UC Irvine

Coronaviruses, like nearly all studied positive sense single-stranded RNA viruses, rely on host cell internal membranes to produce structures that support viral replication. These structures form in the cytoplasm and are called double membrane vesicles (DMVs) and convoluted membranes (CMs). DMVs are ~100-300 nm in diameter and are usually found in clustered groups. They are interconnected with regions of CM that can vary in size and extent. These membranous structures are the localized sites of viral genome replication and transcription within the infected cell. DMVs and CMs are distinct from structures that form later in the infection life cycle for purposes of virion assembly and budding. Ultrastructural studies have confirmed that the walls of DMVs and CMs are composed of two lipid bilayers in close proximity to each other and that the membranes are at least somewhat continuous with the membrane of the endoplasmic reticulum (ER). However, they do not stain for canonical ER markers. Despite the work that has been performed to discern the intricacies of DMV and CM assembly in coronavirus infected cells, the specific mechanisms and pathways involved have remained unclear. We present here the results of work done to elucidate both the viral proteins responsible for DMV and CM formation as well as work to decipher which host cellular proteins and pathways might be important. We used plasmid constructs expressing nonstructural proteins (nsps) from the severe acute respiratory syndrome (SARS) coronavirus to determine that the combination of nsp3, nsp4, and nsp6 is sufficient for producing structures similar to DMVs and CMs induced during SARS-coronavirus infection. We also characterized structures induced by each of the three nsps individually and in paired combinations to theorize a general putative mechanism by which DMV and CM structures form. Additionally, we used inhibitors and siRNA technology to determine that host cellular fatty acid synthase (FASN) is an important protein for murine hepatitis virus replication and that, in the absence of FASN activity, DMVs and CMs are not observed. This work adds to the body of evidence characterizing membrane rearrangement by coronaviruses and will aid ultimately in a full description of the process.