Altered Transfer of Cholesteryl Esters and Phospholipids in Plasma From Alcohol Abusers
Abstract The net mass transfer (NMT) of cholesteryl esters (CEs), triglycerides (TGs), and phospholipids (PLs) between lipoproteins was measured after incubation of fresh plasma for up to 2 hours from 18 male alcohol abusers and 17 male volunteer control subjects. In alcohol abusers the mean value of CE NMT was 3.7 nmol·mL−1·h−1 from apolipoprotein B–containing lipoproteins (apoB-containing lipoproteins) to HDL and in control subjects 8.7 nmol · mL−1 · h−1 from HDL to apoB-containing lipoproteins. The NMT of PL was higher in alcohol abusers than in control subjects (35.0 vs 11.6 nmol·mL−1·h−1 from apoB-containing lipoproteins to HDL, respectively), and plasma PL transfer protein (TP) activity was 33% higher (P<.05) in alcohol abusers than in control subjects. The lack of correlation between the NMTs and CETP and PLTP activities suggests that the NMT could more closely reflect the role of lipoprotein properties in reverse cholesterol transport in vivo, whereas in vitro activities reflect the total capacity of transfer but not its direction. The rate of CE NMT from HDL to apoB-containing lipoproteins was dependent on the VLDL TG concentration. Moreover, at low VLDL TG levels, the increased HDL cholesterol concentration in alcohol abusers reversed the direction of CE NMT. This situation could be reconstructed in the plasma of control subjects by adding autologous HDL or VLDL to mimic the lipoprotein profiles of the alcohol abusers. Addition of VLDL enhanced the CE NMT from HDL to apoB-containing lipoproteins, whereas addition of HDL had an opposite effect, and at higher HDL levels, even reversed the direction of CE NMT. In conclusion, the NMT of CE and PL in alcohol abusers differs from that in control subjects. The concentrations of HDL and VLDL seem to be the major determinants of the direction of CE NMT in alcohol abusers.
- cholesteryl ester transfer protein
- reverse cholesterol transport
- phospholipid transfer protein
- Received January 3, 1997.
- Accepted May 30, 1997.
The two lipid-transfer proteins responsible for lipid-transfer reactions between lipoprotein classes in human plasma are CETP and PLTP. CETP facilitates the transfer of CEs, TGs, and to some extent, PLs between lipoproteins.1 CETP activity in normolipidemic subjects results in an NMT of CEs from HDL to apoB-containing lipoproteins (ie, VLDL and LDL) with a reciprocal NMT of TGs to HDL.2 CETP may play a role in reverse cholesterol transport, a process in which unesterified cholesterol is removed from peripheral tissues by HDL, then esterified by LCAT, transferred to apoB-containing lipoproteins by CETP, and finally taken up in the liver by apoB/E and apoE receptors.3 On the other hand, because apoB-containing lipoproteins may carry cholesterol to peripheral tissues, the role of CETP in reverse cholesterol transport remains an open question, and CETP may not be essential for reverse cholesterol transport.3 PLTP catalyzes two thirds of the plasma PL transfer between lipoprotein classes, but it does not catalyze the transfer of neutral lipids, ie, CEs and TGs.4 PLTP could play an important role in the catabolism of TG-rich lipoproteins because it transfers the surface PLs of chylomicrons to HDL.5
Alcohol consumption is known to be associated with an elevated plasma HDL cholesterol level.6 A negative correlation between moderate alcohol consumption and coronary heart disease has also been observed.7 The mechanisms by which alcohol elevates HDL cholesterol remain unclear, but at least part of it may be mediated by a reduction in plasma CETP activity8 . Because alcohol reduces plasma CETP activity and mass, it might inhibit reverse cholesterol transport.
The higher HDL concentration in alcohol drinkers may influence lipid-transfer reactions and reverse cholesterol transport. Several studies indicate that lipoprotein substrate concentrations and their lipid contents influence lipid-transfer reactions. For example, supplementation of normolipidemic plasma with autologous VLDL increases the CE NMT from HDL to VLDL,11 whereas an increase in HDL concentration decreases CE transfer.12 13 Studies of HDL enriched with phosphatidylcholine14 15 have confirmed an increase in CE transfer, most probably due to increased binding of CETP to HDL. Furthermore, unesterified cholesterol enrichment of plasma lipoproteins stimulates CE transfer from HDL to VLDL and reciprocal transfer of TGs from VLDL to HDL.16
The purpose of this study was to elucidate the effect of alcohol on reverse cholesterol transport by investigating the NMT of CEs and other lipids as well as the effects of lipoprotein concentrations on transfer reactions.
The study was carried out in 22 male alcohol abusers who had been referred by their general practitioners and were admitted as inpatients to the Alcoholism Treatment Unit in Oulu for withdrawal therapy. This unit treats ambulatory patients seeking assistance for terminating their dependence on alcohol and having no signs or symptoms of other diseases. Patients who were acutely ill with alcohol-related or other diseases were referred to other clinics. All 22 alcohol drinkers had been drinking daily for at least 1 week before admission. The subjects were included if their nutritional status was sound and they did not have any clinical signs of kidney or heart dysfunction or severe liver damage, such as spiders, jaundice, or hepatomegaly. Four of the 22 subjects were excluded on the basis of highly elevated liver enzymes (exclusion criteria were alanine aminotransferase values >160 U/L, alkaline phosphatase >750 U/L, or gamma glutamyltransferase >1200 U/L), leaving a total of 18 study subjects.
Through an extensive interview by a trained physician, alcohol intake among both control and alcohol-abusing subjects was assessed: the amount of beer, wine, and strong alcoholic beverages (distilled) consumed during the previous 2 weeks was ascertained and recorded. Alcohol intake was calculated and expressed in grams of pure alcohol per day. Information was also obtained about smoking habits (pipe, cigar, and cigarettes), drug abuse, and past medical history. None of the alcohol abusers were skid row alcoholics or dependent on narcotics. None of the alcohol abusers or control subjects used prescription drugs or had diseases that might influence lipid metabolism.
Seventeen healthy men volunteered as control subjects. They were interviewed in the same manner as the alcohol abusers and had similar inclusion/exclusion criteria, except for alcohol drinking. Six of these men were teetotalers and the others reported occasional, moderate alcohol intake. The study was approved by the Ethics Committee of the University of Oulu.
Venous blood samples were drawn into EDTA-containing evacuated tubes after an overnight fast from both the alcohol abusers and control subjects. Blood samples from the alcohol abusers were taken on the first day after admission into the treatment unit. The tubes were immediately chilled on ice and centrifuged at 2°C to separate the plasma. The plasma was kept on ice and the experiments immediately started after plasma separation. The samples for CETP and PLTP activity were stored at −70°C.
Cholesterol Esterification and Lipid NMT Reactions
Five samples of each subject’s plasma were incubated at 37°C for 0, 0.5, 1, 1.5, and 2 hours according to the method of Fielding et al.17 All incubations were performed using fresh plasma within 2 hours after venipuncture. After incubation, aliquots of plasma were mixed with a one-tenth volume of 2 mol/L MgCl2 and dextran sulfate (1:1, vol/vol) to precipitate the apoB-containing lipoproteins.18 The supernatant was then separated and kept on ice until measurements of total and free cholesterol, TGs, and PLs were made by enzymatic colorimetric methods. The CE concentrations were calculated as the difference between total and free cholesterol concentrations.
The NMT of CEs was calculated as the rate of change in CEs in apoB-containing lipoproteins and cholesterol esterification as the rate of change in total plasma CEs. In addition, the NMTs of TGs and PLs were measured in 9 alcohol abusers and 10 control subjects and calculated as the rate of change in TGs or PLs in apoB-containing lipoproteins. Thus, the NMT of lipids from HDL to VLDL and LDL has a positive sign and the NMT from VLDL and LDL to HDL a negative sign.
The CV was 0.8% for total plasma CEs and 1.5% for apoB-containing lipoprotein CEs. CVs for CE NMT were 23.8% for very high negative transfer rates (<−100 nmol · mL−1 · h−1), 36.8% for low negative esterification rates (−20 to −60 nmol · mL−1 · h−1), 30.0% for low positive transfer rates (20 to 60 nmol · mL−1 · h−1), and 35.2% for high positive transfer rates (≥60 nmol · mL−1 · h−1). For PL and TG NMTs, the CVs were similar to those of CE NMT. For cholesterol esterification, the CVs were 13.3% for high rates of esterification (≥80 nmol · mL−1 · h−1) and 16.5% for medium rates of esterification (20 to 80 nmol · mL−1 · h−1). The NMT analysis was performed twice on 2 different days for three control plasma samples. The SD for these measurements was 8.7 nmol · mL−1 · h−1 for CE NMT, 26.1 nmol · mL−1 · h−1 for TG NMT, 10.8 nmol · mL−1 · h−1 for PL NMT, and 6.1 nmol · mL−1 · h−1 for cholesterol esterification. In some experiments, LCAT activity was inhibited by adding 1.5 mmol/L iodoacetic acid prior to incubation. No major effect of LCAT inhibition on lipid NMTs was seen in either the alcohol abusers or control subjects.
Experiments With Altered HDL Concentrations
The effect of HDL cholesterol concentration on NMT reactions was studied by raising the HDL cholesterol concentration by a factor of 2 or 3 with respect to the original level in the plasma. The HDL fraction (d=1.090 to 1.21 g/mL) was isolated by ultracentrifugation from 130 mL of a control subject’s plasma. After dialysis against 0.15 mol/L NaCl–0.01% EDTA, pH 7.4, at 4°C for at least 3 hours, the cholesterol content in HDL was measured, and an aliquot of the HDL fraction was added to the same subject’s plasma sample to increase its HDL cholesterol level twofold or threefold. The addition of HDL to the plasma diluted it by as much as 25%. The possible effect of dilution on NMTs was tested by adding 0.15 mol/L NaCl–0.01% EDTA, pH 7.4, to the plasma in a volume ratio of 1:3. In the plasma from these three control subjects, no change in NMTs was observed (data not shown).
Experiments With Altered VLDL Concentrations
The effect of VLDL concentration on the NMT of lipids was studied by removing VLDL from the plasma by ultracentrifugation or by raising the VLDL cholesterol concentration twofold or threefold with respect to the original plasma VLDL cholesterol level. The VLDL fraction (d<1.006 g/mL) was isolated by ultracentrifugation and added to the same subject’s plasma sample to increase its VLDL concentration by a factor of 2 or 3. Experiments with VLDL-free plasma obtained from the VLDL isolation procedure were also performed for the same three control subjects. Because addition of VLDL to plasma caused some dilution, we added the same volume of 0.15 mol/L NaCl–0.01% EDTA, pH 7.4, to the original plasma as that of VLDL. VLDL-free plasma was also diluted in the same manner, and therefore in all experiments with VLDL addition, the concentrations of lipoproteins other than VLDL remained the same.
Total plasma cholesterol, free cholesterol, TG, and PL concentrations were determined by enzymatic colorimetric methods using a Kone specific selective chemistry analyzer (Kone Oy) and kits from Boehringer Diagnostica (catalogue Nos. 236691, 310328, and 701912) and Wako Chemicals (No. 990-54009), respectively. CVs for the determination of plasma total cholesterol, free cholesterol, TGs, and PLs were 2.1±1.3%, 2.4±1.3%, 5.3±3.1%, and 0.9±0.6%, respectively.
To analyze VLDL and LDL concentrations separately, 1 mL of the VLDL-free fraction isolated by ultracentrifugation (as described above) was mixed with 25 μL of 2.8% (wt/vol) heparin and 25 μL of 2 mol/L MnCl2 and centrifuged at 1000g and 4°C for 30 minutes. Aliquots of the supernatant were taken for lipid analysis. The LDL lipid concentrations were calculated by subtracting the cholesterol concentration in HDL from that in the VLDL-free fraction, and the VLDL lipid concentrations were analyzed in the VLDL fraction.
Determination of CETP Activity
CETP activity was determined in (VLDL+LDL)-free plasma by a modification of the method of Groener et al19 as described earlier8 9 10 by detecting the exchange of radioactive CEs between labeled LDL and unlabeled HDL, both isolated from control subjects. This method reflects the CETP protein concentration.9
Determination of PLTP Activity
PLTP activity was determined with radiolabeled phosphatidylcholine–containing liposomes as described earlier20 using 10 μmol of egg phosphatidylcholine (Sigma), 1 μCi of [14C]dipalmitoylphosphatidylcholine (Amersham), and 20 nmol BHT (Sigma). The assays were performed as described earlier21 in a volume of 400 μL containing HDL3 (250 μg of protein); liposomes (150 nmol of [14C]phosphatidylcholine-labeled liposomes); buffer containing 150 mmol/L NaCl, 10 mmol/L Tris, and 1 mmol/L EDTA, pH 7.4; and a plasma sample (6.0 μL diluted 1:10 in buffer). “Blank” tubes (with all of the aforementioned constituents except the plasma sample), “control” tubes (containing plasma from a healthy volunteer), and “total” tubes (without precipitation of liposomes) were included in each series. The tubes were incubated in a water bath at 37°C for 1.5 hours. The reaction was stopped by adding 300 μL of 500 mmol/L NaCl, 215 mmol/L MnCl2, and 140 U heparin. The tubes were vortexed and, after allowing them to stand for 10 minutes at room temperature, centrifuged for 10 minutes at 13 000 rpm. Aliquots of 500 μL were taken from the supernatant for determination of radioactivity. The sample volume and incubation time were chosen so that they were within the linear range of the volume and incubation time curves (the reaction was linear from 0.5 to 2 hours and from 0.1 to 1.0 μL in both plasma and lipoprotein-free plasma from the alcohol abusers and control subjects).
The results are given as mean±SEM. Differences between means were calculated with Student’s t test or ANOVA using the spss program. The Levene test was used to explore the distribution of CE NMT data between the alcohol abusers and control subjects. The rate of cholesterol esterification and NMTs were estimated by linear regression. Because total plasma cholesterol concentrations were assumed to remain constant during these short incubations, total cholesterol values were used at each time point to correct the other lipid values for potential changes caused by evaporation, condensation, or drift in lipid determinations. At any time point, the total cholesterol concentration was within 97.5% to 102.5% of the mean during the incubation.
For analyzing the effects of lipoprotein concentrations on NMT rates, we used the natural logarithmic of the reciprocal of the variance of the regression coefficient to weight the variables and minimize the effects of the less exact determinations of NMT. These results did not differ from those calculated using unweighted NMT values. The differences between regression lines were calculated using the Confidence Interval Analysis program (based on the methods of Gardner and Altman).22
Clinical Characteristics and Lipid Values
Clinical characteristics of the alcohol abusers and control subjects are shown in Table 1⇓. The mean daily alcohol consumption was 189 g in the alcohol abuser group and 7 g in the control group. Seventy-four percent of alcohol abusers and 6% of control subjects were smokers (data not shown). Mean age and body mass index were almost identical between the alcohol abusers and control subjects. Liver enzyme levels were higher in the alcohol abusers (Table 1⇓). The concentrations of HDL cholesterol, HDL free cholesterol, HDL CEs, and HDL PLs, as well as total plasma TGs, apoB-containing lipoprotein TGs, and total PLs, were significantly higher in the alcohol abusers than the control subjects (Table 2⇓).
The rate of cholesterol esterification in plasma, calculated as the rate of change in plasma CE concentration, was somewhat but not significantly higher in the alcohol abusers than control subjects (53.9±7.3 nmol · mL−1 · h−1 versus 39.4±5.6 nmol · mL−1 · h−1, respectively, P=.13; Fig 1⇓).
NMT of CEs Between Lipoproteins
The mean value of CE transfer was −3.7 nmol · mL−1 · h−1 (ie, from apoB-containing lipoproteins to HDL) in the alcohol abusers and 8.7 nmol · mL−1 · h−1 (ie, from HDL to apoB-containing lipoproteins) in the control subjects (P=NS; Fig 1⇑). Both groups included individuals who had values for CE transfer from HDL to apoB-containing lipoproteins and individuals whose values were in the opposite direction (n=11 and 7, respectively, in the alcohol abusers versus 12 and 5, respectively, in control subjects). However, the range of CE transfer values was larger in the alcohol abusers (from −141.7 to 82.4 nmol · mL−1 · h−1) than in control subjects (from −65.9 to 57.4 nmol · mL−1 · h−1). The variances were statistically significantly different between the alcohol abusers and control subjects when tested with the Levene test (P<.001). The amount of alcohol consumed seemed to have no effect on CE NMT among the alcohol abusers or control subjects (Fig 2⇓).
NMT of TGs Between Lipoproteins
Alcohol abuse did not influence the TG NMT. The rate and direction of TG NMT were similar in the alcohol abusers and control subjects (mean value of −20.1 versus −19.6 nmol · mL−1 · h−1 from apoB-containing lipoproteins to HDL, respectively; Fig 1⇑). During incubation of the plasma, lipolysis of TG-rich particles could liberate free glycerol, which would appear in the supernatant after precipitation and lead to erroneously high NMT rates from apoB-containing lipoproteins to HDL. To control for this possibility, we measured HDL TG concentration in two samples after removal of free glycerol by dialysis. The rate of lipolysis (ie, the decrease in TG concentration in the total plasma sample during incubation) was negligible compared with the TG NMT (calculated as the rate of change in TGs in apoB-containing lipoproteins).
NMT of PLs Between Lipoproteins
Alcohol abuse had a marked impact on the PL NMT. The direction of PL NMT in both the alcohol abusers and control subjects was from apoB-containing lipoproteins to HDL, but the NMT rate in the alcohol abusers was almost fourfold than that in the control subjects (mean value of −39.2 versus −11.6 nmol · mL−1 · h−1, respectively; Fig 1⇑). The PL NMT was correlated with cholesterol esterification in the alcohol abusers (r=.66, P=.05) but not in the control subjects (data not shown). Enzymatic or nonenzymatic degradation of PLs (phosphatidylcholine, lysophosphatidylcholine, or sphingomyelin) during incubation could lead to the appearance of free choline in the supernatant after the dextran-sulfate precipitation step, and the presence of free choline would result in falsely elevated levels of HDL PLs in the choline oxidase–based determination of PLs and in an apparent NMT of PLs from apoB-containing lipoproteins to HDL. This possibility was ruled out by dialysis of two samples, which showed that free choline was not detected after incubation and that PL degradation therefore had no significant effect on the determination of PL NMT.
CETP and PLTP Activities and Their Correlations With Corresponding Lipid NMTs
We made a novel observation on PLTP activity in alcohol abusers. The alcohol abusers had a 33% higher PLTP activity than the control subjects (P<.05) in this sample of 18 alcohol abusers and 17 control subjects (Table 3⇓). We also determined the PLTP activity in 17 alcohol abusers and 19 control subjects from a previous study,10 and those results confirmed the higher PLTP activity in alcohol abusers (26% higher than that of control subjects, P<.05). In the total series of 35 alcohol abusers and 36 control subjects from both studies combined, PLTP activity was 29% higher in alcohol abusers than in control subjects (P=.001). However, no significant correlation was observed between PLTP activity and the NMT of PLs, although a trend was observed in control subjects (r=−.61, P=.062 for control subjects and r=−.38, P=.31 for alcohol abusers). CETP activity was 31% lower in the alcohol abusers than in the control subjects (P<.001; Table 3⇓), but no significant correlation was observed between CETP activity and CE NMT, although a trend was observed in the alcohol abusers (r=.21, P=.42 for control subjects and r=.42, P=.08 for the alcohol abusers).
Correlations Between Lipids and Lipid NMTs
To investigate the role of major alcohol-induced lipoprotein changes in the lipid NMT reactions of the alcohol abusers, we plotted HDL cholesterol and VLDL TGs against the lipid NMTs and cholesterol esterification. A strong, negative correlation was found between HDL cholesterol and CE NMT as well as between HDL cholesterol and cholesterol esterification in the alcohol abusers (r=−.667, P<.01 and r=−.736, P<.001, respectively; Table 4⇓), suggesting that higher HDL cholesterol concentrations could inhibit cholesterol esterification and the NMT of CEs to apoB-containing lipoproteins. A negative correlation was also seen between HDL cholesterol and PL NMT in the alcohol abusers (r=−.767, P<.05), whereas the control subjects showed an opposite trend (Table 4⇓). VLDL TGs had a strong, positive correlation with the CE NMT and cholesterol esterification in the alcohol abusers (r=.574, P<.05 and r=.657, P<.01, respectively) and with cholesterol esterification in the control subjects (r=.567, P<.05) (Table 4⇓), indicating the key role of VLDL TGs in determining the direction of CE NMT.
We used linear regression to evaluate the impact of key lipid parameters on lipid NMT and cholesterol esterification. HDL cholesterol was the sole significant predictor of CE NMT and cholesterol esterification in the alcohol abusers, accounting for 44% of the CE NMT and 54% of cholesterol esterification. HDL cholesterol and VLDL TGs together explained 95% of the PL NMT in the alcohol abusers, although HDL cholesterol was a more important predictor for PL NMT. Interestingly, VLDL TGs accounted for 35% of the variation in cholesterol esterification in the control subjects, but the lipid NMTs were not explained by the key lipid parameters in the control subjects.
Effect of HDL Concentrations on Lipid NMTs
The results shown in Table 4⇑ point to the role of HDL cholesterol as the determinant of the rate or direction of lipid NMT. Therefore, we further studied the role of ethanol-induced HDL changes by increasing the number of HDL particles in the plasma samples of control subjects (Fig 3⇓). Our aim was to test whether the effect of high HDL levels on lipid NMTs observed in the plasma from alcohol abusers could also be reproduced in the control subjects’ plasma by increasing its HDL level. Addition of an isolated HDL fraction into control plasma before incubation at a twofold or threefold higher concentration reversed the direction of CE NMT in the control subjects, so that the direction of NMT was from apoB-containing lipoproteins to HDL (P<.01) under conditions of increased HDL, similar to what we observed in the alcohol abusers (Fig 3A⇓). Cholesterol esterification and the TG and PL NMTs were not significantly affected by the increased HDL concentration (Fig 3A⇓ and 3B⇓).
Because the HDL concentration seemed to be the most important determinant of CE NMT, we plotted the HDL concentration–predicted CE NMT against the observed NMT. The equation calculated from this HDL increase–affected reverse in the direction of CE NMT (Fig 3⇑) was used to calculate the predicted values. Fig 4⇓ shows how the predicted CE NMTs were correlated with those observed. The correlation coefficients between predicted and observed NMTs were .67 (P<.01) for the alcohol abusers, .11 (P=.67) for control subjects, and .53 (P<.001) for both the alcohol abuser and control groups together.
Effect of VLDL Concentration
An increase in VLDL TGs is another ethanol-induced disturbance in lipoproteins. An approach similar to that described for HDL was adopted to investigate the effect of VLDL concentration on lipid NMT reactions (Fig 5⇓). The CE (Fig 5A⇓) and TG (Fig 5B⇓) NMT rates in VLDL-depleted plasma were found to be almost undetectable and cholesterol esterification was slower than in plasma, whereas the PL NMT (Fig 5A⇓) was not affected by changes in VLDL concentrations. Twofold or threefold higher concentrations of VLDL further increased the CE NMT (Fig 5A⇓) from HDL to apoB-containing lipoproteins (P<.05) but slowed the cholesterol esterification rate (Fig 5B⇓), indicating the role of VLDL in determining the rate and direction of CE NMT.
Because the low plasma CETP activity in alcohol drinkers9 might reduce the rate of reverse cholesterol transport, we studied the NMTs of CEs, TGs, and PLs between lipoproteins in alcohol abusers. The present study shows that alcohol drinking may affect the CE NMT through its effect on VLDL and HDL concentrations. Furthermore, the alcohol abusers had a higher rate of PL NMT than the control subjects.
The mean value of CE NMT in the present control subjects is in accordance with the postulated CE transfer step of reverse cholesterol transport and with previous NMT studies,2 11 although the NMT of CEs toward HDL has also been reported.23 Previously, van Tol et al24 described induction of the NMT of CEs and TGs by moderate wine consumption with dinner. The difference between their results and our present findings may be due to the amount of alcohol consumed, since moderate consumption of alcohol by social drinkers may have a more marked effect on VLDL than on HDL and the direction of NMT may be therefore from HDL to apoB-containing lipoproteins.
The correlations observed between HDL cholesterol or VLDL TGs and CE NMT or cholesterol esterification point to the role of HDL and VLDL concentrations as determinants of the CE NMT process in alcohol abusers. The importance of VLDL in CE transfer has been previously reported,11 25 and an effect of HDL has also been described for normal subjects.12 13 We added autologous lipoprotein fractions to the plasma sample of the control subjects to mimic the lipoprotein profiles seen in the alcohol abusers. The addition of VLDL enhanced the CE NMT from HDL to apoB-containing lipoproteins, whereas addition of HDL had an opposite effect, and higher HDL levels even reversed the direction of CE NMT, which means that control subjects with increased HDL concentrations mimicked the alcohol abusers. Our results indicate that lipoprotein concentrations play a key role in the regulation of CE NMT, although alcohol-induced changes in lipoprotein composition, ie, the higher TG content in apoB-containing lipoproteins or higher HDL CE content,26 may also affect lipid NMTs, a question that is presently being studied in our laboratory. Moreover, the alcohol-induced increase in HDL may be a self-promoting process, ie, the higher the HDL cholesterol level, the more CEs are transferred from apoB-containing lipoproteins to HDL. Thus, CETP may be less important than lipoprotein concentrations per se, and it seems that the low concentration of CETP may not be rate limiting in alcohol abusers despite their high VLDL TG levels. This finding is in contrast with previous results, which showed that a normal CETP level may not be sufficient in the presence of a very high VLDL TG concentration.11
Our study is the first in which PL NMT has been determined in both alcohol abusers and control subjects. In the present study, the PL NMT was almost fourfold higher in the alcohol abusers than in the control subjects, and the NMT rate was surprisingly high in both groups. Our observations agree with recent studies in which a PL NMT rate of 27.2 nmol · mL−1 · h−1 was demonstrated in healthy volunteers27 and a high rate of PL exchange was observed in an isotopic assay.28 In accordance with a recent study, in which Lagrost et al29 observed a reduction in PLTP activity by alcohol withdrawal, our present study demonstrates for the first time that alcohol abusers have higher plasma PLTP activity than do control subjects. Moreover, the NMT rate of PLs from apoB-containing lipoproteins to HDL was increased, especially in the alcohol abusers with a high NMT rate of CEs in the same direction. The increased CE content in the core of HDL particles due to the increased NMT would require a larger surface area, which in turn could create a demand for surface lipids, including PLs. In alcohol abusers with higher cholesterol esterification rates, more (unesterified) cholesterol may move from the surface to the core during esterification, further increasing the requirement for surface components to be transferred from other lipoprotein particles. The results suggest that more attention should be given to the role of PLs to reveal whether they are nonparticipants or have a more active role in lipid metabolism than has been previously assumed.
Severe liver damage might affect the lipid NMTs and cholesterol esterification in alcohol abusers.30 31 In this study, the cholesterol esterification rate was higher in alcohol abusers than in the control subjects, indicating that our exclusion of alcohol abusers with liver failure was successful. The higher LCAT activity in alcohol abusers might be due to induction of LCAT synthesis in the liver.32 In accordance with a previous study,33 a strong correlation was found between cholesterol esterification and CE NMT. Because both LCAT and CETP are present in the plasma in vivo, inhibition of LCAT activity in the NMT analysis could remove an essential step in reverse cholesterol transport, and since the aim of this study was to investigate the events of reverse cholesterol transport in an experimental setting resembling the in vivo situation as closely as possible, we did not inhibit LCAT activity.17 It must be borne in mind, however, that the first (supply of premature HDL particles with low CE content) and the last (uptake of HDL particles or selective uptake of CEs) steps of reverse cholesterol transport were absent even in these ex vivo incubation experiments, although experiments of this kind are often referred to as in vivo experiments. More complicated methods, such as the one described by Mindham et al,34 35 may be required to completely reconstruct reverse cholesterol transport in an experimental setting.
In the present study, it was not possible to evaluate the roles of VLDL and LDL separately in the NMT process, since the MgCl2–dextran sulfate precipitation used in this study does not separate VLDL from LDL. Previous studies with separation of LDL from VLDL revealed an NMT of CEs from HDL and LDL to VLDL and a reciprocal NMT of TGs.25 36 . Because VLDL appears to be the major acceptor of CEs and donor of TGs, the effect of pooled VLDL and LDL on the results of the present study may not be significant. Further studies may be required to confirm this, however, since ethanol affects the binding of CETP to various lipoprotein fractions by increasing the proportion of plasma CETP bound to VLDL, at least in experiments with perfused rabbit livers.39
The previously described lack of correlation between CETP activity and CE NMT may be due to the basic differences between the two methods widely used for evaluation of CETP function.23 11 CETP activity is determined by measuring the exchange of radiolabeled CEs between an excess of exogenous lipoproteins,19 which gives an accurate estimate of CETP protein concentration.9 The ex vivo NMT analysis is performed under more physiological conditions with endogenous lipoproteins, and in contrast to the CETP activity assay, also gives the direction of transfer. However, in spite of these methodological reservations, the CE NMT measured in the ex vivo plasma samples could more accurately reflect the role of lipoprotein particles in reverse cholesterol transport than does CETP activity, which reflects only the total capacity of transfer but not its direction. Similar methods are used for the determination of PLTP activity20 and PL NMT, and no correlation was observed between the two methods of PL transfer. Therefore, with the same kind of methodological reservations as for CETP, the two methods can be used for the determination of PL transfer activity.
In conclusion, transfer of CEs and PLs is altered in the alcohol abuser. Plasma PLTP activity is higher and CETP activity lower in alcohol abusers than in control subjects, but neither CETP activity nor PLTP activity are correlated with their corresponding lipid NMTs. The rate and direction of CE transfer may be predominantly due to alcohol-induced changes in HDL and VLDL concentrations rather than to the effects of alcohol on CETP itself.
Selected Abbreviations and Acronyms
|CETP||=||CE transfer protein|
|NMT||=||net mass transfer|
|PLTP||=||PL transfer protein|
This work was supported by the Finnish Foundation of Alcohol Studies (to M.J.L., M.J.S.), the Finnish Foundation for Cardiovascular Research (to M.J.L., M.L.H., M.J.S.), the Academy of Finland (all authors), and the Deaconess Institute of Oulu (M.J.L.). We wish to thank Drs Riitta Heikkilä and Marita Paassilta for their help in collecting the blood samples. We are also very grateful to the staff of the Kiviharju Alcoholism Treatment Unit for their cooperation. The skillful technical assistance of Saija Kortetjärvi and Eila Saarikoski is greatly appreciated.
Hesler CB, Tall AR, Swenson TL, Weech PK, Marcel YL, Milne RW. Monoclonal antibodies to the Mr 74,000 cholesteryl ester transfer protein neutralize all the cholesteryl ester and triglyceride transfer activities in human plasma. J Biol Chem. 1988;263:5020-5023.
Chajek T, Fielding CJ. Isolation and characterization of a human serum cholesteryl ester transfer protein. Proc Natl Acad Sci U S A. 1978;75:3445-3449.
Fielding CJ, Fielding PE. Molecular physiology of reverse cholesterol transport. J Lipid Res. 1995;36:211-228.
Tall AR, Abreu E, Shuman J. Separation of a plasma phospholipid transfer protein from cholesterol ester/phospholipid exchange protein. J Biol Chem. 1983;258:2174-2180.
Tall AR. Studies on the transfer of phosphatidylcholine from unilamellar vesicles into plasma high density lipoproteins in rat. J Lipid Res. 1980;21:354-363.
Hannuksela M, Marcel YL, Kesäniemi YA, Savolainen MJ. Reduction in the concentration and activity of plasma cholesteryl ester transfer protein by alcohol. J Lipid Res. 1992;33:737-744.
Mann CJ, Yen FT, Grant AM, Bihain BE. Mechanism for plasma cholesteryl ester transfer in hyperlipidemia. J Clin Invest. 1991:88:2059-2066.
Lagrost L, Gandjini H, Athias A, Guyard-Dangremont V, Lallemant C, Gambert P. Influence of plasma cholesteryl ester transfer activity on the LDL and HDL distribution profiles in normolipidemic subjects. Arterioscler Thromb. 1993;13:815-825.
Tollefson JH, Liu A, Albers JJ. Regulation of plasma lipid transfer by the high-density lipoproteins. Am J Physiol. 1988;255:E894–E902.
Sammett D, Tall AR. Mechanism of enhancement of cholesteryl ester transfer protein activity by lipolysis. J Biol Chem. 1985;260:6687-6697.
Tall AR, Granot E, Brocia RE, Tabas I, Hesler C, Williams K, Denke M. Accelerated transfer of cholesteryl esters in dyslipidemic plasma: role of cholesteryl ester transfer protein. J Clin Invest. 1987;79:1217-1225.
Morton RE. Free cholesterol is a potent regulator of lipid transfer protein function. J Biol Chem. 1988;263:12235-12241.
Fielding PE, Jackson EM, Fielding CJ. Chronic dietary fat and cholesterol inhibit the normal postprandial stimulation of plasma cholesterol metabolism. J Lipid Res. 1989;30:1211-1217.
Ogawa Y, Fielding CJ. Assay of cholesterol ester transfer activity and purification of a cholesteryl ester transfer protein. Methods Enzymol. 1985;111:275-285.
Groener JEM, Pelton RW, Kostner GM. Improved estimation of cholesterol ester transfer/exchange activity in serum or plasma. Clin Chem. 1986;32:283-286.
Pussinen P, Jauhiainen M, Metso J, Tyynelä J, Ehnholm C. Pig plasma phospholipid transfer protein facilitates HDL interconversion. J Lipid Res. 1995;36:975-985.
Gardner MJ, Altman DG. Statistics With Confidence. London, England: British Medical Journal; 1989.
Van Tol A, Scheek LM, Groener JEM. Net mass transfer of cholesteryl esters from low density lipoproteins to high density lipoproteins in plasma from normolipidemic subjects. Arterioscler Thromb. 1991;11:55-63.
Yen FT, Deckelbaum RJ, Mann CJ, Marcel YL, Milne RW, Tall AR. Inhibition of cholesteryl ester transfer protein activity by monoclonal antibody: effects of cholesteryl ester formation and neutral lipid mass transfer in human plasma. J Clin Invest. 1989;83:2018-2024.
Lagrost L, Athias A, Herbeth B, Guyard-Dangremont V, Artur Y, Paille F, Gambert P, Lallemant C. Opposite effects of cholesteryl ester transfer protein and phospholipid transfer protein on the size distribution of plasma high density lipoproteins: physiological role in alcoholic patients. J Biol Chem. 1996;271:19058-19065.
Horton RC, Owen JS. LCAT activity is a prognostic liver function test. Lancet. 1990;336:249-250.
Lieber CS. The metabolism of alcohol. Sci Am. 1976;234:25-33.
Murakami T, Michelagnoli S, Longhi R, Gianfranceschi G, Pazzucconi F, Calabresi L, Sirtori CS, Franceschini G. Triglycerides are major determinants of cholesterol esterification/transfer and HDL remodeling in human plasma. Arterioscler Thromb Vasc Biol. 1995;15:1819-1828.
Mindham MA, Mayes PA. Application of simultaneous spleen and liver perfusion to the study of reverse cholesterol transport. Biochem J. 1994;302:207-213.
Liu XQ, Bagdade JD. Neutral lipid mass transfer among lipoproteins in plasma from normolipidemic subjects is not an equimolar heteroexchange. J Lipid Res. 1995;36:2574-2579.
Hannuksela ML, Rantala M, Kesäniemi YA, Savolainen MJ. Ethanol-induced redistribution of cholesteryl ester transfer protein (CETP) between lipoproteins. Arterioscler Thromb Vasc Biol. 1996;16:213-221.