Animal Experiments

All animal studies were approved by the Institutional Animal Care and Use Committee of The Hamner Institutes for Health Sciences and fully complied with the guidelines from the National Institutes for Health. Animals were housed under the usual day (12-h daylight) and night (12-h darkness) circadian and fed ad libitum. The following animal experiments were included: 1) C57BL/6 (B6) mice were fed with either the HFD (Research Diets catalog number D12330: 58.0 kcal % fat, 16.0 kcal % protein, and 26 kcal % carbohydrate) or the normal rodent chow diet (ND) for 8 weeks (28); 2) B6 mice were treated with streptozotocin (STZ; 60 mg/kg/day) for 7 days via intraperitoneal injection to induce T1DM as previously described (28); 3) to study the effect of insulin signaling blockade, B6 mice were similarly fed with either ND or HFD as described above for 4 weeks. During the last week of the ND or HFD, some mice were administered the phosphatidylinositol 3-kinase inhibitor LY294002 (10 mg/kg body weight, intraperitoneal injection, once a day at 9 a.m., 7 days) during day time, when mice normally sleep and do not eat.

Cells

Mouse primary hepatocytes were isolated from B6 mice that were fed with the ND and cultured in Williams' medium E supplemented with 10% fetal bovine serum as previously described (10, 29,–35). Mouse hepatoma 1c1c7 cells were maintained in high glucose Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum.

Calculation of Insulin Sensitivity

Insulin sensitivity was evaluated by measuring levels of blood glucose and plasma insulin and calculated as previously described (36): quantitative insulin sensitivity check index (QUICKI) = 1/log(I0 + log(G0)) and homeostasis model assessment (HOMA) = (G0 × I0)/22.5), where I0 represents fasting insulin (microunits/ml) and G0 represents fasting glucose (mg/dl) for QUICKI, and for HOMA, G0 is in mmol/liter.

DNA Transfection and Adenoviral Infection

DNA plasmids were introduced into the indicated cells by Lipofectamine 2000 transfection agents (Invitrogen). Adenoviruses encoding the nuclear form of FoxO1 (ADA-FoxO1) or empty adenovirus, amplified in HEK-293 cells, were applied to infect primary hepatocytes at 10⁸ plaque-forming units/well in 6-well plates as described (35).

Immunoblotting

Cells were lysed in Nonidet P-40 lysis buffer (1% Nonidet P-40, 150 mm NaCl, 10% glycerol, 2 mm EDTA, 20 mm Tris (pH 8.0), 1 mm dithiothreitol, 1 mm sodium orthovanadate, 1 mm phenylmethylsulfonyl fluoride, 2 μg/ml leupeptin, and 10 μg/ml aprotinin). Cell lysates (10–15 μg/ lane) were resolved in 4–20% Tris-glycine gels (Invitrogen) and transferred to nitrocellulose membranes (Bio-Rad). Target proteins were detected by immunoblotting with primary antibodies as indicated and alkaline phosphatase-conjugated secondary antisera. Fluorescent bands were visualized with a Typhoon 9410 variable mode Imager from GE Healthcare and then quantified by densitometry analysis using ImageQuant version 5.2 software from GE Healthcare.

RNA Extraction and Real-time PCR

Total RNAs were extracted from cells or tissues with an RNeasy minikit (Qiagen) and reverse transcribed into cDNAs, which were quantified by TaqMan® real-time PCR with specific primers listed in Table 1 and normalized to levels of 36B4 transcripts.

TABLE 1

Full names and sequences of primers for real-time PCR used in this study

Gene name	DNA sequence
beclin1
Forward	5′-ggccaataagatgggtctga-3′
Reverse	5′-cactgcctccagtgtcttca-3′
vps34
Forward	5′-tgtcagatgaggaggctgtg-3′
Reverse	5′-ccaggcacgacgtaacttct-3′
ulk2
Forward	5′-cagccctggatgagatgttt-3′
Reverse	5′-ggatgggtgacagaaccaag-3′
atg12
Forward	5′-ggcctcggaacagttgttta-3′
Reverse	5′-cagcaccgaaatgtctctga-3′
gabarapl1 (GEC1)
Forward	5′-catcgtggagaaggctccta-3′
Reverse	5′-atacagctggcccatggtag-3′
36B4
Forward	5′-gcagacaacgtgggctccaagcagat-3′
Reverse	5′-ggtcctccttggtgaacacgaagccc-3′

Open in a separate window

Confocal Microscopy

Hepatoma 1c1c7 cells were plated onto glass bottom cell culture chamber (MatTek, Ashland, MA). The next day, LC3-GFP or GFP was introduced into cells by Lipofectamine 2000 transfection reagent (Invitrogen). Some cells were then cultured in regular Dulbecco's modified Eagle's medium with 10% fetal bovine serum at 37 °C. Cells were next pretreated with 1 nm insulin for 30 min, followed by incubation in the nutrient starvation medium (Earle's balanced salt solution (EBSS)) (Invitrogen) (37) or regular medium plus 100 nm glucagon for 8 h. To test whether insulin suppresses the formation of mitochondrial autophagy (mitophagy) in the presence of insulin resistance, some cells were pretreated with insulin (10 nm) for 24 h to induce insulin resistance as described (27). Mitochondria were labeled with the mitochondrion-specific dye (tetramethylrhodamine methyl ester, 25 nm) (catalog number T668, Invitrogen) at room temperature for 30 min. Cells were subsequently imaged live at room temperature in Dulbecco's modified Eagle's medium without phenol (Invitrogen) using a Zeiss LSM 510Meta confocal imaging system with a 30-milliwatt argon laser and a ×63 1.4 numerical aperture oil immersion objective at ×2 zoom. GFP was detected using a 488-nm argon laser and detected with a 515–540-nm band pass filter. Tetramethylrhodamine methyl ester fluorescence was excited with a 563-nm helium-neon laser attenuated and detected at >570 nm. The images were manipulated with Zeiss LSM software.

Hepatic Expression of Key Genes Involved in Activation of Autophagy Is Suppressed in the Presence of the HFD-induced Insulin Resistance and Hyperinsulinemia

The autophagy machinery includes two main ubiquitin-like complexes (1, 2). In complex I (Fig. 2), Atg12 is activated by Atg7 and then transferred by Atg10 to Atg5. The Atg12-Atg5 dimer will be joined by Atg16 to form an Atg12-Atg5-Atg16 multimer complex, which is localized to membranes of early autophagosomes and will be subsequently joined by the complex II. The formation of autophagy complex II is initiated when LC3-I (a mammalian homologue of Atg8) is cleaved by Atg4 to produce LC3-II. LC3-II will then be conjugated to phosphatidylethanolamine by Atg7 and Atg3. The unconjugated LC3-I stays in the cytosol, but the conjugated LC3-II binds to the autophagosomal membrane, following the formation of the complex I. Complex I will eventually detach from mature autophagosomes, whereas LC3-II stays on the membrane until it is degraded by the lysosomes. That is why LC3-II has been widely used as a marker for monitoring activation of autophagy. Additionally, several other proteins are required for the initiation of autophagy. These proteins include at least Ulks (a mammalian homologue of Atg6), Beclin1 (another homologue of Atg6), and Vps34 (a class III phosphatidylinositol 3-kinase).

Open in a separate window

FIGURE 2.

Diagram of the autophagy pathway.

To further examine the autophagy process in the presence of insulin resistance/hyperinsulinemia, mRNA levels of some of the above described autophagy genes were examined in the livers of the mice detailed in Fig. 1. As shown in Fig. 3, the mRNA levels of vps34, atg12, and gabarapl1 (protein name GEC, another homologue of Atg8 (41)) genes were significantly decreased by the HFD, although Beclin1 and Ulk2 gene transcripts were not affected. Because vps34, atg12, and gabarapl1 genes are involved in the initiation and formation of complexes I and II of autophagy, respectively, these results further demonstrate that hepatic autophagy level is decreased in the presence of insulin resistance/hyperinsulinemia induced by the HFD.

Open in a separate window

FIGURE 3.

Expression levels of some key autophagy genes in the livers of mice were suppressed by the HFD. mRNAs from the liver samples of mice described in Fig. 1 were extracted, converted into cDNA, and used for quantifications of target genes. Transcripts of ulk2 (A), beclin1 (B), vps34 (C), atg12 (D), and gabarapl1 (E) were measured by using real-time PCR and normalized to the level of 36B4 transcript. *, p < 0.05; **, p < 0.01; ND (n = 10) versus HFD (n = 11).

To explore whether or not insulin signaling was responsible for the suppressed autophagy in the presence of insulin resistance/hyperinsulinemia, insulin resistance in mice was induced by the HFD (4 weeks) as we previously described (28). Some mice were administered the phosphatidylinositol 3-kinase inhibitor LY294002 during day time (once a day for the last week), when mice normally sleep and do not eat. The basal level of insulin signaling was evaluated by measurements of Akt phosphorylation. Autophagy markers were detected as detailed above. Results showed that the basal level of Akt phosphorylation was increased by the HFD as we previously described (28) but reversed by the treatment with LY294002 (supplemental Fig. 1, A and B). The autophagy marker p62 in the liver was increased by the HFD, and the increase was largely prevented by LY294002 (supplemental Fig. 1C). Transcript levels of key autophagy genes (vps34, atg12, and gebarapl1) in the liver were decreased by the HFD but reversed by the treatment with LY294002 (supplemental Fig. 1, D–F). These results imply that insulin is responsible for the suppressed autophagy in the presence of insulin resistance and hyperinsulinemia.

Autophagy and Expression of Key Autophagy Genes Are Increased in the Livers of Mice with Insulin Deficiency Induced by STZ

It has previously been shown that the autophagy level is increased in STZ-induced insulin deficiency, which resembles but is not an equivalent of T1DM (42). However, the associated mechanism has not been studied. Because the basal Akt-dependent insulin signaling is increased in the presence of insulin resistance/hyperinsulinemia, as shown in Fig. 1 and supplemental Fig. 1, and expression of some key autophagy genes was decreased by the HFD (Fig. 3), it is possible that insulin suppresses autophagy through inhibiting expression of key autophagy genes. To further examine the effect of insulin on expression of key autophagy genes, insulin secretion in mice was deleted by STZ, followed by measurements of levels of autophagy and transcripts of some key autophagy genes. As shown in Fig. 4, A and B, application of STZ significantly damaged the pancreatic insulin secretion, leading to overt hyperglycemia. As predicted, the LC3-II/LC3-I ratio in the liver was increased by the treatment with STZ (Fig. 4C), whereas levels of both p62 proteins and polyubiquitin were decreased (Fig. 4, D and E). Together, these results demonstrate that autophagy was activated in the livers of mice when insulin secretion was crippled by STZ.

Open in a separate window

FIGURE 4.

Autophagy was increased in the livers of mice with T1DM. T1DM was induced in B6 mice by using STZ (60 mg/kg body weight, once a day, intraperitoneally, 7 days). Levels of fasting blood glucose (A, top), plasma insulin (A, bottom), and blood glucose during the fed state (B) were measured. Levels of LC3-I and -II (C), p62 (D), and polyubiquitin (Poly-Ub) (E) in the liver were measured by immunoblotting and normalized to the level of β-actin. **, p < 0.01; ***, p < 0.001; STZ group (n = 5) versus vehicle group (n = 5).

Next, transcript levels of vps34, atg12, and gabarapl1 genes were quantified in the livers of mice with STZ-induced insulin deficiency. As shown in Fig. 5, expression of all of these genes was significantly increased in the livers of mice with insulin deficiency. These results further support the notion that insulin suppresses autophagy by inhibiting expression of key autophagy genes directly or indirectly.

Open in a separate window

FIGURE 5.

Expression of key autophagy genes in the livers of mice with T1DM was increased. mRNAs in the liver samples from the mice described in the legend to Fig. 4 were extracted, converted into cDNA, and used for quantifications of target genes. Transcripts of vps34 (A), atg12 (B), and gabarapl1 (C) were determined by using real-time PCR and normalized to the level of 36B4 transcript. *, p < 0.05; **, p < 0.01; STZ group (n = 5) versus vehicle group (n = 5).

Insulin Inhibits Autophagy and Expression of Key Autophagy Genes in Cultured Hepatocytes

To investigate the mechanism of insulin suppression of autophagy, we used a cellular system (primary hepatocytes). Autophagy was activated by either nutrient starvation or glucagon, followed by evaluating the expression levels of key autophagy genes. As shown in Fig. 6, A and B, the LC3-II/LC3-I ratio was significantly increased by the nutrient starvation, indicating increased autophagy. The increased LC3-II/LC3-I ratio was reversed by the treatment with insulin. The effect of insulin was largely prevented by the pretreatment with the phosphatidylinositol 3-kinase inhibitor LY294002. Akt phosphorylation was significantly elevated by the treatment with insulin but largely prevented by LY294002. Similarly, in Hepa1c1c7 hepatoma cells stained by immunofluorescence, both nutrient starvation and treatment with glucagon stimulated LC3-II aggregation as intense immunofluorescence dots, and the aggregation of these dots was reduced by insulin (Fig. 6, C–I). These results demonstrate the induction and inhibition of autophagy in our culture system.

Open in a separate window

FIGURE 6.

Autophagy is suppressed by insulin in cultured hepatocytes. A and B, primary hepatocytes isolated from mice were incubated in either the regular Dulbecco's modified Eagle's medium or nutrient-deprived starvation medium (EBSS) for 8 h. Insulin (1 nm) was added into some cells for 8 h as noted. The phosphatidylinositol 3-kinase inhibitor LY294002 (100 nm) was added to some cells prior to insulin as noted. Levels of LC3-I and II, phospho-Akt and total Akt, and β-actin were measured by immunoblotting, quantified, and normalized. * and #, p < 0.05. C–H, Hepa1c1c7 cells were plated onto glass slips in culture chambers for 12 h, followed by the introduction of GFP-labeled LC3 (LC3-GFP) into the cells via transfection. Cells were incubated in either the starvation medium (EBSS) or regular medium with glucagon (100 nm) in the presence or absence of 1 nm insulin for 8 h. LC3-GFP was visualized with confocal microscopy as detailed under “Materials and Methods.” Scale bar, 10 μm. I, quantification analysis of autophagy dots in hepatocytes. Data represent mean ± S.E. from three independent experiments of C–H. ***, p < 0.001.

Next, we examined the effect of insulin on the expression level of key autophagy genes. As shown in Fig. 7, mRNA levels of vps34, atp12, and gabarapl1 were elevated by nutrient starvation, but the elevations were significantly reversed by insulin. The effect of insulin was largely prevented by the phosphatidylinositol 3-kinase inhibitor LY294002. Together, these results show that insulin suppresses expression of key autophagy genes in cultured hepatocytes.

Open in a separate window

FIGURE 7.

Expression of key autophagy genes (vps34 (A), atg12 (B), and gabarapl1 (C)) was suppressed by insulin in cultured cells. Primary hepatocytes were treated as detailed in the legend to Fig. 6A. Transcript levels of target genes were determined by real-time PCR and normalized to the level of 36B4 transcript. Results represent mean ± S.E. from three independent experiments. *, p < 0.05 insulin plus starvation versus starvation only. #, p < 0.05 LY294002-treated cells versus cells treated with insulin plus starvation.

Insulin Inhibits Autophagy and Expression of Key Autophagy Genes in a FoxO1-dependent Manner

To examine the mechanism by which insulin inhibits expression of autophagy genes, the constitutive nuclear form of FoxO1 was introduced into Hepa1c1c7 cells prior to the treatment with insulin as noted (34), followed by measurements of LC3-II and mRNA levels of key autophagy genes. As shown in Fig. 8A, nutrient starvation increased the autophagy marker, LC3-II, but the increase was reversed by insulin, which stimulated Akt phosphorylation. However, in the presence of the overexpressed constitutive nuclear form of FoxO1, the effect of insulin was prevented, suggesting that insulin suppression involved FoxO1-mediated gene transcription. To determine the role of FoxO1 in insulin-mediated suppression of autophagy genes, the effect of the constitutive nuclear form of Foxo1 was examined. As shown in Fig. 8B, nutrient starvation-induced expression of vps34, atg12, and gabarapl1 was blocked by insulin. Nevertheless, in the presence of the overexpressed constitutive nuclear form of FoxO1, insulin failed to suppress expression of vps34, atg12, and gabarapl1 genes. Together, these results demonstrate that the FoxO1-mediated transcription of key autophagy genes is an intricate part of insulin suppression of autophagy.

Open in a separate window

FIGURE 8.

Insulin-suppressed expression of key autophagy genes was mediated by FoxO1. The constitutive nuclear form of FoxO1 (ADA-FoxO1) encoded by the recombinant adenovirus or the empty adenoviruses was introduced into Hepa1c1c7 cells via infection. Thirty-six hours later, cells were cultured in the starvation medium (EBSS) in the presence or absence of insulin (1 nm) for an additional 8h. A, levels of LC3, phosphorylated Akt, and β-actin were measured by immunoblotting. Levels of FoxO1 and PARP1 in nuclear protein extracts were measured by immunoblotting with the antibodies as noted. *, p < 0.05. B, transcripts of vps34, atg12, and gabarapl1 genes were determined by using real-time PCR and normalized to the level of 36B4 transcript. Results represent mean ± S.E. from three independent experiments. *, p < 0.05; **, p < 0.01 insulin plus starvation versus starvation only. #, p < 0.05; ###, p < 0.001 ADA-FoxO1 infection plus insulin plus starvation group versus starvation only group.