BackgroundDeconjugation of ubiquitin and/or ubiquitin-like modified protein substrates is essential to modulate protein-protein interactions and, thus, signaling processes in cells. Although deubiquitylating (deubiquitinating) enzymes (DUBs) play a key role in this process, however, their function and regulation remain insufficiently understood. The "loss-of-function" phenotype studies can provide important information to elucidate the gene function, and zebrafish is an excellent model for this goal.ResultsFrom an in silico genome-wide search, we found more than 90 putative DUBs encoded in the zebrafish genome belonging to six different subclasses. Out of them, 85 from five classical subclasses have been tested with morpholino (MO) knockdown experiments and 57 of them were found to be important in early development of zebrafish. These DUB morphants resulted in a complex and pleiotropic phenotype that, regardless of gene target, always affected the notochord. Based on the huC neuronal marker expression, we grouped them into five sets (groups I to V). Group I DUBs (otud7b, uchl3 and bap1) appear to be involved in the Notch signaling pathway based on the neuronal hyperplasia, while group IV DUBs (otud4, usp5, usp15 and usp25) play a critical role in dorsoventral patterning through the BMP pathway.ConclusionWe have identified an exhaustive list of genes in the zebrafish genome belonging to the five established classes of DUBs. Additionally, we performed the corresponding MO knockdown experiments in zebrafish as well as functional studies for a subset of the predicted DUB genes. The screen results in this work will stimulate functional follow-up studies of potential DUB genes using the zebrafish model system.

Enterovirus 71 (EV71) is a neurotrophic virus that causes hand, foot and mouth disease (HFMD) and occasional neurological infection among children. It infects primate cells but not rodent cells, primarily due to the incompatibility between the virus and the expressed form of its receptor, scavenger receptor class B member 2 (SCARB2) protein, on rodent cells (mSCARB2). We previously generated adapted strains (EV71:TLLm and EV71:TLLmv) that were shown to productively infect primate and rodent cell lines and whose genomes exhibited a multitude of non-synonymous mutations compared with the EV71:BS parental virus. In this study, we aimed to identify mutations that are necessary for productive infection of murine cells by EV71:BS. Using reverse genetics and site-directed mutagenesis, we constructed EV71 infectious clones with specific mutations that generated amino acid substitutions in the capsid VP1 and VP2 proteins. We subsequently assessed the infection induced by clone-derived viruses (CDVs) in mouse embryonic fibroblast NIH/3T3 and murine neuroblastoma Neuro-2a cell lines. We found that the CDV:BS-VP1K98E,E145A,L169F with three substitutions in the VP1 protein—K98E, E145A and L169F—productively infected both mouse cell lines for at least three passages of the virus in murine cells. Moreover, the virus gained the ability to utilize the mSCARB2 protein to infect murine cell lines. These results demonstrate that the three VP1 residues cooperate to effectively interact with the mSCARB2 protein on murine cells and permit the virus to infect murine cells. Gain-of-function studies similar to the present work provide valuable insight into the mutational trajectory required for EV71 to infect new host cells previously non-susceptible to infection.

Since its identification in 1969, Enterovirus 71 (EV71) has been causing periodic outbreaks of infection in children worldwide and most prominently in the Asia-Pacific Region. Understanding the pathogenesis of Enterovirus 71 (EV71) is hampered by the virus’s inability to infect small animals and replicate in their derived in vitro cultured cells. This manuscript describes the phenotypic and genotypic characteristics of two selected EV71 strains (EV71:TLLm and EV71:TLLmv), which have been adapted to replicate in mouse-derived NIH/3T3 cells, in contrast to the original parental virus which is only able to replicate in primate cell lines. The EV71:TLLm strain exhibited productive infection in all primate and rodent cell lines tested, while EV71:TLLmv exhibited greater preference for mouse cell lines. EV71:TLLmv displayed higher degree of adaptation and temperature adaptability in NIH/3T3 cells than in Vero cells, suggesting much higher fitness in NIH/3T3 cells. In comparison with the parental EV71:BS strain, the adapted strains accumulated multiple adaptive mutations in the genome resulting in amino acid substitutions, most notably in the capsid-encoding region (P1) and viral RNA-dependent RNA polymerase (3D). Two mutations, E167D and L169F, were mapped to the VP1 canyon that binds the SCARB2 receptor on host cells. Another two mutations, S135T and K140I, were located in the VP2 neutralization epitope spanning amino acids 136–150. This is the first report of human EV71 with the ability to productively infect rodent cell lines in vitro.