ApoE Genotype Does Not Predict Lipid Response to Changes in Dietary Saturated Fatty Acids in a Heterogeneous Normolipidemic Population
Abstract Recent studies have suggested that variations in apoE genotypes may influence the magnitude of plasma lipid changes in response to dietary interventions. We examined the ability of apoE genotype to predict plasma lipid response to reductions in percent of calories from total fat (TF) and saturated fat (SF) in a normolipidemic study population (n=103) heterogeneous with respect to age, gender, race, and menopausal status. Three diets, an average American diet (34.3% TF, 15.0% SF), an AHA Step 1 diet (28.6% TF, 9.0% SF), and a low saturated fat (Low-Sat) diet (25.3% TF, 6.1% SF) were each fed for a period of 8 weeks in a three-way crossover design. Cholesterol was kept constant at 275 mg/d; monounsaturated and polyunsaturated fat were kept constant at approximately 13% and 6.5% of calories, respectively. Fasting lipid levels were measured during each of the final 4 weeks of each diet period. Participants were grouped by apoE genotype: E2 (E2/2, E2/3, E2/4); E3 (E3/3); E4 (E3/4, E4/4). Relative to the average American diet, both the Step 1 and Low-Sat diets significantly reduced total cholesterol, LDL cholesterol, and HDL cholesterol in all three apoE genotype groups. No evidence of a significant diet by genotype interaction, however, could be identified for any of the measured lipid and lipoprotein end points. Additional analysis of the data within individual population subgroups (men and women, blacks and whites) likewise provided no evidence of a significant diet by genotype interaction. Thus, in a heterogeneous, normolipidemic study population, apoE genotype does not predict the magnitude of lipid response to reductions in dietary saturated fat.
- Received April 10, 1997.
- Accepted July 15, 1997.
Replacing dietary saturated fat with carbohydrate decreases total cholesterol, LDL cholesterol, and HDL cholesterol and increases TGs.1 However, even in well-controlled diet studies, there exists considerable heterogeneity in lipid response among subjects given identical dietary treatments. Through repeated challenges with dietary cholesterol and/or various dietary fatty acids, the existence of individuals who are consistently hyporesponders or consistently hyperresponders has been established.2 3 4 Gender,5 6 7 8 9 age,8 10 BMI,8 9 11 12 prior dietary history,2 and lipoprotein composition and concentration8 12 13 have all been suggested to influence the magnitude of lipid and lipoprotein response to various dietary challenges. Additional studies have focused on the influence of genetic factors. Known polymorphisms in genes encoding for apolipoproteins and lipid transport proteins have been examined in this regard, with the common variants of apoE having been most extensively studied.5 6 9 10 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
The apoE gene is a likely candidate for the study of diet-gene interactions because this apolipoprotein plays a pivotal role in lipid and lipoprotein metabolism. ApoE is a structural component of both chylomicron and VLDL remnants and is thought to mediate their binding and uptake by both the LDL receptor and the LDL receptor-related protein.32 33 Three common genetic variants for apoE have been described: ε2 (112cys, 158cys); ε3 (112cys, 158arg); and ε4 (112arg, 158arg).34 From cross-sectional studies, the resulting six genotypes may account for 7% of the population variance in total and LDL cholesterol levels.35 Relative to the common ε3 allele, normal individuals carrying the ε2 allele have lower LDL cholesterol levels, while those carrying the ε4 allele have higher LDL cholesterol levels.36 Differences in LDL receptor affinity,37 distribution of apoE across lipoprotein subclasses,38 absorption of dietary cholesterol,39 and/or cholesterol synthesis31 40 associated with each of the alleles may underlie the observed differences in plasma and LDL cholesterol levels.
Numerous studies have also shown that apoE genotype can influence the magnitude of total and LDL cholesterol response to changes in dietary cholesterol alone,26 28 changes in dietary fat composition alone,25 or combined changes in dietary cholesterol and dietary fat composition.6 9 14 15 16 31 In these studies, carriers of the ε4 allele demonstrated a greater lipid response to dietary changes than individuals not possessing the ε4 allele. However, an almost equal number of studies5 17 18 19 20 21 22 23 24 27 29 30 have failed to find any association between apoE genotype and lipid response.
The DELTA study (Dietary Effects on Lipoproteins and Thrombogenic Activity) is the first well-controlled diet study to use a multicenter approach to examine the impact of changes in dietary fat on risk factors for atherosclerotic cardiovascular disease. A key feature of the DELTA study is that the relatively large study population allows for meaningful comparisons of subpopulation diet responses. In the present report, we determined the lipid response to isocaloric replacement of dietary saturated fat with carbohydrate at constant dietary cholesterol levels in individuals with different apoE genotypes.
Detailed descriptions of the DELTA study design and main dietary effects appear in a separate article.41 Briefly, four research centers (Columbia University, Pennington Biomedical Research Center, Pennsylvania State University, and the University of Minnesota) each enrolled 25 to 30 participants between the ages of 22 and 65. Participants were eligible if (1) their plasma total cholesterol was between the 25th and 90th percentile; their plasma TGs were below the 90th percentile; and their plasma HDL cholesterol was above the 10th percentile, all adjusted for age, gender, and race, using NHANES III data (unpublished data provided courtesy of the National Center for Health Statistics); (2) they were in good health; (3) they were not taking medications known to affect lipids or thrombogenic factors; and (4) they did not engage in extreme levels of exercise (eg, >7 h/wk). Recruitment goals were aimed at achieving a final study population that was comprised of 60% females, with an equal number of premenopausal and postmenopausal females; 30% blacks; and an equal number of males above and below 40 years of age. All participants provided written informed consent. The experimental protocol was approved by each research center’s institutional review board.
Diets and Protocol
Three diets, differing in total and saturated fat, were studied: an AAD that provided 34.3% of calories from fat with 15.0% saturated, 12.8% monounsaturated, and 6.5% polyunsaturated; an AHA Step 1 diet providing 28.6% of calories from fat with 9.0% saturated, 12.9% monounsaturated, and 6.7% polyunsaturated; and a Low-Sat diet providing 25.3% of calories as fat, with 6.1% saturated, 12.4% monounsaturated, and 6.7% polyunsaturated. All of the diets provided approximately 15% of calories as protein with the remainder of the diet as carbohydrate. Dietary cholesterol averaged 275 mg/d in all three diets.
The three diets were provided to the individual subjects in a randomized, double-blind, three-way crossover design, with each diet period lasting 8 weeks. Breaks of 4 to 6 weeks were provided between each diet period. With the exception of an optionally self-selected Saturday evening meal that was required to meet the National Cholesterol Education Program Step 1 dietary recommendations,42 all diets consumed by the subjects were prepared by the research centers. Subjects were required to eat two meals a day on site during the weekdays. The remaining weekday meal, snacks, and weekend meals were packaged for consumption off site. The design, validation, preparation, presentation, and monitoring of these diets are described in detail elsewhere.43
Twelve-hour fasting blood samples were obtained once per week during the last 4 weeks of each diet period. Standardized blood sampling and processing procedures were validated and used at all four research centers. Plasma and serum were isolated by centrifugation for 30 000g×min immediately after collection. Aliquots of each were stored in cryovials at −80°C until the end of the study, when all samples were assayed. Buffy coats were removed for DNA analysis and separately frozen in cryovials at −80°C.
Each research center determined serum concentrations of total cholesterol, LDL cholesterol, HDL cholesterol, and TGs using enzymatic assays. HDL cholesterol was determined after precipitation of apoB-containing lipoproteins with dextran-sulfate (molecular weight 50 000). All laboratories participated in a special lipid standardization protocol administered by the Centers for Disease Control (CDC). Briefly, pooled serums for total cholesterol and HDL cholesterol were sent to each research center by CDC; results were analyzed by CDC. The within-laboratory coefficients of variation were <1.9% for cholesterol and <2.5% for HDL cholesterol. The interlaboratory coefficients of variation were <2.8% for cholesterol and <6.1% for HDL cholesterol. The demonstrated precision and accuracy of each research center laboratory were adequate to allow the data to be combined.
ApoA-I and apoB were determined by rate immunonephelometry (Beckman Array). The coefficients of variation of the apolipoprotein assays were less than 6%. ApoE genotype was determined by polymerase chain reaction amplification of leukocyte DNA followed by Hha I digestion and product characterization essentially as described by Hixson and Vernier.44 All apolipoprotein assays and apoE genotypings were conducted at a single laboratory (Mary Imogene Bassett Research Institute, Cooperstown, NY).
Individuals were assigned to one of three apoE subgroups on the basis of their apoE genotype: E2 (E2/2, E2/3, E2/4); E3 (E3/3); or E4 (E3/4, E4/4). To compare apoE genotype subgroups with respect to responsiveness to experimental reductions in dietary saturated fat, statistical analyses were performed for each of six response variables: total cholesterol, LDL cholesterol, HDL cholesterol, TGs (natural log scale), apoB, and apoA-I. For each of these variables, the primary a priori hypotheses tested were (1) each genotype subgroup is responsive to dietary changes and (2) some of the genotype subgroups are more responsive than others.
Well-established procedures45 46 47 48 49 50 for estimation and testing were applied to take full advantage of all of the longitudinal data. The linear statistical model, the set of primary hypotheses, the strategy for controlling type I error, and the estimation procedures were all specified a priori. The mean of the conditional distribution of assay values was assumed to be a linear function of six categorical factors (number of levels shown in parentheses): apoE genotype (3), diet (3), race (2), gender-age group (4), research center (4), feeding period (3), and interaction of diet with genotype, race, gender-age group and research center. The variance of the conditional distribution of assay values was assumed to be constant across all factor levels and occasions. The correlation between any two of an individual’s assay values was assumed to be larger for same-diet pairs, smaller for different-diet pairs, but otherwise invariant. This model was represented and interpreted as a components-of-variance model, with the residual variance being the sum of three components: interindividual variance of the individuals’ overall mean levels (“subject”), interindividual variance of the individuals’ diet-specific mean levels (“diet by subject”), and intraindividual variation (“within-subject”). The necessary statistical computations for estimation and testing were performed using the mixed-model procedure of the SAS software system.45
Secondary analyses were used to determine the sensitivity of the main results to perturbations of the modeling assumptions and to address secondary research questions. For example, to address secondary questions about the roles of BMI, age, and gender, auxiliary analyses were performed. To address secondary questions about variance and covariance, several kinds of alternative covariance structures were also examined. Also, the data were examined for evidence of any departures from the modeling assumptions.
On the basis of estimates of variance components, power computations were performed to verify previous perceptions about the magnitude of statistical power available in this study and similar studies.
The race-specific distribution of apoE genotypes in the DELTA Study population, along with the predicted distribution, is shown in Table 1⇓. All six genotypes were present in our study population. Relative to the predicted genotype distribution based on published race-specific gene frequencies,36 51 our study population had fewer than predicted E2/3 genotypes and more than predicted E3/4 genotypes. However, none of the differences between predicted and observed frequencies was statistically significant. For analysis, our study contained 11 subjects in the E2 group, 57 subjects in the E3 group, and 35 subjects in the E4 group.
Baseline Characteristics and Plasma Lipids on AAD
Baseline characteristics of subjects in each genotype group are shown in Table 2⇓. Age did not differ substantially among the genotype groups. While not statistically significant, there were proportionately more males in the E3 group (49%) than in the other two groups (E2, 36%; E4, 40%) and the average BMI for the E2 group tended to be higher than for either the E3 or E4 groups. Total cholesterol, LDL cholesterol, and apoB, measured while subjects were consuming the AAD, increased across the genotype groups in the expected order: E2<E3<E4. However, none of these differences was statistically significant, even after adjustment for age, gender, and race. HDL cholesterol and apoA-I were lower in the E4 group relative to the E3 group. For apoA-I, this difference was significant (P=.006). No significant difference across genotype groups was identified for TGs.
ApoE Genotype and Diet Response
In Fig 1⇓, we show the interaction between apoE genotype groups and plasma lipid response to the lower fat diets. Data are presented as the mean difference between the AAD and the Step 1 or Low-Sat diets. Consumption of either the Step 1 diet or the Low-Sat diet resulted in statistically significant reductions (P<.001 as indicated by asterisks) in both total and LDL cholesterol levels within each apoE genotype group. For example, in response to the Step 1 diet, the E4, E3, and E2 groups experienced LDL reductions of 6%, 7%, and 11%, respectively. There was no substantial evidence that the genotype groups varied in responsiveness; the diet by genotype interaction was not significant for either total cholesterol (P=.77) or LDL cholesterol (P=.15) across all diets. While there was the appearance of a trend across the apoE genotype groups with the response to the Step 1 diet, this trend was in a direction opposite that previously reported for apoE (ie, lower relative response with apoE4 rather than greater response) and was not maintained with the Low-Sat diet.
Similar results were obtained with both the HDL cholesterol and TG responses (Fig 1⇑). With the exception of the E2 group in the Step 1–AAD comparison, HDL cholesterol fell significantly in each apoE genotype as dietary saturated fat was replaced by carbohydrate. Increases in TG levels were significant in the E3 and E4 groups in the Step 1–AAD comparison and in the E4 group in the Low-Sat–AAD comparison. As with total cholesterol and LDL cholesterol, there was no evidence of a significant effect of apoE genotype group on the magnitude of either the HDL cholesterol (P=.80) or TG (P=.54) response.
Fig 2⇓ shows the apolipoprotein responses by apoE genotype group. Reductions in saturated fat significantly reduced apoA-I levels in all apoE genotype groups with the exception of the E2 group in the Step 1–AAD comparison. As with HDL cholesterol, there was no evidence of a diet by genotype interaction (P=.14). Significant reductions in apoB in the apoE genotype groups were observed in the E3 group in the Step 1–AAD comparison and in all three genotype groups in the Low-Sat–AAD comparison. Despite the apparent large differences in apoB response between the genotype groups, they were not statistically significant (P=.44). Furthermore, like LDL cholesterol, it was the E2 group, not the E4 group, that showed the greatest apparent reductions in apoB with the Step 1 diet.
A recent study by Lopez-Miranda et al14 suggested that the apoE genotype effect on diet response may be more pronounced in males than in females. Since our study population was 55% females, the apoE genotype effect on LDL cholesterol response was examined separately in males and females (Fig 3⇓). In both groups, the LDL cholesterol response to diet was very similar when examined across apoE genotype groups. The formal test of the hypothesized diet by genotype by gender interaction was not significant (P=.75).
In addition to gender, we also examined the diet by genotype interactions as a function of race. In Fig 4⇓, results are shown for LDL cholesterol response by genotype for both blacks and nonblacks (predominantly whites). Again, for both racial groups, the LDL cholesterol response was similar across genotype, with no significant diet by apoE genotype interaction in either racial group (P=.60).
Secondary Statistical Analyses
We examined the robustness of our negative findings by considering a number of auxiliary statistical models. In previous studies, it has been suggested that age, BMI, and baseline lipid levels (ie, levels on AAD) may influence response to diet. To remove the potential confounding effects, if any, of these parameters, BMI and age and their interactions with diet were initially entered into our previous statistical model (Table 3⇓, statistical model 2). Age was predictive of total cholesterol (P=.0007), LDL cholesterol (P=.0004), and TG (P=.044) levels, independent of diet treatment, while BMI was predictive of total cholesterol (P=.018) and LDL cholesterol (P=.037). However, neither age nor BMI were predictive of the magnitude of diet response for any of the measured end points. Furthermore, inclusion of BMI, age, and their diet interaction terms in the model did not alter the negative findings regarding apoE genotype and diet response.
Inclusion of AAD lipid values as covariates in the model (Table 3⇑, statistical model 3) revealed that AAD total cholesterol levels were predictive of total cholesterol response (P=.0002); AAD LDL cholesterol levels were predictive of LDL cholesterol response (P=.012); AAD HDL cholesterol levels were predictive of HDL cholesterol response (P<.0001); and AAD ln(TG) values were marginally predictive of ln(TG) response (P=.048). However, in this statistical model, apoE genotype was still not predictive of diet response for any of the lipid end points.
Our a priori model assumed that individual genotypes within the E2 (E2/2, E2/3, E2/4) and E4 groups (E3/4, E4/4) responded to diets in a similar manner. Failure of this assumption would lead to higher variation in diet response within the E2 and E4 groups and therefore might mask true genotype effects. Consequently, we examined a statistical model in which the E2, E3, and E4 groups contained only the genotypes of E2/3, E3/3, and E3/4, respectively. When the narrower definition for inclusion into the apoE genotype groups was used, the diet by genotype interaction remained not significant (Table 3⇑, statistical model 4).
Finally, in our last model, we limited our comparisons to the two most frequent genotypes, E3/3 and E3/4. This direct comparison between these two genotypes also did not demonstrate a significant diet by genotype interaction (Table 3⇑, statistical model 5).
An analysis of statistical power was performed, taking into account the research protocol used (study design, recruitment strategy, sample size, model assumptions, statistical algorithms) and the estimated values of the variance components. In comparing the genotypes with respect to LDL cholesterol responsiveness to the Low-Sat diet, the research protocol provides a 90% chance of detecting any 0.33 mmol/L difference in response among the three genotype groups. The result of this power analysis suggests that either a rare event occurred in the present study (diet by genotype interactions are very large but were not detected), or the actual differences in LDL cholesterol responsiveness among apoE genotype groups are smaller than 0.33 mmol/L. Analogous results hold for total cholesterol and other end point variables considered.
Structural polymorphisms in apoE were first recognized two decades ago,52 and the underlying genetic basis for these polymorphisms has since been identified.34 The three identified apoE alleles, ε2 (112cys, 158cys), ε3 (112cys, 158arg), and ε4 (112arg, 158arg), profoundly affect lipoprotein metabolism and consequently the fasting levels of both total and LDL cholesterol.36 It was therefore not unexpected that early work aimed at identifying the genetic basis of diet responsiveness would focus on apoE gene polymorphisms. Miettinen et al28 first reported that apoE polymorphisms affect the magnitude of total cholesterol and LDL cholesterol response to changes in dietary cholesterol levels with individuals possessing the ε2 allele displaying smaller responses than those possessing the ε4 allele. While some investigators were able to replicate these findings,26 28 others could not.19 21 27 29 30
A number of diet studies have focused on the interaction between apoE genotype and lipid response to combined changes in dietary cholesterol levels and fatty acid content and composition or to changes in fatty acid levels alone. These studies are summarized in Table 4⇓. In approximately half the reported studies, apoE genotype was shown to affect significantly the magnitude of total plasma cholesterol and LDL cholesterol response to a wide range of changes in dietary total fat, saturated fatty acids, and cholesterol. In studies showing a significant apoE interactive effect, individuals possessing the ε4 allele had the greatest response to dietary changes compared with the remaining genotypes. For LDL cholesterol, the variously defined “E4” groups experienced, on average, a 74% greater response (range, 49% to 95%) to reductions in saturated fat and cholesterol than that observed in the remaining apoE genotype groups combined.
However, in an almost equal number of studies, apoE genotype was not found to have any effect on diet responsiveness. From the summary in Table 4⇑, studies that reported significant diet by apoE genotype interactions tended to have studied predominantly normocholesterolemic subjects and tended to examine greater changes in total fat, saturated fat, and cholesterol than studies where no significant interactions were found. However, despite the differences in the magnitude of dietary changes between these studies, the average decline in LDL cholesterol observed in studies reporting a significant diet by apoE genotype interaction (0.56 mmol/L) was identical to that observed in studies where evidence for a diet by apoE genotype interaction was not found. Interestingly, of the seven studies reporting a significant apoE genotype effect on diet response, three studies were conducted with only male participants, while three others found significant effects in males but not females. In the one study where an apoE genotype effect was found in males and females,16 a later analysis of the same data20 failed to reveal a significant diet by apoE genotype interaction.
The DELTA study is the first well-controlled dietary investigation to employ a multicenter approach. The large number of subjects enrolled, coupled with the high level of dietary control and diet monitoring, provided an excellent opportunity to evaluate the impact of apoE genotype on lipid response to changes in the amounts of dietary saturated fatty acid consumed. The study population, by design, was diverse in both age and gender and included a substantial proportion of blacks. All six possible apoE genotypes were represented; however, their distribution differed slightly from predicted values. Our study population contained fewer than predicted E2/3 genotypes and more than predicted E3/4 genotypes. However, these differences were not significant and therefore were unlikely to have affected the outcome of our overall study.
In the present study, substantial evidence of a diet by apoE genotype interaction was absent for all lipid, lipoprotein, and apolipoprotein end points. In fact, the relative order of total and LDL cholesterol response to the Step 1 diet across the apoE genotype groups was opposite that previously reported in studies showing a diet by apoE genotype interaction (see Table 4⇑). In our study, individuals with the E2 genotype group had the greatest changes in total and LDL cholesterol response to both the Step 1 and Low-Sat diets relative to both the E3 and E4 groups, although these differences were small and not significant.
The magnitude of changes in dietary total fat and saturated fat in the present study may partially explain our negative findings. Changes in total fat (5.7% of calories) and saturated fat (6.0% of calories) between the AAD and the Step 1 diet are smaller than those used in published studies demonstrating an effect of apoE genotype (Table 4⇑). It is only in the comparison of the AAD with the Low-Sat diet that differences in total fat (9.0% of calories) and saturated fat (8.9% of calories) approach the range found in studies reporting significant apoE genotype effects. On the Low-Sat diet, the E4 group did experience a 23% greater reduction in LDL cholesterol relative to the E3 group; however, the magnitude of this difference was well below the reported effects of E4 on LDL cholesterol response.
Given the evidence supporting a diet by apoE genotype interaction, we were somewhat surprised by our negative results. In particular, our results stand in marked contrast to a recently published retrospective analysis of three combined well-controlled dietary studies by Lopez-Miranda et al.14 In their study, after correction for BMI and age, males with the E3/4 genotype experienced an almost twofold greater decrease in LDL cholesterol than did males with the E3/3 genotype in response to reductions in both dietary saturated fatty acids and cholesterol. However, in the same analysis, genotype effects in females were not significant. As previously indicated, the findings of a gender effect by Lopez-Miranda are consistent with other published reports of significant diet by apoE genotype interactions in studies of only men or of significant diet by apoE genotype interactions in men but not women in studies including both genders. This prompted us to analyze our data taking gender into consideration. In men, LDL cholesterol reductions with the Low-Sat diet were 46% greater in the E4 group than in the E3 group. This contrasts with the findings in women in whom the LDL cholesterol response was only 18% greater in the E4 group than the E3 group. Thus, while we observed trends in our data similar to those previously published regarding the effects of E4 in men, these trends were only evident with the Low-Sat diet, were not as great as those previously published, and did not achieve statistical significance.
Additional analyses considered the effects of age, BMI, baseline lipid levels, race, and genotype group definitions on potential diet by apoE genotype interactions. However, these additional analyses did not reconcile the differences in results of the present study with those of other published reports in that we still saw no effect of apoE genotype on the lipid and lipoprotein response to diet (Table 3⇑).
Other differences between our study and those reporting significant apoE genotype-diet interactions warrant further consideration. In the present study, fasting lipid levels of subjects on the AAD approximated the 50th percentile for middle-aged Americans. Furthermore, in our study, differences in total and LDL cholesterol levels across apoE genotype groups while consuming the AAD were not marked or significant. This contrasts with other studies whose subjects were either moderately hypercholesterolemic14 15 16 31 and/or had significant and substantial differences in baseline lipid levels between the apoE genotype groups.14 16 25 It is possible that any diet by apoE genotype interaction is only manifest in subjects who are susceptible to hypercholesterolemia as a consequence of other genetic or environmental causes.
Finally, in the present study, only saturated fatty acid levels were changed across diet. In the majority of other studies, there were concomitant changes in both dietary fatty acids and cholesterol (Table 4⇑). Only one study25 has demonstrated a moderate effect of apoE genotype on LDL cholesterol response in a design in which changes in dietary fat levels were achieved (reductions in both saturated and polyunsaturated fatty acids) without significant changes in dietary cholesterol levels. Previous reports have shown that cholesterol absorption is related to apoE genotype39 and may account for the reported differences in apoE genotype response to isolated changes in dietary cholesterol. If the primary interaction is between apoE genotype and dietary cholesterol, then the negative findings in our study would not be unexpected, since dietary cholesterol was held constant.
In conclusion, the present study has shown that every apoE genotype group responded to changes in dietary saturated fat with significant reductions in total and LDL cholesterol. We have previously reported41 similarly significant reductions in total and LDL cholesterol in various subgroups of this same study population when categorized by age, gender, menopausal status, and race. Therefore, at present, we have not identified any significant population subgroup that does not, on average, experience beneficial changes in both total and LDL cholesterol levels in response to reductions in dietary saturated fat.
Selected Abbreviations and Acronyms
|AAD||=||average American diet|
|BMI||=||body mass index|
|Low-Sat||=||low saturated fat diet|
DELTA Research Group
The investigators comprising the multicenter DELTA research group are listed below.
Henry N. Ginsberg, MD, Principal Investigator; Rajasekhar Ramakrishnan, ScD; Wahida Karmally, MS, RD; Lars Berglund, MD; Maliha Siddiqui, MS, RD; Niem-Tzu Chen, MS; Steve Holleran, BS; Colleen Johnson, RD; Roberta Holeman; Karen Chirgwin; Kellye Stennett; Lencey Ganga; Tajudeen T. Towalawi, MBA; Minnie Myers, BS; Colleen Ngai, BS; Nelson Fontenez, BS; Jeff Jones, BS; Carmen Rodriguez; Norma Useche.
Pennington Biomedical Research Center
Michael Lefevre, PhD and Paul S. Roheim, MD, Co-Principal Investigators; Donna H. Ryan, MD; Marlene M. Windhauser, PhD, RD; Catherine M. Champagne, PhD, RD; Donald Williamson, PhD, Richard Tulley, PhD; Ricky Brock, RN; Deonne Bodin, BS, MT; Betty Kennedy, MPA; Michelle Barkate, MS, RD; Elizabeth Foust, BS; Deshoin York, BS.
Pennsylvania State University
Penny Kris-Etherton, PhD, Principal Investigator; Satya S. Jonnalagadda, PhD; Janice Derr, PhD; Abir Farhat-Wood, MS; Vikkie A. Mustad, MS; Kate Meaker, MS; Edward Mills, PhD; Mary-Ann Tilley, MS, RD; Helen Smiciklas-Wright, PhD; Madeline Sigman-Grant, PhD, RD; Jean-Xavier Guinard, PhD; Pamela Sechevich, MS; C. Channa Reddy, PhD; Andrea M. Mastro, PhD; Allen Cooper, MD.
University of Minnesota
Patricia Elmer, PhD, Principal Investigator; Aaron Folsom, MD; Nancy Van Heel, MS, RD; Christine Wold, RD; Kay Fritz, MA, RD; Joanne Slavin, PhD; David Jacobs, PhD.
University of North Carolina at Chapel Hill
Barbara H. Dennis, PhD, Principal Investigator; Paul Stewart, PhD; C.E. Davis, PhD; James Hosking, PhD; Nancy Anderson, MSPH; Susan Blackwell, BS; Lynn Martin, MS; Hope Bryan, MS; W. Brian Stewart, BS; Jeffrey Abolafia, MA; Malachy Foley, BS; Conroy Zien, BA; Szu-Yun Leu, MS; Marston Youngblood, MPH; Thomas Goodwin, MAT; Monica Miles; Jennifer Wehbie.
Mary Imogene Bassett Research Institute
Tom Pearson, MD, PhD; Roberta Reed, PhD.
University of Vermont
Russell Tracy, PhD; Elaine Cornell, BS.
Virginia Polytechnic and State University
Kent K Stewart, PhD; Katherine M. Phillips, PhD.
Bernestine B. McGee, PhD; RD; Brenda Williams, BS.
Beltsville Agricultural Research Center
Gary R. Beecher, PhD; Joanne M. Holden, MS; Carol S. Davis, BS.
National Heart, Lung, and Blood Institute
Abby G Ershow, ScD; David J. Gordon, MD; Michael Proschan, PhD; Basil M. Rifkind, MD, FRCP.
This study was supported by the following grants: NHLBI grants 5-U01-HL49644, 496448, 49649, 49651, and 49659 and M01 RR 006645. The DELTA Investigators express thanks to the following contributors: AARHUS, Bertoli, Best Foods, Campbell Soup Company, Del Monte Foods, General Mills, Hershey Foods Corporation, Institute of Edible Oils and Shortenings, Kraft General Foods, Land O’Lakes, McCormick Incorporated, Nabisco Foods Group, Neomonde Baking Company, Palm Oil Research Institute, Park Corporation, Proctor and Gamble, Quaker Oats, Ross Laboratories, Swift-Armour and Eckrick, Van Den Bergh Foods, Cholestech, and Lifelines Technology, Incorporated.
Grundy SM, Denke MA. Dietary influences on serum lipids and lipoproteins. J Lipid Res. 1990;31:1149-1172.
Katan MB, Beynen AC. Characteristics of human hypo-and hyperresponders to dietary cholesterol. Am J Epidemiol. 1987;125:387-399.
Katan MB, Berns MAM, Glatz JFC, Knuiman JT, Nobels A, de Vries JHM. Congruence of individual responsiveness to dietary cholesterol and to saturated fat in humans. J Lipid Res. 1988;29:883-892.
Schaefer EJ, Lichtenstein AH, Lamon-Fava S, Contois JH, Li Z, Rasmussen H, McNamara JR, Ordovas JM. Efficacy of a National Cholesterol Education Program Step 2 diet in normolipidemic and hypercholesterolemic middle-aged and elderly men and women. Arterioscler Thromb Vasc Biol. 1995;15:1079-1085.
Clifton PM, Nestel PJ. Influence of gender, body mass index, and age on response of plasma lipids to dietary fat plus cholesterol. Arterioscler Thromb. 1992;12:955-962.
Clifton PM, Abbey M, Noakes M, Beltrame S, Rumbelow N, Nestel PJ. Body fat distribution is a determinant of the high-density lipoprotein response to dietary fat and cholesterol in women. Arterioscler Thromb Vasc Biol. 1995;15:1070-1078.
Cole TG, Bowen PE, Schmeisser D, Prewitt TE, Aye P, Langenberg P, Dolecek TA, Brace LD, Kamath S. Differential reduction of plasma cholesterol by the American Heart Association Phase 3 Diet in moderately hypercholesterolemic, premenopausal women with different body mass indexes. Am J Clin Nutr. 1992;55:385-394.
Howard BV, Hannah JS, Heiser CC, Jablonski KA, Paidi MC, Alarif L, Robbins DC, Howard WJ. Polyunsaturated fatty acids result in greater cholesterol lowering and less triacylglycerol elevation than do monounsaturated fatty acids in a dose-response comparison in a multiracial study group. Am J Clin Nutr. 1995;62:392-402.
Grundy SM, Vega GL. Plasma cholesterol responsiveness to saturated fatty acids. Am J Clin Nutr. 1988;47:822-824.
Lopez-Miranda J, Ordovas JM, Mata P, Lichtenstein AH, Clevidence B, Judd JT, Schaefer EJ. Effect of apolipoprotein E phenotype on diet-induced lowering of plasma low density lipoprotein cholesterol. J Lipid Res. 1994;35:1965-1975.
Tikkanen MJ, Huttunen JK, Ehnholm C, Pietinen P. Apolipoprotein E4 homozygosity predisposes to serum cholesterol elevation during high fat diet. Arterioscler Thromb. 1990;10:285-288.
Glatz JFC, Demacker PNM, Turner PR, Katan MB. Response of serum cholesterol to dietary cholesterol in relation to apolipoprotein E phenotype. Nutr Metab Cardiovasc Dis. 1991;1:13-17.
Cobb MM, Teitlebaum H, Risch N, Jekel J, Ostfeld A. Influence of dietary fat, apolipoprotein-E phenotype, and sex on plasma lipoprotein levels. Circulation. 1992;86:849-857.
Hunninghake DB, Stein EA, Dujovne CA, Harris WS, Feldman EB, Miller VT, Tobert JA, Laskarzewski PM, Quiter E, Held J, Taylor AM, Hopper S, Leonard SB, Brewer BK. The efficacy of intensive dietary therapy alone or combined with lovastatin in outpatients with hypercholesterolemia. N Engl J Med. 1993;328:1213-1219.
Zambón D, Ros E, Casals E, Sanllehy C, Bertomeu A, Campero I. Effect of apolipoprotein E polymorphism on the serum lipid response to a hypolipidemic diet rich in monounsaturated fatty acids in patients with hypercholesterolemia and combined hyperlipidemia. Am J Clin Nutr. 1995;61:141-148.
Dreon DM, Fernstrom HA, Miller B, Krauss RM. Apolipoprotein E isoform phenotype and LDL subclass response to a reduced-fat diet. Arterioscler Thromb Vasc Biol. 1995;15:105-111.
Martin LJ, Connelly PW, Nancoo D, Wood N, Zhang ZJ, Maguire G, Quinet E, Tall AR, Marcel YL, McPherson R. Cholesteryl ester transfer protein and high density lipoprotein responses to cholesterol feeding in men: relationship to apolipoprotein-E genotype. J Lipid Res. 1993;34:437-446.
Ginsberg HN, Karmally W, Siddiqui M, Holleran S, Tall AR, Rumsey SC, Deckelbaum RJ, Blaner WS, Ramakrishnan R. A dose-response study of the effects of dietary cholesterol on fasting and postprandial lipid and lipoprotein metabolism in healthy young men. Arterioscler Thromb. 1994;14:576-586.
Ginsberg HN, Karmally W, Siddiqui M, Holleran S, Tall AR, Blaner WS, Ramakrishnan R. Increases in dietary cholesterol are associated with modest increases in both LDL and HDL cholesterol in healthy young women. Arterioscler Thromb Vasc Biol. 1995;15:169-178.
Miettinen TA, Gylling H, Vanhanen H, Ollus A. Cholesterol absorption, elimination, and synthesis related to LDL kinetics during varying fat intake in men with different apoprotein-E phenotypes. Arterioscler Thromb. 1992;12:1044-1052.
Mahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science. 1988;240:622-630.
Rall SC Jr, Weisgraber KH, Mahley RW. Human apolipoprotein E: the complete amino acid sequence. J Biol Chem. 1982;257:4171-4178.
Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis. 1988;8:1-21.
Weisgraber KH, Innerarity TL, Mahley RW. Abnormal lipoprotein receptor-binding activity of the human apo E apoprotein due to cysteine-arginine interchange at a single site. J Biol Chem. 1982;257:2518-2521.
Gregg RE, Zech LA, Schaefer EJ, Stark D, Wilson D, Brewer HB Jr. Abnormal in vivo metabolism of apolipoprotein E4 in humans. J Clin Invest. 1986;78:815-821.
Kesaniemi YA, Ehnholm C, Miettinen TA. Intestinal cholesterol absorption efficiency in man is related to apoprotein E phenotype. J Clin Invest. 1987;80:578-581.
Ginsberg HN, Dennis B, Elmer PJ, Ershow A, Lefevre M, Pearson TA, Roheim PS, Ramakrishnan R, Reed R, Anderson N, Stewart K, Phillips K, Stewart P, Kris-Etherton PM, for the DELTA Investigators. Effects of reducing dietary saturated fatty acids on plasma lipids and lipoproteins in healthy subjects: DELTA Protocol-1. Arterioscler Thromb Vasc Biol. 1998. In press.
Grundy SM, Bilheimer D, Chait A, Clark LT, Denke M, Havel RJ, Hazzard WR, Hulley SB, Hunninghake DB, Kreisberg RA, Kris-Etherton P, McKenney JM, Newman MA, Schaefer EJ, Sobel BE, Somelofski C, Weinstein MC, Brewer HB, Cleeman JI, Donato KA. Summary of the Second Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult-Treatment-Panel-II). JAMA. 1993;269:3015-3023.
Dennis BH, Kris-Etherton PM, Stewart K, Karmally W, Van Heel N, Champagne C, Windhauser MM, Ershow A, Phillips K, Farhat-Wood A, for the DELTA Investigators. Diet design for a multicenter controlled feeding trial: the DELTA Study. J Am Diet Assoc. 1998. In press.
Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with Hha I. J Lipid Res. 1990;31:545-548.
SAS Institute Inc. SAS/STAT User’s Guide, Release 6.07. Cary, NC: SAS Institute Inc; 1992. Technical Report P-229.
Harville DA. Maximum likelihood approaches to variance component estimation and to related problems. JASA. 1977;72:320-340.
Harville DA. Mixed-model methodology: theoretical justification and future directions. In: Annual Proceedings of the Statistical Computing Section. Alexandria, Va: American Statistical Association; 1988.
Jennrich RI, Schlueter MD. Unbalanced repeated-measures models with structured covariance matrices. Biometrics. 1986;42:8805-8820.
Helms RW, Edwards L. Analysis of irregular and/or incomplete data from crossover designs using mixed models with linear covariance structure. In: Annual Proceedings of the Biopharmaceutical Section. Alexandria, Va: American Statistical Association; 1991.
Hixson JE. Apolipoprotein E polymorphisms affect atherosclerosis in young males. Arterioscler Thromb. 1991;11:1237-1244.
Gylling H, Miettinen TA. Cholesterol absorption and synthesis related to low density lipoprotein metabolism during varying cholesterol intake in men with different apoE phenotypes. J Lipid Res. 1992;33:1361-1371.