Hydrogen sulfide (H₂S), known for its distinct smell of rotten eggs, is recognized as the third endogenous gaseous signaling molecule, alongside nitric oxide and carbon monoxide. Often described as a double-edged sword, H₂S exhibits both cytoprotective and cytotoxic properties depending on the biological context. A 2018 study suggested that H₂S enhances the efficacy of doxorubicin (Dox), an anticancer drug, by promoting apoptosis and reducing colony formation in HepG2 cells, even restoring drug sensitivity in resistant cells. However, my preliminary experiments indicated a protective role of H₂S in HepG2 cells under stress, particularly when treated with NaSH (an H₂S donor). Rather than inducing apoptosis, H₂S appears to support cell proliferation and regulate reactive oxygen species (ROS) production. My research project aims to identify H₂S -dependent pathways in HepG2 cells under oxidative stress. Using Dox as a stress inducer, I conducted viability and cytotoxicity assays, demonstrating that supplementation with 250 µM NaSH at 0 and 12 hours significantly restored cell survival. To investigate the molecular mechanisms, RNA-seq analysis identified 2,996 differentially expressed genes in the H₂S + Dox group compared to Dox alone. Principal component analysis (PCA) revealed distinct transcriptomic profiles, while KEGG enrichment analysis highlighted significant alterations in genes within the PI3K-Akt pathway. To further validate these findings, I plan to perform flow cytometry and western blot analysis. While the role of H₂S continues to be debated, my data suggest a protective function in liver cells against Dox-induced stress via the PI3K-Akt pathway. Understanding these mechanisms could pave the way for new therapeutic strategies aimed at maintaining or increasing H₂S levels to support cell health in diseases characterized by oxidative stress, such as cancer and diabetes.

Influenza viruses are a causative agent of seasonal flu outbreaks, which are mitigated through routine vaccination. Due to antigenic drift, many illness-causing strains evolve slower and are therefore, well-characterized. However, new strains occasionally emerge from animal reservoirs through antigenic shift, which can evade pre-existing immunity and cause lethal pandemics. Currently, H5N1 strains are of global health concern. Influenza viruses have two major antigenic surface glycoproteins: hemagglutinin (HA) and neuraminidase (NA), which have opposing functions and depend on a host cellular receptor, sialic acid. HA binds sialic acid for virus entry while NA cleaves sialic acid for viral release. NA is a dimer of dimers with several distinct domains, and two of particular interest: a head domain with sialidase activity and a flexible, hypervariable stalk domain. It is suggested that stalk length alters the range of accepted substrate-enzyme geometries of the NA head. As such, it is hypothesized that stalk length influences NA expression levels, sialic acid cleavage, and head tilting. Recent literature also demonstrates that shorter NA stalks result in reduced viral fitness in human hosts. Characterizing the structural effects of different NA stalk truncation constructs will provide valuable insight into influenza host-virus interactions. HDX-MS is an excellent tool for determining the structural dynamics of NA head regions by measuring local backbone amide solvent accessibility. MS data provides a detailed profile of deuterium uptake kinetics, effectively identifying differences in NA head flexibility across constructs. Additionally, we will use negative stain electron microscopy to observe differences in NA quarternary configuration and head tilting. We plan to correlate structural changes across constructs to changes in NA native function using a variety of NA activity assays in further experiments. This ongoing study aims to inform about how NA stalk length affects the influenza replication cycle, pathogenicity, and broader implications on host immunity.

Liposomes are synthetic vesicles composed of phospholipids that are used as both a model biological membrane and drug-delivery system. Doxil® is a widely used liposome-based chemotherapy drug used to treat ovarian cancer, multiple myeloma, and Kaposi’s sarcoma. Liposome stability affects drug-delivery efficacy. Cholesterol is a key component of membranes that has been shown to regulate membrane fluidity, permeability, and overall structure. Electrostatic interactions between phospholipid headgroups also can impact liposome stability and are impacted by buffer conditions. While it is known that inclusion of cholesterol and electrostatic interactions can impact liposome stability, how these changes influence membrane structure and stability is poorly understood. Cryo-electron tomography (CryoET) is an electron microscopy technique that produces high resolution 3-dimensional images of macromolecular structures, allowing detailed visualization of lipid bilayers and membranes. Cryo-ET can be used to preserve native hydration of membranes in order to maintain lipid organization. Using Cryo-ET, we plan to study how inclusion of different cholesterol concentrations and phospholipid compositions can influence membrane architecture and stability. We hypothesize that we will be able to directly visualize and analyze structural changes in membrane leaflets and membrane fine structure, which will enhance our understanding of lipid membrane architecture. An in-depth understanding of how cholesterol concentrations in liposomes under various buffer conditions influences membrane architecture will provide insight into how these factors directly impact membrane architecture and thus liposome stability. This knowledge is crucial for optimizing liposomes as drug delivery systems, improving their stability and efficiency, and enhancing their use as model membranes for studying biological processes.

In the U.S., Colorectal Cancer (CRC) is the third most common cancer in both men and women and the second leading cause of cancer-related death among individuals over 60. In the past year alone, 53,010 deaths were caused by CRC. While several post-diagnoses treatments exist, preventative treatments are notably limited. A significant contributor to CRC development is enterotoxigenic Bacteroides fragilis (ETBF), a mutant gut bacterium that secretes B. fragilis toxin (BFT). BFT alters signaling pathways in the intestine, producing reactive oxygen species, DNA damage, and carcinogenesis. Patients with Ulcerative Colitis and Inflammatory Bowel Disease exhibit higher quantities of ETBF in their intestine, placing them at higher risk for CRC. There is an urgent need to develop an effective and low cost therapeutic that eliminates the carcinogenic effects of BFT protein and mitigates CRC development in at-risk populations. In my research, I am using deep learning (DL) methods to design cyclic peptide inhibitors targeting BFT. Preliminary data from our lab has identified a promising BFT binding site that has informed my design process. Using RFpeptides, a DL based protein design software, I generated thousands of cyclic peptide backbones for this site. Next, I used ProteinMPNN, a DL sequence design tool to generate 10 optimized sequence variations per backbone resulting in ~10,000 potential binders. Finally, I filtered these design models using AlphaFold, a machine learning based structure prediction tool which assessed the efficacy of proteins to fold and bind as designed. I am chemically synthesizing the top selected binders and characterizing their binding affinity and kinetics towards BFT. If the designed binders inhibit BFT, they will serve as a basis for an effective and low cost preventative therapy for CRC in at-risk populations, reducing incidence and potentially saving thousands of lives.