Memory has been widely studied for its crucial role in learning and its diverse range of expression. Although the means of acquisition differ, it is generally accepted that memory goes through three encoding stages: sensory, short-term and long-term memory. The retention of memory is important as it enables us to act with the wisdom of past experience. However, although one could not survive without memory, remembering everything is also devastating. In fact, forgetting is an important cognitive feature that allows us to adapt to the constantly changing environments. Despite its importance for cognition, little is known about the molecular nature of forgetting. Here, we investigated the genes behind forgetting by studying an olfactory memory in the nematode C. elegans. Like human, worms can modify their behavior upon acquiring unpleasant experience – their movement towards preferred odor is significantly reduced after prolonged exposure to the odor during starvation. Upon returning to food (e. coli OP50), the odor attraction slowly returns to the worms within 3-4 hours, indicating a diminishing impact of the starvation experience. By contrast, the forgetting process was significantly accelerated by 1-2 hours, when worms were cultivated on pathogenic bacteria, pseudomonas aeruginosa PA14. Our genetic studies showed that a null mutation to the daf-16 gene restored recovery time to 3-4 hours, despite of the exposure to PA14. These data indicate that daf-16 plays a positive role in accelerated memory loss upon pathogen ingestion. Because DAF-16 is involved in innate immunity and stress response, our results provide a potential connection that couples the memory to environmental stressors.

Environmental stressors trigger physiological adaptations in organisms that allow them to dynamically remodel tissues. As many of these processes in the gastrointestinal tract of Drosophila melanogaster are studied in the context of stem cell proliferation, we instead chose the novel approach of investigating visceral muscle adaptations in response to stressors to the gut. The three stress conditions we employed were starvation, damage, and aging. For starvation, we treated adult flies with only water following nutrient-enriched recovery. For epithelial damage, we fed them with the chemical damaging agent, bleomycin. Finally, for aging, the flies were subjected to different prolonged time periods on normal media. The flies were then dissected and processed for confocal imaging and phenotypic analysis, allowing us to categorically define and quantify the phenotypes in order to measure the degree to which the visceral muscle has changed. We discovered that various responses of the visceral muscle are induced by the types of stressors, and will further look into whether it is adaptive remodeling, muscle repair, or other undescribed mechanisms. We will investigate the molecular mechanisms that contribute to the remodeling of the visceral muscle using RNAi approach. This research would provide valuable resources as a model system for studying complex tissue responding to environmental challenges and understanding of its mechanisms utilizing the vast genetic tools of Drosophila.

Hydrogen sulfide (H₂S) is a toxic gas in the environment, but it is also an important cellular signaling molecule. Our goal is to understand the genes and pathways that mediate the physiological effects of H₂S. Previous work from the Miller lab has shown RHY-1 to be a component of a pathway that mitigates H₂S toxicity. RHY-1 is an integral membrane protein with predicted acyltransferase activity that localizes to the endoplasmic reticulum. We are attempting to identify proteins that work with RHY-1 to promote survival in H₂S. In this project, we have optimized biotinylation by antibody recognition (BAR), a proximity-labeling approach, to identify proteins that may physically interact with RHY-1. In these experiments, we introduced an antibody conjugated to horseradish peroxidase (HRP) into C. elegans that expresses the epitope-tagged RHY 1::FLAG::GFP protein, so that upon addition of biotin peroxide, biotin radicals were formed only in proximity to RHY-1. These biotin radicals react with other proteins that are localized near RHY-1. We visualized biotinylation using a fluorescent label and showed that the biotinylated proteins colocalize with RHY-1, indicating a successful BAR reaction. To our knowledge, this is the first use of BAR in C. elegans. Going forward, we will use mass spectrometry to identify the proteins that are biotinylated, and then test the functional role of these proteins in mitigating H₂S toxicity. Identifying the genes and pathways downstream of RHY-1 that promote survival in H₂S will reveal new factors that can modulate H₂S signaling in cells and may lead to treatments for H₂S poisoning.

IMPDH1 catalyzes the rate limiting step of de novo guanine synthesis. cGMP is a critical signaling molecule involved in phototransduction in rod and cone photoreceptors. In humans, nine mutations in IMPDH1 lead to Retinitis Pigmentosa and Leber's Congenital Amaurosis; however, the cause of retinal degeneration is unknown. In zebrafish, IMPDH1a is the major variant in the retina, exclusively expressed in rod and cone photoreceptors. To understand the function of IMPDH1, we utilized an IMPDH1a knock-out (KO) zebrafish line. Loss of IMPDH1a does not lead to retinal degeneration, and cGMP levels remain unchanged. However, retinas lacking IMPDH1a show a 61.6% reduction in guanine. Since photoreceptors do not undergo cell division, mitochondrial DNA (mtDNA) synthesis would be a major use of guanine during mitochondrial biogenesis and turnover. Under normal conditions, the amount of mtDNA between the KO and WT zebrafish was the same. To eliminate the potential that KO fish could supplement guanine from their diet, fish were starved for 24 hours. Starved IMPDH1a KO showed a trend of increased mtDNA compared to WT fish. Interestingly, starved fish had about four times the amount of mtDNA for both WT and KO compared to fed fish. Starvation may increase mitochondrial DNA copy number to increase efficiency of ATP production by increased utilization of oxidative phosphorylation. Reducing guanine levels by knocking out IMPDH1a in zebrafish retina does not affect mtDNA or nDNA levels, and the photoreceptor cells appear healthy. Although more research is needed, silencing the gene encoding for IMPDH1 could be a possible therapy for those suffering from the effects of its mutation.

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 hypothesized 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, my project sought to determine if ASNS expression could be used to identify those cancers that are most amenable to aspartate suppression therapies. To test this I generated cell lines that express ASNS in different ways and treated them with multiple electron transport chain inhibitors. Preliminary results suggest that cells that express ASNS to a higher degree are more susceptible to mitochondrial inhibitors. More broadly, 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.

Mitosis produces two genetically identical daughter cells, each inheriting their own nucleus and a full set of replicated chromosomes from the parent cell. Inaccurate chromosome segregation can result in severe consequences like cancer and developmental defects. Microtubules are dynamic cytoskeletal components that provide the forces necessary to segregate chromosomes into their respective daughter cells during mitosis. The kinetochore is an assembly of proteins and protein complexes located on the centromere that binds to microtubule ends to attach chromosomes to the force-generating microtubules. The accurate segregation of 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 chromosome segregation. Emerging cellular data suggests that multiple Ndc80 complexes interact with one microtubule end to facilitate chromosome separation. In vivo data suggests multiple Ndc80 complexes are arranged around microtubules. To closely model the native kinetochore-microtubule interface, we have begun to assemble a structured particle of multiple Ndc80 complexes in vitro using designed proteins that form oligomers. The particles have different geometries and stoichiometries. A method to couple Ndc80 complex to the designed protein was developed. We then tested the coupling efficiency under different temperatures and concentrations to optimize the reaction and ensure complete particle assembly. We found that the reaction goes nearly to completion with a 3:1 ratio of Ndc80 complex to designed protein at room temperature with a reaction time of thirty minutes. We will now measure the stoichiometries of the particles, which are designed to have four, five, or six Ndc80 complexes. Successful formation of these organized particles will allow us to measure the effect of geometry and stoichiometry on the ability of the Ndc80 complex to make strong attachments to microtubules.

Photosynthetic bacteria are ideal systems for the sustainable production of chemicals, but their chemical production is limited by the efficiency with which the bacteria can capture and transfer energy. For this project, I have designed light-harvesting proteins that bind chlorophyll molecules that absorb light of the wavelength reflected by the photosynthetic bacteria Rhodobacter sphaeroides. The LH1 and LH2 light-harvesting complexes found in R. sphaeroides bind chlorophyll molecules in a manner that promotes excitonic coupling and minimizes the energy lost during transfer from the complexes to the reaction centers. The proteins I have engineered are modeled after LH2 and are designed to bind a symmetric ring of pi-stacking chlorophyll dimers, promoting high-efficiency energy transfer between light-harvesting complexes. After characterizing these light-harvesting proteins, they will be further engineered to enable their expression in R. sphaeroides, which will allow these bacteria to absorb a broader range of the solar spectrum and improve their energy transfer efficiency. Incorporating this de novo designed protein into metabolically engineered R. sphaeroides will enhance these bacteria’s ability to convert carbon dioxide to useful chemicals. These engineered light-harvesting proteins could also later be adapted for use in hydrogel-based organic solar cells and forming a 2-D sheet of light-harvesting proteins. In my presentation, I will discuss how I have used the Rosetta protein modeling suite to design hexameric chlorophyll-binding proteins. These proteins are designed to bind twelve chlorophyll molecules, and their cyclical symmetry simplifies the design process. I will present data from characterization experiments such as circular dichroism and fluorescence spectroscopy, which are methods that allow for quantification of the proteins’ binding affinities towards the ZnPPaM chlorophyll, and small-angle x-ray scattering (SAXS) and Size Exclusion Chromatography/multiple-angle light scattering (SEC-MALS), which provide information about the predominant oligomeric states of the designed proteins.

Our group, the King Lab at the University of Washington, computationally designs self-assembling protein nanoparticles for therapeutic applications. These nanoparticles make great candidates for therapeutics because of their ability to both encapsulate molecules as “cargo” and display antigens on their surface. One of the problems we face in vaccine and therapeutic design is targeted nanoparticle delivery, or making sure nanoparticles are delivered to a specific location, in vivo. Understanding where certain nanoparticles localize in vivo will be useful for determining what influences a nanoparticle’s interactions inside an organism. One approach our group is taking to examine nanoparticle biodistribution is miniprotein library display on synthetic nucleocapsids. These miniproteins are 20-50 amino acids long and were designed to produce stable, folded structures with surface patches that could facilitate binding to target receptors. Synthetic nucleocapsids, or nanoparticles designed to encapsulate their own RNA genomes, display these miniproteins in the library. The library was injected into healthy and tumor-ridden mice. RNA sequences were then obtained from blood, brain, heart, spleen, kidney, liver, lung, tumor, and dose samples. Using unsupervised and supervised learning algorithms such as Principal Component Analysis, Hierarchical Clustering, and Random Forests, I am building mathematical models that can analyze the biochemical properties of these library variants and determine why certain nanoparticles vary in biodistribution. I will also be analyzing another synthetic nucleocapsid library to answer similar questions about nanoparticle biodistribution. By constructing these models, I hope to provide tools that aid in experiments regarding targeted drug design.

Humans are incapable of regenerating a majority of their major tissues following traumatic injury. Xenopus tadpoles have the ability to regenerate a variety of complex tissues quickly following tail amputation, but lose this regenerative competency during metamorphosis. Though tadpoles have been extensively used to study regeneration, we do not yet understand the roles that stress signals from injury play in directing regenerative gene expression. We have found that inhibition of the stress responsive transcription factor Hypoxia Inducible Factor 1α (Hif1α) with the Hif1α inhibitors 2-methoxyestradiol (2ME) and Echinomycin (Ech) prevents regeneration. In particular, inhibition of Hif1α decreases Wnt mediated gene expression. Wnt is known to be one of the primary signaling processes necessary for proper Xenopus regeneration, specifically tail regeneration. While we have shown that Hif1α and Wnt regulate expression of similar gene programs in regeneration, we predicted that there are unique processes that Hif1α regulates to facilitate growth. To investigate how Hif1α and Wnt regulate regeneration, we utilized the two Hif1α antagonists, as well as the Wnt antagonist, IWR-1 (IWR) which I have previously shown inhibits regeneration. To determine Hif1α and Wnt regulated genes, we performed RNA-sequencing 24 hours post amputation. I then identified genes downregulated in Hif1α inhibited tadpoles, which are likely Hif1α dependent. I then removed genes downregulated by Wnt in order to isolate genes uniquely regulated by Hif1α. With the 250 genes uniquely regulated by Hif1α, I used PANTHER to perform gene set enrichment based on Gene Ontology terms. The main biological process of interest found to be regulated by Hif1α during regeneration was DNA replication. By determining how Hif1α uniquely regulates DNA replication during regeneration, we will continue to enhance our understanding of the roles that stress signals play in directing how regenerating cells meet increased proliferative demands post traumatic injury.

SARS-CoV-2 spreads across the globe, infecting more than 128 million people and claiming over 2.7 million lives with an absence of definitive treatment up to date. Therefore, there is an immediate need to develop treatments fighting against the COVID-19 global pandemic. The goal of my project is to generate and assess an innovative treatment for the SARS-COV-2 virus infection. Our treatment formulates computationally designed proteins, and we want to evaluate its therapeutic effects using human induced pluripotent stem cell (h-iPSC) derived cell lines and organoids. The designed protein is a combinatorial cage (mosaic cage) containing spike binders previously shown to significantly inhibit SARS-CoV-2 viral infection and F-domains that were shown to activate the Tie2 pathway. The Tie2 pathway is a key regulator of vascular stability, where active Tie2 can strengthen cell-cell junctions and enhances endothelial cell survival, thus enhancing blood vessel stability. We hypothesize that the designed protein would neutralize the spike protein to block viral entry and activate the Tie2 pathway to alleviate sepsis in COVID-19 infected patients. We will test spike-binding activity and determine the activation level of the Tie2 pathway of this mosaic cage in iPSC-derived spike-overexpressing endothelial cells. We expect to measure a strong spike-binding affinity of designed proteins and strong downstream pathway signals pAKT, pERK, pFAK in designed protein-treated iPSC-derived endothelial cells. We also plan to test the mosaic cage’s activities using Kidney Organoids. If our hypothesis is correct, we will apply the findings clinically for their potential intranasal administration as a COVID-19 therapeutic.

The spatial distribution and degree of colocalization for two or more proteins, based on their fluorescence intensities, are useful metrics for phenotyping cells and informing on biological function. Yet, most commercial and open-source image analysis tools report global colocalization statistics at the image-level, and offer limited analysis at the individual cell level. To address this, we are developing a python-based image analysis pipeline to quantify robust per-cell metrics of colocalization. Our pipeline stacks image tiff files acquired on high-throughput automated fluorescence microscopes into multichannel 3D image stacks. The images are then restored via deconvolution algorithms, and cropped to remove out-of-focus image planes using Laplacian variance algorithms. For image thresholding and cell segmentation, the pipeline incorporates scripts from the Allen Institute for Cell Sciences to threshold and segment individual cells. Finally, the pipeline calculates per cell colocalization metrics based on the fluorescent intensity of each voxel in each cell. CLARITY was used to quantify the spatial relationships of the peroxisomal biogenesis protein Pex3 with the endoplasmic reticulum protein Sec61 and peroxisomal membrane protein Pmp70. Pex3 colocalized with both Sec61 and Pmp70 and this colocalization could be manipulated by treatment with different kinase inhibitors. In several instances, these differences in localization contributed to a large variance in the measured Pearson’s Correlation Coefficient of cells within the same image. Morphometric analysis showed the volume of peroxisomes per cell negatively correlated with the number of peroxisomes per cell. The automation of per-cell image analysis leveraged in this pipeline will allow for systems-level phenotyping and data mining from fluorescent microscopy images.

Over the past fifty years, survival rates for most cancers have risen as innovative treatments have been developed. However, Diffuse Intrinsic Pontine Glioma (DIPG), a rare pediatric brainstem tumor, has seen no such improvement and remains one of the deadliest cancers. DIPGs are genetically distinguished from adult gliomas by a lysine-to-methionine mutation in a variant of histone H3 called H3.3 (H3.3K27M), found in 80% of DIPG tumors. This mutation is absent in canonical H3.1 and H3.2, yet it triggers a global reduction in the levels of polycomb repressive complex 2 (PRC2) mediated H3K27me3 (tri-methylation) marks, which is traditionally considered a driving event in tumorigenesis. However, genome-wide studies of H3K27me3 marks in DIPG cells have revealed that this global loss of H3.3K27me3 is accompanied by a sharp increase in H3K27me3 repressive marks at certain genes. These trimethylation spikes represent regions with high transcriptional repression and may be key in understanding and treating DIPG. Our lab has previously described the inhibition of PRC2 using a computationally designed protein EBdCas9. Using a complementary guide-RNA tiled to the promoter region of a gene, I can target EBdCas9 to that specific region on the genome, remove existing H3K27me3 marks, and increase gene transcription. Identification of candidate genes which are repressed in DIPG cells under these H3K27me3 spikes remains an ongoing project that I am working on. Results of one target, tumor-suppressor gene p16, demonstrate that I can transfect primary DIPG cells with EBdCas9 plasmid and a p16-specific gRNA to trigger a 20-fold increase in p16 expression. I hypothesize that restoration of critical cell-cycle genes like p16 will reduce DIPG viability and offer potential therapeutic targets. I’m currently working to improve transfection efficiency in order to increase p16 expression, and future steps include in-vivo testing of this system, either via mouse model or a brainstem organoid.

Learning and forgetting are two key processes that keep our memories in balance. In the past few decades, we have learned a great deal about mechanisms associated with memory formation and consolidation. However, little is known about the molecular mechanisms of forgetting, despite its importance in human health. Here, we take advantage of the nematode C. elegans – a living animal with a simple nervous system of 302 neurons – to explore the mechanisms behind forgetting. In particular, we focus on the decay of associative olfactory learning and the regulation of this decay after experience of pathogenic bacteria. Previous studies have shown that worms acquire an associative memory linking starvation experience and the olfactory response. After prolonged exposure to a preferred odor during starvation, worms exhibit a diminished response towards the preferred odor. However, upon returning to a food source, the attractive response toward the preferred odor recovers within 3-4 hours, indicating the loss of the associative olfactory memory. We found that the rate of memory loss, quantified by measuring the time course of recovery of the olfactory response, depends on the type of food source (bacterial strain) that worms experience. Specifically, exposing worms to pathogenic bacteria PA14, compared to the regular food source OP50, leads to a quicker loss of the associative olfactory memory. Our results further show that the acceleration of memory loss is mediated by a conserved transcription factor DAF-16/FOXO, as daf-16 mutants exhibited similar rates of memory loss regardless of OP50 or PA14 experience. Together, these findings demonstrate an unexpected role of DAF-16/FOXO in memory decay induced by exposure to pathogens. 

Cell quiescence is defined as the reversible state of a cell in which it does not divide but retains the ability to re-enter cell proliferation. Quiescence can either be programmed or injury-induced. The phenomenon is influenced by mTORC1 signaling, a target for cancer therapeutics. Improved understanding of the mechanisms behind stem cell quiescence holds potential for future therapeutics against tumorigenesis. Cancer biology research in the past have suggested a correlation between mTORC1 activity and mitochondrial fusion and biogenesis. To pinpoint how mitochondrial dynamics influence stem cell quiescence, I radiated, dissected, and imaged UAS-Gal4 Drosophila models with specific mitochondrial gene knockdowns. As hypothesized, Drosophila ovaries with mitochondrial biogenesis and inner-membrane fusion knockdowns failed to exit quiescence after radiation. To further confirm our hypothesis, I am now directly quantifying the mTORC1 activity of mitochondrial knockdown Drosophila lines. Once collected, our data will contribute to ongoing research in cancer therapeutics.