Epigenetic proteins modify the chromatin structure to manipulate gene expression, and dysregulated epigenetic modification in cells has been linked to cancer formation. Previous studies on young female Drosophila have shown that after injury, germline stem cells (GSC) are capable of entering and exiting a protective state called quiescence. When exposed to ionizing radiation (IR), the apoptotic differentiating daughter cells send a protective signal to GSC, resulting in GSC quiescence. This survival behavior of GSCs validates them as a potential model for cancer stem cells, which are a subset of tumor cells that are capable of withstanding traditional chemotherapy through reversible quiescence, resulting in future tumor relapse. To identify genes required for GSC survival, we performed a spatially restricted RNA interference (RNAi) screen. Here we show that two members of the repressive epigenetic regulator complex PRC1, Pc and Sce, are required for entry, while demethylase Utx is required for exit of GSC quiescence. Notably, PRC2 dependent H3K27me3 marks are required for PRC1 function, and Utx is required to erase these PRC2 dependent H3K27me3 marks. Importantly, we detected around a 3-fold increase in H3K27me3 marks in GSC following IR, suggesting that the repressive PRC1-PRC2 dependent complex is critical for entry, and elimination of PRC2 dependent marks is critical for exit from the quiescence state. Furthermore, we show that Trx, a writer enzyme which promotes euchromatin formation through H3K4me1 addition, is required for GSC exit from the quiescent. These data suggest that reversible quiescence in GSC is controlled by specific epigenetic states. In the future, more work is needed to investigate gene specificity of the epigenetic regulation.

Humans are incapable of regenerating a majority of their major organs and tissues following traumatic injury, often resulting in an irreversible loss of function. Tadpoles of the frog genus Xenopus can regenerate multiple tissue types in response to injury, however this capability is lost after metamorphosis. This stage-specific regenerative capacity makes Xenopus a uniquely powerful model for studying factors that promote regeneration. Tadpoles develop ex-utero and do not develop mouths until days after fertilization. Before tadpoles are able to ingest exogenous food, they rely instead on maternal yolk stores for sustenance. A regenerative refractory period has been described in tadpoles of Xenopus laevis, in which regenerative capacity is transiently lost. In this study we describe a similar refractory period in the closely related Xenopus tropicalis, and observe that the onset of the refractory period aligns with the transition independent feeding. Based on this observation, we hypothesized that the lapse in regenerative capacity could be due to a lack of metabolic fuel. We used immunohistochemistry (IHC) against the yolk protein vitellogenin (vit) to study the utilization of maternal yolk stores during tadpole development. We find that yolk localization is dynamic over the course of development, and that it is ultimately is depleted by the onset of the refractory period. We additionally used IHC against phospho-Histone 3 (pH3), a marker of mitosis, to study proliferation during development and regeneration. We found that proliferation declines across development heading into the refractory period, in both uninjured and amputated contexts. Lastly, we successfully rescued both regeneration and proliferative rates by feeding tadpoles after they develop the ability to eat. As a whole, this work articulates that nutritive stress may contribute to the loss of regenerative capability in the refractory period, and that alleviation of this stress promotes regenerative ability in this context.

This research focuses on dissecting the molecular mechanism of cardiac regeneration in the animal model, zebrafish, upon a myocardial infarction like injury. Zebrafish are one of the few vertebrates that can fully regenerate their hearts after an injury in 30 days. This phenomenon is not seen in humans, who generate scar tissue after this injury with reduced circulatory efficiency. However, there is evidence that neonatal mice under 7 days old can regenerate their hearts, but this is lost upon adulthood. Determining this pathway is the first step to develop therapeutics in order to provide relief to people suffering from cardiac injuries. In this research, we used chemically ablated transgenic zebrafish to generate a 30% injury. We determined that upon an injury, both the Wnt pathway and the mTOR pathway are sequentially activated and upregulated to restart cardiac proliferation to regenerate the heart. Wnt pathway proteins like Axin and β-catenin are activated 3 days post injury and mTOR proteins like pS6 are activated gradually over 7 days post injury. The inhibition of the Wnt pathway using DKK showed a downregulation of the mTOR pathway and downregulation of cardiomyocyte proliferation. Inhibition of the mTOR pathway using Rapamycin also stopped cardiomyocyte proliferation from occurring. Mass spectrometry data showed a decrease in glutamine and an increase in leucine during the proliferative phase. Since leucine is one of the activators of the mTOR pathway, we see that the glutamine-leucine transporter is also upregulated post-injury. Thus, we show that heart regeneration in adult zebrafish occurs via cardiomyocyte proliferation by using the Wnt and mTOR pathways to upregulate cardiomyocyte proliferation upon injury.

The Polycomb Repressive Complex 2 (PRC2) is an important epigenetic remodeler in developmental transitions and cell fate determinations. PRC2 is responsible for the addition of H3K27me3 marks that repress developmental gene expression. The catalytic subunit of PRC2 is the methyltransferase (Enhancer of Zeste 2) EZH2 which binds to EED (Embryonic Ectoderm Development) to methylate H3K27 on gene promoter regions. To investigate the requirement of PRC2 in different developmental transitions, a computationally designed protein was utilized to inhibit EED-EZH2 interaction. The novel designed protein is named EED binder (EB) and competes over endogenous EZH2 on the EED binding cleft with 300 times greater affinity than endogenous EZH2. We cloned EB-GFP under heatshock inducible promoter and injected this construct to one cell zebrafish embryos to generate a germ line transmissible insertion. To study the requirement of PRC2 in early developing embryos (0-3dpf), we applied heatshock (HS) on EB-GFP positive and negative embryos. Western blot analysis revealed global downregulation of EZH2 and H3K27me3 in EB-GFP positive, but not control embryos. Additionally, Co-Immunoprecipitation experiments showed EB-GFP binding to EED. Finally, to test the requirement of PRC2 in caudal fin regeneration, adult (5 month old) EB-GFP positive and negative animals were fin-amputated and the regeneration growth rate was measured for 14 days. Our results show that EB-GFP positive fish were able to regenerate their fins faster, resulting in a large fin size compared to either negative or non-HS clutch mate. Overall, we have developed a computer designed inducible PRC2 inhibitory system to study PRC2 function in Zebrafish, at the whole animal level. In the future, we will utilize EB-GFP to explore PRC2 and other epigenetic modifiers that are required for tissue and organ regeneration before and after injury.

Viral pathogens, especially those that undergo rapid mutagenesis, pose a significant threat to public health. Viruses that exemplify this issue include influenza, HIV, and Ebola. Given the low efficacy of seasonal vaccines for influenza, our project focuses on improving existing influenza vaccinations. Instead of using conventional methods, such as injecting inactivated pathogens or viral subunits, mRNA sequences encoding the viral hemagglutinin (HA) fused to our recently-developed self-assembling I53-dn5 nanoparticle platform will be administered in vivo. The organized array of the protein platform can lead to stronger B-cell crosslinking and a robust immune response. However, characterization of I53-dn5 in vitro is critical before use in vaccination studies. My work focused on optimizing the expression, secretion, and assembly of the I53-dn5 protein platform. To mimic in vivo conditions, I transfected DNA encoding HA-fused I53-dn5 into HEK293F cells. Past experiments have shown that when dn5A and dn5B are transfected separately, they express at disproportional concentrations. To resolve this, we encoded both components onto one DNA plasmid for transfection. However, with this new approach, we also needed to cleave the two components after expression. To do so, we incorporated different cleaving peptides, such as T2A and Furin cleavage sites. Through western blots, SDS page electrophoresis, SCC protein purification, and electron microscopy, I analyzed how these cleaving peptides impacted assembly and secretion of the protein platform. Once we are able to consolidate an effective model, we will be able to start in vivo studies. Furthermore, if effective, our model can be used to create vaccinations against other viral illnesses, including HIV and coronavirus.

My research project involves the AP-3 (adaptor protein) complex, which plays a key role in membrane trafficking within cells. Vesicles (small, membrane bound structures) mediate the transport of proteins and lipids among cellular organelles. These vesicles are created in various ways by proteins throughout the cell, with some including a protein "coat" around the vesicle as it travels to its destination. AP-3 is a coat protein complex that mediates vesicular transport from the trans-Golgi network to the lysosome. We have utilized a gene reporter system called GNSI to analyze AP-3 function in different genetic backgrounds of Saccharomyces cerevisiae (baker’s yeast). One output of the reporter system is a colorimetric assay that measures the intensity of colored halos around yeast colonies grown on an agar plate. We found that targeted truncations of proteins of the AP-3 subunits Apl6 and Apl5 at the C-terminus of AP-3 (specifically the “ear domains”) resulted in increasing defects in AP-3 trafficking ability. However, it is unknown whether the defects are in membrane recruitment to the trans-Golgi network, or whether recruitment occurs but vesicle budding defects arise. Therefore, the current aim of the project is to begin analysis of the truncation defects in AP-3 by fusing Apl5 trunctions with a C-terminal mNeonGreen fluorescent protein and use live cell fluorescence microscopy to further examine AP-3 localization These basic studies will further our understanding of membrane trafficking and may provide insight into diseases linked to AP-3 function, including HIV-1 particle assembly and human genetic disorders.

Natural proteins often assemble into various complex geometric structures based on their interactions with each other. The King Lab at the University of Washington's Institute for Protein Design uses the way these proteins behave to develop computational models that enable the design of novel self-assembling protein cages, or nanoparticles. The designed particles are capable of holding and transporting molecules or displaying antigens on their surface, making them effective vaccine candidates. My project involves recovering the solubility of one of these protein cages known as T33_dn2. T33_dn2 is a tetrahedral protein cage comprised of four copies each of two trimeric components known as T33_dn2A and T33_dn2B. While both components can be expressed individually through E.coli before being assembled in vitro, they can also be expressed bicistronically and assemble in vivo. Currently, the use of T33_dn2 as a vaccine scaffold is limited because T33_dn2B is insoluble, and only seems to be stabilized in solution when associating with T33_dn2A. When expressed bicistronically, however, the cage has an extremely low yield. For a protein to be developed into a vaccine, it must be soluble. To recover the solubility and yield of T33_dn2B, I am testing ten plasmid variants of bicistronic T33_dn2. The “original” plasmid consists of one gene coding for a high-expressing cleavable SUMO protein attached to T33_dn2A and another coding for T33_dn2B. The additional nine variants have single point mutations at specific locations on the T33_dn2A gene intended to affect binding strength. After expression, introducing wildtype T33_dn2A in vitro will allow for the formation of T33_dn2. I will be presenting the results of these expression, purification, and assembly tests.

Humans are incapable of regenerating a majority of their major tissues following traumatic injury. Tadpoles from the frog species Xenopus tropicalis have the ability to regenerate lost spinal cord, vasculature, muscle, and cartilage within a few days following injury. The regulatory mechanisms of gene expression necessary for regeneration have not yet been well defined. My primary interest is in understanding the relationship between stress signaling and gene expression during regeneration. The lab has shown that the stress responsive transcription factor Hypoxia Inducible Factor 1α (Hif1α) is necessary for the expression of Wnt target genes, one of the primary signaling processes necessary for regeneration. While we have found that Hif1α is necessary for Wnt target gene expression, we do not know the epistatic relationship between Hif1α and Wnt. In order to test this relationship, I utilized the drug IWR to antagonize Wnt and found that tadpoles treated with IWR have reduced tail regeneration 72 hours post amputation (hpa). I then supplemented these tadpoles with DMOG to stabilize Hif1α and found that DMOG is sufficient to rescue tail regeneration, suggesting that Hif1α is downstream of Wnt. In order to determine if Hif1α is sufficient for Wnt target gene expression, I extracted RNA from regenerating tails 24 hPa and used quantitative PCR (qPCR) to determine relative gene expression. I also utilized in situ hybridization to see if expression of these genes is restricted to regenerating tissues. As Wnt is a known regulator of neural and muscle development, I investigated how inhibiting Hif1α would impact complex tissue regeneration and found that Hif1α is necessary for regeneration of axons and muscle specifically. By determining the epistatic relationship between Hif1α and Wnt through the analysis of specific gene expression, we continue to improve our understanding of how regenerative organisms convert stress signals to cell fate signals.

The link between protein sequence and structure is not always apparent. The dogma is that sequence determines structure, but it is not clear how very different sequences can give rise to the same structure. Here, we employ high temperature molecular dynamics unfolding simulations to probe the pathways and specific interactions that direct the folding and unfolding of the SH3 domain, a family of small proteins consisting of two β-sheets arranged to form a barrel. SH3 domain proteins are involved in various functions including protein binding, cell signaling, and nucleic acid modification. The SH3 metafold consists of 753 proteins with the same structure but varied sequence and function. To investigate the relationship between sequence and structure, we selected 17 SH3 proteins with an average pairwise sequence identity of only 27%. Six unfolding simulations were performed for each protein and unfolding transition states were determined, revealing two unfolding/folding pathways. Transition states were also expressed as mathematical graphs of contacts between chemical groups, and three positions in the transition state structure were consistently more connected to the rest of the graph than other nearby positions. These positions represent a folding hub connecting different portions of the structure in the transition state. Analysis of the multiple sequence alignment and covariation also highlighted positions with high conservation due to packing constraints and long-range contacts. This study demonstrates that the SH3 domain can fold through two distinct pathways, but certain folding/unfolding characteristics are conserved independent of sequence and unfolding pathway. By identifying similar interactions, we demonstrate how different sequences can have the same influence on folding pathway and final structure.

Foldit is a citizen science game developed at the University of Washington that allows people to predict and design protein structures for use in scientific research. The game challenges ordinary people to map the complex energy landscape of a protein sequence to find the lowest energy structure as determined by the protein modeling software Rosetta. The lowest energy structure is the one most likely to fold. While Rosetta’s software uses random sampling to predict a protein’s structure from its sequence, Foldit takes advantage of people’s intuition about things like avoiding steric clashes and large cavities to produce structures that can rival or exceed the accuracy of what Rosetta’s software can produce. However, there are still some flaws in Foldit that can cause a large fraction of player designs to not fold correctly. One of the biggest issues concerns an excess of buried unsatisfied polar atoms in player designs caused by imperfect modelling of the electrical potential energy of these atoms in the game. These atoms can prevent a protein from folding by eliminating the energy gap between folded and unfolded states. We have introduced a score penalty which penalizes designs with buried polar atoms. We are using existing scientific software to determine these atoms, but we have had to modify the relevant code to make it fast enough to be playable while retaining accuracy. We have also tested our calculations against known ways of calculating these atoms, and against previous Foldit puzzles to see if adding the penalty improves designs. We hope to even test this filter on some recent Coronavirus binder puzzles to get better solutions. An increase in design quality would mean that such a score penalty is warranted and it would allow for Foldit structures to be more useful across a range of protein design applications.

Cannabidiol (CBD), an active compound in marijuana that provides diverse health benefits such as treating epilepsy, anxiety, inflammation, and chronic pain, is increasingly used in the United States. However, little is known about the pharmacological effects of CBD on neurological diseases. Although the chemical-induced dimerization (CID) system, in which dimerization binder and anchor binder dimerize only in the presence of small molecules, has been well established, very few studies have applied it as a biosensor, especially to detect CBD. This hinders our exploration of the medical values of CBD. Here we utilize the nanobody-based CID system to develop a novel CBD biosensor, consisting of split nano luciferase - a reporter protein, CBD induced CID - a sensor protein, and glycine-serine linker - a protein linker, which ensures the biosensor efficiency. We validated the biosensor between Dimerization binder 1(DB1) and CBD anchor 14 (CA-14) both in vitro and in vivo. From our previous in vitro data, we predict that the biosensor containing Dimerization binder 4 (DB4), an analog to DB1, can detect CBD at a higher concentration in vivo than DB1, which detects better at lower CBD concentrations. We use split-luciferase assays to test the binding affinities of DB4 for a better biosensor with a broader range of drug concentration-response evaluation. This study demonstrates an effective method to maximize the CBD biosensor system, extending the applications onto detecting drugs in several brain regions, different cell populations, and even subcellular components, thus, furthering the understanding of physiological mechanisms and therapeutic potentials of CBD.

Mitosis results in two genetically identical daughter cells, each containing their own nucleus and set of replicated chromosomes from the original parent cell. Inaccurate chromosome segregation can result in severe consequences like cancer and developmental defects. During mitosis, the replicated chromosomes line up at the center of the cell in preparation to be pulled apart by microtubules. Microtubules are dynamic cytoskeletal components that provide the forces necessary to pull the chromosomes towards their respective daughter cells. The ends of the microtubules attach to the kinetochore, which is an assembly of protein complexes located on the chromosome. Accurate segregation of these chromosomes relies on the ability of the kinetochore to strongly bind chromosomes to microtubule ends. Ndc80 complex is an outer-kinetochore component that binds microtubule ends and is required for proper segregation. Emerging cellular data suggests that multiple Ndc80 complexes interact with one microtubule end. We seek to assemble a particle of multiple Ndc80 complexes in vitro, which may model the native kinetochore-microtubule interface more closely. We utilized a designed protein, HBRP, that forms a hexamer in solution and was modified to couple with and artificially oligomerize any protein of interest. We optimized the coupling rate and completion degree of three different variants of the HBRP protein with Ndc80 complex to ensure a complete hexamer particle assembly. Successful formation of this particle assembly will allow us to better understand the binding mechanism of the kinetochore to microtubule ends.

 Chemically induced dimerization (CID) systems, in which a pair of proteins only dimerizes in the presence of a specific ligand, have wide use for the biosensing of small molecules, such as drugs, metabolites, and signaling molecules. However, few CID systems are currently available, and thus, until recently, little work has been done to design effective and highly specific CID-based biosensors, limiting the diversity of our small molecule detection toolkit. Previously, we established a highly efficient and generalizable method for de novo engineering of new CID systems, and demonstrated its effectiveness by designing a CID system specific to cannabidiol (CBD). Here, we engineer circularly permutated green fluorescent proteins (cpGFP’s) to have the cannabidiol CID system flanking it, allowing for small molecule detection via protein dimerization to be translated into measurable changes in fluorescent signals. As an experimental platform to screen high-performance designs, we employ the yeast display technique, where yeast cells are engineered to display on their surfaces any cpGFP-CID biosensor constructs we wish to screen. The lengths and compositions of the linker regions connecting the cpGFP and the dimerization proteins play a crucial role in the efficiency of the biosensor. Currently, we seek to understand what linker lengths and contributions help optimize biosensor performance. Initially, we introduced linkers with lengths of about 20 residues that are composed of alternating glycine and serine residues. Biosensor constructs with these linkers exhibited minimal, baseline performance. We generate biosensor construct libraries by mutating the linker regions, displaying this library on yeast cells, and performing high-throughput screening to identify optimal linker lengths and compositions for our biosensor constructs. Afterwards, we will validate the biosensor’s performance in vivo and in vitro. Our research will not only contribute to fluorescent and CID-based biosensor design and study, but will also aid in the expansion of clinical small-molecule detection toolkits.

Computational protein design is an emerging field that takes advantage of first principles derived from biological protein-protein interactions and explores the protein space that nature has yet to evolve. The Baker Lab developed a software called Rosetta that enables researchers to explore this space and create brand new proteins more stable than those produced in biological systems via evolution. This software has been adapted to take advantage of a concept that exists everywhere in nature- symmetry. Using this concept, I am building higher-order protein assemblies including protein nanocages and three-dimensional protein crystals. Researchers at the Baker Lab have developed a hierarchical approach to engineer these highly symmetric, complex structures. This hierarchical approach involves combining protein building blocks with different symmetric topologies multiple times to facilitate higher-order symmetric assembly of a three-dimensional protein crystal. By breaking up crystal symmetries into their constituent building blocks, we can design these higher-order symmetries with greater accuracy and troubleshoot experimental difficulties by pin-pointing structural deviations along the way. Here, I will describe my experience using this approach to create a protein crystal from symmetric building blocks.

Cancers are broadly characterized by changes in cell metabolism. Tumor cells typically exhibit functional respiration and inhibition of electron transport chain can impair cancer cell proliferation. However, certain neuroendocrine cancers can arise from loss of function (LOF) mutations in succinate dehydrogenase (SDHx/complex II), which plays a key role in the TCA cycle and in mitochondrial respiration. SDH, which catalyzes the conversion of succinate to fumarate, comprises four subunits: A, B, C and D. LOF mutations in subunits B, C, and D can promote tumorigenesis and mutations in subunit B (SDHB) are particularly associated with malignant and metastatic neoplasms. Interestingly, SDHB impaired cells show an accompanied loss of activity in complex I, implying that unlike the majority of cancer cells, respiration is not essential and may even be antagonistic for SDHB mutant cancer cell proliferation. Indeed, preliminary experiments indicate that inhibition of complex I can restore proliferation to cells treated with an SDH/complex II inhibitor. However, the molecular mechanisms behind this phenomenon are not well understood. We aim to investigate the metabolic mechanisms by which dysfunctional respiration is essential for the proliferation of SDH impaired cells. We hypothesize that inhibition of respiration in these cells can prevent oxidation of NADH to NAD+ at complex I and alter the redox homeostasis in the mitochondria to support proliferation. Specifically, we will test to see if increasing the NADH/NAD+ ratio is the required function of complex I inhibition that rescues cell proliferation in SDH impaired cells. In addition, we will characterize the metabolic consequences of specific alterations SDH, complex I, and mitochondrial redox state. Results from this study should allow us to delineate the importance of metabolic alterations in SDH mutant cancer cells and potentially help identify metabolic vulnerabilities for treatment of SDH impaired cancers.

 Cancerous cells have a modified metabolism that supports their demands for increased proliferation. One of the essential molecules in cancer cell metabolism and proliferation is the amino acid aspartate. Aspartate is not only incorporated into proteins, but is also a substrate for nucleotides and other amino acids, including asparagine. Aspartate availability can constrain tumor growth rate, and the consumption of aspartate to generate downstream products can alter aspartate levels. One gene that draws from the aspartate pool is asparagine synthetase (ASNS). ASNS converts aspartate into asparagine, which is used in the production of proteins, but does not increase cell proliferation. Thus we hypothesize that ASNS expression and activity can affect aspartate levels. With this, we aimed to determine if ASNS expression could alter aspartate availability and change sensitivity to aspartate suppressing therapies. Since cancer cells express ASNS to varying degrees, this project sought to determine if ASNS expression could be used to identify those cancers that are most amenable to aspartate suppression therapies. This research sought to better understand the conditions that determine aspartate levels, and how to exploit those conditions to inhibit tumor growth in association with asparagine synthetase.

Ras is classified as a group of proteins that is involved in many signaling processes in the cell. Ras plays key roles in homeostatic cell proliferation, differentiation, and movement. However, the oncogenic point mutation from glycine to valine in the 12th position of H-Ras protein (denoted by RasV12) induces tumors and gives rise to metastatic cell behaviors. Cancer metastasis is characterized in collection of several steps, but the initial step – dissemination of cancer cells from the tumor of origin into the bloodstream – is not well studied. In this project, we use Drosophila Melanogaster as a model organism to screen for genes involved in the process of dissemination. Overexpression of RasV12 in intestinal epithelial cells caused them to disseminate from the intestine. Knockdown of genes known to be critical for the process suppressed the cell dissemination phenotype. Our image analysis in observing relative quantities of GFP-expressing RasV12 cells in the intestine allowed us to discern the importance of a gene on the cell dissemination process. In result, we identified several kinases and phosphatases involved in different steps of the cell dissemination process.