<p>  </p> <p>Infection by Marburgvirus (MARV) or the closely related Ebolavirus (EBOV) results in potentially lethal hemorrhagic fever in humans characterized by uncontrolled viremia, a systemic pro-inflammatory response, and multi-organ failure. Currently, there are no approved countermeasures to treat or prevent MARV infection, which leaves a critical need for development of antiviral therapies. One approach to develop antiviral therapies is exploit a virus’s dependency on the host cell and disrupt critical human-viral interactions. While multiple studies identified host-viral interactions involved in EBOV infection, we are currently limited in our knowledge of host-viral interactions that occur during MARV infection and how these interactions influence the viral replication cycle. Thus, the purpose of this research was to identify and further characterize the biological significance of human-MARV protein-protein interactions that occur during infection.</p> <p>Here, we used genome-wide yeast two-hybrid (Y2H) screens to identify directly interacting human-viral proteins. We identified 431 putative interactions with MARV and used a combination of a novel NanoLuc Y2H assay and confidence criteria to prioritize a final set of 396 interactions. Bioinformatic analysis revealed that the molecular functions of the interacting human genes were significantly enriched in RNA binding, cell adhesion, and cytoskeleton binding. </p> <p>MARV and EBOV have many similarities in their genomic organization, sequence, and protein structures that could facilitate interactions to common host factors during infection. Thus, to identify shared interactions between these related viruses, we compared the MARV interactions to EBOV interactions identified in a parallel Y2H screen. We identified 145 human proteins targeted by both MARV and EBOV. The majority (77%) of shared interactions occurred between homologous viral proteins. Additional bioinformatic analyses comparing MARV and EBOV interactions revealed that these viruses interact with different host factors with similar molecular functions (RNA binding, DNA binding, actin and microtubule binding. Together, these data support the notion that while MARV and EBOV target common host factors there are still differences in protein interactions that support functions specific to each virus. </p> <p>To investigate the biological significance of the identified interactions, we focused on host interactions with the viral matrix protein, VP40. VP40 is a multifunctional protein that facilitates viral assembly and budding from the host cell. Y2H assays using VP40 mutants revealed that the WW-containing host protein MAGI1 interacted with the late domain of VP40. This interaction was validated in mammalian cells using coimmunoprecipitation and GFP complementation assays. Based on multiple reports of WW-containing host proteins interacting with VP40, we predict that the interaction between MAGI1 and VP40 regulates viral budding.</p> <p>In conclusion, the work presented here successfully identified 396 novel human-MARV interactions, which furthers the field’s understanding of host factors involved in MARV infection. Additionally, we identified interactions shared by MARV and EBOV, which could be beneficial in the development of a broad antiviral therapy against filoviruses. Lastly, we validate the interaction between MAGI1 and VP40, which has a potential role in viral budding </p>

Multifunctional proteins that undergo major structural changes to perform different functions are known as “Transformer Proteins”, which is a recently identified class of proteins. One such protein that shows a remarkable structural plasticity and has two distinct functions is the transcription antiterminator, RfaH. Depending on the interactions between its N-terminal domain and its C-terminal domain, the RfaH CTD exists as either an all-α-helix bundle or all-β-barrel structure. Another example of a transformer protein is the Ebola virus protein VP40 (eVP40), which exists in different conformations and oligomeric states (dimer, hexamer, and octamer), depending on the required function.I performed Molecular Dynamics (MD) computations to investigate the structural conversion of RfaH-CTD from its all-a to all-b form. I used various structural and statistical mechanics tools to identify important residues involved in controlling the conformational changes. In the full-length RfaH, the interdomain interactions were found to present the major barrier in the structural conversion of RfaH-CTD from all-a to all-b form. I mapped the energy landscape for the conformational changes by calculating the potential of mean force using the Adaptive Biasing Force and Jarzynski Equality methods. Similarly, the interdomain salt-bridges in the eVP40 protomer were found to play a critical role in domain association and plasma membrane (PM) assembly. This molecular dynamic simulation study is supported by virus like particle budding assays investigated by using live cell imaging that highlighted the important role of these saltbridges. I also investigated the plasma membrane association of the eVP40 dimer in various PM compositions and found that the eVP40 dimer readily associates with the PM containing POPS and PIP2 lipids. Also, the CTD helices were observed to be important in stabilizing the dimer-membrane complex. Coarse-grained MD simulations of the eVP40 hexamer and PM system revealed that the hexamer enhances the PIP2 lipid clustering at the lower leaflet of the PM. These results provide insight on the critical steps in the Ebola virus life cycle.

<p>  </p> <p>Ebola virus (EBOV) infection remains to be a challenge to human health by its high mortality rate. Though it has been discovered for almost 50 years, there are only two antibody-based therapies approved today, and the mortality rate is still greater than 30% with the treatment. Authentic EBOV studies are strictly limited to biosafety level-4 (BSL-4) labs, which slows the development of treatment. While more simple and safer systems have been developed to understand different stages of EBOV infection, such as the matrix protein (VP40) virus-like particle (VLP) and minigenome systems, we still lack a systematical view of EBOV infection. On the other hand, mathematical modeling has been used to assist biological and medical studies for many years, as it has the advantage of integrating data and providing quantitative insight to a biosystem. In our study, we took advantage of mathematical modeling and build the primary ordinary differential equation-based (ODE-based) model of EBOV at subcellular level step by step. We built the budding pathway of EBOV VP40 first, calibrated and validated our model with experimental data. We proposed that phosphatidylserine (PS) can directly influence the stability of VP40 filaments and the budding process of VLPs. Also, the oligomerization of VP40 filaments may follow the nucleation-elongation process. Next, we conducted in-silico simulation to evaluate the treatment efficiency of fendiline, a drug lowering cell membrane PS level, in treating EBOV. We found that while in general, fendiline can decrease VLP production, there can be fendiline-induced VLP production increases at certain time points due to slow filament growth or fast VLP budding rates. Also, we concluded that fendiline is relatively more effective when applied in the budding stage of EBOV life cycle. Moreover, fendiline efficacy may increase when applied with a VLP budding step targeted treatment. Finally, we integrated nucleoprotein (NP) into our model. We reproduced the two-stage interaction between NP and VP40 and predict that NP increases VLP production through influencing filament oligomerization and VLP budding steps. Also, the dual-effect of NP on VLP production may exist, as a too high NP/VP40 production ratio can decrease VLP production. From the aspect of protein expression time, we found that a bit earlier NP production than VP40 production is beneficial for both inclusion body-containing (IB-containing) VLP production and prevention in energy waste on production of VLPs without IBs. Overall, we have built a solid foundation towards a mathematical EBOV model and demonstrated the value of models in assisting experimental EBOV studies.</p>

<p dir="ltr">The Ebola virus (EBOV) is a filamentous RNA virus which causes severe hemorrhagic fever. It is one of the most dangerous known pathogens with a high fatality rate. Multiple outbreaks of EBOV have occurred since the 1970s with the most widespread outbreak starting in December 2013. This outbreak continued through May of 2016 and had a fatality rate of approximately 50%. EBOV outbreaks are recurrent because the virus is still present in animal reservoirs. Despite multiple EBOV outbreaks we still lack a clear understanding of how new viral particles are formed and spread through virus assembly and release. Given the widespread global travel, EBOV now poses a threat to the entire world. EBOV encodes for the matrix protein, VP40, which is one of the most conserved viral proteins. VP40 can form different structures leading to different functions of the protein in different stages of the EBOV life cycle. The VP40 dimer traffics to the inner leaflet of the plasma membrane to facilitate assembly and budding. The VP40 octameric ring has been implicated in transcriptional regulation. This thesis focuses on understanding in further detail the determinates of VP40 plasma membrane assembly and exit from an infected cell.</p><p dir="ltr">The assembly and trafficking of VP40 to the plasma membrane requires a network of protein-protein and lipid-protein interactions (PPIs and LPIs). Studying these interfaces is important for understanding how VP40 structure and function regulates trafficking and assembly and can shed light on therapeutic strategies to target EBOV. The alteration of host cell Ca²⁺ levels is one of the strategies that viruses use to perturb the host cell signaling transduction mechanism in their favor. Evidence has emerged demonstrating that Ca²⁺ is important for the assembly and budding of EBOV in a VP40-dependent manner. The relationship between intracellular Ca²⁺ levels and EBOV matrix protein VP40 function is still unknown. In this work we utilize biophysical techniques to study the role of LPIs and intracellular Ca²⁺ on VP40 dynamics at the plasma membrane and key residues for assembly and budding. This work highlights the sensitivity of slight electrostatic changes on the VP40 surface for assembly and budding and a critical interaction between Ca²⁺ and the VP40 dimer that are important for lipid binding at the plasma membrane.</p>

<p dir="ltr">The recent global pandemic, COVID-19, has revealed to all the importance of understanding the complex relationship between viruses and hosts. Before COVID-19, I started my study of viral protein-host lipid interactions in the hemorrhagic fevers Ebola and Marburg viruses. These viruses contain a matrix protein that interacts with the plasma membrane to facilitate the formation of both authentic viruses and virus-like particles. My goal was to understand the limitations of their specific host lipid interactions. However, when the COVID-19 pandemic began, so to be our swift response in the development of a biosafety level 2 compatible model. This model can be used for studying severe acute respiratory distress syndrome 2 (SARS-CoV-2) assembly, egress, and entry. This model enabled exponentially greater access to more facilities to study the intricacies of SARS-CoV-2 assembly. With more access to studying the virus in a safe model, our goal is to push the understanding of viral assembly faster. I then began to take apart the individual pieces of the model and started to look at understanding the roles that they play independently. The membrane protein is the most abundant structural protein and I studied the specific lipid interactions of the soluble fraction of the protein. Physicians observed nucleocapsid protein mutations in the clinic with the increasing number of SARS-CoV-2 variants that are on the rise. The microscopy data collected can give us more insight into perhaps how the nucleocapsid protein induces the formation of filopodia structures at the plasma membrane. The envelope protein proved to be a challenge, but I determined a specific envelope and ceramide interaction in cells. The envelope protein was also causing the formation of microvesicles for an undefined function. I was able to determine the subcellular localization of the protein to the mitochondria. The localization to the mitochondria appears to induce depolarization of the mitochondria membrane action potential and induces the increase in mitochondria dysfunction signal, cytochrome c. Although the mitochondria were dysfunctional, there was no increase in apoptosis signal in the presence of the protein alone.</p>