Mouse hepatitis virus (MHV), a coronavirus, utilizes a discontinuous transcription mechanism for subgenomic mRNA synthesis. Previous studies (C.-L. Liao and M. C. C. Lai, J. Virol. 68:4727-4737, 1994) have demonstrated that an upstream cis-acting leader sequence serves as a transcriptional enhancer, but the mechanism of transcriptional regulation is not clear. In this study, we constructed a series of defective interfering (DI) RNAs containing the chloramphenicol acetyltransferase (CAT) gene behind a differentially expressed transcription initiation (intergenic) sequence (for mRNA2-1). These DI RNAs had different copy numbers of the UCUAA pentanucleotide sequence at the 3' end of the leader. Transfection of these DI RNA constructs into cells infected with a helper MHV, which contains either two or three UCUAA copies at the 3' end of the leader, resulted in differential expression of CAT activities. We demonstrated that the copy number of UCUAA repeats in the leaders of both helper viral and DI RNAs affected the level of CAT activity, suggesting that MHV leader RNA could regulate both in trans and in cis the transcription of subgenomic mRNAs. The leader RNA of subgenomic mRNAs was derived from either the trans- or the cis-acting leader. Furthermore, insertion of a UA-rich sequence (UUUAUAAAC) immediately downstream of the leader in DI RNA, to match the sequence of helper viral RNA, enhanced the CAT activity by threefold, suggesting that this nine-nucleotide sequence is a cis-acting element. Interestingly, when the nine-nucleotide sequence was absent in DI RNA, the leaders of subgenomic mRNAs were exclusively derived from the helper virus. In contrast, when the nine-nucleotide sequence was present in DI RNA, the leaders were derived from both helper viral and DI RNAs. These results suggest that the nine-nucleotide sequence either is required for the leader RNA to initiate mRNA synthesis or, alternatively, serves as a transcription terminator for the leader RNA synthesis. However, when a constitutively expressed intergenic sequence (for mRNA7) was used, no difference in transcription efficiency was noted, regardless of the copy number of UCUAA in the DI RNA and helper virus. This study thus indicates that MHV subgenomic RNA transcription requires the interaction among the intergenic (promoter) sequence, a trans-acting leader, and a cis-acting leader sequence. A novel model of transcriptional regulation of coronavirus subgenomic mRNAs is presented.

Positive-sense RNA viruses, such as dengue virus (DENV), hijack the intracellular membrane machinery for their own replication. The Rab18 protein, a member of the Rab GTPase family, key regulators of membrane trafficking, is located on the organelles involved in DENV infection, such as the endoplasmic reticulum (ER) and lipid droplets (LDs). In this study, we addressed the potential involvement of Rab18 in DENV infection by using cells overexpressing the wild-type, GTP-bound active form, or GDP-bound inactive form of Rab18 and cells with Rab18 knockdown. DENV replication, measured by viral protein, viral RNA, and viral progeny production, as well as LD induction, was reduced in cells with inactive Rab18 and in cells deprived of Rab18 expression, suggesting a positive role of Rab18 in the DENV life cycle. Interestingly, the interaction of fatty acid synthase (FASN), a key lipogenic enzyme in lipid biosynthesis, with DENV NS3 protein relied on the conversion of the GDP-bound to the GTP-bound form of Rab18. Furthermore, the targeting of FASN to sites participating in DENV infection, such as the ER and LDs, depends on functional Rab18. Thus, Rab18-mediated membrane trafficking of FASN and NS3 facilitates DENV replication, probably by ensuring a sufficient and coordinated lipid supply for membrane proliferation and arrangement. IMPORTANCE Infection by dengue virus (DENV), an important mosquito-borne virus threatening ϳ40% of the world's population, can cause mild dengue fever or severe dengue hemorrhagic fever and dengue shock syndrome. The pathogenesis mechanisms of DENVrelated diseases are not clear, but high viral replication is believed to be a risk factor for the severe form of DENV infection. Thus, understanding the detailed mechanism of DENV replication might help address this devastating virus. Here, we found that Rab18, a small GTPase involved in vesicle trafficking and located in the endoplasmic reticulum network and on the surfaces of lipid droplets, positively regulates DENV replication. The functional machinery of Rab18 is required to recruit the enzyme fatty acid synthase to sites of DENV replication and to interact with DENV NS3 protein to promote fatty acid biosynthesis. Thus, DENV usurps Rab18 to facilitate its own replication. Dengue virus (DENV), a member of the genus Flavivirus of the family Flaviviridae, is the causative agent of dengue fever, dengue hemorrhagic fever, and dengue shock syndrome (1). Other members of the family, such as hepatitis C virus (HCV), yellow fever virus, West Nile virus, and Japanese encephalitis virus, are also involved in human diseases. DENV is enveloped and contains a single-stranded positive-sense RNA genome. Like other positive-sense RNA viruses, DENV modifies host cytoplasmic membranes to form platforms for viral replication (2, 3). DENV replication occurs in close association with virus-induced membrane structures derived from the endoplasmic reticulum (ER) network, with spatial coupling between sites of viral replication and virion assembly (4). The memb...

Previously reported findings by our group showed that non-structural protein 3 (NS3) of Japanese encephalitis virus (JEV) was localized mainly in the JEV-induced convoluted membrane (CM), which has been proposed to originate from rough endoplasmic reticulum (rER), Golgi apparatus or the trans-Golgi network (TGN), and serves as a reservoir for viral proteins during virus assembly. Earlier findings indicated that NS3 of Kunjin virus interacts with microtubules. In addition, one of the Golgi-associated proteins, tumour susceptibility protein 101 (TSG101), associates with microtubules and is required for budding of retroviral particles. To clarify the association of NS3 with microtubules or with TSG101 during JEV assembly, we applied immunofluorescence, co-immunoprecipitation and immunoelectron microscopic methods. Virus infection, as well as transfection with an NS2B-NS3 expression plasmid, induced microtubule rearrangement. When cells were treated with colchicine, which interferes with microtubule polymerization, NS3 still associated with tubulin and TSG101. Furthermore, tubulin and TSG101 were co-localized with NS3 in the CM by immunogold labelling. Our observations indicate that microtubules and TSG101 associate with NS3, which is incorporated into the JEV-induced structure during JEV replication.

Infection with Japanese encephalitis virus (JEV), a mosquito-borne flavivirus, may cause acute encephalitis in humans and induce severe cytopathic effects in various types of cultured cells. We observed that JEV replication rendered infected baby hamster kidney (BHK-21) cells sensitive to the translational inhibitor hygromycin B or α-sarcine, to which mock-infected cells were insensitive. However, little is known about whether any JEV nonstructural (NS) proteins contribute to virus-induced changes in membrane permeability. Using an inducibleEscherichia coli system, we investigated which parts of JEV NS1 to NS4 are capable of modifying membrane penetrability. We found that overexpression of NS2B-NS3, the JEV protease, permeabilized bacterial cells to hygromycin B whereas NS1 expression failed to do so. When expressed separately, NS2B alone, but not NS3, was sufficient to alter bacterial membrane permeability. Similarly, expression of NS4A or NS4B also rendered bacteria susceptible to hygromycin B inhibition. Examination of the effect of NS1 to NS4 expression on bacterial growth rate showed that NS2B exhibited the greatest inhibitory capability, followed by a modest repression from NS2A and NS4A, whereas NS1, NS3, and NS4B had only trivial influence with respect to the vector control. Furthermore, when cotransfected with a reporter gene luciferase or β-galactosidase, transient expression of NS2A, NS2B, and NS4B markedly reduced the reporter activity in BHK-21 cells. Together, our results suggest that upon JEV infection, these four small hydrophobic NS proteins have various modification effects on host cell membrane permeability, thereby contributing in part to virus-induced cytopathic effects in infected cells.