542
Effect of Dietary Fat Selection on
Plasma Cholesterol Synthesis in Older,
Moderately Hypercholesterolemic Humans
Peter J.H. Jones, Alice H. Lichtenstein, Ernst J. Schaefer, Gayle L. Namchuk
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Abstract To study factors controlling plasma cholesterol
levels, the effect of dietary fat type on cholesterol synthesis
was examined in 15 hypercholesterolemic subjects (low-density lipoprotein [LDL] cholesterol > 130 mg • dL"1) consuming
over a period of 32 days (1) a baseline diet (36% kcal as fat:
15% saturated, 15% monounsarurated, and 6% poh/unsaturated fat; 180 mg cholesterol • 1000 kcal"1) and diets meeting
National Cholesterol Education Program step 2 criteria (30%
kcal as fat, £ 7 % saturated fat, 80 to 85 mg cholesterol/Meal),
where two thirds of the fat was either (2) olive, (3) corn, or (4)
canola oil. Plasma total, LDL, and high-density lipoprotein
(HDL) cholesterol and trigh/ceride levels were determined at
the end of each period. Cholesterol fractional synthesis rate
(FSR) was also measured as the deuterium (D) incorporation
into plasma total cholesterol relative to body D2O level (1.2 g
D 2 O • kg"1 estimated body water) over 24 hours. Absolute
synthesis rates (ASRs) were determined as the product of FSR
and rapid turnover cholesterol pool size. Plasma total and
LDL cholesterol levels declined significantly (/)<.005) on all
plant-oil diets compared with the baseline diet; however,
triglyceride levels were not different. FSRs were higher
(P<.05) for the corn oil (0.0665 ±0.0097 pool • d"1) compared
with baseline (O.0412±0.0060 pool • d"1) and olive oil
(0.0409+0.0052 pool • d"1) but not canola oil (0.0492 ±0.0072
pool-d" 1 ) diets. Mean ASR for the corn oil diet (1697±271
mg • d"1) was elevated (P<.05) relative to baseline (1081 ±170
mg • d"1) and olive oil (1034±140 mg • d"1) but not canola oil
(1169+137 mg • d"') diet phases. These data suggest a more
rapid rate of cholesterol synthesis with consumption of corn oil
versus olive oil diets, indicating differential mechanisms that
control circulating cholesterol level control across plant oil
types. (ArterioscUr Thromb. 1994;14:542-548.)
Key Words • cholesterol • synthesis • deuterium •
humans • corn oil • olive oil • canola oil • fatty acids
C
etary fat on cholesterol synthesis could potentially be
attributable to factors including fatty acid composition
or the presence of plant sterols or other constituents.
The present objectives were to examine (1) the influence of the ratio of MUFA to PUFA on circulating
cholesterol levels and plasma cholesterol synthesis rates
in moderately hypercholesterolemic subjects and (2)
associations between fatty acid and plant sterol contents
of the oils consumed and plasma cholesterol synthesis
rate. Cholesterol synthesis was determined as the fractional synthesis rate (FSR) of the rapid-turnover body
cholesterol pool and as a calculated estimate of absolute
synthesis rate (ASR). Hypercholesterolemic subjects
were examined because they are the group for whom the
current dietary recommendations are targeted and are
individuals who would derive the greatest physiological
benefit from any changes in plasma lipids.
ompelling evidence links plasma cholesterol levels
to risk of human coronary heart disease.1 Both
earlier2-4 and more recent5-9 studies indicate that
cholesterol levels are raised with consumption of fats
containing saturated fatty acids (SAFAs) and reduced
with fats high in monounsaturated (MUFA) and polyunsaturated fatty acid (PUFA). Therefore, the type of
dietary fat selected may influence heart disease risk.
Despite this knowledge, the mechanism by which
dietary fats influence circulating cholesterol levels remains to be completely understood. In animals, studies
comparing the effects of various dietary fats on both the
removal of cholesterol from the circulating compartment 1011 and cholesterogenesis1214 suggest higher rates
with PUFA consumption. In humans, although consumption of SAFA versus PUFA fat does not influence
cholesterol synthesis as measured by sterol balance15
and mononuclear leukocytes,16 enhanced fecal sterol
excretion has been reported with PUFA feeding,1719
with resultant negative sterol balance.19 The only previous systematic study in humans has shown no effect of
plant oils on cholesterol synthesis.20 Any effect of diReceived September 3, 1993; revision accepted January 12,
1994.
From the Division of Human Nutrition, University of British
Columbia, Vancouver (PJ.HJ., G.L.N.), and the Lipid Metabolism Laboratory, US Department of Agriculture, Human Nutrition
Research Center on Aging at Tufts University, Boston, Mass
(A.H.L., EJ.S.).
Reprint requests to Peter J.H. Jones, PhD, Division of Human
Nutrition, 2205 East Mall, University of British Columbia, Vancouver, BC V6T 1W5, Canada.
Methods
Subjects
Fifteen healthy volunteers with low-density lipoprotein
(LDL) cholesterol levels >130 mg-dL" 1 were screened for
chronic illness including hepatic, renal, and cardiac dysfunction before admission to the study. Subjects were nonsmokers
and were not taking lipid-lowering drugs, 0-blockers, diuretics,
or hormones. All women were postmenopausal. Protocols
were approved by the Human Investigation Review Committee of New England Medical Center and Tufts University.
Protocol
Subjects consumed four experimental diets composed of
solid foods typical of North American intakes. Each diet was
consumed for a 32-day period, separated by a washout
Jones et al
TABLE 1.
Cholesterol Synthesis Response to Dietary Oil Selection
543
Fatty Acid Composition and Nonsaponlflable Upld Contents of Dietary Oils
Olive Oil
Com Oil
Canola Oil
14:0
Trace
Trace
Trace
16:0
13.3
10.6
4.3
16:1 n7
1.9
Trace
0.3
18:0
1.9
1.0
2.0
18:1 n9
70.8
27.6
62.3
18:2n6
11.2
60.2
19.8
18:3n3
1.0
1.1
9.7
Fatty acid, % composition
20:0
Trace
Trace
20:1 n9
Trace
Trace
0.80
5.28
PUFA/SAFA
0.5
1.5
4.68
Nonsaponifiable lipids: plant sterols, mg/100 g
Sitosterol
107
655
348
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Campesterol
ND
121
152
Stigmasterol
ND
29
3
Others
107*
Total
215
ND
805
63
CCfi
PUFA/SAFA indicates polyunsaturated to saturated fatty acid ratio; ND, not detected.
•Contained 40% to 50% stigmastanol.
interval of 7 to 14 days. Subjects first consumed, over a
period of 32 days, a baseline diet that contained 16.5% and
35.4% of calories from protein and fat, respectively, and a
cholesterol content of 128 mg • 1000 kcar 1 . The fat contained 15% SAFA, 15% MUFA, and 6% PUFA. Subjects
then consumed experimental diets consistent with National
Cholesterol Education Program guidelines21 containing 30%
of kcal as fat, with two thirds of dietary fat derived from
olive, corn, or canola oil and 80 to 85 mg • 1000 kcal"1
cholesterol. These three diets were provided in randomized
order according to a double-blinded study design as part of a
larger study involving other dietary treatments not included
in this report. Foods and beverages containing the targeted
energy, cholesterol, and fatty acid contents were provided by
the Metabolic Research Unit of the US Department of
Agriculture Human Nutrition Center on Aging for consumption on site or packaged for take-out. Subjects reported to
the unit on at least three occasions per week. Energy intakes
of subjects were tailored to individual requirements, as
verified by constant body weight. Adjustments to energy
intakes were made only during the initial 10-day period.
PUFA to SAFA ratios for baseline and olive, corn, and
canola oil diets were 0.616, 0.558, 1.625, and 1.239, respectively. A detailed description of diets has been published
previously.22 Fatty acid composition and nonsaponifiable
plant sterol contents of the dietary oils are shown in Table 1.
During the final week of each dietary phase, fasting blood
samples were obtained for lipid level determinations and
deuterium (D) incorporation into cholesterol. D incorporation
was measured after oral bolus administration of 1.2 g deuterium oxide (D2O) per kilogram body water, estimated as 60%
of body weight, at about 8 AM. Blood samples were collected
just before and 24 hours after dosing for plasma total cholesterol and water D enrichment measurement.
Plasma Lipid Levels
Fasting blood samples from week 5 were collected in tubes
containing EDTA (0.1%). Plasma was separated and assayed
for total, LDL, and high-density lipoprotein (HDL) choles-
terol and triglyceride levels by enzymatic procedures previously described.22
Cholesterol Synthesis Determinations
Lipids were extracted from 2 to 3 mL plasma in duplicate.
Extracts were saponified, dissolved in chloroform, and separated on thin-layer chromatography (TLC) silica gel plates
(20x20 cm, 250 mm, Whatman Inc) as described.17 Plates were
developed in petroleum ether/diethyl ether/acetic acid
(135:15:1.5, vol/vol/vol) for 60 minutes and air-dried. Lipid
fractions were visualized in iodine vapor; free cholesterol was
removed from the plate and eluted from the silica by use of
hexane/chloroform/diethyl ether (5:2:1, vol/vol/vol). Extracts
containing approximately 2 mg cholesterol were transferred to
Pyrex (Corning Glass Works) combustion tubes (12 cmx6
mm) containing 0.5 g ground cupric oxide and a 2.5-cm length
of 1-mm silver wire. Chloroform was removed, and tubes were
sealed under vacuum. Tubes were placed in an oven at 520°C
for 4 hours to combust the cholesterol. Combustion product
water was transferred by vacuum distillation into a second
Pyrex tube containing 60 mg of zinc reagent (Biogeochemical
Laboratories).23
To measure plasma water D enrichment, plasma samples were
diluted sevenfold to reduce the D enrichment to within the range
of working standards. Baseline plasma samples were not diluted.
Plasma samples were distilled into Pyrex tubes containing zinc
and sealed under vacuum. Water from cholesterol and plasma
samples was reduced at 520°C for 30 minutes before analysis of
product hydrogen gas D enrichment by isotope ratio mass
spectrometry (VG Isomass 903D). Mean internal and external
precision (SD) levels of the mass spectrometer were 0.17 and 2.1
per mil (%o), respectively, for mean enrichment changes of about
150%o. The sample H3+ contribution was checked dairy, and
appropriate corrections were applied. The instrument was calibrated against water standards of known isotopic composition.
Samples for each subject were analyzed concurrently by use of a
single set of standards.
Calculations of Cholesterol Synthesis Rates
Cholesterol FSRs were determined as incorporation of precursor D into plasma total cholesterol relative to the maximum
544
Arteriosclerosis and Thrombosis
TABLE 2.
Vol 14, No 4 April 1994
Plasma Cholesterol Levels and Fractional Synthesis Rates of Study Subjects
Plasma Cholesterol, mg/dL
Fractional Synthesis Rate, pools/d
Sex
Age, y
Height, cm
Weight, kg
Baseline
Olive
Corn
Canola
Baseline
Olive
Com
Canola
1
F
69
156
55.3
259
230
220
218
0.0228
0.0153
0.0345
0.0337
2
F
52
158
57.9
230
202
193
207
0.0148
0.0427
0.0474
0.0410
3
F
75
178
59.9
224
196
198
192
0.0176
0.0526
0.0278
0.1293
4
F
78
157
63.3
247
226
216
221
0.0635
0.0296
0.0373
0.0542
5
F
71
160
65.3
230
216
202
200
0.0116
0.0707
0.0570
6
F
71
167
66.7
232
217
201
190
0.0329
7
F
68
155
82.2
174
169
159
153
0.0926
0.0412
0.0808
0.0402
Subject
0.0716
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
8
F
78
166
92.6
214
194
183
174
0.0597
0.0355
0.0517
0.0299
9
M
50
167
66.3
287
231
222
210
0.0314
0.0000
0.0338
0.0309
10
M
46
163
75.4
255
223
199
214
0.0362
0.0476
0.0766
0.0549
11
M
43
177
81.7
193
195
193
179
0.0333
0.0229
0.1015
0.0399
12
M
75
178
85.3
190
184
196
212
0.0712
0.0485
0.0870
0.0594
13
M
51
159
88.4
209
200
190
188
0.0316
0.0407
0.0579
0.0516
14
M
48
178
88.8
189
177
160
167
0.0566
0.0652
0.1657
0.0339
15
M
51
187
99.3
189
216
172
186
0.0418
0.0606
0.0721
0.0182
Mean
62
167
75.2
221
205
194
194
0.0412
0.0409
0.0665
0.0492
SEM
3
3
3.6
8
5
5
5
0.0060
0.0052
0.0097
0.0072
theoretical enrichment with the linear-regression model. Considerations of the technique, which is based largely on the
tritiated water uptake method developed in animals, have been
described previously.23-24 With humans, the present technique is
restrictive, in that other than plasma, tissue cholesterol cannot
be sampled over time. However, the 24-hour measurement
period enables equilibration of D-labeled cholesterol across
plasma and tissue pools. The peak and nadir of the diurnal
rhythm in D incorporation have been shown to accord well with
those obtained when synthesis was measured by use of circulating precursor levels.23 The suggestion that plasma free cholesterol D enrichment reflects any immediate change in synthesis is
supported by data indicating very high plasma-compartment
turnover of free cholesterol with liver and other pools.26 Since
the central, rapid-turnover M, cholesterol pool requires months
to attain plateau D enrichment in humans, the initial 24-hour
incorporation rate closely represents the initial turnover value.
Maximum attainable enrichment was calculated as the body
water pool enrichment corrected for the fraction of protons on
de novo synthesized cholesterol that derive from water relative
to nonwater sources.23-24 Data were expressed both as D incorporation as a percentage of plasma water enrichment and as
FSR. The equation used to calculate FSR was
del(%o) Cholesterol
FSR
=
del(%o) Plasma Water x 0.478
where del refers to the difference in D enrichment over 24
hours.
In addition, ASR (in grams per day) was derived by multiplying FSR by an estimate of the volume of the rapid-turnover
compartment for cholesterol. This compartment was calculated from the equation of Goodman et al,27 based on body
weight and circulating cholesterol and trigh/ceride levels.
Oil Fatty Acid and Nonsaponifiable Lipid Contents
Fatty acid compositional analysis of olive, corn, and canola
oil was carried out by gas-liquid chromatography (model 5890,
Hewlett Packard) after lipid extraction and transesterification.28 A 25-mxO.2-mm diphenyl-.dimethyl polysiloxane col-
umn (HP-5) was used with flame ionization detection. Verification of individual fatty acid methyl esters was carried out by
use of authentic standards (Supelco).
Nonsaponifiable lipids of oils were analyzed after lipid
extraction, saponification with KOH, and TLC separation.
Bands containing the nonsaponifiable lipid components were
scraped from TLC plates, eluted, and methylated with trimethylsiloxane reagents.29 Levels were measured by gas-liquid
chromatography (model 5890, Hewlett Packard) with a 30mxO.25-mm-internal diameter dimethyl polysiloxane column
(Restek Corp). Sitosterol, campesterol, and other peaks were
identified with authentic standards and quantified by comparison with a-5-cholestane internal standard.
Statistical Analyses
ANOVA procedures were used to test FSR and ASR data
for differences attributable to feeding group. Unpaired
Tukey"s / test procedures were applied post hoc to detect
intergroup differences. Pearson correlation coefficients were
determined in comparisons of dietary linoleic acid (18:2 n6)
and plant sterol levels against FSR and ASR. Results are
expressed as mean±SEM.
Results
The mean ± SEM age of the subjects was 62 ±3 years.
Mean weight and height were 75.2±3.6 kg and 167±3
cm, respectively (Table 2). Mean caloric intake of the
subjects was 2700±153 kcal • d~\ Cholesterol intakes
were 346±20 and 224±12.7 mg • d"1 during baseline and
plant oil-enriched diets, respectively.
Plasma lipid levels of subjects consuming the test diets
are displayed in Table 2 and Fig 1. Total cholesterol levels
during olive (205±5 mg-dL" 1 ), corn (194±5 mg-dL" 1 ),
and canola (194±5 mg • dL"1) oil diets were significantly
(P<.005) lower than those on the baseline diet (221 ±8
mg • dL"1) (Table 2). Total cholesterol levels when subjects consumed the olive oil diet were higher (P<.01) than
corn and canola phases. LDL levels during consumption
Jones et al Cholesterol Synthesis Response to Dietary Oil Selection
260
i
I Tool Choi
CZ1 LDL Choi
L_J HDL Choi
I m Pool
rlglycerlde
545
-1 2000
200
- 1S00
£•
ISO
$
Q
700
50
s
1S
%
'0
§
5 5
Baseline
Olive
Corn
Canola
Dietary Oil
FIG 1. Bar graph showing plasma total cholesterol (Choi),
low-density lipoproteln (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, and triglyceride levels In subjects consuming baseline and olive, corn, and canola oil-enriched diets.
Values are mean±SEM. *Dlfference relative to baseline diet
(P<.005 for total, P<.001 for LDL, and P<.01 for HDL, respectively). "Difference relative to olive oil diet (P<.01).
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
of olive (132+5 mg-dL- 1 ), corn (125±5 mg-dL"'), and
canola (126±5 mg-dL" 1 ) oil diets were lower (P<.001)
than those on the baseline (152±8 mg • dL"1) diet. For
HDL, whereas levels on the olive oil diet (46±2 mg • dL"1)
were not different from those on the baseline diet (48 ±3
mg • dL"1), levels were reduced on corn (44±2 mg • dL"1)
(P<.01) and canola (44±3 mg-dL" 1 ) oil (P<.05) versus
baseline. Plasma triglyceride levels were not different
between baseline (107±8 mg-dL" 1 ) and olive (112±8
mg-dL" 1 ), corn (108±8 mg-dL" 1 ), and canola (109±7
mg-dL"1) oil diets.
D incorporation rates are shown in Fig 2 for subjects
consuming each diet. Rates are expressed both as
percent precursor incorporation relative to plasma water enrichment and as calculated FSR. Percent incorporation for baseline and olive, corn, and canola oil
diets was 1.97±0.29%, 1.96±0.24%, 3.18±0.44%, and
2.35±0.33%, respectively. FSR values for baseline and
olive, corn, and canola oil diets were 0.0412±0.0060,
0.0409±0.0052, 0.0665±0.0097, and 0.0492±0.0072
pool-d" 1 , respectively (Table 2). In each case, rates
were significantly (P<.05) higher for the corn oil phase
I D Uptake vs Plasma
Baseline
Olive
g^g FSR
Corn
Canola
Dietary Oil
FIG 2. Bar graph showing plasma cholesterol deuterium (D)
Incorporation rates expressed as percent of plasma water deuterium enrichment (Enr) (solid bars) and as fractional synthetic
rate (FSR) (hatched bars) in subjects consuming baseline and
olive, corn, and canola oil-enriched diets. Values are
mean±SEM. *Significant difference relative to corn oil diet
(P<.05).
Baseline
Olive
Corn
Canola
Dietary Oil
FIG 3. Bar graph showing rapid exchange M, cholesterol pool
size (solid bars) and absolute cholesterol synthesis rates (ASR)
(hatched bars) in subjects consuming baseline and olive, corn,
and canola oil-enriched diets. Values are mean±SEM. *Significant difference relative to corn oil diet (P<.05).
compared with olive oil and baseline diet phases. The
FSR for the canola oil phase tended to be similar to
those of olive oil and baseline phases; however, the
difference from the corn oil phase did not reach statistical significance.
Cholesterol pool size and ASRs are shown for subjects
on each dietary phase in Fig 3. There were no differences
in calculated rapid turnover M, pool sizes among the
baseline (25.4±0.9 g) and olive (24.7±0.9 g), corn
(24.4±0.9 g), and canola (24.4±0.9 g) oil groups. ASRs for
baseline and olive, corn, and canola oil diets were
1080±170,1034±140,1697±271, and 1169±137 mg • d"\
respectively. For ASR, corn oil feeding was associated
with an elevation (/><.O5) in synthesis relative to the
baseline and olive oil phase. As with FSR, rates during the
canola phase were similar to those observed during baseline and olive oil phases. Expressed per unit body weight,
synthesis rates were, for baseline, olive, corn, and canola
diet phases, 13.9±1.9, 13.4±1.6, 21.3±2.8, and 16.1±1.3
mg • kg"1 • d"1, respectively.
Fatty acid composition and nonsaponifiable lipid
contents of dietary oils are compared in Table 1. Olive
and canola oils were relatively rich in oleic acid, whereas
corn oil contained higher levels of linoleic acid. Canola
oil contained greater concentrations of linoleic acid and
less palmitic acid compared with olive oil. Combined
nonsaponifiable lipid levels were highest in corn oil,
with lowest levels in olive oil. Campesterol levels were
virtually undetectable in the olive oil sampled. No
lanosterol was found in any of the oils studied.
Both FSR and ASR were in part related to the 18:2
n6 and plant sterol fractions of the oils. Correlation
analyses comparing FSR against total plant sterol level,
sitosterol content, and percent dietary 18:2 n6 yielded
coefficients of #=.119 (P=.O25), # = . 1 2 9 (/)=.O19),
and # = . 1 2 9 (.P=.020), respectively. Comparison of
ASR against total plant sterol level, sitosterol content,
and percent dietary 18:2 n6 yielded coefficients of
# = . 1 0 8 (/"=.059), #=.121 (/)=.024), and # = . 1 2 5
(/>=.O22), respectively. No other significant associations
were detected between cholesterol synthesis and fatty
acid content.
Discussion
Present findings, indicating enhanced cholesterol synthesis in humans consuming corn relative to olive and
546
Arteriosclerosis and Thrombosis
Vol 14, No 4 April 1994
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
perhaps canola oil, suggest a fundamental difference in
the mechanism by which these different plant oils elicit
their cholesterol-lowering effect. Positive correlations
were observed between both the 18:2 n6 and plant
sterol of the three plant oils tested but not with other oil
constituents. These associations suggest that fatty acid
or plant sterol levels of dietary oils consumed may play
a role in the regulation of cholesterogenesis.
Differential regulation of circulating cholesterol levels among various plant oils has been suggested previously by two lines of evidence. First, whereas plant oils
containing MUFAs and PUFAs lower human total
circulating cholesterol concentrations in comparison
with SAFAs, lipoprotein distribution may differ with the
type of plant oil consumed. Although not without
exception,9-30 33 some studies7-8-34 report that oils high in
MUFAs lower LDL but not HDL cholesterol concentrations, whereas those high in PUFAs result in a
reduction of both lipoprotein subspecies. Differences in
the mechanism of action may underlie the distinct
effects observed in lipoprotein profile with feeding of
PUFA- versus MUFA-containing oils.
Second, dietary fat-dependent variations in both
lipoprotein clearance and cholesterol synthesis rates
have been reported in animal and human studies. In
guinea pigs, high PUFA corn oil-containing diets were
shown to increase specific binding of LDL by elevating
LDL receptor numbers compared with diets containing
olive oil or lard.11-35 In hamsters, an increase in hepatic
LDL clearance has been observed in animals consuming
diets containing safflower oil compared with coconut
oil10-13-36 and olive oil.13 Elevated LDL removal, associated with enhanced hepatic elimination, may reduce
circulating cholesterol levels and result in a compensatory enhancement of synthesis. In support of this mechanism, Fernandez et al12 demonstrated that guinea pigs
incorporate tritium into hepatic digitonin-precipitable
sterols more slowly in olive oil-fed animals than in those
fed corn oil. Similarly, cholesterol synthesis, measured
by use of tritiated water incorporation, was shown to
be higher in hamsters fed diets with PUFAs versus
MUFAs 13 and in rats fed PUFAs versus SAFAs.37 Also,
enhanced apolipoprotein B secretion after addition of
linoleic versus oleic acid has been reported in CaCo-2
cells in vitro.38 These results are not unequivocal; other
studies report no differences in hepatic cholesterol
synthesis as a result of feeding oils high in PUFA versus
MUFA in rats39 or hamsters.40 Lack of agreement in
results may be a result of interstudy variation in dietary
composition of fatty acids, cholesterol content, or other
nonsaponifiable components of the oils.
Data specifically comparing the effect of diets enriched with PUFA versus MUFA on cholesterol synthesis in humans are limited. In a single subject, there was
no influence of diets with PUFA on synthesis measured
by sterol balance compared with diets containing
MUFA. 20 Other studies using sterol balance15 and
mononuclear leukocytes16 similarly have shown no difference in cholesterol synthesis rates between diets rich
in PUFA versus SAFA. However, other reports in
humans suggest that diets with PUFA versus SAFA,
usually as formula, result in enhanced fecal elimination
of body cholesterol, 1719 with negative sterol balance
ensuing.19 In the latter study, it was speculated that
cholesterogenesis may increase to compensate for the
enhanced removal of sterol relative to intake. Similarly,
and consistent with findings of animal studies in which
LDL clearance was increased with PUFA feeding^o.u,13.3536 increased body elimination of sterol
would be expected to result in more rapid clearance of
lipoprotein cholesterol, as has been observed in humans
consuming PUFA diets.41 The present results are consistent with the concept that in humans, feeding corn oil
causes enhanced cholesterogenesis, either in response
to increased whole-body excretion and plasma removal
of cholesterol or as a consequence of oil fatty acid
profile or content of some nonsaponifiable component.
The second objective of the present study was to
examine whether levels of fatty acids or some other
component of dietary oils may be responsible for the
effects observed. Fatty acids may directly alter cholesterol synthesis rates in a structure-specific manner.
Increasing evidence suggests that PUFAs are preferentially used for oxidation versus retention in vivo.4244 It is
conceivable that upregulated /J-oxidation of PUFA results in greater availability of acetyl coenzyme A as a
substrate for cholesterol synthesis, although other
mechanisms also possibly account for the differential
effects of the dietary fatty acids contained in the fats
studied here. Alternatively, plant sterols or other compounds such as squalene found in plant oils may upregulate cholesterogenesis indirectly by depressing the
absorption of cholesterol or directly by serving as a
synthesis precursor. Squalene levels were not measured
in the present study. However, sitosterol and total plant
sterol concentrations in the three oils tested varied
directly with cholesterol synthesis, supporting the possibility of their interference with cholesterol absorption.
Given that subjects were consuming a diet reduced in
fat compared with average North American intakes, it
would be predicted that any such effects of dietary fat
would be even greater at levels typically consumed.
The canola oil tested can be considered intermediate
between corn and olive oil in composition, both in
PUFA-to-MUFA ratio and in sitosterol and total plant
sterol content (Table 1). The cholesterol-lowering effects of canola oil were more typical of corn oil than of
olive oil, as has been reported previously.9-22'30 The
intermediate effect of addition of this oil on cholesterol
synthesis rate may therefore be explained through its
plant sterol or fatty acid composition.
Methodological considerations are important in valid
interpretation of the data from this study. Valid results
require subject compliance with diet and procedural
aspects of the study. Subject testing was conducted on
an outpatient basis, with subjects taking out a variable
proportion of meals for consumption outside the Metabolic Research Unit. Since consumption of each meal
was not verified by unit staff, the compliance level of
individual study volunteers could not be confirmed.
However, plasma fatty acid patterns reflected those of
the diet, as has been previously reported,22 suggesting
good compliance.
The deuterated water incorporation technique we used
provides a novel tool for simple, direct measurement of
human cholesterol synthesis. The model shares many
assumptions of the tritium uptake procedure developed in
animals,24-45-46 enabling cholesterogenesis measurement
over short durations with minimal subject involvement
Orcadian synthesis rate variations seen with D incorpo-
Jones et al
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
ration agree well with periodicity in plasma mevalonate
levels.25 Limitations, assumptions, and procedural issues
have been previously reviewed.24-47
Two issues emerge concerning the capacity of the
D-incorporation method to provide an accurate indication of synthesis in the present application. First, label
incorporation is influenced by both the synthesis and
flux rate of unlabeled cholesterol through the rapidly
miscible pool. With higher dietary intakes, more cholesterol would be absorbed and transported to the liver,
competing with labeled sterol for packaging into verylow-density lipoprotein and thereby diluting the measurable plasma D-containing cholesterol. In this way,
measurable D incorporation would be reduced at a level
proportional to the extent of influx of unlabeled sterol.
In the present study, dietary cholesterol contents were
constant across the three plant oil diets; therefore, the
method should yield accurate relative indices of cholesterol formation, unless cholesterol absorption rates varied across oils. If the greater nonsaponifiable lipid levels
in corn oil caused a decline in cholesterol absorption,
D-incorporation rates might be somewhat increased,
although given the low level of cholesterol in these plant
oil diets overall, it seems unlikely that the observed
differences in cholesterol synthesis could be entirely
explained through differences in absorption. Some of
the potential interoil effects on absorption are accounted for through the calculation of ASR, which
factors in variation in the rapidly miscible pool size.
The second possible limitation of the methodology
concerns the potential impact of dietary fatty acid
composition on the D-incorporation ratio, a critical
ratio in the derivation of FSR values.24 It has been
suggested that the observed disparity between results of
tritium incorporation and sterol balance techniques for
cholesterogenesis measurement in guinea pigs fed fats
differing in fatty acid composition was due to dietary
fat-mediated effects on the metabolic source of
NADPH. 12 Variation in the tracer enrichment of
NADPH caused by the fat type consumed could produce an apparent difference in synthesis, whereas actual
rates remained unchanged. Our calculations have
shown that any such error would be quantitatively
minor24 and thus unlikely to result in changes in synthesis of the magnitude presently observed.
Present results suggest that differences in cholesterol
synthesis in response to dietary fat selection occur in
middle-aged, moderately hypercholesterolemic subjects
consuming diets relatively low in fat content. Whether
these differences occur in normolipidemic or younger
subjects cannot be established from this study. The
intention was to examine the mechanisms of action of
lowering of circulating cholesterol levels in subjects with
a greater risk of heart disease, who would derive the
largest benefit from cholesterol-lowering diets.
In summary, present findings support work done in
animals suggesting that there is a more rapid flux of
central-pool cholesterol associated with enhanced synthesis subsequent to corn versus olive oil feeding. Although other factors not presently identified may be
responsible for this association, 18:2 n6 and plant sterol
content of oils used are possible contributing factors.
Further investigation will be required to delineate the
precise underlying mechanism responsible for this
effect.
Cholesterol Synthesis Response to Dietary Oil Selection
547
Ack nowledgments
This study was supported by grants from the British Columbia and Yukon Heart and Stroke Foundation, the US Department of Agriculture (contract 53-3K06-5-10), the National
Institutes of Health (HL-39326), Uncle Ben's Inc, Dallas, Tex,
and the US Egg Marketing Board. The authors thank the staff
of the Metabolic Research Unit, Human Nutrition Research
Center on Aging at Tufts University, for expert services
provided; gratefully appreciate the cooperation of study subjects; and acknowledge the excellent technical assistance of
Clement Fung, Isabeau Iqbal, Yi Lin, Joost Schulte, and Brian
Toy. Thanks are also extended to Drs Lynne Ausman, Tufts
University, and Mark Bieber, Best Foods Inc, for assistance
with nonsaponifiable lipid analysis, and to Dr Barry Goldin,
Tufts University, for help with the fatty acid analysis.
References
1. AHA Medical/Scientific Statement. The cholesterol facts- a
summary of the evidence relating dietary fats, serum cholesterol,
and coronary heart disease: a joint statement by the American
Heart Association and the National Heart, Lung, and Blood
Institute. Circulation. 199O;81:1721-1733.
2. Hegsted DM, McGandy RB, Myens ML, Stare FJ. Quantitative
effects of dietary fat on serum cholesterol in man. Am J Clin Nutr.
1965;17:281-295.
3. Kinsell LW, Micheals GD. Letter to the editor. Am J Clin Nutr.
1955^3:247-253.
4. Ahrens EH, Insull W Jr, Blomstrand R, Hirsch J, Tsaltas TT,
Peterson ML. The influences of dietary fats on serum-lipid levels
in man. Lancet 1957;l:943-953.
5. Grundy SM, Florentin L, Nix D, Whelan MF. Comparison of
monounsaturated fatty acids and carbohydrates for reducing raised
levels of plasma cholesterol in man. Am J Clin Nutr. 1988;47:
965-969.
6. Mensink RP, Katan MB. Effect of a diet enriched with monounsaturated or polyunsaturated fatty acids on levels of low-density
and high-density lipoprotein cholesterol in healthy women and
men. N EnglJ Med. 1989;321:436-441.
7. Grundy SM. Monounsaturated fatty acids and cholesterol metabolism: implications for dietary recommendations. / Nutr. 1989;119:
529-533.
8. Baggio G, Pagnan A, Muraca M, Martini S, Opportuno A,
Bonanome A, Ambrosio GB, Ferrari S, Guarini P, Piccolo D,
Manzato E, Corrocher R, Crepaldi G. Olive-oil-enriched diets:
effect on serum lipoprotein levels and biliary cholesterol saturation. Am J Clin Nutr. 1988;47:960-964.
9. McDonald BE, Gerrard JM, Bruce VM, Corner EJ. Comparison of
the effect of canola oil and sunflower oil on plasma lipids and
lipoproteins and on in vivo thromboxane A2 and prostacyclin
production in healthy young men. Am J Clin Nutr. 1989;50:
1382-1388.
10. Woollen LA, Spady DK, Dietschy JM. Saturated and unsaturated
fatty acids independently regulate low density lipoprotein receptor
activity and production rate. / Lipid Res. 1992;33:77-88.
11. Fernandez ML, Lin ECK, McNamara DJ. Regulation of guinea
pig plasma low density lipoprotein kinetics by dietary fat unsaturation. / Lipid Res. 1992;33:97-109.
12. Fernandez ML, Yount NY, McNamara DJ. Whole body and
hepatic cholesterol synthesis rates in the guinea-pig: effect of
dietary fat quality. Biochim Biophys Acta. 1990;1044:340-348.
13. Spady DK, Dietschy JM. Interaction of dietary cholesterol and
trigtycerides in the regulation of hepatic low density lipoprotein
transport in the hamster. J Clin Invest 1988;81:300-309.
14. Mercer NJH, Holub BJ. Measurement of hepatic sterol synthesis
in the Mongolian gerbil in vivo using [3H] water: diurnal variation
and effect of type of dietary fat. / Lipid Res. 1981;22:792-799.
15. Grundy SM, Ahrens EH. The effects of unsaturated dietary fats on
absorption, excretion, synthesis, and distribution of cholesterol in
man. J Clin Invest 1970,49:1135-1152.
16. McNamara DJ, Kolb R, Parker TS, Batwin H, Samuel P, Brown
CD, Ahrens EH. Heterogeneity of cholesterol homeostasis in man.
J Clin Invest. 1987;79:1729-1739.
17. Nestel PJ, Havenstein N, Homma Y, Scott TW, Cook LJ. Increased
sterol excretion with polyunsaturated fat-high cholesterol diets.
Metabolism. 1975^4:189-198.
18. Grundy SM. Effects of polyunsaturated fats on lipid metabolism in
patients with hypertrigryceridemia. J Clin Invest. 1975^5:269-282.
548
Arteriosclerosis and Thrombosis
Vol 14, No 4 April 1994
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
19. Connor WE, Witiak DT, Stone DB, Armstrong ML. Cholesterol
balance and fecal neutral steroid excretion and bile acid excretion
in normal men fed dietary fats of different fatty acid composition.
J Clin Invest 1969;48:1363-1375.
20. Spritz N, Ahrens EH, Grundy S. Sterol balance in man as plasma
cholesterol concentrations are altered by exchanges of dietary fats.
J Clin Invest. 1965;44:1482-1493.
21. The Expert Panel. Report of the National Cholesterol Education
Program on detection, evaluation and treatment of high blood
cholesterol in adults. Arch Intern Med. 1988;148:36-39.
22. Lichtenstein AH, Ausman LM, Carrasco W, Gualtieri LJ, Jenner
JL, Ordovas JM, Schaefer EJ. Effects of canola, com and olive oil
on fasting and postprandial lipoproteins in humans as part of a
National Cholesterol Education Program step 2 diet. Arterioscler
Thromb. 1993;13:1533-1542.
23. Jones PJH, Scanu AM, Schoeller DA. Plasma cholesterol synthesis
using deuterated water in humans: effect of short-term food
restriction. / Clin Lab Med. 1988;111:627-633.
24. Jones PJH, Leitch CA, U Z-C, Connor WE Human cholesterol
synthesis measurement using deuterated water theoretical and
procedural considerations. Arterioscler Thromb. 1993;13:247-253.
25. Jones PJH, Papu AS, IUingworth DR, Leitch CA. Correspondence
between plasma mevalonic acid levels and deuterium uptake in
measuring cholesterol synthesis. EwJ Clin Invest. 1992^22:609-613.
26. Schwartz CC, Zech LA, VandenBroek JM, Cooper PS. Cholesterol
kinetics in subjects with bile fistula. / Clin Invest. 1993;91:923-938.
27. Goodman DS, Smith FR, Seplowitz AH, Ramakrishnan R, Dell RB.
Prediction of parameters of whole body cholesterol metabolism in
humans. / LJpid Res. 198021:699-713.
28. Bannon CD, Craske JD, Hai NT, Harper NL, O'Rourke KL.
Analysis of fatty acid methyl esters with high accuracy and reliability, II: methyiation of fats and oils with boron trifluoridemethanol. J Chromatogr. 1982;247:63-69.
29. Miettinien TA. Serum methyl sterols and their distribution
between major lipoprotein fractions in different clinical conditions.
Ann Clin Res. 1971^3:264-271.
30. Chan JK, Bruce VM, McDonald BE. Dietary alpha-linolenic acid
is as effective as oleic and linoleic acid in lowering blood cholesterol in normolipidemic men. Am J Clin Nutr. 1991;53:
1230-1234.
31. Ginsberg HN, Barr SL, Gilbert A, Karmalry W, Deckelbaum R,
Kaplan K, Ramakrishnan R, Holleran S, Dell RB. Reduction of
plasma cholesterol levels in normal men on an American Heart
Association step I diet or a step I diet with added monounsaturatedfat. NEnglJMed. 1990;322:574-579.
32. Wardlaw GM, Snook JT, Lin M-C, Puangco MA, Kwon JS. Serum
lipid concentrations in healthy men on diets enriched in either
canola oil or safflower oil. Am J Clin Nutr. 1991^4:104-110.
33. Wahrburg U, Martin H, Sandkamp M, Schulte H, Assmann G.
Comparative effects of a recommended lipid-lowering diet vs a diet
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
rich in monounsaturated fatty acids on serum lipid profiles in
healthy young adults. Am J Clin Nutr. 1992^6:678-683.
Sirtori CR, Tremoli E, Gatti E, Montanari G, Sirton M, Colli S,
Gianfranceschi G, Maderna P, Dentone CZ, Testolin G, Galli C.
Controlled elevation of fat intake in the Mediterranean diet: comparative activities of olive oil and corn oil on plasma lipids and
platelets in high risk patients. Am J Clin Nutr. 1986;44:635-642.
Fernandez ML, McNamara DJ. Dietary fat-mediated changes in
hepatic apoprotein B/E receptor in the guinea pig: effect of potyunsaturated, monounsaturated, and saturated fat. Metabolism.
1989;8:1094-1102.
Spady DK, Dietschy JM. Dietary saturated triacylglycerols suppress
hepatic low density lipoprotein receptor activity in the hamster. Proc
Nad Acad Sci U S A. 1985^2:4526-4530.
Avignan J, Steinberg D. Effects of saturated and unsaturated fat on
cholesterol metabolism in the rat. Proc Soc Biol Med, 1958;97:
814-816.
de Bruin TWA, van Greevenbroek M, Bouritius H, van Meer G,
Erkelens DW. Linoleic acid is a very potent apolipoprotein B48/
apoBlOO secretagogue in polarized CaCo-2 cell monolayers. Circulation. 1992;86(suppl I):I-748. Abstract.
Triscari J, Hamilton JG, Sullivan AC. Comparative effects of saturated and unsaturated lipids on hepatic lipogenesis and cholesterogenesis in vivo in the meal-fed rat. J Nutr. 1978;108:815-825.
Jones PJH, Ridgen JE, Benson AP. Influence of dietary fatty acid
composition on lipid synthesis and esterification in hamsters.
Lipids. 1990;25:815-820.
Demacker PNM, Reijnen IGM, Katan MB, Stuyt PMJ, Stalenhoef
AFH. Increased removal of remnants of trigh/ceride-rich lipoproteins on a diet rich in poryunsaturated fatty acids. Eur J Clin
Invest. 1991;21:1997-2003.
Shimomura Y, Tamura T, Suzuki M. Less body fat accumulation in
rats fed a safflower oil diet than in rats fed a beef tallow diet.
JNutr. 1990;120:1291-1296.
Mercer SW, Trayhurn P. Effect of high fat diets on the thermogenic activity of brown adipose tissue in cold acclimated mice.
JNutr. 1987;117:2147-2153.
Jones PJH, Schoeller DA. Potyunsaturated: saturated ratio of diet fat
influences energy substrate utilization in the human. Metabolism.
1988^7:145-151.
Lakshmanan MR, Veech R L Measurement of rate of rat liver
sterol synthesis in vivo using tritiated water./Biol Chem. 1977;252:
4667-4673.
Jeske DJ, Dietschy JM. Regulation of rates of cholesterol synthesis
in vivo in the liver and carcass of the rat measured using [3H)water.
J Lipid Res. 1980;21:364-376.
Wong WW, Hachey DL, Feste A, Leggitt J, Clarke LL, Pond WG,
Klein PD. Measurement of in vivo cholesterol synthesis from 2H2O:
a rapid procedure for the isolation, purification and isotopic assay
of erythrocyte cholesterol. J Lipid Res. 1991;32:1049-1056.
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Effect of dietary fat selection on plasma cholesterol synthesis in older, moderately
hypercholesterolemic humans.
P J Jones, A H Lichtenstein, E J Schaefer and G L Namchuk
Arterioscler Thromb Vasc Biol. 1994;14:542-548
doi: 10.1161/01.ATV.14.4.542
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 1994 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/14/4/542
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which permission
is being requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answerdocument.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/