Mapping Protein Interactions between Dengue Virus and Its Human and Insect Hosts

BackgroundDengue fever is an increasingly significant arthropod-borne viral disease, with at least 50 million cases per year worldwide. As with other viral pathogens, dengue virus is dependent on its host to perform the bulk of functions necessary for viral survival and replication. To be successful, dengue must manipulate host cell biological processes towards its own ends, while avoiding elimination by the immune system. Protein-protein interactions between the virus and its host are one avenue through which dengue can connect and exploit these host cellular pathways and processes.Methodology/Principal FindingsWe implemented a computational approach to predict interactions between Dengue virus (DENV) and both of its hosts, Homo sapiens and the insect vector Aedes aegypti. Our approach is based on structural similarity between DENV and host proteins and incorporates knowledge from the literature to further support a subset of the predictions. We predict over 4,000 interactions between DENV and humans, as well as 176 interactions between DENV and A. aegypti. Additional filtering based on shared Gene Ontology cellular component annotation reduced the number of predictions to approximately 2,000 for humans and 18 for A. aegypti. Of 19 experimentally validated interactions between DENV and humans extracted from the literature, this method was able to predict nearly half (9). Additional predictions suggest specific interactions between virus and host proteins relevant to interferon signaling, transcriptional regulation, stress, and the unfolded protein response.Conclusions/SignificanceDengue virus manipulates cellular processes to its advantage through specific interactions with the host's protein interaction network. The interaction networks presented here provide a set of hypothesis for further experimental investigation into the DENV life cycle as well as potential therapeutic targets.Author SummaryDengue virus (DENV) represents a major disease burden in tropical and subtropical regions of the world, and has shown an increase in the number of cases in recent years. DENV is transmitted to humans through the bite of an infected mosquito, typically Aedes aegypti, after which it begins the infection and replication lifecycle within human cells. To perform the molecular functions required for invasion, replication, and spread of the virus, proteins encoded by DENV must interact with and alter the behavior of protein networks in both of these hosts. In this work, we used a computational method based on protein structures to predict interactions between DENV and its human and insect hosts. We predict numerous interactions, with many involved in known cell death, stress, and immune system pathways. Further investigation of these predicted protein-protein interactions should provide targets to combat the clinical manifestations of this disease in humans as well as points of intervention focused within the mosquito vector