Total Antioxidant Performance Is Associated with Diet and Serum Antioxidants in Participants of the Diet and Physical Activity Substudy of the Jackson Heart Study^,

Abstract

Total antioxidant performance (TAP) measures antioxidant capacities in both hydrophilic and lipophilic compartments of serum and interactions known to exist between them. Our objective was to assess TAP levels in a subset of Jackson Heart Study (JHS) participants and to examine associations with dietary and total (diet + supplement) intakes of α-tocopherol, γ-tocopherol (diet only), β-carotene, vitamin C, fruit, vegetables, and nuts, and serum concentrations of α-tocopherol, γ-tocopherol, and β-carotene. We conducted a cross-sectional analysis of 420 (mean age 61 y; 254 women) African American men and women participating in the Diet and Physical Activity Sub-Study of the JHS in Jackson, Mississippi. In multivariate-adjusted models, we observed positive associations between total α-tocopherol, total and dietary β-carotene, and total vitamin C intakes and TAP levels (P-trend < 0.05). Positive associations were also observed for vegetable, fruit, and total fruit and vegetable intakes (P-trend < 0.05). For serum antioxidant nutrients, α-tocopherol but not β-carotene was associated with serum TAP levels. There were inverse associations for serum γ-tocopherol and TAP levels. Associations for α-tocopherol were seen at intake levels much higher than the current Recommended Dietary Allowance. It may, therefore, be prudent to focus on increasing consumption of fruit, vegetables, nuts, and seeds to increase total antioxidant capacity.

Introduction

Oxidative stress is an imbalance between the production of reactive oxygen radicals and the ability of the organism's natural protective mechanisms to cope with these radicals and to prevent adverse effects (1). Cells in the body are exposed to reactive oxygen species under normal circumstances via the leakage of electrons from the electron transport chain, phagocytic cells, and endogenous enzyme systems (2). The oxidation of lipids, nucleic acids, or protein by these reactive oxygen species is thought to be associated with the etiology of several age-related chronic diseases, including cancer (3), cardiovascular disease (4), cataract (5), and age-related macular degeneration (6). These chronic diseases account for a high percentage of morbidity and mortality and preventing them is a major public health priority.

Arrays of defense systems protect the body from the deleterious effects of oxidative stress. These include antioxidant enzymes and radical-scavenging antioxidants (7). Antioxidant nutrients such as vitamin E, carotenoids, and fruits and vegetables rich in such antioxidants have been associated with a lower risk of diseases caused by oxidative stress (8–10). Whereas individual actions of antioxidants have been reported, a large number of studies have indicated that cooperative/synergistic interactions exist among antioxidants in plasma (11). Therefore, studying the overall antioxidant status may be more biologically relevant than studying a single antioxidant (12). A relatively recent method called total antioxidant performance (TAP),⁸ developed by Aldini et al. (13) and validated by Beretta et al. (14), measures not only antioxidant capacity in both the hydrophilic and lipophilic compartments of the biological system, but also their synergistic/ cooperative interactions. Given the potential for nutrients and foods to contribute to the prevention of oxidative stress-associated chronic diseases and the unique ability of TAP to measure total antioxidant status, the objectives of our study were to: 1) measure serum antioxidant capacity using the TAP assay in a subset of Jackson Heart Study (JHS) participants; 2) examine associations between dietary and total intakes of α-tocopherol, γ-tocopherol (diet only), β-carotene, vitamin C, vegetable, fruit and nut intake, and TAP levels; and 3) examine associations between serum α- and γ-tocopherol, β-carotene, and TAP levels.

Participants and Methods

Study population.

The men and women in this cross-sectional analysis were participants of the Diet and Physical Activity Sub-Study (DPASS) of the JHS. The JHS is a single-site prospective epidemiological investigation of cardiovascular disease among African Americans from the Jackson, Mississippi metropolitan area. A detailed description of the original study has been published elsewhere (15).

Study sample selection.

A subset of participants (n = 499) from the JHS cohort (n = 5301) was selected for the JHS DPASS. As participants were enrolled in the JHS, investigators recruited participants for DPASS to include an equal number of men and women from younger (34–64 y) and older (≥65 y) age groups, from lower and higher socioeconomic status, and from lower and higher physical activity groups. All eligible participants were invited to be part of DPASS until each of the enrollment strata were filled. The aim of DPASS was to provide data for validation of the diet and physical activity instruments used for the entire cohort of the JHS.

Dietary assessment.

The Lower Mississippi Delta Nutrition Intervention Research Initiative conducted a telephone survey in the Delta region to collect representative dietary data using 24-h dietary recalls. These data were used to develop a new FFQ designed for use in the LMD region. Details regarding development of this regional FFQ are available elsewhere (16). This FFQ, called the Delta NIRI FFQ (283 items), and its shortened version, the Delta NIRI JHS FFQ (158 items), which was specifically developed for use in the JHS, were then used as dietary assessment tools in the DPASS. We have previously validated both FFQ against the mean of 4 24-h recalls (17), serum tocopherols (18), and carotenoids (19).

Laboratory analyses.

Participants provided blood samples on the day of the baseline clinic interview, which took place on the day of administration of the short FFQ. Blood samples from fasting (12 h) participants were collected in vacutainer tubes and centrifuged at 3000 × g for 10 min at 4°C. Samples were frozen at −70°C until analyzed. Serum TAP was determined by the method first developed by Aldini et al. (13) to measure total antioxidant capacity in both the hydrophilic and lipophilic compartments of serum and validated by Beretta et al. (14) for the application to high throughput studies. This method measures the rate of oxidation of 4, 4-difluoro-5-(4-phenyl-1, 3-butadienyl)-4-bora-3a,4a-diaza-s-indacene-3-undecanoic acid (BODIPY 581/591), a lipid-soluble fluorescent probe, and uses the lipid-soluble radical initiator 2,2′-azobis(4-methoxy-2,4-dimethylvaleronitrile). Oxidation is determined by monitoring the appearance of green fluorescence of the oxidation product of BODIPY (λ_ex = 500 nm, λ_em = 520 nm) using a 1420 multilabel counter (Wallac Victor 2, Perkin Elmer Life Sciences). The results are expressed as TAP values, which represent the percentage of inhibition of BODIPY oxidation in human serum with respect to that occurring in a control sample consisting of BODIPY 581/591 in phosphatidylcholine liposomes. For the measurement of serum carotenoids and tocopherols, analyses were performed using HPLC, as described by Yeum et al. (20,21). After standard lipid extraction with chloroform:methanol (2:1) followed by hexane, samples were analyzed for carotenoids and tocopherols using a reverse phase HPLC system consisting of a 600S controller (Millipore), Waters 616 pump, Waters 717 autosampler, Waters 996 photodiode array detector, and C30 carotenoid column (3 μm, 150 × 4.6 mm, YMC). Millenium32 was the operating system. The programmable photodiode array detector was set at 445 and 455 nm for carotenoids and 292 nm for tocopherols. Carotenoids and tocopherols were quantified by determining peak areas in the HPLC chromatograms, calibrated against known standards. Serum cholesterol and uric acid concentrations for the cohort were determined according to methods described previously (22).

Other covariates.

Information on covariates was obtained at either the initial home visit or the JHS baseline clinic visit. Age was computed from self-reported date of birth. Information regarding presence of self-reported hypertension was assessed from the medical history questionnaire. Smoking status was derived from a set of questions regarding cigarette use. Participant height and weight were measured by trained technicians at the clinic visit using physician quality measurement scales in an exam gown with no shoes. Detailed information regarding the procedures used for the anthropometric procedures conducted at the clinic visit have been published elsewhere (15). BMI was calculated as weight/height² (kg/m²).

Statistical analyses.

Several dietary assessment instruments were used in the DPASS of the JHS. However, we used data from the Delta NIRI JHS FFQ (158 item) for this analysis, as this questionnaire was administered on the day of the blood draw and is the only dietary questionnaire that was used in the main JHS cohort. We excluded participants who reported energy intake outside the plausible range of ≤600 or ≥4000 kcal/d⁹ on the FFQ (n = 34) or whose FFQ had 5 or more food items blank (n = 2). DPASS participants without blood samples for the TAP analysis (n = 43) were also excluded, leaving a sample of 420 individuals. For analyses with serum nutrients, due to missing data, n = 416 for β-carotene and 417 for α-tocopherol. Two participants had γ-tocopherol concentrations below detectable levels; therefore, the number of observations for those analyses was 418.

We examined associations between TAP levels and sample characteristics. Variables included dietary intakes of antioxidants, fruit and vegetables, and nuts and serum measures of dietary antioxidants, uric acid, and cholesterol. Servings of vegetables per day were estimated from questions on the FFQ. These included vegetable groups such as orange vegetables, sweet potato, tomato and tomato products, green leafy vegetables, root vegetables, and other vegetables but did not include white potato and white potato products. Also, vegetables that were part of mixed dishes, e.g. mixed dishes with meat were not included. Servings of fruit were also estimated from questions on the FFQ. These included all citrus and noncitrus fruit and fruit juice, but not fruit drinks.

Associations of TAP with nuts, including peanuts, peanut butter, and pecans, were also examined. As mentioned previously, the FFQ was developed using 24-h recall data from the Delta region. The serving sizes are therefore reflective of this region.

As dietary and biological measures of antioxidant nutrients and foods were skewed, they were log transformed prior to analysis. Dietary intake data were energy adjusted using the residual method (23). Log-transformed dietary measurements were regressed on log-transformed total energy intake to compute residuals. The predicted value of the dietary measurement for the mean of total energy intake was added back to each residual, and the antilogarithm of this value was taken. We tested for interactions between the main dietary and serum predictors (in continuous form) and sex with respect to TAP levels. As none were significant, we present the results for men and women together. We estimated least squares means for TAP levels by quartile of dietary as well as total (diet + supplement) intake of α-tocopherol (mg/d), γ-tocopherol (diet only, mg/d), β-carotene (μg/d), vitamin C (mg/d), fruit, vegetables, and nuts (all servings per day). Because TAP is a new measure of serum antioxidant status, we examined several variables that were available for the cohort as potential covariates and confounders of the association with antioxidants, fruit and vegetables, and nuts and serum measures of dietary antioxidants and adjusted for these in the models. After analysis with a simple model, a second model was further adjusted for age (y), sex, BMI (kg/m²), energy intake (kJ/d), supplement use (yes/no), current smoking status (yes/no), and self-reported hypertension (yes/no). Serum uric acid, which is mainly produced endogenously but is also affected by dietary intake, was strongly associated with TAP levels. Serum uric acid was also a significant negative confounder between total α- tocopherol intake and TAP. Therefore, for the analyses with α-tocopherol, we examined the association between a residual measure of TAP after adjustment for serum uric acid and α-tocopherol intake.

For serum antioxidant analysis, we examined the associations between TAP and serum β-carotene and α and γ-tocopherol levels. Two models were examined, the first adjusting for age (y), sex, BMI (kg/m²), serum cholesterol concentrations (mmol/L), and serum uric acid concentrations (μmol/L) and the second model further adjusted for current smoking status (yes/no) and presence of self-reported hypertension (yes/no).

We also present the P-values for test for trend, which were calculated by assigning subjects the median value of the category of nutrient or food group or serum antioxidant nutrient being considered and including this as a continuous variable in the model. All statistical tests were 2-sided with a significance level of 0.05. SAS version 9.1 (SAS Institute) was used for all statistical analyses.

Results

Age was positively associated with serum TAP levels (P-trend < 0.01) (Table 1). A higher percentage of women were in the lower quartile of TAP (75 compared with 45% in the highest quartile; P < 0.0001). BMI was positively associated with TAP (P-trend < 0.05). Although there was a difference in supplement use across the TAP quartiles (with an increasing trend up to quartile 3), the percentage of supplement users in the highest quartile was 46% compared with 50% in the lowest quartile. Smoking status did not differ across TAP quartiles. A higher percentage of participants in the highest quartile of TAP reported hypertension (66 compared with 48% in the lowest quartile; P < 0.001). Serum β-carotene, γ-tocopherol, and cholesterol concentration were not associated with TAP. Serum uric acid and α-tocopherol were associated across TAP levels (P-trend ≤ 0.0001). There were no associations between intakes of energy, total or dietary α- or γ- tocopherol, and serum TAP levels. Total and dietary β-carotene intake and vegetable and fruit intake were positively associated with serum TAP levels. There were weak associations between dietary and total vitamin C intake and serum TAP level.

TABLE 1.

Sample characteristics of the JHS diet and physical activity substudy participants by quartiles of TAP¹

	TAP2(% protection)
Variables	Q13	Q2	Q3	Q4	P4
Participants' characteristics
Age, y	59.2 ± 0.92	60.0 ± 0.92	62.4 ± 0.92	62.1 ± 0.92	0.007
Women, %	75.2	60.0	62.0	44.8	0.0001
BMI,5kg/m2	29.6 ± 0.66	29.9 ± 0.64	31.7 ± 0.64	31.4 ± 0.64	0.013
Current smoker, %	5.7	10.5	16.2	8.6	0.282
Supplement user, %	49.5	57.1	69.5	45.7	0.003
Self reported hypertension, %	47.6	64.8	78.1	66.4	0.001
Serum β-carotene,5μmol/L	0.64 ± 0.06	0.73 ± 0.06	0.67 ± 0.06	0.61 ± 0.06	0.421
Serum α-tocopherol,5μmol/L	27.0 ± 1.38	31.0 ± 1.34	34.4 ± 1.33	32.6 ± 1.33	0.001
Serum γ-tocopherol,5μmol/L	5.78 ± 0.41	5.39 ± 0.40	5.25 ± 0.40	6.39 ± 0.40	0.585
Serum uric acid,5μmol/L	275 ± 7.38	324 ± 7.14	361 ± 7.15	430 ± 7.11	0.0001
Serum cholesterol,5mmol/L	5.18 ± 0.10	4.97 ± 0.10	5.06 ± 0.10	5.29 ± 0.10	0.986
Food/nutrient intakes
Energy,6kJ/d	8198 ± 306	8440 ± 296	8045 ± 297	8168 ± 295	0.663
Total α-tocopherol,67mg/d	55.6 ± 11.8	89.4 ± 11.4	96.0 ± 11.4	64.4 ± 11.4	0.249
Dietary α-tocopherol,6mg/d	6.98 ± 0.21	6.67 ± 0.20	6.59 ± 0.20	7.27 ± 0.20	0.568
Dietary γ-tocopherol,6mg/d	13.9 ± 0.45	13.3 ± 0.44	13.3 ± 0.43	14.4 ± 0.43	0.562
Total β-carotene,67μg/d	2928 ± 113	3115 ± 109	3302 ± 109	3317 ± 109	0.005
Dietary β-carotene,6μg/d	2666 ± 106	2811 ± 103	2937 ± 103	3085 ± 102	0.002
Total vitamin C,67mg/d	184 ± 19.1	178 ± 18.5	220 ± 18.5	208 ± 18.4	0.06
Dietary vitamin C,6mg/d	112 ± 6.57	109 ± 6.36	120 ± 6.37	124 ± 6.33	0.07
Vegetables,68servings/d	1.15 ± 0.04	1.13 ± 0.04	1.12 ± 0.04	1.30 ± 0.04	0.009
Fruit,69servings/d	1.39 ± 0.09	1.40 ± 0.09	1.55 ± 0.09	1.61 ± 0.09	0.021
Vegetables + fruit,689servings/d	2.54 ± 0.10	2.53 ± 0.10	2.67 ± 0.10	2.91 ± 0.10	0.002
Nuts,610servings/d	0.29 ± 0.04	0.24 ± 0.04	0.23 ± 0.04	0.27 ± 0.04	0.332

¹Values are means ± SE or %, n = 420, 416 (serum β-carotene), 417 (serum α-tocopherol and uric acid), 418 (serum γ-tocopherol) or 419 (hypertension).

²Median total antioxidant performance values for Q1 through Q4 were: 64.0, 71.7, 76.5, and 81.2% protection, respectively.

³Quartile.

⁴Continuous variables were examined using linear regression (SAS Proc GLM) with test for trend across median values of total antioxidant performance in each quartile. Categorical variables were examined with chi-square analysis.

⁵Adjusted for age and sex and, for serum antioxidant concentrations, also for serum cholesterol. Values for serum antioxidant, cholesterol, and uric acid concentrations are presented in original scale.

⁶Adjusted for sex, age, and energy intake (except for energy intake itself). Dietary, total nutrient, and food intakes are presented in original scale.

⁷Total = dietary + supplement intake.

⁸Vegetable intakes were calculated from specific questions about vegetable consumption on the FFQ. We were unable to include the vegetables part of mixed recipes in this calculation.

⁹Fruit intakes were calculated from specific questions asking about fruit and fruit juice consumption on the FFQ.

¹⁰Nut intakes were calculated from specific questions asking about nuts and nut butters on the FFQ.

Unlike dietary α - and γ-tocopherols and vitamin C intakes, the highest quartile of dietary β-carotene intake with a median intake of 4.0 mg/d had ∼4% higher serum TAP compared with the lowest quartile of intake, with a median intake of 1.75 mg/d; tests for trend were also significant (Table 2). The highest quartile of total β-carotene intake, with a median intake of 4.3 mg/d, was also associated with 3.2% higher serum TAP levels compared with the lowest quartile with a median intake 1.9 mg/d and tests for trend were significant. The highest quartile of total vitamin C intake was also associated with an ∼2.5% increase in TAP levels compared with the lowest.

TABLE 2.

Adjusted mean total antioxidant performance of JHS DPASS participants by antioxidant nutrient intakes^1–3

	Model
Antioxidant nutrient4	Basic	Adjusted5
Total α-tocopherol intake, mg/d
4.6	72.2 ± 0.53a	72.0 ± 0.73a
7.4	72.0 ± 0.53a	72.0 ± 0.69a
23.7	73.7 ± 0.53ab	73.4 ± 0.74a,b
285	75.2 ± 0.54b	74.7 ± 0.76b
P-trend6	<0.0001	0.01
R2	0.06	0.09
Dietary α-tocopherol intake, mg/d
4.7	72.8 ± 0.55	72.4 ± 0.65
5.7	73.4 ± 0.55	73.1 ± 0.65
6.6	73.5 ± 0.55	73.2 ± 0.66
8.1	73.1 ± 0.55	72.7 ± 0.69
P-trend6	NS7	NS
R2	0.002	0.07
Dietary γ-tocopherol intake,2mg/d
7.8	72.7 ± 0.74	72.3 ± 0.85
10.8	74.1 ± 0.74	74.1 ± 0.86
13.6	72.1 ± 0.74	72.6 ± 0.87
17.1	73.8 ± 0.74	74.6 ± 0.88
P-trend6	NS	NS
R	0.01	0.14
Total β-carotene intake, μg/d
1920	72.0 ± 0.74a	72.0 ± 0.89a
2767	73.2 ± 0.71a,b	73.4 ± 0.85a
3389	74.0 ± 0.71a,b	73.8 ± 0.87a
4319	74.5 ± 0.71b	74.3 ± 0.87a
P-trend6	0.01	0.03
R2	0.02	0.13
Dietary β-carotene intake, μg/d
1750	71.3 ± 0.74a	71.8 ± 0.86a
2351	73.5 ± 0.74a,b	73.7 ± 0.86a
3063	73.8 ± 0.74a,b	73.8 ± 0.87a
4010	74.3 ± 0.74b	74.1 ± 0.86a
P-trend6	0.01	0.04
R2	0.02	0.14
Total vitamin C intake, mg/d
69.0	73.1 ± 0.74a,b	73.3 ± 0.83a,b
108	72.1 ± 0.74a	72.2 ± 0.90a
155	72.8 ± 0.74a,b	72.6 ± 0.89a,b
498	74.9 ± 0.74b	75.0 ± 0.90b
P-trend6	0.01	0.01
R2	0.02	0.14
Dietary vitamin C intake, mg/d
58.4	72.7 ± 0.74	73.0 ± 0.83
86.0	73.3 ± 0.74	73.7 ± 0.87
123	73.2 ± 0.74	73.1 ± 0.87
173	73.7 ± 0.74	74.0 ± 0.91
P-trend6	NS	NS
R2	0.002	0.13

¹Values are means ± SE. Within category and column, means with superscripts without a common letter differ after Tukey's adjustment for multiple comparisons, P < 0.05.

²TAP values for α-tocopherol (total and dietary) are residuals, adjusted for serum uric acid.

³n = 420 (basic model) and n = 418 (adjusted model).

⁴Median value of energy adjusted nutrient for quartile.

⁵Adjusted for age, sex, BMI, energy intake, supplement use, current smoker, and self-reported hypertension.

⁶Test for trend was calculated across median values of intake in each quartile.

⁷Not significant.

Serum β-carotene and TAP levels were not associated (Table 3). Individuals in the highest quartile of serum α-tocopherol concentrations had significantly higher serum TAP compared with those in the lowest quartile (∼5.6% higher). Tests for trend were also significant. γ-Tocopherol concentrations and TAP levels were inversely associated. No associations were seen for a further adjusted model.

TABLE 3.

Adjusted mean total antioxidant performance of JHS DPASS participants by serum antioxidant concentrations¹²

	Model
Serum antioxidant3	Adjusted 14	Adjusted 25
Serum α-tocopherol, μmol/L
19.4	71.4 ± 0.55a	71.6 ± 0.61a
25.3	72.5 ± 0.53a,b	72.2 ± 0.60a
31.7	73.7 ± 0.54b	73.3 ± 0.60a,b
45.4	75.7 ± 0.55b	75.1 ± 0.66b
P-trend6	<0.0001	<0.0001
R2	0.51	0.51
Serum γ-tocopherol, μmol/L
1.88	75.1 ± 0.55a	74.2 ± 0.66
3.82	73.0 ± 0.54a	72.5 ± 0.60
6.21	72.0 ± 0.54a	72.0 ± 0.60
9.75	73.0 ± 0.57a	73.1 ± 0.63
P-trend6	0.02	NS7
R2	0.49	0.50
Serum β-carotene, μmol/L
0.215	72.4 ± 0.57a	72.4 ± 0.60
0.376	73.5 ± 0.55a	73.2 ± 0.60
0.638	72.8 ± 0.55a	72.4 ± 0.60
1.17	74.5 ± 0.58a	73.8 ± 0.66
P-trend6	0.02	NS
R2	0.48	0.49

¹Values are mean ± SE. Within category and column, means with superscripts without a common letter differ after Tukey's adjustment for multiple comparisons, P < 0.05.

²α-Tocopherol n = 416 (Model 1) and 415 (2); γ-tocopherol n = 417 (Model 1) and 416 (2); β-carotene n = 415 (Model 1) and 414 (2).

³Median value of serum antioxidant concentrations are in original scale.

⁴Adjusted for age, sex, BMI, serum cholesterol, and uric acid.

⁵Adjusted for above plus current smoker and self reported hypertension.

⁶Test for trend was calculated across the median value of serum antioxidant concentration in each quartile.

⁷Not significant.

The highest quartile of vegetable intake, with a median intake of 1.7 servings/d, was associated with ∼4% higher TAP values compared with the lowest quartile, with a median intake of 0.8 servings/d (Table 4). Although serum TAP did not differ between quartiles of fruit intake, tests for trend were significant (P < 0.05). The strongest associations were for combined fruit and vegetable intake (for both models), with significant associations between the lowest quartile of intake and the subsequent quartiles, as well as for trend. Nuts, which are a good source of vitamin E, were also examined. However, only small amounts of nuts were consumed in this population and there were no associations with TAP levels.

TABLE 4.

Adjusted mean total antioxidant performance of JHS DPASS participants by fruit and vegetable intakes^1–3

	Model
Food group3	Basic	Adjusted4
Vegetable intake, servings/d
0.82	72.0 ± 0.74a	72.5 ± 0.84a
0.96	72.6 ± 0.74a	72.6 ± 0.86a,b
1.20	73.2 ± 0.74a,b	73.4 ± 0.88a,b
1.66	75.1 ± 0.74b	75.2 ± 0.87b
P-trend6	0.001	0.003
R2	0.02	0.14
Fruit intake, servings/d
0.75	72.1 ± 0.74a	72.3 ± 0.85a
1.11	73.0 ± 0.74a	73.4 ± 0.85a
1.57	73.7 ± 0.74a	73.6 ± 0.89a
2.22	74.1 ± 0.74a	74.6 ± 0.89a
P-trend6	0.03	0.04
R2	0.01	0.13
Vegetable + fruit intake, servings/d
1.72	71.3 ± 0.73a	71.6 ± 0.83a
2.28	73.1 ± 0.73a,b	73.5 ± 0.85a,b
2.81	73.6 ± 0.73a,b	73.5 ± 0.89a,b
3.63	75.0 ± 0.73b	75.2 ± 0.87b
P-trend6	0.0007	0.0006
R2	0.03	0.15
Nut intake, servings/d
0.006	73.1 ± 0.74	72.9 ± 0.88
0.05	73.1 ± 0.74	73.1 ± 0.89
0.18	73.3 ± 0.74	73.4 ± 0.85
0.51	73.4 ± 0.74	73.8 ± 0.86
P-trend6	NS	NS
R2	0.0003	0.12

¹Values are mean ± SE. Within category and column, means with superscripts without a common letter differ after Tukey's adjustment for multiple comparisons, P < 0.05.

²n = 420 (basic model) and n = 418 (multivariate model).

³Median value of energy adjusted food for quartile.

⁴Adjusted for age, sex, BMI, energy intake, supplement use, current smoker, and self-reported hypertension.

⁵Test for trend was calculated across median values of intake in each quartile.

Discussion

A variety of assays have been developed to measure antioxidant capacity in plasma. These include the oxygen radical absorbance capacity assay (ORAC) (24), the Randox Trolox-equivalent antioxidant capacity assay (25) and the ferric reducing ability of plasma assay (26). However, most of these assays use either a hydrophilic or a lipophilic approach, thereby not capturing valuable information on the interactions that exist between antioxidants of the 2 compartments (27).

The TAP assay measures the activity of both the hydrophilic and lipophilic antioxidants in serum/plasma along with their interactions (14). The results of this study show that, in African Americans, total α-tocopherol, total vitamin C, total and dietary β-carotene, fruit, vegetable, and fruit plus vegetable intakes are associated with serum TAP levels. Serum α-tocopherol was also associated with this measure of overall antioxidant status.

α-Tocopherol in serum is a potent lipid-soluble antioxidant (28). In this study, we did not find an association between increasing quartiles of dietary α-tocopherol intake and TAP levels, but we did see significantly greater TAP levels in the highest vs. lowest quartile of total α-tocopherol intake (including supplements) and a linear trend across quartiles. This association remained after adjustment for several covariates. The median intake of total α-tocopherol intake in the highest quartile was 285 mg/d, whereas the mean daily dietary intake of α-tocopherol in this population was ∼7 mg/d, which is considerably lower than the Estimated Average Requirement of 12 mg/d for adults (29). The highest quartile of total α-tocopherol intake in this population, therefore, represents supplement use. In accordance with the correlation between plasma TAP and α-tocopherol in a human plasma model (13) and in plasma samples from a small number of participants previously reported by Beretta et al. (14), there was an association between α-tocopherol and TAP values but only for those taking supplements.

TAP levels did not differ across γ- tocopherol intake quartiles. The inverse association between serum γ- tocopherol and TAP levels for 1 of the statistical models examined was surprising. A study that examined the effect of smoking and environmental tobacco exposure on plasma antioxidant status in smokers, passive smokers, and nonsmokers found that whereas several other plasma antioxidant concentrations were significantly lower in smokers and passive smokers compared with nonsmokers, the levels of γ-tocopherol were elevated even after adjusting for dietary intakes (30). These results warrant further investigation into the role of γ- tocopherol in oxidative stress.

Both dietary and total β-carotene and total vitamin C intakes were associated with serum TAP. Intervention studies have shown that β-carotene and vitamin C supplements alone or in combination with other known antioxidants have been shown to improve antioxidant capacity (31) or lower markers of oxidative stress (32). Although, we did see significantly higher TAP values with higher β-carotene intake and vitamin C (total intake only) intakes, results were not reflected in serum β-carotene and TAP associations. Because serum vitamin C has not been estimated in the cohort, we were not able to examine its associations with serum TAP levels. Considering that TAP was significantly (positively) associated with intake of fruit and vegetables and that this remained after adjustment for dietary intakes of β-carotene and vitamin C (data not shown), the association of these food groups with TAP is not solely because of these well-documented antioxidant vitamins. Intervention studies have demonstrated increases in overall antioxidant status with increases in fruit and vegetable intakes (33–35). On the other hand, single or combinations of antioxidant supplement interventions have not shown improvements of antioxidant status or oxidative stress measures (36–38). It is interesting to note that observational studies have also suggested that higher intake of foods that are rich sources of these phytochemicals are associated with lower risk of chronic disease morbidity and mortality (39–42).

We are not aware of any other study examining associations between this particular measure of antioxidant status (serum TAP levels) and dietary intake of antioxidants and antioxidant-rich foods. The results from our study are consistent with others that have examined the associations between diet, dietary estimates of total antioxidant capacity, and other biological estimates of total antioxidant capacity. In a recent study, Rautiainen et al. (43) concluded that fruit and vegetables were major contributors to FFQ-based total antioxidant capacity measures such as ORAC, total radical-trapping antioxidant parameters, and ferric reducing ability of plasma assay. These dietary estimates were also positively correlated with plasma measures of ORAC and total radical-trapping antioxidant parameters. Results from the ATTICA epidemiologic study (44) demonstrated that, in an apparently healthy population, plasma total antioxidant capacity was positively associated with consumption of fruit and vegetables. Our results disagree with a recent crossover intervention study examining the effects of an intervention of a high and low total antioxidant capacity diet on markers of antioxidant status, systemic inflammation, and liver dysfunction (45). This study did not observe significant differences in serum total antioxidant capacity measures associated with the 2 interventions. Unlike our study, where there was a wide variation in antioxidant and fruit and vegetable intakes, the intervention study had fewer participants and a similar range of fruit and vegetable intake due to its study protocol. This may have prevented them from detecting differences in serum total antioxidant capacity between the 2 study groups.

In the past there has been criticism leveled against the concept of “total antioxidant capacity” (46). Most assays use either a hydrophilic or lipophilic approach and are unable to capture interactions between fat- and water-soluble antioxidants that exist or antioxidant enzymes that contribute to the antioxidant defense systems in the body. Yeum et al. (47) recently demonstrated that the TAP assay captures the synergistic protective associations between both water- and fat-soluble antioxidants. However, it is important to note that, similar to the shortcomings of other assays, TAP cannot capture direct reactive oxygen or nitrogen scavenging activities.

In the present study, information on nutritional intake was obtained by FFQ. In the past there have been several criticisms levied against this method of dietary assessment (48). However, the FFQ is the method of choice for dietary assessment in large epidemiological studies for ranking nutritional intakes of individuals (49). Also, the FFQ used in the present study was region specific and designed to capture the intake patterns of the study population and had been previously validated for total α-tocopherol and β-carotene intakes using corresponding measures in serum as biomarkers (18,19). Our study participants were from a population-based cohort and were not disease free. TAP may be influenced by the presence of disease and hence final models were adjusted for presence of self-reported hypertension, because it was one of the most commonly reported heath conditions. Although we adjusted for a number of covariates, all observational studies may have residual confounding.

In recent years, several antioxidant supplement trials have shown lack of benefit and, in some cases, potential harm (50,51). This has shifted the focus to overall dietary patterns and food intake rather than single nutrient intakes (52). Although higher total antioxidant status was associated with antioxidant supplement use, our study also clearly demonstrates that modest increases in fruit and vegetable intake (1 serving/d of vegetables or 2 servings/d of fruit and vegetables) were positively associated with total antioxidant status. Antioxidant nutrients that are associated with increased fruit and vegetable consumption include carotenoids and vitamin C (53). Beneficial effects of increased fruit and vegetable consumption have been attributed in part to their antioxidant flavonoids. A review of intervention studies examining the effect of fruit and vegetables on measures of antioxidant capacity attributed postintervention increases to a spike in serum/plasma concentrations of uric acid. According to the authors, fructose intake (mainly due to fruit consumption) results in purine nucleotide degradation from fructose catabolism, resulting in increased uric acid concentrations and, thereby, increased antioxidant capacity measures (54).

In general, we found ∼4% greater serum TAP levels across extreme quartile categories of intakes of total α-tocopherol, β-carotene (total and dietary), total vitamin C, fruit and vegetables, and serum measures of α-tocopherol. This seemingly small increase may be due to the fact that endogenous serum components (e.g. protein, uric acid, etc.) contribute greatly to overall antioxidant capacity, and dietary micronutrients may, thus, play a relatively small role. The clinical importance of this increase in serum TAP levels is not presently known. Further studies measuring TAP in other populations and examining its associations with clinical outcomes are warranted.