Correlation between vitamin A, E, coenzyme Q(10) and degree of insulin resistance in obese and non-obese subjects.

The aim of the present study was to investigate correlation between plasma vitamin A, vitamin E, serum coenzyme Q(10) levels and degree of insulin resistance in obese and normal weight people. The study was performed on 98 (21 Male, 77 Female) obese people and 78 (20 Male, 58 Female) control subjects. Vitamin A, E and coenzyme Q(10) levels were adjusted to the lipid levels. Adjusted vitamin A and E and coenzyme Q(10) levels of the obese female group were significantly lower than those of the control female group. Adjusted vitamin A and coenzyme Q(10) levels of the obese male group were significantly lower than those of the control male group. Insulin resistance level of the obese female and male groups were significantly higher than that of the control female and male groups. There were no significant correlations between serum coenzyme Q(10), plasma vitamin A and E levels and insulin resistance in obese and control subjects. Our findings show that it is essential to use the lipid adjusted levels of lipid soluble nutrients in obesity. Also, we have found no association between insulin resistance and vitamin A, vitamin E and coenzyme Q(10) levels in obese subjects.

Introduction

The prevalence of obesity has increased in world with a major impact on public health because of its role as risk factor for many diseases including cardiovascular disease and cancer,( with a consequent reduction in life expectancy. Obesity has been associated with low-grade inflammation,( playing a role in the pathogenesis of insulin resistance( and cardiovascular risk( and with oxidative stress, an imbalance between the production of reactive oxygen species and antioxidant defences that is also involved in the pathogenesis of cardivascular disease and cancer.(

The antioxidants α-tocopherol, the dominant component of vitamin E, vitamin A and coenzyme Q10 (CoQ10) contribute to the body’s defense agaist reactive oxygen species.( Epidemiologic studies have consistently shown that higher intakes of vitamin A, vitamin E and CoQ10 are associated with reduced risk of several chronic diseases, including heart failure, age-related macular degeneration and some cancers.( These micronutrients exhibit multiple biological actions that may protect against disease. For example, vitamin E is a chain breaking antioxidant that protect cell membranes from damage caused by lipid peroxidation and also inhibits cell proliferation and platelet adhesion.( Vitamin A is essential for differentiation of epithelial cells and maintenance of cell signaling and communication.( CoQ10 is well known for its role as an electron transporter in the mitochondrial respiratory chain. Also, CoQ10 is a powerful antioxidant even stronger than α-tocopherol.(

The aim of the present study was to investigate correlation between vitamin A, vitamin E, CoQ10 leves and homeostasis model assesment (HOMA-IR) in obese and normal weight subjects. Also, we examined associations of these parameters with body mass index (BMI), waist circumference and waist-to-hip ratio in the subjects.

Materials and Methods

Patients

This study was performed on 98 (21M, 77F) obese people aged 18–65 years and 78 (20M, 58F) control subjects aged 18–65 years. There were no complaints and symptoms of the obese subjects other than obesity. BMI was used as an obesity criteria. BMI of the obese subjects was more than 30 kg/m2 and that of healty controls was less than 25 kg/m2. Diatery patterns of the subjects were estimated using a food questionnaire. Less than 5% of the obese and healty control participants reported tobacco use. None of the subjects were taking a HMG-CoA reductase inhibitor (statin) treatment and vitamin supplementation. The study was approved by Hospital Ethics Committee, and informed consent was obtained from all participants. Blood samples were obtained from the subjects between 8 A.M. and 9 A.M. after a 12-h overnight fast on plain and EDTA tubes. Plasma and sera of the samples were separated and frozen at −80°C until the day of study and thawed only once.

Anthropometric assessment

All anthropometric measurement were made with participants wearing light clothing and no shoes. BMI was measured in all participants. BMI was calculated as weight (in kilograms) divided by height (in meters) squared, and participants waists were measured with a soft tape midway between the lowest rib and the iliac crest. The hip circumferences was measured at the widest part of the gluteal region.

Sample analyses

Serum CoQ10 levels were measured using the ImmuChrom reagent kit for high-performance liquid chromatography (HPLC) technique with UV-detector. Plasma (in EDTA-blood) vitamin A and vitamin E levels were measured using the Chromsystems (Munich, Germany) reagent kit for HPLC technique with UV-detector. Serum total cholesterol, triglycerides, high density lipoprotein (HDL) cholesterol and blood glucose was measured by commertially available kits based by routine methods on the Synchron LX System (Beckman Coulter, Fullerton CA). Low density lipoprotein (LDL) cholesterol was calculated using the formula by Friedewald et al.( Insulin levels were determined by routine chemiluminesans method on E170 analyzer (Roche Diagnostics).

Calculations

HOMA-IR was used to detect the degree of insulin resistance. The resistance can be assessed from the fasting blood glucose and insulin concentrations by the formulae:

HOMA-IR scores greater than ⩾2.5 shows low insulin sensitivity (insulin resistance).( Adjusted vitamin A and vitamin E were calculated as vitamin A/(total cholesterol + triglycerides) (µmol/mmol) and vitamin E/(total cholesterol + triglycerides) (µmol/mmol).

Statistical analysis

All data are expressed as mean ± standard deviation (SD). Statistical analyses were done using SPSS ver. 17.0 (SPSS Inc., IL). The normal distribution of variables were examined with Independent-Samples t test, and non-normally distributed variables were examined by Mann-Whitney U test. The correlations between variables were performed by Pearson’s Correlation test. Differences were considered significant at a probability level of p<0.05.

Results

Clinical and analytical characteristics of the obese and control subjects are presented in Table 1. As seen, BMI, waist circumference, waist-to-hip ratio, systolic and diastolic blood pressure, triglycerides, fasting blood glucose, fasting blood insülin and HOMA-IR values of the obese female and male subjects were significantly higher than those of the controls (p<0.001). Total-cholesterol and LDL-cholesterol of the obese female subjects were significantly higher than those of the controls (p<0.001 for total-cholesterol and p<0.01 for LDL-cholesterol) whereas HDL-cholesterol level of obese female group was lower than that of the control group (p<0.001). There were no significant differences between total-cholesterol, LDL-cholesterol and HDL-cholesterol levels of the male groups.

Table 1

Clinical and analytical characteristics of the control and the obese subjects

	Control (mean ± SD1)		Obese (mean ± SD1)
	Female (n = 58)	Male (n = 20)	Female (n = 77)	Male (n = 21)
Age (years)	36.15 ± 10.1	39.20 ± 14.4	37.65 ± 12.4	39.47 ± 14.1
BMI (kg/m2)	20.66 ± 2.4	22.10 ± 2.2	43.61 ± 6.4**	39.85 ± 6.0**
Waist circumference (cm)	71.15 ± 7.4	84.70 ± 5.3	117.85 ± 11.5**	123.85 ± 11.4**
Waist-to-hip ratio	0.74 ± 0.05	0.84 ± 0.03	0.84 ± 0.06**	0.95 ± 0.07**
Systolic blood pressure (mmHg)	110.13 ± 4.1	117.1 ± 2.8	130.13 ± 3.4**	137.13 ± 3.4**
Diastolic blood pressure (mmHg)	70.1 ± 2.5	73.4 ± 1.0	82.1 ± 3.0**	84.2 ± 2.4**
Total cholesterol (mmol/l)	4.62 ± 0.9	4.91 ± 0.9	5.14 ± 1.1**	5.0 ± 1.0
Triglycerides (mmol/l)	0.80 ± 0.4	1.23 ± 0.5	1.77 ± 1.0**	1.98 ± 0.9**
HDL-cholesterol (mmol/l)	1.39 ± 0.3	1.01 ± 0.2	1.13 ± 0.2**	0.92 ± 0.2
LDL-cholesterol (mmol/l)	2.86 ± 0.8	3.15 ± 1.0	3.19 ± 0.9*	3.34 ± 0.8
Fasting blood glucose (mg/dl)	84.11 ± 7.4	83.8 ± 10.6	99.81 ± 12.9**	100.1 ± 11.3**
Fasting blood insülin (µU/ml)	5.65 ± 2.7	6.26 ± 3.9	15.94 ± 8.4**	22.07 ± 11.7**
HOMA-IR	1.16 ± 0.6	1.31 ± 0.9	3.75 ± 3.2**	4.64 ± 2.9**

1 standart deviation. BMI, body mass index. HDL-cholesterol, high density lipoprotein-cholesterol. LDL-cholesterol, low density lipoprotein-cholesterol. HOMA-IR, homeostasis model assesment. *p<0.05, **p<0.01.

Vitamins and CoQ10 concentrations of the obese and the control subjects are presented in Table 2. As seen, adjusted vitamin A (p<0.001), adjusted vitamin E (p<0.05), CoQ10/total cholesterol (p<0.05), CoQ10/LDL (p<0.05), CoQ10/triglyceride (p<0.001) and CoQ10/total cholesterol + triglyceride (p<0.001) values of the obese female subjects were significantly lower than those of the control group. Adjusted vitamin A (p<0.05) and CoQ10/triglyceride (p<0.05) values of the obese male subjects were significantly lower than those of the control group.

Table 2

Vitamins and CoQ10 concentrations in the control and the obese subjects

	Control (mean ± SD1)		Obese (mean ± SD1)
	Female (n = 58)	Male (n = 20)	Female (n = 77)	Male (n = 21)
Vitamin A (retinol) (µmol/l)	1.84 ± 0.6	2.33 ± 0.5	2.01 ± 0.5	2.25 ± 0.5
Adjusted Vitamin A (µmol/mmol) (vitamin A/(total cholesterol + triglycerides))	0.34 ± 0.1	0.39 ± 0.1	0.29 ± 0.08**	0.32 ± 0.09*
Vitamin E (α-tokoferol) (µmol/l)	30.46 ± 8.5	32.55 ± 8.3	33.47 ± 8.8	29.75 ± 8.1
Adjusted Vitamin E (µmol/mmol) (vitamin E/(total cholesterol + triglycerides))	5.53 ± 1.1	4.86 ± 0.9	5.07 ± 1.2*	4.67 ± 1.2
CoQ10 (µmol/l)	1.58 ± 0.7	1.46 ± 0.5	1.49 ± 0.6	1.38 ± 0.4
CoQ10/total cholesterol (µmol/mmol)	0.34 ± 0.1	0.29 ± 0.1	0.29 ± 0.1*	0.27 ± 0.07
CoQ10/LDL (µmol/mmol)	0.58 ± 0.3	0.44 ± 0.1	0.48 ± 0.2*	0.47 ± 0.1
CoQ10/triglycerides (µmol/mmol)	2.50 ± 2.0	1.28 ± 0.5	1.05 ± 0.6**	0.89 ± 0.5*
CoQ10/(total cholesterol + triglycerides) (µmol/mmol)	0.30 ± 0.1	0.024 ± 0.08	0.22 ± 0.08**	0.20 ± 0.05

1 standart deviation. CoQ10, coenzyme Q10. LDL, low density lipoprotein-cholesterol. *p<0.05, **p<0.01.

Simple correlation analysis were performed to investigate the association of plasma vitamins (adjusted and unadjusted) and CoQ10 measures with BMI, waist circumference, waist-to-hip ratio, age and HOMA-IR. As shown in Table 3, vitamin A (r = −0.222, p<0.05) and adjusted vitamin A (r = −0.269, p<0.01) levels were negatively correlated with BMI and adjusted vitamin A was negatively correlated with waist circumference (r = −0.202, p<0.05) in obese group. Vitamin A was positively correlated with age (r = 0.258, p<0.05) in obese group.

Table 3

Correlations of plasma vitamin A and adjusted vitamin A levels with clinical and biochemical variables of the control and the obese subjects

	Control (n = 78)		Obese (n = 98)
	Vitamin A	Adjusted vitamin A	Vitamin A	Adjusted vitamin A
BMI	0.132	–0.085	–0.222*	–0.269**
Waist circumference	0.336	0.060	–0.161	–0.202*
Waist-to-hip ratio	0.320	0.082	–0.019	–0.059
HOMA-IR	0.148	0.135	0.073	0.092

BMI, body mass index. HOMA-IR, homeostasis model assesment. *p<0.05, **p<0.01.

Adjusted vitamin E was negatively correlated with BMI (r = −0.237, p<0.05), waist circumference (r = −0.380, p<0.01) and waist-to-hip ratio (r = −0.271, p<0.05) in control group (Table 4). On the other hand, adjusted vitamin E was negatively correlated with waist circumference (r = −0.233, p<0.05), waist-to-hip ratio (r = −0.273, p<0.01) in obese group. Furthermore, vitamin E (r = 0.246, p<0.05) was positively correlated with age in obese group.

Table 4

Correlations of plasma vitamin E and adjusted vitamin E levels with clinical and biochemical variables of the control and the obese subjects

	Control (n = 78)		Obese (n = 98)
	Vitamin E	Adjusted vitamin E	Vitamin E	Adjusted vitamin E
BMI	0.068	–0.237*	0.095	–0.044
Waist circumference	0.060	–0.380**	0.047	–0.233*
Waist-to-hip ratio	0.080	–0.271*	–0.121	–0.273**
HOMA-IR	–0.082	–0.130	0.134	–0.003

BMI, body mass index. HOMA-IR, homeostasis model assesment. *p<0.05, **p<0.01.

CoQ10 was positively correlated with age (r = 0.240, p<0.05) in obese group. CoQ10/total cholesterol ratio was negatively correlated with age (r = −0.228, p<0.05) in control group (Table 5). CoQ10/LDL ratio was negatively correlated with waist circumference (r = −0.337, p<0.01), waist-to-hip ratio (r = –0.285, p<0.05) and age (r = −0.295, p<0.01) in the control group. CoQ10/triglyceride ratio was negatively correlated with waist circumference (r = −0.387, p<0.01), waist-to-hip ratio (r = −0.285, p<0.05) in the control group. CoQ10/total cholesterol + triglyceride ratio was negatively correlated with waist circumference (r = −0.281, p<0.01), waist-to-hip ratio (r = −0.250, p<0.05) and age (r = −0.233, p<0.05) in the control group.

Table 5

Correlations of serum CoQ10 levels with clinical and biochemical variables of the control and the obese subjects

	Control					Obese
	BMI	Waist circumference	Waist-to-hip ratio	Age	HOMA-IR	BMI	Waist circumference	Waist-to-hip ratio	Age	HOMA-IR
CoQ10	0.110	–0.065	–0.050	–0.151	0.065	0.140	0.085	0.101	0.240*	–0.016
CoQ10/total cholesterol	–0.026	–0.217	–0.185	–0.228*	0.054	0.142	0.067	0.095	0.126	–0.063
CoQ10/LDL	–0.147	–0.337**	–0.285*	–0.295**	0.057	0.168	0.083	0.084	0.111	–0.050
CoQ10/triglyceride	–0.155	–0.387**	–0.285*	–0.211	–0.060	–0.097	–0.089	–0.022	0.111	–0.055
CoQ10/total cholesterol + triglyceride	–0.059	–0.281*	–0.250*	–0.233*	0.017	0.064	0.003	0.049	0.108	–0.047

BMI, body mass index. LDL-cholesterol, low density lipoprotein-cholesterol. HOMA-IR, homeostasis model assesment. CoQ10, coenzyme Q10. *p<0.05, **p<0.01.

CoQ10 was positively correlated with vitamin E in obese group (r = 0.272, p<0.01) (data not shown). On the other hand, there were no correlations between HOMA-IR values and adjusted and unadjusted vitamin E, vitamin A and CoQ10 levels in the obese and the control subjects.

Discussion

In the present study, adjusted serum vitamin E concentration was significantly higher in the obese female group than in the control group. However, there was no significant difference between unadjusted vitamin E levels of the groups. Also, there was no correlation between unadjusted vitamin E levels and BMI while adjusted vitamin E level was negatively correlated with BMI in the control group and waist circumference and waist-to-hip ratio in the obese and the control group. Since vitamin E is transported in lipoprotein particules we believe that it is more accurate to use the lipid-adjusted vitamine E levels to evaluate vitamine E status of the obese subjects. Indeed, it has been reported that the strongest and most consistent predictor of all serum fat-soluble nutrients were serum cholesterol.(

In the literature, there are conflicting findings about the correlation of vitamin E levels with obesity. For example, adjusted( and unadjusted( plasma vitamin E level was found to be negatively associated with BMI. In contrast, adjusted vitamin E level was not associated in one study( and positively associated with BMI in another study.(

We believe that the reason for the differences between these findings is the usage of the adjusted and unadjusted levels of the vitamin. Another reason might be the differences between the degree of obesity, age and dietary habituate of the investigated subjects.

In our study there was no significant difference between unadjusted plasma vitamin A levels of the obese and control subjects. However, adjusted vitamin A level of the obese female and male subjects was significantly lower than that of the control group. Also, adjusted vitamin A level was negatively correlated with BMI and waist circumference in the obese group. Since retinol is considered an antioxidant and anti-inflammatory agent( our results suggest that obesity is associated with a lesser capacity of anti-inflammatory and antioxidant potential. Although there are some conflicting findings concerning plasma vitamin A levels of obese subjects, generally the findings of the other investigors are in consisting with our findings. For example, Viroonudomphol et al.( have found a negative correlation between serum unadjusted retinol and BMI, waist-hip circumferences in both overweight and obese subjets. Botella-Carretero et al.( have found a negative correlation of serum unadjusted retinol with BMI in morbid obese people. Neuhouser et al.( have found that obesity was associated with low serum unadjusted retinol concentration. Switzer et al.( have reported that plasma adjusted retinol concentrations was not associated with BMI in older African American women.

The present study demonstrated that there was no difference between unadjusted plasma CoQ10 levels of the obese female and male subjects and the control subjects. However, lipid adjusted CoQ10 levels were significantly lower in the obese female and male subjects than in the controls. Since, CoQ10 is a powerful antioxidant, our lipid adjusted CoQ10 levels show that there is an antioxidant defficiency in obese subjects.

Mancini et al.( have found no difference between unadjusted plasma CoQ10 levels of obese and control subjects. Also, Menke et al.( have found that unadjusted plasma CoQ10 levels of obese and normal weight children were not different. Butler et al.( have reported that unadjusted plasma CoQ10 level was significantly lower in obese subjects than in control subjects. Since, CoQ is transported in lipoproteins, it is clear that lipid-corrected levels are more informative. Therefore, we believe that our results provide more accurate information about CoQ status of the obese subjects.

On the other hand, CoQ10 originates from endogenous synthesis as well as from food intake.( Therefore, the ethnic difference in lifestyle and nutritional patterns may result in differences in plasma CoQ10 levels found in different studies.(

In this study, both unadjusted plasma CoQ10 and unadjusted vitamin E levels positively correlated to cholesterol, triglycerides and HDL concentrations in the obese subjects. Also, unadjusted plasma CoQ10 levels were positively correlated to unadjusted vitamin E in obese people (data not shown). This may indicate that vitamine E and endogenous CoQ10 synthesis may compensate the greater demand of lipophilic antioxidants in the plasma in situations like obesity where plasma lipids are increased.(

Both vitamin E and CoQ10 possess distinct lipoprotective antioxidant properties in biological membranes. Their combined antioxidant activity, however, is markedly synergistic when both are present together.( These data provide direct evidence for an interactive effect between vitamin E and CoQ10 in obese people.

In our study, we have found no correlation between HOMA-IR, CoQ10, vitamin A and E values in both the obese and the control subjects. Botella-Carretero et al.( have found no correlation between HOMA-IR and unadjusted plasma vitamin A and E levels in obese subjects, a finding which supports our findings. We have found no study investigating an asociation between (adjusted/unadjusted) CoQ10 levels and HOMA-IR values in obese subjects. Therefore, our study is the first one investigating such a correlation.

As seen from our data (Table 2), the findings of the obese female and the control female are different from those of the findings of the obese male and the control male. Although the underlying mechanism of these differences are not known we believe that this may be due to metabolic differences between male and female subjects and the lower number of male subjects in our study.

In conclusion, our findings show that lipid-adjusted vitamin A, E and CoQ10 levels of the obese subjects were significantly lower than those of the controls. Also, we have found no association between HOMA-IR and vitamin A, E and CoQ10 levels in the obese subjects.