Modulation of Lipoprotein B Binding to the LDL Receptor by Exogenous Lipids and Apolipoproteins CI, CII, CIII, and E
Abstract We have recently shown that apo B–containing lipoproteins isolated by immunoaffinity chromatography bind to the LDL receptor with an affinity dependent on their apo E or apo CIII content. However, these lipoproteins—LpB:E, LpB:CIII, and LpB:CIII:E—isolated from whole plasma have variable lipid and apolipoprotein contents, and it is difficult to consider each parameter separately, particularly because an increase in the apo CIII content is always associated with an increase in the content of other C apolipoproteins. Therefore, we used affinity-purified LpB free of other apolipoproteins. Lipid content of LpB was increased by incubation with a lipid emulsion, and this triglyceride-enriched LpB was named TG-LpB. Free apo CI, apo CII, apo CIII, and apo E were added to LpB and TG-LpB and their associations to the lipoprotein were assessed by gel filtration, nondenaturing electrophoresis, and immunoblotting. Molar ratios of 6 (apo E), 30 (apo CII), 20 (apo CIII), and 30 (apo CI) for 1 apo B were obtained. The association of apo CII to LpB and TG-LpB induced modifications to the LpB structure and a redistribution of lipids and apolipoproteins on the lipoprotein particles. The binding of these LpBs and TG-LpBs with and without added apo CI, CII, CIII, and E was tested at 4°C on the LDL receptors of HeLa cells. The increased content of lipids reduced TG-LpB binding to the LDL receptor. Addition of apo CIII to LpB decreased its affinity, although this decrease was lower than that observed with LpB:CIII prepared from whole plasma. Apo CIII almost completely abolished the interaction of TG-LpB with the receptor, indicating a synergistic effect of lipids and apo CIII. The apo CIII effect was specific and cannot be obtained with apo CI. With apo CII, an inhibitory effect can also be obtained but to a lesser extent than with apo CIII. At 37°C the C apolipoproteins decreased the catabolism of LpB and TG-LpB by the LDL receptor of fibroblasts. Addition of apo E to either LpB or TG-LpB had a small effect on the binding of the enriched lipoproteins at 4°C but markedly increased their catabolism at 37°C.
- Received June 6, 1994.
- Accepted April 6, 1995.
Apo B100 is responsible for the interaction of LDLs with their receptor.1 The areas of apo B involved in the interaction with the LDL receptor remain to be identified. Studies with monoclonal antibodies suggest that the region of apo B between amino acids 2800 and 4000 contains the receptor binding domain2 and that several domains of apo B distributed along the protein may contribute to the apo B conformation required for interaction with the receptor.3 4 Nevertheless, it is known that a point mutation leading to an arginine-to-glutamine substitution at amino acid 3500 of apo B abolishes the LDL receptor interaction5 and that subtle changes in the lipid surroundings may modify the functional and immunological properties of the apo B receptor domain.6 7 8
Two studies by Bradley and coworkers9 10 have shown that in large triglyceride (TG)-rich lipoproteins, apo B does not interact with the LDL receptor and that the LDL receptor–binding determinants switch from apo E to apo B during conversion of these VLDL to LDL. In large TG-rich VLDL, the surroundings of apo B impair its binding to the LDL receptor, this effect being probably due to the lipid content of the lipoproteins but also to the other apolipoproteins present at the surface. We demonstrated the inhibitory effect of C apolipoproteins on the LDL receptor binding of apo B–containing lipoproteins purified by immunoaffinity.11 However, naturally occurring lipoproteins are very heterogeneous, whatever the different purification approaches, such as ultracentrifugation,8 12 heparin Sepharose chromatography,13 or immunoaffinity.11 Even if all studies suggest an important role for apo E in the catabolism of TG-rich lipoproteins and an inhibitory effect of lipids and C apolipoproteins for apo B binding to the LDL receptor, no one can definitively determine the individual effect of each factor. Likewise, it has already been demonstrated that apo E14 15 and apo C14 become associated with VLDLs or TG-rich lipoproteins by simple incubation at 37°C and modify their functional properties, but the use of VLDLs containing apo B, apo C, and apo E before exogenous addition makes the interpretation of these effects difficult.
The purpose of our study was to assess the possible role of lipid and apolipoprotein composition on the binding affinity of particles containing apo B as the sole apolipoprotein. To this end we purified B lipoproteins (LpBs) by immunoaffinity16 and enriched these lipoproteins either in lipids or in apolipoproteins CI, CII, CIII, or E. The specific effect of each factor on the LDL receptor–binding properties of apo B was studied.
Apolipoproteins CI, CII, and CIII were prepared from delipidated human plasma VLDL by gel filtration on Sephacryl S200 followed by chromatography on a MonoQ cation exchange column (Pharmacia Sweden) as previously described.17 Three fractions of apo CIII were thus obtained (CIII0, CIII1, and CIII2). Only apo CIII2 was used. Apo E3 was obtained from VLDL from a hypertriglyceridemic E3/3 subject according to the procedure described by Rall et al.18 Apo E phenotype was determined from total plasma and verified on purified apo E by electrofocusing followed by immunoblotting.19 20 The purity of apolipoproteins was checked by SDS–polyacrylamide gel electrophoresis (SDS-PAGE) and urea PAGE.
LpB was prepared by sequential immunoaffinity from human plasma derived from fasting normolipemic subjects.16 In brief, all apo B–containing lipoproteins were retained by an anti–apo B immunosorbent, then passed through anti–apo CIII, anti–apo E, and anti–apo(a) immunosorbents. All lipoproteins containing apo CIII also contain apo CI and apo CII and vice versa (V. Clavey, PR, et al, unpublished data, 1994). The LpB was extensively dialyzed against sodium phosphate buffer (0.01 mol/L, pH 7.0) containing 0.15 mol/L NaCl and 0.01% EDTA (PBS-EDTA). The LpB obtained was free of other apolipoproteins, as checked by SDS-PAGE and immunoenzymatic assay for apo E, apo(a), and apolipoproteins AI, AII, CI, CII, and CIII.21 Apo B content of LpB was determined by nephelometry (Behring) and compared with the total protein content determined by the modified Lowry method of Petersen.22 In all preparations the two values (apo B and total proteins) were consistent with the absence of all other proteins.
LpB was enriched in vitro with TGs by incubation with lipid emulsion in the presence of lipoprotein-deficient serum.23 The lipid emulsion (Endolipide 20%, Bruneau) had been washed with an equal volume of 0.15 mol/L NaCl and centrifuged at 50 000g for 30 minutes at 20°C; the lipids of the supernatant were then collected and 0.15 mol/L NaCl was added to obtain the initial volume. A 1.5-mL volume of LpB (2 mg/mL apo B) dialyzed against 0.15 mol/L NaCl, pH 8.0, was incubated for 17 hours at 37°C with 600 μL of the lipid emulsion and 15 mL of lipoprotein-deficient serum to initiate lipid transfer. The modified TG-enriched LpB, named TG-LpB, was reisolated by zonal ultracentrifugation in a discontinuous gradient with a Beckman SW28 rotor (24 hours, 100 000g, 10°C). TG-LpB was then dialyzed against PBS-EDTA, concentrated on a Centricon 100 (Amicon), and sterilized by passage through a 0.22-μm Millipore filter. As checked by SDS-PAGE and Coomassie Blue staining, TG-LpB was free of C and E apolipoproteins, and apo B100 remained intact after treatment (data not shown).
Reassociation of Apolipoproteins With Lipoproteins
Association of apolipoproteins was obtained by a 1-hour incubation at 37°C of LpB and TG-LpB (0.3 to 1 mg/mL apo B) with different concentrations of apolipoproteins CI, CII, CIII2, or E3 to obtain molar ratios of about 3, 10, or 30 apolipoproteins added for 1 LpB particle.
Gel Filtration Studies
We used the fast protein liquid chromatography system (Pharmacia Sweden). Lipoproteins were separated on a 10×30-mm Superose 6 HR column (Pharmacia) equilibrated with PBS-EDTA buffer containing 1.5 10−3 mol/L NaN3. The absorbance of the eluate was monitored at 280 nm. In each run, 100 to 300 μg apoB in a maximal volume of 0.5 mL was injected and was eluted at a constant flow rate of 12 mL/h.
Lipoprotein and Apolipoprotein Labeling
125I-labeled LpB was prepared by the iodine monochloride method of MacFarlane modified by Bilheimer et al24 (specific activity between 20 and 600 cpm/ng). Addition of lipids, apolipoproteins, or both was always performed on prelabeled LpB, and for binding studies lipoproteins were not reisolated by gel filtration and free apolipoproteins were not removed except when indicated. Studies with LpB have shown us that free apolipoproteins do not interfere with apo B binding to the LDL receptor (Fig 1⇓).
Apolipoproteins CI, CII, CIII, and E were labeled by the chloramine-T method25 before addition to unlabeled LpB or TG-LpB (specific activity, 5 to 20 cpm/ng).
The HeLa cell line was cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing penicillin (100 U/mL), streptomycin (100 μg/mL), and 10% (vol/vol) fetal calf serum (FCS). Human skin fibroblasts were cultured from punch biopsies from normal volunteers. Subcultures were used between 4 and 15 passages. Fibroblasts or HeLa cells were plated in 35-mm dishes in DMEM with 10% FCS. When cultures were subconfluent the medium was changed to DMEM supplemented with 10% lipoprotein-deficient serum for 48 hours to upregulate LDL receptors (lipoprotein-deficient serum was obtained from FCS after centrifugation at d=1.21 kg/L).
Binding and Cell Association of 125I-Labeled LpB
Binding studies were performed on HeLa cells as previously described.11 Increasing concentrations of 125I-LpB or modified 125I-LpB (1 to 25 μg apo B/mL) were incubated for 2 hours at 4°C with HeLa cells chilled on ice in the presence of fresh lipoprotein-deficient serum (10%) in the medium with or without an excess (30-fold) of unlabeled LDL to determine the nonspecific binding. Cells were then washed and cell-associated radioactivity was counted. To determine the uptake of 125I-LpB or modified 125I-LpB by fibroblasts,26 cells were incubated for 4 hours at 37°C with labeled lipoproteins (10 μg/mL apo B) in the absence or presence of an excess of unlabeled LDL. Cells were then washed and cell-associated radioactivity (binding and internalization) was counted.
Results are expressed as nanograms of apo B bound or associated (bound and internalized) per milligram of cellular protein after subtraction of nonspecific values. A control of LpB incubated in the same conditions as for TG-LpB but without lipid addition was done. Its binding and cellular association were not significantly different from those of fresh LpB.
Enzymatic kits for determination of cholesterol, phospholipids, and TGs were from Boehringer Mannheim GmbH. For TGs, free glycerol was eliminated from the sample in a preliminary reaction.
Association of apo CIII and apo E to LpB was checked by nondenaturing electrophoresis on ready-to-use 4% to 30% polyacrylamide gels (Pharmacia). Twenty micrograms of LpB was spotted in each well. After electrophoresis, the proteins were transblotted electrophoretically (26 hours, 100 mA) on nitrocellulose paper27 and then incubated overnight at 4°C in 20 mmol/L Tris (pH 8)/90 mmol/L NaCl/Tween 20 (1 g/L)/SDS (1 g/L) with mouse monoclonal antibodies against apo E20 or with rabbit polyclonal antibodies against apo CIII. After being washed they were finally incubated with peroxidase-labeled sheep anti-mouse or goat anti-rabbit immunoglobulins (Pasteur Diagnostic) for 2 hours at room temperature. The enzymatic reaction was started by addition of 1 mmol/L 4-chloro-1-naphthol and 50 μL H2O2.
TG Enrichment of LpB
Incubation of LpB with the TG-rich emulsion and reisolation by gradient ultracentrifugation induced a reproducible modification of the lipid content of LpB (Table 1⇓). The recovery after one or two isolation steps was 85±2.8% (mean±SD) of total apo B (mean of six experiments). Reisolation of TG-LpB by a further step of gradient ultracentrifugation did not modify the TG-LpB composition or the apo B yield. The lipid content of TG-LpB depends on the initial LpB lipid content, which is variable16 but always corresponds to a 10- to 15-fold increase in TGs, associated with a 50% to 60% decrease in cholesterol, without significant change in phospholipid content. The cholesterol decrease corresponds to a decrease of both esterified and free cholesterol to the same extent.
The gel filtration profile of LpB was modified by TG enrichment, as shown in Fig 2⇓, but there was no modification of the elution volume of each peak. The cholesterol-TG ratio in LpB was higher in the small LpB. After TG enrichment, the cholesterol-TG ratio was the same in small and large LpB, indicating that both were enriched in TGs. However, the modification of the profile reflects a partial redistribution of LpB particles between small and large sizes.
Interaction of TG-LpB with the LDL receptor was tested on HeLa cells at 4°C (Fig 3⇓). For these experiments LpBs were radiolabeled before TG enrichment, so the specific radioactivity was the same for LpBs and TG-LpBs. For each apo B concentration used, we observed a decrease of the apoB binding to the LDL receptor induced by triglyceride addition.
Apo CIII and Apo E Association to LpB
LpB was incubated for 1 hour at 37°C with apo CIII or apo E in a ratio of 30 apo CIII for 1 apo B or 8 apo E for 1 apo B. Free apo CIII and apo E were removed by filtration on a Centricon 100. The association of apo CIII or apo E was demonstrated by nondenaturing PAGE (4% to 30% gel) and immunoblot of the gel with polyclonal anti-CIII or anti-E (Fig 4⇓).
The gel filtration profile on the 10×30-mm Superose 6 HR column did not show any difference between LpB (Fig 2A⇑) and reconstituted LpB-E and LpB-CIII (data not shown).
Fig 1⇑ shows the interaction of LpB-CIII (30 apo CIII added for 1 LpB) and LpB-E (8 apo E added for 1 LpB) with the LDL receptor of HeLa cells at 4°C. LpB was radiolabeled before incubation with apo CIII or apo E. Free apolipoproteins were removed by ultrafiltration. Apo CIII decreased the binding of LpB with the LDL receptor, whereas apo E had the opposite effect of increasing the affinity. These variations were similar but less than those observed for natural LpB-E or LpB-CIII isolated from plasma by immunoaffinity.11 The same experiment, performed without removal of free apolipoproteins, produced the same results, indicating that free apolipoproteins do not interfere with apo B binding to the LDL receptor (Fig 1⇑). A control with apo CIII added extemporaneously to 125I-LpB in the binding medium shows that no effect can be obtained without previous incubation with the lipoproteins. Free apo E does not interfere with the LpB binding, as also shown in Fig 1⇑ and as previously reported.28
Association of Apolipoproteins CI, CII, CIII, and E to TG-LpB
As for LpB, association of C or E apolipoproteins with TG-LpB could be obtained by a simple incubation at 37°C. We tried to quantitatively evaluate the level of associated apolipoproteins when increasing concentrations of free apolipoprotein were added to TG-LpB. For this purpose, apolipoproteins CI, CII, CIII, and E were radiolabeled and then incubated for 1 hour at 37°C with LpB. The mixture was injected onto a Superose 6 HR column. The distribution of lipoprotein was checked at 280 nm and fractions were counted for radioactivity. In Table 2⇓ the results obtained for the different preparations are presented. Apo CI and CIII remained associated with TG-LpB almost to the same ratio, whereas larger quantities of apo CII could be associated. A large excess of apo E (30 apo E for 1 apo B) did not allow the association of more than 7 apo E for 1 apo B.
When radiolabeled apo CIII and unlabeled apo E were added together to TG-LpB, apo E did not impair the apo CIII association; conversely, when apo E was radiolabeled, addition of apo CIII decreased the apo E’s association with TG-LpB (from 4.14 to 2.09 or 2.62). A very interesting observation was also made regarding the gel filtration profile. Neither apo CIII nor apo E modified the A280 gel filtration profile of TG-LpB, and labeled apolipoproteins were distributed as apo B between large and small TG-LpBs (Fig 5A⇓, 5B⇓, and 5C⇓). Apo CI did not modify the lipoprotein distribution but was more associated with the large TG-LpBs than apo CIII or apo E (Fig 5F⇓). However, when apo CII was added, there was a different distribution of large and small TG-LpBs, with an increased peak of large TG-LpB. An equivalent repartition of apo CII between the two peaks was obtained (Fig 5D⇓). The increase of the A280 peak of large TG-LpB with addition of CII could be due to the different repartition of TG-LpBs between large and small particles or to the formation of large particles with only lipids and apo CII. To choose between the two hypotheses, we injected onto the column TG-LpB labeled before TG enrichment and preincubated with (Fig 5E⇓) or without (Fig 5A⇓) apo CII (30 apo CII for 1 apo B) and counted the apo B radioactivity eluted. The comparison of Fig 5A⇓ and 5E⇓ shows that apo B distribution between large and small particles was strongly modified by apo CII, suggesting that the large particles contained both apo B and apo CII.
Interaction at 4°C With the LDL Receptor of TG-LpB Associated With Apolipoproteins CI, CII, CIII, and E
To study apo B binding to the cellular LDL receptor, we used HeLa cells, which are described as having LDL receptor expression at their surface independent of the passage number.11 LpB was radiolabeled, enriched with lipids, and then incubated with apolipoproteins CI, CII, CIII, and E. Preliminary studies indicated that free apolipoproteins did not interfere with apo B binding to the LDL receptor (V. Clavey, PR, et al, unpublished data, 1994), so we did not remove free apolipoproteins. These modified lipoproteins were incubated with HeLa cells for 2 hours at 4°C. The specificity of the interaction with the LDL receptor was always verified by incubation with a 30-fold excess of unlabeled LDL. Nonspecific binding was less than 15% of maximal TG-LpB binding. As shown in Fig 3⇑, the binding of TG-LpB was always less than the binding of LpB. Apo E did not significantly modify the TG-LpB binding to the LDL receptor (Fig 6A⇓). Apo CI had a minimal effect in decreasing TG-LpB binding, even with a 30-fold excess of apo CI (30 apo CI for 1 apo B) (Fig 6B⇓). Apo CII (Fig 6C⇓) and apo CIII (Fig 6D⇓) produced a drastic inhibition of TG-LpB interaction with the LDL receptor, but apo CIII was slightly more effective than apo CII. The inhibitory effect increased with the apo CIII:apo B ratio.
Effect of Lipids and Apolipoproteins on the Apo B Internalization by the LDL Receptor on Fibroblasts
After we had demonstrated the effect of lipids and apolipoproteins on apo B binding to the LDL receptor, we wanted to see whether any effect was observed on the cellular uptake of these complex particles. We chose to study fibroblast cells, which are widely used for this purpose.26 After preincubation with apolipoproteins, labeled LpB or TG-LpB was incubated for 4 hours at 37°C with cells whose LDL receptor was upregulated. Cell surface binding and internalization were measured together (cell association). Results are shown in Fig 7⇓. Lipid association to LpB decreased cell association by 49±6% in either the absence or the presence of other apolipoproteins. C Apolipoproteins decreased cell association of LpB and TG-LpB, with a maximal effect for apo CII, while apo E increased it by about 210%. Simultaneous addition of apo CIII and apo E impaired the stimulating effect of apo E, especially in TG-LpB.
The aim of this study was to prepare and characterize LpBs enriched with lipids, apolipoproteins, or both and to evaluate the effect of these modifications on their interactions with the LDL receptor.
In agreement with our previous experiments,16 immunoaffinity-purified LpBs (containing apo B as the sole apolipoprotein) were heterogeneous in size (Fig 2⇑). The small first peak corresponds to large particles of VLDL size and the bigger peak contains most of the LpBs, including the small particles of LDL size. In most preparations, a small intermediary peak was found that could contain IDLs.
After incubation of LpB with lipid emulsions, stable TG enrichment was obtained for small and large LpBs. An increase of the peak of large LpBs was found without change in the elution volume of each peak (Fig 2B⇑). The TG enrichment was associated with an important decrease of LpB binding to the LDL receptor at 4°C and of LpB internalization at 37°C (Figs 3⇑ and 7⇑). It was not possible to obtain enough large LpBs and large TG-LpBs to evaluate whether the binding inhibition was the same for large and small LpBs. Thus, we estimate that the overall effect was probably due to the enrichment of the small LpBs, which were in large excess. Previous studies with TG-rich VLDL9 10 13 or with LDL8 indicated that lipids and particularly TGs have by themselves an inhibitory effect on the apo B binding to the LDL receptor; our experiments, allowing the evaluation of the specific effect of lipids, confirmed these findings for LpB. Discrete changes in apo B conformation and in apo B–lipid interaction may explain the difference of interaction with the LDL receptor. The temperature phase transition of LDL that Deckelbaum et al29 demonstrated to occur at 30°C did not seem to affect this conformational change, because the decrease in the apo B–receptor interaction was obtained at both 4°C and 37°C. A recent publication by Galeano et al30 showed that small TG-rich LDLs exist in vivo and have a low affinity for the LDL receptor.
Association of C apolipoproteins to LpB and TG-LpB was demonstrated by immunoblotting after electrophoresis under nondenaturing conditions and by gel filtration of a mixture of TG-LpBs and labeled apolipoproteins. Apo CIII preferentially associated with small TG-LpBs. By contrast, apo CI and apo CII were largely associated with large TG-LpBs. The addition of apo CII produced an enlargement of the first peak, suggesting an increasing amount of large particles. This was confirmed by measuring the apo B distribution in the absence (Fig 5A⇑) or presence (Fig 5E⇑) of apo CII, the latter of which induced an increase of apo B in the first peak. The comparison of Fig 5D⇑ and Fig 5E⇑ shows that the ratio of apo CII to apo B in large LpBs was higher than in small LpBs. The specific modification of apo B distribution between large and small TG-LpBs after apo CII addition was reproducible and very surprising. Several hypotheses can be advanced: one would be an aggregation of small TG-LpBs. Another would be a lipid redistribution between the two populations: this was not confirmed by lipid analysis, which did not show any modification of the TG-cholesterol ratio in the two peaks after apo CII addition (results not shown). Another hypothesis would be a fusion of small particles induced by apo CII addition, leading to larger lipoproteins.
Association of C apolipoproteins with LpB or TG-LpB was found to alter the apo B binding to the LDL receptor. In previously published studies apo CIII has often been implicated in the inhibition of lipoprotein interaction with the LDL receptor,13 31 but the influence of other C apolipoproteins was never demonstrated. For the different C apolipoproteins we did not find a correlation between the level of their association and their inhibitory effect on the apo B–LDL receptor interaction: apo CI association, which was almost in the same range as that of apo CIII association, did not inhibit TG-LpB binding to the LDL receptor at 4°C but partially impaired its catabolism at 37°C. Apo CII, which was associated to a greater degree with TG-LpB than was apo CIII, did not produce a higher level of inhibition than apo CIII. One possible explanation was that apo CII was mostly associated with the large TG-LpBs. All of these results suggest that the C apolipoproteins did not randomly bind to the surface of apo B–containing lipoproteins. Association of each type of apo C must be specific and also must specifically modify the apo B accessibility to the LDL receptor.
Reconstituted lipoproteins were a good tool in the study of the specific effect of each apolipoprotein, but the physiological importance of our results have still to be demonstrated. In vivo, such lipoproteins with only apo B and apolipoproteins CI, CII, or CIII are either not present in large amounts or have only a very short half-life. Indeed, when we prepared LpB:CIII by immunoaffinity,11 apo CI and apo CII were always associated with LpB:CIII, which would then be better called LpB:Cs. In vivo, C apolipoproteins are transferred from HDL to VLDL and vice versa during TG-rich lipoprotein metabolism, and our results show that apo Cs, when present, can inhibit the catabolism of these lipoproteins by the LDL receptor pathway; this catabolism, then, can be increased after the transfer of apo Cs from TG-rich lipoproteins to HDL. It is possible that in vivo, the distribution of apo Cs is not the same among the different lipoproteins and that distribution would play a role in individual lipoprotein catabolism during and after lipolysis.
Association of apo E to LpB and to TG-LpB was much lower than that of C apolipoproteins. About 6 apo Es remained associated to 1 apo B when excess apo E was added. Prepared by immunoaffinity from plasma, the LpB:E without C apolipoproteins is also very poor in apo E,11 but complex lipoproteins can be obtained that contain 1 apo B, 13 apo CIII, and 20 apo E and also apo CI and apo CII. We tried to add apo E and apo CIII simultaneously to TG-LpB. Data in Table 2⇑ show that in vitro, apo CIII partially impaired the apo E association to TG-rich lipoproteins, indicating that these reconstituted lipoproteins did not mimic exactly the naturally occurring particles.
Previously we demonstrated that the apo E content of lipoproteins (LpB:E or LpB:C:E) could increase their affinity for the LDL receptor.11 This was also seen with reconstituted LpB-E or TG-LpB-E. But the poor association of apo E with these lipoproteins in vitro was not enough to promote high cooperative binding to the LDL receptor at 4°C. However, at 37°C the effect of apo E on the LpB or TG-LpB internalization in fibroblasts was much more pronounced. This activation of LpB internalization was partially impaired by apo CIII. That is probably because apo CIII was shown to decrease the apo E association with the lipoproteins (Table 2⇑). A direct inhibition of apo E binding to the LDL receptor by apo CIII cannot be excluded. Recently, Sehayek and Eisenberg12 demonstrated that VLDLs that lack the capacity to interact with the receptors could be internalized if enriched with exogenous apo E3. They have also shown that the degree of cell metabolism is determined by the ratio of apo E to apo C. They concluded that apo C dramatically depressed the interaction of IDL with the fibroblast receptor through apo E but had a moderate effect on apo B100. In our study we have shown that apo CII and CIII have a direct inhibitory effect on the apo B binding at 4°C and at 37°C and that this effect is synergistic with the lipid effect. In contrast with previous experiments done elsewhere, our approach presents the advantage of allowing us to dissociate the effect of lipids and apolipoproteins without interference of endogenous apolipoproteins other than apo B.
In summary, the use of isolated LpBs enriched in vitro by TGs and apolipoproteins represents an interesting approach for studying the lipoprotein-receptor interaction. The results demonstrated that the LDL receptor interaction of apo B was impaired and suggested that the receptor domain of apo B may be masked either by lipids or by C apolipoproteins (particularly apo CII and apo CIII). The addition of apo E to such modified lipoproteins could restore their catabolism by the LDL receptor, but then this catabolism could become apo E dependent and not apo B dependent.
We thank Dr Dallongeville for valuable discussions, B. Derudas for his excellent technical assistance, and P. Kelly for rewriting the manuscript.
Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science. 1986;232:34-47.
Milne R, Theolis R Jr, Maurice R, Pease RJ, Weech PK, Rassart E, Fruchart JC, Scott J, Marcel YL. The use of monoclonal antibodies to localize the low density lipoprotein receptor-binding domain of apolipoprotein B. J Biol Chem. 1989;264:19754-19760.
Fantappié S, Corsini A, Sidoli A, Uboldi P, Granata A, Zanelli T, Rossi P, Marcovina S, Fumagalli R, Catapano AL. Monoclonal antibodies to human low density lipoprotein identify distinct areas on apolipoprotein B100 relevant to the low density lipoprotein receptor interaction. J Lipid Res. 1992;33:1111-1121.
Soria LF, Ludwig EH, Clarke HRG, Vega GL, Grundy SM, McCarthy BJ. Association between a specific apolipoprotein B mutation and familial defective apo B100. Proc Natl Acad Sci U S A. 1989;86:587-591.
Marcel YL, Innerarity TL, Spilman C, Mahley RW, Protter AA, Milne RW. Mapping of human apolipoprotein B antigenic determinants. Arteriosclerosis. 1987;7:166-175.
Aviram M, Lundkatz S, Phillips MC, Chait R. The influence of the triglyceride content of low density lipoprotein on the interaction of apolipoprotein B100 with cells. J Biol Chem. 1988;263:16842-16848.
Nigon F, Lesnik P, Rouis M, Chapman MJ. Discrete subspecies of human low density lipoproteins are heterogeneous in their interaction with the cellular LDL receptor. J Lipid Res. 1991;32:1741-1753.
Bradley WA, Hwang SC, Karlin JB, Lin AH, Prasad SC, Gotto AM, Gianturco SH. Low density lipoprotein receptor binding determinants switch from apolipoprotein E to apolipoprotein B during conversion of hypertriglyceridemic very low density lipoproteins to low density lipoproteins. J Biol Chem. 1984;259:14728-14735.
Bradley WA, Gianturco SH. Apo E is necessary and sufficient for the binding of large triglyceride-rich lipoprotein to the LDL receptor: apo B is unnecessary. J Lipid Res. 1986;27:40-48.
Agnani G, Bard JM, Candelier L, Delattre S, Fruchart JC, Clavey V. Interaction of LpB, LpB:E, LpB:C-III and LpB:C-III:E lipoproteins with the low density lipoprotein receptor of HeLa cells. Arterioscler Thromb. 1991;11:1021-1029.
Sehayek E, Eisenberg S. Abnormal composition of hypertriglyceridemic very low density lipoprotein determines abnormal cell metabolism. Arterioscler Thromb. 1990;10:1088-1096.
Evans AJ, Huff MH, Wolfe BM. Accumulation of an apo E-poor subfraction of very low density lipoprotein in hypertriglyceridemic men. J Lipid Res. 1989;30:1691-1701.
Eisenberg S, Friedman G, Vogel T. Enhanced metabolism of normolipidemic human plasma very low density lipoprotein in cultured cells by exogenous apolipoprotein E-3. Arteriosclerosis. 1988;8:480-487.
Sehayek E, Eisenberg S. Mechanism of inhibition by apolipoprotein C of apolipoprotein E-dependent cellular metabolism of human triglyceride-rich lipoproteins through the low density lipoprotein receptor pathway. J Biol Chem. 1991;266:18258-18267.
Mezdour H, Vu Dac N, Parra HJ, Luc G, Luyeye I, Fruchart JC. A selective bisite immunoenzymometric procedure for human Lp(a) particle quantification using monoclonal antibodies against apo (a) and apo B. J Lipid Res. 1989;30:1437-1443.
Fruchart JC, Fievet C, Puchois P. Methods of enzymatic analysis: apolipoproteins. Bergmeyer 3rd ed. Weinheim, FRG: Verlag, Chemie; 1985:126-138.
Petersen GL. A simplification of the protein assay method of Lowry which is more generally applied. Anal Biochem. 1977;83:343-356.
Goldstein JL, Brown MS. Binding and degradation of low density lipoproteins by cultured human fibroblasts: comparison of cells from a normal subject and from a patient with homozygous familial hypercholesterolemia. J Biol Chem. 1974;249:5153-5162.
Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheet: procedure and some applications. Proc Natl Acad Sci U S A. 1979;76:4350-4354.
Deckelbaum RJ, Shipley GG, Small DM. Structure and interactions of lipids in human plasma low density lipoproteins. J Biol Chem. 1977;252:744-752.
Galeano NF, Rumsey SC, Preud’Homme DL, Gleeson A, Deckelbaum RJ. Small LDL from hypertriglyceridemic subjects have increased non-LDL receptor low density affinity cell binding. Circulation. 1993;88(suppl I):I-366. Abstract.
Ginsberg HN, Le NA, Goldberg IJ, Gibson JC, Rubinstein A, Wang-Iverson D, Norum R, Brown WV. Apolipoprotein B metabolism in subjects with deficiency of apolipoprotein C-III and A-I. J Clin Invest. 1986;78:1287-1295.