PGC-1α controls hepatitis B virus through nutritional signals

  1. Amir Shlomai,
  2. Nir Paran*, and
  3. Yosef Shaul†
  1. Department of Molecular Genetics, The Weizmann Institute of Science, Rehovot 76100, Israel

  1. Communicated by Bruce M. Spiegelman, Dana–Farber Cancer Institute, Boston, MA, September 7, 2006 (received for review August 1, 2006)

 
Next Section

Abstract

Hepatitis B virus (HBV) is a 3.2-kb DNA virus that replicates preferentially in the liver. Liver-enriched nuclear receptors (NRs) play a major role in the HBV life cycle, operating as essential transcription factors for viral gene expression. Notably, these NRs are also key players in metabolic processes that occur in the liver, serving as central transcription factors for key enzymes of gluconeogenesis, fatty acid β-oxidation, and ketogenesis. However, the association between these metabolic events and HBV gene expression is poorly understood. Here we show that peroxisome proliferator-activated receptor-γ coactivator 1α (PGC-1α), a major metabolic regulator and a coactivator of key gluconeogenic genes, robustly coactivates HBV transcription. We further demonstrate that the liver-enriched NR hepatocyte nuclear factor 4α that binds HBV plays an important role in this process. Physiologically, we show that a short-term fast that turns on the gluconeogenic program robustly induces HBV gene expression in vivo. This induction is completely reversible by refeeding and depends on PGC-1α. We conclude that HBV is tightly regulated by changes in the body's nutritional state through the metabolic regulator PGC-1α. Our data provide evidence for nutrition signaling to control viral gene expression and life cycle and thus ascribe to metabolism an important role in virus–host interaction.

Hepatitis B virus (HBV) is a 3.2-kb DNA virus that possesses four major ORFs and five promoters. These promoters control the synthesis of the six major viral transcripts, designated long-X RNA (lxRNA) and short-X RNA (sxRNA), both initiated at the X promoter and encode for the X protein (1), the pregenomic (pg) and the precore (pc) transcripts, and the preS1 and preS2/S transcripts, encoding for the viral surface proteins (2). Two enhancers, named enhancer I (EnhI) and enhancer II (EnhII), have been identified in the HBV genome (3–5). However, the activity of EnhII depends on a functional EnhI (1).

Liver-enriched nuclear receptors (NRs) play a central role in the regulation of the HBV transcriptional program by binding to both EnhI and EnhII (1, 6–8). Interestingly, these liver-enriched NRs are also central mediators of metabolic processes in the liver. A prominent example for such a process is gluconeogenesis, which is required for the maintenance of a normal blood glucose level during starvation. NRs, such as glucocorticoid receptor (GR), hepatocyte nuclear factor 4α (HNF4α), and peroxisome proliferator-activated receptors (PPARs), bind and activate the promoter of the phosphoenolpyruvate carboxykinase (PEPCK) gene, a key gluconeogenic enzyme. Notably, NRs are also involved in fatty acid β-oxidation, ketogenesis, and bile-acid homeostasis, other essential metabolic events that occur in the liver (9, 10).

Peroxisome proliferator-activated receptor-γ coactivator 1α (PGC-1α) was originally identified in brown fat as a PPARγ-interacting protein, with the ability to coactivate PPARγ in response to external stimuli such as cold (11). In vitro and in vivo studies have demonstrated that PGC-1α interacts with and coactivates a variety of NRs, such as GR, HNF4α, and PPARs (12) in various tissues. The induction of PGC-1α in those tissues usually links external stimuli, such as cold in brown adipocytes and mechanical stress in muscle, to the activation of central metabolic and energy-related genes (13).

A major breakthrough was the discovery that PGC-1α has a central role in the regulation of gluconeogenesis in the liver, by coactivation of key enzymes in the gluconeogenesis pathway, through binding to NRs such as HNF4α and GR (14, 15). The notion that PGC-1α is sufficient to activate nearly all of the components of hepatic fasting response, including gluconeogenesis, fatty acid β-oxidation, and ketogenesis by coactivating liver-enriched NRs (12), establishes the role of PGC-1α as an important metabolic regulator, reacting to external nutritional stimuli.

The important role of liver-enriched NRs in HBV gene expression and the fact that this virus exclusively replicates in the liver led us to investigate a possible association between major metabolic processes taking place in the liver and HBV gene expression.

Here we show that nutritional signals, such as a short-term fast that turns on the gluconeogenic program, robustly induce HBV gene expression in vivo. This induction is completely reversible by refeeding and depends on PGC-1α, which is shown here to serve as a coactivator of HBV transcription.

Previous SectionNext Section

Results

PGC-1α Coactivates HBV Gene Expression.

PGC-1α is a strong transcription coactivator of the genes of the key gluconeogenic enzymes PEPCK and G6Pase, acting through the liver-enriched NR HNF4α (15). PGC-1α was also reported to bind and to coactivate PPARα and GR, as well as the forkhead transcription factor FOXO1 (FKHR) (12, 14). Because HBV is transcriptionally regulated by HNF4α and the heterodimer PPARα/RXR, which bind to NR response elements residing in its two enhancers (Fig. 1 A), we asked whether PGC-1α coactivates HBV. For this aim, HepG2 cells were cotransfected with HBV and PGC-1α expression plasmids. Northern blot analysis revealed a significant increase in all HBV transcripts in a PGC-1α dose-dependent manner. This effect was abolished by knocking-down PGC-1α with a specific siRNA (Fig. 1 B). Western blot analysis of the HBV core protein confirmed that HBV up-regulation was maintained at the protein level (Fig. 1 C). To quantify the effect of PGC-1α on HBV, HepG2 cells were transfected with the HBV-luciferase (HBV-Luc) plasmid, a “gutless” HBV construct containing a luciferase ORF under the HBV core promoter (Fig. 1 D Upper). Knocking-down PGC-1α reduced luciferase activity by ≈30%, attributing a significant role for the endogenous PGC-1α in HBV gene expression in the basal state. Overexpression of PGC-1α resulted in a ≈3-fold increase in luciferase activity, an effect that was significantly blocked by the PGC-1α siRNA. Notably, the effect of PGC-1α siRNA was comparable to that of siRNA targeting luciferase, whereas the control siRNA targeting the core region, a region which is deleted in the “gutless” HBV-Luc construct, had no effect (Fig. 1 D Lower).

Fig. 1.
View larger version:
Fig. 1.

Overexpression of PGC-1α results in HBV coactivation. (A) A schematic representation of the liver-enriched NRs binding sites on EnhI and on EnhII of the HBV genome. RXR, retinoid X receptor; Cp, core promoter; PCp, precore promoter; Xp, X promoter. (B) HepG2 cells were seeded on 10-cm dishes and were transfected with 1.3x wtHBV plasmid (12 μg) together with increasing amounts (2–8 μg) of PGC-1α plasmid, with or without pSUPER PGC-1α plasmid. Three days after transfection, cells were harvested and RNA was analyzed by Northern blot for HBV transcripts. 18S and 28S rRNAs were analyzed for equal loading control. pg/pc, pregenomic and precore HBV RNAs; Surface, surface RNA; X, X RNA. (C) The same transfection protocol as in B was performed, this time with a HA-tagged PGC-1α construct. Three days after transfection, cells were harvested for protein analysis (Core, HBV core protein; IB, immunoblot). (D Upper) A schematic representation of the 1.3x HBV-Luc construct [P(A)S, poly (A) signal; Cp, core promoter; RV, EcoRV restriction site]. (D Lower) HepG2 cells were cotransfected with 1.3x HBV-Luc construct with the indicated plasmids. Sixty hours after transfection, cells were harvested and analyzed for luciferase activity. Results are presented as folds of activation after normalization to Renilla. The experiment was repeated three times in triplicate. HBc, HBV core protein; *, P < 0.05 (calculated by using Student's t test). (E) Huh7 cells were transfected with 10 μg of 1.3x wtHBV plasmid together with increasing amounts (1–6 μg) of PGC-1α plasmid with or without pSUPER PGC-1α (4–10 μg). Six days after transfection, cells were harvested and analyzed for HBV replicative-intermediates (RC, relaxed circular DNA; DL, double-stranded linear DNA; SS, single-stranded DNA).


Next, the effect of overexpressed PGC-1α on HBV replication was investigated. Analysis of the viral replicative forms from the cytoplasm of HBV-transfected cells revealed that in both Huh7 and HepG2 cells (Fig. 1 E and data not shown), overexpression of PGC-1α resulted in a significant increase in HBV replication. This effect was abolished by knocking-down PGC-1α. These data indicate that overexpression of the metabolic regulator PGC-1α robustly coactivates HBV transcription.

Induced Endogenous PGC-1α Supports HBV Transcription.

To investigate whether the endogenous PGC-1α controls HBV gene expression, we treated HepG2 2.2.15 cells, which constitutively express HBV (16), with forskolin, a cAMP inducer, and dexamethasone. These compounds are known for their synergistic effect on PGC-1α induction through the CREB response element and the glucocorticoid response element, respectively, both residing on the PGC-1α promoter (15). Indeed, treatment with forskolin and dexamethasone resulted in a robust induction of both PGC-1α and HBV transcripts (Fig. 2 A). To investigate whether HBV induction by dexamethasone and forskolin is PGC-1α-dependent, HepG2 cells were transfected with HBV-expressing plasmid, with or without siRNA targeting PGC-1α. As expected, treatment of HBV-transfected cells with forskolin alone, and to a greater degree in combination with dexamethasone, resulted in a robust induction of both PGC-1α and HBV transcription. Remarkably, HBV induction was PGC-1α-dependent because PGC-1α knockdown completely blocked the effect of dexamethasone and forskolin on HBV expression (Fig. 2 B). Similar results were obtained with the HBV-Luc reporter. Here again, up-regulation of the luciferase activity was abolished by knocking-down PGC-1α (Fig. 2 C). Overall, these results strongly indicate that the endogenous PGC-1α supports HBV transcription.

Fig. 2.
View larger version:
Fig. 2.

Induction of the endogenous PGC-1α results in HBV coactivation. (A Left) HepG2 and HepG2 2215 cells were treated as indicated. Fourteen hours later, cells were harvested and RNA was analyzed by semiquantitative RT-PCR. (A Right) RNA from the same experiment was analyzed by real-time RT-PCR. (G2-HepG2, 2215-HepG2 2215 cell lines). (B) HepG2 cells were cotransfected with 1.3x wtHBV construct and the indicated plasmids. Twenty-four hours after transfection, treatment was performed as indicated. Fourteen hours later, cells were harvested and analyzed by semiquantitative RT-PCR. (C) HepG2 cells were cotransfected with 1.3xHBV-Luc construct and the indicated plasmids. After transfection (≈40 h), treatment was performed as indicated. Fourteen hours later, cells were harvested and analyzed for luciferase activity. The experiment was repeated three times in triplicate. Forsk, forskolin; Dex, dexamethasone.


HNF4α Mediates Transcription Coactivation of HBV by PGC-1α.

HNF4α is essential for PEPCK transcription coactivation by PGC-1α (17). To investigate whether HNF4α is also required for HBV transcription coactivation by PGC-1α, we used the Chang (CCL-13) cells. In these cells, HBV EnhII activity depends on ectopically expressed HNF4α (1). Chang cells transfected with HBV predominantly express the 3.9-kb long early (le)RNA and the 0.7-kb X RNA, transcripts that are regulated by EnhI, but not the EnhII-regulated pg (pregenomic) and pc (precore) RNA species (Fig. 3 A). Consequently, the HBV core protein encoded by these transcripts was not produced to the detectable level (Fig. 3 B). As expected, ectopic expression of HNF4α resulted in the restoration of EnhII-directed HBV transcripts and the HBV core protein (Fig. 3 A and B). Quantitative analysis revealed that HNF4α alone increased luciferase activity of the HBV-Luc construct 9-fold. Interestingly, cotransfection of PGC-1α and HNF4α had a synergistic effect, resulting in an up to 28-fold increase in luciferase activity (Fig. 3 C). Notably, PGC-1α alone, without HNF4α, was sufficient to partially restore EnhII-directed transcriptional activity (Fig. 3 A–C). Overall, these results suggest that HNF4α is an important but not the sole target for HBV coactivation by PGC-1α.

Fig. 3.
View larger version:
Fig. 3.

Transcription coactivation of HBV by PGC-1α is mediated through HNF4α. (A) Chang cells were transfected with 10 μg of 1.3x wtHBV plasmid with or without HNF4α-expressing plasmid (200 ng) and PGC-1α-expressing plasmid (2–6 μg). Two days after transfection, cells were harvested and RNA was extracted and analyzed by Northern blot for HBV transcripts. HBV-transfected HepG2 cells were used as a positive control. Analysis of the GAPDH transcripts was used for equal loading control. pg/pc, pregenomic and precore HBV RNAs; le, long early RNA; Surface, surface RNA; X, X RNA. (B) Chang cells were transfected as indicated. Two days after transfection, cells were harvested for protein analysis. IB, immunoblot. (C) Chang cells were transfected with 1.3x HBV-Luc construct together with the indicated plasmids. Two days after transfection, cells were harvested and analyzed for luciferase activity. Results are presented as folds of activation after normalization to Renilla (open columns, no PGC-1α; filled columns, increasing amounts of PGC-1α).


Starvation Induces HBV Gene Expression in Vivo.

Upon starvation, PGC-1α expression in the liver is robustly induced, resulting in a full activation of the gluconeogenic program (15). Given the ability of PGC-1α to coactivate HBV, we asked whether HBV transcription is affected by changes in the nutritional status in vivo. We used the HBV-Luc construct and injected it to mice using the hydrodynamic injection method (18). Two days after injection, all animals showed a clear basal luciferase activity in their livers (Fig. 4 A Top and B). Interestingly, a significant increase in luciferase activity of ≈2-fold was observed in all mice already after a 7-h fast (Fig. 4 A Middle and B), whereas no increase was detected in ad libitum-fed control mice. Importantly, upon refeeding for 12 h, the luciferase activity in the 7-h fasting mice returned to the baseline level (Fig. 4 A Bottom and B), suggesting that the positive effect on HBV expression was fasting-dependent. Overall, these results strongly indicate that HBV is robustly induced during starvation and that this induction is reversible by refeeding.

Fig. 4.
View larger version:
Fig. 4.

Fasting-induced activation of HBV expression in vivo. (A) Mice were tail-injected with 6 μg of HBV-Luc plasmid. Forty-eight hours later, mice were divided into two groups: a control group (n = 5) that was allowed continuous free feeding and an experimental group (n = 8) that was subjected to a 7-h fast and to a subsequent 12-h refeeding. In vivo luciferase analysis of all mice was performed at baseline, after 7 h, and after an additional 12 h. Shown are the luciferase images of three representive animals from the experimental group at the indicated times. (B) A quantitative analysis of the experiment described in A. Red bars indicate the experimental (Exp) group and blue bars indicate the control (Cont) group. Error bars indicate SD. *, P = 0.002 (calculated by using Student's t test).


Starvation-Induced HBV Gene Expression in Vivo Is PGC-1α-Dependent.

Next, we investigated whether starvation-induced activation of HBV is PGC-1α-dependent. Two days after the injection of the wild-type (wt) HBV, mice were put on either a 14-h fast or were allowed to feed freely. Analysis of livers from 14-h fasting mice revealed a significant increase in both PGC-1α and HBV transcription as compared with ad libitum-fed mice (Fig. 5 A and B). Western blot analysis revealed that PGC-1α induced the expression of the HBV core protein, as well (Fig. 5 C). Remarkably, the induction in HBV expression was significantly abolished when PGC-1α siRNA was coinjected (Fig. 5 A–C).

Fig. 5.
View larger version:
Fig. 5.

Fasting-induced coactivation of HBV in vivo is PGC-1α-dependent. (A) Semiquantitative RT-PCR analysis of total RNA extracted from livers of mice injected with the indicated constructs and subjected to either 14-h fast or normal feeding. (B) The same experiment described in Fig. 5 A was performed. Total RNA extracted from mice livers was subjected to a Northern blot analysis. pg/pc, pregenomic and precore HBV RNAs; HBsAg, HBV surface antigen-encoding RNAs. (C) The same experiment described in Fig. 5 A was performed. Protein was extracted and analyzed by Western blotting. IB, immunoblot. (D) A ChIP analysis of livers from either fasting or fed mice using either anti-PGC-1α antibody or only beads as a control. The immunoprecipitated chromatin was amplified by primers spanning the HNF4α-binding site region in EnhII (Right) or another unrelated region along the HBV genome (Left). Numbers on the arrows indicate nucleotide number (5′ to 3′, EcoRI site = nt 0) from which the primers originate. Cp, core promoter; PCp, precore promoter; RXRα, retinoid X receptor α; pX, X promoter.


To further establish the direct role of PGC-1α in coactivation of HBV transcription under starvation, we looked for HBV DNA-associated PGC-1α. To this end we performed ChIP experiments using antibodies against PGC-1α. As shown in Fig. 5 D, upon starvation, PGC-1α is preferentially recruited to the HNF4α binding site on EnhII. As a negative control we analyzed another region of the HBV, located just upstream to EnhI. This region did not recruit PGC-1α, neither in the fed nor in the fasting state (Fig. 5 D Left). Overall, these results suggest that upon starvation, PGC-1α is recruited to associate with the HBV EnhII and to coactivate HBV transcription.

Previous SectionNext Section

Discussion

In the present study, we investigated the interplay between metabolic processes in the liver that are largely mediated through NRs in response to nutritional cues (13, 19–21) and HBV life cycle. Our study provides solid evidence that links nutritional signals to HBV expression, both in cell culture and in animal model, and offers a molecular mechanism for this linkage.

Thanks to the recently introduced high-pressure injection technique (18), we were able to express the HBV reporter construct in mice livers and monitor the transcriptional activity of HBV under certain nutritional conditions, in “real time.” These experiments highly suggest that HBV transcription is intimately coupled to the body's nutritional state, being robustly up-regulated during a short-term starvation, an effect that is completely reversible upon refeeding. At the molecular level, we present a line of evidence linking the metabolic coactivator PGC-1α that is robustly induced during starvation and turns on the gluconeogenic program (15), to the regulation of HBV gene expression. In tissue culture, we show that PGC-1α coactivates HBV transcription and significantly increases its replication. Furthermore, in a mouse model we show that upon starvation, PGC-1α is recruited to the HBV genome and robustly coactivates its transcription. This effect completely depends on PGC-1α, because knocking-down PGC-1α, or alternatively, in ad libitum-fed mice, where PGC-1α level in the liver is low, PGC-1α is not recruited to the HBV genome and does not coactivate its transcription. Interestingly, HBV behaves like the key gluconeogenic genes, such as PEPCK and G6Pase, although unlike these genes a certain level of HBV coactivation by PGC-1α can take place independently of HNF4α. We speculate that other yet unidentified factor(s) that are potential targets for PGC-1α may bind HBV and mediate its coactivation by PGC-1α.

Being an integral part of the metabolic machinery, HBV represents a novel mechanism for virus–host interaction that includes a tight interplay between viral expression and major metabolic processes in its host cell. We speculate that by coupling its gene expression with external metabolic signals, the virus ensures a continuous supply of essential transcription factors as long as the hepatocyte is viable and metabolically active. Furthermore, tuning its gene expression according to changes in environmental stimuli enables the virus to take advantage of certain physiological states for its own needs. Upon starvation, or during other stress conditions in which the immune response naturally wanes, the virus can “afford” a relatively robust gene expression, without facing the risk of an effective immune response.

From the clinical aspect, our results may pave the way toward a better understanding of the dynamic and the natural history of HBV infection. Fundamental questions regarding the fluctuating nature of the disease (22), the not-yet-fully understood mechanism behind HBV flare-ups after stress conditions and immunosuppression (23), and the reason for the geographical diversity and the much more aggressive course of the disease in certain countries (24), could be now approached from a different point of view. Keeping in mind that in terms of regulation and response to nutritional stimuli, HBV is very similar to metabolic genes, one can attribute certain dynamic changes in the natural history of HBV not only to certain mutations or genotypic diversity of the virus (24) but also to changes in environmental/nutritional conditions, or alternatively, to preexisting pathologic states that affect the host's metabolism. For example, in patients with certain mutations in the maturity-onset diabetes of the young 1 (MODY1) gene, HNF4α loses its ability to bind DNA and to transactivate transcription, resulting in overt diabetes during childhood or young adulthood (25, 26). Given the synergistic effect of HNF4α and PGC-1α on HBV transcriptional activity, it will be interesting to investigate the natural history of HBV infection in patients who suffer from MODY1 and who are concurrently infected with HBV. Given the fact that PGC-1α coactivates HBV to a certain level even in the absence of HNF4α (see Fig. 3), we speculate that in those patients, HBV gene expression will be completely suppressed in the fed state but will be only partially suppressed during starvation, when the PGC-1α level is elevated.

In summary, we show that nutritional signals of fed versus fasting states regulate HBV gene expression through the metabolic coactivator PGC-1α. Our results suggest a previously unrecognized mechanism of virus–host interaction and identifies PGC-1α as a central player in HBV life cycle, making it a potential target for future anti-HBV therapy.

Previous SectionNext Section

Materials and Methods

Cell Culture and Treatments.

Cells were maintained in DMEM with 8% FBS. Cells were seeded at ≈60% confluence 4–6 h before transfection, which was carried out by the CaPO4 method. For hormonal treatments, cells were maintained in DMEM supplemented with 8% charcoal-treated serum for 3 days before transfection. Treatments were done with 10 μM of forskolin (Sigma, St. Louis, MO; catalog no. F6886) and 100 nM dexamethasone (Sigma; catalog no. D8893).

Plasmids.

pSport-PGC-1α plasmid was kindly provided by B. M. Spiegelman (Dana–Farber Cancer Institute, Boston, MA). To express siRNA targeting PGC-1α, we used a pSUPER PGC-1α plasmid that was created as described in ref. 27 using the following primers: 5′-GATCCCCGGTGGATTGAAGTGGTGTATTCAAGAGATACACCACTTCAATCCACCTTTTTGGAAA-3′ (sense) and 5′-AGCTTTTCCAAAAAGGTGGATTGAAGTGGTGTATCTCTTGAATACACCACTTCAATCCACCGGG-3′ (antisense). The 1.3x HBV-Luc and the 1.3x wtHBV constructs were described in ref. 1.

Animal Experiments.

Seven-week-old female BALB/C mice were used for animal experiments, which were performed according to the institutional guidelines. For the whole HBV analysis, mice were injected on day 0 with 1.5 ml of normal saline (0.9% NaCl) by using the high-pressure hydrodynamic method (18), containing either 1.3x wtHBV plasmid with empty pSUPER vector or 1.3x wtHBV together with pSUPER PGC-1α plasmid. Mice injected with normal saline alone were used as controls. Fifty-five hours after injection, mice were divided into two groups, which were let for either free feeding or a fast (free access to water was allowed). The animals were killed 14 h later, and liver was collected and immediately was put in liquid nitrogen for further analysis. For in vivo imaging experiments, mice were injected with 1.3x HBV-Luc construct and analysis was performed 48 h later. In this experiment, we used a shorter starvation time (7 h), which has been shown to be sufficient for PGC-1α induction (15), to avoid a significant ATP depletion that might affect the luciferase activity. Immediately before in vivo luciferase analysis, animals were anesthetized with isoflurane, and 100 μl of 3 mg/ml d-Luciferin (Xenogen, Alameda, CA) was injected i.p. Visualization of luciferase activity was performed by using the IVIS 100 machine (Xenogen). Data analysis including quantification was performed by using the Living Image software (Xenogen). Luciferase activity of every animal from two independent experiments was quantified at all time points and was expressed as the ratio relative to the baseline activity.

Protein Analysis.

Proteins were extracted from cells by using RIPA buffer [150 mM NaCl/1% Nonidet P-40 (vol/vol)/0.5% sodium deoxycholate (vol/vol)/0.1% SDS (vol/vol)/50 mM Tris·HCl (pH 8)/1 mM DTT/1 μg/ml each leupeptin, aprotinin, and pepstatin (Sigma mixture)] and were analyzed on 10% polyacrylamide-SDS gel. Antibodies used in our experiments included monoclonal mouse anti-HBV core antigen (clone 22, diluted 1:5,000), monoclonal mouse anti-HA [BabCO (Richmond, CA) and COVANCE (Richmond, CA), diluted 1:2,000], goat polyclonal antibodies against HNF-4α (Santa Cruz Biotechnology, Santa Cruz, CA; catalog no. 6556), IgGs of mouse anti-β-tubulin (clone no. TUB2.1; Sigma; diluted 1:10,000), and monoclonal mouse anti-actin (clone AC-40; Sigma; diluted 1:10,000). As secondary antibody we used goat anti-mouse antibody conjugated with HRP (Jackson ImmunoResearch, West Grove, PA; diluted 1:10,000), or donkey anti-goat IgG antibodies (Jackson ImmunoResearch; diluted 1:10,000).

Isolation and Analysis of Viral RNA.

Total RNA was extracted from either transfected cells or from livers of mice (after homogenization) by TRI REAGENT (MRC, Inc., Cincinnati, OH). RNA from each sample (10 μg) was analyzed by Northern blotting as described (1). HBV transcripts were detected by using radioactive probes prepared from the X gene region, labeled by random priming protocol. RT-PCR analysis was performed by using SuperScript II reverse transcriptase (Invitrogen Life Technologies, Carlsbad, CA) according to the manufacturer's instructions. PCR amplification was performed with the ReadyMix master mix (Abgene, Surrey, U.K.) by using the following primers: HBV (transfected), 5′-CTCAGCTCTGTATCGAGAAGCC-3′ (sense), 5′-CAGTGAGAGGGCCCACAAATTG-3′ (antisense); HBV (HepG2 2215), 5′-CTCAGCTCTGTATCGGGAAGCC-3′ (sense), 5′-CTGTGAGTGGGCCTACAAACTG-3′ (antisense); PGC-1α (mouse), 5′-GAAGTGGTGTAGCGACCAATC-3′ (sense), 5′-AATGAGGGCAATCCGTCTTCA-3′ (antisense); PGC-1α (human), 5′- TCCTCACAGAGACACTAGACAG-3′ (sense), 5′- CTGGTGCCAGTAAGAGCTTCT-3′ (antisense); PEPCK (mouse), 5′-CTGCATAACGGTCTGGACTTC-3′ (sense), 5′-CAGCAACTGCCCGTACTCC-3′ (antisense); Actin (mouse), 5′-ACACTGTGCCCATCTACGAG-3′ (sense), 5′-AGGGGCCGGACTCGTCATACT-3′; Actin (human), 5′-ACCGCGAGAAGATGACCCAG-3′ (sense), 5′-CCATCTCGTTCTCGAAGTCCA-3′ (antisense).

Real-time PCR was performed by using the LightCycler 480 (Roche, Gipf-Oberfrick, Switzerland). Results were normalized to GAPDH mRNA levels.

HBV Replication Intermediates Assay.

Analysis of HBV replication intermediates was performed as described in ref. 28. Briefly, 5 days after transfection with HBV-expressing plasmids, cells underwent cytoplasmic extraction in the presence of DNaseI to eliminate the transfected DNA. The lysates were subsequently treated with proteinase K, and after phenol extraction, the encapsidated viral DNA was ethanol precipitated. Total DNA was fractioned on a 1.7% agarose-TAE gel, followed by denaturation and Southern blotting to a Hybond-N nylon membrane (Amersham, Piscataway, NJ). Viral DNA was detected by hybridization with a 32P random primed HBV probe.

ChIP Analysis.

Liver tissue (≈30 mg) from either fed or fasting mice was taken and cross-linked with 1% formaldehyde for 15 min with subsequent adding of glycine to a final concentration of 0.125 M to stop cross-linking. The tissue was washed once with cold PBS and was later disaggregated with a Daunce homogenizer. The subsequent steps were performed according to a published protocol (29). Rabbit polyclonal anti-PGC-1α antibody (6 μg; SC 13067; Santa Cruz Biotechnology) was used for immunoprecipitation. The immunoprecipitated chromatin was amplified by using HBV-specific pairs of primers as follows: EnhII, 5′-CAATGTCAACGACCGACCTTGAG-3′ (sense), 5′-GGCAGAGGTGAAAAAGTTGCATG-3′ (antisense); negative control primers, 5′-GAAGTTGGGGAACTTTGCCACAG-3′ (sense), 5′-GGAGCAGCAAAGCCCAAAAGAC-3′ (antisense).

Previous SectionNext Section

Acknowledgments

We thank Dr. Vyacheslav (Slava) Kalchenko for his excellent help with the in vivo imaging experiments and Sylvia Budilovsky for her technical help. This work was partially supported by the Sylvia and David Salzberg Cancer Research Fund.

Previous SectionNext Section

Footnotes

  • †To whom correspondence should be addressed. E-mail: yosef.shaul{at}weizmann.ac.il

  • Author contributions: A.S. and Y.S. designed research; A.S. and N.P. performed research; A.S. and Y.S. analyzed data; and A.S. and Y.S. wrote the paper.

  • ↵*Present address: Israel Institute for Biological Research, P.O. Box 19, 74100 Ness-Ziona, Israel.

  • The authors declare no conflict of interest.

  • Abbreviations:
    HBV,
    Hepatitis B virus;
    NR,
    nuclear receptor;
    GR,
    glucocorticoid receptor;
    PPAR,
    peroxisome proliferator-activated receptor;
    HNF4α,
    hepatocyte nuclear factor 4α;
    PEPCK,
    phosphoenolpyruvate carboxykinase;
    PGC-1α,
    peroxisome proliferator-activated receptor-γ coactivator 1α;
    HBV-Luc,
    HBV-luciferase;
    wt,
    wild type.
Previous Section
 

References

  1. ↵
    1. Doitsh G,
    2. Shaul Y
    (2004) Mol Cell Biol 24:1799–1808.
    Abstract/FREE Full Text
  2. ↵
    1. Seeger C,
    2. Mason WS
    (2000) Microbiol Mol Biol Rev 64:51–68.
    Abstract/FREE Full Text
  3. ↵
    1. Hu KQ,
    2. Siddiqui A
    (1991) Virology 181:721–726.
    CrossRefMedlineISI
  4. ↵
    1. Antonucci TK,
    2. Rutter WJ
    (1989) J Virol 63:579–583.
    Abstract/FREE Full Text
  5. ↵
    1. Yuh CH,
    2. Ting LP
    (1990) J Virol 64:4281–4287.
    Abstract/FREE Full Text
  6. ↵
    1. Raney AK,
    2. Johnson JL,
    3. Palmer CN,
    4. McLachlan A
    (1997) J Virol 71:1058–1071.
    Abstract
  7. ↵
    1. Tang H,
    2. McLachlan A
    (2001) Proc Natl Acad Sci USA 98:1841–1846.
    Abstract/FREE Full Text
  8. ↵
    1. Yu X,
    2. Mertz JE
    (2001) J Virol 75:11354–11364.
    Abstract/FREE Full Text
  9. ↵
    1. Beaven SW,
    2. Tontonoz P
    (2006) Annu Rev Med 57:313–329.
    CrossRefMedlineISI
  10. ↵
    1. Hall RK,
    2. Sladek FM,
    3. Granner DK
    (1995) Proc Natl Acad Sci USA 92:412–416.
    Abstract/FREE Full Text
  11. ↵
    1. Puigserver P,
    2. Wu Z,
    3. Park CW,
    4. Graves R,
    5. Wright M,
    6. Spiegelman BM
    (1998) Cell 92:829–839.
    CrossRefMedlineISI
  12. ↵
    1. Lin J,
    2. Handschin C,
    3. Spiegelman BM
    (2005) Cell Metab 1:361–370.
    CrossRefMedlineISI
  13. ↵
    1. Finck BN,
    2. Kelly DP
    (2006) J Clin Invest 116:615–622.
    CrossRefMedlineISI
  14. ↵
    1. Puigserver P,
    2. Rhee J,
    3. Donovan J,
    4. Walkey CJ,
    5. Yoon JC,
    6. Oriente F,
    7. Kitamura Y,
    8. Altomonte J,
    9. Dong H,
    10. Accili D,
    11. Spiegelman BM
    (2003) Nature 423:550–555.
    CrossRefMedline
  15. ↵
    1. Yoon JC,
    2. Puigserver P,
    3. Chen G,
    4. Donovan J,
    5. Wu Z,
    6. Rhee J,
    7. Adelmant G,
    8. Stafford J,
    9. Kahn CR,
    10. Granner DK,
    11. Newgard CB,
    12. Spiegelman BM
    (2001) Nature 413:131–138.
    CrossRefMedline
  16. ↵
    1. Sells MA,
    2. Chen M-L,
    3. Acs G
    (1987) Proc Natl Acad Sci USA 84:1005–1009.
    Abstract/FREE Full Text
  17. ↵
    1. Rhee J,
    2. Inoue Y,
    3. Yoon JC,
    4. Puigserver P,
    5. Fan M,
    6. Gonzalez FJ,
    7. Spiegelman BM
    (2003) Proc Natl Acad Sci USA 100:4012–4017.
    Abstract/FREE Full Text
  18. ↵
    1. Lewis DL,
    2. Wolff JA
    (2005) Methods Enzymol 392:336–350.
    MedlineISI
  19. ↵
    1. Glass CK
    (2006) J Clin Invest 116:556–560.
    CrossRefMedlineISI
  20. ↵
    1. Semple RK,
    2. Chatterjee VKK,
    3. O'Rahilly S
    (2006) J Clin Invest 116:581–589.
    CrossRefMedlineISI
  21. ↵
    1. Zelcer N,
    2. Tontonoz P
    (2006) J Clin Invest 116:607–614.
    CrossRefMedlineISI
  22. ↵
    1. McMahon BJ
    (2005) Semin Liver Dis 25:3–8.
    CrossRefMedlineISI
  23. ↵
    1. Shibolet O,
    2. Ilan Y,
    3. Gillis S,
    4. Hubert A,
    5. Shouval D,
    6. Safadi R
    (2002) Blood 100:391–396.
    Abstract/FREE Full Text
  24. ↵
    1. Zeng G,
    2. Wang Z,
    3. Wen S,
    4. Jiang J,
    5. Wang L,
    6. Cheng J,
    7. Tan D,
    8. Xiao F,
    9. Ma S,
    10. Li W,
    11. et al.
    (2005) J Viral Hepat 12:609–617.
    CrossRefMedlineISI
  25. ↵
    1. Navas MA,
    2. Munoz-Elias EJ,
    3. Kim J,
    4. Shih D,
    5. Stoffel M
    (1999) Diabetes 48:1459–1465.
    Abstract
  26. ↵
    1. Stoffel M,
    2. Duncan SA
    (1997) Proc Natl Acad Sci USA 94:13209–13214.
    Abstract/FREE Full Text
  27. ↵
    1. Brummelkamp TR,
    2. Bernards R,
    3. Agami R
    (2002) Science 296:550–553.
    Abstract/FREE Full Text
  28. ↵
    1. Amir Shlomai YS
    (2003) Hepatology 37:764–770.
    CrossRefISI
  29. ↵
    1. Ainbinder E,
    2. Revach M,
    3. Wolstein O,
    4. Moshonov S,
    5. Diamant N,
    6. Dikstein R
    (2002) Mol Cell Biol 22:6354–6362.
    Abstract/FREE Full Text
« Previous | Next Article »Table of Contents

This Article

  1. Published online before print October 16, 2006, doi: 10.1073/pnas.0607837103 PNAS October 24, 2006 vol. 103 no. 43 16003-16008
  1. AbstractFree
  2. Figures Only
  3. » Full Text
  4. Full Text (PDF)

Classifications

Navigate This Article

About Our New Site Design >>