![]()
|
|
||||||||
Volume 272, Number 50, Issue of December 12, 1997 pp. 31348-31354
(Received for publication, July 25, 1997, and in revised form, September 12, 1997)
,From INSERM Unité 391, Rennes 35043, France and § Serlia and INSERM Unité 325, Lille 59019, France
ApoC-III overexpression in mice results in severe hypertriglyceridemia due primarily to a delay in the clearance of triglyceride-rich lipoproteins. We have, in primary cultures of rat hepatocytes, characterized a lipolysis-stimulated receptor (LSR). The apparent number of LSR that are available on rat liver plasma membranes is negatively correlated with plasma triglyceride concentrations measured in the fed state. We therefore proposed that the primary physiological role of the LSR is to contribute to the cellular uptake of triglyceride-rich lipoproteins. We have now tested the effect of apoC-III on the binding of triglyceride-rich lipoproteins to LSR. Supplementation of 125I-very low density lipoprotein (VLDL) with apoC-III inhibited the LSR-mediated binding, internalization, and degradation of 125I-VLDL in primary cultures of rat hepatocytes. Studies using isolated rat liver plasma membranes showed that enrichment of human VLDL and chylomicrons with synthetic or purified human apoC-III decreased their binding to the LSR by about 40%. Supplementation of triglyceride-rich lipoproteins under the same conditions with human apoC-II had no such inhibitory effect, despite the fact that this apoprotein bound as efficiently as apoC-III to these particles.
Preincubation of LDL with apoC-III did not modify its binding to LSR. Partitioning studies using 125I-apoC-III showed that this lack of effect was due to apoC-III's inability to efficiently associate with LDL. Purified human apoC-III1 was as efficient as the synthetic nonsialylated form of apoC-III in inhibiting binding of VLDL to LSR. However, despite a 2-fold greater binding of apoC-III2 to VLDL, this isoform was a less efficient inhibitor of the binding of VLDL to LSR than apoC-III1 or nonsialylated apoC-III. Desialylation of apoC-III2 by treatment with neuraminidase increased the inhibition of VLDL binding to LSR to a level similar to that observed with apoC-III1 and nonsialylated apoC-III. We propose that apoC-III regulates in part the rate of removal of triglyceride-rich particles by inhibiting their binding to the LSR, and that the level of inhibition is determined by the degree of apoC-III sialylation.
Apolipoprotein (apo)C-III1 has been shown to modulate the plasma triglyceride (TG) concentrations. ApoC-III is a 79-amino acid polypeptide in the plasma that readily exchanges between the different lipoprotein fractions (1). In subjects with normal plasma lipid levels, 30% of apoC-III is distributed among triglyceride-rich lipoproteins (TGRL), i.e. chylomicrons and very low density lipoprotein (VLDL), and about 60% is bound to high density lipoprotein (1).
Biochemical studies showed that apoC-III is an inhibitor of lipase activity (2-4), and it was established in the early 1980's that supplementation of small chylomicrons or apoE-containing triglyceride emulsions with apoC-III inhibited their uptake in perfused rat livers (5-7). An inhibitory effect was also observed with apoC-I and C-II, but was less pronounced and not consistently observed (5, 6). Genetic linkage map studies suggest that polymorphism of apoC-III is associated with hypertriglyceridemia (8-11). Human subjects with defects of apoC-III and apoA-I genes had low plasma TG levels, attributed primarily to an accelerated clearance of TGRL (12, 13). In transgenic mice, suppression of apoC-III expression by homologous recombination increased the removal rate of chylomicron remnants and thus reduced the postprandial triglyceridemia (14). In contrast, overexpression of apoC-III increased the plasma concentrations of TG in a dose-dependent manner, and thereby caused a reduction of the catabolic rate of TGRL (15, 16).
The mechanisms through which apoC-III regulates the catabolism of TGRL are complex and involve both lipolysis and cellular uptake of these particles. Tissue lipase activities of apoC-III overexpressor mice were not different when compared with those of normal mice, nor were their VLDL defective as substrates for lipolysis (16). The hypertriglyceridemia of transgenic apoC-III overexpressors was corrected when these animals were cross-bred with mice overexpressing apoE, consistent with the hypothesis that apoC-III interferes with an apoE-mediated process (17). ApoE has been shown to mediate the binding of TGRL to heparan sulfate proteoglycans (for review, see Ref. 18), and it was recently reported that the VLDL isolated from apoC-III overexpressors were defective in binding to heparin-Sepharose, suggesting that the excess apoC-III inhibits their binding to heparan sulfate proteoglycans on endothelial cells (19). Because most of the plasma lipolytic enzymes, i.e. lipoprotein and hepatic lipases, are bound to heparan sulfate proteoglycans, this docking defect of apoC-III-enriched TGRL could subsequently reduce the rate of lipolysis.
It is also widely accepted that apoE serves as the primary ligand for
the cellular uptake of TGRL. It is therefore possible that apoC-III's
hypertriglyceridemic effect is due in part to a decreased rate of
cellular uptake of TGRL by liver cells. ApoC-III has been shown to
decrease the binding of VLDL to the low density lipoprotein
(LDL)-receptor expressed in human fibroblasts (20, 21). This inhibitory
effect was observed not only with apoC-III, but also with apoC-I and
C-II, apoC-I being the most effective. In addition, apoC-I, but not
apoC-III, inhibits the binding of apoE-enriched
-VLDL to the
LDL-receptor related protein (LRP) (22).
We have characterized a third candidate receptor, the lipolysis-stimulated receptor (LSR), which is activated by free fatty acids (FFA), and is distinct from the LDL-receptor and the LRP (23, 24). Similar to the LDL receptor, the LSR binds both apoB and apoE (23, 24). However, unlike the LDL receptor, its affinity is maximal for TGRL (23, 24). Furthermore, the affinity for the LSR of VLDL isolated from subjects with type III hyperlipidemia (apoE-2/2 phenotype) is decreased when compared with that isolated from normal individuals (24). The physiological significance of the LSR as candidate chylomicron remnant receptor has remained controversial. Indeed, during in vitro assays, maximal activation of the LSR requires the addition of FFA at concentrations that exceed albumin binding capacity and are therefore at levels not normally observed under physiological conditions. It has been shown that FFA bound to albumin at the physiological molar ratio of 1 to 1 are sufficient to consistently activate the LSR. However, under these conditions, the receptor activation is only 10% of the maximal activity. We therefore speculate that under physiological conditions, the LSR is not activated primarily by FFA circulating in plasma bound to albumin, but rather by the FFA generated at the site of lipolysis. When intestinally-derived lipoproteins arrive in the liver sinusoid and are hydrolyzed by hepatic lipase (25) and possibly also lipoprotein lipase (26, 27), it is likely that relatively high local concentrations of FFA are produced. Consistent with the hypothesis that LSR participates in the clearance of chylomicrons is the finding that the apparent number of LSR expressed in rat liver plasma membranes strongly (p < 0.001) and negatively correlates with plasma TG concentrations measured in the physiological postprandial stage of these animals (28). To further examine the potential physiological significance of the LSR, we have now characterized the effect of apoC-III on the binding of the different LSR ligands. Data presented here show that apoC-III is a potent and specific inhibitor of TGRL, but not LDL binding to LSR.
Materials
Na[125I] was purchased from Amersham (Les Ulis, France). Oleic acid (0-1008), bovine serum albumin (BSA) (A2153), neuraminidase from Clostridium perfringens (N2876), leupeptin, and benzamidine were obtained from Sigma (St. Quentin, Fallavier, France). Dulbecco's modified Eagle's medium, penicillin-streptomycin, glutamine, and fetal bovine serum were purchased from Life Technologies, Inc. (Eragny, France). Sodium suramin was purchased from FBA Pharmaceutical (Westhaven, CT).
Methods
AnimalsMale Sprague-Dawley rats (Janvier Breeding Center, Le Genest-St.-Isle, France) were housed in an animal care facility equipped with a 12-h light/dark cycle. Animals were given water and regular chow ad libitum, except for 15 h prior to liver isolation when they received only water. For isolation of livers, animals were anesthetized with ether, and the livers were perfused through the portal vein with ice-cold Hepes-buffered saline solution (5 mM Hepes, 150 mM NaCl, 2 mM EDTA, pH 7.4). The livers were then immediately excised, and used for the preparation of liver plasma membranes.
Preparation of LipoproteinsAll plasma samples used for lipoprotein isolation were supplemented with 0.01 volumes of the following inhibitor mixture: leupeptin (1 mg/ml), benzamidine (10 mg/ml), and sodium azide (300 mM) prior to lipoprotein isolation. VLDL (density (d) < 1.006 g/ml) and LDL (1.025 < d < 1.055 g/ml) were prepared from plasma of overnight fasted human subjects (29). Chylomicrons were prepared by ultracentrifugation (30,000 × g, 1 h, 10 °C, Beckman Ti50.2 rotor) of plasma drawn from human subjects 2 h after a breakfast consisting of 3 croissants with butter and jam, and 250 ml of whole milk. Contaminating albumin was removed from the chylomicron fraction by incubation for 30 min at room temperature with an equivalent volume of swollen Blue Sepharose CL-6B gel (Pharmacia Biotech, Orsay, France). This was verified by SDS-polyacrylamide gel electrophoresis separation of chylomicron protein on 4-18% gradient gels. Image analysis (ImageMaster, Pharmacia Biotech) of Coomassie Blue-stained gels indicated that after this treatment, albumin represented less than 1% of total protein. All lipoproteins were stored in the dark at 4 °C under N2 for use within 4 days of their isolation.
Isolation of ApoC PeptidesHuman apoC-III was synthetically
prepared using solid phase methodology (30) as described previously
(31), and provided the source of non-sialylated apoC-III. The apoC-III
was stored at
20 °C as a lyophilized powder and was dissolved into
0.1 M phosphate-buffered saline, pH 7.4, on the day of use.
Human apoC-III preparations containing 1 and 2 sialyl residues were
purified from plasma obtained from the local blood bank, as described
(32), except that chromatographic steps were conducted by fast protein liquid chromatography using a Superdex 75 HR10/30 column (Pharmacia Biotech) for gel filtration, and a Bio-Scale Q20 column (Bio-Rad, Ivry-Sur-Seine, France) for anion-exchange.
Purity of apoC-III1 and C-III2 preparations was
verified by isoelectric focusing on PhastGel IEF polyacrylamide gels
(Pharmacia Biotech) using a broad separation range of pI 3-10 expanded
at the 3-6 end. Image analysis of these gels stained with Coomassie Blue showed the purity of the apoC-III1 and
C-III2 preparations to be >96%. ApoC-II was purified from
human plasma as described (32), and stored at concentrations >1 mg/ml
at
80 °C until the day of use.
Purified apoC-III2 was desialylated using a method modified from Morell et al. (33). Briefly, apoC-III2 was incubated at 37 °C for 16 h in a shaking water bath with 2 units/mg neuraminidase in phosphate-buffered saline, pH 7.4, containing 0.1% (w/v) CaCl2 and 0.01 volume of protease inhibitor mixture. The buffer of this mixture was then exchanged for 10 mM Tris containing 2 mM EDTA, pH 8, using a Fast Desalting HR 10/10 column (Pharmacia Biotech), prior to reisolation of the apoC-III by anion exchange employing a Mono Q PC 1.6/5 SMART column (Pharmacia Biotech) using a gradient of 0-250 mM NaCl. Purity of the recovered desialylated apoC-III was verified by isoelectric focusing on PhastGel IEF polyacrylamide gels.
RadiolabelingChylomicrons, VLDL, and LDL were radioiodinated by Bilheimer's modification of McFarlane's method (34), as described previously (35). 125I-Labeled lipoproteins were used within 2 days of preparation, and 125I-LDL was filtered (0.2 µm filter, Gelman, Ann Arbor, MI) immediately prior to use. Specific activities of 125I-labeled lipoproteins preparations ranged between 102 and 192 cpm/ng protein.
In experiments designed to determine the precise amount of the different apoC-III isoforms associated with TGRL, apoC-III preparations were iodinated using IODO-BEADS (Pierce, Asnieres, France) according to manufacturer's instructions, and used within 1 week of preparation.
CellsPrimary cultures of rat hepatocytes were prepared as described previously (28), and used 48 h after plating in 36-mm dishes. The cells were preincubated at 37 °C for 30 min in Dulbecco's modified Eagle's medium containing 0.2% (w/v) BSA, 5 mM Hepes, 2 mM CaCl2, and 10 ng/ml recombinant mouse leptin,2 pH 7.5, followed by incubation at 37 °C for 4 h with 20 µg/ml 125I-VLDL supplemented or not with apoC-III (see below) in the same media. After which, the cells were washed and the amounts of 125I-lipoprotein bound, internalized, and degraded were determined as previously reported (23).
Rat Liver Plasma Membrane IsolationPlasma membranes were prepared from livers of fasted rats as described (28, 36), stored in the dark at 4 °C under N2 in the presence of the inhibitor mixture, and used within 7 days of their preparation.
Preparation of Lipoproteins Supplemented with ApoCChylomicrons, VLDL, or LDL (0.15 mg protein) and apoC-II or C-III (0.15 mg) were mixed and incubated at 37 °C for 30 min in phosphate-buffered saline, pH 7.4, containing 2 mM EDTA and 3 mM sodium azide (buffer A) in a volume not exceeding 500 µl. After cooling for 2 min on ice, the sample was applied to a 50 mm × 10-mm gel filtration column containing Sepharose CL-4B (Pharmacia Biotech) pre-equilibrated in buffer A. The sample was eluted at a flow rate of 500 µl/min, and collected in 250-µl fractions. Measurements of the 125I radioactivity and absorbance at 280 µm were used to separate the lipoprotein-bound from the free apoC fractions. All lipoprotein fractions were collected onto ice within 3-7 min, and used immediately. No difference between the 125I to TG ratio of VLDL were found before and after such treatment, indicating that incubation with the apoC peptides did not cause major losses of radiolabeled endogenous apoproteins. Samples were also applied on 4-25% SDS gradient polyacrylamide gels, which were then stained with Coomassie Blue. Analysis of these gels showed that the gel filtration step did not modify the TGRL apolipoprotein profile. It must be emphasized, however, that the gel filtration step is critical to observe the specificities regarding both: 1) the type of ligand affected, i.e. the lack of effect of apoC-III on LDL binding to LSR, and 2) the type of C apoprotein, i.e. the lack of effect of apoC-II on the binding of TGRL to LSR. Indeed, when the gel filtration chromatography step was omitted prior to testing, all apoC fractions inhibited LDL binding to the LSR. We speculate that this inhibitory effect proceeds through mechanisms distinct from those responsible for the apoC-III specific inhibition of TGRL binding to LSR, and most probably involves the apoC's that are free in solution.
Measurement of LSR Activity in Isolated Rat Liver Plasma MembranesLSR activity was measured under conditions that
maximally activated the receptor (28). All buffers were prepared
without Ca2+, and were supplemented with 2 mM
EDTA, to suppress the activities of the LDL receptor (37) and the LRP
(38). Briefly, aliquots of membranes (100 µg of protein/tube) were
incubated at 37 °C for 30 min in the absence or presence of 1000 µM oleate adjusted to a final volume of 250 µl with 0.1 M phosphate buffer containing 350 mM NaCl and 2 mM EDTA, pH 8.0 (buffer B). The membranes were then washed
by 3 series of centrifugation (35,000 × g, 15 min, 4 °C) and resuspension into 250 µl of buffer B by brief sonication (power 1.0, 90% pulse, 5 s); at the final wash, the membrane
pellets were resuspended in 150 µl of buffer B. After this, the
membranes were incubated at 4 °C for 60 min with the different
preparations of apoC-supplemented lipoproteins in a final volume of 250 µl. Membrane-bound 125I-lipoproteins were separated from
unbound 125I-lipoproteins by layering a 200-µl aliquot
over a 600-µl cushion of 5% (w/v) BSA in buffer B, and centrifuging
(35,000 × g, 20 min, 4 °C). After careful
aspiration of the supernatants, the bottoms of the tubes containing the
membrane pellets were cut and counted in a
-counter (Pharmacia 1470 Wizard). The need to thoroughly wash the FFA not bound to the membrane
prior to adding the 125I-lipoprotein to the media must be
emphasized. Indeed, relatively high concentrations of FFA cause the
precipitation of protein when subjected to centrifugation. This leads
to a significant increase in the nonspecific binding.
Protein contents of lipoproteins and membranes were measured by Markwell's modification of Lowry's procedure (39), using BSA as standard. TG concentrations were determined using an enzymatic kit (Boehringer Mannheim, Meylan, France).
We first tested the effect of apoC-III on the binding to LSR of
the different lipoprotein fractions that had previously been identified
as potential ligands for this candidate receptor (23, 24). Radiolabeled
chylomicrons, VLDL, or LDL were supplemented with synthetic apoC-III,
and then separated by gel filtration chromatography to remove unbound
apoprotein. Fig. 1 shows that both
chylomicrons and VLDL, but not LDL, supplemented with apoC-III (open squares) displayed, in the presence of oleate, a
decreased binding to rat liver plasma membranes when compared with
those not supplemented with apoC-III (Fig. 1, closed
squares). In membranes incubated without oleate, the binding of
chylomicrons, VLDL, and LDL was less than 10% of that measured in the
presence of oleate. ApoC-III had no detectable inhibitory effect on
this nonspecific binding (closed versus open circles). Due
to the dilution effect caused by passing the apoC-III supplemented
lipoprotein through gel filtration columns, the concentrations of
lipoproteins used in these experiments did not exceed 20 µg/ml, a
level well below that required to achieve saturation of the receptor
(28). Therefore, data of Fig. 1 differ from the typical saturation
curves previously reported for the binding of these lipoproteins to LSR
(28). We chose not to test at concentrations greater than 20 µg/ml
since this would require the addition of a concentration step which would have altered the apoprotein profile of the apoC-III-supplemented TGRL.
[View Larger Version of this Image (15K GIF file)]
We next sought to determine the amount of apoC-III that bound to TGRL or LDL under these experimental conditions. Unlabeled VLDL or LDL were supplemented with 125I-apoC-III exactly as described for the experiments shown in Fig. 1. After this, 125I-apoC-III bound to lipoprotein was separated from free 125I-apoC-III by gel filtration chromatography. Under these conditions, a maximum of 10% of added 125I-apoC-III associated with VLDL, while 90% remained in solution. In contrast, only 0.3% of 125I-apoC-III was associated with LDL, while >99% remained unbound (data not shown). This low level of apoC-III bound to LDL could thus explain the lack of significant effect of apoC-III on LDL binding to LSR.
We next tested, in experiments similar to those described in Fig. 1,
whether the inhibitory effect of apoC-III on the binding of TGRL to LSR
was specific for this apolipoprotein. Fig.
2 shows that supplementation of
chylomicrons and VLDL with apoC-II had no significant effect on the
binding of these lipoproteins to rat liver membranes incubated with
oleate. This lack of effect of apoC-II was not due to a defect of
binding of apoC-II to TGRL. Indeed, analysis of Coomassie Blue-stained
gels showed enrichment of VLDL after incubation with either apoC-II
(Fig. 3, third lane from left)
or apoC-III (Fig. 3, fourth lane from left).
[View Larger Version of this Image (16K GIF file)]
[View Larger Version of this Image (43K GIF file)]
The next objective was to establish if the specific inhibitory effect
of apoC-III on the binding of TGRL to LSR expressed on the plasma
membranes caused a decrease of the uptake and degradation of these
particles in intact cells. 125I-VLDL supplemented or not
with apoC-III was incubated with primary cultures of rat hepatocytes in
the presence and the absence of oleate. ApoC-III caused a significant
decrease in oleate-induced binding (46%) of VLDL (Fig.
4, panel A). As a consequence,
uptake and degradation of 125I-VLDL were reduced to a
similar or slightly greater extent (Fig. 4, panels B and
C, respectively).
[View Larger Version of this Image (14K GIF file)]
In experiments described thus far, a synthetic nonsialylated form of
human apoC-III was used. However, under physiological conditions,
apoC-III exists as 3 different isoforms designated as
C-III0, C-III1, and C-III2, where
the subscript indicates the number of sialic residues bound to each of
the peptides. We subsequently undertook the purification of the
different isoforms of apoC-III to determine which was the most
effective inhibitor of LSR. As shown in Fig.
5, apoC-III1 showed an
inhibitory effect similar to that for nonsialylated
apoC-III0 (Fig. 1). However, apoC-III2 demonstrated a diminished ability to inhibit VLDL binding to LSR. This
decreased inhibitory effect was not caused by a reduction in the
binding of apoC-III2 to VLDL. Indeed, partitioning
experiments showed that the amount of
125I-apoC-III2 bound to VLDL was 2-fold greater
as compared with the other apoC-III isoforms (Fig. 5,
inset). To determine whether the number of sialic acid
residues were an important determinant of the degree of inhibition,
apoC-III2 was treated with neuraminidase. The top
panel of Fig. 6 shows that treatment
of apoC-III2 with neuraminidase altered its migration after
isoelectrofocusing electrophoresis to a level comparable with
nonsialylated apoC-III0. This desialylation significantly
enhanced the ability of the apoC-III2 to inhibit VLDL
binding to a level similar to that observed with apoC-III0 (Fig. 6, bottom panel). Taken together, these data indicate
that apoC-III supplementation of TGRL significantly inhibits their binding to LSR, and that this inhibition is dependent on the degree of
sialylation of this apoprotein.
), or in the presence of 0.15 mg
of nonsialylated apoC-III (apoC-III0,
), apoC-III1 (
), or apoC-III2 (
) in buffer
A. The VLDL were then reisolated by gel filtration, and the
oleate-induced binding of increasing concentrations of these different
125I-VLDL was then measured using rat liver plasma
membranes as described in the legend to Fig. 1. Each point represents
the mean of duplicate determinations. Inset, comparison of
the affinity of apoC-III isoforms for VLDL. Eighty µg of VLDL were
incubated at 37 °C for 30 min with 80 µg of
125I-apoC-III0,
125I-apoC-III1, or
125I-apoC-III2 (specific activities, 241, 285, and 884 cpm/ng, respectively) in a final volume of 500 µl of buffer
A. VLDL containing bound 125I-apoC-III was then separated
from unbound 125I-apoC-III by gel filtration as described
under "Methods." Aliquots of each fraction were counted for
125I content, which revealed two peaks: lipoprotein
associated 125I-apoC-III and unbound
125I-apoC-III. The amount of
125I-apoC-III0 (open bar),
125I-apoC-III1 (hatched bar), and
125I-apoC-III2 (solid bar)
associated with VLDL is expressed as percentage of total radioactivity
eluted from the column. Column recoveries were
90%.
[View Larger Version of this Image (18K GIF file)]
[View Larger Version of this Image (26K GIF file)]
The data reported in this study show that the enrichment of VLDL or chylomicrons with apoC-III significantly decreased their ability to bind to LSR. This inhibition was specific for TGRL and was not found with apoC-II. Based on partitioning experiments, up to 10% of the added apoC-III remained associated with VLDL after gel filtration. In normal human subjects, apoC-III represents about 30% of VLDL protein mass. Therefore, supplementation experiments described here caused a 30-40% increase in VLDL apoC-III. This led to a 40% inhibition of its binding to LSR. ApoE was present on VLDL supplemented with apoC-III (Fig. 3). Thus, the inhibitory effect was not caused by a massive displacement of apoE, but rather by a masking or a change in the conformation of the apoE that can serve as ligand for LSR (28). The possibility that apoC-III also causes a conformational change in apoB must be considered (40). Indeed, the lack of effect of apoC-III on LDL binding to LSR does not rule out this hypothesis, because it was primarily due to apoC-III inability to associate in significant amounts to purified LDL. It is also possible that apoC-III inhibitory effect occurs through interactions with other apolipoproteins. Among which is apoA-I that is present on most TGRL and remains to be tested for its ability to bind the LSR.
The inhibition of the binding of TGRL to LSR by apoC-III is dependent on the number of sialic residues present on the apolipoprotein. Despite a greater ability of apoC-III2 to associate with VLDL, this isoform was a less efficient inhibitor as compared with nonsialylated apoC-III0 or C-III1. We therefore speculate that the charge provided by the additional sialic residue interferes with the ability of apoC-III to modulate apoE and possibly apoB binding to LSR. This observation opens a new perspective regarding the regulation of TG removal rate by apoC-III. On the basis of data presented here, one could predict that the least sialylated isoform of apoC-III would be the most efficient inhibitor for the removal of TGRL. The physiological mechanism might, however, be more complex. Indeed, a modulation of the partitioning of apoC-III between TGRL and high density lipoprotein as a function of the degree of sialylation must also be considered. We have shown here that the higher degree of sialylation (i.e. 2 residues/molecule) increased the binding of apoC-III to VLDL. It would be of interest to define whether apoC-III sialylation modulates its affinity for high density lipoprotein in the same or the opposite direction. The degree of sialylation of apoC-III in transgenic animals has thus far not been reported. At this stage, we can only speculate that the degree of sialylation of the apoC-III is an important determinant of its ability to modulate the clearance of TGRL.
Most experiments reported here relied on a membrane binding assay that requires very high concentrations of FFA to maximally recruit LSR. We do not believe that such high concentrations of FFA exist in vivo, but are nevertheless confident that although unphysiological, this method provides a valuable tool to test the activity of the LSR for the following reasons. 1) Measurement of LSR apparent number using this method led to the finding of a highly significant linear and negative correlation between the apparent number of receptors and postprandial plasma TG levels (28). 2) The relatively high concentrations of FFA that are used do not exert detergent effects because the pH of the incubation media does not allow formation of micelles (41). Therefore, the plasma membranes are not denatured by the FFA. 3) Removal of FFA bound to the plasma membranes after the washes by treatment with albumin lead to the reversible deactivation of the receptor (28). 4) The same activation and deactivation of LSR by adding and removing oleate to and from plasma membranes can be repeated several times without significant loss of activity (28). 5) The binding assay described with isolated plasma membranes retains the same ligand specificity than that observed after incubation of intact cells with FFA in the presence of albumin (28). 6) The oleate-induced binding of LDL to rat liver plasma membranes was inhibited by more than 80% by a polyclonal antibody raised against purified LSR protein and by more than 50% by a polyclonal antibody raised against a peptide synthesized on the basis of protein sequences predicted by LSR mRNA.3 Thus, although performed under nonphysiological conditions, the LSR membrane binding assay is robust and likely to reflect a physiologically significant phenomenon.
ApoC-III inhibition of the LSR was also observed in experiments using
primary cultures of rat hepatocytes. In these cells, both the LDL
receptor and the LRP are expressed and functional. We do not believe,
however, that either of these receptors significantly contribute to the
binding, uptake, and degradation of VLDL. Indeed, the LDL receptor has
been shown not to bind normal human VLDL (42) and to be inhibited by
oleate (43). LRP, on the other hand, recognizes
-VLDL supplemented
with large amounts of apoE (44); such high apoE concentrations are
unlikely to be found in the hepatocyte incubation media within the
short period of time of this incubation. Finally, the oleate-induced
binding, uptake, and degradation of lipoprotein in primary cultures of rat hepatocytes was inhibited by more than 85% by a polyclonal anti-LSR antibody.3
The specificity of the inhibitory effect of apoC-III on the binding of TGRL to LSR required experimental conditions in which no free C apoproteins were left in the incubation media. When the apoC/LDL mixture was used directly in the LSR activity assay, apoC-I, C-II, and C-III all inhibited LDL binding to LSR (data not shown). The mechanism of this inhibition is not clear. Partitioning experiments using radiolabeled apoC-III indicated that under these experimental conditions, about 90% of total apoC-III remained free in solution. One possible mechanism is therefore that the hydrophobic residues of the free apoC's trapped the FFA thus inhibiting the conformational shift needed to activate LSR (28). It is also possible that free apoC directly binds to the membrane phospholipid or protein, thereby masking the receptor.
At least three different mechanisms can be proposed to explain the hypertriglyceridemia observed when apoC-III concentrations are moderately increased. First, the increased apoC-III could cause a defective binding of the TGRL to heparan sulfate proteoglycans and hence a resistance to lipolysis (19). It remains to be determined whether such a defect occurs in the absence of a displacement of the apoE that is responsible for the docking of TGRL to the endothelium (18). The recent finding by Ebara et al. (45) that apoC-III overexpression increases the number of large TG-rich particles in apoE-deficient mice is consistent with the notion that apoC-III is capable of inhibiting the lipolytic process even in the absence of apoE.
Second, the binding of apoC-III-enriched TGRL to the LDL receptor is decreased. This effect accounts for most of the reduced uptake of apoC-III-enriched TGRL observed in cultured fibroblasts or HeLa cells (20, 21). Indeed, in these experiments, the expression of the LDL receptor was maximized by incubation in lipoprotein-deficient serum. Because no FFA were added to the incubation media of these cells, it is unlikely that any LSR activity would account for these observations. Nevertheless, the LDL receptor model is not sufficient to explain all experimental data. Indeed apoC-I is a more potent inhibitor of the LDL receptor than apoC-III. In mice, the LDL receptor is estimated to account for the clearance of LDL and for about 50% of the removal of intestinally-derived chylomicron remnants (46). Consistent with these observations is the finding that apoC-I causes in transgenic mice an increase not only of plasma TG, but also plasma cholesterol (47), whereas apoC-III overexpression causes a significant increase in plasma TG and only limited changes in cholesterol (15, 16). A recent report has shown that in homozygous LDL receptor knock-out mice, overexpression of apoC-III increased VLDL concentrations 10-fold (48).
Third, apoC-III hypertriglyceridemic effect may be due in part to the inhibition of TGRL binding to the LSR. The finding reported here provides yet another line of circumstantial evidence in support of the notion that the LSR represents a physiological rate-limiting step for the removal of TGRL.
To whom correspondence should be addressed: INSERM U391, 2 Ave du
Prof. Léon Bernard, 35043 Rennes, France. Tel.: 33-2-99-33-69-40; Fax: 33-2-99-33-62-08.
We thank Michael J. Kluk for technical assistance.
This article has been cited by other articles:
![]() |
J.-F. Mauger, P. Couture, N. Bergeron, and B. Lamarche Apolipoprotein C-III isoforms: kinetics and relative implication in lipid metabolism J. Lipid Res., June 1, 2006; 47(6): 1212 - 1218. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. G Jackson, E. J Wolstencroft, P. A Bateman, P. Yaqoob, and C. M Williams Greater enrichment of triacylglycerol-rich lipoproteins with apolipoproteins E and C-III after meals rich in saturated fatty acids than after meals rich in unsaturated fatty acids Am. J. Clinical Nutrition, January 1, 2005; 81(1): 25 - 34. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Wanner, U. Bahner, R. Mattern, D. Lang, and J. Passlick-Deetjen Effect of dialysis flux and membrane material on dyslipidaemia and inflammation in haemodialysis patients Nephrol. Dial. Transplant., October 1, 2004; 19(10): 2570 - 2575. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. S. Cohn, B. W. Patterson, K. D. Uffelman, J. Davignon, and G. Steiner Rate of Production of Plasma and Very-Low-Density Lipoprotein (VLDL) Apolipoprotein C-III Is Strongly Related to the Concentration and Level of Production of VLDL Triglyceride in Male Subjects with Different Body Weights and Levels of Insulin Sensitivity J. Clin. Endocrinol. Metab., August 1, 2004; 89(8): 3949 - 3955. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Ando, T. Shimizugawa, S. Takeshita, M. Ono, M. Shimamura, R. Koishi, and H. Furukawa A decreased expression of angiopoietin-like 3 is protective against atherosclerosis in apoE-deficient mice J. Lipid Res., June 1, 2003; 44(6): 1216 - 1223. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Olin-Lewis, R. M. Krauss, M. La Belle, P. J. Blanche, P. H. R. Barrett, T. N. Wight, and A. Chait ApoC-III content of apoB-containing lipoproteins is associated with binding to the vascular proteoglycan biglycan J. Lipid Res., November 1, 2002; 43(11): 1969 - 1977. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. C. Jong, P. C. N. Rensen, V. E. H. Dahlmans, H. van der Boom, T. J. C. van Berkel, and L. M. Havekes Apolipoprotein C-III deficiency accelerates triglyceride hydrolysis by lipoprotein lipase in wild-type and apoE knockout mice J. Lipid Res., October 1, 2001; 42(10): 1578 - 1585. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Batal, M. Tremblay, P. H. R. Barrett, H. Jacques, A. Fredenrich, O. Mamer, J. Davignon, and J. S. Cohn Plasma kinetics of apoC-III and apoE in normolipidemic and hypertriglyceridemic subjects J. Lipid Res., May 1, 2000; 41(5): 706 - 718. [Abstract] [Full Text] |
||||
![]() |
F. T. Yen, M. Masson, N. Clossais-Besnard, P. Andre, J.-M. Grosset, L. Bougueleret, J.-B. Dumas, O. Guerassimenko, and B. E. Bihain Molecular Cloning of a Lipolysis-stimulated Remnant Receptor Expressed in the Liver J. Biol. Chem., May 7, 1999; 274(19): 13390 - 13398. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. C. Jong, M. H. Hofker, and L. M. Havekes Role of ApoCs in Lipoprotein Metabolism : Functional Differences Between ApoC1, ApoC2, and ApoC3 Arterioscler. Thromb. Vasc. Biol., March 1, 1999; 19(3): 472 - 484. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||