Study Design and Intervention
A 19 wk, randomized crossover trial was conducted in 21 generally healthy men and postmenopausal women ≥50 y to study the dose-response effects of chronic walnut intake on biomarkers of antioxidant activity, oxidative stress, and nutrient status. Subjects were asked to refrain from eating any other tree nuts, seeds, peanuts, or ellagitannin-rich foods (including berries, wine, cocoa, tea, and pomegranates) during the entire 19 wk study, and other polyphenol-rich foods (including most fruits, vegetables, beans, legumes, whole grains, olive oil and coffee) for 2 d prior to each study visit. The trial consisted of a 1 wk run-in period, two 6 wk intervention phases, and a 6 wk washout period between interventions. Subjects were randomized to receive either 21 or 42 g packaged raw walnuts/d during the first intervention phase and the alternate dose during the second intervention phase. The doses were based on the FDA qualified health claim that the regular consumption of 1.50 oz (42 g)/d walnuts decreases the risk of heart disease. Packaged walnuts were provided by the California Walnut Commission (Folsom, CA). Fasting plasma samples were collected at the baseline and end of each 6 wk phase. Plasma samples were also collected 1 h post-walnut consumption (included in a standard meal) at each of these visits. Randomization was stratified by gender according to a computer generated list. Study personnel were blinded to the treatment assignment for the duration of the intervention and sample analysis. The only exception was the study dietitian who was responsible for distributing the walnuts and dietary instructions to eligible subjects at randomization, and assessing compliance. During each visit, subjects reported to the Jean Mayer USDA Human Nutrition Research Center on Aging (HNRCA) at Tufts University after fasting for 12 h. At each visit, subjects were queried regarding interval changes in health, as well as use of prescription medications, tobacco, and dietary supplements.
Twenty-one non-smoking men and postmenopausal women, age ≥50 y, with body mass index (BMI) 18.5-35 kg/m2 were recruited from the greater Boston area via newspaper advertisements, direct mailings, and clinic postings. Subjects were excluded if they had a history or known allergy to nuts of any kind; regularly consumed ≥5 oz (140 g) nuts/wk within 6 wk of study admission; were taking estrogen, oral steroids or cholesterol-lowering medications; had renal, endocrine or gastrointestinal disease, rheumatoid arthritis; presented with systolic blood pressure >150 mm Hg and/or diastolic blood pressure >95 mm Hg; had usual ethanol intake ≥2 drinks (28 g)/d or used illicit drugs; presented with EKG or standard clinical laboratory values outside acceptable parameters. Subjects were also excluded if they used dietary supplements containing either fish oils, high doses (≥3× U.S. Daily Value) of vitamins C, E, β-carotene or selenium, and/or phenolic compounds (e.g., herbal or berry-containing preparations) within 6 wk of study admission.
In total, 23 volunteers were recruited for an initial baseline screening visit between February and October 2007. Two subjects declined to participate after the dietary restrictions were reviewed, but before the start of the first 6 wk period. No subject dropped out of the study prior to completion. The study design was approved by the Institutional Review Board of Tufts University Health Sciences Campus and Tufts Medical Center. All subjects signed a written informed consent agreement before participating.
Sample Collection and Preparation
Collected samples were assessed for selected measures of antioxidant activity (ORAC, ORAC with perchloric acid (pca) precipitation, FRAP, and Total Antioxidant Performance [TAP]), lipid peroxidation (MDA), biomarkers of antioxidant status (total thiols, phenols, carotenoids, and glutathione peroxidase [GPX]), micronutrients (α- and γ-tocopherol, folate, pyridoxal phosphate, vitamin C, magnesium), RBC fatty acid profiles, and lipid profile (total, low density lipoprotein [LDL] and high density lipoprotein [HDL] cholesterol, and triglycerides) in fasting plasma at the baseline and end of each walnut intervention phase. Data were also collected on ORAC, ORAC-pca, FRAP, total phenols, and thiols in plasma 1-h after walnut consumption at each visit during the intervention.
Blood samples for the analysis of FRAP, TAP, MDA, total thiols, phenols, fatty acids, lipids, folate, PLP, carotenoids, and vitamins C and E were collected in EDTA-containing evacuated tubes and centrifuged within 15 min of drawing (3000 × g, 15 min, 4°C) with a SUR-Sep cap (Organon Teknika, Durham, NC). Samples for the ORAC and plasma glutathione peroxidase assays were collected in heparin tubes, while samples for the analysis of magnesium were collected in trace metal-free tubes (Becton, Dickinson, and Company, Franklin Lakes, NJ) and processed similarly. Samples for MDA analysis were prepared by adding 10 μL of 50 μM butylated hydroxytoluene (BHT) to 250 μL of plasma. An equal volume of cold 0.35 M perchloric acid was mixed with the plasma sample for vitamin C analysis, followed by centrifugation (2500 × g, 10 min, 4°C). For the fatty acid analyses, packed red blood cells (RBC) were washed with saline and BHT added at 0.4 mg/mL to prevent oxidation. Sample aliquots were stored in 2 mL NUNC tubes (Vanguard Cryotubes, Neptune, NJ) at -80°C. Aliquots for the magnesium analysis were stored in acid-washed storage tubes also at -80°C. All samples for each subject were analyzed within the same run for every assay performed.
Antioxidant activity markers
The ORAC values of whole plasma and perchloric acid treated protein-free plasma were determined according to the method of Huang et al.  and Cao et al. . The assay was carried out on a FLUOstar OPTIMA plate reader using a fluorescence filter with an excitation wavelength of 485 nm and an emission wavelength of 520 nm with the data expressed as μmol Trolox Equivalents (TE)/L. The FRAP value of whole plasma was determined with the spectrophotometric method of Benzie  with the data expressed as μmol TE/L. Serum TAP was determined by the method developed by Aldini et al.  to measure total antioxidant capacity in both the hydrophilic and lipophilic compartments of serum, and validated by Beretta et al.  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) (MeO-AMVN). Oxidation is determined by monitoring the appearance of green fluorescence of the oxidation product of BODIPY with an excitation wavelength of 500 nm and emission wavelength of 520 nm using a 1420 multilabel counter (Wallac Victor 2, Perkin Elmer Life Sciences, MA). 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.
Plasma MDA was measured by HPLC with fluorometric detection  with the data expressed as μmol/L.
Total thiols in plasma were determined spectrophotometrically according to Hu  and the data expressed as mmol/L. The total phenol content of plasma was determined by the Folin-Ciocalteu reaction according to Singleton et al.  with the data expressed as μmol gallic acid equivalents (GAE)/L. Total carotenoids were determined spectrophotometrically according to the method described by Roels et al. . GPX activity was assessed using a kinetic assay with UV detection according to Pleban et al. .
A reverse-phase gradient HPLC method was used for the simultaneous determination of α- and γ-tocopherol . PLP was measured with a radioenzymatic assay described by Camp et at. . Folic acid was assessed with a chemiluminescent immunoassay (Diagnostic Products Corporation, Los Angeles, CA). Vitamin C was measured with an isocratic reverse-phase HPLC procedure according to Behrens et al. . Magnesium was determined with a clinical chemistry analyzer (Olympus AU400, Center Valley, PA) according to the manufacturers' instructions.
RBC fatty acid profiles were quantified using an established gas chromatography method as previously described [26, 27]. Peaks of interest were identified by comparison with authentic fatty acid standards (Nu-Check-Prep, Elysian, MN) and the data expressed as molar percentage (mol %) proportions of fatty acids relative to the internal standard.
Plasma concentrations of total cholesterol, LDL, HDL, and triglycerides were determined with a clinical chemistry analyzer (Olympus AU400, Center Valley, PA) according to the manufacturers' instructions.
Dietary assessments were made at the baseline and end of each 6 wk intervention phase using a validated food frequency questionnaire (Fred Hutchinson Cancer Research Center Food Frequency Questionnaire Version 06.10.88, Cancer Prevention Research Program, Fred Hutchinson Cancer Research Center, Seattle, WA) to determine usual nutrient intakes and detect any significant changes that may have occurred during the intervention periods.
Compliance was measured by having each subject keep track of their walnut intake using a daily diary chart, and by having the study dietitian count opened bags at each visit. Compliance with the walnut regimen was determined to be >90%.
Statistical analyses were performed using SPSS version 15.0 (SPSS, Inc., Chicago, IL). All outcome variables were tested for sequence and dose effects using a linear mixed model; none were found. Paired t-tests were used to compare changes from baseline values for each respective walnut dose at 1 h post-consumption and after 6 wk of daily consumption, and to compare changes between the two walnut doses achieved after each 6 wk intervention. Regression was used to assess the independent predictive capability of the plasma micronutrients on antioxidant activity measures. Student's t test and chi-square analyses were used to compare baseline characteristics between the two groups at randomization. Unless otherwise noted, results are expressed as either mean ± SD or % change from baseline with respective P values for within and between group differences. P values < 0.05 were considered statistically significant.