PLoS One. 2016 Jul 26;11(7):e0159324. doi: 10.1371/journal.pone.0159324.
The phospholipid:diacylglycerol acyltransferase Lro1 is responsible for hepatitis C virus core-induced lipid droplet formation in a yeast model system
Shingo Iwasa, Naoko Sato, Chao-Wen Wang, Yun-Hsin Cheng, Hayato Irokawa, Gi-Wook Hwang, Akira Naganuma and Shusuke Kuge
Department of Microbiology, Faculty of Pharmaceutical Sciences, Tohoku Medical and Pharmaceutical University, Japan
Laboratory of Molecular and Biochemical Toxicology, Graduate School of Pharmaceutical Sciences, Tohoku University, Japan
Institute of Plant and Microbial Biology, Academia Sinica, Taiwan
Chronic infection with the hepatitis C virus frequently induces steatosis, which is a significant risk factor for liver pathogenesis. Steatosis is characterized by the accumulation of lipid droplets in hepatocytes. The structural protein core of the virus induces lipid droplet formation and localizes on the surface of the lipid droplets. However, the precise molecular mechanisms for the core-induced formation of lipid droplets remain elusive. Recently, we showed that the expression of the core protein in yeast as a model system could induce lipid droplet formation. In this study, we probed the cellular factors responsible for the formation of core-induced lipid-droplets in yeast cells. We demonstrated that one of the enzymes responsible for triglyceride synthesis, a phospholipid:diacylglycerol acyltransferase (Lro1), is required for the core-induced lipid droplet formation. While core proteins inhibit Lro1 degradation and alter Lro1 localization, the characteristic localization of Lro1 adjacent to the lipid droplets appeared to be responsible for the core-induced lipid droplet formation. RNA virus genomes have evolved using high mutation rates to maintain their ability to replicate. Our observations suggest a functional relationship between the core protein with hepatocytes and yeast cells. The possible interactions between core proteins and the endoplasmic reticulum membrane affect the mobilization of specific proteins.
Nearly 3% of the world’s population is chronically infected with the hepatitis C virus (HCV), which is a major risk factor for liver cirrhosis and hepatocellular carcinoma. Chronic infection with HCV causes the abnormal accumulation of significant levels of liver lipids (steatosis), which is associated with hepatocellular carcinomas in HCV-infected patients. Steatosis is characterized by the accumulation of liver lipid droplets (LDs), which are essential for RNA replication and HCV particle formation.
The positive-stranded RNA genome of HCV encodes 10 viral proteins . A structural proteins core is believed to have an important role in pathogenesis, as shown by the formation of steatosis and hepatocellular carcinomas in core-transgenic mice. A triglyceride (TAG) synthetic enzyme, acyl-CoA:diacylglycerol O-acyltransferase (DGAT), is required for the core to translocate from the ER to LDs. However, the activity of DGAT1 is not affected by the core . However, the mechanism underlying the effect of the core on these factors, such as the interactions between the core and the endoplasmic reticulum (ER) membrane that lead to the altered translocation of these factors, remains obscure.
Yeast (Saccharomyces cerevisiae) is a viable model system to study neutral lipid homeostasis for higher eukaryotes. The core protein is localized to the cytoplasmic side of the ER and enhances LD formation in yeast cells suggesting that a functional analogy of the core between hepatocytes and yeast cells, namely in the intrinsic characteristics of the core.
LDs form when sterol ester (SE) and/or TAG accumulate and are surrounded by phospholipid monolayers. The syntheses of cellular SE and TAG are catalyzed by the acyltransferase family of proteins. In yeast, TAG is synthesized by a DGAT homologue, encoded by the gene DGA1, and a phospholipid:diacylglycerol O-acyltransferase (PDAT), encoded by the gene LRO1. Additionally, two acyl-CoA:cholesterol O-acyltransferase (ACAT)-related enzymes (Are1 and Are2) are responsible for SE synthesis.
Fig 1. The core-induced LD level is mediated by the gene responsible for TAG synthesis.
The LDs in live yeast cells were stained with BODIPY 493/503 and analyzed by fluorescent microscopy. The differential interference contrast images (DIC) and BODIPY 493/503 fluorescent images (BODIPY) are shown. Scale bars: 5 μm. (Data adapted from PLOS ONE DOI:10.1371/journal.pone.0159324)
Lro1 is responsible for core-induced LD formation
Induction of the expression of the core protein for 3 h significantly enhanced LD levels in the fluorescent images (Fig 1, Wild type, Core). The yeast strain lacking all four genes (dga1, lro1, are1 and are2, 4Δ), which was previously shown to be defective in forming LDs, also failed to form LDs by the core expression. Interestingly, the LRO1 single knockout (lro1Δ) had a markedly reduced level of the core-induced LD formation.
The TAG level was enhanced in wild-type and dga1Δ cells by core expression, whereas the levels of phospholipids (PLs), ergosterol (ERG) and SE were unaffected (see original paper). Most of the TAG species that were upregulated by the core expression were mainly Lro1-dependent. Our results suggested that Lro1 was the enzyme most responsible for the core-dependent induction of the upregulation of TAG levels, resulting in the induction of LDs.
The Lro1 distribution changed in response to core expression
Although the quantitative enhancement of Lro1 by core-induced ERAD inhibition did not seem to be directly responsible for the core-dependent increase in LD levels, we found that Lro1-mCherry, but not Dga1-mCherry, was relocalizaed in response to core expression. As shown in Figs 2A, mCherry-fused Lro1 was perinuclear and cortical shaped, especially hrd1Δ cells, in which the fluorescent level of Lro1-mCherry was enhanced due to disruption of ERAD system. The Lro1-mCherry fluorescence was shown as punctuated and laminar structures in response to core expression in the proximity of the induced LDs.
As shown in Fig 2B, tunicamycin treatment in the absence of the core protein successfully induced LD formation but had no apparent effect on the Lro1-mCherry distribution. These results suggested that the relocalization of Lro1-mCherry seems to be determined by Core expression but not formation of LDs by tunicamycin treatment (Fig. 2B).
Fig 2. Changes in the distribution of Lro1-mCherry, but not Dga1-mCherry, in response to the core.
(A) Changes in the distribution of Lro1 by the core expression. Images of mCherry, BODIPY 493/503, and the overlay of mCherry (red) and BODIPY (green), and DIC are shown. (B) The expression of the core, but not of tunicamycin, altered the distribution of Lro1-mCherry in wild-type cells expressing Lro1-mCherry. The cells carrying the empty vector were treated with tunicamycin (4 μg/ml) for 3 h. (Data adapted from PLOS ONE DOI:10.1371/journal.pone.0159324)
The core D2 region is responsible for LD formation and its localization close to LDs and Lro1
We examined whether the core D2 region (the D2 domain and the residual C-terminus, see Fig. 3A) could induce and co-localize with LDs, as observed in mammalian cells. As shown in Fig 3B, DsRed-core (Fig. 3B) successfully induced LDs and accumulated as punctate structures and localized close to the BODIPY 493/503-stained LDs. The DsRed-core colocalized with LDs, which was induced by tunicamycin in lro1Δ cells. Thus, we concluded that Lro1 might be unnecessary for the punctate localization of DsRed-core on the adjacent surface of LDs.
Furthermore, we found that the GFP-core was partially colocalized with Lro1-mCherry. Fig 4A shows that Lro1-mCherry was localized in larger patches and the GFP-core surrounded Lro1 patches in punctate structures. The GFP-core expression also altered the localization of the ER membrane protein Hmg2-mCherry (Fig 4B). The Hmg2-mCherry fluorescence surrounding the ER membrane was altered from the laminar structures to the punctate structures by the GFP-core expression. In contrast to Lro1-mCherry, the GFP-core did not colocalized with Hmg2-mCherry (Fig 4B). Collectively, the core protein might affect the ER membrane and affect the localization of the ER proteins (Hmg2 and Lro1). However, the adjacent localization with a punctate structure of core protein was specific to Lro1, but not to Hmg2 and Dga1.
We showed that the expression of the core protein might cause changes in the localization of the ER surface protein Lro1 to a position adjacent to LDs (Fig 2). Despite the Lro1 changes, the distribution of another protein responsible for TAG synthesis, Dga1, was unaffected. Additionally, the localization of the punctate core was not affected by the loss of Lro1 (Fig 3C). Thus, we speculated that the punctate expression of the core protein on the ER surface might guide the accumulation of Lro1 as patches, which results in LD accumulation (Fig 4C). The mechanism by which Lro1 accumulation occurs remains unknown. Nevertheless, our present data suggest that the mobilization of Lro1 may be regulated by a specific mechanism: the core accumulation on the cytoplasmic side of the ER membrane changes various aspects of ER homeostasis, namely, the alteration of the distribution of ER membrane proteins and the Lro1-dependent accumulation of TAGs (LDs) (see Fig 4C). Additional investigations to determine the mechanism responsible for the changes in Lro1 localization by the ER-core interaction in this yeast model system may provide important insights into understanding how the core protein affects the homeostasis of ER surface proteins.
RNA virus genomes evolve quickly because of their high mutation rates to adapt to different circumstances to maintain effective replication. Our observations suggested a functional relationship between the core protein of HCV with hepatocyte and yeast cells, which in turn implicates an intrinsic interaction of the core D2 region with the ER membrane and possible alterations of the ER membrane.
Fig 3.Localization of DsRed-core and LDs in wild-type and tunicamycin-treated lro1Δ cells.
(A) The structure of a full-length, DsRed- and GFP-fused core. The C-terminal region of the core protein is further processed by a host signal peptidase to generate a mature core (amino acid 1 to 177, here called core), which then enables the translocation to LDs. The D1 domain is responsible for RNA binding, whereas the D2 domain anchors to the surfaces of the ER and LDs [3, 4]. (B) Wild-type cells expressing DsRed-core were stained with BODIPY 493/503. The images of DsRed, BODIPY 493/503, the overlay of DsRed (red) and BODIPY (green), and DIC are shown. (C) Lro1Δ cells expressing DsRed-core were were stained with BODIPY 493/503. Fluorescent images were taken as described above. (Data adapted from PLOS ONE DOI:10.1371/journal.pone.0159324)
Fig 4. Effect of GFP-core on localization of Lro1-mCherry and Hmg2-mCherry.
(A) The colocalization of GFP-core and Lro1-mCherry was observed. Images of Lro1-mCherry, GFP, the overlay of GFP (green) and mCherry (Red) in Hrd1Δ LRO1-mCherry cells expressing GFP (Control) or GFP-core are shown. (B) The expression of GFP-core induced the redistribution of Hmg2-mCherry. Because Hmg2 is also a substrate for the Hrd1-dependent ERAD pathway, we used the hrd1Δ genotype for this experiment. Fluorescent images of GFP and Hmg2-mCherry were taken. (C) A model for core-induced LD formation. The expression of core accumulated at a specific locus on the ER, which triggers Lro1 accumulation. The accumulation of Lro1 induced LD formation. (Data adapted from PLOS ONE DOI:10.1371/journal.pone.0159324)
This study was supported by the Program for Promotion of Fundamental Studies in Health Sciences of the National Institute of Biomedical Innovation (NIBIO) and by Grants-in-Aid for Scientific Research (KAKENHI 17028003 and KAKENHI 20059003) from MEXT.
Shusuke Kuge, Ph.D.
Department of Microbiology, Faculty of Pharmaceutical Sciences, Tohoku Medical and Pharmaceutical University, Sendai, Miyagi 981-8558, Japan
- Bartenschlager R, Frese M, Pietschmann T. Novel insights into hepatitis C virus replication and persistence. Adv Virus Res. 2004;63:71-180. Epub 2004/11/09. doi: S0065352704630028 [pii]10.1016/S0065-3527(04)63002-8. PubMed PMID: 15530561.
- Herker E, Harris C, Hernandez C, Carpentier A, Kaehlcke K, Rosenberg AR, et al. Efficient hepatitis C virus particle formation requires diacylglycerol acyltransferase-1. Nat Med. 2010;16(11):1295-8. Epub 2010/10/12. doi: nm.2238 [pii]10.1038/nm.2238. PubMed PMID: 20935628; PubMed Central PMCID: PMC3431199.
- Okamoto K, Mori Y, Komoda Y, Okamoto T, Okochi M, Takeda M, et al. Intramembrane processing by signal peptide peptidase regulates the membrane localization of hepatitis C virus core protein and viral propagation. J Virol. 2008;82(17):8349-61. Epub 2008/06/20. doi: JVI.00306-08 [pii]10.1128/JVI.00306-08. PubMed PMID: 18562515; PubMed Central PMCID: PMC2519675.
- Boulant S, Montserret R, Hope RG, Ratinier M, Targett-Adams P, Lavergne JP, et al. Structural determinants that target the hepatitis C virus core protein to lipid droplets. J Biol Chem. 2006;281(31):22236-47. Epub 2006/05/18. doi: M601031200 [pii]10.1074/jbc.M601031200. PubMed PMID: 16704979.