Atherosclerosis and Lipoproteins |
From the Metabolic Research Laboratory, Oxford Centre for Diabetes, Endocrinology and Metabolism, Nuffield Department of Clinical Medicine, University of Oxford, Radcliffe Infirmary, Oxford, UK.
Correspondence to Dr G.F. Gibbons, Metabolic Research Laboratory, Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HE, UK. E-mail geoff.gibbons{at}mrl.ox.ac.uk
| Abstract |
|---|
|
|
|---|
Key Words: primary hepatocytes apolipoprotein B insulin phosphoinositide 3-kinase triacylglycerol
| Introduction |
|---|
|
|
|---|
Synthesis of VLDL from apoB is thought to occur in at least 2 distinct stages. The first stage involves stabilization of translocating apoB, locking it into a conformation that favors subsequent lipid addition.11,12 This conformation effectively requires interaction between newly translocating/translocated apoB and lipids, possibly in the endoplasmic reticular membrane.11,12 The second stage requires transfer of a large quantity of TAG to this stable VLDL-precursor form of apoB, resulting in a fully mature particle of VLDL.1113 It is not known which of these stages is inhibited by insulin.
The fungal metabolite brefeldin A (BFA) interrupts anterograde vesicular transfer between intracellular organelles in secretory cells14,15 and acts primarily by inhibiting the activity of a GTP-exchange protein associated with ADP-ribosylation factor-1, a small GTPase.16 In addition, BFA, at a lower concentration, specifically inhibits the bulk transfer of lipid to the small apoB-containing VLDL precursor with the density of HDL. However, the formation and secretion of apoB-HDL is largely unaffected.11,12,17,18
The use of BFA to selectively interrupt VLDL assembly in this way was exploited to determine whether insulin suppressed the overall process of VLDL assembly by specific inhibition of the bulk lipid addition stage and, if so, whether PI3-K was involved. LY 294002 was the PI3-K inhibitor chosen for these experiments.10
| Methods |
|---|
|
|
|---|
| Results |
|---|
|
|
|---|
|
|
The reason for this additional effect of LY 294002 on the secretion of apoB-48 as VLDL and HDL is not clear, and we cannot rule out the possibility that LY 294002 affects a process(s) other than PI3-Kmediated insulin signaling. The mechanism of VLDL apoB-48 particle assembly differs subtly from that of VLDL apoB-100.18 Perhaps the different mechanism that characterizes the former process is susceptible to an LY 294002dependent event not involving PI3-K, which may also be responsible for the trend toward an increased HDL apoB output in the presence of LY 294002 alone (Figure 2).
When Early and Late Stages of VLDL Assembly Are Uncoupled, Addition of LY 294002 During Late Stage Abolishes Overall Inhibitory Effect of Insulin
In the experiment described above, LY 294002 was added only at the end of the initial 30-minute chase. During this first chase, the initial but not the final stage of the overall VLDL assembly process had been completed. Thus, PI3-K was inhibited only during the subsequent 4-hour chase encompassing the bulk lipid addition step. This experiment suggested that insulin signaling through PI3-K inhibited only the last stage of VLDL assembly. For verification, 2 further experiments were carried out in which BFA was used to specifically suppress any bulk lipid addition that might have occurred in the first experiment during the pulse label and 30-minute chase. A low (0.2 µg/mL) concentration of BFA, which inhibited only the maturation phase of VLDL synthesis but not the entire secretory process, was chosen.11 When BFA was present during the pulse and subsequent chase periods, the secretion of VLDL apoB-48 and apoB-100 was inhibited by 82.3±5.2% and 86.3±3.0%, respectively, of control values, which were 1169±691 and 1267±557 dpm, respectively (P<0.001 and P<0.002, n=3). The secretion of HDL apoB-48 and apoB-100 were not significantly suppressed (29.4±10.4% and 22.2±29.8% decrease, respectively, compared with the control values, which were 2657±1274 and 1855±899 dpm, respectively). Neither was the secretion of albumin affected under these conditions (data not shown). Removal of BFA at the end of the first 30-minute chase abolished the inhibition of the maturation phase, and VLDL apoB secretion was reestablished, with a 2.9±0.4-fold greater secretion of apoB-100 and 3.2±1.3-fold greater secretion of apoB-48 VLDL compared with that in the presence of BFA. Cells cultured in the presence or absence of insulin as described above were exposed to BFA during the pulse and first (30-minute) chase. Conditions for the completion of VLDL assembly were then established by removal of BFA. In the first of these experiments, when insulin was removed from cultures that had been pretreated with insulin, the secretion of either form of VLDL apoB was no longer inhibited (Table 1). In fact, a small, but significant, stimulatory effect was observed on the secretion of both apoB isoforms.
|
In the second experiment, insulin signaling via PI3-K was abolished by the addition of LY 294002 to the 78 nmol/L insulincontaining cell cultures simultaneously with the removal of BFA at the start of the 4-hour chase. The secretion of VLDL apoB-48 and VLDL apoB-100 was then compared with that observed when insulin alone was present throughout. Figure 3 shows that 78 nmol/L insulin alone significantly suppressed the secretion of VLDL apoB-48 and apoB-100 by 56.0±8.0% and 53.3±1.9% of control values, respectively. The addition of BFA during the pulse and chase periods further suppressed VLDL apoB secretion. Removal of BFA from the insulin-containing cultures, at the start of the 4-hour chase, restored VLDL apoB secretion to that observed in the presence of insulin alone. Under these conditions, when the PI3-K inhibitor was added, the rate of VLDL apoB secretion exceeded that observed in the presence of insulin alone (P<0.02). Secretion of VLDL apoB-48 was now indistinguishable from that observed in the 0.1 nmol/L insulin controls (Figure 3), although the secretion of apoB-100 remained somewhat (but not significantly) suppressed. Although when insulin was present, removal of BFA during the second chase restored VLDL apoB secretion (Figure 3), this was not the case when insulin was absent. Perhaps, in the presence of BFA, insulin stimulated the formation of the VLDL precursor so that more was available at the start of the maturation phase. This explanation is supported by the data shown in Table 1, in which insulin was present with BFA only during the formation of the VLDL precursor. In this experiment, the output of VLDL apoB was increased compared with that of the low insulin control during the maturation phase.
|
Changes in VLDL ApoB Secretion Are Associated With Reciprocal Changes in Cellular ApoB Degradation
Despite the decreased output of newly synthesized VLDL apoB in the presence of the high insulin concentration, there was no accumulation of labeled apoB within the cell and no compensatory increase in the secretion of HDL apoB (Figure 2). Summation of the amount of labeled apoB secreted and that remaining in the cell between the beginning and the end of the 4-hour chase confirmed that increased intracellular degradation was responsible for the deficit. Compared with the 0.1 nmol/L insulin controls, in which recoverable disintegrations per minute were 3707±1199 and 4883±1828 for apoB-100 and apoB-48, respectively, only 77±5.2% (P<0.05) and 74.7±9.0% of apoB-100 and apoB-48, respectively, were recovered in the presence of 78 nmol/L insulin (Figure 4).
|
Similar calculations revealed that when VLDL apoB secretion was restored by LY 294002, the increased insulin-mediated degradation of apoB was abolished. Thus, in the experiment described in Figure 3, in which BFA was used to uncouple the early and late phases of VLDL assembly, the addition of LY 294002 to the insulin-treated cells did not result in increased degradation compared with the low (0.1 nmol/L) insulin controls treated in an identical manner (Figure 4).
Insulin Inhibits VLDL TAG Secretion, and This Inhibition Is Abolished by LY 294002
VLDL TAG secretion was decreased by 28.7±0.3% (P<0.01) when 78 nmol/L insulin had been present throughout. The addition of LY 294002 during the final (4-hour) chase abolished insulin inhibition, and in this case, secretion was 95.5±5.3% of the control value. Under these conditions LY 294002 alone, in the absence of the high insulin concentration, had no effect on VLDL TAG output (Table 2). In these experiments, insulin inhibited the secretion of TAG less than the secretion of newly synthesized apoB, an effect that possibly resulted in the secretion of more highly lipidated VLDL particles.
|
| Discussion |
|---|
|
|
|---|
Insulin inhibits the overall assembly of VLDL from apoB (see review1) by a mechanism that involves an increase in apoB degradation. Insulin injection directly into the portal vein in humans decreases VLDL output into the hepatic vein.4 A postprandial decrease in splanchnic VLDL output after an oral glucose load has also been reported,20 and increased portal insulin resulting from tolbutamide infusion suppresses VLDL output in human subjects.21 Indirect evidence suggests that in the postabsorptive state after a decrease in portal insulin concentration, there is an increased mobilization of stored hepatic TAG for secretion as VLDL.22,23 Direct measurements in rats after a glucose load when portal insulin levels might be expected to rise also show a decreased VLDL output.24
Presently, no study has ascertained which stage of VLDL assembly is inhibited by insulin. Overall suppression of apoB secretion by insulin in hepatocytes requires an activation of PI3-K,10 as shown by LY 294002 inhibition of this pathway. In the present work, initial experiments showed that insulin inhibited the secretion of pulse-labeled apoB-48 and apoB-100 as VLDL (Figure 1) but not small HDL-like particles, which are similar to, or identical with, the VLDL precursor form (Figure 2). This indicated an inhibitory effect of insulin at a stage after the formation of the VLDL precursor, as did the fact that the addition of the PI3-K inhibitor LY 294002 to insulin-treated cells after precursor formation abolished the inhibitory effect of insulin on the secreted VLDL apoB (Figure 1).
The overall pathway of VLDL synthesis remained intact in the above experiments, and it was not possible to study exclusively 1 particular step of the overall process. This difficulty was overcome by using BFA to isolate the early from late stage(s) of VLDL assembly. Previous work with McArdle RH 7777 cells showed that a low concentration of BFA specifically suppressed the bulk lipid addition step but left the formation of the precursor form unaffected.11 BFA removal, at this stage, reestablished synthesis of the mature VLDL. These observations provided an opportunity to study the role of insulin signaling through PI3-K specifically at the bulk lipid addition step in primary hepatocyte culture. Thus, in the present work, a BFA concentration of 0.2 µg/mL selectively suppressed the secretion of VLDL apoB-48 and apoB-100 but had no effect on the secretion of labeled albumin. There was little effect on the secretion of small HDL apoB particles. Restoration of VLDL assembly was demonstrated by removal of BFA after the pulse and first (0.5-hour) chase, when an
3-fold increase in VLDL apoB-48 and apoB-100 secretion was seen during the subsequent 4-hour chase (Figure 3).
Specific inhibition of the maturation phase of VLDL assembly did not increase the secretion of HDL apoB particles. Instead, apoB degradation was increased by 30±9% for apoB-100 and by 20±2% for apoB-48 compared with that in low-insulin controls (n=3, P<0.05). This probably resulted from retention of apoB in a subcellular compartment accessible to proteases normally responsible for posttranslational degradation of apoB.25,26 When BFA abolished VLDL maturation, insulin did not further increase apoB degradation (results not shown).
Abolition of insulin signaling during this period of maturation after removal of BFA made it possible to distinguish the inhibitory effect of insulin either during VLDL precursor formation and during bulk lipid addition. In the present experiments, insulin signaling was terminated in 2 ways. First, removal of insulin during the maturation phase abolished inhibition of VLDL assembly (Table 1). The second way involved addition of the PI3-K inhibitor LY 294002, in which insulin failed to inhibit the maturation of VLDL, as shown by VLDL apoB output during the 4-hour chase (Figure 3). Net synthesis of apoB was unaffected by insulin during the pulse-labeling period (results not shown). The small stimulation of VLDL apoB secretion, when insulin was removed during the maturation phase, suggested that the early step of VLDL synthesis was actually enhanced by insulin (Table 1).
The present results suggest that the insulin-mediated degradation of apoB shown previously8 resulted from a failure to achieve bulk lipid addition to the VLDL precursor. When this step was restored by preventing the PI3-Kmediated transduction of the insulin signal, enhanced apoB degradation was no longer observed (Figure 4). It is becoming increasingly clear that intracellular apoB is susceptible to degradation, not only translationally and immediately posttranslationally (as was originally thought) but at all stages during its assembly into VLDL.26
If the inhibitory effect of insulin is limited to the maturation phase of VLDL assembly, it is not yet clear whether this results from a defective transfer of TAG into the apoB-free particle or from a defective fusion process per se. The cytosolic TAG storage pool is the major source of VLDL TAG (see review27). Insulin enhances the recycling of fatty acids liberated by lipolysis of cytosolic TAG back into the cytosolic storage pool at the expense of transfer into the secretory pool, the probable source of the apoB-free lipid precursor of VLDL.27 Inhibition of microsomal triglyceride transfer protein has a similar effect.28,29
The molecular mechanisms involved in VLDL maturation remain far from clear. However, recent work has provided evidence of an obligatory role for phosphatidic acid,30 which results from the ADP-ribosylation factor-1mediated stimulation of phospholipase D.31 This mechanism would account for the observed conversion of cellular phosphatidylcholine into VLDL TAG.32
In summary, the present work provides 3 lines of evidence that the inhibitory action of insulin on the overall synthesis of VLDL is limited to the bulk lipid addition stage. First, interference with insulin signal transduction, after the initiation phase of VLDL synthesis had been completed, abolished the normal insulin-dependent inhibition of VLDL assembly. Second, under these conditions, insulin was no longer able to stimulate the intracellular degradation of apoB. Finally, insulin had no inhibitory effect on the synthesis and secretion of dense apoB-containing particles possibly related to the VLDL precursor particles. Further support for the above conclusion was provided by the observation that removal of insulin from the culture medium, after the initial phase of VLDL assembly had been completed, increased the subsequent rate of VLDL apoB secretion.
| Acknowledgments |
|---|
Received March 27, 2001; accepted June 28, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. M. Allister, E. E. Mulvihill, P. H. R. Barrett, J. Y. Edwards, L. P. Carter, and M. W. Huff Inhibition of apoB secretion from HepG2 cells by insulin is amplified by naringenin, independent of the insulin receptor J. Lipid Res., October 1, 2008; 49(10): 2218 - 2229. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Morral, H. J. Edenberg, S. R. Witting, J. Altomonte, T. Chu, and M. Brown Effects of glucose metabolism on the regulation of genes of fatty acid synthesis and triglyceride secretion in the liver J. Lipid Res., July 1, 2007; 48(7): 1499 - 1510. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Tsai, W. Qiu, R. Kohen-Avramoglu, and K. Adeli MEK-ERK Inhibition Corrects the Defect in VLDL Assembly in HepG2 Cells: Potential Role of ERK in VLDL-ApoB100 Particle Assembly Arterioscler. Thromb. Vasc. Biol., January 1, 2007; 27(1): 211 - 218. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. M. den Boer, P. J. Voshol, F. Kuipers, J. A. Romijn, and L. M. Havekes Hepatic glucose production is more sensitive to insulin-mediated inhibition than hepatic VLDL-triglyceride production Am J Physiol Endocrinol Metab, December 1, 2006; 291(6): E1360 - E1364. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. V. Chirieac, N. O. Davidson, C. E. Sparks, and J. D. Sparks PI3-kinase activity modulates apo B available for hepatic VLDL production in apobec-1-/- mice Am J Physiol Gastrointest Liver Physiol, September 1, 2006; 291(3): G382 - G388. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Duvillard, E. Florentin, S. Baillot-Rudoni, M.-L. Lalanne-Mistrich, A. Brun-Pacaud, J.-M. Petit, J.-M. Brun, P. Gambert, and B. Verges Comparison of Apolipoprotein B100 Metabolism between Continuous Subcutaneous and Intraperitoneal Insulin Therapy in Type 1 Diabetes J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5761 - 5764. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Grefhorst, J. Hoekstra, T. G. J. Derks, D. M. Ouwens, J. F. W. Baller, R. Havinga, L. M. Havekes, J. A. Romijn, and F. Kuipers Acute hepatic steatosis in mice by blocking {beta}-oxidation does not reduce insulin sensitivity of very-low-density lipoprotein production Am J Physiol Gastrointest Liver Physiol, September 1, 2005; 289(3): G592 - G598. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Adiels, J. Boren, M. J. Caslake, P. Stewart, A. Soro, J. Westerbacka, B. Wennberg, S.-O. Olofsson, C. Packard, and M.-R. Taskinen Overproduction of VLDL1 Driven by Hyperglycemia Is a Dominant Feature of Diabetic Dyslipidemia Arterioscler. Thromb. Vasc. Biol., August 1, 2005; 25(8): 1697 - 1703. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Qiu, R. K. Avramoglu, N. Dube, T. M. Chong, M. Naples, C. Au, K. G. Sidiropoulos, G. F. Lewis, J. S. Cohn, M. L. Tremblay, et al. Hepatic PTP-1B Expression Regulates the Assembly and Secretion of Apolipoprotein B-Containing Lipoproteins: Evidence From Protein Tyrosine Phosphatase-1B Overexpression, Knockout, and RNAi Studies Diabetes, December 1, 2004; 53(12): 3057 - 3066. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. M. Borradaile, L. E. de Dreu, and M. W. Huff Inhibition of Net HepG2 Cell Apolipoprotein B Secretion by the Citrus Flavonoid Naringenin Involves Activation of Phosphatidylinositol 3-Kinase, Independent of Insulin Receptor Substrate-1 Phosphorylation Diabetes, October 1, 2003; 52(10): 2554 - 2561. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |