Recently, we demonstrated that the anti-viral drug ribavirin (RBV) had the ability to suppress lipogenesis through down-regulation of retinoid X receptor α (RXRα) under the control of the intracellular GTP-level and AMP-activated protein kinase-related kinases, especially microtubule affinity regulating kinase 4 (MARK4). RXRα-overexpression attenuated but did not abolish lipogenesis suppression by RBV, implying that additional factor(s) were involved in this suppressive effect. In the present study, we found that the protein level, but not the mRNA level, of CCAAT/enhancer-binding protein α (C/EBPα) was down-regulated by RBV in hepatic cells. Treatment with proteasome inhibitor attenuated RBV-induced down-regulation of C/EBPα, suggesting that RBV promoted degradation of C/EBPα protein via the ubiquitin–proteasome pathway. Depletion of intracellular GTP through inosine monophosphate dehydrogenase inhibition by RBV led to down-regulation of C/EBPα. In contrast, down-regulation of C/EBPα by RBV was independent of RXRα and MARK4. Knockdown of C/EBPα reduced the intracellular neutral lipid levels and the expression of genes related to the triglyceride (TG) synthesis pathway, especially glycerol-3-phosphate acyltransferase, mitochondrial (GPAM), which encodes the first rate-limiting TG enzyme. Overexpression of C/EBPα yielded the opposite results. We also observed that RBV decreased GPAM expression. Moreover, overexpression of GPAM attenuated RBV-induced reduction in the intracellular neutral lipid levels. These data suggest that down-regulation of C/EBPα by RBV leads to the reduction in GPAM expression, which contributes to the suppression of lipogenesis. Our findings about the mechanism of RBV action in lipogenesis suppression will provide new insights for therapy against the active lipogenesis involved in hepatic steatosis and hepatocellular carcinomas.
Ribavirin (RBV) is a synthetic guanosine analogue and has been used as an anti-viral drug, specifically for patients with chronic hepatitis C. Numerous direct-acting antivirals (DAAs) have also been developed and approved for the treatment of chronic hepatitis C patients [1,2]. More than 90% of patients with chronic hepatitis C achieve a sustained virological response by DAA treatments with or without RBV. Nonetheless, RBV is considered to have various advantages, such as a shorter treatment time and a reduction in viral breakthrough or relapse, compared with the use of DAA treatments alone .
Previous studies have proposed various mechanisms to explain the action of RBV against hepatitis C virus (HCV): the inhibition of the RNA-dependent RNA polymerase activity of HCV NS5B, the induction of mutagenesis into the HCV RNA genome, the induction of interferon-stimulated genes, the depletion of intracellular GTP via inhibition of inosine monophosphate dehydrogenases (IMPDHs) activity, and the immunomodulation of the switching of the Th cell phenotype from type 2 to type 1 [4,5]. In a study using RBV-sensitive HCV RNA-replicating cells, we previously demonstrated that inhibition of IMPDHs activity by RBV caused GTP depletion in cells, which resulted in the inhibition of HCV RNA replication [6,7]. Furthermore, we identified adenosine kinase (ADK), which mono-phosphorylates RBV to a metabolically active form, as the determinant host factor for RBV sensitivity against HCV RNA replication . Finally, we succeeded in the establishment of RBV-resistant HCV RNA-replicating cells from RBV-sensitive ones and showed that the acquisition of RBV-resistance was mainly conferred by host factors, suggesting that RBV played roles in the suppression of HCV RNA replication by acting on host factors and thereby affects cellular physiologies .
RBV has the ability to exert not only anti-viral but also anti-tumorigenesis effects. In acute myeloid leukemia patients with poor prognosis, RBV treatment showed substantial clinical benefit . It appeared that RBV inhibited the translation of oncogenes, such as c-myc and cyclin D1, through inactivation of eukaryotic translation initiation factor 4E. In human prostate cancer cells with high resistance to the anti-tumor drug docetaxel, it was demonstrated that RBV treatment became sensitized to docetaxel by reprograming the gene expression profile of docetaxel-resistant tumor cells into that of docetaxel-sensitive tumor cells . Although the mechanism by which RBV reprograms the gene expression profile has remained to be clarified, it has been suggested that RBV has the ability to alter the cell physiology by dynamically changing the gene expression profile.
We recently demonstrated that RBV had the ability to suppress lipogenesis through the suppression of genes related to fatty acid synthesis, such as sterol regulatory element binding protein 1c (SREBP-1c), fatty acid synthase (FASN), and stearoyl-coenzyme A desaturase (SCD) . We also proposed a mechanism of RBV action for the suppression of lipogenesis, which involved IMPDH inhibition, several AMP-activated protein kinase (AMPK)-related kinases (AMPK-RKs), especially microtubule affinity regulating kinase 4 (MARK4), and down-regulation of retinoid X receptor α (RXRα). In the present study, we further searched for other factors involved in the action of RBV-induced suppression of lipogenesis, and we found that liver-enriched transcription factor CCAAT/enhancer-binding protein α (C/EBPα) was also down-regulated by RBV and its down-regulation led to a decreased expression of glycerol-3-phosphate acyltransferase, mitochondrial (GPAM), which encodes the first rate-limiting enzyme of triglyceride (TG) synthesis. Our results suggest that down-regulation of the C/EBPα-GPAM pathway also contributes to RBV-induced suppression of lipogenesis in addition to down-regulation of the RXRα-fatty acid synthesis pathway.
Materials and methods
RBV and guanosine were purchased from WAKO Pure Chemical Industries, Ltd (Osaka, Japan). MG-132 was purchased from Calbiochem (Darmstadt, Germany). Mycophenolic acid (MPA) was purchased from Sigma–Aldrich (St. Louis, MO, U.S.A.).
HuH-7- and Hep3B-derived cell lines were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, as described previously . ADK-expressing HuH-7A and Hep3B-ADK cells were generated from parental HuH-7 and Hep3B cells, respectively, by using a retrovirus vector system, as described previously . These cells were cultured in a medium supplemented with 2 μg/ml blasticidin S (Funakoshi, Tokyo, Japan) to maintain ADK expression.
Construction of a retrovirus vector plasmid for C/EBPα expression (pCX4pur/C/EBPα) or GPAM expression (pCX4pur/GPAM)
To amplify the open reading frame of the C/EBPα gene (MN_004364.4) or GPAM gene (MN_001244949.1), reverse transcription-polymerase chain reaction (RT-PCR) was performed as described previously , using total RNAs from HuH-7 cells. The primer sets used were as follows: 5′-AGAAGGATCCACCATGGAGTCGGCCGACTTCTACGAG-3′ and 5′-GGTAGGAATTCTCACGCGCAGTTGCCCATGGCC-3′ for C/EBPα; 5′-TTTGGATCCACCATGGATGAATCTGCACTGACC-3′ and 5′-TTTGCGGCCGCTTACTACAGCACCACAAAACTCAG-3′ for GPAM. The amplified PCR product of C/EBPα was cloned into the BamHI and EcoRI sites of pCX4pur . The amplified PCR product of GPAM was cloned into the BamHI and NotI sites of pCX4pur .
Generation of HuH-7A cells stably expressing exogenous RXRα, C/EBPα, or GPAM
A retrovirus vector plasmid for exogenous RXRα (pCX4pur/RXRα) , C/EBPα (pCX4pur/C/EBPα), or GPAM (pCX4pur/GPAM) expression was used. The retrovirus vector for exogenous expression of each protein was introduced into the HuH-7A cells by retroviral transfer, and then cells stably expressing exogenous RXRα, C/EBPα, or GPAM (designated as HuH-7ARα, HuH-7AC/α, or HuH-7AGPAM, respectively) were selected by 2 μg/ml puromycin (Sigma–Aldrich). The pCX4pur plasmid was also used for generation of the control puromycin-resistant HuH-7A cells (designated as HuH-7Acont). These cells were maintained in a medium supplemented with 2 μg/ml blasticidin S (Funakoshi) and 2 μg/ml puromycin (Sigma–Aldrich).
Total RNAs were extracted from cells using ISOGEN (Nippon Gene, Tokyo, Japan). RT-qPCR (RT-quantitative PCR) analysis for the mRNAs of the selected genes was performed using a real-time LightCycler PCR (Roche Diagnostics, Basel, Switzerland), as described previously . The primer sets, GPAM, glycerol-3-phosphate acyltransferase 3 (GPAT3), GPAT4, diacylglycerol acyltransferase 1 (DGAT1), DGAT2, and C/EBPα, are as follows: GPAM, 5′-ggaaagtttatccagtatggcatt-3′ and 5′-tgatatcttcctggtcatcgtg-3′; GPAT3, 5′-ggtgctgggcgtcatagt-3′ and 5′-cccaatgaaagccaaggtaa-3′; GPAT4, 5′-aggacttgtggacctgctgt-3′ and 5′-cccacgatcatcttgctgt-3′; DGAT1, 5′-actaccgtggcatcctgaac-3′ and 5′-ataaccgggcattgctca-3′; DGAT2, 5′-gaggggtctgggagatgg-3′ and 5′-ttggacctattgagccaggt3′; C/EBPα, 5′-cggtggacaagaacagcaac-3′ and 5′-tcactggtcagctccagcac-3′. Other gene primer sets used in this study were described previously . These gene expression levels were normalized to the levels of ATP synthase, H+ transporting, mitochondrial Fo complex subunit B1 (ATP5F1) mRNA.
Western blot analysis
The cells were harvested in cell lysis buffer [1% sodium dodecyl sulfate, 10 mM Tris–HCl (pH 7.5)] with protease inhibitor cocktail (Roche Diagnostics). The protein concentration measurement, preparation of samples, and immunoblotting analysis were performed, as described previously . Rabbit anti-C/EBPα and -MARK4 antibodies were purchased from Cell Signaling Technology (Danvers, MA, U.S.A.). Rabbit anti-RXRα, mouse anti-GPAM, and mouse anti-DGAT1 antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, U.S.A.). Mouse anti-β-actin antibody (Sigma–Aldrich) was used as the control for the amount of protein loaded per lane.
Small interfering RNAs (siRNAs) targeting C/EBPα (siC/EBPα) (M-006422-03), siGPAM (M-009946-01), and siDGAT1 (M-003922-02) were purchased from Dharmacon Inc. (Lafayette, CO, U.S.A.). Non-targeting siRNAs (D-001206-13) were used as a control (designated as siCont). Other siRNAs used in this study were described previously . Cells were transfected with the indicated siRNAs using lipofectamine RNAiMAX reagent (Invitrogen, Carlsbad, CA, U.S.A.). The knockdown efficiency of these siRNAs was determined by RT-qPCR using the corresponding primers.
Measurement of intracellular neutral lipids
Cells were treated with BODIPY493/503 (Thermo Fisher Scientific) for the staining of neutral lipids, and the fluorescence intensity in each cell was measured by a flow cytometer (FACS Calibur; BD Biosciences, Franklin Lakes, NJ, U.S.A.), as described previously . The mean fluorescence intensities in cells were calculated.
Data are presented as the means ± standard deviation from three or four independent experiments. Determination of the significance of differences among groups was assessed using the Student's t-test with a two-sided test. P < 0.05 was considered statistically significant.
The C/EBPα protein level is decreased by RBV via the proteasome degradation pathway
We recently demonstrated that down-regulation of RXRα by RBV was involved in the decreased expression of genes related to fatty acid synthesis, such as SREBP-1c, FASN, and SCD, and the suppression of lipogenesis . However, we noted that down-regulation of RXRα was not sufficient for the RBV-induced suppression of lipogenesis, since overexpression of RXRα attenuated but did not abolish it. In an effort to search for other factor(s) involved in suppression of lipogenesis by RBV, we found that C/EBPα, a transcription factor playing important roles in lipogenesis, was decreased at the protein level but not the mRNA level by RBV treatment in ADK-expressing HuH-7 (HuH-7A) (Figure 1A,B) and Hep3B (Hep3B-ADK) cells (Figure 1C), both of which highly express C/EBPα . But, no alteration of the protein level of either C/EBPα or RXRα was observed in control HuH-7 (HuH-7cont) cells expressing only endogenous ADK. These data indicated that the C/EBPα protein level was decreased by RBV-metabolites via post-transcriptional mechanism(s). It is known that the C/EBPα protein level is regulated in its protein stability by proteasome [16,17]. Thus, we examined whether the inhibition of proteasome activity would block the RBV-induced reduction in C/EBPα protein levels. As shown in Figure 1D, the proteasome inhibitor MG-132 attenuated the RBV-induced reduction in C/EBPα protein, suggesting that down-regulation of C/EBPα resulted from its protein degradation via the proteasome activity promoted by RBV.
Down-regulation of C/EBPα by RBV.
Next, we investigated whether the intracellular GTP depletion caused by RBV-mediated inhibition of IMPDHs was involved in the down-regulation of C/EBPα. For this purpose, we analyzed C/EBPα expression in HuH-7A cells treated with RBV along with guanosine to replenish GTP pools through the salvage pathway. As shown in Figure 1E, down-regulation of C/EBPα by RBV was canceled by co-treatment with guanosine in a similar manner as down-regulation of RXRα, indicating that GTP-deletion by RBV down-regulates not only RXRα but also C/EBPα. This notion was supported by the observation that treatment with MPA, a non-nucleoside inhibitor of IMPDH, also down-regulated C/EBPα expression (Figure 1F).
Down-regulation of C/EBPα by RBV is independent of RXRα and MARK4
To investigate whether down-regulation of C/EBPα by RBV was under the control of RXRα, we analyzed the expression of C/EBPα in RXRα-overexpressing HuH-7A cells. As shown in Figure 2A, the protein level of C/EBPα was not affected by RXRα overexpression, and down-regulation of C/EBPα by RBV was also seen in RXRα-overexpressing cells (HuH-7ARα) as in control cells (HuH-7Acont), indicating that down-regulation of C/EBPα by RBV was independent of RXRα expression.
Down-regulation of C/EBPα by RBV is independent of RXRα and MARK4.
We previously demonstrated that one of the AMPK-RKs, MARK4, was required for RBV to down-regulate RXRα expression . To investigate whether MARK4 also regulates C/EBPα expression, we examined the effect of MARK4 knockdown on RBV-induced down-regulation of C/EBPα. The results showed that down-regulation of C/EBPα by RBV was also seen in cells transfected with siRNAs targeting MARK4, while down-regulation of RXRα was attenuated (Figure 2B), indicating that down-regulation of C/EBPα was not under the control of MARK4.
Taken together, our data revealed that down-regulation of C/EBPα was mediated by RBV-induced GTP-depletion and independent of MARK4 and RXRα expressions (Figure 2C).
Down-regulation of C/EBPα leads to a reduction in intracellular neutral lipids without affecting genes related to the fatty acid synthesis pathway
C/EBPα is known to positively regulate lipogenesis [18–21]. We thus assumed that down-regulation of C/EBPα would be associated with RBV-induced suppression of lipogenesis. We performed C/EBPα-knockdown or -overexpression experiments to clarify the roles of C/EBPα in lipogenesis in HuH-7A cells. We first assessed the amounts of neutral lipids in cells by staining with BODIPY493/503 and measuring of the signals using a flow cytometer. The results revealed that knockdown of C/EBPα significantly reduced the BODIPY signal intensity, while overexpression of C/EBPα significantly increased it (Figure 3A,B), suggesting that C/EBPα positively regulated lipogenesis in HuH-7A cells and thus down-regulation of C/EBPα by RBV contributed to the suppression of lipogenesis.
Down-regulation of C/EBPα leads to a reduction in intracellular neutral lipids without affecting genes related to the fatty acid synthesis pathway.
We then examined whether C/EBPα regulated the expression of genes related to fatty acid synthesis, such as SREBP-1c, FASN, and SCD. RBV treatment significantly reduced the expression of these genes in C/EBPα-knockdown cells as in control cells (Figure 3C). In C/EBPα-overexpressing cells, FASN and SCD mRNAs were decreased by RBV as in the control cells (Figure 3D). At the very least, these data showed that down-regulation of C/EBPα was not involved in the RBV-induced reduction in these fatty acid-synthesis genes, although the expression level of SREBP-1c was lower in C/EBPα-overexpressing cells compared with control cells, with unknown reason.
RBV suppresses GPAM expression through down-regulation of C/EBPα
Since C/EBPα positively regulates the expression of genes related to the TG synthesis enzyme , we examined the effect of C/EBPα expression on these genes in HuH-7A cells. We found that knockdown of C/EBPα significantly reduced the expression of GPAM and DGAT1, but not DGAT2 (Figure 4A). But, the expressions of GPAM and DGAT1 were significantly increased in C/EBPα-overexpressing cells (Figure 4B). We next examined whether RBV influenced the expression of these TG synthesis enzyme genes in HuH-7A cells, and we found that only GPAM expression was significantly suppressed by RBV treatment (Figure 4C and Supplementary Figure S1). Moreover, RBV-induced reduction in GPAM expression was largely abolished by simultaneous knockdown of C/EBPα (Figure 4D), indicating that RBV suppressed GPAM expression mainly by decreasing the level of its transcription factor, C/EBPα.
RBV suppresses GPAM expression.
We analyzed the effect of MARK4 on lipogenic gene expression with RBV treatment. As shown in Figure 4E, RBV-induced reduction in GPAM expression was not altered by knockdown of MARK4, while the RBV-induced reductions in SREBP-1c, FASN, and SCD were attenuated, indicating that MARK4 was not involved in the RBV-induced down-regulation of C/EBPα-GPAM pathway.
RBV-induced reduction in GPAM contributes to lipogenesis suppression
We examined whether C/EBPα-overexpression canceled the RBV-induced reduction in GPAM expression. However, the results revealed that the reduction in GPAM expression by RBV was also observed in C/EBPα-overexpressing cells as in control cells, although basal GPAM expression in C/EBPα-overexpressing cells was higher than that in the control cells (Figure 5A). C/EBPα-overexpression also did not cancel the effect of RBV on the amount of neutral lipids (Figure 5B). We then analyzed the level of C/EBPα protein in the C/EBPα-overexpressing cells by Western blotting, and found that the overexpressed C/EBPα protein was also sharply down-regulated by RBV treatment (Figure 5C), suggesting that the overexpressed C/EBPα protein was also degraded by RBV in the same manner as endogenous C/EBPα, so that it accompanied the reduction in the high-level expression of GPAM in C/EBPα-overexpressing cells. Thus, we analyzed GPAM expression when down-regulation of the C/EBPα protein was blocked by treatment with proteasome inhibitor. As shown in Figure 5D, the reduction in GPAM expression by RBV was attenuated in cells co-treated with proteasome inhibitor MG-132, whereas genes related to the fatty acid synthesis were not affected, suggesting that RBV-induced degradation of C/EBPα led to a decrease in the expression of GPAM.
Reduction in GPAM expression under the control of C/EBPα is important for lipogenesis suppression.
To investigate whether the RBV-induced reduction in GPAM expression is associated with suppression of lipogenesis, we analyzed the amount of neutral lipids in cells with GPAM-knockdown. We assessed BODIPY signal intensities to determine the amounts of neutral lipids in HuH-7Acont or HuH-7AC/α cells transfected with siRNAs targeting GPAM (Figure 5E). We found that GPAM-knockdown did not affect the amount of neutral lipids in HuH-7Acont cells. But, GPAM-knockdown significantly reduced the amount of neutral lipids in HuH-7AC/α cells, in which the expression level of GPAM was up-regulated by exogenous C/EBPα expression, implying that reduction in GPAM expression under the control of C/EBPα contributed to the RBV-induced reduction in neutral lipids.
Given the data showing that GPAM-knockdown had no effect on the lipogenesis in HuH-7Acont cells, we speculated that alternative pathways of TG synthesis, which are under the control of RBV, compensated for the reduced GPAM activity. To clarify this possibility, we examined the effect of GPAM-knockdown on the expression of TG synthesis enzyme genes in HuH-7Acont cells. As shown in Figure 6A, the mRNA level of DGAT1, the protein of which functions in TG synthesis by re-esterification of partial glycerides produced by lipolysis of TG with preformed fatty acids , was significantly increased by GPAM-knockdown. This increase was also confirmed in the protein level (Figure 6B). Thus, we next investigated whether the increase in DGAT1 canceled the effect of GPAM-knockdown on lipogenesis. When siRNAs targeting GPAM were co-transfected with siRNAs targeting DGAT1 to preserve nearly the basal level of DGAT1, the level of neutral lipids was significantly reduced in HuH-7Acont cells (Figure 6C). These data suggested that GPAM-knockdown led to the increase in DGAT1, which abrogated the effect of GPAM-knockdown on lipogenesis. DGAT1 was not increased by RBV in HuH-7A cells, even though GPAM was decreased (Figure 4C). It is likely that RBV also has the ability to block up-regulation of DGAT1, resulting in lipogenesis suppression by the reduction in GPAM.
RBV-induced reduction in GPAM expression contributes to lipogenesis suppression.
We next examined the effect of GPAM overexpression on the level of neutral lipids in RBV-treated cells (Figure 6D,E). As shown in Figure 6E, ectopic expression of GPAM attenuated RBV-induced reduction in neutral lipids, indicating that RBV-induced GPAM reduction contributed to lipogenesis suppression.
Taken together, these results suggest that RBV-mediated down-regulation of C/EBPα leads to the lowered expression of GPAM, which in turn is involved in the RBV-induced suppression of lipogenesis (Figure 6F).
We recently demonstrated that RBV-induced down-regulation of RXRα led to the decreased expression of genes related to fatty acid synthesis . In the present study, we further demonstrated that down-regulation of C/EBPα and subsequent decreased expression of GPAM were also important for the actions of RBV to suppress lipogenesis. These studies suggest that RBV suppresses lipogenesis in hepatic cells by suppressing at least two pathways, the fatty acid synthesis pathway and the TG synthesis pathway regulated by RXRα and C/EBPα, respectively (Figure 6F). In addition to cell culture system, animal experiments will be needed to clarify RBV action for lipogenesis. It was previously reported that SREBP-1c positively regulated GPAM expression in HepG2 cells . Thus, it is possible that RBV-induced down-regulation of RXRα, which leads to decreased expression of SREBP-1c, also contributes to the reduction in GPAM expression. In the future, it would be useful to examine the cross-talk between RXRα and C/EBPα in RBV-induced suppression of lipogenesis.
C/EBPα plays versatile roles as a transcription factor to maintain cellular physiology. It was previously demonstrated that C/EBPα formed a complex with p300 to activate the expression of genes encoding TG synthetic enzymes . Although C/EBPα appeared to positively regulate GPAM and DGAT1, each of which encodes a rate-limiting enzyme of TG synthesis, in HuH-7A cells according to C/EBPα-knockdown and -overexpression experiments in the present study, only GPAM expression was decreased by RBV treatment. We speculate that this has been due to a difference in the amount of C/EBPα required to activate each gene; a large amount of C/EBPα is required for GPAM expression, but only a small amount is required for DGAT1 expression. The resulting C/EBPα expression level by RBV treatment may be sufficient to maintain DGAT1 expression, but not GPAM expression. C/EBPα-binding sites have been identified in the mouse genome for the TG synthesis enzyme genes, Dgat1 and Dgat2 , but the C/EBPα-binding sites in the human genome for the TG synthesis enzyme genes have not yet been identified. To delineate the difference in RBV-sensitivity or C/EBPα-dependency between GPAM and DGATs in human hepatic cells, it will be necessary to clarify the mechanism of activation of human TG synthesis genes by C/EBPα.
AMPK-RKs have been shown to regulate various types of metabolism, including lipogenesis , and we previously showed that some AMPK-RKs, especially MARK4, controlled the expression of SREBP-1c, FASN, and SCD in hepatic cells through the regulation of RXRα expression . But, neither C/EBPα nor GPAM expression was regulated by MARK4 in the present study (Figures 2B and 4E). Hence, we examined whether other AMPK-RKs expressed in hepatic cells were involved in the regulation of GPAM expression. To this end, we analyzed the effect of knockdown of either of the AMPK-RKs on RBV-induced GPAM reduction in HuH-7-derived OR6Ac cells ; however, neither AMPK-RK knockdown had a significant effect on the GPAM reduction (data not shown). Previously it was demonstrated that phosphorylated threonine residues at positions 222 and 226 in the human C/EBPα protein were targeted by ubiquitin ligase Fbxw7 for C/EBPα degradation . Thus, it is possible that a kinase(s) for C/EBPα phosphorylation, which has not yet been clarified, is responsible for the RBV-induced degradation of C/EBPα. The mechanism underlying the down-regulation of C/EBPα, as distinct from the down-regulation of RXRα, under the GTP-depletion by RBV thus remains to be revealed.
Studies of mice with deleted GPAT1, mouse homologue of human GPAM, and primary hepatocytes from these mice demonstrated that GPAT1 comprises 30–50% of total GPAT activity in the liver and is required to incorporate de novo synthesized fatty acid into TG [25,26]. Our study unanticipatedly revealed that GPAM-knockdown in HuH-7Acont cells up-regulated DGAT1 expression, which appeared to compensate for the reduction in GPAM activity. We observed that in HuH-7AC/α cells GPAM-knockdown also increased DGAT1 at the same level to that in HuH-7Acont cells (data not shown). Since the basal level of the intracellular neutral lipids was increased by up-regulation of GPAM in HuH-7AC/α cells (Figure 5E), increased DGAT1 resulting from GPAM-knockdown might not be sufficient to compensate for GPAM reduction in HuH-7AC/α cells. Previous study demonstrated that DGAT1 functioned in the re-esterification of partial glycerides generated by intracellular lipolysis, using preformed or exogenous fatty acids . Thus, we assume that human hepatic cells or some kind of particular hepatic cells have the machinery to preserve the intracellular TG levels by mutual interaction with GPAM- and DGAT1-mediated TG synthesis pathways. It is interesting to clarify the mechanisms how to respond to GPAM-knockdown or reduction in de novo lipogenesis and induce the up-regulation of DGAT1. Moreover, animal experiments examining whether this compensation mechanism for TG synthesis occurs in the normal or pathological liver will be needed.
Aberrant lipogenesis is a characteristic feature of tumor cells, in which large amounts of lipids are required for aberrant proliferation and transformed phenotypes. It was previously reported that up-regulation of C/EBPα expression was observed in a subset of hepatocellular carcinomas (HCC) and predicted poorer HCC patient prognosis . The authors observed that knockdown of C/EBPα in C/EBPα-highly-expressing HCC cell lines, including HuH-7, decreased cell proliferation and blocked the generation of xenograft tumors in nude mice. Another study reported that mice deficient in GPAT1 had low contents of TG in the liver and reduced susceptibility to liver cancer induced by the carcinogen diethylnitrosamine, suggesting that reduction in GPAM expression blocked the promotion of liver tumorigenesis . We previously reported that RBV suppressed the proliferation of OR6Ac cells , suggesting that down-regulation of C/EBPα and GPAM by RBV contributed to the inhibition of cell proliferation. Since intermediate molecules in TG synthesis, such as lysophosphatidic acid, phosphatidic acid, and diacylglycerol, are thought to be capable of initiating signaling pathways , it is possible that RBV suppresses cell proliferation by altering the amount of these intermediates. Thus, we will analyze not only TG but also these intermediates in order to clarify the mechanism of RBV action for cell proliferation and tumorigenesis.
Arendt et al.  performed a comprehensive gene expression analysis using liver biopsies among patients with simple steatosis, those with nonalcoholic steatohepatitis (NASH), and healthy controls. In that study, both C/EBPα and GPAM expressions were up-regulated more than 2-fold in both patient groups compared with healthy controls. Jin et al. reported that activity of the C/EBPα-p300-TG synthesis pathway was elevated in the livers of patients with nonalcoholic fatty liver disease (NAFLD), and thus considered that C/EBPα-p300 pathway might be a therapeutic target to prevent the development of hepatic steatosis . According to these studies and ours, RBV appears to be a potential drug for the treatment of patients with HCC and NAFLD/NASH, since RBV has the ability to down-regulate C/EBPα expression, leading to reduced expression of GPAM, which is involved in the TG synthesis pathway. However, RBV is known to have some side effects, such as anemia. Therefore, an RBV-replacement drug should be developed. Our studies showing the mechanism of RBV action for the suppression of lipogenesis could help to identify target molecules for therapeutic approaches to prevent hepatic pathogeneses that involve aberrant lipogenesis, such as NAFLD/NASH and HCC.
AMP-activated protein kinase
CCAAT/enhancer-binding protein α
fatty acid synthase
glycerol-3-phosphate acyltransferase, mitochondrial
hepatitis C virus
inosine monophosphate dehydrogenase
microtubule affinity regulating kinase 4
nonalcoholic fatty liver disease
retinoid X receptor α
stearoyl-coenzyme A desaturase
small interfering RNAs
sterol-regulatory element-binding protein-1c
S.S. and N.K. designed the research. S.S. performed most of the experiments. D.O. made the retrovirus vectors for C/EBPα and GPAM expression. S.S., D.O., Y.U., H.D., M.H., S.K., and N.K. analyzed the data. S.S. wrote the paper. All authors reviewed the manuscript.
The present study was supported, in part, by a grant for Practical Research on Hepatitis from the Japan Agency for Medical Research and Development [no. JP18fk0210305], a JSPS KAKENHI grant [no. JP18K07972 to S.S.], and a research grant from Gilead (to S.S.).
We thank Rimi Nonoyama, Yuka Nishimori, and Takashi Nakamura for their technical assistance.
The Authors declare that there are no competing interests associated with the manuscript.