AMPK and Akt/mTOR signalling pathways participate in glucose‐mediated regulation of hepatitis B virus replication and cellular autophagy

A growing consensus indicates that host metabolism plays a vital role in viral infections. Hepatitis B virus (HBV) infection occurs in hepatocytes with active glucose metabolism and may be regulated by cellular metabolism. We addressed the question whether and how glucose regulates HBV replication in hepatocytes. The low glucose concentration at 5 mM significantly promoted HBV replication via enhanced transcription and autophagy when compared with higher glucose concentrations (10 and 25 mM). At low glucose concentration, AMPK activity was increased and led to ULK1 phosphorylation at Ser 555 and LC3‐II accumulation. By contrast, the mTOR pathway was activated by high glucose concentrations, resulting in reduced HBV replication. mTOR inhibition by rapamycin reversed negative effects of high glucose concentrations on HBV replication, suggesting that low glucose concentration promotes HBV replication by stimulating the AMPK/mTOR‐ULK1‐autophagy axis. Consistently, we found that glucose transporters inhibition using phloretin also enhanced HBV replication via increased AMPK/mTOR‐ULK1‐induced autophagy. Surprisingly, the glucose analogue 2‐deoxy‐D‐glucose reduced HBV replication through activating the Akt/mTOR signalling pathway also at the low glucose concentrations. Our study reveals that glucose is an important factor for the HBV life cycle by regulating HBV transcription and posttranscriptional steps of HBV replication via cellular autophagy.

worldwide. Patients with chronic HBV infection (CHB) often suffer from severe liver diseases including fibrosis, cirrhosis, and hepatocellular carcinoma (Ganem & Prince, 2004). The intrinsic mechanisms of liver disease progression in CHB are under active investigation. Nevertheless, there is no treatment available for effective eradication of CHB yet. Therefore, it is essential to further investigate the complex host-HBV interaction to gain a deeper understanding of the mechanisms of HBV pathogenesis and to identify new therapeutic targets.
Glucose is the most important nutrition and energy source for organisms and is the raw material for glycolysis. To be utilised, glucose is taken up via glucose transporters (GLUTs) and converted to glucose 6-phosphate by hexokinases (HKs), the first and rate-limiting step in glycolysis (Adeva-Andany, Perez-Felpete, Fernandez-Fernandez, Donapetry-Garcia, & Pazos-Garcia, 2016). Glucose 6phosphate may enter various metabolic pathways, including glycolysis, glycogen synthesis, the hexosamine pathway, the pentose phosphate pathway, and oxidative routes (DeBerardinis & Chandel, 2016). Glycolysis is an essential cellular pathway in all cell types.
During this process, glucose is metabolized to pyruvate, with subsequent releases of two molecules of adenosine triphosphate (ATP) and two nicotinamide adenine dinucleotides. Then, pyruvate is reduced to lactate under anaerobic conditions. However, pyruvate may transfer to mitochondria and be oxidized to produce acetylcoenzyme A under aerobic conditions, sustaining the tricarboxylic acid cycle. Several small molecule inhibitors can be used to interfere with glucose uptake and glycolysis. Phloretin is a phenolic compound found in apples (Lee, Kim, Kim, Lee, & Lee, 2003) and strawberries (Hilt et al., 2003) and is an inhibitor of GLUTs. A glucose analogue, 2-deoxy-D-glucose (2-DG), has a 2-hydroxy group substituted by hydrogen and can also enter the cell through GLUTs. However, 2-DG is phosphorylated to 2-DG-6-P by HKs, which in turn blocks HK activity and inhibits glycolysis. Cellular energy metabolism can globally alter other processes, such as transcription and autophagy (Lindqvist, Tandoc, Topisirovic, & Furic, 2018). Glucose metabolism directly influences the abundance of related transcription factors as well as downstream gene expression (Metallo & Vander Heiden, 2013;Vaulont, Vasseur-Cognet, & Kahn, 2000). Under nutrient-rich conditions, mTORC1 is activated to support cell growth and to block autophagy via inhibition of ULK1. Low cellular energy metabolism activates AMPK that subsequently phosphorylates ULK1 on amino acid residue Ser 555, thereby inducing autophagy. Glycolysis inhibition by 2-DG has indirect impact on many cellular pathways. As an example, glucose deprivation or glycolysis inhibition by 2-DG results in decreased ATP levels and increased adenosine monophosphate (AMP) levels in cells.
An elevated AMP/ATP ratio leads to AMPK activation and triggers downstream pathways.
In response to viral infections, glycolysis may be increased or decreased, with beneficial results for viruses, for example, in the case of herpes simplex virus 1 and adenovirus infections (Abrantes et al., 2012;Thai et al., 2014). Studies found that HBV infection modulated host liver metabolic pathways, resulting in upregulation of glucose metabolism (e.g., gluconeogenesis, aerobic oxidation of glucose, and pentose phosphate pathway; Liu et al., 2015;Shin et al., 2011) and lipid metabolism (e.g., fatty acids, phospholipids, and cholesterol biosynthesis; Chen, Liang, Ou, Goldstein, & Brown, 2004;M. D. Wang et al., 2016;Yang et al., 2008). These reports suggest a complex relationship between HBV infection and metabolic changes in hepatocytes. Nevertheless, the impact of the changes in hepatic metabolisms on HBV replication has not been studied so far in detail. Therefore, we aimed to provide experimental evidence for the hypothesis that glucose metabolism regulates HBV replication.
Indeed, the low glucose concentration in cell cultures and inhibition of GLUTs led to activation of AMPK-mTOR-ULK1-autophagy axis in hepatocytes. In recent years, many studies demonstrated that HBV replication depended on cellular autophagy, and the AMPK-Akt/mTOR-ULK1-induced autophagy pathway significantly regulates HBV replication (Lin et al., 2017;Lin et al., 2019;J. Wang et al., 2019;Xie et al., 2016). In the present study, we found that HBV replication and gene expression were modulated by varying glucose concentrations in the cell culture medium. Treatment with 2-DG decreased glycolysis and activated the Akt/mTOR pathway, resulting in the inhibition of HBV replication, even at a low glucose concentration.

| Low glucose concentration enhances HBV replication and gene expression in hepatocytes
Given glucose as one of the most important metabolic substrates in living organisms, we asked whether and how external glucose supply regulates HBV replication. Under the standard condition, HepG2.2.15 cells with stable HBV replication were grown in complete RPMI-1640 medium with 10-mM glucose. We first examined how the glucose concentration in the culture medium affected HBV replication and gene expression. HepG2.2.15 cells were cultured with different glucose concentrations (5, 10, and 25 mM).
HBV RIs were prepared on Day 4 and subjected to Southern blotting analysis. The amount of HBV RIs decreased significantly with increasing glucose concentrations ( Figure 1a). The levels of encapsidated and secreted HBV DNA (in HBV virions) were determined using real-time PCR. Consistently, a low glucose concentration of 5 mM in the cell culture medium led to increased levels of both intracellular and secreted HBV DNA (Figure 1a). The levels of intracellular and secreted HBsAg and HBeAg were measured by chemiluminescent microparticle immunoassay. The levels of intracellular and secreted HBsAg but not HBeAg were substantially higher under the low glucose condition (Figure 1a).
To explain the changes in HBV replication activity at different glucose concentrations, the effects of glucose concentrations on HBV gene expression were judged by determining the levels of HBV RNAs using real time RT-PCR and the HBV promoter activity using luciferase reporter assays. The levels of HBV RNAs were highest at the glucose concentration of 5 mM, compared with those at higher glucose concentrations (10 and 25 mM; Figure 1b). However, HBV RNA levels changed less than twofold among the different glucose concentrations in the cell cultures. The luciferase reporter assays clearly showed that single HBV promoter activity dropped slightly (Figure 1c), though the levels of some specific transcription factors, such as PGC1α, CREB, FIGURE 1 Low glucose concentration enhances HBV replication and gene expression in hepatocytes. (a,b) HepG2.2.15 cells were cultured in medium with the indicated glucose concentrations (5, 10, and 25 mM) for 72 hr. (a) The HBsAg and HBeAg levels in the culture supernatants and intracellular HBsAg and HBeAg from cell lysates were quantified by chemiluminescent microparticle immunoassay. The levels of intracellular HBV DNA and that in the supernatants were determined by real-time PCR. Encapsidated HBV RIs were detected by Southern blotting. (b) Realtime RT-PCR was performed to determine the HBV RNA levels in HepG2.2.15 cells. (c) HepG2.2.15 cells were cotransfected with HBV promoters, pSP1, pSP2, pCp and pXP, and pRenilla. Six hours post transfection, cells were cultured in medium with the indicated glucose concentrations (5, 10, and 25 mM) for 48 hr. The Dual-Glo luciferase report assays were performed to detect the activity of HBV promotes. The results were normalized to the control samples (pGL3) and shown as fold-change. (d) HepG2.2.15 cells were cultured in medium with the indicated glucose concentrations (5, 10, and 25 mM) and harvested after 48 hr. PGC1α, CREB, and ChREBP expression levels were analysed by Western blotting, using beta-actin as a loading control. *p < .05; **p < .01; ***p < .001; # p < .05; ## p < .01; ns, not significant. HBV, Hepatitis B virus; RC, relaxed circular DNA; S/CO, signal to cutoff ratio; SS, single-stranded DNA and ChREBP, were strongly reduced at the higher glucose concentrations ( Figure 1d). These results suggest that the glucose concentration modulates HBV transcription. Nevertheless, the regulation at the level of transcription was apparently not the only mechanism to determine the changed magnitude of HBV replication at different glucose concentrations.
Low glucose concentration may cause insufficient nutrition of cells.
Therefore, we added pyruvate as a supplement at different glucose concentrations in the culture medium. Nevertheless, adding pyruvate did not affect HBV replication at low glucose concentration ( Figure   S1a, Supporting information). Next, cell proliferation was measured using a CCK8 assay at indicated time points from 6 to 72 hr after cul-  (Galle et al., 1994), indicating that PHHs fully support HBV replication.
Collectively, the data suggest that low glucose concentration enhances HBV replication in hepatoma cells and PHHs. However, the changes in HBV transcription do not fully explain the difference on HBV replication activity at low and high glucose concentrations.
Other cellular mechanisms may participate in the modulation of HBV replication.

| Enhanced HBV replication at low glucose concentration depends on increasing autophagic flux
AMPK is active at low glucose concentrations and interacts with ULK1, subsequently phosphorylating ULK1 at amino acid residue Ser 555, thereby initiating autophagy (Kim, Kundu, Viollet, & Guan, 2011). Furthermore, the Akt/mTOR pathway is inhibited at low glucose concentrations (Harada et al., 2009;Ryu, Lee, Yun, & Han, 2010). In addition, previous studies reported that the Akt/mTOR pathway is negatively associated with HBV replication (Bagga, Rawat, Ajenjo, & Bouchard, 2016;Lin et al., 2017;Rawat & Bouchard, 2015). Therefore, we proposed that glucose modulated HBV replication through the AMPK-and Akt/mTOR/ULK1-induced autophagy. The levels of LC3 at the indicated glucose concentrations were measured using immunofluorescence staining and Western blotting analysis. At low glucose concentration (5 mM), the numbers of endogenous LC3-positive autophagic puncta as well as HBsAg FIGURE 2 Glucose regulates Hepatitis B virus (HBV) infection, and the AMPK-, Akt/mTOR-autophagy axis in primary human hepatocytes (PHHs). (a-d) PHHs were infected with HBV virions (multiplicity of infection = 30). Ten days post infection, PHHs were cultured in Dulbecco's Modified Eagle medium with the indicated glucose concentrations (5 and 10 mM) and harvested after 48 hr. (a) The HBsAg and HBeAg levels in the culture supernatants were determined as described above. (b-d) Western blotting analysis was performed to detect the levels of LC3, p62, AMPK, p-AMPK, ULK1, p-ULK1, Akt, p-Akt, mTOR, and p-mTOR using beta-actin as a loading control. (e) PHHs were infected with HBV virions (multiplicity of infection = 30). The cell lysates were collected at the indicated time points. The expression of albumin was detected by Western blotting using beta-actin as a loading control. *p < .05; **p < .01. LC3, microtubule-associated protein 1 light chain 3 beta; S/CO, signal to cutoff ratio expression ( Figure 3a) and the expression levels of LC3-II and p62 (Figure 3b) were markedly higher than those at higher glucose concentrations in HepG2.2.15 cells. Consistently, the levels of LC3-II and p62 in PHHs were higher at 5 mM glucose than at 10 mM ( Figure 2b). In an additional experiment, HepG2.2.15 cells were cultured at three indicated glucose concentrations with or without chloroquine, an inhibitor of autolysosomal cargo degradation. The 5-mM glucose further permitted LC3-II accumulation ( Figure S2), suggesting a stronger autophagic flux at the low glucose concentration.
To further investigate the involvement of autophagy in the modulation of HBV replication at different glucose concentrations, the autophagy-related gene ATG5 was silenced, and HBV replication, intracellular HBsAg and HBeAg levels, and HBsAg and HBeAg levels in the supernatants were measured. ATG5 silencing decreased the LC3-II levels in HepG2.2.15 cells at all three indicated glucose concentrations used for cell culture ( Figure 3c). Furthermore, HBV IRs, intracellular HBV DNA levels, intracellular HBsAg levels, and HBsAg levels in the supernatants were also lower at all used glucose concentrations ( Figure 3d). In summary, low glucose concentration increases autophagic flux, which is associated with enhanced HBV replication.

| Glucose changes HBV replication through regulating AMPK-Akt/mTOR-dependent autophagy
Next, we examined in detail how glucose concentration modulates cellular signalling pathways and thereby regulates HBV replication. Previous studies established that autophagy is regulated by the AMPK and FIGURE 3 Enhanced HBV replication at low glucose concentration depends on increasing autophagic flux. (a) HepG2.2.15 cells were cultured in medium with the indicated glucose concentrations (5, 10, and 25 mM) and harvested after 48 hr. The cells were fixed and incubated with a primary rabbit anti-LC3B and horse anti-HBsAg antibodies and then stained with an Alexa Fluor 488-conjugated anti-rabbit and Alexa Fluor 594conjugated anti-horse secondary antibody IgG, respectively. The distribution of LC3 was imaged by immunofluorescence microscopy. Scale bar, 5 μm. (b) HepG2.2.15 cells were cultured in medium with the indicated glucose concentrations (5, 10, and 25 mM) and harvested after 48 hr. The LC3 and p62 expression levels were analysed by Western blotting using beta-actin as a loading control. (c) HepG2.2.15 cells were transfected with siATG5 or a control siRNA (siR-C) at 40 nM, 24 hr post transfection, the cells were cultured in medium with the indicated glucose concentrations (5, 10, and 25 mM) and harvested after 72 hr. Western blotting analysis was used to detect the ATG5 silencing effect. The LC3 and p62 expression levels were analysed by Western blotting, using beta-actin as a loading control. (d) HepG2.2.15 cells were treated as in (c). The HBsAg and HBeAg levels in the culture supernatants and intracellular HBsAg levels from cell lysates were determined as described above. The HBV DNA levels in intracellular was detected by real-time PCR. Encapsidated HBV replicative intermediates were detected by Southern blotting. *p < .05; **p < .01; ***p < .001; ns, not significant. DAPI 4′,6-diamidino-2-phenylindole; HBV, Hepatitis B virus; LC3, microtubule-associated protein 1 light chain 3 beta; S/CO, signal to cutoff ratio mTOR signalling pathways Nwadike, Williamson, Gallagher, Guan, & Chan, 2018;Xie et al., 2016;Xu et al., 2016). Therefore, we tested whether changed glucose concentrations regulated AMPK, Akt, and mTOR activities in host cells. HepG2.2.15 cells were cultured with the indicated glucose concentrations (5, 10, and 25 mM) for 48 hr. The expression of total AMPK, Akt, mTOR, and p70S6K proteins and its phosphorylated forms were detected using Western blotting. Although AMPK and ULK1 were significantly activated, Akt, mTOR, and p70S6K were inactivated at 5-mM glucose, respectively, according to the relative level of their phosphorylated forms (Figure 4 a,b). Due to long-term experiments, the total levels of these proteins often changed with the same tendency of the corresponding phosphorylated forms. In PHHs, 5-mM glucose in the culture medium had a similar impact on the expression levels of these proteins (Figure 2c,d).
Therefore, the AMPK pathway was activated, whereas the Akt/mTOR pathway was inhibited at the low glucose concentration.

FIGURE 4
Low glucose concentration activates AMPK and inactivates mTOR, thereby enhancing HBV replication. (a,b) HepG2.2.15 cells were cultured in medium with the indicated glucose concentrations (5, 10, and 25 mM) and harvested after 48 hr. Western blotting analysis was performed to detect the levels of total or phosphorylated AMPK, ULK1, mTOR, AKT, and p70 S6K, using beta-actin as a loading control. (c) HepG2.2.15 cells were treated with AICAR (0.1 mM) for 72 hr. The intracellular HBsAg and HBeAg levels and the HBsAg and HBeAg levels in the culture supernatants were determined as above described. The levels of intracellular HBV DNA and that in the supernatants were determined by real-time PCR. Encapsidated HBV replicative intermediates were detected by Southern blotting. (d) HepG2.2.15 cells were treated with AICAR (0.1 mM) for 48 hr. Western blotting analysis was used to determine the levels of AMPK, p-AMPK, LC3, and p62 using beta-actin as a loading control. (e,f) PHHs were infected with HBV virions (multiplicity of infection = 30). 10 days post infection, PHHs were treated with AICAR (0.1 mM) and harvested after 48 hr. (e) The HBsAg and HBeAg levels in the culture supernatants were determined as described above. (f) Western blotting analysis was used to determine the levels of AMPK, p-AMPK, LC3, and p62 in the lysates of PHHs using beta-actin as a loading control. *p < .05; **p < .01; ***p < .001; # p < .05; ns, not significant. AICAR, 5-aminoimidazole-4-carboxyamide-1-β-D-ribofuranoside; HBV, Hepatitis B virus; LC3, microtubule-associated protein 1 light chain 3 beta; PHH, primary human hepatocyte; RC, relaxed circular DNA; S/CO, signal to cutoff ratio; SS, single-stranded DNA To confirm the function of AMPK in regulation of HBV replication, the HepG2.2.15 cells were treated with an AMPK agonist AICAR for Taken together, these findings show that low glucose concentration in the medium activates the AMPK pathway but inhibits the Akt/mTOR signalling pathway to induce autophagy, thereby upregulating HBV replication.

| GLUT inhibitor phloretin enhances HBV replication by upregulating AMPK-Akt/mTOR-induced autophagy
Glucose is uptaken by GLUTs into cells. Phloretin is a well-known inhibitor of GLUTs and reduces the speed of glucose uptake into host cells (Xintaropoulou et al., 2015). We assumed that phloretin may modulate  Blocking Akt/mTOR signalling attenuates the promotion effects of low glucose on HBV replication. (a) HepG2.2.15 cells were cultured in medium with the indicated glucose concentrations (5, 10, and 25 mM) with or without inhibitor Akti 1/2 (1 μM) or rapamycin (1 μM) for 72 hr. Encapsidated HBV replicative intermediates were detected by Southern blotting. The levels of intracellular HBV DNA were determined by real-time polymerase chain reaction. The HBsAg and HBeAg levels in the culture supernatants and intracellular HBsAg from cell lysates were determined as described above. (b) HepG2.2.15 cells were cultured in the medium with the indicated glucose concentrations (5, 10, and 25 mM) with or without inhibitor Akti 1/2 (1 μM) or rapamycin (1 μM) for 48 hr. The LC3 and p62 expression levels were analysed by Western blotting, using betaactin as a loading control. *p < .05; ***p < .001; ns, not significant. HBV, Hepatitis B virus; LC3, microtubule-associated protein 1 light chain 3 beta; PHH, primary human hepatocyte; RC, relaxed circular DNA; S/CO, signal to cutoff ratio; SS, single-stranded DNA (Figure 6c), proven that phloretin activated AMPK but inhibited Akt/mTOR. Furthermore, phloretin treatment increased the expression of p62 and LC3-II in HepG2.2.15 cells.
Similarly, the phloretin treatment led to slightly enhanced HBsAg production in PHHs. PHHs were infected with HBV and treated with phloretin at 20 or 50 μM up to 10 days post infection. After phloretin treatment, the HBsAg but not HBeAg levels were slightly increased in the culture supernatants (Figure 6e). Consistently, the levels of phosphorylated AMPK and ULK1 and LC3-II in PHHs were elevated after phloretin treatment, whereas decreased levels of phosphorylated Akt and mTOR were detected (Figure 6f). Thus, phloretin treatment upregulates AMPK-Akt/mTOR-induced autophagy in hepatoma cells and PHHs.
In addition, phloretin slightly enhanced glycolysis, as indicated by increased lactate production in the culture medium ( Figure 6d). Finally, the presence of pyruvate did not alter the effect of phloretin on the production of HBsAg and HBeAg ( Figure S4a). The proliferation of HepG2.2.15 cells was measured using the CCK8 assay, confirming that phloretin treatment did not affect cell proliferation at any indicated time point ( Figure S4b). The HBsAg and HBeAg levels in the culture supernatants were determined as described above. (f) Western blotting analysis was performed to detect the levels of total or phosphorylated AMPK, Akt, mTOR, LC3, and p62 using beta-actin as a loading control. *p < .05; **p < .01; ***p < .001; ## p < .01; ns, not significant. HBV, Hepatitis B virus; LC3, microtubule-associated protein 1 light chain 3 beta; PHH, primary human hepatocyte; RC, relaxed circular DNA; S/ CO, signal to cutoff ratio; SS, single-stranded DNA Taken these data together, phloretin treatment enhances HBV replication, upregulates AMPK/mTOR-ULK1 autophagy, and increases glycolysis in hepatoma cells.

| 2-DG decreases HBV replication and gene expression
2-DG, an analogue of glucose, is best known as an inhibitor of glycolysis and blocks the cellular HK enzymes (Brown, 1962). We addressed the question whether blocking of glycolysis affects HBV replication. Therefore, we treated the HepG2.2.15 cells with 2-DG for 72 hr. As shows in Figure 7a

| 2-DG treatment leads to AMPK and Akt/mTOR phosphorylation but decreased HBV replication independently on glucose concentrations
It has been reported that 2-DG treatment results in an increase in intracellular AMP/ATP ratio, thereby activating AMPK. Consistently, the level of phosphorylated AMPKα subunit (Thr172) increased, when hepatoma cells were treated with 2-DG (1, 5, and 10 mM) for 48 hr (Figure 8a). However, the levels of phosphorylated Akt (Ser473), which is required for Akt to activate its downstream targets, was enhanced, as well as mTOR phosphorylation and p70 S6K expression (Figure 8a). 2-DG also promoted the p62 and LC3-II expression in HepG2.2.15 cells, consistent with the results of previous reports (Jeon, Kim, Park, & Yun, 2015;Xi et al., 2011).
Then, we asked whether the Akt/mTOR pathway participated in the regulation of HBV replication in the presence of 2-DG. Akt and mTOR were blocked using the inhibitors Akti 1/2 and rapamycin after 2-DG treatment, respectively. Both inhibitors abrogated the suppressive effect of 2-DG on HBV replication (Figure 8c), suggesting an involvement of Akt/mTOR signalling pathway in the regulation of HBV replication in the presence of 2-DG. The upregulated AMPK and Akt/mTOR phosphorylation by 2-DG may explain decreased HBV replication and gene expression under these culture conditions. Due to the complexity of the functions controlled by these pathways, other mechanisms may also contribute to HBV suppression and need to be considered in future study.

| DISCUSSION
Viruses depend on the supply of energy and building blocks for their replication (Fontaine, Sanchez, Camarda, & Lagunoff, 2015;Mahmoudabadi, Milo, & Phillips, 2017). Indeed, various metabolic pathways are essential for efficient viral replication. Here, we found that HBV replication was significantly influenced by the glucose concentration in the medium and by glucose uptake via GLUTs at the step of transcription and via AMPK-Akt/mTOR-ULK1-induced autophagy.
However, its analogue 2-DG suppressed HBV replication by inhibiting the glycolytic pathway and activating the Akt/mTOR signalling pathway despite the increased AMPK and autophagic activity (Figure 10).
Defining the cellular metabolic processes related to viral infection may reveal new therapeutic targets and contribute to the development of safe and effective therapies against viral infections (Ikeda & Kato, 2007). HBV is thought to be a "metabolovirus" (Shlomai & Shaul, 2008), and its transcription is largely dependent on hepatic metabolic controls (Bar-Yishay, Shaul, & Shlomai, 2011;Tacke, Liedtke, Bocklage, Manns, & Trautwein, 2005) and cellular transcription factors (Ondracek & McLachlan, 2011;Quasdorff & Protzer, 2010;Ramiere et al., 2008). Previous studies in HBV transgenic mice have illustrated, for example, that fasting decreases glucose levels but increases HBeAg synthesis in serum. PGC1α transcripts are induced by fasting in HBV transgenic mice (L. Li, Oropeza, Kaestner, & McLachlan, 2009). In this study, the expression of transcription factors, including PGC1α, CREB, and ChREBP, was found to be altered by these indicated glucose concentrations in the cell cultures and correlated with changes in the HBV RNAs levels. This is consistent with previous studies regarding the importance of HBV transcriptional control. HBV transcriptional activity is regulated by HBV promoters and two additional enhancers and the abundance of specific hepatic transcription factors. The expression of relevant transcription factors is inversely FIGURE 9 2-DG decreases Hepatitis B virus replication and gene expression by upregulating the Akt/mTOR signalling pathway in primary human hepatocytes (PHHs). Primary human hepatocytes were infected with Hepatitis B virus virions (multiplicity of infection = 30). Ten days post infection, PHHs were treated with 2-DG (1, 5, and 10 mM) and harvested after 48 hr. (a) The HBsAg and HBeAg levels in the culture supernatants were determined as described above. (b) Western blotting analysis was performed to detect the levels of total or phosphorylated AMPK, Akt, mTOR, LC3, and p62 using beta-actin as a loading control. *p < .05; **p < .01; ***p < .001; ns, not significant. 2-DG, 2-deoxy-Dglucose; LC3, microtubule-associated protein 1 light chain 3 beta; S/CO, signal to cutoff ratio determined by glucose concentrations and correlated to the steady state levels of HBV RNAs. However, the posttranscriptional control of HBV replication at the steps of assembly, release, and degradation by autophagy has emerged as relevant and effective mechanisms that require attention in future studies on the HBV life cycle. In a number of studies, the production of HBeAg has been considered as a marker of HBV replication. However, autophagy significantly promotes HBsAg and virion production but not HBeAg production, as shown in several early studies (Lin et al., 2017;Lin, Wu, Wang, Kemper, et al., 2019;Lin, Wu, Wang, Liu, et al., 2019). HBeAg production occurs in the endoplasmic reticulum-Golgi compartment and does not have a connection with autophagy. In this study, HBsAg but not HBeAg production was significantly regulated by glucose concentrations, again consistent with the major role of autophagy under these conditions.
AMPK stimulation is reported to inactivate mTOR by AMPK-mediated phosphorylation of both TSC2 and Raptor. ULK1 can subsequently interact with and be phosphorylated by AMPK to initiate autophagy (Corradetti, Inoki, Bardeesy, DePinho, & Guan, 2004;Gwinn et al., 2008;Kim et al., 2011;Nwadike et al., 2018). Consistent with this notion, AMPK activation occurred at low glucose concentration and in the presence of the GLUT inhibitor phloretin, along with inhibited mTOR activity, ULK1 phosphorylation, and autophagy induction.
Taken together, our results reveal a novel regulatory mechanism by which glucose supply regulates HBV replication.
Glucose is an essential nutrient and energy source in living organisms. Maintaining energy homeostasis is very important in mammalian physiology. Glycolysis is considered as a "central" carbon metabolic pathway because it is the backbone of several metabolic pathways and is pivotal for energy homeostasis (Akram, 2014;Dashty, 2013;Kornberg, 2000). Herein, HBV replication decreased after 2-DG treatment in HepG2.2.15 cells. This observation highlights the notion that HBV replication requires glycolysis in host cells. Treatment with 2-DG induced AMPK and Akt/mTOR activation, two processes that may regulate HBV replication in opposite ways. Activation of the Akt/mTOR signalling pathway negatively regulates HBV replication (Guo et al., 2007;Rawat & Bouchard, 2015;Xiang & Wang, 2018) and blocks glycolysis through the regulation of GLUT transport (Altomare & Khaled, 2012), whereas AMPK acts through downstream autophagic pathway to regulate HBV replication. Treatment with 2-DG strongly inhibited HBV replication, but promoted autophagy . We assume that other pathways in the downstream of AMPK and Akt/mTOR are also involved in the regulation of HBV replication. Previously, we tested several mTOR-related pathways and found that SREBP1 was also partly activated to regulate HBV replication (Lin et al., 2017). Therefore, a detailed analysis is needed to completely dissect the functions of different pathways and their relative contributions to the control of HBV replication. On the other hand, 2-DG may markedly inhibit glycolysis that is essential for viral replication (Gardner, Abcouwer, Losiewicz, & Fort, 2015;Maus et al., 2006;Muaddi et al., 2010). Our results in this study did not support the role of reduced nutrient supply in HBV suppression as the cellular protein synthesis was not affected by 2-DG. However, these questions remain to be answered in future studies.

| Detection of HBV gene expression and replication
The method for the detection of HBV progeny DNA in the culture supernatants has been described previously (Lin et al., 2017;Zhang et al., 2011). HBV RNA levels in cells were measured by real-time reverse transcriptase polymerase chain reaction (RT-PCR) assays (Qiagen, 204154) using the primers 5′-CCGTCTGTGCCTTCTCATCT-3′ (forward) and 5′-TAATCTCCTCCCCCAACTCC-3′ (reverse). The mRNA levels were normalized to the beta-actin mRNA level. These primers cover all four HBV RNAs as described previously (Zhang et al., 2011). The detection of intracellular HBsAg and HBeAg levels from cell lysates and both in the culture supernatants was performed by chemiluminescent microparticle immunoassay (; Abbott Laboratories, Chicago, IL, USA). HBV replicative intermediates (RIs) from intracellular core particles were extracted and detected by Southern blotting as described previously (Lin et al., 2017;J. Wu et al., 2007;Zhang et al., 2011). To extract HBV RIs from intracellular core particles, cells were lysed with lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM MgCl2, 1% NP-40) for 10 min. The cytoplasmic fraction was separated from the nuclear fraction by centrifugation. The supernatants were taken to the new Eppendorf tubes, mixed with 10 mM MgCl2 and 500 μg/ml DNaseI (Roche, Germany), and incubated for 1 hr at 37°C. The reaction was terminated using 25 mM EDTA. Sodium dodecyl sulfate and proteinase K (QIAGEN) were added separately and incubated for 2 hr at 56°C. Thereafter, the DNA was extracted with phenol/chloroform (1:1) and precipitated with isopropanol. The precipitated nucleic acid was resuspended in 15-μl TE buffer. The preparation containing HBV DNA was then subjected to electrophoresis on a 0.8% agarose gel, followed by blotting onto a Hybond-N+ membrane. Blots were hybridised with 32 P-labelled HBV DNA probes, which were prepared using a random priming labelling kit (GE Healthcare, RPN1633) in hybridization buffer (G-Biosciences, 786-160). The signals were visualised and analysed using a phosphoimager (Cyclon, Packard Instrument).

| Western blotting assays
The methods for preparing whole cell protein lysates and Western blotting have been described previously (Lin et al., 2017). Antibodies

| Immunofluorescence staining
Immunofluorescence staining was performed as described previously (Lin et al., 2017). Briefly, HepG2.2.15 cells were grown on coverslips and treated as indicated in each experiment. Then, the cells were fixed in 4% paraformaldehyde and permeabilised with 0.1% Triton X-100 and incubated with anti-LC3B antibodies, staining with Alexa Fluor 488-conjugated Goat anti-Rabbit IgG (H + L). The distribution of LC3B protein was observed with a Zeiss ELYRA PS.1 SIM/PAL-M/ STORM/TIRF and LSM710 microscope (Zeiss, Jena, Germany). The number of LC3B puncta in cells was quantified as described previously (Lin et al., 2017).

| Cell proliferation assay
Cell proliferation was determined by a cell counting Kit-8 assay kit (Sigma-Aldrich, 96992) according to the manufacturer's protocol.

| Lactate assay
Accumulation of lactate in the culture medium was measured by a lactate colorimetric/fluorometric assay kit (BioVision Inc, K607) according to the manufacturer's protocol.

| Statistical analyses
Data are shown in mean ± standard error of the mean. Statistical analyses were performed using Graph Pad Prism software version 7 (La Jolla, CA, USA). ANOVA with two-tailed Student's t test or by one-way ANOVA with a Tukey posttest was used to determine significant differences. Differences were considered statistically significant when p < .05. All experiments were repeated independently at least three times.