Articles |
Presented in part at the 65th Scientific Sessions of the American Heart Association, New Orleans, La, and published in abstract form (Circulation. 1992;86[suppl I]:I-71.)
From the Laboratoire de Biochimie des Lipoprotéines, INSERM CJF 93-10, Faculté de Médecine, Dijon, France.
Correspondence to Laurent Lagrost, Laboratoire de Biochimie Médicale, Hôpital du Bocage, 21034 Dijon, France.
| Abstract |
|---|
|
|
|---|
VLDL+LDL activity) (r=.46;
P<.05) but not with the transfer of cholesteryl ester from
LDL toward plasma HDL (CETLDL
HDL activity)
(r=-.05; NS). The high binding capacity of
albumin for NEFA was used to investigate whether
lipoprotein-bound NEFAs were implicated in the modulation of the
cholesteryl ester transfer reaction. As compared with nonsupplemented
controls, the addition of an excess of fatty acidfree
albumin (8 g/L) to total normolipidemic plasmas reduced
CETHDL
VLDL+LDL activity (18.3±5.5% versus 9.8±3.1%;
P<.0001) but not CETLDL
HDL activity
(22.3±4.5% versus 23.3±5.1%; NS). Moreover,
CETHDL
VLDL+LDL and CETLDL
HDL activities
correlated negatively when measured in native plasma
(r=-.45; P<.05) but positively when measured in
albumin-supplemented plasma (r=.40;
P<.05). In long-term incubation experiments,
lipoprotein-bound NEFA increased the net mass transfer of
cholesteryl esters from HDL toward VLDL+LDL but reduced the net mass
transfer of triglycerides in the opposite direction, from
VLDL+LDL toward HDL. Taken together, data of the present study
brought strong and concordant arguments in favor of a dual effect of
plasma NEFA in modulating both the mass and the activity of CETP in
vivo.
Key Words: plasma lipoprotein cholesteryl ester triglyceride cholesteryl ester transfer protein
| Introduction |
|---|
|
|
|---|
The goals of the present study were to determine whether NEFAs in fasting human plasma can modulate CETP-mediated cholesteryl ester transfers and to investigate their effects on the reciprocal exchange of cholesteryl esters and triglycerides between HDL and apo Bcontaining lipoproteins.
| Methods |
|---|
|
|
|---|
Lipoprotein Preparation
Lipoprotein fractions were isolated from total normolipidemic
plasmas by sequential ultracentrifugation at
100 000 rpm (350 000g) in a TLA-100.2 rotor in a Beckman
TL-100 ultracentrifuge. LDL was isolated as the plasma
fraction 1.019<d<1.055 g/mL with one 4-hour spin at the
lowest density and two 5-hour spins at the highest density.
HDL3 was isolated as the plasma fraction
1.13<d<1.21 g/mL with one 5-hour spin at the lowest
density and two 7-hour spins at the highest density. Lipoproteins were
recovered by tube slicing and were then dialyzed overnight against a
buffer of 10 mmol/L Tris, 150 mmol/L NaCl (pH 7.4) containing 5 mmol/L
EDTA-Na2 and 3 mmol/L NaN3 (TBS buffer).
Purification of CETP
CETP was partially purified from citrated human plasma according
to the sequential procedure previously described.31
Briefly, the ultracentrifugally isolated d>1.21 g/mL plasma
proteins were fractionated successively by hydrophobic interaction
chromatography on a phenyl-Sepharose CL-4B column,
affinity chromatographies on Heparin-Ultrogel A4R and
Blue-Trisacryl columns, and anion-exchange
chromatography on a Mono-Q HR 5/5 column.
Chromatographic separations were performed by using a fast
protein liquid chromatography system (Pharmacia). The
preparation of CETP was deficient in LCAT activity and PLTP
activity.31
Radiolabeling of LDL and HDL3
HDL3 was biosynthetically labeled according to the
general procedure previously described.32 A
d>1.13 g/mL plasma fraction was obtained after
ultracentrifugation of 20 mL total normolipidemic
plasma, dialyzed against TBS, and incubated with 10 nmol [1
,
2
(n)-3H]cholesterol (specific activity, 46
Ci/mmol; Amersham) for 24 hours at 37°C in a shaking water bath.
Subsequently, the 1.019<d<1.055 g/mL fraction obtained
from 10 mL total normolipidemic plasma (about 15 mg LDL
cholesterol) was added to the incubated mixtures. The
incubation was then prolonged for a 6-hour period to allow the exchange
of radiolabeled esterified cholesterol between lipoprotein
substrates. That procedure presents two main advantages: (1)
radiolabeled HDL3 and LDL are prepared in the same batch
and (2) HDL3 is replete with unlabeled nonesterified
cholesterol during the 6-hour prolongation of the
incubation in the presence of isolated LDL. At the end of the
incubation, the LDL and HDL3 fractions were recovered by
sequential ultracentrifugation as described above.
Typical labeled preparations of LDL and HDL3 obtained with
this procedure had specific activities of approximately 7000 and
28 000 cpm/nmol of cholesterol, respectively. As judged by
thin-layer chromatography, more than 95% of total
radioactivity of both lipoprotein substrates resided in the cholesteryl
ester moiety.
Isotopic Assays of Plasma CET Activity
The CET activity of a sample was measured as its capacity
to promote the transfer of radiolabeled cholesteryl esters from donor
to acceptor lipoprotein substrates. In native plasma, total CET
activity was evaluated by measuring the rate of radiolabeled
cholesteryl esters transferred either from HDL3 to
endogenous apo Bcontaining plasma lipoproteins
(CETHDL
VLDL+LDL activity) or from LDL to
endogenous plasma HDL (CETLDL
HDL activity).
For that purpose, a tracer dose (2.5 nmol of cholesterol)
of either radiolabeled HDL3 or radiolabeled LDL was added
to 25 µL of plasma in a final volume of 50 µL. Plasma LCAT activity
was blocked by adding iodoacetate (final concentration, 1.5 mmol/L). In
some experiments, fatty acidfree albumin (Sigma) was
added (final concentration, 8 g/L). Duplicate mixtures were incubated
for 3 hours at 37°C in an incubator. At the end of the incubation,
the tubes were immediately placed on ice. A volume of 45 µL of the
incubated mixtures was adjusted to density 1.07 g/mL Kbr in a final
volume of 2 mL in Quickseal centrifugation tubes
(Beckman). The tubes were then sealed and subjected to
ultracentrifugation for 7 hours at 50 000 rpm
(269 000g) in a 50.4 Ti rotor in an L7
ultracentrifuge (Beckman). The plasma fractions of
d<1.068 g/mL (plasma apo Bcontaining lipoproteins) and of
d>1.068 g/mL (HDL-containing plasma fraction) were
recovered in a volume of about 1 mL and transferred into counting
vials. A volume of 2 mL scintillation fluid (OptiScint Hisafe 3,
Pharmacia) was added to each vial, and the radioactivity was assayed
for 5 minutes in a Wallac 1410 liquid scintillation counter
(Pharmacia). The recovery of total radioactivity in the
d<1.068 and d>1.068 g/mL fractions was
constantly greater than 95%. In nonincubated controls containing
radiolabeled HDL3, the radioactivity recovered in
the d<1.068 g/mL fraction did not exceed 4% of the total.
In nonincubated controls containing radiolabeled LDL, less than 10% of
the radioactivity was recovered in the d>1.068 g/mL
fraction. CET activity was measured as the rate of total radiolabeled
cholesteryl esters transferred from the lipoprotein tracer to the
d<1.068 g/mL or the d>1.068 g/mL acceptor
fraction during a 3-hour incubation at 37°C after deduction of blank
values of control mixtures kept at 4°C.33 Results were
expressed in percentage of radiolabeled cholesteryl esters
transferred.
Cholesteryl Ester and Triglyceride Mass Transfer
Assay
The mass transfer of cholesteryl esters and
triglycerides between HDL and apo Bcontaining
lipoproteins was measured during the long-term incubation of total
plasma at 37°C in the presence of an LCAT
inhibitor.34 For that purpose, 100 µL
aliquots of total plasma were supplemented with iodoacetate (final
concentration, 1.5 mmol/L) and incubated at 37°C for 24 hours. At the
end of incubation, apo Bcontaining lipoproteins were selectively
precipitated with a phosphotungstic acid/MgCl2 reagent
(Boehringer Mannheim) as recommended by the manufacturer.
Unesterified cholesterol, total cholesterol,
and triglyceride concentrations present in the
supernatant were measured, and the mass of esterified
cholesterol and triglycerides transferred were
calculated as compared with control samples maintained at 4°C.
CETP ELISA
CETP concentration in total human plasma was measured by using a
competitive ELISA on a Biomek 1000 Biorobotic System (Beckman
Instruments).35 Calibration was carried out by comparison
with a standard curve obtained with a frozen plasma
standard.35 Calculation of CETP concentrations was
realized by using a data analysis software (Immunofit EIA/RIA
Data Analysis Software, Beckman). The CETP concentration value
for each plasma sample was calculated by averaging quadruplicate
determinations. Intra-assay and interassay coefficients of
variation were 4% and 6%, respectively.
Specific CET Activity
CETHDL
VLDL+LDL and CETLDL
HDL
specific activities were calculated as the ratio of plasma CET activity
to plasma CETP mass concentration and expressed in percentage of total
radiolabeled cholesteryl esters transferred per microgram of CETP.
Precipitation of the Total Plasma Lipoprotein
Fraction
Lipoproteins were removed from total human plasma by using a
single-step adsorption method based on the ability of FSD (SERVA
Feinbiochemica) to remove large particles, such as lipid emulsions and
lipoprotein particles, from aqueous media.36 37 38 Briefly,
total plasmas were treated for 10 minutes at room temperature with
various concentrations of FSD according to the general procedure
previously described.38 At the end of the incubation,
precipitates were removed by low-speed
centrifugation, and remaining amounts of NEFA,
cholesterol, apo B, apo A-I, and albumin were
assayed in the supernatants.
Other Analytical Methods
All chemical assays were performed on a Cobas-Fara Centrifugal
Analyzer (Roche). Total cholesterol, unesterified
cholesterol, and triglyceride concentrations
were measured by enzymatic methods using Boehringer reagents.
HDL cholesterol was determined after precipitation of apo
Bcontaining lipoproteins with phosphotungstic acid/MgCl2
as described above. Concentrations of plasma apo B, apo AI, and apo AII
were determined by immunoturbidimetry39 with anti-apo
B, anti-apo AI, and anti-apo AII antibodies purchased from
Behring. Apo B and apo AI standards were purchased from Behringwerke
AG. Apo AII standard was purchased from Immuno AG. NEFA concentrations
were determined by using both a commercially available enzymatic kit
(Wako Pure Chemical Industries) and a capillary gas
chromatography procedure.40 By use of the
statistical method of Bland and Altman,41 a good agreement
was observed between enzymatic and gas chromatography
plasma NEFA determinations. Albumin concentrations were
measured by using the A-Gent Albumin Test from Abbott.
Statistical Analysis
Student's t test was used to determine the
significance of the difference between data means. Coefficients of
correlation were calculated by using linear regression
analysis. Multiple regression analysis was used to
determine significant contributions of triglycerides, HDL
cholesterol, VLDL+LDL cholesterol, and NEFA to
the prediction of CET activity in human plasma.
| Results |
|---|
|
|
|---|
Correlation of Plasma NEFA Concentration With CET
Activity
CETP activity was evaluated by measuring the transfer of tritiated
cholesteryl esters from a tracer dose of exogenous radiolabeled
lipoproteins toward endogenous plasma lipoprotein fractions
(see "Methods"). As shown in Fig 1
, the addition
of increasing concentrations of purified CETP to total normolipidemic
plasmas resulted in CET assays that gave linear response and permitted
the measurement of plasma CETP activity over a wide range of
values.
|
Among a population of 37 normolipidemic subjects (plasma total
cholesterol range, 121 to 240 mg/dL; plasma
triglyceride range, 27 to 142 mg/dL; plasma NEFA range,
0.07 to 1.02 mmol/L), the rate of transfer of cholesteryl esters from
radiolabeled HDL3 to the d<1.068 g/mL fraction
correlated positively with total NEFA concentrations (r=.46,
P<.005) (Fig 2
), while no significant
relation was observed with plasma albumin levels (results not
shown).
|
To investigate further the effect of NEFA on the bidirectional
transfer of cholesteryl esters between HDL and VLDL+LDL plasma
fractions, CET activity was measured in 27 normolipidemic plasmas
(total cholesterol range, 121 to 234 mg/dL; plasma
triglyceride range, 27 to 84 mg/dL) in two ways, either
from a tracer dose of radiolabeled HDL3 toward plasma
apo Bcontaining lipoproteins (CETHDL
VLDL+LDL activity)
or from a tracer dose of LDL toward plasma HDL
(CETLDL
HDL activity). As shown in Table 1
, CETHDL
VLDL+LDL activity correlated
negatively with HDL cholesterol and esterified
cholesterol:triglyceride ratio in HDL and
correlated positively with VLDL+LDL cholesterol,
VLDL+LDL:HDL cholesterol ratio, plasma
triglyceride, and NEFA concentrations. Conversely,
CETLDL
HDL activity correlated negatively with VLDL+LDL
cholesterol, VLDL+LDL:HDL cholesterol ratio,
and triglyceride levels and positively with HDL
cholesterol levels. In contrast to
CETHDL
VLDL+LDL activity, CETLDL
HDL
activity did not correlate significantly with NEFA concentrations
(Table 1
). Multiple regression analysis revealed that plasma
triglyceride, HDL cholesterol, VLDL+LDL
cholesterol, and NEFA concentrations, when combined in a
four-variable model, accounted for 75% of the variability in
CETHDL
VLDL+LDL activity, with only NEFA reaching a
significant level (P=.025), and for 68% of the variability
in CETLDL
HDL activity, with both HDL
cholesterol and VLDL+LDL cholesterol levels
reaching significant levels (P=.033 and P=.010,
respectively). Correlations between plasma NEFA concentration and
specific CET activities calculated as the ratio of plasma CET activity
to plasma CETP mass concentration did not reach the significance
level.
|
Distribution of NEFA in Normolipidemic Human Plasma
To evaluate the relative proportions of
lipoprotein-bound and albumin-bound NEFA, lipoproteins
were removed from total human plasma by using FSD, as described under
"Methods." As shown in Fig 3
, precipitated amounts
of total plasma cholesterol, apo B, and apo AI increased
gradually while raising the final concentration of FSD. When
present at a concentration of 2.5 mg/mL, FSD removed the totality
of plasma lipoprotein components, whereas only 5% of plasma
albumin was coprecipitated (Fig 3
). Interestingly, specific
removal of plasma lipoproteins was accompanied by an approximately 20%
reduction in plasma NEFA concentration (Fig 3
). NEFA content of the
plasma lipoprotein fraction was determined in 14 fasting plasmas with
NEFA levels ranging from 0.16 to 1.07 mmol/L. A significant, positive
correlation was observed between total plasma NEFA concentrations and
NEFA amounts in the plasma lipoprotein fraction (Fig 4
).
In good agreement with previous data of Shafrir,42
parameters of linear regression analysis indicated
that approximately 20% of plasma NEFA localized in the lipoprotein
fraction, independent of NEFA levels in total plasma (Fig 4
).
|
|
Among a subset of 10 plasmas (cholesterol range, 145 to 240
mg/dL; triglyceride range, 49 to 142 mg/dL; NEFA range,
0.16 to 1.02 mmol/L; albumin range, 4.4 to 5.7 mg/dL), a
positive correlation between the amount of NEFA bound to the plasma
lipoprotein fraction and CETHDL
VLDL+LDL activity was
observed (r=.65; P<.05).
Effect of Albumin on Plasma CET Activity
Albumin, the major carrier of NEFA in human plasma, is
able to remove NEFA from plasma lipoprotein substrates when added in
excess.42 In the present study, we made use of the
high binding capacity of albumin for NEFA to confirm further
that plasma NEFAs were implicated in the modulation of the CET
reaction. For that purpose, we investigated the effect of the
supplementation of plasmas with fatty acidfree albumin
(final concentration, 8 g/L) on the rate of CET between the
d<1.068 g/mL and d>1.068 g/mL plasma fractions.
As assessed by using denaturing polyacrylamide gradient gel
electrophoresis and specific anti-apolipoprotein immunoassays, and
as indicated by the supplier (Sigma), the purified albumin
fraction used in the present study was at least 99% pure (results
not shown). The addition of albumin induced a significant
reduction in CETHDL
VLDL+LDL activity (18.3±5.5% versus
9.8±3.1%; n=27; P<.0001). In contrast,
CETLDL
HDL activity was not significantly modified in the
presence of albumin (22.3±4.5% versus 23.3±5.1%; n=27; NS).
As shown in Fig 5
, CETHDL
VLDL+LDL
activity did not correlate anymore with HDL cholesterol
levels in the presence of albumin. In contrast, a significant
positive correlation of CETHDL
VLDL+LDL activity with
VLDL+LDL cholesterol levels subsisted and was even more
pronounced after albumin supplementation (Fig 5
). Whereas
CETLDL
HDL activity correlated positively with HDL
cholesterol levels (r=.447; P<.05)
and negatively with VLDL+LDL cholesterol levels
(r=-.483; P<.05), the addition of
albumin to the incubation mixture abolished the significant
correlations (r=.176, NS, and r=.239, NS,
respectively).
|
In native plasmas, regression analysis revealed the existence
of a significant negative correlation between
CETHDL
VLDL+LDL activity and CETLDL
HDL
activity (r=-.45, P<.05). By contrast, when
CETs were measured in plasma supplemented with albumin, we
observed no more a negative but a significant positive correlation
between the transfers in both directions (r=.40,
P<.05).
Net Mass Transfers of Cholesteryl Esters and
Triglycerides Between HDL and VLDL+LDL Plasma
Fractions
To investigate whether nonesterified fatty acids could
influence not only the velocity of CETs but also the capacity of
lipoprotein substrates to accommodate or donate neutral lipids, total
plasmas, containing an LCAT inhibitor (iodoacetate, 1.5
mmol/L), were incubated at 37°C for 24 hours in the presence or in
the absence of albumin. As shown in Table 2
,
incubation of plasmas in the absence of albumin significantly
decreased the esterified cholesterol content of HDL but
increased its triglyceride content (variations compared
with plasmas kept at 4°C: -4.5±1.4 mg/dL, P<.001, and
+14.4±4.7 mg/dL, P<.001, respectively). The depletion in
HDL esterified cholesterol induced by the incubation of
total plasma was significantly lower in the presence of albumin
than in nonsupplemented samples (Table 2
). Conversely, the increase in
HDL triglycerides was significantly higher in the presence
of albumin than in nonsupplemented samples (Table 2
).
|
| Discussion |
|---|
|
|
|---|
In the normolipidemic population studied, we searched first for relations between plasma CETP mass concentration, as determined by using a specific ELISA,35 and plasma lipid parameters. Plasma CETP levels were positively and significantly correlated only with NEFA levels and the VLDL+LDL:HDL cholesterol ratio, whereas no significant relations were observed with other plasma lipoprotein components. The observation of a positive correlation between plasma NEFA and CETP mass concentrations, together with the previously described ability of fatty acids to enhance in vitro the secretion of CETP by cultured cells,16 17 suggest that NEFA might upregulate the synthesis and/or secretion of CETP in vivo. Furthermore, the positive relations between NEFA and CETP levels might account, at least in part, for recent observations of Moulin and coworkers,47 who reported increased plasma CETP concentrations in dyslipidemic patients with nephrotic syndrome, a pathology shown to be associated with abnormally elevated NEFA content of plasma lipoproteins.42
To determine further whether NEFA can affect not only the mass
but also the activity of CETP, we chose to assay CET activity in total,
native plasma. In previous studies, combinations of radiolabeled (donor
particles) and unlabeled (acceptor particles) lipoprotein substrates
have been used widely to evaluate the lipid transfer rates in
plasma.9 20 As they have been performed either with total
plasma, delipidated plasma, or with isolated plasma fractions, plasma
lipid transfer assays are somewhat different, and all of them have
their limitations.9 20 In vitro studies have demonstrated
that CET activity is dependent on many factors including, in addition
to CETP mass,18 19 the amount,48 49 the
apoprotein composition,32 50 51 52 and the lipid
composition50 53 54 of lipoprotein substrates (see
Reference 99 for a review). Since in the present study CETP activity
was measured in freshly collected native plasmas without prior
treatment, values reflect the activity of the CETP as modulated by the
endogenous plasma factors. That latter procedure led to
results that are in good agreement with previously reported data.
Indeed, the negative correlation we observed between the rate of CET
toward VLDL+LDL and both HDL cholesterol levels and
esterified cholesterol:triglyceride ratio in
HDL confirmed the results from previous studies,55 56 57 58
which have shown that high levels of plasma CETP activity in vivo cause
lower HDL cholesterol levels. Positive correlations with
VLDL+LDL cholesterol concentrations, VLDL+LDL:HDL
cholesterol ratio, and triglyceride
concentrations observed in the present study confirmed the direct
relationship that has recently been described in normolipidemic
subjects between plasma VLDL concentrations and net CET from HDL to apo
Bcontaining lipoproteins.58 While plasma CET activity
has been measured usually in only one direction, it is
noteworthy that CETP is actually an exchange protein that
promotes the transfer of cholesteryl esters in several
directions.9 20 The diversity in CETs is important to
consider since a recent study in dyslipoproteinemic patients has shown
that increased transfer of cholesteryl esters toward plasma VLDL+LDL
can be associated with decreased transfer toward HDL.59
Thus, in the present study, plasma CET activity was evaluated not
only by measuring the transfer of radiolabeled cholesteryl esters from
HDL toward VLDL+LDL (CETHDL
VLDL+LDL activity) but also
from LDL toward HDL (CETLDL
HDL activity). Globally, the
relations between plasma lipid parameters and
CETLDL
HDL activity were in contrast with those observed
with CETHDL
VLDL+LDL activity. Indeed,
CETLDL
HDL activity correlated positively with HDL
cholesterol but negatively with plasma
triglycerides, VLDL+LDL cholesterol, and
VLDL+LDL:HDL cholesterol ratio. It must be noted that the
fraction of radioactivity transferred may be underestimated in the
CETLDL
HDL assay due to undetected transfers from LDL
toward VLDL. In particular, because of the negative correlation between
VLDL and HDL concentrations in plasma, a reduced transfer from LDL
toward HDL might be associated with an increased transfer from LDL
toward VLDL. That latter point might account in part for the
significant correlations of CETLDL
HDL activity with HDL
cholesterol, plasma triglycerides, VLDL+LDL
cholesterol, and VLDL+LDL:HDL cholesterol
ratio.
One of the novel findings from the present study is the significant
positive correlation of plasma NEFA concentration with
CETHDL
VLDL+LDL activity. In contrast, no significant
correlation with CETLDL
HDL activity was observed,
indicating again that the choice of both donor and acceptor lipoprotein
substrates constitutes one source of variation in CETP activity
measurement. Plasma NEFA concentration did not correlate significantly
with transfer activities expressed relative to CETP mass, ie, specific
CET activities, indicating that (1) as expected, CETP mass constitutes
one major determinant of plasma neutral lipid transfers and (2) when
CETP mass is taken into account, NEFAs do not constitute the single,
most important modulators of the transfer reaction. In fact, it is well
known that CETP activity can be regulated by many other plasma
factors,9 and the lack of significance of the correlation
of NEFA levels with specific transfer activities does not rule out the
potential role of NEFA in modulating CETP activity. Depletion of NEFA
content of lipoproteins in the presence of fatty acidfree
albumin sustains that latter view. Indeed, since in vivo
albumin binds the major part of plasma NEFA,42 60
it is likely to play a determinant role in modulating the interactions
between NEFA and CETP in plasma. In agreement with observations of
Shafrir,42 we demonstrated that, despite the fact that
plasma NEFAs are mainly bound to albumin, a small fraction is
normally carried by lipoprotein particles. To confirm that this latter
fraction is available for interaction with CETP, we studied the effect
of the supplementation of plasma with fatty acidfree
albumin, an experimental protocol that has been shown
previously to remove NEFA from plasma lipoprotein
particles.42 When total plasmas were supplemented with
albumin (20% increase in plasma albumin content), a
significant reduction in CETHDL
VLDL+LDL activity was
observed. Under identical experimental conditions, we reported that the
positive relation between plasma CETP mass and CETP activity was
improved in the presence of albumin, indicating further that
plasma NEFA can modulate the activity of CETP.35 Taken
together, these observations brought strong functional arguments in
favor of the role of lipoprotein-bound NEFA in the modulation of
CET activity in normolipidemic plasma. The effect of albumin
supplementation is more likely to relate to its ability to bind NEFA
rather than to a putative, direct effect on the transfer reaction.
Indeed, as demonstrated in previous studies,22 23 24
albumin supplementation abolished the ability of NEFA to
modulate the CETP-mediated exchange of cholesteryl esters between
isolated lipoprotein substrates. On the contrary, in the absence of
addition of NEFA, albumin alone did not affect the
CETP-mediated CET reaction.24 Since the inhibition of CET
by LTIP would concern mainly VLDL-LDL exchange,61
resulting in a concomitantly greater exchange between VLDL and HDL, one
can speculate that NEFAs might also affect LTIP activity in addition to
their ability to modulate directly the interaction of CETP with
lipoprotein substrates. Whether that mechanism might account in part
for differences in the relations between plasma NEFA concentrations and
either CETHDL
VLDL+LDL or CETLDL
HDL
activities remains to be established.
In support of a role of lipoprotein-bound NEFA in modulating CETs
in total native plasma, correlations observed between CETP activity and
lipoprotein concentrations were mostly suppressed in the presence of an
excess of albumin. Indeed, after albumin
supplementation, CETLDL
HDL activity no longer correlated
with plasma HDL and VLDL+LDL cholesterol levels. Similarly,
the correlation of CETHDL
VLDL+LDL activity with plasma
HDL cholesterol levels was suppressed. Only a significant
positive correlation of CETHDL
VLDL+LDL activity with
plasma VLDL+LDL cholesterol levels was still apparent in
the presence of albumin. These results indicate that
lipoprotein-bound NEFA may account in part for the significant
correlations between plasma CETP activity and the concentration of
various lipoprotein fractions. However, in accordance with previous
studies,2 48 49 58 the VLDL+LDL concentration remains a
strong, independent determinant of plasma neutral lipid transfers, even
when the modulating effect of lipoprotein-bound NEFA is
cancelled.
Since under certain circumstances CETP can catalyze the rapid exchange of cholesteryl esters between lipoprotein particles without promoting any net mass transfer,62 we sought after the ability of plasma NEFA to alter the CETP-mediated mass redistribution of cholesteryl esters and triglycerides between plasma lipoproteins. To this end, HDL cholesteryl ester and triglyceride masses were measured in incubated plasmas, supplemented or not with albumin. Long-term incubation of native, nonsupplemented plasmas induced the transfer of cholesteryl esters to VLDL+LDL with a reciprocal transfer of triglycerides to HDL. This observation is consistent with the existence of an interdependence between cholesteryl ester and triglyceride transfer in plasma. However, results obtained after incubation of plasmas in the presence of albumin compared with plasmas without supplementation indicated that NEFA increased the mass of cholesteryl esters transferred from HDL to apo Bcontaining lipoproteins but, simultaneously, decreased the mass of triglycerides transferred from apo Bcontaining lipoproteins to HDL. These observations are consistent with a recent study30 that demonstrated that NEFA can dissociate the CETP-mediated heteroexchange of cholesteryl esters and triacylglycerol between HDL and triacylglycerol-rich lipoproteins. Our data indicated further that such an effect of NEFA may occur in native plasma.
Taken all together, the results of the present study suggest that plasma NEFA can modulate neutral lipid transfers through a dual effect on both the level and the activity of CETP. By increasing CETP mass concentrations and dissociating both the homoexchange of cholesteryl esters and the heteroexchange of cholesteryl esters and triglycerides between HDL and VLDL+LDL fractions, NEFAs globally favor the redistribution of neutral lipids toward plasma apo Bcontaining lipoproteins. The control of neutral lipid transfers by plasma NEFA could have some important implications, in particular under physiopathological conditions associated with abnormally low plasma albumin concentrations and high NEFA content of plasma lipoproteins, ie, analbuminemia63 and nephrotic syndrome.42 In support of that latter view, increased ratios of plasma NEFA to albumin have also been observed in patients with coronary heart disease.64 Whether these observations relate at least in part to the ability of NEFA to enhance the CETP-mediated transfer of cholesteryl esters from the "anti-atherogenic" HDL to the "atherogenic" VLDL and LDL fractions remains to be clarified.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
Received March 7, 1995; accepted June 26, 1995.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. E. Morton and D. J. Greene CETP and lipid transfer inhibitor protein are uniquely affected by the negative charge density of the lipid and protein domains of LDL J. Lipid Res., December 1, 2003; 44(12): 2287 - 2296. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. J. A. M. Jonkers, A. H. M. Smelt, H. Hattori, L. M. Scheek, T. van Gent, F. H. A. F. de Man, A. van der Laarse, and A. van Tol Decreased PLTP mass but elevated PLTP activity linked to insulin resistance in HTG: effects of bezafibrate therapy J. Lipid Res., August 1, 2003; 44(8): 1462 - 1469. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. V. Sekhar, F. Jahoor, A. C. White, H. J. Pownall, F. Visnegarwala, M. C. Rodriguez-Barradas, M. Sharma, P. J. Reeds, and A. Balasubramanyam Metabolic basis of HIV-lipodystrophy syndrome Am J Physiol Endocrinol Metab, August 1, 2002; 283(2): E332 - E337. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Gander, P. Eller, S. Kaser, I. Theurl, D. Walter, T. Sauper, A. Ritsch, J. R. Patsch, and B. Foger Molecular characterization of rabbit phospholipid transfer protein: choroid plexus and ependyma synthesize high levels of phospholipid transfer protein J. Lipid Res., April 1, 2002; 43(4): 636 - 645. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Guerin, W. Le Goff, T. S. Lassel, A. Van Tol, G. Steiner, and M. J. Chapman Proatherogenic Role of Elevated CE Transfer From HDL to VLDL1 and Dense LDL in Type 2 Diabetes : Impact of the Degree of Triglyceridemia Arterioscler. Thromb. Vasc. Biol., February 1, 2001; 21(2): 282 - 288. |