(740d) Comparative Mapping of Dengue Virus-Host Interactions Using Global Proteomics

Introduction: Dengue virus (DENV) is transmitted by mosquitoes and infects nearly 400 million people annually. Like other viruses, DENV interacts with the host to orchestrate myriad biochemical processes required for replication. A unique characteristic of vector-borne viruses such as DENV is that they use cellular machinery in two very different hosts to successfully replicate. This raises the question, how does DENV hijack cellular processes for replication in two hosts that have diverged over hundreds of millions of years. 

Materials and Methods: We investigated this question using complementary systems biology techniques. We first sought to systematically identify DENV-human and DENV-mosquito protein interactions for all 10 DENV proteins using affinity purification/mass spectrometry (AP/MS). To identify DENV-human and DENV-mosquito protein interactions, individual Strep-tagged DENV proteins were over-expressed in human HEK293T or A. aegypti Aag2 cells, and subjected to AP/MS. High confidence interactions were determined using a statistical algorithm called MS Interaction STatistics (MiST) to filter the resulting MS data based on specificity, reproducibility and abundance of the identified peptides. We then sought to systematically link physical interactions to functional outcomes during DENV replication using RNAi knockdown and a reporter virus system in human Huh7 cells. 

Results and Discussion: From our proteomic work, we recovered many previously reported physical interactions, included interactions between NS1 and Complement C4, NS5 and FAM192A, and NS5 and Importin-a/b. We also found a significant enrichment for orthologous interactions that are shared between the human and mosquito hosts, potentially providing conserved physical interfaces by which functional modules can be manipulated during replication. The functional genetic characterization of 349 DENV-interacting human factors identified 95 genes that significantly impacted viral replication, including previously reported replication factors SEC61G and UBE3A. To further understand how shared interactions facilitate replication in mosquitoes, we are testing the impact DENV-interacting insect factors on DENV replication in the mosquito A. aegypti Aag2 cells. This comparative approach will elucidate how conserved functional modules are manipulated by a virus to facilitate replication two evolutionarily distant hosts. Upon integration of the proteomic and genetic data, we identified several nuclear human factors that impact viral replication and physically interact with DENV. We are currently investigating how these DENV-host interactions impact global gene expression using ChIP-seq and RNA-seq. Our preliminary data suggest that DENV inhibits the expression of antiviral interferon stimulated genes through a protein interaction between the viral polymerase and transcription elongation machinery.

Conclusions: Using proteomics and genetics, we identified DENV-host protein interactions in both human and mosquito systems. We further determined the impact of these host factors on viral replication, and identified interactions between DENV proteins and host nuclear factors involved in the regulation of gene expression. Our data support a model in which a physical interaction between the DENV polymerase and transcription elongation machinery inhibits the antiviral innate immune response.

Acknowledgements: We thank Dr. Michel Tassetto and Dr. Jason Wojcechowski for feedback. This work was supported by DARPA 10-93-Prophecy-PA-008 to RA and NJK, and NIH F32AI112262-01A1 to PSS.