Hepatic very low density lipoprotein-ApoB overproduction is associated with attenuated hepatic insulin signaling and overexpression of protein-tyrosine phosphatase 1B in a fructose-fed hamster model of insulin resistance.

A fructose-fed hamster model of insulin resistance was previously documented to exhibit marked hepatic very low density lipoprotein (VLDL) overproduction. Here, we investigated whether VLDL overproduction was associated with down-regulation of hepatic insulin signaling and insulin resistance. Hepatocytes isolated from fructose-fed hamsters exhibited significantly reduced tyrosine phosphorylation of the insulin receptor and insulin receptor substrates 1 and 2. Phosphatidylinositol 3-kinase activity as well as insulin-stimulated Akt-Ser473 and Akt-Thr308 phosphorylation were also significantly reduced with fructose feeding. Interestingly, the protein mass and activity of protein-tyrosine phosphatase-1B (PTP-1B) were significantly higher in fructose-fed hamster hepatocytes. Chronic ex vivo exposure of control hamster hepatocytes to high insulin also appeared to attenuate insulin signaling and increase PTP-1B. Elevation in PTP-1B coincided with marked suppression of ER-60, a cysteine protease postulated to play a role in intracellular apoB degradation, and an increase in the synthesis and secretion of apoB. Sodium orthovanadate, a general phosphatase inhibitor, partially restored insulin receptor phosphorylation and significantly reduced apoB secretion. In summary, we hypothesize that fructose feeding induces hepatic insulin resistance at least in part via an increase in expression of PTP-1B. Induction of hepatic insulin resistance may then contribute to reduced apoB degradation and enhanced VLDL particle assembly and secretion.

Insulin resistance is an extremely common pathophysiological condition that is implicated in the development of a number of important human diseases including type 2 diabetes, atherosclerosis, hypertension, and dyslipidemia (1)(2)(3)(4)(5)(6)(7). The atherogenic dyslipidemia commonly associated with insulinresistant states consists of hypertriglyceridemia, high levels of very low density lipoproteins (VLDL), 1 low levels of high den-sity lipoprotein cholesterol (8), and elevated small, dense low density lipoprotein (9). It has been suggested that the most fundamental defect in these patients is resistance to the cellular actions of insulin, particularly resistance to insulin-stimulated glucose uptake, leading to hyperinsulinemia, enhanced VLDL secretion by the liver, and hypertriglyceridemia (10). Assembly and secretion of VLDL is a complex process involving the interaction of apolipoprotein B (apoB) with both core and surface lipids to form a lipoprotein particle (for reviews, see Refs. 11 and 12). An important determinant of apoB secretion appears to be intracellular stability of the protein, and a substantial amount of newly synthesized apoB is known to be subjected to intracellular degradation (for a review, see Ref. 13). Although the major mode of apoB degradation appears to involve the ubiquitin-proteasome system (14 -16), other proteases have been implicated in apoB degradation. One candidate protease is ER-60, a cysteine protease first purified from the lumen of the ER of rat hepatocytes (17)(18)(19)(20). We previously reported that in HepG2 cells, ER-60 is intracellularly associated with apoB and possibly involved in its degradation in the ER lumen (21).
Insulin may regulate VLDL secretion via multiple mechanisms including effects on apoB synthesis and degradation (reviewed in Ref. 22), modulation of MTP expression (23), apoB phosphorylation (24), and apoB mRNA editing (25)(26)(27). Early studies of apoB expression suggested that apoB is under regulatory control of insulin (22). Insulin has two profound effects on the synthesis and secretion of apoB. Acute incubation of rat hepatocytes (28 -34), isolated human hepatocytes (35), and HepG2 cells (36) suppressed apoB synthesis and secretion. Sparks et al. (37) also found increased sensitivity to the inhibitory action of insulin on apoB secretion in hepatocytes derived from partial hepatectomy rats compared with that of normal rats. The inhibitory effect of insulin was attributed to reduced synthesis and increased degradation of apoB in cells incubated with the hormone (24,37). Patsch et al. (30) showed that insulin exerted this effect through its receptor. Studies in our laboratory, using a cell-free system, have shown that insulin attenuates the rate of apoB mRNA translation (38,39). ApoB availability may thus be a limiting factor in VLDL assembly and secretion in insulin-treated hepatocytes (40). Furthermore, Sparks and co-workers (41) demonstrated that in rat hepatocytes, insulin-mediated inhibition of apoB was a phosphatidylinositol 3-kinase (PI 3-kinase)-dependent process. They also reported (42) that in rat hepatocytes, PI 3-kinase activity was necessary for insulin-dependent inhibition of apoB secretion and showed insulin-induced activation and localization of PI 3-kinase and insulin receptor substrate-1 (IRS-1) in an ER fraction containing apoB. Insulin can also affect the phosphorylation of apoB (24). Insulin may also exert its suppressive effect on VLDL secretion through regulating microsomal triglyceride transfer protein (MTP). The promoter region of the MTP gene has an insulin response element, which is negatively regulated by the hormone (23).
In contrast to acute effects, chronic exposure of primary rat hepatocytes (34) and HepG2 cells (43) to insulin increased apoB secretion. Inui et al. (44) reported that in obese rats, hepatic fatty acid synthesis and apoB transcription increased compared with lean rats. The same group later showed that in obese rats fed a high sucrose diet, apoB mRNA levels remained constant, while hepatic fatty acid synthesis and apoA-IV gene expression were elevated (44). Bourgeois et al. (45) and Sparks and Sparks (46) reported that hepatocytes prepared from obese Zuker rats were resistant to the inhibitory effects of insulin. Sparks et al. (32) also demonstrated that in cultured hepatocytes from diabetic rats, insulin failed to inhibit apoB secretion. Thus, hepatic insulin resistance may contribute to insensitivity of apoB secretion to the inhibitory actions of insulin.
There is a significant gap of knowledge concerning the molecular mechanisms that lead to deregulation of VLDL/apoB secretion in insulin-resistant states. We have recently studied the molecular mechanisms of VLDL overproduction in an animal model of diet-induced insulin resistance, the fructose-fed Syrian golden hamster (47). High fructose feeding induced whole body insulin resistance, which was associated with hyperinsulinemia, hepatic MTP overexpression, and enhanced intrahepatic stability and assembly of apoB containing lipoproteins, leading to VLDL oversecretion (47). In the current study, we have attempted to investigate the link between possible impairment of the insulin signaling pathway in the liver to VLDL-apoB overproduction in hepatocytes isolated from insulin-resistant hamsters. Using a combination of ex vivo and in vitro experiments, we provide evidence for significant downregulation of hepatic insulin signaling in the fructose-fed, insulin-resistant hamster model as documented by attenuated phosphorylation of the insulin receptor, IRS-1, and IRS-2; decreased activity of phosphotyrosine-associated PI 3-kinase; reduced phosphorylation of Akt/PKB; and overexpression and overactivity of PTP-1B. The potential link between down-regulation of hepatic insulin signaling and factors involved in the synthesis and secretion of VLDL-apoB including MTP and ER-60 protease was also explored.

MATERIALS AND METHODS
Male Syrian golden hamsters (Mesocricetus auratus) weighing 100 -120 g were purchased from Charles River (Montréal, Canada). Fetal bovine serum (certified grade), liver perfusion medium, hepatocyte wash medium, liver digest medium, hepatocyte attachment medium, and Williams' medium E were obtained from Life Technologies, Inc. All surgical disposable materials were obtained from Johnson & Johnson Medical Inc. (Arlington, TX). Rabbit anti-human insulin receptor ␤-subunit, rabbit anti-human IRS-1, and rabbit anti-p85 subunit of PI 3-kinase polyclonal antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Anti-IRS-2 rabbit polyclonal IgG was obtained from Upstate Biotechnology, Inc. (Lake Placid, NY). Mouse anti-PTP-1B monoclonal antibody (Ab-1) was obtained from Oncogene Research Products (Boston, MA). Rabbit anti-total Akt, anti-phospho-Akt (Ser 473 ), and anti-phospho-Akt (Thr 308 ) polyclonal antibodies were purchased from Cell Signaling Technology (Beverly, MA). Rabbit anti-hamster apoB antiserum was prepared by Lampire Biological Laboratories (Pipersville, PA) using hamster low density lipoprotein prepared in our laboratory. Specificity of this commercial preparation of anti-apoB polyclonal antibody and lack of any cross-reactivity to other hamster apolipoproteins (apoAI or apoE) was confirmed by immunblotting analysis of purified plasma lipoprotein fractions. Mouse anti-phosphotyrosine monoclonal antibody was obtained from Santa Cruz Biotechnology. pp60 c-src C-terminal phosphoregulatory peptide (TSTEPQpYQPGENL; where pY represents phosphotyrosine) and Biomol Green reagent were purchased from Biomol (Plymouth Meeting, PA).
Animal Protocols-All animals were housed individually and given free access to food and water. Normal chow was given for 2 days before blood samples were drawn and animals were placed on either the control diet (normal chow) or fructose-enriched diet (pelleted hamster diet containing 60% fructose; Dyets Inc., Bethlehem, PA). The diet was continued for 3 weeks, and hamster weight was monitored every 2 days.
Liver Perfusion and Isolation of Primary Hamster Hepatocytes-At the end of the 3-week feeding period, hamsters were fasted overnight, and blood samples were collected from the orbital sinus for measurement of a number of analytes in plasma. Hamsters were then fed for another day and were anesthetized by isoflurane. After achieving complete general anesthesia, the liver was perfused as described (48). Hepatocytes released from digested liver tissue were washed three times in hepatocyte wash medium and transferred into culture medium (hepatocyte attachment medium containing 5% FBS, 10 g/ml insulin) and seeded in collagen-coated plates (1.5 ϫ 10 6 cells/35-mm plate). After 4 h, medium was changed to Williams' medium E containing 5% FBS, 0.0015 g/ml insulin, and antibiotics, and these cells were used for experiments either immediately or after overnight incubation at 37°C, 5% CO 2 . In some experiments, the effects of chronic incubation with high insulin concentrations (1-10 g/ml) were studied using hepatocytes isolated from control hamsters.
Metabolic Labeling of Intact Primary Hamster Hepatocytes-Primary hamster hepatocytes were preincubated in methionine-free minimum essential medium at 37°C for 1 h and labeled with 100 Ci/ml [ 35 S]methionine for 45-60 min. Following the labeling pulse, the cells were washed twice and chased in Williams' medium E or hepatocyte attachment medium supplemented with 10 mM unlabeled methionine. At various chase times, duplicate or triplicate dishes were harvested, and cells were lysed in solubilization buffer (phosphate-buffered saline containing 1% Nonidet P40, 1% deoxycholate, 5 mM EDTA, 1 mM EGTA, 2 mM PMSF, 0.1 mM leupeptin, 2 g/ml N-acetyl-leucinyl-leucinyl-norleucinal). The lysates were centrifuged for 10 min at 4°C in a microcentrifuge (12,000 rpm), and supernatants were collected for immunoprecipitation.
For the vanadate dose-response studies, primary hepatocytes isolated from fructose-fed hamsters were incubated with culture medium containing 5% FBS, 1 g/ml insulin, and 0, 10, 20, 40, and 80 M vanadate for 6 h, and then culture medium was replaced with methionine-free ␣-minimum essential medium (with insulin and 0 -80 M vanadate) for 1 h and labeled with 100 Ci/ml [ 35 S]methionine for 90 min. Media and cell lysates were then subjected to apoB immunoprecipitation.
Evaluation of Tyrosine Phosphorylation Status of Insulin Signaling Cascade Proteins-In order to detect tyrosine phosphorylation of insulin receptor ␤-subunit, IRS-1, and IRS-2, hepatocytes derived from control and fructose-fed hamsters were incubated for 5 h in a serumand insulin-free medium. Half of the cells in each group were then stimulated with 100 nM insulin for 10 min at room temperature. Cells were then washed once with phosphate-buffered saline and lysed with a buffer containing a phosphatase inhibitor mixture (150 mM NaCl, 10 mM tris(hydroxymethyl)aminomethane (pH 7.4), 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 1% Nonidet P 40, 2 mM PMSF, 10 g/ml aprotinin, 10 g/ml leupeptin, 100 mM sodium fluoride, 10 mM sodium pyrophosphate, and 2 mM sodium orthovanadate) and subjected to immunoprecipitation with a specific polyclonal antibody against either insulin receptor ␤-subunit, IRS-1, or IRS-2 (1 g of antibody/0.5 mg of total cell lysate) using 50 l of 10% protein A-Sepharose (for each sample). Immunoprecipitates were then washed three times at 4°C using wash buffer (phosphate-buffered saline containing 100 mM sodium fluoride, 10 mM sodium pyrophosphate, 2 mM sodium orthovanadate, 0.1% Nonidet P 40, and 0.1% Triton X-100). Immunoprecipitates were used for immunoblotting with monoclonal antibody ␣pY (1:1000 dilution) using enhanced chemiluminescence as described below.
PI 3-Kinase Activity Assays-The PI 3-kinase assays were performed as described (49,50). Briefly, hepatocytes isolated from control and fructose-fed hamsters were incubated in serum-and insulin-free medium for 5 h and then exposed to 100 nM insulin for 10 min at room temperature. Cells were washed and lysed in solubilizing buffer con-taining 150 mM NaCl, 10 mM tris(hydroxymethyl)aminomethane (pH 7.4), 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 1% Nonidet P-40, protease inhibitors (2 mM PMSF, 10 g/ml leupeptin, and 10 g/ml aprotinin) and phosphatase inhibitors (100 mM sodium fluoride, 10 mM sodium pyrophosphate, and 2 mM sodium orthovanadate). Cell lysates were subjected to overnight immunoprecipitation with anti-p85 or antiphosphotyrosine antibodies at 4°C. PI 3-kinase activity was measured on p85 or phosphotyrosine immunoprecipitates. All immunoprecipitates were washed and incubated for 5 min with 20 g of phosphatidylinositol. The reaction was initiated by the addition of 5 l of [␥-32 P]ATP (1 mM final concentration, 10 Ci/reaction) in 20 mM MgCl 2 and then stopped after 10 min by the addition of 8 M HCl. The lipids were extracted with 160 l of chloroform/methanol (1:1). 50 l of the lower phase was applied to a silica gel 60 F 254 thin layer chromatography plastic sheet (Merck), and lipids were separated in a chloroform/methanol/water/ammonium hydroxide (60:47:12:2) solvent system. Radiolabeled lipids were quantitated using a Storm 840 PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA).
PTP-1B Activity Assay-The in vitro PTP activity assay was conducted based on a protocol previously published by Cho et al. (51) with some modifications. Hepatocytes isolated from control and fructose-fed hamsters were lysed in solubilization buffer (phosphate-buffered saline containing 1% Nonidet P-40, 1% deoxycholate, 5 mM EDTA, 1 mM EGTA, 2 mM PMSF, 0.1 mM leupeptin, 2 g/ml N-acetyl-leucinyl-leucinyl-norleucinal). The lysates were centrifuged for 10 min at 4°C in a microcentrifuge, and supernatants were collected for immunoprecipitation. Prior to immunoprecipitation, cell lysates were subjected to preclearing with nonimmune serum and protein A-Sepharose for 15 min at 4°C. Equal quantities of each sample (750 g of total protein) were then subjected to immunoprecipitation with anti-PTP-1B antibody (Ab-1; Oncogene Research Products) at 4°C overnight. PTP-1B immunocomplexes were precipitated with protein A-Sepharose at 4°C for an additional 2 h. Immunoprecipitates were washed in PTP assay buffer (100 mM HEPES (pH 7.6) 2 mM EDTA, 1 mM dithiothreitol, 150 mM NaCl, 0.5 mg/ml bovine serum albumin). The pp60 c-src C-terminal phosphoregulatory peptide (TSTEPQpYQPGENL; Biomol) was added to a final concentration of 200 M in a total reaction volume of 60 l in PTP assay buffer, and the reaction was allowed to proceed for 1 h at 30°C. At the end of the reaction, 40-l aliquots were placed into a 96-well plate, 100 l of Biomol Green reagent (Biomol) was added, and absorbance was measured at 630 nm.
Chemiluminescent Immunoblot Analysis-Cell samples either directly or after immunoprecipitation against a target protein were subjected to chemiluminescent immunoblotting for the insulin receptor ␤-subunit, IRS-1, IRS-2, and Akt/PKB, their phosphorylated forms, and other proteins such as ER-60 and PTP-1B. Samples were analyzed by SDS-PAGE using 8 or 10% polyacrylamide minigels (8 ϫ 5 cm). Following SDS-PAGE, the proteins were transferred electrophoretically overnight at 4°C onto polyvinylidene difluoride membranes using a Bio-Rad Wet Transfer System. The membranes were blocked with a 5% solution of fat-free dry milk powder, incubated with relevant antiserum, washed, and then incubated with a secondary antibody conjugated to peroxidase. Membranes were then incubated in an enhanced chemiluminescence detection reagent (Amersham Pharmacia Biotech) for 60 s and exposed to Eastman Kodak Co. Hyperfilm. Films were developed, and quantitative analysis was performed using an Imaging Densitometer.
Immunoprecipitation, SDS-PAGE, and Fluorography-Immunoprecipitation was performed as described previously (48). Immunoprecipitates were washed with wash buffer (10 mM Tris-HCl (pH 7.4), 2 mM EDTA, 0.1% SDS, 1% Triton X-100) and prepared for SDS-PAGE by resuspension and boiling in 100 l of electrophoresis sample buffer. SDS-PAGE was performed essentially as described (52). The gels were fixed and saturated with Amplify (Amersham Pharmacia Biotech) before being dried and exposed to Dupont autoradiographic film at Ϫ80°C for 1-4 days. ApoB bands were excised from the gel, digested in hydrogen peroxide/perchloric acid, and associated radioactivity was quantitated by liquid scintillation counting.

Overproduction of VLDL-apoB in Hepatocytes Isolated from
Fructose-fed Hamsters-To establish whether VLDL-apoB overproduction was induced by fructose feeding in the present study, we performed ex vivo steady state labeling experiments in which hepatocytes from control and fructose-fed hamsters were radiolabeled for a 2-h period. Radiolabeled apoB associated with VLDL particles secreted into the culture medium was immunoprecipitated and analyzed by SDS-PAGE and fluorography. Fig. 1 shows the immunoprecipitable VLDL-apoB secreted by control and fructose-fed hepatocytes. There was a highly significant (ϳ3.6-fold) elevation in the amount of VLDL-apoB secreted into the medium by hepatocytes derived from fructose-fed hamsters. Increased VLDL-apoB levels suggest the secretion of a considerably higher number of VLDL particles by hepatocytes isolated from fructose-fed hamsters compared with control hepatocytes and confirm our previous observation (47).
Fructose-fed hamsters exhibiting VLDL-apoB overproduction also showed evidence of whole body insulin resistance based on elevated plasma levels of insulin, triglyceride, and free fatty acids (data not shown). Although euglycemic-hyperinsulinemic clamp studies were not performed in the animals used in the present study, this technique was used previously to document the induction of whole body insulin resistance in the fructose-fed hamster model (47).

Effect of Fructose Feeding on the Phosphorylation Status and Protein Mass of the Insulin Receptor, IRS-1, and IRS-2 in
Hamster Hepatocytes-Phosphorylation status of the insulin receptor and its substrate proteins was assessed in hepatocytes derived from control and fructose-fed hamsters. As depicted in Fig. 2A, basal insulin receptor phosphorylation (in the absence of insulin) in hepatocytes isolated from control and fructose-fed hamsters was too weak to be detected by immunoprecipitation and immunoblotting methods. However, in control hepatocytes, insulin-induced phosphorylation of its receptor was about 2-fold higher (n ϭ 5, p ϭ 0.001) than that induced in hepatocytes isolated from fructose-fed hamsters. To examine endogenous substrate phosphorylation, immunoprecipitated IRS-1 was subjected to Western blotting with an anti-phosphotyrosine antibody as described under "Materials and Methods." Densitometric analysis (Fig. 2B) revealed that in hepatocytes isolated from fructose-fed hamsters, basal and insulin-stimulated phosphorylation of IRS-1 were lower by more than 5-fold (n ϭ 3, p ϭ 0.01) and 11-fold (n ϭ 3, p ϭ 0.009), respectively, compared with that in control hepatocytes. Moreover, in control hepatocytes, insulin increased IRS-1 phosphorylation approximately 2-fold (199.7 Ϯ 34.5% of basal, n ϭ 3, p ϭ 0.04), whereas, in hepatocytes isolated from fructose-fed hamsters, insulin failed to induce phosphorylation of IRS-1 (91.35 Ϯ 12.2% of basal; n ϭ 3, p ϭ 0.073). We also examined the effects of fructose feeding on the protein mass and phosphorylation level of IRS-2. As depicted in Fig. 2C, in control hepatocytes, stimulation with 100 nM insulin caused a 54.5 Ϯ 3.4% (n ϭ 3, p ϭ 0.023) increase in tyrosine phosphorylation of IRS-2 compared with its basal level, whereas in hepatocytes isolated from fructose-fed hamsters, insulin increased phosphorylation only by 20 Ϯ 7.4% (n ϭ 3, p ϭ 0.029) of basal level, suggesting a significant impairment in IRS-2-mediated insulin signal transduction following fructose feeding. Reduced IRS-1 and IRS-2 phosphorylation in hepatocytes from fructose-fed hamsters is consistent with the data on insulin receptor tyrosine phosphorylation and further supports the induction of hepatic insulin resistance in this model.
To investigate whether down-regulation of insulin receptor, IRS-1, and IRS-2 phosphorylation in hepatocytes isolated from fructose-fed hamsters was related to changes in the intracellular mass of these signaling molecules, we examined cellular levels of insulin receptor, IRS-1, and IRS-2 by immunoblotting. As shown in Fig. 2D, in hepatocytes isolated from fructose-fed hamsters, the mass of insulin receptor was 92.4 Ϯ 12% of that in control hepatocytes, suggesting no significant change in receptor protein mass with fructose feeding. Fig. 2E shows protein mass levels of IRS-1. IRS-1 protein levels in hepatocytes isolated from fructose-fed hamsters were 31.8 Ϯ 1.0% (n ϭ 3, p ϭ 0.01) of that in control hepatocytes. As shown in Fig.  2F, IRS-2 protein mass in insulin-resistant hepatocytes was also dramatically reduced to 57.8 Ϯ 7.1% (n ϭ 4, p Ͻ 0.001) of that in control hepatocytes.
PI 3-Kinase Activity in Hepatocytes Isolated from Control and Fructose-fed Hamsters-PI 3-kinase activity was assessed in two ways. Total activity was measured in immunoprecipitates generated using an antibody specific to the p85 subunit of PI 3-kinase. Fig. 3A demonstrates total PI 3-kinase activity (normalized to total protein) as a percentage of the activity in the control hepatocytes. There was no significant difference in total PI 3-kinase activity between cells derived from control and fructose-fed livers. The second assay was designed to assess activity associated with insulin receptor substrates and therefore available for involvement in insulin signal transduction. These experiments were performed on cell lysates immunoprecipitated with an anti-phosphotyrosine antibody. As shown in Fig. 3B, PI 3-kinase activity associated with tyrosinephosphorylated proteins was about 25% lower in hepatocytes isolated from fructose-fed hamsters (74.7 Ϯ 3.3% of control, n ϭ 3, p ϭ 0.03).
Evidence That Intracellular Level and Activity of PTP-1B Are Enhanced in Hepatocytes Isolated from Fructose-fed Hamsters-Protein-tyrosine phosphatases, particularly PTP-1B, play an important role in regulating the phosphorylation status of proteins involved in insulin signaling (for a recent review, see Ref. 53). To investigate the possible role of PTP-1B in the impairment of signal transduction in liver of fructose-fed hamsters, experiments were conducted using a specific anti-PTP-1B polyclonal antibody. Fig. 3C shows a representative Western blot for PTP-1B and the quantification of data obtained from three independent experiments. These experiments revealed that PTP-1B protein levels in hepatocytes isolated from fructose-fed hamsters were significantly higher (147.03 Ϯ 22.6%, mean Ϯ S.D., p ϭ 0.004, n ϭ 3) compared with that of control hepatocytes.
We also conducted experiments to measure PTP-1B activity in hepatocytes isolated from control and fructose-fed hamsters. PTP-1B activity in hepatocytes isolated from fructose-fed hamsters was significantly increased by almost 2-fold (193 Ϯ 51.9%) compared with that in control hepatocytes (n ϭ 8, p ϭ 0.0021) ( Fig. 3D). These results parallel the above observation of an elevated protein mass of PTP-1B and together suggest enhanced expression and activity of this phosphatase in liver of fructose-fed hamster.
Impaired Akt/PKB Serine and Threonine Phosphorylation in Hepatocytes Isolated from Fructose-fed Hamsters-In order to investigate insulin signaling status downstream of PI 3-kinase, we examined the phosphorylation status of serine 473 and threonine 308 of Akt/PKB, a key serine/threonine kinase, which mediates many metabolic effects of insulin. Fig. 3, E and F, show serine and threonine phosphorylation of Akt/PKB in hepatocytes isolated from control and fructose-fed hamsters. Fructose feeding reduced insulin-stimulated phosphorylation levels of serine 473 and threonine 308 to 29 Ϯ 15% (n ϭ 4, p ϭ 0.002) and 42.6 Ϯ 20.8% (n ϭ 5, p ϭ 0.001) of that of control hepatocytes, respectively, indicating that phosphorylation, and therefore activity, of Akt/PKB were significantly compromised with fructose feeding. Immunoblotting for Akt mass (Fig. 3G) showed a small but significant increase (39.3 Ϯ 1%, n ϭ 3, p ϭ 0.03) in Akt/PKB protein expression levels in hepatocytes isolated from fructose-fed hamsters, suggesting a possible compensatory response to the suppressed phosphorylation status of the protein.

Insulin Signaling Status in Hamster Hepatocytes Exposed to
High Insulin Levels ex Vivo-Fructose-fed hamsters were previously shown to have an elevated plasma insulin level (47), apparently due to peripheral insulin resistance as documented by euglycemic-hyperinsulinemic clamps. Such hyperinsulinemia may in turn be responsible for the down-regulation of insulin signaling in the liver and the induction of hepatic insulin resistance observed in experiments above. To further examine this hypothesis, we directly incubated control hepatocytes with high insulin (1.0 g/ml) for up to 3 days and measured the basal phosphorylation level of the insulin receptor following each day of incubation. Fig. 4A shows a representative immunoblot of the tyrosine-phosphorylated insulin receptor following various periods of insulin exposure. For the purpose of these experiments, the basal level of insulin receptor phosphorylation was taken to be that measured in freshly isolated, and therefore untreated, hepatocytes (day 0). In cells incubated with high insulin, insulin receptor phosphorylation initially decreased from day 0 to day 1 and then increased from day 1 to day 2 but was substantially reduced by day 3 (856 Ϯ 24, 520 Ϯ 9, 852 Ϯ 48, and 285 Ϯ 69 arbitrary densitometric units/mg of total cell protein ϫ 10 Ϫ3 on days 0, 1, 2, and 3 of incubation, respectively). As depicted in Fig. 4A, following 3

FIG. 3. Fructose feeding reduces hepatic PI 3-kinase and Akt/PKB phosphorylation and increases the protein mass and activity of PTP-1B.
A and B, cell lysates were subjected to immunoprecipitation with anti-p85 or anti-phosphotyrosine antibodies. PI 3-kinase activity was measured on p85 (A) or phosphotyrosine immunoprecipitates (B), as described under "Materials and Methods." The PI 3-kinase activity was normalized for total protein content of the samples and expressed as a percentage of activity detected in control hepatocytes. Data represent mean Ϯ S.D., n ϭ 3. *, significant difference (p ϭ 0.03). C, hepatocytes isolated from control and fructose-fed hamsters were solubilized and subjected to SDS-PAGE and immunoblotting for PTP-1B as described under "Materials and Methods." Corresponding bands were quantified by densitometry and expressed as a percentage of PTP-1B in control hamster hepatocytes. Shown is one representative immunoblot and combined quantitation of immunoblots from three experiments performed in triplicate. Data represent mean Ϯ S.D. **, significant difference (p ϭ 0.004). D, hepatocytes isolated from control and fructose-fed hamsters were lysed in solubilization buffer. The lysates were centrifuged, and supernatants were collected for immunoprecipitation with anti-PTP-1B antibody. PTP-1B immunocomplexes were used to measure phosphatase activity using the pp60 c-src C-terminal phosphoregulatory peptide (TSTEPQpYQPGENL; Biomol) as substrate. PTP-1B activity was expressed relative to the activity detected in control hepatocytes. Activity was measured in five control and eight fructose-fed hamster hepatocyte preparations, and data are shown as mean Ϯ S.D. ***, significant difference (p ϭ 0.0021). E and F, lysates from basal and insulin-stimulated cells were immunoblotted using polyclonal antibodies against phospho-Ser 473 -Akt (E) or phospho-Thr 308 -Akt (F). Representative immunoblots are shown along with densitometric quantification. Net intensity of the bands was normalized for the total protein content of the samples. Data were collected from four experiments performed in duplicate. All data are shown as mean Ϯ S.D. (*, p ϭ 0.002 for E; *, p ϭ 0.001 for F). G, immunoblot analysis of total Akt mass. A representative immunoblot is shown with combined quantitated densitometry data from three experiments performed in duplicate or triplicate. Mean Ϯ S.D.; *, p ϭ 0.03. days of high insulin exposure phosphorylation of the insulin receptor decreased to 33.3 Ϯ 8.1% of control, suggesting desensitization of the receptor at high insulin concentrations.
Since the observed differences in insulin receptor phosphorylation status may be attributed to possible changes in the expression level of the insulin receptor, we also measured insulin receptor protein levels under the above experimental condition. Phosphotyrosine immunoblots were stripped and reprobed with a polyclonal antibody against the insulin receptor ␤-subunit as described under "Materials and Methods" (Fig.  4B). Insulin receptor band intensity under the control condition (day 0) was measured as 1136 Ϯ 16 arbitrary units/mg of protein. High insulin incubation resulted in a slight but consistent reduction in the receptor protein mass, suggesting possible down-regulation of the receptor. The insulin receptor protein mass was significantly decreased on the third day of incubation and reached 56.8 Ϯ 14.1% (p ϭ 0.025) of the basal level (day 0).
An arbitrary index (specific insulin receptor phosphorylation index (SIRPI)) was also calculated to better compare the phosphorylation status of the insulin receptor under control and high insulin exposed conditions. The index is calculated as the ratio of phosphorylated insulin receptor to the total insulin receptor protein mass, as measured by densitometric quantifi-cation of Western blots, normalized for total cellular protein content. Fig. 4C reevaluates the data presented above in terms of the SIRPI. The SIRPI value for untreated, day 0 cells was 0.75 Ϯ 0.01. After exposure to high insulin concentrations, the index remained constant and approximately equal to control (0.62 Ϯ 0.02 and 1.01 Ϯ 0.24 on days 1 and 2 of incubation, respectively); however, it dropped significantly to 0.4 Ϯ 0.003 (p ϭ 0.01) at the end of 3 days of incubation, suggesting the induction of insulin resistance.
Chronic High Insulin Exposure Induces PTP-1B Expression in Hamster Hepatocytes-In order to investigate whether changes in the insulin signaling pathway may be related to alterations in expression of PTP-1B in hamster hepatocytes, we investigated protein levels of PTP-1B in control hamster hepatocytes incubated with 1 g/ml insulin for 3 days. As shown in Fig. 4D, immunoblotting of equal amounts of cell lysate (10 g of total cell protein) revealed that PTP-1B levels in hepatocytes incubated with high insulin gradually increased from day 0 (day of hepatocyte preparation), reached a plateau on day 2, and remained constant on day 3 of incubation. Cellular levels of PTP-1B on days 1, 2, and 3 were 108.3 Ϯ 0.15%, 138.3 Ϯ 2.76% (p ϭ 0.003), and 134.8 Ϯ 4.33% (p ϭ 0.008) of the basal level (day 0), respectively. Thus, elevation in cellular levels of PTP-1B occurred earlier on day 2 and appeared to precede the FIG. 4. Hepatic insulin signaling status in hepatocytes chronically exposed to high insulin concentrations. Primary hepatocytes isolated from control (chow-fed) hamsters were incubated in culture medium containing high insulin (1.0 g/ml) and 5% FBS for up to 3 days, and the phosphorylation status and mass of insulin receptor were assessed on a daily basis. Hepatocytes were solubilized and immunoprecipitated for insulin receptor ␤-subunit and immunoprecipitates were then subjected to immunoblotting with a monoclonal antibody against phosphotyrosine. The same membrane was then stripped of antibody and reprobed for insulin receptor mass using a polyclonal antibody against the insulin receptor ␤-subunit. A, representative immunoblot for tyrosine-phosphorylated insulin receptor and its quantification. Data are shown as mean Ϯ S.D. ***, significant difference (p ϭ 0.001). B, representative immunoblot for insulin receptor ␤-subunit and its quantification (per mg of protein). Data are shown as mean Ϯ S.D. * and **, significant differences (p ϭ 0.045 and 0.025, respectively). C, SIRPI, an arbitrary parameter used to assess the phosphorylation status of the insulin receptor, taking into account variability in insulin receptor protein mass levels. The value consists of the ratio of phosphorylated insulin receptor (measured as described above) to total insulin receptor mass, as measured by densitometric quantitation of Western blots, normalized for total cellular protein mass. SIRPI is a dimensionless arbitrary parameter. Data are shown as mean Ϯ S.D. **, significant difference (p ϭ 0.01). D, equal amounts of cell lysate were subjected to SDS-PAGE (8%) and immunoblotting for PTP-1B as described under "Materials and Methods." Corresponding bands were scanned and quantified. This figure shows a representative immunoblot for PTP-1B and densitometric quantification of immunoblots from three experiments. Data are shown as mean Ϯ S.D. ***, significant difference compared with day 0 (p ϭ 0.003).
down-regulation of insulin receptor phosphorylation in hamster hepatocytes on day 3 (as shown above).
Chronic Exposure of Hepatocytes to High Insulin Induces ApoB Oversecretion-In order to assess the effect of long term high insulin incubation of control hamster hepatocytes on apoB biogenesis, hepatocytes were incubated with culture medium supplemented with 5% FBS and 1 g/ml insulin for up to 3 days. After each day of treatment, cells were pulsed with [ 35 S]methionine, and apoB secreted into the medium was analyzed by immunoprecipitation, SDS-PAGE, and fluorography. As depicted in Fig. 5A, secreted apoB was slightly increased on day 1 and then remained unchanged on day 2 of exposure. The apparent elevation in secreted apoB from day 0 to days 1 and 2 was not statistically significant (p ϭ 0.22). On the third day of incubation, secreted apoB was significantly increased by ϳ2.5fold (p ϭ 0.04) over the basal level (day 0). Importantly, total protein synthesis did not change significantly during the incubation period (8,474,400 Ϯ 871,500, 6,431,700 Ϯ 1,216,500,  9,685,200 Ϯ 1,297,500, 8,779,200 Ϯ 933,600 cpm/dish on days 0, 1, 2, and 3, respectively). Thus, long term exposure to high insulin levels appeared to cause a significant hypersecretion of apoB, which coincided with desensitization of the cells to insulin and down-regulation of the insulin signaling pathway (as above).
ER-60 Protein Mass in Hepatocytes Isolated from Fructosefed Hamsters-We have previously shown that ER-60, an ER lumen-localized cysteine protease, is associated with apoB in-tracellularly and may be involved in apoB degradation (21). In order to investigate possible changes in expression of ER60 following fructose feeding, we conducted immunoblotting experiments using a rabbit polyclonal antibody against rat ER-60 that readily cross-reacts with hamster ER-60. As depicted in Fig. 5B, ER-60 protein levels in hepatocytes isolated from fructose-fed hamsters were 14.7 Ϯ 6.8% (n ϭ 8; p Ͻ 0.001) of that in control hepatocytes, indicating marked down-regulation of ER-60 in insulin-resistant hamster livers. In order to determine whether the suppression of ER-60 expression was a result of an impairment of insulin signal transduction in fructose-fed hamsters, we repeated the above experiments using hepatocytes isolated from fructose-fed hamsters treated with rosiglitazone, an insulin-sensitizing drug (20 mol/kg body of weight daily for 3 weeks). As depicted in Fig. 5B, rosiglitazone significantly increased ER-60 protein levels relative to untreated fructose-fed animals, although it could not restore the ER-60 levels to the level observed in control hepatocytes. Control experiments showed that rosiglitazone treatment was also capable of enhancing insulin receptor phosphorylation in hamster livers (data not shown).
ER-60 Suppression Can Be Induced by High Insulin Exposure ex Vivo-To investigate whether ER-60 protein levels change in response to insulin exposure, we incubated control hamster hepatocytes with a high concentration of insulin for up to 3 days and monitored expression levels of ER-60. As shown in Fig. 5C, ER-60 protein levels increased from 100 Ϯ 14.4% on FIG. 5. Effect of chronic high insulin exposure of hamster hepatocytes on apoB secretion and ER-60 protein mass. A, control hamster hepatocytes were incubated with a high concentration of insulin (1 g/ml) for up to 3 days. Cells from each day of incubation were pulsed for 90 min with [ 35 S]methionine (100 Ci/ml) as described under "Materials and Methods." Culture medium was collected and subjected to immunoprecipitation using an anti-hamster apoB antibody. Immunoprecipitates were subjected to SDS-PAGE and fluorography. ApoB bands were excised, and their radioactivity was quantified by scintillation counting and expressed as labeled apoB/total labeled cellular protein. Shown is a representative gel and quantification of apoB bands. Data are presented as mean Ϯ S.D. *, significantly different from day 0 (p ϭ 0.04). B, equal amounts of hepatocyte cell lysates from control, fructose-fed, and fructose-fed/rosiglitazone-treated animals were subjected to SDS-PAGE (8%) and immunoblotted for ER-60 as described under "Materials and Methods." Corresponding bands were scanned and quantified. Shown is a representative immunoblot for ER-60 and densitometric quantification of immunoblots from four experiments performed in duplicate or triplicate, expressed as mean Ϯ S.D. * and **, significant differences (p ϭ 0.004 and 0.04, respectively). C, control hamster hepatocytes treated for 0 -3 days with 1 g/ml insulin as in A were analyzed for ER-60 protein mass at each time point. A representative immunoblot is shown along with combined data from three experiments. Data represent mean Ϯ S.D. *, significant difference (p Ͻ 0.008). D, control hamster hepatocytes treated for 0 -3 days with 1 g/ml insulin as in A were analyzed for a control ER-resident protein (ERp72) using a monospecific rabbit anti-mouse ERp72 polyclonal antibody (Stressgen) at each time point. A representative immunoblot is shown (repeated twice). day 0 to 149.17 Ϯ 13.27% on day 1 (p ϭ 0.01) and remained constant on day 2 (153.72 Ϯ 8.18%; p Ͻ 0.01 compared with basal). Interestingly, on day 3 of incubation, ER-60 protein mass dramatically decreased to 18.39 Ϯ 1.9% (p ϭ 0.007) of basal (day 0), concomitant with increased PTP-1B levels and induction of insulin resistance (as shown above), suggesting an association between the induction of insulin resistance and ER-60 suppression. As a control, we also probed for another ER-resident protein, ERp72, using a rabbit anti-mouse ERp72 polyclonal antibody. As shown in Fig. 5D, there were no appreciable changes in the protein mass of ERp72 over the 3 days of insulin incubation, arguing against a global change in ER protein mass under these conditions.
Vanadate Improves Tyrosine Phosphorylation of Insulin Receptor in Hepatocytes Isolated from Fructose-fed Hamsters in a Dose-dependent Manner-Sodium vanadate is a known phosphatase inhibitor and insulin-mimetic agent that improves insulin signal transduction. We investigated whether treatment of hepatocytes from fructose-fed hamsters with vanadate can improve insulin signaling and reduce apoB secretion. Hepatocytes isolated from fructose-fed hamsters were incubated in serum-and insulin-free Williams' medium E containing 0, 10, 40, and 80 M activated sodium orthovanadate for 6 h and then subjected to 100 nM insulin stimulation for 10 min. Equal amounts of cell lysates were immunoprecipitated for insulin receptor ␤ subunit and then immunoblotted with anti-phosphotyrosine antibody as described under "Materials and Methods." Fig. 6A depicts the dose-response curve of the effect of vanadate on insulin receptor phosphorylation and a representative im-munoblot. Exposure to vanadate increased tyrosine phosphorylation of the insulin receptor from 100 Ϯ 6.4% in untreated cells to 365.2 Ϯ 20.5% (p ϭ 0.0009), 474.3 Ϯ 9.9%, and 488.4 Ϯ 19%, at 10, 40, and 80 M vanadate, respectively. These results indicate that vanadate increased phosphorylation of the insulin receptor in a dose-dependent manner, reaching a plateau at 40 M. Membranes were later stripped and reprobed for insulin receptor mass, which showed no significant change (data not shown).

Vanadate Reduces Synthesis and Secretion of ApoB from Insulin-resistant Hepatocytes in a Dose-dependent Manner-To
investigate whether the vanadate-induced enhancement in insulin receptor phosphorylation influences the synthesis and secretion of apoB, we incubated hepatocytes isolated from fructose-fed hamsters with culture medium containing 0, 10, 20, 40, and 80 M vanadate for 6 h. Cells were then pulsed with [ 35 S]methionine for 90 min. Vanadate (0 -80 M) and insulin (1 g/ml) were present throughout the experiments. Cellular, secreted, and total (secreted plus cellular) apoB, are shown in Fig. 6, B-D. Exposure of the cells to vanadate reduced cellular apoB from 100 Ϯ 7% at 0 M to 97.6 Ϯ 14.1% (p ϭ 0.788), 92.9 Ϯ 9.3% (p ϭ 0.26), 75.9 Ϯ 13.2% (p ϭ 0.050), and 62.8 Ϯ 7.1% (p ϭ 0.003) at 10, 20, 40, and 80 M, respectively, indicating that vanadate did not affect cellular apoB significantly until the dose of 40 M, which caused about a 25% suppression of cellular apoB (which increased to 37.2% at 80 M). However, vanadate had a more significant suppressive effect on apoB secretion at lower doses such that at 10 M vanadate reduced apoB secretion to 52 Ϯ 19% (p ϭ 0.015) of control. ApoB secretion was FIG. 6. Vanadate treatment of fructose-fed hamster hepatocytes stimulates tyrosine-phosphorylation of the insulin receptor and reduces apoB secretion. A, hepatocytes isolated from insulin-resistant fructose-fed hamsters were incubated with serum-and insulin-free Williams' medium E containing 0, 10, 40, and 80 M activated vanadate for 6 h, and cells were then subjected to 100 nM insulin stimulation for 10 min. Equal amounts of cell lysates (0.5-1 mg) were immunoprecipitated for the insulin receptor ␤-subunit and then immunoblotted with an anti-phosphotyrosine antibody, as described under "Materials and Methods." Corresponding bands were quantified by densitometry and normalized for total protein. One representative immunoblot is shown along with densitometric quantification. Data were collected from three experiments performed in duplicate. All data are shown as mean Ϯ S.D. *, significantly different from control (p Ͻ 0.01). B-D, tissue culture dishes (60 mm) containing 3 ϫ 10 6 hepatocytes isolated from fructose-fed hamsters were incubated with vanadate for 6 h as above and pulsed with [ 35 S]methionine for 90 min. Vanadate (0 -80 M) and insulin (1 g/ml) were present throughout the experiment. Cell lysates and media were collected and subjected to immunoprecipitation for apoB and fluorography. Quantitated labeled apoB was normalized against total radiolabeled protein (counts/min). *, significantly different from control (p Ͻ 0.05). B, a representative autoradiogram of cellular apoB is shown as well as combined quantitated data for three experiments performed in duplicate. C, apoB detected in the medium. D, sum of cellular and secreted apoB (total apoB). All data are shown as mean Ϯ S.D. *, significantly different from control (p Ͻ 0.01).
There are a few studies on the molecular mechanisms that mediate inhibition of apoB secretion following acute insulin exposure; however, the role of chronic hyperinsulinemia and insulin resistance in VLDL overproduction has been understudied. We recently investigated the cellular and molecular mechanisms involved in apoB overproduction in an animal model of insulin resistance, the fructose-fed hamster (47). In the present study, ex vivo and in vitro experiments were performed to establish whether hepatic insulin resistance is a component of the metabolic syndrome observed in the fructosefed hamster model and to examine how changes in the insulin signaling pathway in the liver may impact hepatic apoB secretion.
It has previously been documented that fructose feeding in rodents, including hamsters (54), results in chronic hyperinsulinemia, an insulin-resistant state, and hyperlipidemia. The fructose protocol employed in the current study is similar to those previously shown to induce insulin resistance in rats (55)(56)(57)(58) and hamsters (47). We have previously documented the induction of whole body insulin resistance in this model following fructose feeding (47). Induction of insulin resistance at the level of adipocytes and muscle tissues has also been well documented by others (for reviews, see Refs. 5, 7, and 59), whereas the induction of hepatic insulin resistance is less well established. It was important to determine whether resistance to insulin action develops in livers of fructose-fed hamsters and whether hepatic insulin resistance plays a direct role in deregulation of VLDL-apoB secretion. In order to document the induction of hepatic insulin resistance, we quantified the tyrosine phosphorylation status of the insulin receptor, IRS-1, and IRS-2 ex vivo in control and fructose-fed hamsters. Data obtained from these experiments showed a significant reduction in the phosphorylation of these key proteins of the insulin signaling pathway in hepatocytes of fructose-fed hamsters, suggesting down-regulation of insulin signaling in the liver. Our observations on insulin receptor and IRS-1 phosphorylation in the liver are in agreement with Bezerra et al. (60), who reported a reduction in IRS-1 and insulin receptor phosphorylation in muscle and livers of high fructose-fed rats. Our data are also in agreement with the findings of Jiang et al. (61), who reported significant reductions in phosphorylation and protein mass of both IRS-1 and IRS-2 in the aorta and microvessels of obese hyperinsulinemic Zucker (fa/fa) rats. Similar results have been reported in Zucker fatty rats (62) and ob/ob mice (63).
Fructose feeding of hamsters also appeared to cause suppressed insulin signaling downstream of IRS-1/IRS-2. Thus, we found that PI 3-kinase activity associated with tyrosine-phosphorylated proteins was decreased in fructose-fed insulin-resistant hepatocytes, while p85-associated activity remained unchanged. These observations indicate that, although intrinsic activity of the enzyme was intact, recruitment to the insulin signaling pathway was decreased, most likely as a result of the reduced phosphorylation of upstream signaling components. We also examined the impact of fructose feeding on phosphorylation of Ser 473 and Thr 308 of Akt, a key enzyme downstream of PI 3-kinase. In hepatocytes isolated from fructose-fed hamsters, insulin-stimulated phosphorylation of both Ser 473 and Thr 308 was significantly reduced, suggesting suppressed activity of Akt in fructose-fed hamster hepatocytes, most likely due to the observed decrease in tyrosine phosphorylation of upstream insulin signaling proteins and reduced PI 3-kinase activity. Our observations are in agreement with the findings of Krook et al. (64) and Rondinone et al. (65) who reported impaired Akt phosphorylation in muscle and adipocytes of insulin resistant diabetic subjects, respectively. Several other groups have reported similar observations in skeletal muscle of insulin resistant rats (66,67) and muscle and adipose tissues of db/db mice (68). However, the effect on hepatic Akt/PKB phosphorylation status was not investigated in these previous studies. Our findings appear to be the first report of suppressed Akt/ PKB phosphorylation in a carbohydrate-induced model of insulin resistance.
We also investigated whether the suppression of insulin signaling in hepatocytes could be related to the hyperinsulinemia observed in fructose-fed hamsters. To assess the chronic effect of hyperinsulinemia, we incubated control hepatocytes with high insulin levels for up to 3 days and monitored changes in key components of the insulin signaling pathway. We found that exposure to high insulin initially reduced receptor expression and eventually caused a significant down-regulation of insulin receptor phosphorylation. These data appear to suggest that hepatic insulin resistance in fructose-fed hamster may be secondary to elevated plasma insulin levels. In support of this hypothesis, Yoshino et al. (69) showed that in streptozotocintreated, diabetic, hypoinsulinemic rats, fructose feeding increased plasma insulin levels. More recently, Suga et al. (70) reported that dietary fructose caused substantial insulin resistance and hyperinsulinemia in both ventromedial hypothalamic-lesioned obese and sham-operated lean rats. Dirlewanger et al. (71) also showed that in healthy humans, fructose infusion induced hepatic and extrahepatic insulin resistance.
Since protein-tyrosine phosphatases play a crucial role in the insulin signaling pathway as negative regulators of signal transduction, we examined both protein levels and activity of PTP-1B in the current model and found significant elevations in the cellular mass and activity of this protein in hepatocytes isolated from fructose-fed hamsters. Our findings in fructosefed hamsters are in agreement with previous reports in livers of obese, diabetic ob/ob mice (72), streptozotocin-induced diabetic rats, and genetically diabetic BB rats (73) as well as muscles of nondiabetic, glucose-intolerant subjects (74). Recent reports indicate that PTP-1B is capable of dephosphorylating the insulin receptor and IRS-1 (75-77) with higher efficiency than other tyrosine phosphatases (78). In the fructose-fed hamster, PTP-1B overexpression in the liver may contribute to dephosphorylation of the insulin receptor and IRS-1, leading to hepatic insulin resistance. It has been shown that following insulin stimulation, PTP-1B is compartmentalized with IRS-1 in the microsomal membrane and acts as a regulator of insulin signaling (79,80). Phung et al. (42) have also shown that, with acute exposure of hepatocytes to insulin, PI 3-kinase and IRS-1 were localized to the microsomal membrane close to the site of apoB synthesis. They postulated that localization of these two key proteins may be involved in the suppression of apoB secretion from these cells. Whether co-localization of PTP-1B with IRS-1 and other signaling molecules alters insulin signaling status and indirectly influences the apoB biosynthetic pathway is unclear and requires further investigation. In the current study, incubation of insulin-resistant hepatocytes with increasing doses of vanadate improved the phosphorylation status of the insulin receptor in a dose-dependent manner. The improvement in insulin receptor phosphorylation coincided with a marked reduction in cellular, secreted, and total apoB, suggesting possible involvement of protein-tyrosine phosphatases such as PTP-1B in modulating apoB production. Our observations confirmed previous observations by Jackson et al. (81) about inhibitory effects of vanadate on apoB biogenesis in rat hepatocytes, although they used the phosphatase inhibitor in normal hepatocytes rather than fructose-fed/insulin-resistant hepatocytes.
In experiments involving chronic exposure of normal hepatocytes to high insulin concentrations, overexpression of PTP-1B and reduction in insulin receptor phosphorylation appeared to coincide with an increase in the synthesis and secretion of apoB, suggesting a possible link between the impairment in hepatic insulin signaling and apoB overproduction. This observation compares well with our previous in vivo and ex vivo observations in the fructose-fed hamster model and supports the hypothesis that induction of hepatic insulin resistance may play an important role in VLDL-apoB overproduction. It is unclear, however, how changes in the insulin signaling pathway lead to alterations in hepatic apoB metabolism. Insulin may directly alter the phosphorylation status of apoB as shown previously in rat hepatocytes (24). It is also possible that reduced phosphorylation of apoB due to impaired insulin signaling may protect apoB from degradation in insulin-resistant hepatocytes.
An ER-localized protease, ER-60, has been previously implicated in intracellular degradation of apoB in HepG2 cells (21). Interestingly, we observed a marked suppression of ER-60 protein expression in hepatocytes isolated from fructose-fed hamsters compared with that of control counterparts, suggesting that increased stability of apoB in insulin-resistant hepatocytes may be partly attributed to the suppression of ER-60 levels. This suppression was partially restored in hepatocytes isolated from fructose-fed hamsters treated with rosiglitazone, an insulin-sensitizing agent, suggesting that ER-60 may normally be positively regulated by insulin, an effect that may be lost in insulin resistance. In support of this view, we observed that the expression of ER-60 protein was markedly suppressed with chronic high insulin exposure of control hamster hepatocytes.
In summary, hepatic VLDL-apoB overproduction in the fructose-fed hamster model appears to be closely associated with the development of insulin resistance in hamster hepatocytes. Down-regulation of hepatic insulin signaling was linked to overexpression of PTP-1B and correlates with changes in the cellular level of this phosphatase. Hepatic insulin resistance was in turn associated with suppression of ER-60 and oversecretion of apoB. Further studies are in progress to more directly link changes in hepatic insulin signaling status to key components of the VLDL assembly and secretion process.