INTRODUCTION

Lipoprotein lipase (LPL)¹ catalyzes the hydrolysis of triglycerides in plasma chylomicrons and very low density lipoproteins (VLDL) (1, 2, 3, 4). In humans, deficiency of either LPL or apolipoprotein (apo) CII, an essential cofactor for LPL's lipolytic activity, results in severe elevation in plasma chylomicrons and large VLDL (5). In addition to its lipolytic actions, LPL is also a ligand for receptors belonging to the family of low density lipoprotein (LDL) receptors, including LDL receptor-related protein (LRP)/₂-macroglobulin (₂M) receptor, GP330/LRP-2, and VLDL receptors (6, 7, 8, 9, 10, 11, 12, 13). LPL can influence lipoprotein catabolism by simultaneously binding to both lipoproteins and these cell surface receptors (6, 7, 8, 9, 10, 11, 12, 13). LPL promotes VLDL catabolism in LDL receptor-lacking mutant fibroblasts, and ligands specific for LRP inhibit LPL-stimulated VLDL degradation (9). LPL promotes lipoprotein binding to purified LRP (9) and GP330/LRP-2 (11) in solid-phase assays. Argraves et al. (12) demonstrated that transfection of PEA13 cells with VLDL receptors enhances their ability to internalize and degrade ¹²⁵I-LPL. Takahashi et al. (13) showed that in chinese hamster ovary cells co-transfected with the VLDL receptor and LPL, binding and degradation of ¹²⁵I-VLDL is much greater than in cells transfected with VLDL receptor alone. These studies demonstrate that LPL-dependent VLDL catabolism is mediated by LRP (6, 7, 8, 9, 10), GP330/LRP-2 (10, 11), and VLDL receptors (12, 13). LPL also binds to cell surface heparan sulfate proteoglycans (3, 14, 15, 16). LPL binding to either heparan sulfate proteoglycans or receptors can be separated from its lipolytic activity as indicated by studies with LPLC, the carboxyl-terminal, noncatalytic domain of LPL (8, 17, 18).

There are conflicting data regarding the ability of LPL to promote lipoprotein catabolism via LDL receptors (19, 20, 21, 22, 23). LDL receptors are the major mediators of LDL clearance from plasma, but they have also been implicated as mediators of VLDL clearance via apoE (24). Aviram et al. (19) concluded that LPL enhances LDL uptake and degradation in both fibroblasts and human monocyte-derived macrophages via LDL receptors. They suggest that LPL's lipolytic activity is necessary for its ability to enhance LDL degradation, but they did not study the effect of LPL on the catabolism of VLDL, which is a better substrate for LPL than LDL (25). Several investigators determined that LPL-mediated LDL degradation is independent of the LDL receptor and is not regulated by factors affecting LDL receptor expression (20, 21, 22). These studies also focused on the effect of LPL on LDL rather than VLDL catabolism. Mulder et al. (23) concluded that internalization of LPL·VLDL complexes in HepG2 cells was mediated by LDL receptors, since it was negligible in LDL receptor-lacking fibroblasts. However, they did not study lipoprotein degradation or the ability of LPL itself to bind to LDL receptors. The relative contributions of LRP versus LDL receptors to the effects of LPL have not been clearly shown in cells expressing both receptors.

The present studies were aimed at resolving these discrepancies. Normal human foreskin fibroblasts (FSF cells) were studied in which both LRP and LDL receptor-dependent pathways are active. Data were also obtained in LDL receptor-lacking (FH) human fibroblasts, HUVECs, and normal (MEF) and LRP-deficient (PEA13) murine embryonic fibroblasts. None of these cell lines is known to express GP330/LRP-2, but HUVECs express VLDL receptors. Our results establish a significant role for LDL receptors in LPL-mediated binding and degradation of VLDL. In addition, both solid-phase assays using immobilized LDL receptors and internalization studies using protein-free lipid emulsions demonstrate that LPL binds directly to LDL receptors and thereby may contribute to the catabolism of native VLDL via the LDL receptor pathway.

EXPERIMENTAL PROCEDURES

VLDL with S_f 100-400 were isolated as described previously by ultracentrifugation of plasma from fasted normolipidemic human subjects with the most common apoE phenotype (E3/3) (26). These are the largest, most triglyceride-rich particles present in the fasted state (27). Bovine milk LPL was isolated using a published method (28). The carboxyl-terminal fragment of human LPL (amino acid residues 313-448) was produced in Escherichia coli as a fusion protein with glutathione S-transferase (GST) as described previously (18) and designated as GST-LPLC. Using site-directed mutagenesis, tryptophan residues at positions 393 and 394 were changed to alanine to generate GST-LPLC_ww (17, 18) and tryptophans 390, 393, and 394 were replaced with alanine to generate GST-LPLC_www. Recombinant human receptor-associated protein (RAP) was produced as described previously (29). A monoclonal antibody against the catalytic domain of LPL (mAb7) was prepared from a hybridoma kindly provided by Dr. Louis Smith (Methodist Hospital, Houston, TX) (30). Monoclonal antibodies IgG-C7 and IgG-4A4 directed against the first cysteine-rich repeat and the cytoplasmic terminal 14 amino acids, respectively, of the LDL receptor were obtained as described previously (31). Polyclonal rabbit antihuman LRP antibody Rb777 was produced as described previously (7). Human ₂M was isolated and activated by methylamine treatment as described previously (32). Monoclonal antibodies and ₂M were ¹²⁵I-labeled using IODO-GEN (Pierce) as described previously (18). Iodination of S_f 100-400 particles to specific activities of 300-500 cpm/ng was done by the iodine-monochloride method (33). Protein-free particles with S_f 100-400 were isolated from a 10% intralipid emulsion (Travenol) by ultracentrifugal flotation (26), and their triglyceride content was estimated by an enzymatic calorimetric assay (Sigma) (34). They were labeled with [³H]cholesteryl oleyl ether, a nondegradable marker of cellular uptake (21). A glass tube containing 0.5 ml of minimum Eagle's medium, 4 mg/ml BSA and 35 µCi of [³H]cholesteryl oleyl ether (Amersham Corp.) was sonicated for 10 min at room temperature. Intralipid particles with S_f 100-400 containing 3-4 mg of triglycerides were added, and the mixture was incubated at 37 °C for 20 min and then returned to room temperature. This treatment resulted in the incorporation of [³H]cholesteryl oleyl ether in the emulsion. The tritiated lipid emulsions were stored at 4 °C overnight before use. The sizes of S_f 100-400 particles from plasma and those from Intralipid were similar as indicated by scanning electron microscopy on a Hitachi S-4000 instrument (data not shown) using previously described techniques (35).

Normal human foreskin fibroblasts (FSF cells) were cultured as described (36, 37). Mutant skin fibroblasts that do not express LDL receptors (FH cells) (38), were obtained from the National Institute of General Medical Sciences Human Genetic Mutant Cell Repository (GM00486A), Camden, NJ. MEF and mutant MEF cells that lack LRP (PEA13) (39) were provided by Dr. Joachim Herz (Dallas, TX). HUVECs were isolated from human umbilical cords as described previously (40). LDL receptors were usually up-regulated by incubation prior to the assay for 48 h with media containing 2 mg/ml lipoprotein-deficient serum (LPDS) and 24 h in the presence of 1 µg/ml of lovastatin (41, 42). In some studies with murine fibroblasts, lovastatin was omitted and the duration of LPDS treatment was reduced to 24 h. In a few experiments, as indicated in the figure legends, LDL receptors were down-regulated by the addition of 20 µg/ml LDL (41). Surface binding to metabolically inactive cells was studied after incubating cells with ¹²⁵I-ligands for 3 h at 4 °C as described previously (36, 37). Steady-state ligand internalization and degradation were measured after incubating cells with radiolabeled ligands at 37 °C for 5 h (36, 37). Degradation was defined as the trichloroacetic acid-soluble radioactivity in the incubation medium. Internalization was defined as radioactivity that remained cell-associated after incubating cells at 4 °C in buffer containing 10 mg/ml tripolyphosphate. Total protein in each well as determined by the Lowry assay (43) varied by less than 15% within each experiment. Total DNA in each well as determined spectrofluorometrically using the DNA-binding dye Hoechst 33258 (44) varied by <10% within each experiment. Wells treated with lovastatin or LPDS contained 30-60% of the protein and DNA amounts present in untreated wells. Thus, when the two treatments were compared, results were corrected for protein or DNA per well. When LDL receptors were not up-regulated, results are presented in units of ng/well.

LDL receptors were partially purified by fractionation of total cell extracts over DE52-cellulose (Whatman) as described previously (45). Microtiter wells (96-well plates, Immulon 2, Dynatech) were coated for 30 min at 37 °C with 100 µl of buffer containing 3 µg of anti-LDL receptor antibody, IgG-4A4. The wells were then blocked with 1% BSA at 37 °C for 1 h. LDL receptors from DE52 eluants were specifically immobilized by incubating IgG-coated wells for 16 h at 4 °C with 0-100 µg of DE52 eluant protein in 100 µl of buffer containing 50 mM Tris-HCl, pH 8.0, 2 mM CaCl₂, 0.5% BSA (46). To examine the direct binding of LPL to LDL receptors, the wells were washed and incubated first for 3 h at 4 °C with buffer containing increasing concentrations of LPL then for 2 h at 4 °C with 5 µg/ml of ¹²⁵I-mAb7. After washing, bound ligands were desorbed in 0.1 N NaOH and quantitated for radioactivity. The extent of LPL binding to LDL receptors was reflected by the amount of ¹²⁵I-mAb7 bound. In some experiments, immobilized LDL receptors were incubated with ¹²⁵I-S_f 100-400 in the absence or presence of LPL or GST-LPLC.

LPL binding to LDL receptors was also detected by an enzyme-linked immunoabsorbant assay as described previously (46). Wells were coated with 0-10 µg of LPL for 5 h at 4 °C, blocked with 1% BSA for 2 h at 4 °C, washed, and incubated for 16 h at 4 °C with 100 µl of buffer with or without 1 µg/µl DE52 eluant (partially purified LDL receptor preparation). Then, wells were washed and incubated for 2 h at 4 °C with 100 µl of buffer containing 30 µg/ml of anti-LDL receptor antibody (IgG-C7) followed by another 2-h incubation at 4 °C with 100 µl of a 1:10,000 dilution of alkaline phosphatase-conjugated rabbit anti-mouse IgG (Sigma). Wells were then washed and 100 µl of substrate (1 mg/ml p-nitrophenyl phosphate, Sigma) was added. After 30 min at 37 °C, relative LDL receptor binding was estimated by measuring the absorbance at 405 nm. As a control, nonimmune mouse IgG was used instead of IgG-C7. Standard curves prepared simultaneously by coating wells with known amounts of IgG-C7 were linear (correlation coefficients >0.98). All data from cellular and solid-phase assays were measured in duplicate.

RESULTS

Normal foreskin fibroblasts (FSF cells) were used to investigate the role of LDL receptors in LPL-stimulated degradation of VLDL particles. Lovastatin treatment of FSF cells up-regulated their LDL receptors and increased ¹²⁵I-LDL binding and degradation by 8-10-fold (data not shown and Ref. 36). Lovastatin had no effect on LRP expression as indicated by catabolism of ¹²⁵I-labeled activated ₂M (₂M*) (data not shown and Ref. 47). In five control experiments, we found that expression of catabolic data in terms of cellular DNA instead of protein gave results with identical interpretation. This was true in the presence or absence of lovastatin/LPDS treatment. Half-maximal LPL-induced ¹²⁵I-VLDL binding to cell surfaces at 4 °C shifted to a ~5-fold lower LPL concentration when LDL receptors were up-regulated by lovastatin (Fig. 1A). This surprising result suggests that binding affinity was dependent on the number of LDL receptors, as would be expected if LPL simultaneously interacted with multiple LDL receptors. Another interesting observation in Fig. 1A is that GST-LPLC, a fusion protein containing the carboxyl-terminal segment (residues 313-448) of LPL, was also able to enhance VLDL binding in a dose-dependent and lovastatin-regulated manner, albeit with a lower maximal binding. This indicates that the carboxyl-terminal domain of LPL, which lacks catalytic activity, contributes to LPL's interaction with LDL receptors. LPL also significantly promoted degradation of ¹²⁵I-VLDL by LDL receptors in a dose-dependent manner (Fig. 1B). Up-regulation of LDL receptors enhanced LPL-dependent degradation up to 4-fold. Either with or without lovastatin treatment, significantly higher concentrations of GST-LPLC than LPL were required to enhance ¹²⁵I-VLDL degradation, and the magnitude of stimulation was much lower than that obtained with LPL.

To further examine the contribution of LDL receptors to LPL-dependent VLDL degradation, studies were performed in cells lacking LRP. Results in Fig. 2 show degradation of ¹²⁵I-LDL and ¹²⁵I-₂M* by wild-type murine embryonic fibroblasts (MEF cells) and by mutant MEF cells lacking LRP (PEA13 cells). Unlike MEF cells, PEA13 cells are completely devoid of LRP activity and are unable to catabolize ¹²⁵I-₂M* (Fig. 2A). In contrast, ¹²⁵I-LDL degradation is comparable in both cell lines and is stimulated 5-7-fold by lovastatin treatment (Fig. 2B). This indicates that the lack of LRP in PEA13 cells does not affect the level of LDL catabolism or its up-regulation by lovastatin. As expected, catabolism of ¹²⁵I-₂M* was not affected by lovastatin treatment.

I-

We compared the effect of LPL and lovastatin treatment in normal (FSF) and LDL receptor-lacking (FH) human fibroblasts and wild-type (MEF) and LRP-deficient (PEA13) murine embryonic fibroblasts (Fig. 3). LPL promoted ¹²⁵I-VLDL degradation in a dose-dependent manner in untreated FSF cells. LPL stimulated VLDL degradation up to 15-fold in lovastatin-treated normal fibroblasts. In three separate experiments, up-regulation of LDL receptors with lovastatin resulted in a 4-5-fold increase in LPL-induced ¹²⁵I-VLDL degradation by FSF cells (Fig. 3A). In control experiments (data not shown), catabolism of ¹²⁵I-LDL in the absence of LPL was increased by 8-10-fold by lovastatin treatment. These data suggest that LDL receptors mediate the vast majority of LPL-dependent ¹²⁵I-VLDL degradation in lovastatin-treated cells.

Previously, we showed that LPL induced uptake of ¹²⁵I-VLDL via LRP (7, 9), and this finding was confirmed here (Fig. 3B) with a 10-20-fold induction of ¹²⁵I-VLDL catabolism by gradually increasing the concentration of LPL. However, compared with normal fibroblasts, the magnitude of ¹²⁵I-VLDL degradation was ~10-fold lower and not significantly altered by lovastatin treatment. Even in the absence of lovastatin treatment, normal fibroblasts displayed 2.5-4-fold higher catabolism in the presence of LPL than FH cells (Fig. 3, A and B). This difference probably reflects degradation by the LDL-receptor pathway, since degradation of ¹²⁵I-₂M* by normal and FH cells was nearly identical, indicating that both cell lines express similar levels of LRP (data not shown). LPL-dependent ¹²⁵I-VLDL degradation by LRP-deficient PEA13 cells was up-regulated by lovastatin treatment (Fig. 3C). In the presence or absence of lovastatin treatment, LPL-induced ¹²⁵I-VLDL degradation by PEA13 cells was comparable with that seen in MEF cells. Taken together, these results suggests that a majority of LPL-dependent VLDL catabolism occurs via the LDL receptor pathway.

We next studied the catabolism of increasing concentrations of ¹²⁵I-VLDL in the presence of a fixed LPL concentration (Fig. 4). At all concentrations of ¹²⁵I-VLDL, degradation was enhanced severalfold by the presence of 1 µg/ml LPL in FSF, HUVECs, MEF, and PEA13 cells. LPL-dependent degradation of ¹²⁵I-VLDL was strongly up-regulated by lovastatin treatment in all four cell lines, suggesting a role for LDL receptors. Unlike normal fibroblasts, HUVECs express very low levels of LRP and are unable to catabolize significant amounts of ¹²⁵I-₂M* (data not shown and Ref. 12). Although HUVECs express VLDL receptors, these receptors are not up-regulated by lovastatin. The up-regulation of LDL receptors in HUVECs with lovastatin caused an increase in LPL-dependent VLDL catabolism that was similar to the other cell lines. LPL-induced VLDL catabolism in PEA13 cells was very similar to that in MEF cells despite the absence of LRP in the former.

In the basal state, LRP expression is constitutive and LDL receptor expression is low but not absent. To measure the relative contribution of LRP and LDL receptors to LPL-induced VLDL catabolism under basal conditions, we compared wild-type murine fibroblasts to the LRP-deficient PEA13 cells (Fig. 5). When cells were incubated with 20 µg/ml LDL to down-regulate LDL receptors prior to the assay, MEF cells displayed ~30% greater LPL-induced VLDL degradation (Fig. 5). This agrees with results shown in Fig. 3C. But in Fig. 4, C and D, LPL-induced VLDL degradation was slightly greater in PEA13 cells than in MEF cells. Thus the difference between MEF and PEA13 was too small to be seen in all experiments, despite the fact that in most experiments MEF cells displayed slightly greater LPL-dependent VLDL catabolism than PEA13 cells.

The receptor-mediated component of VLDL catabolism was measured by the ability of RAP to compete for the effects of LPL. RAP has previously been shown to inhibit catabolism via all members of the LDL receptor family (46, 48, 49, 50). As shown in Fig. 5, RAP inhibited >50% of the catabolism in both MEF and PEA13 cells. The ability of RAP to inhibit degradation in LRP-lacking PEA13 cells provides evidence that the LDL receptor pathway contributes to LPL-dependent VLDL degradation, because no other members of the LDL receptor family are expressed by these cells.

In control experiments (data not shown) LPL-stimulated ¹²⁵I-VLDL degradation was partially inhibited by anti-LDL receptor monoclonal antibody IgG-C7, anti-LPL antibody mAb7, and RAP, but not by nonimmune murine IgG or anti-LRP antibody Rb777. However, Rb777 completely inhibited degradation of ¹²⁵I-₂M*.

To determine if the ability of LPL to induce ¹²⁵I-VLDL catabolism via LDL receptors could be due to a direct interaction between LPL and LDL receptors, we studied catabolism of protein-free triglyceride and phospholipid emulsions. In these experiments, neither apoE nor apo B100 were present. Likewise, the activator of LPL, apoC-II, was not present. The emulsion was subjected to ultracentrifugation to isolate particles with S_f 100-400, which were labeled with [³H]cholesteryl oleyl ether, a nondegradable marker of cellular binding or uptake. Results in Fig. 6 show that in FSF cells, basal uptake of the emulsion was low, but LPL promoted the internalization of these particles in a dose-dependent manner, even in the absence of any apolipoproteins. This suggests a direct interaction between LPL and cell surface receptors. FSF cells treated with lovastatin were able to take up almost twice as many emulsion particles than untreated cells, suggesting an involvement of the LDL receptor pathway. The average induction by lovastatin was 1.7-fold ± 0.3 in three separate experiments. The fold induction was the same even when the data were corrected for DNA instead of protein. Likewise MEF and PEA13 cells also internalized protein-free lipid emulsion in the presence of LPL (data not shown). Internalization was ~2-fold higher in MEF and PEA13 cells treated with LPDS to up-regulate LDL receptors (data not shown).

We showed previously that in the absence of LPL, VLDL particles bind to LDL receptors but not to LRP (24). We have also shown that GST-LPLC contains both lipoprotein and receptor binding domains (17). Two tryptophans at positions 393 and 394 are required for binding lipoproteins (17). A site-directed double mutant, GST-LPLC_ww, with both tryptophans changed to alanines, and a triple mutant, GST-LPLC_www, with tryptophans at positions 390, 393, and 394, replaced with alanine were tested for their effects on VLDL binding and catabolism. GST-LPLC_ww (17) and GST-LPLC_www (data not shown) do not bind VLDL. After lovastatin treatment, the vast majority of ¹²⁵I-VLDL binding and degradation in FSF cells is LDL receptor-mediated (24). GST did not have any effect on either binding at 4 °C or degradation at 37 °C. On the other hand, GST-LPLC_ww and GST-LPLC_www inhibited both ¹²⁵I-VLDL binding (Fig. 7A) and degradation (Fig. 7B) in a dose-dependent manner. Similarly, both mutants completely inhibited ¹²⁵I-LDL degradation as well (data not shown).

Next, we tested the ability of LPL to bind directly to LDL receptors in solid-phase assays. Microtiter plates coated with increasing amounts of LPL were incubated with partially purified LDL receptors. The amount of receptor bound was then quantitated using IgG-C7, a specific anti-LDL receptor antibody. Fig. 8A shows that the amount of LDL receptor bound is directly proportional to the amount of immobilized LPL. When IgG-C7 was exposed to LPL-coated wells in the absence of added LDL receptors, there was relatively little binding. Also, when nonimmune mouse IgG was used as a control instead of IgG-C7, there was no detectable binding (data not shown). In another approach, LDL receptors from partially purified cell extracts were specifically immobilized on microtiter wells using anti-LDL receptor antibody IgG-C7 (46). The wells were subsequently incubated with increasing amounts of LPL, and the amount of LPL bound was quantitated using ¹²⁵I-labeled IgG mAb7, a monoclonal antibody against LPL (Fig. 8B). In the presence of LDL receptors, the amount of LPL bound increased with the dose of LPL added; binding of LPL in the absence of LDL receptors was significantly lower.

We investigated the ability of LPL or GST-LPLC to promote binding of ¹²⁵I-VLDL to LDL receptors in solid-phase assays (Fig. 9). These assays were done at 4 °C, a temperature which inhibits lipolytic activity. In the absence of either protein, ¹²⁵I-VLDL bound to LDL receptor-coated wells (Fig. 9A), as shown previously (24). Both LPL and GST-LPLC stimulated the binding of ¹²⁵I-VLDL to immobilized LDL receptors (Fig. 9A). Nonspecific binding to BSA-coated wells was <5% and has been subtracted from these data. In Fig. 9B, wells coated with anti-LDL receptor antibody IgG-4A4 were used to immobilize increasing amounts of LDL receptors from DE52 eluants (LDL receptor preparation) of normal fibroblasts. At a fixed concentration of ¹²⁵I-VLDL, binding was dependent on the amount of LDL receptors added and was enhanced by 2-3-fold in the presence of LPL (Fig. 9B).

DISCUSSION

It is well established that LPL promotes the catabolism of triglyceride-rich lipoproteins by a variety of cultured cells. Previous studies show that LPL promotes lipoprotein degradation via LRP (6, 7, 8, 9, 10). The VLDL receptor and GP330/LRP-2 are implicated in the process as well (10, 11, 12, 13). Here we demonstrate that LDL receptors also contribute to the LPL-dependent enhancement of lipoprotein catabolism.

The present data establish that LPL stimulates VLDL catabolism via LDL receptors and binds directly to LDL receptors. LPL induces catabolism of normal VLDL in FSF, FH, HUVECs, MEF, and PEA13 cells. Since PEA13 cells do not express either LRP, GP330, or VLDL receptors (39, 51), LPL-dependent VLDL catabolism by these cells is probably mediated by LDL receptors. Treatment with lovastatin to up-regulate LDL receptors greatly enhances LPL-stimulated VLDL degradation in all four cell lines possessing an intact LDL receptor pathway. Up-regulation of LDL receptors results in a 2-fold increase in LPL-dependent binding and uptake of protein-free intralipid emulsion by normal fibroblasts, MEF, and PEA13 cells. Since intralipid emulsion do not contain any apolipoprotein, this result provides evidence for a direct interaction between LPL and LDL receptors. We confirmed the direct binding of LPL to LDL receptors in solid-phase assays. LDL receptors bind to LPL-coated microtiter wells in a dose-dependent manner. Both LPL and a fusion protein containing its carboxyl-terminal domain (GST-LPLC) promote binding of ¹²⁵I-VLDL to LDL receptor-coated wells. A ligand specific for LRP, ¹²⁵I-₂M*, does not bind to these wells (46), eliminating the possibility that the LPL effect may be due to LRP contamination.

Studies with various competitors provide further evidence that LPL stimulates VLDL degradation via LDL receptors. Mutations of tryptophan to alanine in GST-LPLC at positions 393, 394, and 390 (GST-LPLC_www) or at positions 393 and 394 (GST-LPLC_ww) abolish its binding to VLDL without affecting binding to LDL receptors or LRP (data not shown and Refs. 16 and 17). Both mutants competitively inhibit degradation of ¹²⁵I-VLDL by normal fibroblasts in which the LDL receptor pathway is up-regulated. RAP also competes for LPL-induced ¹²⁵I-VLDL degradation in both MEF and PEA13 cells, because it is a ligand for LDL receptors (46). Similarly, antibodies against the LDL receptor, but not LRP, partially compete for LPL-mediated lipoprotein degradation (data not shown).

Lipolysis is not required for LPL-dependent promotion of VLDL binding and catabolism. This is suggested by solid-phase and cell experiments at 4 °C when the enzymatic activity is negligible. Moreover, LPL promotes the internalization of protein-free intralipid emulsions that are devoid of apoC-II, an essential co-factor for LPL's lipolytic activity. Rumsey et al. (21) also showed that LPL enhances the internalization of protein-free triglyceride emulsion particles by macrophages. GST-LPLC, a fusion protein containing the carboxyl-terminal noncatalytic domain of LPL, also stimulates ¹²⁵I-VLDL binding at 4 °C and degradation at 37 °C as was shown previously in studies on LRP (17, 18). However, GST-LPLC was significantly less potent than LPL. This may be explained by the inability of GST-LPLC to undergo dimerization (17) or optimal protein folding. Nykjaer et al. (52) showed that dimeric LPL is much more potent than monomeric LPL at stimulating ¹²⁵I-VLDL binding to immobilized LRP. The effect of LPL appears to depend on the ratio of VLDL to LPL. Thus, increasing concentrations of LPL to ~30 nM or GST-LPLC to ~700 nM induces degradation of 3.5 nM VLDL in a dose-dependent manner. At higher concentrations, when LPL and GST-LPLC are present in even larger molar excess over VLDL, they behave as competitive inhibitors of LDL receptor-mediated catabolism (data not shown) as observed previously in studies of LRP (18).

Despite the fact that lipolysis is not required for LPL-dependent VLDL catabolism, it clearly enhances catabolism. Lipolysis may expose or activate the receptor-recognition domains of apoE or apoB100 and contribute to LPL's effect (53, 54, 55, 56, 57). This is consistent with the finding that the lipid component of lipoprotein particles may modulate the reactivity of the apoprotein component (58). Also, Takahashi et al. (12) showed that the structurally unrelated Pseudomonas lipase can enhance VLDL catabolism via the VLDL receptor. Thus, GST-LPLC's lack of catalytic activity may be partly responsible for its relatively lower potency than LPL to the stimulate VLDL catabolism at 37 °C. In our study, internalization of protein-free emulsion increased only 2-fold following up-regulation of LDL receptors in FSF cells, whereas under similar conditions VLDL catabolism was increased up to 5-fold. Thus, the magnitude of LPL-induced catabolism of the emulsion is less than that obtained with native VLDL. This suggests that the effect of LPL on VLDL catabolism is only partially dependent on direct interaction of LPL with the receptor.

Our studies indicate that the affinity of LPL for LDL receptors is lower than its affinity for LRP, GP330/LRP-2, or VLDL receptors. The K_D values reported for these receptors are, respectively, 34, 6, and 1 nM (11, 12, 17). However, due to the low affinity of LPL for LDL receptors as compared with its high nonspecific binding, we could not accurately determine its affinity in solid-phase or cell surface assays. Also, we were unable to demonstrate LPL binding to LDL receptors by ligand-blotting due to high nonspecific binding of LPL to the membrane support at the concentrations of LPL required.

Our studies indicate a major role for LDL receptors as compared with LRP in LPL-dependent VLDL catabolism. By comparing FSF and MEF cells, which express both LDL receptors and LRP, to their mutant counterparts that lack either LDL receptors or LRP, we were able to demonstrate that LDL receptors contribute to the majority of LPL-dependent VLDL catabolism. Only when LDL receptors were not up-regulated could a significant contribution of LRP to LPL-dependent VLDL catabolism be demonstrated. However, this may reflect the situation in vivo where the LDL receptor pathway is usually down-regulated (59).

Although we studied VLDL, it is likely that these observations apply to chylomicron catabolism as well, because the remnants of large VLDL and chylomicrons appear to share the same catabolic pathway (60). The relative contributions of LDL receptors and LRP to remnant catabolism in vivo are not clearly defined, but evidence presented here and elsewhere implicates both receptors (61, 62, 63, 64, 65, 66, 67). Accumulation of triglyceride-rich precursors of LDL has been reported in humans with severe LDL receptor deficiency (61). Treatment of hypertriglyceridemic subjects with lovastatin to increase their LDL receptor number results in a reduction of their VLDL levels (62). Furthermore chylomicron remnant catabolism is impaired in Watanabe heritable hyperlipidemic rabbits, which lack functional LDL receptors and in cholesterol-fed rabbits with low LDL receptor activity (63, 64). Willnow et al. (65) implicated LRP in the catabolism of remnant particles when they found that over-expression of RAP, an inhibitor of LRP, inhibits chylomicron remnant clearance by transgenic mice lacking LDL receptors. However, this effect was not seen in mice having normal LDL receptor activity, suggesting a role for LDL receptors in chylomicron remnant clearance (66). Consistent with this idea, Willnow et al. recently found that chylomicron remnants accumulate in RAP knock-out mice that have deficient LRP activity (67). However, the remnants accumulate only if the mice also lack LDL receptors.

The observation that LPL can bind directly to LDL receptors as well as to other members of the LDL receptor family is consistent with other studies that provide evidence for an apolipoprotein-like role for LPL in remnant catabolism (68, 69, 70, 71). LPL is present in nanomolar concentrations in human plasma and is associated with lipoproteins (68). Nordenstrom et al. (69) found that plasma LPL activity directly correlates with plasma triglyceride levels during infusion of triglyceride emulsion in humans. Peterson et al. (70) demonstrated that plasma LPL activity dramatically increases in rats after intravenous administration of intralipid emulsions. Elevated plasma LPL caused by heparin injection or intravenous infusion of Intralipid results in increased LPL activity in the liver. Moreover, hepatectomy of hypertriglyceridemic rats causes a progressive increase in plasma LPL activity (70). Vilaro et al. (71) confirmed these findings by showing that injection of intralipid in starved rats enhances LPL activity in the liver. These data suggest that a fraction of LPL can bind to triglyceride-rich lipoproteins under normal circumstances and undergo transport to the liver where LPL may function as a recognition signal for the hepatic uptake of lipoproteins.

In addition to its potential role in hepatic clearance of triglyceride-rich particles, another interesting potential role for LPL is its participation in atherogenesis. Corey and Zilversmit (72, 73) showed a coordinate increase in the LPL activity and cholesterol content of rabbit aortas with cholesterol feeding and proposed that atherogenesis may result from abnormal postprandial lipolytic processing of triglyceride-rich particles. LPL is synthesized by macrophages and macrophage-derived foam cells present in atherosclerotic plaques and to a lesser degree by smooth muscle cells as well (74). Rutledge and Goldberg (75) recently reviewed evidence that in these tissues, LPL may play a role in the pathogenesis of atherosclerosis by promoting the retention of VLDL in the arterial wall. Furthermore Rapp et al. (76) and Chung et al. (77) have found intact VLDL and surface remnants of triglyceride-rich particles in human atherosclerotic plaques. Studies in HUVECs reported here indicate that LPL induces catabolism of VLDL via LDL receptors in endothelial cells. Other cells present in atheromas, including macrophages and smooth muscle cells also participate in LPL-induced uptake of lipoproteins (21, 78, 79, 80). The potential role of vascular LPL in atherogenesis is an important area for future research.