium (2nd tertile), and 2 = low (1st tertile) |
Vitamin C | 0 = low (1st tertile), 1 = medium (2nd tertile), and 2 = high (3rd tertile) |
Retinol | 0 = low (1st tertile), 1 = medium (2nd tertile), and 2 = high (3rd tertile) |
Carotene | 0 = low (1st tertile), 1 = medium (2nd tertile), and 2 = high (3rd tertile) |
Physical activity (Phy-MET) | 0 = low (1st tertile), 1 = medium (2nd tertile), and 2 = high (3rd tertile) |
Smoking | 0 = current smoker, 1 = former smoker, and 2 = never smoker |
Alcohol | 0 = heavy drinker (≥50 g/day), 2 = nonheavy drinker (<50 g/day) |
As expected, intake of antioxidants including vitamin C, carotene, and retinol was higher among participants with higher OBS values (data not shown). Contrary to expectation, intakes of iron were higher in the upper OBS quartile group. Participants in the higher OBS quartiles were also more likely to never be smokers, nonregular drinkers, and higher physical activity (data not shown). Individual component level of OBS between non-MetS and MetS was provided in Table
Individual component level of OBS between non-MetS and MetS groups.
Characteristics | Non-MetS | MetS |
---|---|---|
Nutrients | ||
Iron, mg/day | | |
Vitamin C, mg/day | | |
Retinol, | | |
Carotene, | | |
Alcohol consumption, g/day | | |
Nonregular drinker | 4575 | 1470 |
Regular drinker | 248 | 101 |
Smoking status, | ||
Never | 2944 | 1052 |
Former | 977 | 238 |
Current | 911 | 285 |
Physical activity, MET/day | | |
We firstly confirmed that there is no association between each OBS component and MetS (data not shown). For categorical analyses, participants in the highest quartile of all three OBS by weighting methods were less likely to be at risk for MetS than those in the lowest quartile, with statistical significance (Table
Association of the OBS with metabolic syndrome by weighing method
OBS | Number of cases (control) | OR (95% CI) | | |
---|---|---|---|---|
OBS-equal weight (AUC = 0.823) | ||||
Quantile 1 | 547 (1522) | 1 | <0.01 | |
Quantile 2 | 314 (855) | 0.94 (0.78–1.15) | 0.56 | |
Quantile 3 | 510 (1685) | 0.81 (0.68–0.97) | 0.02 | |
Quantile 4 | 186 (724) | 0.65 (0.51–0.82) | <0.01 | |
OBS-equal weight | ||||
Quantile 1 | 353 (1351) | 1 | <0.01 | |
Quantile 2 | 205 (774) | 0.91 (0.72–1.14) | 0.40 | |
Quantile 3 | 328 (1513) | 0.79 (0.64–0.97) | 0.02 | |
Quantile 4 | 118 (644) | 0.60 (0.45–0.81) | <0.01 | |
OBS-beta coefficient (AUC = 0.824) | ||||
Quantile 1 | 378 (1204) | 1 | <0.01 | |
Quantile 2 | 397 (1161) | 0.65 (0.52–0.81) | <0.01 | |
Quantile 3 | 431 (1209) | 0.67 (0.49–0.90) | <0.01 | |
Quantile 4 | 355 (1212) | 0.56 (0.76–0.41) | <0.01 | |
OBS-beta coefficient | ||||
Quantile 1 | 239 (1089) | 1 | <0.01 | |
Quantile 2 | 272 (1047) | 0.66 (0.50–0.87) | <0.01 | |
Quantile 3 | 376 (1050) | 0.66 (0.47–0.91) | 0.01 | |
Quantile 4 | 218 (1107) | 0.56 (0.38–0.82) | <0.01 | |
OBS-PCA (AUC = 0.824) | ||||
Quantile 1 | 375 (1203) | 1 (reference) | <0.01 | |
Quantile 2 | 389 (1174) | 0.64 (0.51–0.79) | <0.01 | |
Quantile 3 | 419 (1189) | 0.68 (0.50–0.92) | 0.01 | |
Quantile 4 | 372 (1216) | 0.55 (0.40–0.75) | <0.01 | |
OBS-PCA | ||||
Quantile 1 | 233 (1087) | 1 | <0.01 | |
Quantile 2 | 251 (1068) | 0.71 (0.55–0.92) | <0.01 | |
Quantile 3 | 292 (1037) | 0.66 (0.46–0.96) | 0.03 | |
Quantile 4 | 228 (1098) | 0.62 (0.43–0.91) | 0.01 |
Table
Associations of the OBS with metabolic related disorders by OBS quantile.
Characteristics | Q1 | Q2 | Q3 | Q4 | |
---|---|---|---|---|---|
Metabolic syndrome, | 547 (26) | 314 (27) | 510 (23) | 186 (20) | <0.001 |
Metabolic components, | |||||
Abdominal obesity | 686 (30) | 416 (36) | 753 (34) | 305 (34) | 0.071 |
Hypertriglyceridemia | 748 (36) | 382 (33) | 631 (29) | 235 (26) | <0.001 |
Low HDL cholesterol | 1087 (53) | 667 (57) | 1294 (59) | 539 (59) | 0.005 |
High blood pressure | 486 (24) | 287 (25) | 432 (20) | 161 (18) | 0.267 |
High fasting glucose | 465 (23) | 244 (21) | 392 (18) | 131 (14) | 0.163 |
MetS score | | | | | <0.001 |
CRP (mg/dL) | | | | | 0.013 |
WBC (103/ | | | | | <0.001 |
CRP, C-reactive protein; WBC; white blood cell count;
Oxidative stress and inflammation have been associated with MetS and oxidative stress can increase inflammation and vice versa. Confirming the role of OBS as oxidative stress indicator, we tested association of OBS with inflammatory markers. The CRP and WBC in blood serve as inflammatory markers although these markers have nonspecific features [
Association of the OBS with inflammatory markers
OBS | Beta coefficient | 95% CI | | | Beta coefficient | 95% CI | | |
---|---|---|---|---|---|---|---|---|
CRP (mg/dL) | WBC (103/ | |||||||
Quantile 1 | 0 | (Reference) | <0.01 | 0 | (Reference) | <0.01 | ||
Quantile 2 | −0.22 | −0.30 to −0.14 | <0.01 | −0.77 | −0.92 to −0.63 | <0.01 | ||
Quantile 3 | −0.12 | −0.23 to −0.01 | 0.04 | −0.82 | −1.02 to −0.61 | <0.01 | ||
Quantile 4 | −0.28 | −0.40 to −0.17 | <0.01 | −0.98 | −1.19 to −0.77 | <0.01 |
With 352,228 SNPs having
To conduct the biological pathway analysis, we used all SNPs that were used in GWAS analysis. As shown in Table
Pathway-based analysis for interaction between OBS and genetic variation for metabolic syndrome.
Resources | Biological process | Description | | FDR | Significant genes/selected genes/all genes |
---|---|---|---|---|---|
KEGG | VEGF signaling pathway | Genes involved in VEGF signaling pathway. | <0.001 | 0.020 | 25/54/70 |
KEGG | Glutathione metabolism | Genes involved in glutathione metabolism | <0.001 | 0.022 | 12/26/39 |
Biocarta | Rac-1 pathway | Rac-1 is a Rho family G protein that stimulates formation of actin-dependent structures | <0.001 | 0.045 | 14/20/22 |
Biocarta | Rho pathway | Rac-1 is a Rho family G protein that stimulates formation of actin-dependent structures such as filopodia and lamellipodia | 0.001 | 0.062 | 14/25/31 |
Biocarta | TNFR1 pathway | Tumor necrosis factor alpha binds to its receptor TNFR1 and induces caspase-dependent apoptosis | 0.005 | 0.085 | 13/23/29 |
In this study, we examined the possibility of OBS as a predictor of MetS risk and found that higher OBS, which indicates predominance of antioxidant exposures, was associated with significant reduction of the risk of MetS. Considering different contribution of each component on risk of MetS, we used different weighting scheme for combination of pro- and antioxidant exposure into a single score. Dash et al. mentioned that combination of antioxidants and prooxidants into a single score may be more powerful measurement of oxidative stress than approaches that use a single antioxidant or prooxidant [
CRP and WBC are inflammation markers which have been shown in multiple prospective epidemiological studies to predict the risk of cardiovascular disease and MetS [
Pathway analysis on MetS focuses on the combined effects of multiple SNPs within a gene and multiple genes within a pathway that are grouped according to their shared biological function. Current GWAS-derived pathway analysis can provide insights into mechanism of disease and biological pathways considering interaction between genes and environment factors [
VEGF (vascular endothelial growth factor) and its receptor, VEGFR, have been shown to play major roles not only in physiological but also in most pathological angiogenesis. Angiogenesis requires initiation by proangiogenic factors, such as VEGF, and mediated the Rho GTPases Rac-1. In addition angiogenesis via VEGF involves the main mechanism of oxidative stress [
Glutathione (GSH)/glutathione disulfide is the major redox couple in animal cells and effectively scavenges free radicals and other reactive oxidative species directly and indirectly through enzymatic reactions. GHS has critical role in regulating lipid, glucose, and amino acid utilization [
Rac-1, small GTP binding protein, plays many important biological functions in cells adhesion, migration, and inflammation. Rac-1 is a mediator of VEGF signaling pathway that involves permeability and cell migration. Rac-1 is associated with adiponectin and has a direct connection with hyperglycemia and ß-cell apoptosis [
Several studies using OBS mentioned their limitations for lacking of endogenous factors that modify oxidative stress [
In summary, we identified that individuals with high OBS may have lower risk of MetS. In addition, we showed interactions between genetic polymorphism and OBS through several signaling pathways. Such information would provide scientific knowledge of comprehensive biological contribution by complex exposures to pro- and antioxidants. Further research is needed to understand that modification of oxidative balance could be preventive strategy for the development of MetS.
The authors declare that they have no competing interests.
This work was supported by the Bio-Synergy Research Project of the Ministry of Science, ICT and Future Planning through the National Research Foundation (2013M3A9C4078158).