The production of 85kDa N-terminal fragment of apolipoprotein B in mutant HepG2 cells generated by targeted modification of apob gene occurs by ALLN-inhibitable protease cleavage during translocation

To study the mechanism of low levels of full length and truncated apoB in individuals heterozygous for apoB truncation, a non-sense mutation was introduced in one of the three alleles of apob gene of HepG2 cells by homologous recombination. Despite very low levels of apoB-82 (1–2%) in the media, a prominent N-terminal apoB protein of 85 kDa (apoB-15) was secreted that fractionated at d > 1.065 in density gradient ultracentrifugation. The mechanism of production of this short protein was studied by 35 S-methionine pulse–chase experiment. Oleate prevented presecretory degradation of apoB-100 in the cell and resulted in increased secretion of newly synthesized apoB-100 with decreases in the apoB-15, suggesting that rescue of pre-secretary intracellular degradation of apoB restricted the production and secretion of apoB-15. Further investigation on the degradation of transmembrane forms of apoB, in the presence and absence of a cysteine protease inhibitor, N-acetyl-leucyl-leucyl-norleucinal (ALLN), showed appearance of detectable levels of newly synthesized apoB-82 in the cell and the media together with increased apoB-100 secretion, and reduction in the secretion of apoB-15. Compared to ER membrane, the levels of apoB were higher in the luminal content, and presence of both oleate and ALLN had additive effect on apoB secretion. These results suggest that the presence of improper folding of apoB during translocation led to the cleavage of both apoB-100 and apoB-82 by ALLN-sensitive protease and generation of 85 kDa N-terminal fragment of apoB. Keywords Familial hypobetalipoproteinemia Apolipoprotein B Gene targeting Translocational degradation Cysteine protease ApoB-15 1 Introduction Familial hypobetalipoproteinemia (FHBL) is characterized by low levels of apolipoprotein B (apoB) and LDL-C [1,2] . Heterozygotes of FHBL are usually asymptomatic, but homozygotes may experience dietary fat malabsorption and develop fatty liver [3,4] . The hypobetalipoproteinemic individuals have mutations in the apob gene that result in premature termination of translation and the formation of truncated apoB. A number of nonsense mutations have been described in human kindreds that cause formation of truncated forms of apoB [5,6] ranging from as low as B-31 and as big as B-89. apoB-100 is a 550 kD protein expressed in and secreted by mammalian livers as VLDL particles [7] . Since apoB-100 also serves as a ligand for the LDL receptor to mediate the clearance of apoB-100-containing lipoproteins from plasma [8] through its LDL receptor binding domain [9] , LDL isolated from patients with the genetic disorder hypobetalipoproteinemia may or may not contain LDL receptor-binding site depending upon the size of the truncated apoB. Interestingly, LDL containing the amino-terminal 67% of apoB-100 [10] and B-70.5 [11] did not bind to the LDL receptor [12] . However, the LDL with truncated apoB containing the amino-terminal 75% (apoB-75) or 89% (apoB-89) bound with an increased affinity to the LDL receptor [13,14] . In vivo metabolic studies in humans have shown that the products of both the normal apoB-100 allele and the apoB truncation allele are produced at lower than expected rates in the FHBL heterozygotes [15,16] . Both in humans as well as in mice, heterozygotes for apoB-83 truncation produce undetectable levels of truncated apoB, B-83 [17,18] . Since shorter apoB, B-31, B-37, B-43.7, B-46 that lack LDL receptor binding domain also show very low levels [5,10,19–21] , it suggests that presecretory intracellular degradation of apoB may be the major pathway for diminished secretion of these truncated apoBs. To study the effects of a truncation mutation on the apoB metabolism, we targeted the apob gene of HepG2 cells by homologous recombination using the dual selection “in and out” strategy [22] . While low production of truncated apoB was demonstrated as a result of low levels of mutant apoB mRNA, the mechanism of low secretion rates of apoB-100 and the generation of an 85 kDa N-terminal fragment of apoB (B-15) remained unexplained. The present study investigated the secretion of newly synthesized full length and the truncated forms of apoB. We also examined the secretion of a shorter 85 kDa N-terminal fragment of apoB (B-15) in the mutant cell line. Our results suggest that rescue of intracellular degradation, either by providing exogenous fatty acid, or by inhibiting cysteine protease, enhanced apoB-100 secretion, and reduced the formation of apoB-15. Thus, intracellular degradation of apoB in cells expressing full length and the truncated forms of apoB appears to be responsible for reduced apoB-100 secretion, and a shorter apoB-15 N-terminal fragment of apoB is produced as a result of cleavage by cysteine protease at a specific site. 2 Materials and methods 2.1 The mutant HepG2 cell line HepG2 cell lines used in the present study were normal non-targeted HepG2 cells, denoted as wild-type, and mutant HepG2 cells in which one of the apob alleles was modified by targeted insertion of stop-codon using homologous recombination as described [22] . 2.2 Secretion of 85 kDa N-terminal fragment of apoB (B-15) in culture media WT and the mutant HepG2 cells were grown in DMEM media containing 10% fetal calf serum to 80% confluency in 75 cm 2 flasks. After 48 h growth in complete media, the spent media were transferred to sterile tubes followed by the addition of aprotinin (100 U). The media were concentrated in a Centriprep-30 (Amicon, Inc., Beverly, MA). Final volumes were made to 1 ml with EDTA saline solution containing aprotinin and fractionated by ultracentrifugation by adjusting the density to 1.065 g/ml with KBr [23] . apoB proteins were analyzed by Western blotting as described [24] . The fractions (LDL fraction, d = 1.065, and fraction having a density of >1.065) obtained by ultracentrifugation were immunoprecipitated using a rabbit antihuman apoB antibody [25] . The samples were run in a 3–12% gradient SDS–polyacrylamide gel. After electrophoresis, proteins were transferred from the gel to PVDF membranes, and the membranes were probed with 125 I-labeled anti-apoB monoclonal antibody C1.4 (an N-terminal antibody) and also with B1B6 (a C-terminal apoB antibody) as described [25] . Membranes were exposed to X-ray film to visualize and quantitate apoB protein bands. 2.3 Pulse–chase experiments To investigate the rates of de novo synthesis and secretion of apoB-100 and B-82, as well as to examine the mechanism of secretion of apoB-15, pulse–chase experiments were carried out in WT and the mutant cells. All experiments were carried out in collagen-treated dishes as described [26] . WT and mutant cells (1 × 10 6 /plate) were grown in complete media till the cells became 80% confluent and formed a monolayer. Next day the cells were starved for 75 min in methionine-free media (MEM) followed by incubation in serum-free media with [ 35 S]-methionine (35 mCi/ml) procured from ICN Biochemicals (15 mCi/ml). The cells were pulsed (20 min) and chased with 1000-fold cold methionine for various time intervals. At indicated time points, the media were removed, aprotinin (100 U) added, and concentrated in Centriprep-30 (Amicon Inc., Beverly, MA). Cells were washed with cold PBS and scraped off the plate and transferred into a 2-ml Eppendorf tube. The processing of the cells was done as described [22] . Based on TCA precipitable counts, equal amounts of counts were used for apoB immunoprecipitation using polyclonal antibody, R197 [25] . To study the effects of oleic acid on the synthesis and secretion of apoB-100 in WT and the mutant cells, oleic acid was complexed with fatty acid-free BSA as described [26] . Fatty acid-free BSA solution was adjusted to pH 7.4, and stored frozen in aliquots. Oleic acid (Sigma Chemical Co., St. Louis, MO) was treated with NaOH, dried under N 2 , and dissolved in serum-free media containing 1.5% BSA to make 0.8 mM oleic acid solution. Fresh oleic acid–BSA complex was prepared each time. Cells grown as monolayers were starved for 75 min in methionine-free media (MEM) and incubation in serum-free media with [ 35 S]methionine (35 mCi/ml) in the presence of 1.5% BSA or 1.5% BSA/0.8 mM oleate. The processing of cells and media was done as described above. For experiments with N-acetyl-leucyl-leucyl-norleucinal (ALLN), cells were grown to 85% confluence, the culture media were changed to methionine-free MEM media, and ALLN (50 μg/ml) was added and one hour later pulsed with [ 35 S]-methionine. The processing of cell and media was done as described above. 2.4 Fractionation of ER membrane and lumen In some experiments cells were fractionated into membrane and lumen following [ 35 S]-methionine pulse and chase as described [27] . Total microsomes were isolated from intact cells and fractionated by ultracentrifugation to isolate a luminal fraction and a membrane-enriched fraction. The membrane and luminal fractions were then subjected to immunoprecipitation and SDS–PAGE as described above. 2.5 Measurements of triglyceride synthesis Triglyceride synthesis was measured by incorporation of [ 3 H]-glycerol as described [26] . Wild-type and the mutant HepG2 cells were grown as described above, spent media removed, and serum-free MEM added followed by the addition of 10 μCi of [ 3 H]-glycerol. After labeling, the media was removed, the cells were washed with cold PBS to remove unincorporated label. The cellular lipids were extracted by organic extraction (hexane:isopropanol, 3:2) and dried under N 2 . The lipids were separated on a silica gel TLC plate, triglyceride band identified by using iodine vapor, and triglyceride containing area was scrapped off from the silica plate and counted in a liquid scintillation counter. 3 Results A mutant HepG2 cell was made by targeted introduction of a nonsense mutation in apob gene that produces a truncated apoB, B-82, albeit in a very low concentrations [22] . The spent media from wild-type and the mutant HepG2 cells were fractionated into LDL-sized particles and smaller particles using density gradient ultracentrifugation. As shown in Fig. 1 , mutant HepG2 cells, but not the WT HepG2 cells, produce an 85 kDa N-terminal fragment of apoB that fractionated in smaller lipoprotein particles. The synthesis and secretion of newly synthesized apoB-100, B-82, and B-15 was studied in cells pulsed with 35 S-methionine. The quantitation of intracellular labeled apoBs from 2 min to 120 min showed a decreasing trend with time ( Fig. 2 A). The levels of apoBs in the mutant cells were less when corrected for the internal standard, albumin ( Fig. 2 A). Concomitant increases of secreted apoB in the media were observed with time, but again the levels of secreted apoB were lower in the mutant HepG2 cells when compared to the wild-type HepG2 cells ( Fig. 2 B). An additional N-terminal fragment of apoB, B-15, was secreted in the media of mutant cells, but not in the WT cells. The mechanism of lower than expected synthesis and secretion of apoBs as well as production of apoB-15 was investigated. First, we carried out studies in the presence and absence of exogenous fatty acid. We reasoned that since majority of apoBs are degraded intracellularly before secretion, providing oleic acid would rescue presecretory degradation of apoB [26] , without altering the levels of smaller fragment, apoB-15. Indeed, providing exogenous oleate caused increased secretion of apoB-100 and decreased secretion of B-15 ( Fig. 3 A). To further have insights into the generation of the apoB-15, we took advantage of the fact that apoB degradation occurs during translocation into the lumen by a cysteine protease cleavage mechanism [28] . We, therefore, used a cysteine protease inhibitor, ALLN, and followed the newly synthesized apoBs in the cell and the media after labeling with 35 S-methionine. First, we designed an experiment to see the effect of secreted apoBs in the media in the presence of oleate or ALLN alone or in combination. The results in Fig. 3 B suggested that there was an additive effect of combination of oleate and ALLN, suggesting that these two agents increased apoB secretion via independent mechanism. A detailed investigation with ALLN showed that ALLN increased apoB-100 both in the WT and the mutant HepG2 cells ( Fig. 4 A and B). The smaller apoB, B-15, showed a decreasing trend. Next, we asked the question whether the generation of B-15 occurs during translocation. HepG2 cells were labeled with 35 S-methionine, and the cells were fractionated into the ER membrane and the lumen. At the same time, media were also processed at 90 min time point to measure secreted apoBs. While the levels of apoB-100 in the luminal fraction of the WT HepG2 cells were higher, the luminal apoB-100 in the mutant cells was about 60% of the WT, which resulted in the decreased secretion (∼60%) of apoB-100 in the mutant cells (data not shown). In order to rule out the role of cellular triglyceride synthesis in the generation of the apoB-15, triglyceride synthesis was carried out using [ 3 H]-glycerol. The extracted lipid fraction was subjected to triglyceride separation in a silica gel thin layer chromatography and the amount of triglyceride determined by counting the triglyceride spots. Our results showed that the triglyceride synthesis in the WT and the mutant HepG2 cells was comparable (data not shown). 4 Discussion Apolipoprotein B (apoB) is the major protein component of very low density lipoprotein (VLDL) and low density lipoproteins, and is essential for the assembly and secretion of nascent VLDL particles [7] . The number of VLDL particles secreted by the liver is a function of the proportion of apoB that escapes degradation [29] . In hypobetalipoproteinemic subjects, the levels of apoB and LDL-C are low, which could either result from increased intracellular degradation or increased clearance from plasma. Studies in mice [30] and humans [6,14] suggested that both the apoB production and clearance are important in determining the apoB and LDL-C levels in hypobetalipoproteinemic subjects. To study molecular mechanism of low apoB in hypobeta subjects, a nonsense mutation was introduced in one of the apob alleles of HepG2 cells by homologous recombination [22] . The mutant cells produced very low levels of truncated apoB, B-82, but significant amount of apoB-15 in the media. The aim of the present study was to investigate the mechanism of production of apoB-15. The low production of apoB-82 agreed with the low levels (1–2%) of B-83 in human heterozygotes [17] as well as in mouse B-81 [30] and B-83 [18] heterozygotes. To study the mechanism of production of apoB-15, we considered a possible linkage between the conformation of apoB during the translocation and assembly process, and its secretion and degradation within the hepatocytes [31–33] . Thus, proper folding of apoB may be a prerequisite for proper sorting for secretion. Since B-15 could result from cleavage at a specific site during translocation arrest across ER membrane, which can translocate and resume secretion, our approach to understand the mechanism of B-15 production was designed around this hypothesis. Increased secretion of B-100, but not B-15, in the media in the presence of oleate suggested that rescue of intracellular apoB degradation facilitated the translocation across the ER membrane during apoB assembly and this in turn caused lower secretion of B-15 ( Fig. 3 ). Thus, B-15 possibly resulted from the cleavage of apoB during translocation. Since, a cysteine protease has been implicated in the presecretory intracellular degradation of apoB, next we asked the question whether inhibiting cysteine protease would result in greater secretion of apoB-100 with no changes or lesser secretion of B-15. Indeed, we found that an inhibitor of cysteine protease, ALLN, caused increased secretion of apoB-100, but not B-15 ( Fig. 4 ). These findings again suggest that B-15 is produced during translocation arrest via cleavage by ALLN-inhibitable protease. The mechanism of increased secretion of apoB in the presence of oleate or ALLN is very different. While oleate increases secretion of apoB by providing more fatty acids for the assembly of apoB particles, ALLN increases secretion by inhibiting proteases that degrade apoB during translocation. Therefore, we expected an additive effect of combination of oleate and ALLN on the secretion of apoB, and as shown in Fig. 3 B, increased secretion of apoB-100 was noticed in the presence of oleate and ALLN compared to either one alone. To further corroborate our findings that translocation arrest in the ER membrane triggered apoB cleavage by ALLN-inhibitable protease, pulse–chase experiment was carried out and ER membrane as well as luminal fraction was isolated. While the amount of B-15 within the cells was very low, we quantitated newly synthesized apoB-100 in the ER membrane and the lumen. The luminal B-100 contents were lower in the mutant HepG2 cells as compared to WT, which resulted in the decreased secretion of apoB in the media. Since intracellular triglyceride synthesis could influence the amount of apoB secreted in the mutant cells, we measured triglyceride synthesis, and the results showed that triglyceride synthesis was comparable in the WT and the mutant cell lines. Shorter truncated proteins have been reported to be secreted by non-hepatic cells [34–36] . For instance, endothelial cells secrete 116 kD and 85 kD N-terminal fragment of apoB [34,35] . An 85 kD N-terminal fragment of apoB has also been reported to be secreted by CHO cells transfected with apoB cDNA that synthesizes apoB-53 [36] . While transfected CHO cells expressed apoB-53 intracellularly, but secreted mostly apoB-15 (85 kDa fragment) into the media. This could happen as a result of inability of these cells to translocate and secrete apoB. The arrested apoB in the ER membrane during translocation possibly gives rise to B-15 in CHO and other non-hepatic cells as a result of cleavage at a specific site. In the mutant HepG2 cells, the conformational changes as a result of premature stop codon makes apoB unable to translocate across the ER membrane. At this point, a cysteine protease cleaves ER membrane-bound apoB-100 and/or B-82 at a specific site to enable the cleaved apoB-15 to resume secretion. References [1] M.F. Linton R.V. Farese Jr. S.G. Young Familial hypobetalipoproteinemia J. Lipid Res. 34 1993 521 541 [2] G. Schonfeld Familial hypobetalipoproteinemia: a review J. Lipid Res. 44 2003 878 883 [3] M.V. Clarke A.J. Hooper H.A. Headlam J.H.Y. Wu K.D. Kroft J.R. Burnett Assessment of tocopherol metabolism and oxidative stress in familial hypobetalipoproteimnemia Clin. Chem. 52 2006 1339 1345 [4] T. Tanoli P. Yue D. Yablonskiy G. Schonfeld Fatty liver in familial hypobetalipoproteinemia: roles of the apoB defects, intra-abdominal adipose, and insulin sensitivity J. Lipid Res. 45 2004 941 947 [5] N. Srivastava D. Noto M. Averna J. Pulai R.A.K. Srivastava T.G. Cole M.A. Latour B.W. Patterson G. Schonfeld A new apolipoprotein B truncation (apoB-43.7) in familial hypobetalipoproteinemia: genetic and metabolic studies Metabolism 45 1996 1296 1304 [6] K.G. Parhofer P.H.R. Barrett C.A. Aguilar-Salinas G. Schonfeld Positive linear correlation between the length of truncated apoB and its secretion rate: in vivo studies in human apoB-89, apoB-75, apoB-54.8, and apoB-31 heterozygotes J. Lipid Res. 37 1996 844 852 [7] V.N. Schumaker M.L. Phillips J.E. Chatterton Apolipoprotein B and low density lipoprotein structure: implications for biosynthesis of triglyceride-rich lipoproteins Adv. Protein Chem. 45 1994 205 248 [8] A. Law J. Scott A cross-species comparison of the apoB domain that binds to the LDL receptor J. Lipid Res. 36 1990 1109 1120 [9] T.L. Innerarity R.W. Mahley K.H. Weisgraber T.P. Bersot R.M. Krauss G.L. Vega S.M. Grundy W. Friedl J. Davington B.J. McCarthy Familial defective apolipoprotein B-100: a mutation of apoB that causes hypercholesterolemia J. Lipid Res. 31 1990 1337 1349 [10] F.K. Welty S.T. Hubl V.R. Pierotti S.G. Young A truncated species of apoB (B-67) in a kindred with familial hypobetalipoproteinemia J. Clin. Invest. 87 1991 1748 1754 [11] Z. Chen J.E. Saffitz M.A. Latour G. Schonfeld Truncated apoB-70.5-containing lipoproteins bind to megalin but not the LDL receptor J. Clin. Invest. 103 1999 1419 1430 [12] A.V. Hospattankar S.W. Law K. Lackner H.B. Brewer Jr. Identification of low density lipoprotein receptor binding domains of human apoB-100: a proposed consensus LDL receptor binding sequence of apoB-100 Biochem. Biophys. Res. Commun. 139 1986 1078 1085 [13] E.S. Krul K.G. Parhofer P.H.R. Barrett R.D. Wagner G. Schonfeld ApoB-75, a truncation of apoB associated with familial hypobetalipoproteinemia: genetic and kinetic studies J. Lipid Res. 33 1992 037 1050 [14] K.G. Parhofer P.H. Barrett D.M. Bier G. Schonfeld Lipoproteins containing the truncated apolipoprotein, apoB-89, are cleared from human plasma more rapidly than apoB-100-containing lipoproteins in vivo J. Clin. Invest. 89 1992 1931 1937 [15] C.A. Aguilar-Salina P.H.R. Barrett K.G. Parhofer S.G. Young D. Tessereau J. Bateman C. Quinn G. Schonfeld Apoprotein B-100 production is decreased in subjects heterozygous for truncations in apoB Arterioscler. Thromb. Vasc. Biol. 15 1995 71 80 [16] F.K. Welty A.H. Lichtenstein P.H.R. Barret G.G. Dolnikowski J.M. Ordovas E.J. Schaefer Decreased production and increased catabolism of apoB-100 in apoB-67/B-100 heterozygotes Arterioscler. Thromb. Vasc. Biol. 17 1997 881 888 [17] R.V. Farese Jr. A. Garg V.R. Pierotti G.L. Vega S.G. Young A truncated species of apoB, B-83, associated with hypobetalipoproteinemia J. Lipid Res. 33 1992 569 577 [18] E. Kim C.M. Cgam M.M. Veniant P. Ambroziak S.G. Young Dual mechanism of the low plasma levels of truncated apoB proteins in familial hypobetalipoproteinemia: analysis of a new mouse model with a nonsense mutation in the apolipoprotein B gene J. Clin. Invest. 101 1998 1468 1477 [19] S.G. Young S.T. Hubl R.S. Smith S.M. Snyder J.F. Terdiman Familial hypobetalipoproteinemia caused by a mutation in the apolipoprotein B gene that results in a truncated species of apoB (B-31). A unique mutation that helps to define the portion of the apolipoprotein B molecule required for the formation of buoyant, triglyceride-rich lipoproteins J. Clin. Invest. 85 1990 933 942 [20] S.G. Young S.J. Bertics L.K. Curtiss B.W. Dubois J.L. Witztum Genetic analysis of a kindred with familial hypobetalipoproteinemia. Evidence for two separate gene defects: one associated with an abnormal apolipoprotein B species, apoB-37; and a second associated with low plasma concentrations of apoB-100 J. Clin. Invest. 79 1987 1842 1851 [21] S.G. Young S.T. Hubl D.A. Chappell R.S. Smith F. Claiborne S.M. Snyder J.F. Terdiman Familial hypobetalipoproteinemia associated with a mutant species of apoB (B-46) N. Engl. J. Med. 320 1989 604 1610 [22] R.A.K. Srivastava N. Srivastava M. Averna A.B. Cefalu G. Schonfeld Molecular bases of low production rates of apoB-100 and truncated apoB-82 in a mutant HepG2 cell line generated by targeted modification of the apolipoprotein B gene J. Lipid Res. 40 1999 901 912 [23] R.A.K. Srivastava S. Jiao J. Tang B.A. Pfleger R.T. Kitchens G. Schonfeld In vivo regulation of LDL receptor and apoB gene expressions in inbred strains of mice by dietary fatty acids and cholesterol Biochim. Biophys. Acta. 1086 1991 29 43 [24] J.J. Tang R.A. Srivastava E.S. Krul D. Baumann B.A. Pfleger R.T. Kitchens G. Schonfeld In vivo regulation of apolipoprotein A-I gene expression by estradiol and testosterone occurs by different mechanisms in inbred strains of mice J. Lipid Res. 32 1991 1571 1585 [25] E.S. Krul Y. Kleiman M. Kinoshita B. Pfleger K. Oida A. Law J. Scott R. Pease G. Schonfeld Regional specificities of monoclonal anti-human apoB antibodies J. Lipid Res. 29 1988 937 948 [26] J.L. Dixon S. Furukawa H.N. Ginsberg Oleic acid stimulates secretion of apoB-containing lipoproteins from HepG2 cells by inhibiting early intracellular degradation of apoB J. Biol. Chem. 266 1991 5080 5086 [27] J. Bore´n M. Wettesten A. Sjöberg T. Thorlin G. Bondjers O. Wiklund S.-O. Olofsson The assembly and secretion of ApoB-100-containing lipoproteins in HepG2 Cells. Evidence for different sites for protein synthesis and lipoprotein assembly J. Biol. Chem. 265 1990 10556 10564 [28] J.A. Bonnardel R.A. Davis In HepG2 cells, translocation, not degradation, determines the fate of de novo synthesized apoB J. Biol. Chem. 270 1995 28892 28896 [29] K. Adeli Regulated intracellular degradation of apoB in semipermeable HepG2 cells J. Biol. Chem. 269 1994 9166 9175 [30] R.A.K. Srivastava L. Toth N. Srivastava M.E. Hinsdale N. Maeda A.B. Cefalu M. Averna G. Schonfeld Regulation of the apoB in heterozygous hypobetalipoproteinemic knock-out mice expressing truncated apoB, B-81. Low production and enhanced clearance of apoB cause low levels of apoB Mol. Cell. Biochem. 202 1999 37 46 [31] S.L. Chuck V.R. Ligappa Pause transfer: a topogenic sequence in apoB mediates stopping and restarting of translocation Cell 68 1992 9 21 [32] R.A. Davis R.N. Thrift C.C. Wu K.E. Howell Apolipoprotein B is both integrated into and translocated across the endoplasmic reticulum membrane. Evidence for two functionally distinct pools J. Biol. Chem. 265 1990 10005 100011 [33] R.N. Thrift J. Drisko S. Dueland J.D. Trawick R.A. Davis Translocation of apoB across the endoplasmic reticulum is blocked in a non-hepatic cell line Proc. Natl. Acad. Sc. USA 89 1992 9161 9165 [34] J. Wolle E. Ferguson L.J. Devall R.S. Newton U. Saxena Identification of a novel 85 kD lipoprotein lipase binding protein on human aortic endothelial cell surface Biochem. Biophys. Res. Commun. 216 1995 906 912 [35] P. Sivaram T. Vanni-Reyes I. Goldberg Endothelial cells synthesize and process apoB J. Biol. Chem. 271 1996 15261 15266 [36] E.Z. Du J. Kurth S. Wang P. Humiston R.A. Davis Proteolysis-coupled secretion of the N-terminus of apoB. Characterization of a transient, translocation arrested intermediate J. Biol. Chem. 269 1994 24169 24176