Subjects and research design
Seventy-three children (42 boys and 31 girls) ranging in age from 7 to 13 years, and in body mass index from 13 to 36 kg/m2, were recruited from local elementary schools and home school programs. Inducements included subject stipends and free results of fitness, body composition, and immune function tests. This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects/patients were approved by the university's Institutional Review Board for Human Studies. Written informed consent was obtained from all subjects and a parent. A parent for each child attended all orientation and test sessions, and assumed responsibility for home feeding, dietary recording, health logs, and transportation of their children to the laboratory.
Triceps and subscapular skinfolds were measured in each child and summed using the procedures of Lohman et al. . The skinfolds were measured by one trained technician using a Lange skinfold caliper (Cambridge Scientific Industries, Cambridge, MA).
Subjects were tested for immune function pre-study, and then again two months later following a regimen of micronutrient supplementation through ingestion of a fortified breakfast cereal. The subjects used a daily health log to record symptoms of sickness using number codes. A pneumococcal vaccine was administered halfway through the study, with the antibody response measured one month later (post-study).
Subjects were randomized to one of three groups--low, medium, or high fortification--with breakfast cereals administered in double blinded fashion. Extruded, expanded, puffed corn cereal products with selected micronutrient blends were supplied by General Mills in coded boxes (Minneapolis, MN). For all groups, each 100 g cereal provided 400 kilocalories, 6.7 g protein, 86.7 g carbohydrate, 1.7 g fat, 3.3 g dietary fiber, 133 mg potassium, and 900 mg sodium. The "low" fortification group received the following nutrients per 100 g cereal: 0.8 mg vitamin C, 133 mg calcium, and 0.6 mg iron. The calcium provided in the low fortification cereal ensured that subjects received partial support for growing bones. The "medium" fortification group received the following nutrients per 100 g cereal: 1.3 mg thiamin, 1.4 mg riboflavin, 16.7 mg niacin, 1.7 mg vitamin B6, 5.0 μg vitamin B12, 20.0 mg vitamin C, 1,667 IU vitamin A, 667 μg folate, 500 mg calcium, 27.0 mg iron, and 12.5 mg zinc. The "high" fortification group received the same amount of nutrients as the "medium" group, with higher amounts of the following nutrients per 100 g cereal: 100 mg vitamin C, 8,333 IU vitamin A, 18.5 mg vitamin E (α-tocopherol equivalents), and 25.0 mg zinc. If 100 g cereal were consumed per day, the "high" fortification cereal would have provided approximately 50% of the U.S. Daily Value.
Subjects were given two coded boxes for each week of the study, with instructions to consume two to three measured cups (~50-70 g) per day (anytime of the day) for two months. Intake was recorded in daily logs by the subjects/parents, and all unconsumed cereal was returned to investigators for weighing to determine actual intake. To remain in the study, subjects had to consume a minimum of 1,700 g cereal during the study period. Subjects (with parental supervision) were instructed to avoid all other forms of nutrient supplements during the 2-month study.
Diet intake was estimated through 3-day food records pre-study, and then again after one and two months. The study dietitian provided detailed instructions using food models and volume measuring supplies to subjects and parents regarding methods for recording volume and portion size in the 3-day food records, and then entered the information into a computerized dietary analysis system, Food Processor Plus (ESHA Research, Salem, Oregon).
Blood and saliva sample collection
Children and parents reported to the testing facility between the hours of 7:30-9:30 am having avoided energy intake for the previous 9 h. A 4-min timed saliva sample was first collected followed by a blood sample drawn by trained pediatric phlebotomists. A delayed-type hypersensitivity (DTH) skin test was administered using the Mantoux method with three antigens (Candida albicans, mumps antigen, and tetanus toxoid). Each subject reported back to the testing lab two days later for a 48-h measure of skin induration. One month later, subjects returned to the testing lab, and were given a pneumococcal vaccination by medical personnel. At the end of the two-month period, saliva and blood samples were again collected, and the DTH skin test readministered.
Immune assay measurements
All blood samples were obtained from an antecubital vein from the children while in the supine position after having rested for more than 15 min. Routine complete blood counts were performed by a clinical hematology laboratory (Lab Corp, Burlington, NC), and provided leukocyte subset counts.
The proportions of T cells (CD3+), B cells (CD19+), and NK cells (CD3-CD16+CD56+) were determined in whole blood preparations and absolute numbers calculated using CBC data to allow group comparisons on blood concentrations of cells. Lymphocyte phenotyping was accomplished by two-color fluorescent labeling of cell surface antigens with mouse anti-human monoclonal antibodies conjugated to fluoresceinisothiocyanate (FITC) and phycoerythrin (PE) using Simultest monoclonal antibodies and isotype controls (Becton Dickinson, San Jose, CA). For immunophenotyping, 50 μl aliquots of heparinized whole blood from each sample were added to five wells of a 96 well plate. Five μg (diluted in 50 μl RPMI ) of each antibody or isotype control were added to appropriate wells for 20 min on ice in the dark with orbital shaking (170 rpm). The cell suspension was then lysed using 200 μl FACSlyse solution (Becton Dickinson) for 10 min in the dark on ice with shaking. Plates were then centrifuged for five min (Beckman GS-R6 Centrifuge) at 1500 × g. Samples were kept at 4°C in the dark until analyzed by flow cytometry (FacsCalibur, Becton Dickinson, San Jose, CA)
Natural killer cell activity (NKCA)
NKCA was assessed using the chromium release assay [34
]. Peripheral blood mononuclear cells were isolated from heparinized blood by density gradient centrifugation with Ficoll and sodium diatrivoate (American Red Cross, Washington D.C.). 51
Chromium labeled K562 target cells were then added (1 × 104
) to each of the wells containing effector cells to yield 40:1 and 20:1 effector to target (E:T) ratios. The assay was performed in triplicate in "V" bottom microtiter plates (Costar, Cambridge, MA). The microtiter plates were then incubated for 4 h at 37°C in a 5% C02 (in air) incubator. At the end of the incubation, the plates were centrifuged for 5 min at 1500 rpm, the supernatants harvested onto Skatron harvesting frames (Skatron, Sterling, VA), and the level of radioactivity measured in a Packard Tri Carb Liquid Scintillation Analyzer model 2500 TR series (Packard Instruments Company, Meriden, CT). Total release of 51
Cr was determined by counting an equal aliquot of resuspended cells in 100 μl of 1% triton-X (Sigma, St. Louis, MO). Spontaneous release was determined by counting the radioactivity in the supernatant of labeled target cells cultured in medium alone. The percent lysis was calculated using the mean counts per min (cpm) of triplicate values for each E:T ratio and the following formula:
Granulocyte phagocytosis and oxidative burst activity
The phagocytosis assay utilized a FITC-labeled bacteria (Staphylococcus aureus; Molecular Probes, Eugene, OR) to quantify the degree of phagocytosis by granulocytes, as described in a previous publication . Briefly, to determine the extent of oxidative burst exhibited by granulocytes, we employed 2',7'-dichlorodihydrofluorescein diacetate (DCF-DA; Molecular Probes), a non-fluorescent molecule which is oxidized to green fluorescent dichlorofluorescein (DCF) as oxygen radicals are generated in the oxidative burst to kill unlabeled Staphylococcus aureus. The white blood cell count was acquired using the Becton Dickinson Unopet manual counting protocol. Using two-color flow cytometric immunophenotying (CD45-FITC/CD13,14-PE), the granulocyte percentage was determined. Bioparticle reagents of unlabeled and labeled Staphylococcus aureus were suspended into PBS at a working concentration of 3 × 105 bioparticles/μl. After determining the number of phagocytic cells in 100 μl of whole blood, and adding 15 FITC-labeled bacteria per cell, the mean channel fluorescence (FITC) was analyzed to determine the degree of engulfed bacteria (non-phagocytized bacteria were quenched with ethidium bromide; final concentration of 200 μM). To determine the oxidative burst activities, either DCF-DA (final concentration, 100 μM) (basal activity level), or DCF-DA and unlabeled bacteria (stimulated activity level) were added to 100 μl whole blood. After incubating the samples for 60 min (37°C) in the dark, lysing the RBC, centrifuging, and resuspending the pellets, the samples (10,000 phagocytes) were acquired on the flow cytometer.
Unstimulated saliva was collected for four min into 5 mL plastic, sterilized vials. Participants were urged to pass as much saliva as possible into the vials during the 4-min timed session. Saliva volume was measured to the nearest 0.1 mL, and then frozen at -80°C until analysis. Salivary IgA was measured by enzyme linked immunosorbent assay (34). The data were expressed as concentration of sIgA (μg . mL-1), concentration of sIgA relative to total protein concentration (μg . mg-1), and salivary immunoglobulin secretion rate (μg . min-1).
Delayed-type hypersensitivity (DTH) skin response
The DTH skin response was assessed with use of three antigens, Candida albicans, mumps antigen, and tetanus toxoid (diluted 5:1), by the Mantoux method with needle and syringe (Allermed Laboratories, Inc., San Diego, CA; Aventis Pasteur, Swiftwater, PA). The volar surfaces of the left and right arms were cleansed and labeled (mumps antigen and tetanus toxoid on the left arm, Candida albicans on the right). At each site, a needle was inserted into the skin at a 45 degree angle to a depth of <0.2 mm, and 0.1 mL of antigen injected until a 5 mm pea-sized bleb was produced. After 48 h, subjects returned to the lab, and the DTH response at each test site measured. The extent of the induration response was palpated at the reaction area (manifested as firmness and redness), outlined with a black-ink ballpoint pen, and then removed with scotch tape prior to mounting on the skin test record form. The tape impress for each induration was measured across two diameters and averaged.
Pneumoccocal vaccination and IgG antibody response
Plasma samples were collected one month before and one month after the children were administered a single 0.5 mL dose of PNEUMOVAX 23 intramuscularly (deltoid muscle) (Merck & Co., Inc., West Point, PA). The plasma samples were assayed for specific IgG antibodies against Pneumococcal Capsular Polysaccharide (PCP) (BINDAZYME™ Anti-PCP IgG Enzyme Immunoassay Kit, MK012, The Binding Site LTD, Birmingham, England). The measuring range of the assay for anti-PCP IgG antibodies levels is 3.3-270 mg/l, with an intra-assay precision of 3.1-5.9% CV and an analytical sensitivity of 0.62 mg/l.
Upper respiratory tract infection log
Subjects with parental assistance recorded URTI symptoms on a daily basis in a log using numbered codes. The following health problems were recorded, in accordance with previous investigations by our research team : 1) No health problems; 2) Cold symptoms (runny, stuffy nose, sore throat, coughing, sneezing, colored discharge); 3) Flu symptoms (fever, headache, general aches and pains, fatigue and weakness, chest discomfort, cough); 4) Nausea, vomiting, and/or diarrhea; 5. Muscle, joint, or bone problems/injury; 6) Other health problems. An URTI episode was recorded if cold (#2 item) or flu (#3) symptoms persisted for two days or longer. The primary outcome reported in this study is the total days with URTI symptoms. URTI severity and duration per episode were not monitored or calculated in this study.
Data are reported as mean ± SD, and were analyzed using SPSS 11.5 (SPSS Inc., Chicago, IL). The dietary intake data were analyzed using 3 (groups) × 3 (times of measurement) repeated measures ANOVA, with immune data analyzed using a 3 × 2 repeated measures MANOVA. When Box's M suggested a violation of homoscedasticity assumption in MANOVA, Pillais trace statistic was used as the test statistic because it has been shown to be robust against departures from covariance equality. If the group × time interaction P value was ≤ 0.05, a change variable was calculated and compared between groups using Bonferroni adjusted Student's t-tests. URTI data from the daily logs were combined into group averages and the total number of sick days recorded was compared between groups using oneway ANOVA. Chi-square analysis was used to compare the number of children across groups who reported at least one URTI episode during the study period. Power analysis was performed after the study was conducted using the fpower macro in SAS (SAS Institute, Inc., Cary, NC), and revealed that 41 to 142 subjects would be needed (depending on the immune measure) to achieve 80% power. The power analysis is a conservative estimate and was conducted a posteriori because data on children for all of the immune measures used in this study were not available prior to the study. We acknowledge that the sample size is considered marginal for this study, but subjects were well randomized, and the results are in accordance with what we and other have observed in adults.