JACN
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Djuric, Z.
Right arrow Articles by Heilbrun, L. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Djuric, Z.
Right arrow Articles by Heilbrun, L. K.
Journal of the American College of Nutrition, Vol. 25, No. 3, 178-187 (2006)
Published by the American College of Nutrition

Effects of High Fruit-Vegetable and/or Low-Fat Intervention on Plasma Micronutrient Levels

Zora Djuric, PhD, Jianwei Ren, MD, Olga Mekhovich, PhD, Raghu Venkatranamoorthy, MEng and Lance K. Heilbrun, PhD

Department of Family Medicine, University of Michigan, Ann Arbor (Z.D.)
Barbara Ann Karmanos Cancer Institute, Wayne State University (J.R., O.M., R.V., L.K.H), Detroit, Michigan

Address correspondence to: Dr. Zora Djuric, University of Michigan, 1500 E. Hospital Drive, Room 2150 Cancer and Geriatrics Center, Ann Arbor, MI 48109-0930. E-mail: zoralong{at}umich.edu


   ABSTRACT
TOP
 FOOTNOTES
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Objectives: Higher plasma micronutrient levels have been associated with decreased cancer risks. The objective of this study was to determine the relative effects of reduced fat and/or increased fruit-vegetable (FV) intakes on plasma micronutrient levels.

Methods: Healthy, premenopausal women with a family history of breast cancer (n = 122) were randomized across four diet arms for one year in a 2 x 2 factorial design study: control, low-fat, high fruit-vegetable and combination low-fat/high FV diets. Levels of plasma micronutrients were measured in plasma at 0, 3, 6 and 12 months.

Results: The high FV intervention, regardless of fat intake, significantly increased {alpha}-carotene, ß-carotene and vitamin C levels in plasma. Only the combination high FV, low-fat intervention significantly increased plasma ß-cryptoxanthin and zeaxanthin levels over time. Although {alpha}-tocopherol was not affected, a potential concern is that the low-fat intervention resulted in significantly decreased both {gamma}-tocopherol dietary intakes and plasma levels, regardless of whether or not FV intakes were concomitantly increased.

Conclusions: Unlike {alpha}-tocopherol, {gamma}-tocopherol plasma levels were decreased by a low fat diet, perhaps because {gamma}-tocopherol is not generally added to foods nor widely used in vitamin E supplements. The decreased dietary intakes and plasma levels of {gamma}-tocopherol with a low-fat diet may have implications for health risks since the biological functions of the different tocopherol isomers have been reported to be distinct.

Key words: high fruit vegetable diet, low-fat diet, carotenoids, tocopherols, plasma levels


   INTRODUCTION
TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Increased fruit and vegetable consumption has been suggested to be associated with decreased cancer risks in epidemiological studies, but the findings are not consistent. The protective effect of increased intake of fruits and vegetables (FV) is more readily apparent in case/control studies than in cohort studies [1–3]. In a meta analysis of several large cohort studies, intake of only one of the individual vegetables examined had a significant association with breast cancer risk, and that was a protective effect for spinach intake [1]. There are, however, problems inherent with use of food frequency questionnaires and the question remains whether the range of intakes within a cohort is wide enough to detect a protective effect. Using measured serum levels of ß-carotene, which is a carotenoid found in many types of FV, fairly strong protective effects were found with higher levels in two large prospective studies, and relative risk was decreased by more than half in the groups with the highest ß-carotene levels [4,5]. In the prospective Nurse’s Health Study, risk of breast cancer was 25–35% lower for women in the highest quintiles of plasma carotenoids, versus the lowest quintiles [6]. These studies highlight the importance of understanding better the dietary factors that can affect plasma carotenoid levels.

The increase in FV that is needed to substantially change plasma carotenoid levels remains a question, and this is further complicated by the fact that errors in estimating intakes can be a problem [7]. A large intervention trial with a goal of increasing FV intakes to 5–8 servings/day failed to show more than a 10% increase in serum carotenoids [8]. In our previous study, an increase in FV intake to from 3.3 to 5.2 servings/day (excluding potatoes) during a low-fat intervention was not sufficient to significantly increase blood carotenoid levels [9]. When intake goals are higher, 9 servings/day or greater, then increases in plasma carotenoids can be substantial, and plasma carotenoids can be used as a marker of FV intake [10–12]. For example, in the Women’s Healthy Eating and Living trial that seeks to examine the effects of concomitant lower fat and higher FV intakes on breast cancer recurrence, plasma levels of individual carotenoids increased between 4 and 223% in the intervention arm [13].

Dietary factors, such as fat intake, that can potentially influence absorption of micronutrients found in FV also need to be examined. Vitamin C is a water soluble compound, but carotenoids and tocopherols are fat soluble. A review of several studies has indicated that the amount of fat needed for complete absorption of carotenoids is 3–5 g per meal, which is a modest amount and would be easily achieved by persons following a low-fat diet [14]. The data are, however, sparse and most of the studies were done with carotenoid supplements instead of foods. Absorption of both carotenoids and {alpha}-tocopherol in rats has been shown to be increased by increased dietary lipid in the animal model [15–17]. All micronutrients, however, may not be affected in a similar way. For example, fat content has been shown to affect the bioavailability of lutein esters more than that of {alpha}-carotene, ß-carotene or {alpha}-tocopherol [18]. Another very important aspect of low-fat diets that might influence plasma micronutrients is the possibility of decreased tocopherol intakes, since tocopherols are plentiful in vegetable oils [19]. In an {alpha}-tocopherol supplementation trial, persons consuming a high fat diet had higher plasma levels of {alpha}-tocopherol than those consuming a low-fat diet [20].

In this study, we randomized healthy, pre-menopausal women to one of four diets for 12 months: non-intervention, low-fat, high FV, and combination low-fat, high FV. The low-fat goal was 15% of energy from fat and the high FV goal was 9 servings/day in specific categories to increase variety of intake [21]. Since fat intake can affect absorption of fat-soluble micronutrients, it was of interest to examine the effects of high FV intervention with and without a concomitant decrease in fat intake on plasma levels of carotenoids, tocopherols, and vitamin C.


   METHODS
TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Subjects
The Nutrition and Breast Health Study was approved by the Human Investigation Committee of Wayne State University. Women were recruited through community advertising, and they participated in the trial between April 23, 1997 and October 22, 1999. Pre-menopausal women, ages 21–50, were eligible if they had: 1) at least one first degree relative with breast cancer, 2) a current benign mammogram or breast exam with follow-up recommendation of 1 year or more, 3) no expected changes in use of oral contraceptives, 4) good general health, and 5) no expected changes in lifestyle during the study. The dietary criteria for eligibility were a fat intake of 25% of total energy or greater and a FV intake of 5 or fewer servings per day on 4-day food records. We excluded persons taking supplements containing more than 150% of the RDAs for vitamins and minerals.

Eligible women gave their signed, informed consent to participate in this study. Subjects were stratified by race (Caucasian and non-Caucasian) and randomized across four diet arms: control, low-fat, high FV, and a combination of low-fat and high FV. The study used a 2 x 2 factorial design. Participants all were asked to give blood samples by venipuncture, after an overnight fast, at 0, 3, 6, and 12 months into the study, and they were paid $25 for each blood sample. Of the 160 eligible women, 122 were interested in participating and were enrolled onto the study. Ninety-seven women (80% of those enrolled) completed the one-year intervention. The numbers of plasma samples available for successful analysis of micronutrients was 121 at baseline, 101 at 3 months, 97 at 6 months, and 97 at 12 months. Blood could not be obtained from one woman at baseline. Mean dietary intake of FV at baseline was 3.84 servings/day, and mean fat intake was 32% of energy.

Dietary Counseling
Women in the non-intervention arm received no dietary counseling and were told that they should continue following their own usual diet. They received a one-page Daily Food Guide Pyramid from the National Dairy Council as a guide for healthy eating; however, this information was not discussed. None of the other arms received this information sheet. For the intervention arms, both individualized in-person counseling and monthly group meetings were implemented. The counseling was bi-weekly initially and then monthly once women became adept at meeting dietary goals. In all intervention arms, women were given intake goals using modified American Dietetic Association exchange lists, and the two low-fat arms also received a daily fat-gram goal [21].

The goal of the low-fat arm was to reduce fat intake to 15% of total energy while keeping FV and total energy consumption constant. The percentage of energy from carbohydrates increased to about 70% of total energy, while protein content remained constant. The goal for the high FV arm was nine servings of FVs per day in a specified variety to increase carotenoid intakes: 1 serving of a dark green vegetable high in lutein; 1 serving of a dark orange vegetable high in {alpha}-carotene; 1 serving of a red product high in lycopene; 2 servings of other vegetables; 2 servings of vitamin C rich fruits; and 2 servings of other fruits (one serving was defined as approximately 60 kcal for fruit and 25 kcal for most vegetables) [21]. The carotenoid ß-carotene is more universally present in FV while ß-cryptoxanthin is mainly found in fruits [12]. Counseling for maintenance of baseline energy intakes included emphasis on substitution of other FV for other carbohydrates. For the combination arm, both grams of fat and servings of FVs were enumerated to meet goals of 15% energy from fat and 9 servings/day of FV. This diet resulted in energy from fat largely being replaced with energy from FV.

Dietary Micronutrient Intakes
Self-reported compliance to the diets using 4-day food records was excellent [21], and micronutrient intakes are shown in Table 2. The 4-day food records were kept prior to the visits at 0 and 12 months and were reviewed in person with each participant. Un-announced 24-hour recalls done by telephone in the 2 weeks before the visit and 24-hour recalls were done at the time of the blood draw. These assessments by recall gave very similar means as the 4-day food records shown in Table 2. The recalls and food records were all analyzed using the Nutrition Data System 2.93 software, nutrient database version 24, food database 14A (Nutrition Coordinating Center, University of Minnesota). As reported previously, the 4-day food records indicated that intakes of fat decreased to about 16% of energy in the two low-fat arms while intakes of FV in the high FV arms reached about 11 servings/day using the serving sizes devised for this study [21]. In the two low-fat arms, only four of 47 women had a fat intake above 20% of energy from fat at 12 months on 4-day food records. In the two high FV arms, FV intake ranged from 7.3 to 16 servings/day, with only three of 48 women consuming less than 9 servings/day at 12 months. These increases in FV intakes are reflected in the increased carotenoid and vitamin C intakes in Table 2.


View this table:
[in this window]
[in a new window]
  		Table 2. Dietary Micronutrient Intakes from 4-day Food Records (Total Intake from Foods and Supplements), Values Shown Are Mean and SD underneath in mg/1000 kcal/Day except for ß-Carotene in µg/1000 kcal/Day

 
Plasma Micronutrient Levels
Plasma was prepared immediately after drawing venous blood into heparinized tubes, and plasma was frozen at –80°C under argon until analysis by HPLC. One aliquot of plasma was mixed with 2 volumes of 10% meta-phosphoric acid before freezing for later analysis of vitamin C levels. Plasma samples had been frozen 1–4 years before the analyses were done. All samples from each woman were analyzed on the same day, and laboratory personnel were blinded to diet arm assignment.

All experiments were carried out in subdued light. Plasma samples were thawed in batches together with a quality control plasma sample from one donor (obtained from the Red Cross). Aliquots of plasma, 200 µl, were mixed with 200 µl ethanol containing 0.05% BHT and 10 µl of the internal standard, Tocol (a generous gift of F. Hoffmann-LaRoche, Ltd., Basel, Switzerland). The stock Tocol solution was diluted with ethanol to an absorbance of 1.0 before use. The plasma mixture was vortexed for 30 sec, and extracted with 2 ml hexane containing 0.05% BHT by vortexing for an additional 60 sec and centrifugation at 4°C. The hexane extraction was repeated once, and extracts were combined and dried in a Speed-vac. Samples were reconstituted in 100 µl ethanol containing 0.05% BHT and were sonicated for 5 min. Prior to injection, all extracts were filtered using microfilterfuge tubes (0.2 µm nylon-66, Rainin, Woburn, MA). A 20 µl aliquot was then injected onto HPLC.

HPLC analysis of plasma carotenoids was accomplished using a YMC C-30 column (3 mm, 4.6 x 250 mm) and a C-30 guard cartridge. Mobile phase A consisted of methanol with 20 mM of ammonium acetate, and mobile phase B was 20% methanol, 75% methyl-tert-butyl ether and 5% hexane with 20 mM of ammonium acetate (prepared from 1M solution, pH 4.5). There was an initial 3 minute hold of 90% A, 10% B followed by a linear gradient to 100% B from 3 to 27 minutes, and a 6 minute final hold at 100% B. The flow rate was 1.0 ml/min. An ESA CoulArray electrochemical detector (Chelmsford, MA) was used for measurement of tocopherols and carotenoids. Three electrodes were set up at the following potentials: 310, 390 and 470 mV, thus allowing obtaining multiple electrochemical profiles for each compound of interest. The elution times of each compound were: tocol-7.7 min (internal standard), {gamma}-tocopherol-10.1 min, {alpha}-tocopherol-11.7 min, lutein-14.6 min, zeaxanthin-15.7 min, ß-cryptoxanthin-20.6 min, {alpha}-carotene-23.9 min, ß-carotene-24.6 min and lycopene-30.5 min. Peak areas for the incompletely separated cis and trans isomers of lycopene were added together to determine "total" lycopene concentration.

Calibration curves were constructed for each compound using five different concentrations, and each sample was injected twice. Standard Reference Material (SRM) 968c (National Institutes of Standards Technology, Bethesda, MD), which consists of two vials of lyophilized human serum, with low and high levels of carotenoids, was used in validating the calibration curves. This laboratory participated in the "round-robin" conducted by the NIST to compare our results with those of other laboratories, and we were within 1 standard deviation of the mean for all analytes at the time when these samples were analyzed. The inter-assay coefficients of variation were: 8.7% {gamma}-tocopherol, 12.6% {alpha}-tocopherol, 11.6% lutein, 9.5% zeaxanthin, 3.6% ß-cryptoxanthin, 7.3% {alpha}-carotene, 11.6% ß-carotene and 8.7% for lycopene.

HPLC analysis of vitamin C was also accomplished using electrochemical detection, based on the methods of Margolis et al.[22]. The HPLC column was a Waters µ-Bondapak C18 5 um 3.9 x 300 mm column with a C-18 guard column (Waters, Milford, MA). The mobile phase consisted of 0.1 M sodium phosphate and 2.5 mM disodium-EDTA, filtered through a 0.2 µm membrane filter and adjusted to pH 3.0 with o-phosphoric acid. The mobile phase flow rate was 1.0 ml/min. Three electrodes were set at the following potentials: 50, 100 and 200 mV, and calculations were based on the sum of peak areas for all three potentials. The retention time of ascorbic acid was 4 minutes.

L-ascorbic acid (> 99.0%) was purchased as a standard from Sigma-Aldrich Company (St. Louis, MO). The stock solution in 6.66% m-phosphoric acid was kept at –20°C for less than week. The UV absorbance of the stock at 244 nm was checked prior to use as a HPLC standard. Calibration curves were constructed using 0, 10, 20, 30, 40 and 50 µM ascorbic acid. The intra-assay cv varied from 2 to 7% for high and low concentrations of ascorbic acid, respectively. We also participated in the "round-robin" conducted by the National Institute of Standards Technology (NIST) to validate our results with those of other laboratories, and our values were within 5% of the median value. Serum samples, 100 µl, were mixed with 50 µl 0.5 M Trizma buffer (pH 9.3) containing 2 mM DTT. The mixture was allowed to react at 25°C for 5 minutes, and the reduction reaction was quenched by adding 50 µl of 0.2 M sulfuric acid to the solution on ice. All solutions were filtered using 0.2 µm nylon-66 microfilterfuge tubes (Rainin, Woburn, MA) prior to HPLC analysis of 20 µl aliquots.

Total cholesterol, HDL and triglycerides were determined using Sigma Infinity kits (St. Louis, Mo.). Lipid and triglyceride levels did not change significantly over time in any diet arm (Table 1) and total cholesterol was used for adjusting plasma levels of micronutrients in the analyses.


View this table:
[in this window]
[in a new window]
  		Table 1. Plasma Lipid and Triglyceride Levels with Time in the Four Study Arms (Given in mg/dl as Mean, SD and n)*

 
Statistical Methods
Dietary intakes from the 4-day food records and 24-hour recalls were analyzed by the Nutrition Data System, which generated data for dietary intakes of ß-carotene and individual tocopherols. Using all available data, mean intakes from 4-day food records at baseline and 12 months are shown in Table 2 for each diet arm. Dietary data were transformed as needed to achieve normal distributions prior to analysis (see footnotes to Table 2). Mean baseline and 12 month micronutrient intakes from 4-day food records were compared using time point contrasts as part of incomplete mixed models ANOVA of all available data. The resulting p-values were interpreted using the method of Holm to control for multiple comparisons [23]. Correlations of dietary intakes from records and recalls with plasma levels were also done at both 0 and 12 months. Nutrient intakes were expressed per 1000 kcal and plasma micronutrient levels were expressed per mg cholesterol except for vitamin C which was expressed per volume of plasma. After transformation of the data to achieve normality, Pearson correlation coefficients were calculated and p-values were again interpreted using the method of Holm [23].

For plasma levels, simple descriptive statistics were used to summarize the 9 blood micronutrients by diet intervention arm and by time on study (Figs. 1 and 2). To deal with occasional missing data at any of the 4 time points, incomplete mixed models repeated measures analysis of variance (ANOVA) was used to model the mean levels of each blood micronutrient. This allowed analysis of all available data, consistent with the intention to treat principle. Prior to any modeling, each endpoint required a Normalizing transformation: square root for gamma tocopherol, fifth root for lutein, cube root for zeaxanthin, fourth root for lycopene, and natural logarithm for the remaining 5 blood micronutrients. For lycopene, the distributions at 3 and 6 months remained significantly non-normal despite investigation of 11 different transformations, hence the robustness of the modeling procedure for that micronutrient at 2 of the 4 time points was relied upon. The modeling of each blood micronutrient was conducted using the MIXED procedure in SAS Version 8.2 [24,25].


Figure 1
View larger version (9K):
[in this window]
[in a new window]
  		Fig. 1. Plasma levels of {gamma}-tocopherol, {alpha}-tocopherol and vitamin C with time on study for the four diet groups. Data shown are the mean and SE.

 

Figure 2
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2. Plasma levels of {alpha}-carotene, ß-carotene, lutein, zeaxanthin, lycopene, and ß-cryptoxanthin with time on study for the four diet groups. Data shown are the mean and SE.

 
Consistent with the 2 x 2 factorial study design, the two dietary intervention effect variables were the low-fat effect (yes/no) and the high FV effect (yes/no). We examined their simultaneous individual (i.e., main) effects, and their interaction effect. An interaction effect occurred if the main effect of say, low-fat intake upon a plasma micronutrient depended in a statistically important way upon the level of the other main effect (i.e., presence or absence of high FV intake). An example of an interaction is when a low-fat effect is operative only for the women who were not maintaining high FV intake. In the absence of a significant interaction effect between the two dietary intervention variables, then the main effect of each one was directly interpreted as being similar throughout our study population. For example, in the absence of interaction, the low-fat effect was assessed by comparing two groups of women: the women who did receive a low-fat intervention vs. those who did not. The ‘effect’ of the low-fat dietary intervention on mean plasma carotenoid levels was then significant if there was a sufficient difference in the model-fitted slopes (and/or curvatures) between the women who did follow low-fat intake vs. those who did not. The sign of the model parameter (e.g., the slope parameter) reconfirmed which group (intervention or no intervention) tended to have statistically higher (or lower) plasma micronutrient levels over time. In a similar fashion, the high FV effect on micronutrient levels was assessed by comparing the women who were assigned to a high FV intervention vs. those who were not when there was no significant interaction effect between the two interventions. This analytic approach was a benefit of the 2 x 2 factorial design, and it provided higher power to detect a significant effect of a dietary intervention since it used n/2 vs. n/2 subjects. Comparison of individual diet arms would typically use only n/4 vs. n/4 subjects. In cases where there was a significant interaction effect, the influence of each intervention effect variable (high FV and low-fat) was analyzed after stratifying on the other one, which is equivalent to performing selected contrasts of individual diet arms.

The linear effect of time (i.e., slope), and quadratic (i.e., squared) effects of time (i.e., curvature) were modeled. We included interaction effects of time with the low-fat effect (LFE) and with the high FV effect (HFVE) (and their interaction), as well as the inclusion of interaction effects of time2 with LFE and with HFVE (and their interaction). Each of these transformed, incomplete repeated measures micronutrient variables was modeled using mixed model repeated measures analysis of variance (ANOVA). The 11 predictors were: LFE, HFVE, their interaction (a cross-product term); time, its interactions with LFE, HFVE, and LFE*HFVE; and time2 along with its interactions with LFE, HFVE, and LFE*HFVE. For each micronutrient, the statistical modeling was done after preliminary analysis to find the best of 14 covariance structures for its 4 repeated measures (at 0, 3, 6 and 12 months) based on smallest value of Aikake’s Information Criterion (AIC). For each plasma micronutrient we also adjusted for total plasma cholesterol as a time-dependent covariate (linear term only) in the mixed model for that plasma micronutrient.


   RESULTS
TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
ACKNOWLEDGMENTS
 REFERENCES
 
Blood Lipids and Triglycerides
Plasma levels of total cholesterol, HDL, LDL and triglyceride levels are shown by diet arm in Table 1, and these data were analyzed using mixed model repeated measures analysis of variance in accordance with the intention to treat principle. There were no statistically significant changes over time in mean levels of any of these four plasma lipid variables. For total plasma cholesterol, there was a weak interaction effect of the low-fat effect with time (p = 0.078), and the slopes of the models suggested that total plasma cholesterol decreased in women receiving a low-fat intervention while it increased in those without a low-fat intervention. The only other marginally significant effect was on plasma LDL (p = 0.078), again with a decrease over time in women with a low-fat intervention and an increase for those without low-fat intervention. These trends for decreases with a low-fat diet and increases in the absence of low-fat intervention may be related to changes in body weight that we reported previously [21]. The mean changes in body weight from 0 to 12 months were 0.4 pounds in the control arm, –5.3 pounds in the low-fat arm, 5.6 pounds in the high FV arm and –2.3 pounds in the combination arm.

The Low Fat Effect
The low-fat intervention was associated with significant changes in the levels of only one plasma micronutrient variable: {gamma}-tocopherol. Thus, the statistical main effect of the low-fat intervention upon mean plasma {gamma}-tocopherol levels was independent of the high FV effect. Specifically, the fitted slopes of plasma {gamma}-tocopherol levels for women maintaining a low-fat diet vs. those not doing so were significantly different (p < 0.001). Women randomized to the low-fat intervention had significantly lower mean levels of {gamma}-tocopherol in plasma relative to those without low-fat intervention (Fig. 1). No other plasma micronutrient levels were affected significantly by the low-fat intervention. Mean plasma lycopene levels differed significantly by diet arm, but this was not a time-dependent difference.

The High Fruit and Vegetable Effect
The high FV intervention was associated with significant changes in the levels of 3 plasma micronutrients: {alpha}-carotene, ß-carotene, and vitamin C. The statistical main effect of the high FV intervention upon the mean plasma levels of each of those 3 micronutrients was independent of the low-fat intervention. Specifically, the fitted slopes of women maintaining high FV intake vs. those not doing so were significantly different for plasma {alpha}-carotene (p < 0.001), ß-carotene (p = 0.002), and vitamin C (p = 0.006). The women maintaining high FV intake had significantly higher mean plasma levels of each of these 3 micronutrients than did the women who were not maintaining high FV intake (Fig. 2). The fitted curvatures in the patterns of means over time were significantly different by high FV status for only plasma {alpha}-carotene (p = 0.003) and ß-carotene (p = 0.011).

The only other statistically significant high FV effect was on plasma ß-cryptoxanthin, but it was a conditional association. Due to a significant intervention interaction effect, mean plasma ß-cryptoxanthin levels differed by high FV intervention status, but only for women also maintaining low-fat dietary intake (Fig. 2). Specifically, the fitted slope for plasma ß-cryptoxanthin levels in women on the combination high FV/low-fat arm was significantly different from the slope for women on the low-fat arm (p < 0.001). The fitted curvatures in the patterns of means over time were also significantly different between these two diet arms (p = 0.001). The combination arm was associated with significantly higher mean plasma ß-cryptoxanthin levels over time as compared to the low-fat arm.

Other Effects
Plasma zeaxanthin levels of women on the combination arm had significantly different fitted positive slope (p = 0.033) than did the women on the other three arms combined. Plasma zeaxanthin levels rose sharply at 3 months, then fell somewhat, but always stayed higher than baseline (Fig. 2). Statistically, the levels in the combination arm were significantly higher over time as compared to levels in the other three diet arms combined. Mean plasma levels in the other three diet arms combined tended to stay the same or to decrease slightly.

The only other statistically significant association was for plasma lutein, and it was a conditional association. Due to a significant intervention interaction effect, mean plasma lutein levels differed significantly by individual diet arm. The fitted slope for plasma lutein in women on the combination high FV/low-fat arm was significantly different from the slope in women on the other 3 arms combined (p = 0.005). The combination arm was associated with significantly increased mean plasma lutein levels over time as compared to the other 3 arms combined (Fig. 2). This may be due, at least in part, to higher baseline plasma lutein levels in women randomized to the combination arm.

Correlations of Dietary Intakes with Plasma Levels
Dietary intakes of ß-carotene, vitamin C, {alpha}- and {gamma}-tocopherols were available from three different assessment measures: 4-day food records, un-announced 24-hour recalls and 24-hour recalls done at the time of blood draw (blood recalls). These dietary intakes were expressed per 1000 kcal to control for differences in reported energy intakes. Pearson correlation coefficients were calculated with plasma levels of these micronutrients expressed per mg cholesterol. These revealed stronger correlations for plasma micronutrients with dietary intakes from 4-day food records after intervention versus at baseline (Table 3). Plasma levels of {alpha}-tocopherol did not correlate significantly with intakes from food records or recalls at either time point, but plasma levels of {gamma}-tocopherol were correlated significantly with dietary intakes from all three methods at 12 months (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3. Correlation of Plasma Levels with Dietary Intakes

 

   DISCUSSION
 
This study is unique in that the independent and interactive effects of a high FV and low-fat intervention on plasma micronutrient levels could be examined. There were a number of interesting and potentially important findings. It has been suggested previously that a low-fat diet may result in decreased dietary intakes of tocopherols, which are found in vegetable oils [19]. For example, in one study increased use of reduced-fat foods was shown to decrease intakes of vitamin E [26]. It is, however, possible to maintain {alpha}-tocopherol intake during a reduced-fat diet, depending on the method used to reduce fat intake [27]. Here we show that a low-fat diet, regardless of whether or not it was accomplished with a simultaneous increase in FV consumption, resulted in significantly decreased plasma {gamma}-tocopherol levels (Fig. 1). Plasma levels of {alpha}-tocopherol were not decreased, perhaps because supplement use was allowed up to 150% of the RDA, and {alpha}-tocopherol is often added to foods, such as cereals. Plasma levels of {alpha}- and {gamma}-tocopherols reflected their dietary intakes, and plasma {gamma}-tocopherol (but not {alpha}-tocopherol) intake and levels exhibited significant decreases in women receiving the low-fat intervention (Results, Tables 2 and 3). This decrease in plasma levels of {gamma}-tocopherol with a low-fat intervention is of potential concern since {gamma}-tocopherol has shown promise for cancer prevention. In both experimental systems and epidemiological studies, [28,29].

The high FV intervention resulted in significantly higher {alpha}-carotene, ß-carotene and vitamin C in plasma. The increase in plasma vitamin C was linear over time while the increase for {alpha}- and ß-carotene had both linear and curvilinear components (Figs. 1 and 2). Plasma levels of {alpha}- and ß-carotene increased almost 3-fold, which is similar to the large increases seen in the Women’s Healthy Eating and Living study that used an intervention goal of 8 FV servings/day in addition to 16 ounces of juice [30]. It is important to note that concomitant counseling for a decreased fat intake had no significant effect on the increase in {alpha}- and ß-carotene plasma levels with the high FV intervention, indicating that these carotenoids are similarly bio-available from a high FV diet when fat intake is or is not decreased.

The high FV intervention also resulted in significantly higher plasma ß-cryptoxanthin and lutein levels over time, but this was only true in the combination arm relative to the other three arms (see Results). It is not clear why this might be the case since dietary intakes were similar in the high FV and combination arms (Table 2). Correlation coefficients for usual dietary intakes with plasma levels of carotenoids, however, generally have not been greater than 0.5, including those in our study. This was true for four-day food records or recalls done either un-announced or at the time of blood draw (Table 3). Four-day food records can be used to represent usual intakes, but they have the disadvantage that even if records are reported accurately, the very act of record-keeping can affect intakes [31]. Overall, however, the strength of the associations was similar with all assessment instruments. It was of particular interest to examine correlations with dietary intakes over the 24-hours prior to the blood draw to determine if recent intakes affect blood levels to a greater extent. A recent study, however, indicated that usual intakes from 4-day food records were more closely associated with plasma micronutrients than with dietary intakes from 24-hour recalls taken at the time of the blood draw [32]. In our study this was only true in three instances, namely for ß-carotene and vitamin C at 12 months and {gamma}-tocopherol at baseline.

The generally low correlations of dietary intakes with plasma levels indicate that other factors need to be considered. In the study of Michaud et al., for example, the correlation coefficients differed quite a bit between men and women [33]. In our study, it might have been expected that the low-fat goal might interfere with absorption of carotenoids resulting in lower plasma levels in the combination arm than the high FV arm, but this was not observed. Lutein ester bio-availability, which is similar to that of lutein [34], has been shown to be affected negatively by fat intake more than that of {alpha}- or ß-carotene using carotenoid supplementation [18], and plasma lutein was one of the carotenoids that was specifically higher in the combination arm than the other three arms. This may be due to differences in carotenoid exposure from foods versus supplements. With the high FV diet, however, we reported previously that weight gain of a few pounds did occur over 12 months of intervention [21], and increased body mass index has been shown to be negatively correlated with carotenoid levels [35–37].

Another somewhat unexpected finding was the lack of an increase in plasma lycopene levels with either high FV intervention. Dietary intake of lycopene increased almost two-fold in the two high FV arms (Table 2). The lack of a statistically significant increase in plasma levels of lycopene was therefore un-expected despite the modest sample size of this study. This may be due to differential bio-availability and/or metabolism of lycopene, or due to the nature of the food sources selected by the subjects to satisfy their intake goals. For example, lycopene availability is greatly influenced by food processing, with greater bioavailability from processed tomato products than fresh tomatoes [38]. To obtain an increase in plasma lycopene, the intervention may need to target processed tomato intake rather than any lycopene-containing food. An intervention using tomato sauce increased lycopene levels two-fold in men with prostate cancer [39]. Another consideration is that lycopene is readily oxidized [40,41], and oxidized lycopene products were not measured in this study. Results also may differ when studying populations that have initially low intakes or after a wash-out period to result in lower baseline levels prior to intervention, as is often done in supplementation trials.

In summary, the high FV intervention, regardless of fat intake, increased {alpha}-carotene, ß-carotene and vitamin C in plasma. Only the combination high FV, low-fat intervention, however, increased plasma ß-cryptoxanthin, zeaxanthin and lutein levels, making the combination intervention appear to be more beneficial for increasing plasma carotenoid levels than the high FV intervention. The decrease in plasma {gamma}-tocopherol levels, however, is a concern with the low-fat intervention. The low-fat diet, achieved with either substitution of fat with FV or with other carbohydrates, resulted in significantly decreased {gamma}-tocopherol dietary intakes and plasma levels, which may have impact on use of low-fat diets for cancer prevention. In order to fully realize the benefits of a high FV diet, it may not be prudent to decrease fat intake to 15% of energy unless intakes of all tocopherol isomers can be maintained through specific food choices.


   ACKNOWLEDGMENTS
TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 ACKNOWLEDGMENTS
REFERENCES
 
We thank the women who gave their time to participate in the Nutrition and Breast Health Study. Janice B. Depper and Kathleen M. Poore were dietitians for the study. F. Hoffman-LaRoche Ltd., Basel, Switzerland generously provided the Tocol and lycopene HPLC standards.


   FOOTNOTES
TOP
 FOOTNOTES
ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 ACKNOWLEDGMENTS
 REFERENCES
 
This work was funded by grants UO1CA77297 and P30 CA22453 from the National Institutes of Health.

Received September 7, 2004. Accepted December 19, 2005.


   REFERENCES
TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Smith-Warner SA, Spiegelman D, Yaun SS, Adami HO, Beeson WL, van den Brandt PA, Folsom AR, Fraser GE, Freudenheim JL, Goldbohm RA, Graham S, Miller AB, Potter JD, Rohan TE, Speizer FE, Toniolo P, Willett WC, Wolk A, Zeleniuch-Jacquotte A, Hunter DJ: Intake of fruits and vegetables and risk of breast cancer: a pooled analysis of cohort studies.JAMA285 :769 –776,2001 .[Abstract/Free Full Text]
  2. Gandini S, Merzenich H, Robertson C, Boyle P: Meta-analysis of studies on breast cancer risk and diet: the role of fruit and vegetable consumption and the intake of associated micronutrients.Eur J Cancer36 :636 –646,2000 .[Medline]
  3. Clavel-Chapelon F, Niravong M, Joseph RR: Diet and breast cancer: review of the epidemiologic literature.Cancer Detect Prev21 :426 –440,1997 .[Medline]
  4. Toniolo P, Van Kappel AL, Akhmedkhanov A, Ferrari P, Kato I, Shore RE, Riboli E: Serum carotenoids and breast cancer.Am J Epidemiol153 :1142 –1147,2001 .[Abstract/Free Full Text]
  5. Sato R, Helzlsouer KJ, Alberg AJ, Hoffman SC, Norkus EP, Comstock GW: Prospective study of carotenoids, tocopherols, and retinoid concentrations and the risk of breast cancer.Cancer Epidemiol Biomarkers Prev11 :451 –457,2002 .[Abstract/Free Full Text]
  6. Tamimi RM, Hankinson SE, Campos H, Spiegelman D, Zhang S, Colditz GA, Willett WC, Hunter DJ: Plasma carotenoids, retinol, and tocopherols and risk of breast cancer.Am J Epidemiol161 :153 –160,2005 .[Abstract/Free Full Text]
  7. Byers T: On the hazards of seeing the world through intervention-colored glasses.Am J Clin Nutr78 :904 –905,2003 .[Free Full Text]
  8. Schatzkin A, Lanza E, Corle D, Lance P, Iber F, Caan B, Shike M, Weissfeld J, Burt R, Cooper MR, Kikendall JW, Cahill J: Lack of effect of a low-fat, high-fiber diet on the recurrence of colorectal adenomas. Polyp Prevention Trial Study Group.N Engl J Med342 :1149 –1155,2000 .[Abstract/Free Full Text]
  9. Djuric Z, Uhley VE, Naegeli L, Lababidi S, Macha S, Heilbrun LK: Plasma carotenoids, tocopherols, and antioxidant capacity in a 12-week intervention study to reduce fat and/or energy intakes.Nutrition19 :244 –249,2003 .[Medline]
  10. Martini MC, Campbell DR, Gross MD, Grandits GA, Potter JD, Slavin JL: Plasma carotenoids as biomarkers of vegetable intake: The University of Minnesota Cancer Prevention Research Unit feeding studies.Cancer Epidemiol Biomarkers Prev4 :491 –496,1995 .[Abstract]
  11. Thompson HJ, Heimendinger J, Haegele A, Sedlacek SM, Gillette C, O’Neill C, Wolfe P, Conry C: Effect of increased vegetable and fruit consumption on markers of oxidative cellular damage.Carcinogenesis20 :2261 –2266,1999 .[Abstract/Free Full Text]
  12. Le Marchand L, Hankin JH, Carter FS, Essling C, Luffey D, Franke AA, Wilkens LR, Cooney RV, Kolonel LN: A pilot study on the use of plasma carotenoids and ascorbic acid as markers of compliance to a high fruit and vegetable dietary intervention.Cancer Epidemiol Biomarkers Prev3 :245 –251,1994 .[Abstract]
  13. Pierce JP, Newman VA, Flatt SW, Faerber S, Rock CL, Natarajan L, Caan BJ, Gold EB, Hollenbach KA, Wasserman L, Jones L, Ritenbaugh C, Stefanick ML, Thomson CA, Kealey S: Women’s Healthy Eating and Living (WHEL) Study Group: Telephone counseling intervention increases intakes of micronutrient- and phytochemical-rich vegetables, fruit and fiber in breast cancer survivors.J Nutr134 :452 –458,2004 .[Abstract/Free Full Text]
  14. van Het Hof KH, West CE, Weststrate JA, Hautvast JG: Dietary factors that affect the bioavailability of carotenoids.J Nutr130 :503 –506,2000 .[Abstract/Free Full Text]
  15. Tso P, Lee T, DeMichele SJ: Randomized structured triglycerides increase lymphatic absorption of tocopherol and retinol compared with the equivalent physical mixture in a rat model of fat malabsorption.J Nutr131 :2157 –2163,2001 .[Abstract/Free Full Text]
  16. Clark RM, Furr HC: Absorption of canthaxanthin by the rat is influenced by total lipid in the intestinal lumen.Lipids36 :473 –475,2001 .[Medline]
  17. Mokady S, Ben-Amotz A: Dietary lipid level and the availability of ß-carotene of Dunaliella-bardawil in rats.Nutr Cancer15 :47 –51,1991 .[Medline]
  18. Roodenburg AJ, Leenen R, van het Hof KH, Weststrate JA, Tijburg LB: Amount of fat in the diet affects bioavailability of lutein esters but not of alpha-carotene, beta-carotene, and vitamin E in humans.Am J Clin Nutr71 :1187 –1193,2000 .[Abstract/Free Full Text]
  19. Mueller-Cunningham WM, Quintana R, Kasim-Karakas SE: An ad libitum, very low-fat diet results in weight loss and changes in nutrient intakes in postmenopausal women.J Am Diet Assoc103 :1600 –1606,2003 .[Medline]
  20. Dimitrov NV, Meyer C, Gilliland D, Ruppenthal M, Chenoweth W, Malone W: Plasma tocopherol concentrations in response to supplemental vitamin E.Am J Clin Nutr53 :723 –729,1991 .[Abstract/Free Full Text]
  21. Djuric Z, Poore KM, Depper JB, Uhley VE, Lababidi S, Covington C, Klurfeld DM, Simon MS, Kucuk O, Heilbrun LK: Methods to increase fruit and vegetable intake with and without a decrease in fat intake: compliance and effects on body weight in the nutrition and breast health study.Nutr Cancer43 :141 –151,2002 .[Medline]
  22. Margolis SA, Schapira RM: Liquid chromatographic measurement of L-ascorbic acid and D-ascorbic acid in biological samples.J Chromatogr B Biomed Sci Appl690 :25 –33,1997 .[Medline]
  23. Holm S: A simple sequentially rejective multiple test procedure.Scand J Stat6 :65 –70,1979 .
  24. Littell RC, Milliken GA, Stroup WW, Wolfinger RD: "Statistical Analysis System (SAS) for Mixed Models." Cary, NC: SAS Institute Inc,1996 .
  25. SAS II: "SAS/STAT User’s Guide," Version 8, 8th ed. Cary, NC: SAS Institute,1999 .
  26. Weststrate JA, van het Hof KH, van den Berg H, Velthuis-te-Wierik EJ, de Graaf C, Zimmermanns NJ, Westerterp KR, Westerterp-Plantenga MS, Verboeket-van de Venne WP: A comparison of the effect of free access to reduced fat products or their full fat equivalents on food intake, body weight, blood lipids and fat-soluble antioxidants levels and haemostasis variables.Eur J Clin Nutr52 :389 –395,1998 .[Medline]
  27. Swinburn BA, Woollard GA, Chang EC, Wilson MR: Effects of reduced-fat diets consumed ad libitum on intake of nutrients, particularly antioxidant vitamins.J Am Diet Assoc99 :1400 –1405,1999 .[Medline]
  28. Campbell S, Stone W, Whaley S, Krishnan K: Development of gamma (gamma)-tocopherol as a colorectal cancer chemopreventive agent.Crit Rev Oncol Hematol47 :249 –259,2003 .[Medline]
  29. Hensley K, Benaksas EJ, Bolli R, Comp P, Grammas P, Hamdheydari L, Mou S, Pye QN, Stoddard MF, Wallis G, Williamson KS, West M, Wechter WJ, Floyd RA: New perspectives on vitamin E: gamma-tocopherol and carboxyelthylhydroxychroman metabolites in biology and medicine.Free Radic Biol Med36 :1 –15,2004 .[Medline]
  30. McEligot AJ, Rock CL, Shanks TG, Flatt SW, Newman V, Faerber S, Pierce JP: Comparison of serum carotenoid responses between women consuming vegetable juice and women consuming raw or cooked vegetables.Cancer Epidemiol Biomarkers Prev8 :227 –231,1999 .[Abstract/Free Full Text]
  31. Mela DJ, Aaron JI: Honest but invalid: what subjects say about recording their food intake.J Am Diet Assoc97 :791 –793,1997 .[Medline]
  32. Pollard J, Wild CP, White KL, Greenwood DC, Cade JE, Kirk SF: Comparison of plasma biomarkers with dietary assessment methods for fruit and vegetable intake.Eur J Clin Nutr57 :988 –998,2003 .[Medline]
  33. Michaud DS, Giovannucci EL, Ascherio A, Rimm EB, Forman MR, Sampson L, Willett WC: Associations of plasma carotenoid concentrations and dietary intake of specific carotenoids in samples of two prospective cohort studies using a new carotenoid database.Cancer Epidemiol Biomarkers Prev7 :283 –290,1998 .[Abstract]
  34. Bowen PE, Herbst-Espinosa SM, Hussain EA, Stacewicz-Sapuntzakis M: Esterification does not impair lutein bioavailability in humans.J Nutr132 :3668 –3673,2002 .[Abstract/Free Full Text]
  35. Rock CL, Thornquist MD, Neuhouser ML, Kristal AR, Neumark-Sztainer D, Cooper DA, Patterson RE, Cheskin LJ: Diet and lifestyle correlates of lutein in the blood and diet.J Nutr132 :525S –530S,2002 .[Abstract/Free Full Text]
  36. Al-Delaimy WK, van Kappel AL, Ferrari P, Slimani N, Steghens JP, Bingham S, Johansson I, Wallstrom P, Overvad K, Tjonneland A, Key TJ, Welch AA, Bueno-de-Mesquita HB, Peeters PH, Boeing H, Linseisen J, Clavel-Chapelon F, Guibout C, Navarro C, Quiros JR, Palli D, Celentano E, Trichopoulou A, Benetou V, Kaaks R, Riboli E: Plasma levels of six carotenoids in nine European countries: report from the European Prospective Investigation into Cancer and Nutrition (EPIC).Public Health Nutr7 :713 –722,2004 .[Medline]
  37. Djuric Z, Poore KM, Depper JB, Uhley VE, Lababidi S, Covington C, Klurfeld DM, Simon MS, Kucuk O, Heilbrun LK: Methods to increase fruit and vegetable intake with and without a decrease in fat intake: compliance and effects on body weight in the Nutrition and Breast Health Study.Nutr Cancer43 :141 –151,2002 .[Medline]
  38. Gartner C, Stahl W, Sies H: Lycopene is more bioavailable from tomato paste than from fresh tomatoes.Amer J Clin Nutr66 :116 –122,1997 .[Abstract/Free Full Text]
  39. Bowen P, Chen L, Stacewicz-Sapuntzakis M, Duncan C, Sharifi R, Ghosh L, Kim HS, Christov-Tzelkov K, van Breemen R: Tomato sauce supplementation and prostate cancer: lycopene accumulation and modulation of biomarkers of carcinogenesis.Exp Biol Med (Maywood)227 :886 –893,2002 .[Abstract/Free Full Text]
  40. Woodall AA, Lee SW, Weesie RJ, Jackson MJ, Britton G: Oxidation of carotenoids by free radicals: relationship between structure and reactivity.Biochimica et Biophysica Acta1336 :33 –42,1997 .[Medline]
  41. Chen G, Djuric Z: Carotenoids are degraded by free radicals but do not affect lipid peroxidation in unilamellar liposomes under different oxygen tensions.FEBS Letters505 :151 –154,2001 .[Medline]



This article has been cited by other articles:


Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
Z. Djuric, G. Chen, J. Ren, R. Venkatramanamoorthy, C. Y. Covington, O. Kucuk, and L. K. Heilbrun
Effects of High Fruit-Vegetable and/or Low-Fat Intervention on Breast Nipple Aspirate Fluid Micronutrient Levels
Cancer Epidemiol. Biomarkers Prev., July 1, 2007; 16(7): 1393 - 1399.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Djuric, Z.
Right arrow Articles by Heilbrun, L. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Djuric, Z.
Right arrow Articles by Heilbrun, L. K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS