Influence of Triacylglycerol Biosynthesis Rate on the Assembly of ApoB-100-Containing Lipoproteins in Hep G2 Cells

Transcription

1 1743 Influence of Triacylglycerol Biosynthesis Rate on the Assembly of ApoB-100-Containing Lipoproteins in Hep G2 Cells Jan Boren, Sabina Rustaeus, Margit Wettesten, Maria Andersson, Anna Wiklundt, Sven-Olof Olofsson Apolipoprotein B-100 (apob-100) appears in three forms in the endoplasmic reticulum of Hep G2 cells: (1) tightly bound to the membrane, ie, not extractable by sodium carbonate. This form is glycosylated but protease sensitive when present in intact microsomes, suggesting that it is only partially translocated to the microsomal lumen; (2) extractable by sodium carbonate and present on low-density lipoprotein-verylow-density lipoprotein (LDL-VLDL)-like particles. This form is glycosylated and secreted into the medium; and (3) extractable by sodium carbonate but having a higher density than the LDL-VLDL-like particles. This form, referred to as Fraction I, is glycosylated and protected against proteases when present in intact microsomal vesicles, indicating that it is completely translocated to the luminal side of the microsomal membrane. Fraction I is not secreted into the medium, but it disappears with time from the cell, suggesting that it is degraded. Oleic acid induced a 2.7-fold increase in the rate of the biosynthesis of triacylglycerol but not of phosphatidylcholine in Hep G2 cells. Incubation of the cells with oleic acid had no significant effect on the rate of initiation of the apob-100-containing lipoproteins, nor did it influence the amount of apob-100 that was associated with the membrane or the turnover of apob-100 in the membrane. Instead, it increased the proportion of the nascent apob polypeptides on initiated lipoproteins that was converted into full-length apob-100 on LDL-VLDL-like particles, giving rise to an increased amount of these particles in the lumen of the secretory pathway. Pulse-chase experiments showed that incubation with oleic acid gave rise to an increased formation of LDL-VLDL-like particles on behalf of the formation of Fraction I. This effect of oleic acid could partially explain the protective effect of the fatty acid on apob-100, preventing it from undergoing posttranslational degradation. (Arterioscler Thromb. 1993;13: ) KEY WORDS apob-100 Hep G2 cells triacylglycerol biosynthesis A polipoprotein (apo) B-100 occurs mainly on the / \ liver-derived lipoproteins, ie, very-low-density li- JL A. poproteins (VLDLs), intermediate-density lipoproteins, and low-density lipoproteins (LDLs). ApoB-100 is essential for the assembly of these lipoproteins. The interaction between apob-100 and the lipids can occur cotranslationally and simultaneously with the translocation of the protein to the lumen of the secretory pathway. 1 ApoB-100 could also be bound to the endoplasmic reticulum (ER) membrane to be consigned to posttranslational degradation. 12 The incorporation of apob into lipoprotein particles is first detected when the nascent polypeptide has reached a size of 80 to 100 kda. These nascent polypeptides form particles with a diameter and density of a high-density lipoprotein (HDL) particle. As the nascent polypeptide increases in length, the size and lipid load of the particle that is being formed increase. Thus, the size of the nascent polypep- Received February 24, 1993; revision accepted August 10, From the Department of Medical Biochemistry and the Wallenberg Laboratory, University of Goteborg, Goteborg, Sweden. fdr Wiklund died March 12, This article is dedicated to her memory. Correspondence to Sven-Olof Olofsson, Department of Medical Biochemistry, University of Goteborg, Medicinaregatan 9, S Goteborg, Sweden. tide determines the size of the particle that it has assembled. 13 During the elongation of the nascent polypeptides, these initiated lipoproteins, ie, the immature lipoprotein particles containing nascent polypeptides, are converted to triacylglycerol-rich LDL-VLDL-like particles with full-length apob-100 that are secreted from the cell However, we have also reported 14 that a more dense particle (Fraction I in Reference 4) with fulllength apob-100 may be formed in the secretory pathway. Fraction I does not seem to be secreted from the cell to any significant degree. The question of the nature of this particle is addressed in this article. Incubation of Hep G2 cells with oleic acid increases the secretion of apob without changing the amount of apob-100 mrna in the cell. 46 Instead, it seems as if the oleic acid stabilizes apob-100 and prevents it from undergoing posttranslational degradation. 7 Thus, incubating cells with oleic acid induces an increase in the proportion of the amount of intracellular apob-100 that is used for the assembly and subsequent secretion of LDL-VLDL particles. 4 There are at least two factors of importance for the assembly process that are potential targets for the influence of oleic acid: the biosynthesis of phosphatidylcholine and the biosynthesis of triacylglycerol. Thus, the secretion

2 1744 Arteriosclerosis and Thrombosis Vol 13, No 12 December 1993 of apob-100 is highly dependent on an ongoing biosynthesis of phosphatidylcholine, 8-9 and the assembly of the apob-containing lipoproteins may occur in regions of the ER with high diacylglycerol: acyltransferase activity, 2 ie, regions with a high capacity for the formation of triacylglycerol. Taken together, the observations discussed above suggest that oleic acid could be used to manipulate important steps in the assembly process and thus could be used as a tool in the elucidation of this process. In this article we have characterized one of the mechanisms by which oleic acid could influence the secretion of apob-100 from Hep G2 cells. Methods Materials Eagle's minimal essential medium was from Flow Laboratories, Irwine, UK, and Eagle's minimal essential medium without methionine was from GIBCO. [ 14 C]methylated protein standards ("rainbow standards"), [ 35 S]methionine, [ 3 H]glycerol, and Amplify were purchased from Amersham International, Arnersham, UK. [ 3 H]mannose was from Du Pont-New England Nuclear. Ready Safe was from Beckman, Fullerton, Calif. NADPH and cytochrome c were from Boehringer Mannheim. Triton X-100, oleic acid (cellculture grade), fatty acid-free bovine serum albumin, UDP-galactose, ATP, ovalbumin (fraction V), trypsin (type XIII), phenylmethylsulfonyl fluoride (PMSF), L-1- chloro-3-[4-tosylamido]-7-amino-2-heptanone-hcl (TLCK-HC1), tetracaine, and soybean trypsin inhibitor (Bowman-Birk) were from Sigma. Immunoprecipitin was purchased from Bethesda Research Laboratories. Trasylol (aprotinin) was from Bayer Leverkusen, FRG. Concanavalin A (ConA), Sepharose, and PD-10 columns were from Pharmacia LKB Biotechnology Inc. All chemicals used for sodium dodecyl sulfate (SDS)- polyacrylamide gel electrophoresis were from Bio Rad. All chemicals were of analytical grade. Cell Culture, Metabolic Labeling, and Subcellular Fractionation The Hep G2 cells were grown in Eagle's minimal essential medium as described. 2-4 Labeling with [ 35 S]methionine and chasing with a surplus of cold methionine in the absence or presence of cycloheximide was performed as described. 1 For subcellular fractionation we used the method described by Boren et al. 2 In the cases in which the subcellular fractions were subjected to proteolysis, no protease inhibitors were used in the sucrose gradient. Control experiments indicated that no detectable degradation of apob-100 occurred during centrifugation. Thus, 95 ±4% of the applied apob-100 was recovered from the gradient that did not contain protease inhibitors. To determine the orientation of the ER-derived microsomes, we took advantage of the fact that vesicles with the "wrong" orientation expose mannose-rich carbohydrate structures that bind to ConA. The ERderived microsomes were first desalted by chromatography on a PD-10 column that was equilibrated with Dulbecco's phosphate-buffered saline, ph 7.4. The desalted microsomes were incubated for 1 hour at room temperature with ConA-Sepharose with or without 100 mg a-methyl-d-mannoside. After the incubation the supernatant was collected, and the ConA-Sepharose pellet was washed eight times with 1 ml Dulbecco's phosphate-buffered saline, ph 7.4, which was added to the first supernatant. ApoB-100 and apantitrypsin were recovered by immunoprecipitation and electrophoresis in 3% to 15% polyacrylamide gradient gels containing SDS, and the radioactivity was counted. 10 The luminal content of the isolated vesicles was separated from the membrane by the sodium carbonate method 11 with the modifications described. 12 Protease Treatment of the ER-Derived Microsomes Protease treatment of isolated ER-derived vesicles was performed as described by Davis et al. 13 ER-derived microsomes, 0.2 mg (measured as proteins), were incubated with 240 /xg trypsin for 30 minutes at room temperature. After the incubation, 2.4 mg soybean inhibitor, 100,u.mol/L PMSF, and 100 /xmol/l TLCK- HC1 were added, and the vesicles were extracted with sodium carbonate to separate the luminal content from the membranes. In control experiments the isolated ER-derived vesicles were stabilized by the addition of 3 mmol/l tetracaine-hcl. 14 Estimation of the Rate of Initiation of ApoB-100-Containing Lipoproteins To estimate the initiation rate of lipoproteins, we determined the amount of radioactivity present in nascent apob polypeptides on immature lipoproteins after a 5-minute pulse with [ 35 S]methionine by using the method described by Boren et al. 1 The pulse-labeled cells were incubated with cycloheximide and puromycin and were chased for 2 hours in the presence of cycloheximide and puromycin. The concentrations of puromycin and cycloheximide were those used earlier. 1 The chase medium was collected and analyzed by sucrose gradient ultracentrifugation (see below). The gradients were unloaded from the bottom, and apob was recovered from each fraction by immunoprecipitation and electrophoresis in 3% to 15% polyacrylamide gradient gels (with SDS). The radioactive proteins were visualized by autoradiography, the nascent apob-100 polypeptides were cut from the gel, and the radioactivity was counted. Immunoprecipitation and Sucrose Gradient Ultracentrifugation of the ApoB-Containing Lipoproteins Immunoprecipitation was performed as described elsewhere. 10 ApoB-containing lipoproteins present in subcellular fractions or secreted into the medium were analyzed by gradient ultracentrifugation. The sucrose gradient used was formed by layering the following from the bottom of the tube: 1.5 ml 47% sucrose, 3 ml 25% sucrose, 2 ml 20% sucrose, 3.2 ml of the sample in 12.5% sucrose, 1.9 ml 5% sucrose, and 0.9 ml 0% sucrose. All solutions contained 3 mmol/l imidazole, ph 7.4. The gradients were ultracentrifuged at rpm in a Beckman SW-40 rotor for 65 hours at 4 C and unloaded into 12 fractions. ApoB-100 was recovered from each fraction by immunoprecipitation and electrophoresis in 3% to 15% polyacrylamide gradient gels in

3 Boren et al Triacylglycerol and the Assembly of ApoB-100-Containing Lipoproteins 1745 the presence of SDS, and the radioactivity was determined. To determine the radioactivity in apob polypeptides, the protein was isolated by immunoprecipitation and SDS-polyacrylamide gel electrophoresis. The band corresponding to the protein was cut from the gel and digested, 10 and the radioactivity was counted. Protein Sequencing To determine whether the signal sequence of the membrane-bound apob-100 was cleaved, the protein was labeled with 35 S-methionine, and the membranebound apob-100 was recovered by immunoprecipitation. After electrophoresis in 3% to 15% polyacrylamide gradient gels, the protein was blotted on polyvinylidene difluoride filters. 15 The band corresponding to apob-100 was identified by autoradiography, cut out, and used for amino-acid sequencing. The sequence analyses were performed on an Applied Biosystems gas-phase sequencer, and the 35 S activity was recovered after each cycle was determined. B g / m I Quantification of Lipids and DNA Cellular lipids were extracted as described by Bligh and Dyer. 16 To quantify the lipid pools, we used a combination of thin-layer chromatography and a flameionization detection on an Iatroscan. 4 To estimate the rate of the biosynthesis of triacylglycerol and phosphatidylcholine, the cells were incubated with [ 3 H]glycerol for periods from 0 to 120 minutes. The cellular lipids were extracted and separated by thinlayer chromatography. 1 The fractions corresponding to phosphatidylcholine and triacylglycerol were recovered, and the radioactivity was determined. DNA was determined by the method described by Burton. 17 Statistical Analysis All values are presented as mean±sd (n=5) unless otherwise stated. Statistical significance was tested by paired t test. Results Sodium Carbonate Extract and Membrane-Bound ApoB-100 From the Microsomal Fraction of the Hep G2 Cells Gradient ultracentrifugation of the sodium carbonate extract from the total microsomal fraction (as well as from the ER) showed that apob-100 occurred on particles of the same density as the apob-100-containing particles that were secreted into the medium (Fig 1). We refer to these particles as LDL-VLDL-like particles. In addition, the sodium carbonate extract contained apob-100 that banded within the gradient in fractions of higher density (corresponding to the region of the gradient where the HDL particles would band). These fractions are referred to as Fraction I. 4 No or only trace amounts of Fraction I could be recovered from the medium. However, Fraction I disappeared from the microsomal fraction of the cells (Fig 2), indicating that it was degraded in the cell. Since the LDL-VLDL-like particles are secreted into the medium, it is most likely that they are present in the lumen of the microsomes Fraction FIG 1. Line graphs showing gradient ultracentrifugation of apolipoprotein (apo) B-100-containing lipoproteins recovered from the microsomal content of Hep G2 cells (A) and from the medium (B). Confluent Hep G2 cells cultured in 56-cm 2 culture dishes in the presence of oleic acid were labeled with 250 /uci [ 35 S]methionine medium for 2.5 hours. The total microsomal fraction was recovered and extracted with sodium carbonate, and both this extract and the culture medium were subjected to sucrose gradient ultracentrifugation for 65 hours at 4 C and rpm. The gradients were unloaded into 12 fractions, apob-100 was recovered by immunoprecipitation and electrophoresis in 3% to 15% polyacrylamide gradient gels containing sodium dodecyl sulfate, and the radioactivity was counted. DPM indicates disintegrations per minute. To determine if the apob-100 of Fraction I was present on the luminal side of the ER membrane, we analyzed the sodium carbonate extract after trypsin treatment of the isolated ER vesicles. In these experiments we first took advantage of the observation that Fraction I was the quantitatively dominating form in which apob-100 occurred in the sodium carbonate extract of microsomes from Hep G2 cells that were cultured in the absence of oleic acid (see below). The cells, cultured in 56-cm 2 Petri dishes, were labeled for 2 hours with 250 /ici [ 35 S]methionine, and the recovery of sodium carbonate-extractable apob-100 was deter-

4 1746 Arteriosclerosis and Thrombosis Vol 13, No 12 December Minutes FIG 2. Line graph showing turnover of Fraction I in the total microsomal fraction of the Hep G2 cells. Confluent Hep G2 cells cultured in 56-cm 2 culture dishes were pulse labeled with 400 /uci [ 35 S]methionine for 5 minutes and chased in the presence of cold methionine for periods between 0 and 120 minutes. After each chase period, the total microsomal fraction was recovered and treated with sodium carbonate. The sodium carbonate extract was subjected to sucrose gradient ultracentrifugation, apolipoprotein B-100 was recovered from the fractions corresponding to Fraction I by immunoprecipitation and electrophoresis in 3% to 15% polyacrylamide gradient gels in the presence of sodium dodecyl sulfate, and the radioactivity was determined. Per cent indicates percent of the maximal amount of Fraction I in the total microsomal fraction. mined after trypsin treatment of the ER fraction. A recovery of 104 ±26% indicated that Fraction I, when present in the microsomes, was protected against proteolysis. This was further confirmed in experiments in which we followed the effect of the proteolysis on Fraction I isolated from the sodium carbonate extract by gradient ultracentrifugation. The results showed that 100% (mean of three experiments) of the apob-100 in Fraction I could be recovered after the protease treatment of the ER-derived vesicles. A series of control experiments was performed. The "correct" orientation of the ER-derived vesicles was controlled by adsorption to ConA-Sepharose. The ER fraction from labeled (250 /id [ 35 S]methionine for 2 hours) Hep G2 cells cultured in 56-cm 2 Petri dishes was incubated with ConA-Sepharose. The gel pellet (with the bound vesicles) was removed by low-speed centrifugation. ApoB-100 and a,-antitrypsin were recovered from the supernatant by immunoprecipitation and SDS-polyacrylamide gel electrophoresis, and the radioactivity was determined. The recovery rates of apob-100 and a r antitrypsin were 95±26% and 93±25%, respectively. In a second experiment, we compared the amount of radioactivity present in apob-100 and a^-antitrypsin in the supernatant with that obtained when the incubation with ConA-Sepharose was performed in the presence of 100 mg a-methyl-d-mannoside. The results showed a recovery of 96±28% of apob-100 and 94±26% of a,-antitrypsin. These results indicated that almost all of the apob- 100 and a,-antitrypsin radioactivity was present in microsomes that had the "right" side out, since the microsomes did not bind to ConA-Sepharose. A correct orientation of the vesicles is also supported by the following observations. Eighty-two percent (mean of three experiments) of the total cellular a,- antitrypsin was recovered with the microsomes, and 96±6% of the protein was protease resistant when isolated with the microsomes, ie, the protein was quantitatively recovered on the luminal side of the membrane. All a,-antitrypsin, however, was digested after detergent treatment of the vesicles. These circumstances allowed us to use the recovery of the a r antitrypsin to estimate the recovery of microsomes derived from the total secretory pathway in the supernatant after the ConA treatment. Thus, we estimated that more than 80% of the total microsomes recovered from the secretory pathway had the right orientation. It should be pointed out that the a,-antitrypsin that is present on the outside of the vesicles or any accidentally released from the microsomes during the incubation will bind to the ConA-Sepharose and thus will not appear in the supernatant. To make sure that the enzymes had the capacity to digest all apob-100, we assessed the recovery of apob- 100 after proteolysis of vesicles that were treated with detergents. No apob-100 could be detected after trypsination of detergent-treated microsomes (not shown). To make sure that the vesicles were not leaking, we assessed the recovery of radiolabeled transferrin, a r antitrypsin, and a 2 -macroglobulin after trypsin treatment of the microsomes. The recovery rates of these three secretory proteins were 105±4.5%, 96±6%, and 97±6%, respectively. We also investigated the glycosylation of the sodium carbonate-extractable apob-100. The cells were labeled with [ 3 H]mannose for 2 hours, the sodium carbonate extract was recovered from the total microsomal fraction, and the extracted apob-100 was analyzed by gradient ultracentrifugation. The results (Fig 3) indicated that the apob-100 present in Fraction I (as well as on the LDL-VLDL-like particles) was glycosylated. Taken together, these results indicated that Fraction I was completely translocated to the luminal side of the microsomal membrane. In agreement with the results of other investigators, 1819 we found that the major amount (81 ±3%) of the tightly membrane-associated apob-100 was lost after protease treatment of the ER-derived vesicles. Although the complete recovery of the luminal apob-100 and other secretory proteins after protease treatment of isolated ER-derived vesicles strongly argued against leaking of the membranes, we performed a second type of control experiment in which the membranes of the isolated vesicles were stabilized with tetracaine-hcl. 14 The results showed that 71 ±11% of the membranebound apob-100 was lost during protease treatment of the tetracaine-hcl-stabilized ER-derived vesicles. Thus, even in the presence of tetracaine-hcl, the major

5 Boren et al Triacylglycerol and the Assembly of ApoB-100-Containing Lipoproteins Fractions FIG 3. Autoradiograph showing glycosylation of apolipoprotein (apo) B-100 present in the sodium carbonate extract of microsomes from Hep G2 cells. Confluent Hep G2 cells in 56-cm2 culture dishes were labeled with 2 mci [3H]mannose for 2 hours. The total microsomal fraction was recovered and extracted with sodium carbonate, and the extract was subjected to sucrose gradient ultracentrifugation for 65 hours at 4 C and rpm. The gradients were unloaded into 10 fractions, and apob-100 was recovered by immunoprecipitation and analyzed by electrophoresis in 3% to 15% polyacrylamide gradient gels containing sodium dodecyl sulfate followed by autoradiography. Arrows indicate the position of the 200and 92.5-kDa standards. amount of the membrane-bound apob-100 was susceptible to proteolysis. Experiments with [3H]mannose labeling indicated that the tightly membrane-bound apob-100 was glycosylated (Fig 4). To test the possibility that this glycosylated apob-100 belonged to a subfraction that was not protease sensitive, the following experiment was performed. Cells cultured in 56-cm2 Petri dishes were labeled with 2 mci [3H]mannose for 2 hours, and the ER-derived vesicles were recovered and subjected to proteolysis. The results showed that 79% (mean of two experiments) of the mannose-labeled membrane-bound apob-100 was lost during protease treatment of the microsomes (compare Fig 4). This indicated that the major amount of the glycosylated apob-100 molecules belonged to the protease-sensitive pool of apob-100 molecules in the ER membrane. These results suggested that a portion of the membrane-bound apob-100 was translocated to the lumen of the secretory pathway. This conclusion is also supported by the results from microsequencing (Fig 5). Thus, the first methionine occurred in position four, which is in agreement with a processed signal sequence.2021 Effect of Oleic Acid on Cellular Lipids Culturing the Hep G2 cells in the presence of oleic acid gave rise to a 2.7-fold increase in the rate of biosynthesis of triacylglycerol in the cells, as evaluated by the incorporation of [3H]glycerol into triacylglycerol (Fig 6A), and a substantial increase in the pool of triacylglycerol in the cells (Table). B FIG 4. Autoradiograph showing glycosylation of apolipoprotein (apo) B-100 bound to the endoplasmic reticulum (ER) membrane. Confluent Hep G2 cells in 56-cm2 culture dishes were labeled with 2 mci [3H]mannose for 2 hours. The total microsomal fraction was recovered and subjected to ultracentrifugation in a sucrose gradient. The gradient was assayed for marker enzymes for the ER and Golgi apparatus by using NADPH cytochrome c reductase and galactosyltransferase, respectively. Fractions derived from the ER were combined, and the ER membrane was recovered after disruption of the vesicles by sodium carbonate, without (A) or with (B) a preceding treatment of the vesicles with proteases. ApoB-100 was immunoprecipitated from the membranes and analyzed by electrophoresis in 3% to 15% polyacrylamide gradient gels with sodium dodecyl sulfate followed by autoradiography. Arrows indicate the position of the 200- and 92.5-kDa standards. In agreement with earlier reports, 622 we failed to demonstrate any significant change in the incorporation of [3H]glycerol into phosphatidylcholine (Fig 6B), nor could we detect any increase in the amount of lipid phosphorus in the cell; rather, a significant decrease was noted (Table). Oleic acid did not induce any significant increase in the cellular content of cholesteryl esters in the cells (Table). Effect of Oleic Acid on the Amount of ApoB-100 Present in the Cells To estimate the effect of oleic acid on the amount of apob-100 present in the cells, Hep G2 cells, cultured in the presence or absence of oleic acid in 9.6-cm2 culture dishes, were labeled with 50 /xci [35S]methionine for 2.5 hours.4 The cells were harvested by a rubber policeman and lysed. ApoB-100 was then recovered by immunoprecipitation, electrophoresis was performed in 3% to 15% polyacrylamide gradient gels in the presence of SDS, and the radioactivity was counted. The ratio of the amount of the apob-100 radioactivity recovered from cells cultured in the presence to those cells cultured in the absence of oleic acid was 1.13±0.27 (n=10). Thus, the results suggested that the incubation with oleic acid did not influence the total amount of full-length apob100 in the cells.

6 1748 Arteriosclerosis and Thrombosis Vol 13, No 12 December 1993 Q. Q from the system (cell + medium) was significantly higher in the cells cultured in the absence of oleic acid than in those cultured in the presence of the fatty acid. These results support the possibility that oleic acid prevents the posttranslational degradation of apob-100. During these experiments we also observed that 23±6% (in the absence of oleic acid) and 25±7% (in the presence of oleic acid) of the radiolabeled apob-100 remained in the cells after the 2-hour chase. Even after a 5-hour chase there were still significant amounts of radiolabeled apob-100 in the cells. Thus, 14±6% of the initially radiolabeled apob-100 remained in cells cultured in the absence of oleic acid, whereas 8.5±4% remained in cells that had been grown in the presence of oleic acid Cycle Glu Glu Glu Met FIG 5. Bar graph showing radiosequencing of apolipoprotein (apo) B-100 isolated from the membrane of the endoplasmic reticulum (ER). Hep G2 cells cultured in 56-cm 2 culture dishes were labeled for 15 minutes with [ 35 S]methionine and chased for 15 minutes. The ER fraction was recovered and disrupted by sodium carbonate. ApoB-100 present in the membrane pellet was recovered by immunoprecipitation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis and was blotted on a polyvinylidene difluoride membrane. The band corresponding to apob-100 was cut from the filter and used for sequencing, which was performed on an Applied Biosystems gas-phase sequencer. The 35 S activity recovered after each cycle was then determined. DPM indicates disintegrations per minute. The proportion of the intracellular amount of apob- 100 that was secreted into the medium increased from 22±6% in the absence of oleic acid to 71±12% in the presence of oleic acid. In this experiment, cells cultured in 9.6-cm 2 Petri dishes were labeled for 2.5 hours with 80 /ici [ 35 S]methionine. The cells were then chased in the presence of cold methionine for 3 hours, and the amount of the labeled intracellular apob-100 that was recovered in the medium was determined. A chase period of 3 hours was chosen, since the secretion of apob-100 has by that time reached a plateau level. 4 Effect of Oleic Acid on the Posttranslational Degradation of ApoB-100 The observation discussed above suggested that oleic acid might influence the intracellular posttranslational degradation of apob-100. This possibility was examined in the next experiments, in which we followed the posttranslational degradation in cells cultured in the absence or presence of oleic acid. The cells were labeled with [ 35 S]methionine for 2.5 hours and chased for periods from 0 to 120 minutes. The recovery of apob-100 was measured after each chase period. The results (Fig 7) confirmed our earlier observations 12-4 that apob-100 is subjected to posttranslational degradation. They also indicated that the rate of disappearance of apob-100 Effect of Oleic Acid on the Initiation and Maturation of the ApoB-100-Containing Lipoproteins To estimate the rate of initiation of lipoproteins, we determined the radioactivity in nascent apob polypeptides (ie, partially completed apob polypeptides still attached to the ribosome) that had begun to assemble a lipoprotein. 1 Such nascent polypeptides appear in a density gradient between the density of HDL and that of LDL (Fig 8). To label the nascent polypeptides, the cells were incubated for 5 minutes with [ 35 S]methionine. During this incubation, a certain amount of the initiated apob polypeptides was completed, left the ribosome, and thus appeared as full-length apob-100 on LDL- VLDL-like particles in the gradient (Fig 8). This fulllength apob-100 was not included in the calculation of the radioactivity of the nascent apob-100 polypeptides. There was no significant difference in the total radioactivity of these nascent polypeptides when isolated from cells cultured in the absence or presence of oleic acid (123±6xlO 3 versus 130±5xl0 3 disintegrations per minute). This indicated that the same amount of nascent apob polypeptides was cotranslationally interacting with lipids under the two conditions (Fig 8). Analyses of the distribution of radioactive nascent polypeptides over the density gradient suggested that more nascent polypeptides were involved in the formation of the less dense particles in the cells cultured in the presence of oleic acid (Fig 9). This suggested that more nascent apob-100 polypeptides were being integrated into mature LDL-VLDL-like lipoprotein particles. This observation was supported by results from experiments in which the conversion of the initiated lipoproteins into LDL-VLDL-like particles was estimated. The results (Fig 8) showed that this conversion increased approximately threefold when the cells were cultured in the presence of oleic acid (22±1% versus 68±3%). The observation of an increased formation of LDL- VLDL-like particles was supported by the finding that an increased amount of apob-100 radioactivity could be recovered in the d= g/ml supernatant, ie, the apob-100 present on LDL-VLDL-like particles, from the luminal content of the total microsomal fraction as well as from the ER when the cells were cultured in the presence of oleic acid. Thus, the ratios between the apob-100 radioactivity recovered from the g/ml supernatant of cells cultured in the presence and those cells cultured in the absence of oleic acid were 2.54±0.41 (the luminal content of the total microsomal

7 Boren et al Triacylglycerol and the Assembly of ApoB-lOO-Containing Lipoproteins 1749 B X E s a. a Minutes Minutes FIG 6. Line graphs showing the effect of oleic acid on the incorporation of [ 3 H]glycerol into triacylglycerol (A) and phosphatidylcholine (B). Confluent Hep G2 cells cultured in the absence ( ) or presence ( ) of oleic acid in 9.6-cm 2 culture dishes were incubated with 200 /xci [ 3 H]glycerol for 0, 30, and 60 minutes. After each period of incubation, the cells were harvested, and the lipid was extracted. Unlabeled phosphatidylcholine and triacylglycerol were added, and the lipid extract was fractionated by thin-layer chromatography (see "Methods"). The lipid fractions were visualized by iodine vapor and scraped from the plates, and the radioactivity was measured. Data are mean±sd (n=5). DPM indicates disintegrations per minute. fraction) and 1.72±0.42 (the lumen of the ER fraction). In these experiments confluent Hep G2 cells cultured in 56-cm 2 culture dishes in the presence or absence of oleic acid were labeled with 250 /ici (for the analysis of the lipoproteins in the total microsomal fraction) or 400 /xci (for the analysis of the lipoproteins in the ER fraction) [ 35 S]methionine for 2.5 hours. The luminal contents of the total microsomal fraction and the ER fraction were recovered after disruption of the vesicles with sodium carbonate. The carbonate extracts were ultracentrifuged at d=1.063 g/ml for 48 hours in a Beckman Ti 50 rotor at 4 C. ApoB-100 was recovered from the supernatant by immunoprecipitation and electrophoresis 2 in 3% to 15% polyacrylamide gradient gels in the presence of SDS, and the radioactivity was then measured. Effect of Oleic Acid on Intracellular Pools of Triacylglycerol, Lipid Phosphorus, Cholesterol Esters, and Unesterified Cholesterol Cellular Lipid Triacylglycerol Lipid phosphorous Cholesterol ester Unesterified cholesterol -OA 18.9± ± ± ±4.4 +OA 97.9± ± Lipids were extracted from confluent Hep G2 cells, cultured in 9.6-cm 2 culture dishes in the absence (-OA) or presence (+OA) of oleic acid, and analyzed by a combination of thin-layer chromatography and flame-ionization detection (see "Methods"). The results are given as micrograms lipid per microgram DNA and are mean±sd (n=5 for all except cholesterol ester, when n=10) t i V i r i 0- i i i i i i Minutes I I-- j, FIG 7. Line graph showing posttranslational degradation of apolipoprotein (apo) B-100 in cells cultured in the presence ( ) or absence ( ) of oleic acid. Confluent Hep G2 cells in 9.6-cm 2 culture dishes were labeled with 50 /ici [ 35 S]methionine for 2.5 hours. The cells were then chased for 0, 20, 40, 60, 90, and 120 minutes. ApoB-100 was recovered from cells and the medium by immunoprecipitation and electrophoresis in 3% to 15% polyacrylamide gradient gels in the presence of sodium dodecyl sulfate, and the radioactivity was counted. Results are given as percent of the initial apob-100 radioactivity, ie, the radioactivity after 0 minutes' chase, recovered after each chase period.

8 1750 Arteriosclerosis and Thrombosis Vol 13, No 12 December OA + OA 0 MIN MIN FIG 8. Autoradiographs showing the initiation of apolipoprotein (apo) B-100-containing lipoproteins and the conversion of these initiated lipoproteins to low-density lipoprotein-very-low-density lipoprotein particles in cells cultured in the absence (-OA) or presence (+OA) of oleic acid. Confluent Hep G2 cells cultured in 28-cm2 culture dishes were pulse labeled with 300 ^,Ci [35S]methionine for 5 minutes and chased for 0 or 25 minutes. After each chase period, cycloheximide and puromycin were added to block the elongation of the nascent polypeptides and release them into the secretory pathway. The cells were then chased for 2.5 hours (in the presence of cycloheximide and puromycin) to recover the material that had been released into the secretory pathway in the medium. The medium was then analyzed by sucrose gradient ultracentrifugation. The fractions recovered from this gradient were immunoprecipitated by polyclonal antibodies to apob-100 and analyzed by electrophoresis in 3% to 15% polyacrylamide gradient gels in the presence of sodium dodecyl sulfate, followed by autoradiography. Arrows indicate the position of the 200- and 92.5-kDa standards. Stars indicate the migration of a protein that was nonspecifically precipitated with the antibody. We did not observe any effect of the incubation with oleic acid on the total amount of apob-100 in the luminal content (see above), nor did we find any effect of the incubation with oleic acid on the amount of JL I I I 5 6 ; 8 Fraction FIG 9. Bar graph showing quantification of the radioactivity of nascent apolipoprotein B polypeptides present in the different fractions of the gradients (see Fig 8) obtained from cells cultured in the absence (filled bars) or presence (open bars) of oleic acid. The results are given as mean±sd (n=5). DPM indicates disintegrations per minute membrane-bound apob-100 radioactivity (the ratio with oleic acid/without oleic acid is l.ll±0.07). These results could indicate that the oleic acid increased the formation of the LDL-VLDL-like particles and diverted apob-100 from Fraction I. To test this possibility, we followed the appearance of apob-100 on LDL-VLDL-like particles and on Fraction I in the microsomal lumen in cells cultured in the presence or absence of oleic acid. The cells were labeled for 10 minutes and chased for periods between 0 and 30 minutes. After each chase period, the luminal content of the total microsomal fraction was recovered (by sodium carbonate treatment) and subjected to sucrose gradient ultracentrifugation. ApoB-100 was recovered from each fraction of the gradient by immunoprecipitation and electrophoresis in 3% to 15% polyacrylamide gradient gels containing SDS, and the radioactivity was counted. The results (Fig 10) indicated that apob-100 almost exclusively appeared as Fraction I in the absence of oleic acid. On the contrary, when the cells were cultured in the presence of oleic acid, the dominating amount of pulse-labeled apob-100 appeared on LDLVLDL-like particles, and only relatively small amounts of apob-100 radioactivity were present as Fraction I (Fig 10). No difference was found in the amount of apob-100 radioactivity that was tightly bound to the membrane, nor was there any major difference in the turnover of this radioactivity (Fig 10).

9 Boren et al Triacylglycerol and the Assembly of ApoB-lOO-Containing Lipoproteins min 10 min 15 min s o s 0. o Fraction Fraction Fraction 20 min 30 min Memb s a. a a eoooo Discussion The results indicate that apob-100 occurs in three forms in the secretory pathway of Hep G2 cells: (1) tightly bound to the membrane; (2) extractable with sodium carbonate and present on LDL-VLDL-like particles; and (3) extractable with sodium carbonate but present on particles with a higher density than the LDL-VLDL-like particles. This form is referred to as Fraction I. The LDL-VLDL particles are secreted from the cells and could therefore be expected to be present in the lumen of the microsomes, a localization that is supported by the observation that the particles are both released with sodium carbonate and glycosylated. Also, Fraction I is present on the luminal side of the microsomal membrane, as inferred from the following two observations: the apob-100 of Fraction I is protected against proteolysis when present in isolated microsomes, and the protein is glycosylated. The exact localization of the apob-100 of Fraction I, ie, whether it is associated with the ER membrane or present on lipoproteins in the lumen, cannot be unequivocally established from the present results because sodium carbonate treatment is known to release peripheral membrane proteins, ie, proteins that are not integrated into the membrane. 11 Thus, we conclude that Fraction I represents a completely translocated form of apob-100 that is not tightly bound to the membrane of the secretory pathway (ie, it is sodium carbonate releasable) Minutes FIG 10. Line graphs showing the results of pulse-chase studies of the formation of low-density lipoprotein-very-low-density lipoprotein particles and Fraction I in the secretory pathway in cells cultured in the presence ( ) or absence (A) of oleic acid (OA). Confluent Hep G2 cells cultured in 56-cm 2 culture dishes in the presence or absence of OA were pulse labeled with 500 /tci [ 35 S]methionine for 5 minutes and chased for 0, 10, 15, 20, and 30 minutes. After each chase period, the total microsomal fraction was recovered and extracted with sodium carbonate. The sodium carbonate extract was analyzed by sucrose gradient ultracentrifugation for 65 hours at 4 C and rpm. The gradients were unloaded into 11 fractions, and apolipoprotein B-100 was recovered by immunoprecipitation and electrophoresis in 3% to 15% polyacrylamide gradient gels containing sodium dodecyl sulfate; the radioactivity was then counted. DPM indicates disintegrations per minute; Memb, membrane. In agreement with other investigators, 1318 we observed that the tightly membrane-bound pool of apob- 100 was susceptible to proteolysis, indicating that a large portion of the sequence was exposed on the cytoplasmic surface of the membrane. 13 ' 18 A small amount of fulllength apob-100 could, however, be detected after the proteolysis, suggesting that a portion of the tightly membrane-bound pool of apob-100 is completely translocated to the luminal side of the microsomal membrane. An alternative explanation that we cannot exclude is that this small amount of the full-length apob- 100 represents an incomplete degradation of the membrane-bound apob-100. Our results indicated that the membrane-bound, protease-sensitive apob-100 was glycosylated. Assuming that the protease sensitivity is due to an incomplete translocation of the protein, these results suggest that a portion of the molecule is translocated to the luminal side. In fact, such a translocation could be expected, since the signal sequence 23 of the membrane-bound apob-100 is cleaved. An alternative explanation would be that the glycosylation occurred on a subset of apob- 100 molecules that is completely translocated to the luminal side of the ER. The observation that the mannose-labeled apob-100 was protease sensitive argues against this possibility. In agreement with other investigators, 6-22 we also observed that incubation of the Hep G2 cells with oleic acid increased both the intracellular pool and the biosynthesis of triacylglycerol, whereas oleic acid did

10 1752 Arteriosclerosis and Thrombosis Vol 13, No 12 December 1993 not affect the rate of the biosynthesis of phosphatidylcholine. However, some reports indicate that oleic acid induces an increase in the rate of the biosynthesis of phosphatidylcholine in Hep G2 cells. Thus, Weinhold and coworkers 24 report not only that oleic acid induces an increase in the rate of incorporation of choline into phosphatidylcholine but also that the fatty acid activates the enzyme cytidyltransferase. These results agree with observations in other cell types. 25 ' 26 The reason for the conflicting results with the Hep G2 cells is unclear; however, differences in the experimental procedure may be a tentative explanation. Thus, Weinhold et al 24 used [ 14 C]choline instead of [ 14 C]glycerol to measure the rate of the biosynthesis of phosphatidylcholine. Moreover, whereas Weinhold et al 24 incubated the cells with oleic acid for periods of up to 2 hours, during which time the biosynthesis rate of phosphatidylcholine was measured, we cultured the cells for longer periods (up to 48 hours) before the experiments were performed. Further investigation is needed to clarify whether these differences could explain the differences observed in the rate of the biosynthesis of phosphatidylcholine. Although this uncertainty of the effect of oleic acid on the rate of the biosynthesis of phosphatidylcholine must be taken into account when generalizing from our results, the experimental conditions described in this article appear to offer a model for studies of the assembly of the lipoprotein core. Previous results show that incubation with oleic acid induces a severalfold increase in the secretion of apob-100. This increase is not due to an increased transcription of the apob-100 gene, 4-6 but rather appears to be explained by an increased use of apob-100 for lipoprotein assembly 1 and a decreased intracellular degradation of the protein. 7 These observations were confirmed in the present article, and, as could be expected, the increased rate in the biosynthesis of triacylglycerol appeared to have a fundamental influence on the lipoprotein assembly process. It should, however, be pointed out that not all cells respond to oleic acid and the subsequent increase in the rate of the biosynthesis of triacylglycerol, with an increased rate of apob secretion. Thus, there is no increase in the secretion of apob from cultured rat hepatocytes after the addition of oleic acid to the culture medium. The reason is not known, but one possible explanation is that rat hepatocytes keep their in vivo phenotype during a brief culture Another reason may be that the rat hepatocytes have the capacity to increase the lipid load of the particle, thus forming a triacylglycerol-rich VLDL. Hep G2 cells, on the contrary, have a limited capacity to form lipid-rich VLDL particles. 33 One possible site at which a variation in the rate of the biosynthesis of triacylglycerol may influence the lipoprotein assembly is the cotranslational association between the nascent apob-100 proteins and the lipids, 1 ie, the initiation of lipoproteins. We estimated this rate of lipoprotein initiation as the amount of pulse-labeled nascent apob polypeptides that were associated with lipoproteins. 1 In these experiments, the labeled nascent polypeptides were released to the secretory pathway by puromycin, chased into the medium in the presence of puromycin and cycloheximide, and analyzed by gradient ultracentrifugation. Previous results indicate that the profile of nascent polypeptides recovered in the medium under these conditions corresponds very well with the pattern found in the secretory pathway. 1 The results did not support any major effect on the rate of initiation of lipoproteins; rather, we observed that the increased rate of the biosynthesis of triacylglycerol induced a severalfold increase in the proportion of the initiated lipoproteins that was converted into full-length apob-100 on LDL-VLDL-like particles, ie, particles that were secreted from the cells. These results were supported by the observation that oleic acid induced an increase in the amount of LDL-VLDL-like particles in the lumen of the secretory pathway. Moreover, a pulse-chase experiment demonstrated that the oleic acid induced an increased formation of LDL-VLDL-like particles on behalf of the formation of Fraction I. These results indicated that variation in the formation of the core lipids influences the assembly process at a step distal to the translocation of apob-100 over the ER membrane. Thus, taken together, the results presented in this article indicate that the formation of the lipid core is separate from the process that leads to the translocation of apob-100 to the luminal side of the ER membrane. However, the rate of the biosynthesis of triacylglycerol appears to be of importance for the fate of the proteins that are being translocated. Thus, if apob-100 assembles a particle that reaches the size of an LDL, the particle will be released to the lumen of the secretory pathway to be secreted. On the other hand, if the protein is underlipidated, it will appear as Fraction 1, which is retained in the cell and most likely degraded. The results discussed above, together with our earlier observations,'- 2-4 suggest that during the translocation of apob-100 the cell forms either an LDL-VLDL-like particle or a Fraction I and that the type of particle that is formed depends on the availability of triacylglycerol. This, in turn, would imply that Fraction I is not, under the present experimental conditions, a precursor to the LDL-VLDL-like particles. However, we cannot exclude the possibility that Fraction I under other circumstances could be used for the assembly of a complete lipoprotein. It has, for example, been suggested 3 that lipoproteins are assembled in two steps, the first being the formation of a dense apob particle, which, in the second step, is associated with more triglycerides to form a full-size VLDL. Indeed, Fraction I could correspond to the proposed primordial particle. 3 Since the Hep G2 cell forms relatively small amounts of full-size VLDL particles, it is possible that a precursor-product relation between Fraction I and VLDL could be observed under other experimental conditions than those described in this article. As mentioned above, the influence of oleic acid on the secretion of apob-100 lipoproteins differs between Hep G2 cells and primary cultures of hepatocytes. The Hep G2 cells may thus be a good model system for studies on the effect of variations in the rate of biosynthesis of triacylglycerol on the early events in lipoprotein assembly. However, our results did not, of course, rule out the possibility that the rate of the biosynthesis of other lipids is of importance for the assembly and secretion of lipoproteins. Thus, results from other investigators 9 show that the secretion of VLDL depends on the biosynthesis of phosphatidylcholine. It has also been suggested that the biosynthesis of both cholester-

11 Boren et al Triacylglycerol and the Assembly of ApoB-100-Containing Lipoproteins 1753 ol 34 and cholesteryl esters influences the assembly/ secretion of VLDL. The mechanism by which Fraction I is retained and degraded in the cell is unknown. It appears as if the relation between the size of the particle and the size of the apob polypeptide is of importance. We have demonstrated 1 that particles with a density similar to that of Fraction I but containing C terminal-truncated forms of apob-100 are secreted. It is also well known that rat hepatocytes secrete apob-48 on particles with the density of HDL Together this may suggest that the small, dense Fraction I particle does not allow a fulllength apob-100 to fold correctly, and thus the whole particle may be retained by a mechanism that allows the cell to retain unfolded/misfolded proteins in the cell. Other potential mechanisms for the selective retention and degradation of Fraction I may involve interaction with specific receptors such as the LDL receptor. Such an interaction may occur within the secretory pathway. Such a mechanism would imply that Fraction I would have a high affinity for the LDL receptor. Affirming information is not yet available. It is also possible that Fraction I is secreted into the medium but is rapidly removed from the medium by a receptor-mediated uptake. If this is the case, such an uptake must be very efficient, since we have not been able to detect any significant amount of Fraction I in the culture medium. Moreover, we have failed to detect any removal of newly secreted apob-100 by Hep G2 cells. 2 Similar observations were made for cultured rat hepatocytes. 39 On the other hand, it has been suggested 40 that the region surrounding the cell surface forms a compartment of unstirred water that is not in complete equilibrium with the rest of the culture medium and that this compartment is of importance for the receptormediated uptake of the lipoprotein particles. Thus, it may be possible that Fraction I is secreted into this compartment of unstirred water and quantitatively removed by the receptor and diverted into a degradative pathway. It is obvious that a rather extensive investigation is needed to elucidate the mechanism behind the degradation of Fraction I. In summary, the results of the present experiments may provide one mechanism by which oleic acid, and thus, the rate of the biosynthesis of triacylglycerol, protect apob-100 from posttranslational degradation and increase the secretion of the protein.' 27 ' 4 ' This protection may be of importance for the regulation of the secretion of apob-100, since it has been suggested that variation in the posttranslational degradation may play a fundamental role in regulating VLDL secretion. 19 Acknowledgments The research for this article was supported by grants 7142 and 8862 from the Swedish Medical Research Foundation; the Heart and Lung Foundation; the Swedish Oleo-Margarine Foundation for Nutritional Research; King Gustaf V's Foundation; the Swedish National Board for Technical Development; Nordisk Insulinfond; Torsten and Ragnar Soderbergs Foundation; the Swedish Medical Society; the Goteborg Medical Society; the Swedish Society for Medical Research; the Swedish Nutrition Foundation; and the Odd Fellow Society. We thank Margareta Evaldsson and Anita Magnusson for skillful technical assistance. We also would like to thank Dr Lillemor Mattson for help with the lipid quantifications. References 1. Boren J, Graham L, Wettesten M, Scott J, White A, Olofsson S-O. The assembly and secretion of apob 100 containing lipoproteins in Hep G2 cells: apob 100 is cotranslationally integrated into lipoproteins. J Biol Chem. 1992;267: Boren J, Wettesten M, Sjoberg A, Thorlin T, Bondjers G, Wiklund O, Olofsson S-O. The assembly and secretion of apob 100 containing lipoproteins in Hep G2 cells: evidence for different sites for protein synthesis and lipoprotein assembly. J Biol Chem. 1990;265: Spring DJ, Chen-Liu LW, Chatterton JE, Elovson J, Schumaker VN. Lipoprotein assembly: apolipoprotein B size determines lipoprotein core circumference. J Biol Chem. 1992;267: Bostrom K, Boren J, Wettesten M, Sjoberg A, Bondjers G, Wiklund O, Carlsson P, Olofsson S-O. Studies on the assembly of apo B-100-containing lipoproteins in Hep G2 cells. J Biol Chem 1988;263: Moberly JB, Cole TG, Alpers DH, Schonfeld G Oleic acid stimulation of apolipoprotein B secretion from Hep G2 cells occurs post-transcriptionally. Biochim Biophys Ada. 1990; 1042: Pullinger CR, North JD, Teng B-B, Rifici VA, Ronhild de Bnto AE, Scott J. The apolipoprotein B gene is constitutively expressed in Hep G2 cells: regulation of secretion by oleic acid, albumin, and insulin, and measurement of the mrna half-life. J Lipid Res. 1989;30: Dixon JL, Furukawa S, Ginsberg HN. Oleate stimulates secretion of apolipoprotein B-containing lipoproteins from Hep G2 cells by inhibiting early intracellular degradation of apolipoprotein B. J Biol Chem. 1991;266: Yao Z, Vance DE. Head group specificity in the requirement of phosphatidylcholme biosynthesis for very low density lipoprotein secretion from cultured hepatocytes. / Biol Chem. 1989;264: Yao Z, Vance D. The active synthesis of phosphatidylcholine is required for very low density lipoprotein secretion from rat hepatocytes. / Biol Chem. 1988;263: Wettesten M, Bostrom K, Bondjers G, Jarfeldt M, Norfeldt P-I, Carella M, Wiklund O, Boren J, Olofsson S-O. Pulse-chase studies of the synthesis of apolipoprotein B in a human hepatoma cell line, Hep G2. EurJ Biochem 1985;149: Fujiki Y, Hubbard AL, Fowler S, Lazarow P. Isolation of intracellular membranes by means of sodium carbonate treatment: application to endoplasmic reticulum. J Cell Biol. 1982;93: Bostrom K, Wettesten M, Boren J, Bondjers G, Wiklund O, Olofsson S-O. Pulse-chase studies of the synthesis and intracellular transport of apolipoprotein B-100 in Hep G2 cells. J Biol Chem. 1986;261: Davis RA, Thrift RN, Wu CC, Howell KE. Apolipoprotein B is both integrated into and translocated across the endoplasmic reticulum membrane: evidence for two functionally distinct pools. J Biol Chem. 1990;265: Morimoto T, Arpin M, Gaetani S. Use of proteases for the study of membrane insertion. Methods Enzymol. 1983;96: Matsudaira P. Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes. J Biol Chem. 1987; Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol. 1959;37: Burton K. A study of conditions and mechanism of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic add. J Biochem. 1956;62: Dixon JL, Chattapadhyay R, Huima T, Redman CM, Banerjee D. Biosynthesis of lipoproteins: location of nascent apoai and apob in the rough endoplasmic reticulum of chicken hepatocytes. J Cell Biol. 1992;117: Davis RA, Prewett AB, Chan DCF, Thompson JJ, Borchardt RA, Gallaher WR. Intrahepatic assembly of very low density lipoproteins: immunologic characterization of apolipoprotein B in lipoproteins and hepatic membrane fractions and its intracellular distribution. J Lipid Res. 1989; Knott TJ, Pease RJ, Powell LM, Wallis SC, Rail SC Jr, Innerarity TL, Blackhart B, Taylor WH, Marcel Y, Milne R, Johnson D, Fuller M, Lusis AJ, McCarthy BJ, Mahley RW, Levy-Wilson B, Scott J. Complete protein structure and identification of structural domains of human apolipoprotein B. Nature. 1986;323: De Loof H, Rosseneu M, Yang C-Y, Li W-H, Gotto AM Jr, Chan L. Human apolipoprotein B: analysis of internal repeats and homology with other apolipoproteins. J Lipid Res. 1987;28:

12 1754 Arteriosclerosis and Thrombosis Vol 13, No 12 December Ellsworth JL, Erickson SK, Cooper AD. Very low and low density lipoprotein synthesis and secretion by the human hepatoma cell line Hep G2: effects of free fatty acid. J Lipid Res. 1986;27: Xu C-F, Tikkanen MJ, Huttunen JK, Pietinen P, Butler R, Humphries S, Talmud P. Apolipoprotein B signal peptide insertion/deletion polymorphism is associated with Ag epitopes and involved in the determination of serum triglyceride levels. J Lipid Res. 1990;31: Weinhold PA, Charles L, Rounsifer ME, Feldman DA. Control of phosphatidylchohne synthesis in Hep G2 cells: effect of fatty acids on the activity and immunoreactive content of choline phosphate cytidyltransferase. J Biol Chem. 1991;266: Vance DE. Phosphatidylcholine metabolism: masochistic enzymology, metabolic regulation, and lipoprotein assembly. Bwchem Cell Biol. 1990;68: Vance DE. Regulatory and functional aspects of phosphatidylcholine metabolism. In: Vance DE, ed. Phosphatidylcholine Metabolism. Boca Raton, Fla: CRC Press; 1989: Sparks JD, Zolfaghan R, Sparks CE, Smith SH, Fisher EA. Impaired hepatic apolipoprotein B and apolipoprotein E translation in streptozotocin diabetic rats. J Clin Invest. 1992;89: Davis RA, Boogaerts JR, Borchardt RA, Malone-McNeal M, Archambault-Schexnayder J. Intrahepatic assembly of very low density hpoproteins: varied synthetic response of individual apolipoproteins to fasting. J Biol Chem. 1985;266: Duerden JM, Bartlett SM, Gibbons GF. Regulation of very-lowdensity lipid secretion in hepatocyte cultures derived from diabetic animals. Biochem J. 1989;262: Sparks JD, Sparks CE, Bolognino M, Roncone AM, Jackson TK, Amatruda JM. Effects of nonketoic streptozotocin diabetes on apolipoprotein B synthesis and secretion by primary cultures of rat hepatocytes. J Clm Invest. 1988;82: Sparks JD, Sparks CE, Miller LL. Insulin effects on apolipoprotein B production by normal, diabetic and treated-diabetic rat liver and cultured rat hepatocytes. Biochem J. 1989;261: Strobl W, Gorder NL, Fienup GA, Lin-Lee YC, Gotto AM Jr, Patsch W. Effect of sucrose diet on apolipoprotein biosynthesis in rat liver, increase in apolipoprotein E gene transcription. J Biol Chem. 1989;264: Thrift RN, Forte TM, Cahoon BE, Shore VG. Characterization of Hpoproteins produced by the human liver cell line, Hep G2, under denned conditions. J Lipid Res. 1986;27: Fungwe TV, Cagen L, Wilcox HG, Heimberg M. Regulation of hepatic secretion of very low density lipoprotein by dietary cholesterol. J Lipid Res. 1992;33: Cianflone K, Dahan S, Monge JC, Sniderman AD. Pathogenesis of carbohydrate-induced hypertriglyceridemia using Hep G2 cells as a model system. Artenoscler Thromb. 1992;12: Cianflone KM, Yasruel Z, Rodriguez MA, Vas D, Sniderman AD. Regulation of apob secretion from Hep G2 cells: evidence for a critical role for cholesteryl ester synthesis in the response to a fatty acid challenge. J Lipid Res. 1990;31: Leighton JK, Joyner J, Zamarripa J, Deines M, Davis RA. Fasting decreases apolipoprotein B mrna editing and the secretion of small molecular weight apob by rat hepatocytes: evidence that the total amount of apob secreted is regulated post-transcriptionally. J Lipid Res. 1990;31: Fainaru M, Felker TE, Hamilton RL, Havel RJ. Evidence that a separate particle containing B-apoprotein is present in highdensity Hpoproteins from perfused rat liver. Metabolism. 1977;26: Sjbberg A, Oscarsson J, Bostrom K, Innerarity TL, Eden S, Olofsson S-O. Effect of growth hormone on apolipoproteinb (apob) messenger ribonucleic acid editing, and apob 48 and apob 100 synthesis and secretion in the rat liver. Endocrinology. 1992; 130: Williams KJ, Brocia RW, Fisher EA. The unstirred water layer as a site of control of apolipoprotein B secretion. J Biol Chem. 1990; 265: White AL, Graham L, LeGros J, Pease RJ, Scott J. Oleatemediated stimulation of apolipoprotein B secretion from rat hepatoma cells: a function of the ability of apolipoprotein B to direct lipoprotein assembly and escape presecretory degradation. J Biol Chem. 1992;267: