Articles |
the School of Dietetics and Human Nutrition, McGill University, Montreal, Canada (P.J.H.J., Z.-C.L.), and the Division of Endocrinology, Metabolism, and Clinical Nutrition, Oregon Health Sciences Center University, Portland (A.S.O., L.H., D.R.I., W.E.C.).
| Abstract |
|---|
|
|
|---|
Key Words: cholesterol synthesis deuterium mevalonic acid dietary cholesterol lipoproteins
| Introduction |
|---|
|
|
|---|
Although de novo synthesis accounts for most of the plasma pool of cholesterol in lipoproteins, the response of cholesterol biosynthesis to different levels of dietary cholesterol within the physiological range of intake has not been fully characterized in humans. The depression of cholesterol synthesis by dietary cholesterol that has been observed in various species1 2 20 21 may not be representative of human cholesterol metabolism. In fact, in humans this question remains to be fully answered: dietary cholesterol at various levels has been shown to either modify6 7 or have no effect22 23 24 on synthesis. Moreover, whether the responsiveness of cholesterol synthesis to the addition of modest amounts of dietary cholesterol is influenced by the plasma concentration of TC has not been established. Decreased synthesis has been previously suggested in hypercholesterolemic subjects.25
Disparity between the results of previous studies on the responsiveness of cholesterol synthesis to changes in dietary cholesterol intake in humans may be related to a number of factors, including the amount of dietary cholesterol or fat included in the diet and the sensitivities of the different methodologies employed. Earlier cholesterol intake-balance methods for measuring synthesis depended on comprehensive stool collections and adherence to fixed diets over extended periods.7 22 23 Isotopic kinetic-decay analyses have required prolonged measurement periods.26 27 More recent methods, including determination of deuterium incorporation into cholesterol24 28 29 and mevalonic acid levels in plasma30 31 32 or urine,33 offer the advantage of short-term, noninvasive measurement. However, these techniques have not been systematically compared.
The primary objective of the present study was to examine the response of cholesterol biosynthesis, as measured by deuterium incorporation and urinary mevalonic acid excretion, to modest changes in dietary cholesterol in subjects with low, normal, and elevated plasma cholesterol levels. A second aim was to characterize the associations between and within the two methods of measurement. The null hypothesis tested was that changes in dietary cholesterol intake within the physiological range do not influence rates of cholesterogenesis in a manner that is dependent on initial cholesterol levels.
| Methods |
|---|
|
|
|---|
Protocols and Diets
The studies were conducted on an outpatient basis in the CRC at OHSU. All subjects consumed each of three diets prepared in the CRC metabolic kitchen for 4 weeks; each diet period was separated by a 4-week washout period during which the subjects consumed their typical diets. All three diets were identical except for cholesterol content, which was provided at a level of 50, 350, or 650 mg/2800 kcal for low-, medium-, or high-cholesterol diets, respectively. Energy allowances as determined by each subject's anthropometric characteristics were adjusted to maintain body weight throughout the individual study phases. The diets were designed to contain 15%, 55%, and 30% protein, carbohydrate, and fat, respectively. Dietary fat was calculated to have a fatty acid composition of 12% monounsaturated, 8% polyunsaturated, and 10% saturated fatty acids. Diet composition was calculated from food table values and by direct laboratory analysis of the fatty acid and cholesterol contents of major fat and cholesterol sources. A 7-day rotating menu composed of weighed, whole foods was used, and the order of dietary periods was systematically randomized between subjects. Dietary cholesterol was provided as egg yolk and whole egg incorporated into baked or cooked products such as custards, omelettes, and cookies. On lower-cholesterol diets, egg yolk was partially or completely replaced with egg whites and a mixture of fats similar in fatty acid composition to the fat in egg yolk. Participants typically ate their breakfast at the CRC on Monday through Saturday each week. Lunch, dinner, and all Sunday meals were prepacked for consumption at work or home. During the final week of each diet period, blood was drawn after an overnight fast on three successive mornings for determination of plasma TC levels. Toward the end of the final week, deuterium incorporation into erythrocyte cholesterol was measured for 24 hours after oral administration of 0.7 g 99.8 atom% excess D2O at 8 AM. During this 24-hour period, subjects were provided with drinking water that contained 1.2 g D2O per kilogram water to maintain body-water deuterium enrichment at plateau levels during the incorporation interval. Blood samples were obtained before and 12 and 24 hours after D2O dosing. Urine was collected during two consecutive 24-hour periods for determination of mevalonic acid levels. Plasma TC levels were determined fluorometrically34 on an autoanalyzer (Technicon Instruments).
Deuterium Incorporation Methodology
Incorporation of deuterium into erythrocyte free cholesterol was determined by sequential extraction, isolation, combustion, and reduction as previously described.24 29 Erythrocyte lipids were extracted, dried, and chromatographed on a thin-layer silica system to isolate free cholesterol. The cholesterol was then eluted from silica scrapings and quantitatively transferred to Pyrex combustion tubes containing CuO and silver wire. The tubes were flame-sealed and cholesterol was completely combusted at 520°C to CO2 and water. Combustion water was cryogenically separated from CO2 by vacuum distillation and reduced to H2 over 50 mg hot zinc. Deuterium enrichment of the resultant gas was measured by isotope ratio mass spectrometry (VG Micromass 903D). Plasma-water deuterium enrichment was measured similarly, after dilution of plasma samples taken at 12 and 24 hours with water of known isotopic abundance to bring the enrichment into the working range of the standards.24 29
Deuterium enrichment values in erythrocyte cholesterol over 24 hours were expressed relative to the mean enrichment at 12 and 24 hours of corresponding plasma water samples, after correction for the deuterium-protium ratio in cholesterol, to yield FSRs (in pools per day) for the free cholesterol pool. The FSR index represents that fraction of the free portion of the rapid-turnover cholesterol pool that is synthesized in 24 hours29 as per the formula:
![]() |
refers to deuterium enrichment above baseline level over 24 hours.
Urinary Mevalonic Acid Determinations
Mevalonic acid concentrations in urine samples from 24-hour collections were determined enzymatically with the radioenzymatic technique of Popjack et al,30 which assesses the phosphorylation of mevalonic acid with [32P]ATP and mevalonate kinase to 5-[32P]phosphomevalonate. The thoroughness of 24-hour urine collections was verified by threshold creatinine determinations. Results for the two consecutive 24-hour periods were averaged for each subject and reported in micromoles per day and micromoles per day.
Statistical Analyses
Between-group differences in lipid levels and synthesis indices were identified by two-factor ANOVA procedures with tests for interactions. Subsequent pairwise post hoc tests were used to determine individual group differences. For examination of possible relationships between variables, nontransformed linear regression analyses were conducted. A level of significance of P<.05 was used.
| Results |
|---|
|
|
|---|
|
Effects of diet and cholesterol classification were observed on plasma lipid levels. Plasma TC levels increased with each level of added cholesterol (P<.001). From the lowest to the highest dietary cholesterol level, the mean increase in plasma TC levels were 8.3% (hypocholesterolemic), 8.7% (normocholesterolemic), and 7.1% (hypercholesterolemic). TC levels differed significantly (P<.0001) across subject classifications.
Effect of Dietary and Plasma Cholesterol on Determinants of Cholesterol Synthesis
There was no systematic change in total body-water deuterium enrichment during either 12- or 24-hour intervals. The influence of dietary cholesterol level and plasma cholesterol classification on cholesterogenesis, measured as deuterium incorporation into erythrocyte free cholesterol and mevalonic acid excretion, is illustrated in Fig 1
, with individual data in Table 2
. By the deuterium incorporation method, mean cholesterol synthesis rates in hypocholesterolemic subjects on low-, medium-, and high-cholesterol diets were 0.0711±0.0153, 0.0585±0.0101, and 0.0483±0.0070 pool/d (mean±SEM), respectively. In normocholesterolemic subjects, respective mean synthesis rates were 0.0756±0.0095, 0.0659±0.0060, and 0.0634±0.0104 pool/d and in hypercholesterolemics 0.0628±0.0084, 0.0507±0.0082, and 0.0481±0.0082 pool/d. In all subjects dietary cholesterol levels depressed (P<.05) deuterium incorporation across low- to medium- and low- to high-cholesterol diets but not across medium- to high-cholesterol diets (Table 3
). There was no significant influence of cholesterol category of the subjects on deuterium incorporation into cholesterol. A nonsignificant trend toward higher deuterium incorporation in hypocholesterolemics was observed. No interaction between group and diet-lowering effect on FSR was observed.
|
|
|
For urinary mevalonic acid excretion, hypocholesterolemic individuals on low-, medium-, and high-cholesterol diets excreted 2.325±0.139, 2.021±0.123, and 1.944±0.192 µmol/d, respectively, whereas normocholesterolemic subjects excreted 2.476±0.280, 2.069±0.222, and 1.829±0.189 µmol/d and hypercholesterolemic subjects 1.881±0.118, 1.979±0.272, and 1.573±0.191 µmol/d (Fig 1
and Table 2
). Dietary cholesterol intake influenced urinary mevalonic acid excretion in all subjects. However, the effect occurred only when subjects switched from medium- to high-cholesterol diets (P<.05) and low- to high-cholesterol diets (P<.01) (Table 3
). Differences (P<.05) in mevalonic acid excretion were also observed in the normolipidemic group alone across low- to medium- and low- to high-cholesterol diets but not across medium- to high-cholesterol diets (Table 3
). As with deuterium incorporation, there was no effect of subject group on mevalonate excretion, nor was there any interaction between group and dietary cholesterol effects.
Relationships Between Indices of Cholesterogenesis
Cross-comparison of deuterium incorporation and mevalonic acid excretion by regression analysis revealed no significant associations in individual subgroups or diets or in all subjects and diet phases combined. Associations were also insignificant for comparisons between the degree of change from low- to medium- (P=.200), medium- to high- (P=.559), or low- to high- (P=.168) cholesterol diets in all subjects.
Additional comparisons were performed to examine whether the extent of responsiveness of each synthesis index could be related to the level of synthesis at any diet phase. The degree of change in each synthesis index between low- to medium- and low- to high-cholesterol diets was found to be negatively associated with the absolute level of that indicator during the low-cholesterol diet (Fig 2
). This relation was also found for deuterium uptake between the change in synthesis during medium- and high-cholesterol diets and the absolute synthesis level measured during the medium- (P<.001) and high- (P<.001) cholesterol diets. The change in cholesterol synthesis between low- and high-cholesterol diets was inversely associated (P<.001) with the rate of cholesterol synthesis on the high-cholesterol diet. No other significant associations were observed for urinary mevalonic acid excretion, nor were any significant associations found between the degree of change in each index between low- to high-cholesterol diets and TC levels (W.E.C., et al, 1996, unpublished data) at the end of the low-cholesterol diet. Urinary mevalonic acid excretion data were recalculated on the basis of weight and BMI. There was no improvement in the comparison between synthesis-measuring methods (P=.168, P=.781, and P=.689 for comparison of the change in synthesis from high- to low-cholesterol diets between FSR and mevalonic acid, calculated as micromoles per day, micromoles per day per kilogram body weight, and micromoles per day per BMI, respectively).
|
Relationships between FSR and urinary mevalonic acid excretion were also examined when the three zero-FSR values were removed from the data set. There were no significant associations between net values for synthesis or for comparisons between the degree of change from low- to medium- (P=.251), medium- to high- (P=.632), and low- to high- (P=.463) cholesterol diets in all subjects. The existing relationships held for the comparisons shown in Fig 2
. The degree of change in FSR and mevalonic acid indices of synthesis between low- to medium- (r=.509, P=.01 and r=.631, P=.001, respectively) and low- to high- (r=.642, P=.0007 and r=.671, P=.0003, respectively) cholesterol diets was found to be negatively associated with the absolute level of that indicator during the low-cholesterol diet.
| Discussion |
|---|
|
|
|---|
Resistance of synthesis rate to changes in dietary cholesterol level has also been shown. Grundy et al23 reported that body cholesterol synthesis as measured by the sterol-balance method was not significantly reduced when a large amount of cholesterol was provided to normolipidemic subjects; in these studies cholesterogenesis increased only when cholesterol absorption was suppressed by concurrent feeding with plant sterols. The authors concluded that feedback control of cholesterol synthesis was relatively unimportant in comparison with changes in absorption that occur subsequent to ingestion of higher levels of cholesterol. Kern40 similarly demonstrated that cholesterol synthesis measured by [14C]acetate incorporation into mononuclear cell cholesterol was no different in a subject who consumed 25 eggs per day than it was in normal individuals. In this "eggman," cholesterol absorption was markedly reduced and hepatic conversion of cholesterol to bile acids enhanced by his elevated cholesterol intake. Similarly, our previous work demonstrated that adding 550 mg dietary cholesterol per day to self-selected low-cholesterol diets resulted in no change in the rate of cholesterol synthesis as measured by deuterium incorporation.42 In a different study, although the type of dietary fat influenced dietary cholesterol synthesis as measured by deuterium uptake, no significant effect was observed when 120 mg cholesterol per day was added to a 1000 kcal/d diet.24 However, these two studies involved low-cholesterol diets that contained
250 mg cholesterol per day, a level that may be above the threshold effect on cholesterol synthesis. The results of the current study are thus consistent with previous studies that have used both deuterium uptake and sterol-balance techniques and that have indicated that suppression of cholesterol synthesis occurs primarily at lower levels of cholesterol intake. The magnitude of such changes in synthesis is subtle compared with responses to dietary cholesterol in several animal species,20 21 and it has been suggested that there are fundamental differences in whole-body cholesterol homeostasis in animals and humans.39 43
Zero values for FSR were observed in three cases. These values were retained in the data set because replicate agreement of deuterium enrichments at the beginning and end of the measurement interval fell within acceptable limits. In addition, although unlikely, absence of a positive deuterium enrichment has been previously observed during 24-hour periods in several subjects28 and thus cannot be ruled out as physiologically impossible.
Our studies did not disclose any differential response among three subgroups with low, normal, or elevated plasma cholesterol levels on the rates of cholesterol synthesis as measured by either technique. Similar conclusions were made by Quintao and Sperotto,35 who reported no differences in compensating mechanisms involving synthesis and absorption of cholesterol between normocholesterolemic and hypercholesterolemic individuals. These investigators, in their review of other studies, also found that whole-body cholesterol production rates were similar in these two groups in the absence of added dietary cholesterol. The modest differences in cholesterogenesis observed in the present study in response to dietary cholesterol contrast with the larger shifts generated by other dietary manipulations that have used a similar methodology.28 44 In short-term studies there was a 10-fold reduction in synthesis in response to a 24-hour fast28 and a 30% reduction when meals were provided every 4 hours instead of three times a day.44 Animal data indicate that whereas cholesterol feeding inhibits de novo cholesterol synthesis in hepatic but not extrahepatic tissues, fasting suppresses cholesterogenesis in both tissues by reducing substrate availability.20 39 The results of the present study indicate that in humans the magnitude of inhibition of cholesterol synthesis by dietary cholesterol is quantitatively less than that which occurs in response to fasting28 and is consistent with the recent suggestion that extrahepatic biosynthesis accounts for the larger portion of human whole-body cholesterol production.43
An additional objective of the present work was to study the relationships between the two indices of cholesterogenesis. Deuterium incorporation relies on several assumptions: (1) Newly synthesized cholesterol enters the larger, free-exchange cholesterol pool in which erythrocytes participate. (2) Erythrocyte cholesterol turnover is representative of the larger, exchangeable cholesterol pool in the body. (3) The deuterium-protium incorporation pattern of newly synthesized cholesterol over time is known. (4) The flux of sterol from the free-exchange pool is small during the measurement period.29 In contrast, determination of the sterol precursor mevalonic acid assumes that concentrations of this intermediate in pools such as plasma or urine reflect the overall rate of flux throughout the cholesterol biosynthetic pathways. To be accurate, the mevalonate method also requires complete urine collection. Mevalonic acid levels in plasma30 31 32 and urine33 have been demonstrated to be valid indicators of de novo synthesis when compared with the activity of hepatic hydroxymethylglutaryl coenzyme A reductase levels in rats30 or sterol-balance procedures in humans.31 32 Urinary mevalonic acid excretion offers the advantage of short-term determination of synthesis in contrast to the integrative data obtained by the sterol-balance technique. The former method is also analytically less labor-intensive compared with other approaches but has the disadvantage that concentrations of mevalonic acid do not yield quantitative data on the rates of cholesterol synthesis. The poor agreement between absolute FSRs and the values for 24-hour urinary excretion of mevalonic acid in response to differences in dietary cholesterol level may reflect the inherent limitations of each method. Previous work has shown good agreement in the ability of deuterium incorporation and plasma mevalonic acid level to identify diurnal changes in cholesterol synthesis in healthy humans.45 However, the rates of excretion of urinary mevalonic acid and deuterium incorporation have not been previously cross-compared. It appears that whereas deuterium uptake is sensitive in detecting changes in cholesterol synthesis from low- to moderate- but not moderate- to high-cholesterol diets, mevalonic acid excretion in urine may be more sensitive to shifts between levels of cholesterol in the medium to higher range of normal intakes. The possibility cannot be ruled out that incomplete urine collection may account for some of the lack of agreement between the two methods.
The relationship between the magnitude of change in each index of cholesterol synthesis when subjects changed from low- to medium-cholesterol diets and the net level of cholesterol synthesis on the low-cholesterol diet indicates that the higher the index of cholesterol synthesis on the low-cholesterol diet (when synthesis is highest), the greater the degree of downregulation in response to an increase in the cholesterol level of the diet. This effect was more pronounced by the deuterium incorporation method. However, no relationship was seen between the degree of inhibition of cholesterol synthesis in response to additional dietary cholesterol and the extent to which the plasma TC level was modified. Lack of an association between the mean responsiveness of plasma TC to dietary cholesterol and the change in whole-body synthesis that occurs during the transition from a low- to a high-cholesterol diet was also seen by Katan and Beynen.46 These workers, however, noted that responsiveness was negatively correlated with sterol balance on the low- and high-cholesterol diets and that, as we observed in the present study, cholesterol synthesis was depressed with higher levels of dietary cholesterol.46 Our data indicate that individuals with relatively low rates of cholesterogenesis on a low-cholesterol diet do not show a further reduction in endogenous synthesis in response to an increased intake of dietary cholesterol; in contrast, endogenous cholesterol biosynthesis is reduced under similar dietary conditions in those subjects with inherently higher basal rates of cholesterol biosynthesis. These results were found whether the zero-synthesis data points were used or not.
In summary, our results indicate that although considerable variation can be observed across subjects, changes in dietary cholesterol intake within the physiological range modestly influence cholesterol synthesis, as assessed by two independent indices, in subjects with inherently low, normal, and high levels of plasma TC. These findings support the view that dietary cholesterol causes significant but minimal feedback inhibition of cholesterol biosynthesis in humans.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
| Footnotes |
|---|
Received August 17, 1995;
revision received March 18, 1996;
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. A. Demmers, P. J. H. Jones, Y. Wang, S. Krug, V. Creutzinger, and J. E. Heubi Effects of Early Cholesterol Intake on Cholesterol Biosynthesis and Plasma Lipids Among Infants Until 18 Months of Age Pediatrics, June 1, 2005; 115(6): 1594 - 1601. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. A. Woollett, D. D. Buckley, L. Yao, P. J. H. Jones, N. A. Granholm, E. A. Tolley, and J. E. Heubi Effect of ursodeoxycholic acid on cholesterol absorption and metabolism in humans J. Lipid Res., May 1, 2003; 44(5): 935 - 942. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. S. Pappu, R. D. Steiner, S. L. Connor, D. P. Flavell, D. S. Lin, L. Hatcher, D. R. Illingworth, and W. E. Connor Feedback inhibition of the cholesterol biosynthetic pathway in patients with Smith-Lemli-Opitz syndrome as demonstrated by urinary mevalonate excretion J. Lipid Res., October 1, 2002; 43(10): 1661 - 1669. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Souidi, M. Combettes-Souverain, F. Milliat, E. R. Eckhardt, O. Audas, S. Dubrac, M. Parquet, J. Férézou, and C. Lutton Hamsters Predisposed to Sucrose-Induced Cholesterol Gallstones (LPN Strain) Are More Resistant to Excess Dietary Cholesterol than Hamsters That Are Not Sensitive to Cholelithiasis Induction J. Nutr., June 1, 2001; 131(6): 1803 - 1811. [Abstract] [Full Text] |
||||
![]() |
D. J. McNamara The Impact of Egg Limitations on Coronary Heart Disease Risk: Do the Numbers Add Up? J. Am. Coll. Nutr., October 1, 2000; 19(90005): 540S - 548. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. J. H. Jones, L. M. Ausman, D. H. Croll, J. Y. Feng, E. A. Schaefer, and A. H. Lichtenstein Validation of deuterium incorporation against sterol balance for measurement of human cholesterol biosynthesis J. Lipid Res., May 1, 1998; 39(5): 1111 - 1117. [Abstract] [Full Text] |
||||
![]() |
T. J. Howell, D. E. MacDougall, and P. J. H. Jones Phytosterols partially explain differences in cholesterol metabolism caused by corn or olive oil feeding J. Lipid Res., April 1, 1998; 39(4): 892 - 900. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |