Pulse-chase studies of the synthesis and intracellular transport of apolipoprotein B-100 in Hep G2 cells.

The synthesis and secretion of apolipoprotein B-100 (apoB-100) have been studied in a human hepatoma cell line, the Hep G2 cells. The time needed for the synthesis of apoB-100 was estimated to be 14 min, which corresponds to a translation rate of approximately 6 amino acids/s. ApoB-100 was compared with albumin and alpha 2-macroglobulin as to the distribution between the membrane and the luminal content in the endoplasmic reticulum (ER) and the Golgi apparatus. The results suggested that apoB-100 approximately followed the distribution of these secretory proteins in the Golgi, while the ratios between the percent membrane-bound apoB-100 and percent membrane-bound albumin or alpha 2-macroglobulin were 3-4:1 in the ER. This may suggest that apoB-100 occurs in a membrane-associated form in ER prior to the integration in the lipoproteins. Pulse-chase studies combined with subcellular fractionation was used to investigate the kinetics for the intracellular transfer of apoB-100. A 3-min pulse of [35S]methionine was followed by an increase in apoB-100 radioactivity in the ER during the first 10-15 min of chase. The following 10-15 min of chase were characterized by linear decrease in apoB-100 radioactivity with a decay rate of approximately 6%/min. The residence kinetics for apoB-100 in the ER differed from that of transferrin and probably also from that of albumin. By comparing the time for the pulse maximum in ER with that in the denser Golgi fractions the time needed for the transfer between ER and Golgi could be estimated to be 10 min. The time needed for the secretion of newly synthesized apoB-100 was estimated to be 30 min. This indicates that the transfer of the protein through the Golgi apparatus to the extracellular space requires 20 min.

The synthesis and secretion of apolipoprotein B-100 (apoB-100) have been studied in a human hepatoma cell line, the Hep G2 cells.
The time needed for the synthesis of apoB-100 was estimated to be 14 min, which corresponds to a translation rate of approximately 6 amino acidsls.
ApoB-100 was compared with albumin and az-macroglobulin as to the distribution between the membrane and the luminal content in the endoplasmic reticulum (ER) and the Golgi apparatus. The results suggested that apoB-100 approximately followed the distribution of these secretory proteins in the Golgi, while the ratios between the percent membrane-bound apoB-100 and percent membrane-bound albumin or az-macroglobulin were 3-4:l in the ER. This may suggest that apoB-100 occurs in a membrane-associated form in ER prior to the integration in the lipoproteins.
Pulse-chase studies combined with subcellular fractionation was used to investigate the kinetics for the intracellular transfer of apoB-100. A 3-min pulse of [36S]methionine was followed by an increase in apoB-100 radioactivity in the ER during the first 10-15 min of chase. The following 10-15 min of chase were characterized by linear decrease in apoB-100 radioactivity with a decay rate of approximately 6%/min. The residence kinetics for apoB-100 in the ER differed from that of transferrin and probably also from that of albumin.
By comparing the time for the pulse maximum in ER with that in the denser Golgi fractions the time needed for the transfer between ER and Golgi could be estimated to be 10 min.
The time needed for the secretion of newly synthesized apoB-100 was estimated to be 30 min. This indicates that the transfer of the protein through the Golgi apparatus to the extracellular space requires 20 min.
The major protein component of the low density lipoproteins (LDL)' of human plasma is referred to as apoB-100 (1).

* This study was supported by grants from the Swedish Medical
Research Council (7142 and 4531), the Swedish Oleo-Margarine Foundation of Nutritional Research, the Swedish National Association against Heart and Lung Diseases, the Goteborg Medical Society, and King Gustaf V's and Queen Victoria's Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The abbreviations used are: LDL, low density lipoproteins; apoB-100, apolipoprotein B-100 (for definition see Ref. 1); ER, endoplasmic reticulum; SDS, sodium dodecyl sulphate; VLDL, very low density lipoproteins. This protein has a molecular mass of more than 300 kDa (1)(2)(3) and is synthesized in the liver. It is coded for by a 20kilobase mRNA (4)(5)(6)(7)(8)(9)(10) but in contrast to some other proteins coded for by large messengers, apoB-100 is secreted without undergoing any major proteolytic processing (3) (compare, for example, the insulin receptor (11,12), the C3 (13) It is unclear whether the intracellular transport of apoB-100 follows that of other secretory proteins, as apoB-100 is integrated into a complex particle, the lipoprotein. The processes involved in the assembly and modification (21)(22)(23)(24) of the lipoprotein are still not completely elucidated. In view of these considerations we have undertaken this study to characterize the intracellular transport of apoB-100.
ApoB-100 is a hydrophobic protein (8) with physical-chemical properties that have been suggested to resemble those of membrane proteins more than those of the other apolipoproteins (25). In addition structural analyses based on sequences of cloned cDNA have revealed the presence of hydrophobic fi sheets (5,8) and it has been suggested that the protein is woven in and out of the hydrophobic portion of the lipoprotein similar to an integral membrane protein (5). To investigate whether apoB-100 follows the classical route of a secretory protein (see above) or if it may at some stage of the secretion appear as the membrane proteins, we have compared the distribution of apoB-100 between the membrane and content of the ER and Golgi with that to two secretory proteins, albumin and a2-macroglobulin.
Cell Culture-Hep G2 cells (26,27) were made available by the courtesy of Drs. B. B. Knowles and D. P. Aden (The Wistar Institute, Philadelphia, PA). The cells were cultured as described earlier (3).
Pulse-Chase Studies-The first set of experiments was undertaken to investigate the intracellular transport of apoB-100. Confluent cells in 28-cm2 culture dishes were incubated with methionine-free Eagle's minimum essential medium for 2 h. The cells were then pulsed for 3 min with 500 pCi of [35S]methionine and chased for 0, 2.5, 5, 7.5, 10, 20,30, 40, and 90 min or in a second set of experiment for 0, 10, 15, 20, 25, 30,35, and 40 min (19). After each chase period the cells were lysed and the cell lysate was fractionated on a linear sucrose gradient (see below). ApoB-100 was recovered from each fraction as well as from the culture medium by immunoprecipitation and SDS-polyacrylamide gel electrophoresis (3). The band corresponding to apoB-100 was identified by autoradiography, cut out, and the radioactivity measured as described earlier (3).
The second set of experiments was carried out to compare the residence kinetics for apoB-100 in the ER with that of transferrin and albumin. The cells were cultured and preincubated as described above and pulsed with 300 pCi of [%]methionine for 10 min. The pulse was followed by chase for 0, 10,20,30,40, and 90 min (19), and the cells were recovered and lysed (3) after each chase period. A fraction enriched in membranes derived from the ER was isolated from the cell lysate by centrifugation on a discontinuous sucrose gradient (see below). ApoB-100, transferrin, and albumin were recovered from the ER fraction by consecutive immunoprecipitation (anti-apoB-100 followed by anti-transferrin and finally anti-albumin) and polyacrylamide gel electrophoresis. The radioactivity in the bands corresponding to apoB-100, albumin, and transferrin were determined as described above.
Subcellular Fractionation on a Linear Sucrose Gradient-The cells were harvested and lysed as described by Fries et al. (19) with the exception that all buffers used contained 1 mM phenylmethylsulfonyl fluoride and 0.1 mM leupeptin.
The homogenization (20 strokes with a tight fitting pestle in a Dounce homogenizer at 0 'C) resulted in less than 5% intact cells as judged by dye exclusion test or counting in a cell counter (A. J. Cellcounter 134, Analys Instrument, Lidingo, Sweden). The sucrose concentration of the cell lysate (500 pl) was adjusted to 250 mM. The lysate was then centrifuged at 1900 X g for 10 min at +4 "C, and the supernatant (sup I) was recovered. The pellet was gently suspended (by vortexing) in 300 p l of 250 mM sucrose, and a supernatant (sup 11) was recovered by centrifugation as described above. The pellet was again suspended in 250 pl of 250 mM sucrose and centrifuged as described above. The recovered supernatant was combined with supernatants I and I1 and layered on top of a 3.5-ml sucrose gradient (22-49%, w/w) which in turn was layered on top of a 0.6-ml cushion of 49% (w/w) sucrose. This system was adopted from Beaufay and co-workers (28) but was scaled down to fit the purpose of subfractionation of lysed Hep G2 cells from one 28-cmZ culture dish. Furthermore, the gradient was adjusted to obtain a maximal separation between the fractions containing the NADPH depending cytochrome c reductase activity and the fractions containing the galactosyltransferase activity (cf. Fig. 1).
After centrifugation for 15 h at 35,000 rpm and +4 "C in a Beckman SW 50.1 rotor, the tubes were unloaded from the bottom into 24 fractions (8 drops were recovered in each of the first 13 fractions, 14 drops in each of the next 7 fractions, and finally 20 drops in the last 4 fractions; cf. Fig. 1). Each fraction was assayed for NADPHcytochrome c reductase activity (29), galactosyltransferase activity (30), and RNA (31). Fig. 1 shows the distribution of the marker enzymes and RNA over the gradient. The NADPH-cytochrome c reductase was recovered as a peak between fraction 5 and 14. Electron micrographs (see below) of fraction 6-11 showed that granulated vesicles characteristic of the rough ER were the dominating component present (Fig. 2). Furthermore, polyacrylamide gel electrophoresis of immunoprecipitated apoB-100 showed that fraction 6-11 in addition to apoB-100, contained apoB-100 nascent polypeptides (for identification of apoB-100 nascent polypeptides, see Ref. 3). Taken together these results indicate that fractions 6-11 are enriched in membranes derived from the rough endoplasmic reticulum. We will refer to these fractions as ER in the following presentation. Most of the galactosyltransferase activity was recovered in fraction 16-19 ( Fig. 1). Electron micrographs (Fig. 2) of these fractions showed the presence of smooth membrane structures characteristic of Golgi fractions (32)(33)(34). These results indicate that. fractions 16-19 are enriched in membranes derived from the Golgi apparatus and will be referred to as Golgi in the following presentation.
The membrane and the luminal content were isolated from the ER and Golgi vesicles after disruption of the vesicles in 100 mM sodium carbonate at pH 11.5. The method of Fujiki and co-workers (35) was used with the following modification: (i) the isolated subcellular fractions were not pelleted before sodium carbonate treatment; (ii) bovine serum albumin was added (final concentration 0.5%) to the sample after the 30 min incubation on ice; (iii) the pellet was washed with 50 mM sodium phosphate, pH 7.3, with 150 mM sodium chloride and 0.5% bovine serum albumin. In our hands these modifications (i-iii) increased the recovery of apoB-100 from approximately 10% to more than 90%.
The radioactivity in apoB-100, albumin, and a*-macroglobulin was determined after isolation of the proteins by immunoprecipitation and polyacrylamide gel electrophoresis (3). Subcellular Fractionation on a Discontinuous Sucrose Gradient-We have used the system described by Fries and co-workers (19) with the following exceptions: (i) all buffers contained 1 mM phenylmethylsulfonyl fluoride and 0.1 mM leupeptin and (ii) one fraction corresponding to the bottom one-third of the tube was collected. This fraction contained 69 f 10% (x f S.D., n = 10) of the NADPHcytochrome c reductase_ recovered after the centrifugation, corresponding to 38 f 7% (X f S.D., n = 15) of the activity present in the cell lysate loaded on the gradient. Only small amounts (7 f 4, x f S.D., n = 10) of the galactosyltransferase activity recovered after centrifugation was recovered in this fraction. Based on these findings we regard this fraction I as being enriched in membranes derived from the endoplasmic reticulum.
Estimation of Half-transit Time-Half-transit time for apoB-100 was determined by the double isotope method of Palmiter (36) under conditions to minimize the influence of RNase (37). Cells at confluency in 28-cm2 culture dishes were incubated with 50 pCi of [14C] leucine for 2 h. This was followed by pulses with 75 pCi of (3H]leucine for 18,19, and 20 min. After each pulse period the cells were harvested with a rubber policeman and lysed (37). A postmitochondrial supernatant was recovered (38)  isolated from the supernatant by immunoprecipitation and electrophoresis in 3-15% polyacrylamide gradient gels with SDS. The gels were cut in 1-mm pieces and apoB-100 was eluted from the gel slices by shaking in 10 ml of 3% Protosol in Econofluor at 37 "C for 24 h. The gels were then removed and the 3H-and "C-activity measured (see below).
Radioactivity was determined in a liquid scintillation counter (LKB Rack-Beta) in a system of 3% Protosol in Econofluor (for extracts from polyacrylamide gels) or 10% Protosol in Aquasol 2 (for solubilized trichloroacetic acid precipitate). Channel settings in the liquid scintillation counter had been adjusted to optimize separation between the "C and the 3H channels under the quenching conditions induced by the extraction of the polyacrylamide gels or the solubilization of the trichloroacetic acid precipitates. Quenching levels were checked for each sample with the external standard channels ratio method and was found to be compatible with the 14C?H separation intended. Under the conditions used the "C efficiency was 82% for the extract from the polyacrylamide gels and 73% for the solubilized trichloroacetic acid precipitates. The 3H efficiency was 37% for the extract from the polyacrylamide gels and 22% for the solubilized trichloroacetic acid precipitates.
Electron Microscopy-The subcellular fractions were fixed in glutaraldehyde, post-fixed in osmium tetroxide, and stained en bloc with 1% uranyl acetate in 70% ethanol (33). After dehydration in a graded series of ethanol solutions, the samples were embedded in agar resin 100 and cut into ultrathin sections on an LKB Ultrotome V. The samples were counterstained with an alkaline bismuth solution (41), and examined in a Philips 400 transmission electron microscope.

Pulse-Chase Studies Combined with Subcellular Fractionation on a Linear Sucrose Gradient-Most of the apoB-100
radioactivity was recovered in fractions 6-19 from the linear sucrose gradient, while only little radioactivity was present in fractions 1-5 and 20-24.
The residence kinetics for apoB-100 in the ER fractions (6-11) was characterized by an increase in radioactivity during the first 10 min followed by a rapid decrease during the next 10 min (Fig. 3).
The residence kinetics in E R was further characterized in three different experiments (Fig. 4) where chase periods of 0, 10,15,20,25,30,35, and 40 min were used. The choice of the chase period was based on the observations in Fig. 3 indicating that the pulse maximum in ER appeared after 10 or more than 10 min of chase. The results (Fig. 4) showed a pulse maximum occurring between 10 and 15 min (mean: 11 min) of chase followed by a almost linear decay until 25 min of chase ( Y = 166-5.9X; r = -0.93, n = 11). The curves leveled off after 30 min of chase (compare also Fig. 3).
In the denser Golgi fractions (16 and 17) the apoB-100 radioactivity reached a maximum after 20 min of chase. A comparison with the radioactivity maximum in the E R fraction, suggested that the time needed for the transfer of apoB-100 between these two compartments is approximately 10 min. This was confirmed in three additional experiments using chase periods of 0, 10, 15, 20, 25, 30, 35, and 40 min.
In agreement with our earlier findings (3), apoB-100 could first be detected in the culture medium after 30 min chase (five experiments) (Fig. 3). This indicates that the transfer time of apoB-100 through the cell is approximately 30 min (23). ApoB-100 radioactivity in the medium reached its maximum after 45 min of chase (not shown) and leveled off, suggesting a transfer time through the cell of 35 min. Thus, all these data indicate that the time needed for the intracellular transport of apoB-100 is approximately 30 min. Since we have estimated that 10 min is required for the transfer between ER and the Golgi apparatus, this means that 20 min is required for the transfer through the Golgi apparatus to the extracellular space.

Comparison between the Residence Kinetics for ApoB-100 in ER with That for Transferrin and Albumin-
The apoB-100 radioactivity in the ER fraction, isolated by the discontinuous gradient, increased during the first 10 min of chase. This was followed by a relatively rapid decrease during the next 20 min (Fig. 5). This residence kinetics did not differ substantially from that found in the ER fractions isolated by the linear sucrose gradient. The residence kinetics for apoB-

M I N U T E S
100 in the ER fraction differed from that of transferrin and probably also from that of albumin (Fig. 5). In the latter case, however, it should be noticed that the initial increase in apoB-100 associated radioactivity may impair the comparison between the two curves. Estimation of the Translation Time for apoB-100-A 1' mear increase in 3H:'4C with time was found between 18 and 20 min for apoB-100 recovered from the pool of completed proteins (Fig. 6). The linear extrapolation (36) based on pulse lengths between 18 and 20 min gave an estimated translation time for apoB-100 of 14 min as the mean of four experiments (the range was 11-17 min). Table I , the Hep G2 cells were incubated with [35S]methionine for 2 h, the ER and Golgi vesicles were recovered with ultracentrifugation in a linear sucrose gradient. The vesicles were disrupted and apoB-100, albumin, and az-macroglobulin were recovered from the membrane as well as from the luminal content. Only a minor portion of the albumin and a,-macroglobulin (Table I ) were recovered from the ER membrane, while on the other hand a substantial amount of apoB-100 was associated with the membrane. The ratio between percent membrane bound apoB-100 and percent membrane bound albumin or az-macroglobulin were 3.9 and 2.7, respectively.

Distribution of ApoB-100 Albumin and az-Macroglobulin between the Membrane and Content of the ER and Golgi Vesicles-In the experiment shown in
A somewhat larger portion of albumin and a,-macroglobulin was associated with the membrane in the Golgi fractions than it was in the ER fractions ( Table I ) . The results obtained for albumin are in agreement with those reported by Howell and Palade (34). A significantly lower proportion of apoB-100 was associated with the membrane in the Golgi fractions than in the ER fractions (Table I ) . However, it was still larger than the corresponding values for albumin (the ratio was 1.7) and az-macroglobulin (ratio 1.3). The significance of this observation is difficult to evaluate. It probably represents a slightly recovered by centrifugation on a discontinuous sucrose gradient after each period of chase. ApoB-100, albumin, and transferrin were isolated from the ER fraction by consecutive immunoprecipitations (a-apoB-100 followed by a-albumin and finally a-transferrin) and SDSpolyacrylamide gel electrophoresis. The results are given as percent of the initial radioactivity in the compartment (y axis), as a function of time (minutes on x axis).
higher affinity of apoB-100 than of albumin and a2-macroglobulin to the membrane. However, a contamination of the Golgi fractions with ER could also be of significance.
The results could suggest that a portion of the apoB-100 present in the ER is associated with the membrane, while the distribution of apoB-100 between the membrane and the content of the Golgi apparatus approximately follows that of other secretory proteins. The relationship between apoB-100 and the ER membrane was also investigated with pulse-chase methodology. The cells were pulsed with [35S]methionine for 10 min and chased for 15 min. ER fractions were isolated by ultracentrifugation on a linear sucrose gradient and the radioactivity was determined in apoB-100, albumin, and a2-macroglobulin recovered from the membrane as well as the ER content. The choice of the chase periods was based on the time needed for the synthesis of apoB-100 (14 min) and the time when most apoB-100 had left the ER compartment (20-25 min). The change of methodology from continuous labeling (Table I) to pulse-chase labeling ( Table 11) had no major influence on the proportion of membrane-bound albumin or a2-macroglobulin. Nor did the resuspension of the membrane pellet in 50 mM Tris-HC1 (pH 7.4) with 500 mM KCl, 5 mM MgC12, 250 mM sucrose, and 0.5% bovine serum albumin (34) affect these results. On the contrary the proportion of membranebound apoB-100 increased considerably (compare Tables I  and 11). The ratios between the proportion of membranebound apoB-100 uersus membrane-bound albumin and upmacroglobulin was 4.8 and 3.4, respectively.

DISCUSSION
This paper deals with the synthesis and intracellular transport of apoB-100 (1) in an established human liver cell line (Hep G2) (26,27).

TABLE I1
The amount of radioactivity rec_ouered in the ER membrane (percent of total amount in the fraction X k S.D., n = 5) after a 10-minpulse with 300 pCi of r5S]methionine/28-cm2 culture dish followed by chase for 15 min LS, the membrane pellet was washed with 50 mM sodium phosphate, pH 7.3, with 150 mM NaCl and 0.5% bovine serum albumin. HS, the membrane pellet was washed by suspension in 50 mM Tris-HCl, pH 7.4, with 500 mM KC1, 5 mM MgC12, and 250 mM sucrose, the STKM buffer (34), supplemented with 0.5% bovine serum albumin followed by repelletation. The time needed for the synthesis of apoB-100 was estimated (36) to be 14 min. Information for the calculation of the translation rate could be obtained from recent studies on the structure of apoB-100 mRNA. ApoB-100 cDNA has recently been cloned and to a large extent sequenced (4)(5)(6)(7)(8)(9)(10). Northern blot analysis with such cDNA clones have suggested that apoB-100 mRNA has a size of 15-20 kilobases (4)(5)(6)(7)(8)(9)(10). This messenger appear to contain less than 3% non-coding regions (8, lo), suggesting that it has the capacity to code for a protein of at least 5000 amino acids. Our results would therefore suggest a translation rate of approximately 6 amino acids/s. This translation rate is similar to that reported for the synthesis of apoB in estrogen-treated chick liver cells (42), and well within the range reported for other proteins (36,38,43,44).
The intracellular transport of apoB-100 in Hep G2 cells was studied by combining pulse-chase methodology with subcellular fractionation on a linear sucrose gradient. The apoB-100 radioactivity peaked in the ER after 10-15 min chase. This is in agreement with an estimated translation time of 14 min. An approximately linear decay of apoB-100 radioactivity was found between 11 and 25 min chase. Based on the finding of a translation time for apoB-100 of approximately 14 min, we made the assumption that the inflow of apoB-100 in the E R compartment could be regarded as neglectible during the linear decay. Using this decay, we estimated the efflux of apoB-100 from ER to be 6%/min. It has to be kept in mind that these results may be influenced by the contamination of the ER fractions with Golgi vesicles. To estimate the influence of such a contamination, we made the assumption that the majority of the apoB-100 radioactivity present in the ER fractions after chase periods of 30 min or more (when the curves had leveled off) was due to contamination of Golgi vesicles. Making this assumption, we could estimate that an ER fraction was at the most contaminated with 3% of the apoB-100 radioactivity present in the Golgi vesicles. The maximal theoretical effect of such a contamination was an increase of the slope from 6 to 6.3%/min.
The peak of apoB-100 radioactivity was recovered in the denser Golgi fractions after 20 min chase. This suggests that newly synthesized apoB-100 molecules are transported from ER to Golgi within 10 min. Since the time needed for the intracellular transport of apoB-100 was estimated to be approximately 30 min, our results indicate that approximately one-third of the time needed for intracellular transport of apoB-100 is due to the transfer from ER to Golgi. These results are in agreement with the findings for other secretory proteins by Yeo et al. (20). Stein and Stein (45) have presented results from electron microscopy studies indicating that [3H] palmitate or [3H]glycerol that had been injected into rats could be found in the ER fraction of the liver 2 min after the injection, and in Golgi, in lipoprotein form, 10 min after the injection. These in vivo data on the lipid moiety of the lipoprotein are in good agreement with our results on apoB-100.
The residence kinetics for apoB-100 in ER differed from that of transferrin and probably also from that of albumin. This result is consistent with the report by other authors (18,19,46), indicating that different secretory proteins leave ER with different kinetics. It should be noticed that the results obtained for albumin and transferrin are in agreement with those presented by Fries and co-workers (19) using the same system for subcellular fractionation.
The results obtained indicate that the transfer of apoB-100 through the later part of the secretory pathway accounts for approximately two-thirds of the time needed for the intracellular transport of apoB-100. This relatively long transfer time through the Golgi apparatus to the cell surface may be the explanation for the finding of a pulse maximum after 30 min chase in the less dense Golgi fractions (cf. Fig. 3). It should, however, be noticed that the Hep G2 cells express the LDL receptor (47)(48)(49). Thus it is possible that the recidence kinetics in the Golgi fractions after more than 30 min chase, may be influenced by apoB-100 in endocytosis vesicles. Such vesicles have been shown to be contaminants in the density fractions containing Golgi vesicles (50). In order to estimate the influence of this contamination we carried out the following experiment: Hep G2 cells were labeled with 200 pCi of [35S]methionine/28-cm2 culture dish for 1 h. The medium were changed and the metabolically labeled apoB-100 were chased into the medium for 1 h. The chase medium was transferred to another 28-cm2 culture dish with confluent Hep G2 cells and incubated for 2 h.
Cells and medium were collected and apoB-100 were recovered by immunoprecipitation and polyacrylamide gel electrophoresis and the radioactivity determined. Only 1% of the apoB-100 radioactivity was associated with the cells. Together with the finding of Hornick et al. (50) that 1.6% of the endocytosed LDL was found in the Golgi fractions, our finding suggests a very minor influence of endocytosed apoB-100 radioactivity on the apoB-100 radioactivity recovered from the Golgi fractions. Furthermore, most of the results presented in this paper are based on chase periods shorter than 30 min, i.e. before the apoB-100 radioactivity has appeared in the medium. Even under the conditions where the cells had been labeled continuously for 2 h, less than 1% of the radioactivity recovered in apoB-100 from the Golgi could be derived from endocytosed apoB-100.
The relatively long residence time for apoB-100 in the later part of the secretory pathway may explain the observations from ultrastructural studies (21,45,51) that VLDL particles are concentrated within the Golgi.
The distribution of apoB-100 between the membrane and the content of the Golgi vesicles followed approximately that of the two secretory proteins; albumin and a,-macroglobulin. This was, however, not the case in the ER vesicles where apoB-100 appeared to be associated with the ER membrane to a greater extent. The possibility that the association is a function of the long translation time, i.e. is due to nascent polypeptide chains, is less likely since (i) only the band corresponding to the completed protein have been excised and counted, (ii) the polysomes are stripped from the membranes during the disruption, and (iii) the proportion of membranebound apoB-100 increased rather than decreased when the cells were pulse labeled and chased for 15 min, although the nascent polypeptides disappear after this chase period. The main problem in interpreting the results is, however, the possibility that apoB-100 unspecifically interacts with the membrane during the preparation. It is possible that the finding that a somewhat higher proportion of apoB-100 than of albumin or az-macroglobuiin is associated with the Golgi membrane is due to a higher affinity of apoB-100 for the membranes of the vesicles. The ratios between the proportion of membrane-bound apoB-100 uers'sus albumin or a2-macroglobulin are, however, 2-3 times higher in the ER than in the Golgi. It is therefore less likely that artificial binding explains the entire association between apoB-100 and the ER membrane. The observation of a higher proportion of a membrane associated apoB-100 in ER after 15 min chase than after 2 h continuous labeling of the cells, could be consistent with a model in which apoB-100 is co-translationally integrated into a membrane associated "pool" and from this pool transferred to the lumen. This model, as well as its relation to the lipoprotein assembly process, is now under investigation in our laboratory.