The engineering of cells to be capable of detecting and responding to multiple input signals has major applications in the area of diagnostics, cancer therapy and biological computation. Current biological signal processing methods rely on single input single output computations. This proves ineffective when detecting and responding to multiple inputs such as determining a pathology based on combinations of different microscopic cues. The aim is to build non-linear computation in S.cerevisiae (baker yeast). A well-known example of nonlinear computation is neural networks, a complex framework of parallel computations talking together to come to a common output signal. The goal is to build several yeast strains that work together to perform these parallel computations. Each individual strain would process multiple inputs and release a singular output signal to another yeast strain. Combining many strains together would create a complex decision tree capable of higher-order computation. With this goal in mind, we decided to edit the native alpha-factor pathway within the yeast cells. Alpha-factor, a signaling molecule involved in yeast mating, is secreted by yeast and detected with surface protein STE2 (a protein in the GPCR family). We also inserted, a non-native version of STE2 from C.albicans yeast cells. This allowed our engineered strains to detect two different alpha-factor inputs (S. cerevisiae alpha-factor and C.albicans alpha-factor). When detected both alpha-factors trigger the pFUS1-yeGFP pathway causing the yeast cells to fluoresce. We investigated how different concentrations of STE2 and C.A.STE2 on the cell surface influenced the output fluorescents of the yeast strain. Mathematical models were used to interpret the data and predict signal response. These models can be used to design and predict more complex networks.

The receptor human protein kinase EphA2 has emerged as a therapeutic drug target for cancer and infectious diseases. However, the dearth of selective EphA2 inhibitors in the field is hampering EphA2-based research and therapies. Our lab recently developed a novel selective small molecule EphA2 inhibitor I268 using medicinal chemistry. Surprisingly, we observed the same inhibitory effects on cell proliferation paradoxically when inhibiting EphA2 with I268 or activating EphA2 with its agonist EphrinA1. We hypothesized that both I268 (inhibitor) and EphrinA1 (agonist) may lead to the same protein-protein interaction network of EphA2 that caused this paradoxical inhibition. We used a chemical proteomic method developed in our lab called “co-clickable precipitation (Co-CP)” to study the interaction network of EphA2 in colon cancer cell line HCT-116. We observed that the onset of inhibitory effect by I268 concurred with the association between EphA2 and a protein interactor called c-RAF, a key player in activating proliferation pathways in cancer cells containing Ras mutations such as HCT-116 cells. We are currently performing confirmatory experiments to characterize whether upon inhibition by I268 or activation by EphrinA1, EphA2 sequesters c-RAF in a form unable to transmit proliferative signals downstream. Our results may demonstrate a possible mechanism for inhibitory effects of EphA2 inhibition on cancer cell growth. These results validated EphA2 as a drug target and may provide novel ways for treating cancer cells by inhibiting cell proliferation pathways indirectly using EphA2 inhibitors.

Long non-coding (lnc)RNAs have multiple biological functions, including recruitment of kinases to regulate signaling pathways involved in tumorigenesis and other human diseases. Of particular interest is the proposed interaction between Long Intergenic Noncoding RNA for Kinase Activation (LINK-A) and the membrane component phosphatidylinositol-3,4,5-triphosphate (PIP₃). The proposed interaction between PIP₃-LINK-A would be the first example of a direct interaction between a non-coding RNA and phospholipid. I investigated the proposed interaction between LINK-A and PIP₃ using Nuclear Magnetic Resonance (NMR) spectroscopy. Multiple NMR-based experiments were performed to assess the degree of binding on the basis of line-width broadening of NMR spectra. Wild-type and mutated RNA constructs were titrated into a 100 uM solution of PIP₃, but the NMR data showed no evidence of line-width broadening, indicating that no direct interaction occurs between wild-type or mutated RNA constructs. Presumably, the reported cellular interaction is not direct and might require an additional mediating factor. I determined the 3D structure of the LINK-A RNA hairpin, required to recruit PIP₃, using biophysical molecular modeling and NMR data. My results provide a biophysical foundation to elucidate the functional role of LINK-A in PIP₃ recruitment and kinase activation.

Modern cells discriminate among the amino acids chosen to be included in protein synthesis: some, like leucine, serine, glycine, and alanine appear in most proteins across most cells, whereas other amino acids, like (g)-aminobutyric acid and aminoisobutyric acid do not appear. This raises the question of how selectivity among amino acids arose – does the selectivity rely on modern protein machinery or could it have arisen in the first protocells of the Early Earth? For example, could certain amino acids have, by some mechanism, associated more strongly with protocell membranes, increasing their chances of being integrated into the first peptide chains? To test this hypothesis, our group assembles rudimentary protocells from molecules that would have been present on the early-Earth: decanoic acid (a fatty acid), sodium mono-phosphate, salt, and water. Vesicles of the decanoic acid spontaneously form. We then add different amino acids to the solutions and measure their turbidity to determine whether each amino acid causes the number of lamellae in the vesicles to increase or decrease. Increased lamellarity correlates with a sturdier vesicle. If certain amino acids increase lamellarity of protocells, that could serve as a method of selection for certain amino acids rather than others. Our results are that particular amino acids (most notably serine, glycine, and alanine) do in fact increase the lamellarity of fatty acid vesicles significantly, whereas other, less common, amino acids do not. We are currently exploring the plausibility of a mechanism for this occurrence involving ease of rotation around the alpha carbon of the amino acids, and we are investigating other ways in which interactions between amino acids and fatty acid membranes might be manifested, for example by a shift in the solution’s critical vesicle concentration. Our results will fit into the overarching goal of understanding peptide formation and protocell stability in order to gain insight into the origins of life on Earth.

STUDIO is a peer mentoring program that focuses on connecting University of Washington (UW) undergraduate student mentors with low income, immigrant and refugee youth through Science, Technology, Engineering and Mathematics( STEM). The program is a partnership with the Neighborhood House community center in the High Point Neighborhood of West Seattle. Neighborhood House provides the facilities and is the gathering place for UW mentors and youth mentees on a weekly basis. As STUDIO is offered as an undergraduate course (EDUC 421), the mentors visit Neighborhood House once a week and then attend a weekly seminar class on campus as well. The program began as a research study that focuses on promoting youth interest in STEM. However, over the past few years it has made progress in undergraduate STEM learning research as well. As a lead mentor within the STUDIO program, I have an opportunity to participate in research aspect as well as cultivate mentorship with youth, and supporting other mentors. In this study, I explore how relationships between youth and mentors develop in STUDIO. I am doing this through analysis of documentation of STEM and non-STEM activities, written reflections of mentors and interviews. They are analyzed for types of relationships that mentors develop through this program as well as themes and patterns of development. The goal is to better understand how a mentorship program that focuses on bridging the gap of educational inequity for youth lends possibility for undergraduate learning through relationships building.

Growing reliance on antibiotics in livestock production, commercial agriculture, and human healthcare has created evolutionary pressures on bacteria. These pressures have given rise to a new threat to public health, drug-resistant microbes. In addition, antibiotics eliminate all microbes, including the beneficial ones, and are not advisable for long-term use. Recent advances in the understanding of bacterial virulence via quorum-sensing (QS) has presented the scientific community with a promising new approach for alleviating the ongoing overuse of broad-spectrum antibiotics. Quorum-sensing is a form of coordinated gene expression mediated through detection of specific population density. Bacteria use QS as a way to regulate behaviors including biofilm formation, virulence, and motility. Quorum-sensing inhibition (QSI), is the ability to prevent QS and is used by organisms across kingdoms. This prevalence suggests that disrupting prokaryotic communication is either a common defense tactic against infection or acts as a competitive advantage in resource acquisition. Fungi, a kingdom in direct competition with prokaryotes are prime candidates for broadening our understanding of the mechanisms behind quorum-sensing. With its abundance of endemic fungi, the Puget Sound region provides a unique opportunity to screen species for QSI compounds. In this study, fungi representing diverse ecological niches were collected from local forests around the Puget Sound and isolated on potato dextrose agar plates. To screen fungal isolates we used C. violaceum, a bacterium that forms purple colonies when able to quorum-sense and changes to white when a QSI is present. Our results suggest fungi are an underexploited and unexplored source of novel bioactive molecules that could provide a method to inhibit virulent effects of bacteria without damaging an organism’s microbiome.

In the US more than 100 million people are living with diabetes or pre-diabetes. The economic burden caused by these conditions, including medical costs, is approximately $327 billion annually. Conventionally, transgenic Escherichia coli has been the primary source of commercial insulin production, a process that requires extensive purification to ensure shelf stability and complete removal of contaminants. This study seeks to establish an alternative mode of insulin production using polyethelyne glycol (PEG) and agrobacterium to transform the oyster mushroom, Pleurotus ostreatus, with the human insulin gene. P. ostreatus is a valuable target for genetic transformation due to its lack of endotoxins, rapid growth, and fully sequenced genome. P. ostreatus was transformed using PEG and agrobacterium with a plasmid containing the human insulin gene and a carboxin resistance gene. Transformed cells were selected using carboxin, extracted, and regenerated on plates composed of yeast extract, malt extract, and glucose (YMG). Integration of the human insulin gene in to the mushroom genome was confirmed through PCR analysis of the transformants. Successful transformation of P.ostreatus offers a new avenue for insulin production, potentially diversifying the market and treatment options for diabetics.

Proteins are essential to life, but many structural methods fail to capture the dynamic nature of proteins. This means researchers are left with an incomplete view of how protein function and structure relate. The Stoll lab uses electron paramagnetic resonance (EPR) to study the dynamic structure of proteins. The project I am working on is investigating the way in which the measurement conditions in EPR affect the determined protein structure. We use maltose binding protein (MBP) as a model system, since its structure has been previously well-characterized. My primary role has been to create new mutants for site-directed spin labeling, in which a radical spin label is attached to the protein. This radical is measured by placing the sample in a magnetic field and results in a probability distribution of distances. Common spin labels used in EPR, however, have many rotatable bonds, andd can cause uncertainty in the extracted distance distribution. Our project is investigating the contribution of spin labels to the EPR experiment. I use polymerase chain reaction to create new mutants of MBP by swapping a native residue with a cysteine residue in the mutants. Each mutant has one or two amino acids replaced with a cysteine. After mutating the DNA, I then transform the mutant DNA into wild type cells to grow the mutant protein and purify the protein. The spin label forms a disulfide bond with the cysteine residue in the mutant, providing an unpaired electron for measurement. By using various commonly used spin labels on many different site pairs in MBP, we hope to develop a model in which the contribution of spin labels to the probability distribution of distances is accounted for. This will ensure more accurate results in determining protein structure and will aid in structural characterization of many proteins.

Tin-doped indium oxide (ITO) is a n-type semiconducting material that have been widely used in solar cells, smart windows, and transistors. This material conducts electricity due to excess free charge carriers which charge-compensate the tin dopants. We are working to develop new chemical treatment methods to build optoelectronic devices from ITO nanocrystals. Colloidal nanocrystals are well-suited for solution-processing and low-temperature deposition techniques. These routes can be ideal for fabrication of thin-film electronics, especially on flexible substrates; however, increasing electronic communication between nanocrystals require post-deposition treatments. Few low-temperature methods to create conductive films from oxide nanocrystals have yet been reported. We seek to increase the conductivity of ITO nanocrystal films with ligand-exchange procedures to decrease the distance between each nanocrystal, then chemically crosslinking them. We synthesized ITO nanocrystals, then resuspended in solvents at high concentrations for film deposition. We first modify the surface of our nanocrystals by replacing the native long nonpolar ligands with short polar ligands. The modified crystals were deposited by spin-coating onto glass substrates and a variety of solvents/conditions were tested to optimize uniformity and thickness. Four-point probe measurements resulted in sheet resistance (R_s) ranging from 1 MΩ/square (ligand exchanged nanocrystals) to 2 TΩ/square (untreated nanocrystals). Profilometer measurements probed film thickness that can be used to calculate our films' bulk resistivity. Next, we used zinc and aluminum compounds as chemical solders to build bridges between nanocrystals to increase the material’s conductivity. R_s drops to ~0.15 MΩ/square with aluminum bridges and ~0.10 MΩ/square with zinc bridges. We plan on using other undoped metal oxides with this same scheme and apply them to thin-film field effect transistors (FET) to investigate how these treatments improve FET performance. We are also working on using similar techniques with SnO₂ nanocrystal films for application as electron-transport layers in perovskite solar cells.

Ion Mobility-Mass Spectrometry (IM-MS) is an analytical technique that is useful for analyzing large biomolecules with minimal disruption to their natural structure. One of the most common methods of introducing proteins to the gas phase in IM-MS is electrospray ionization (ESI), whereby a large voltage applied to a sample induces a spray of droplets that quickly evaporate and leave the desired analyte as a gas-phase ion. It is established that the high voltages of this process lead to a buildup of charge via electrochemistry in the sample solution, which may cause changes to the pH of the solution. However, this phenomenon has primarily been characterized in systems with continuously replenishing samples that form a steady state between excess charge formation and incoming sample flow. This work establishes a method to identify these changes in small, non-replenishing systems, which have not yet been characterized and are the standard practice for IM-MS analysis of proteins. Using SNARF-4F, a pH-sensitive fluorescent dye; a series of filters and a camera; and Python scripts for image processing; the rates and spatial position of pH changes under non-equilibrium conditions were determined with high accuracy. Additionally, several methods of preventing or slowing pH changes are examined. These include the use of buffers such as ammonium dihydrogen phosphate and ammonium bicarbonate, or periodic cycling between positive and negative electrospray. Preliminary findings indicate that these buffers can affect pH change, but at high enough concentrations, may also reduce the quality of mass spectra.

Crucial processes in clean energy research, such as the splitting of water into H₂ and O₂ and the reduction of CO₂, require multi-electron redox events throughout a catalytic cycle. Noble metals such as iridium and platinum prefer such events, while more abundant and consequently cheaper base metals prefer single-electron events. Redox-active ligands offer the potential of enabling noble metal behavior in base metals by combining a single-electron transformation at both the metal and the ligand to create an overall two-electron process. While redox non-innocent ligands are typically comprised of organic components, cobalt selenide clusters offer an attractive alternative given their wide variety of accessible oxidation states. Here, I present the synthesis of heteroleptic cobalt selenide clusters [cis-Co₆Se₈(PEt₃)₄(RNHP(C₆H₅)₂)₂ Et = ethyl, R = alkyl, aryl] containing ditopic aminophosphine ligands. The cluster was then metallated using copper(II) triflate and the product was fully characterized by multi-nuclear nuclear magnetic resonance, ultraviolet-visible, and infrared spectroscopies. Further analysis was performed using cyclic voltammetry and the solid-state structure has been solved via single-crystal X-ray crystallography. Upon spectroscopic analysis, it appears that copper(II) is reduced to copper(I) by the cluster. This result is encouraging, as using clusters as redox-active ligands would require facile electron transfer between the metal atom and the cluster core.

Mercury (Hg) is widely known to be a neurotoxin. Mercury in our environment is found in many different forms, and the key difference between them is evident in the way they are absorbed by organisms. Hydrophobic methylmercury (Me-Hg) readily bioaccumulates in the tissues of all organisms, leading to Hg exposures involving higher doses. Recent research shows that sulfate-reducing bacteria (SRBs) play a role in methylating mercury only when they possess the hgcAB gene cluster. Gaining a more comprehensive understanding of the aqueous conditions required for SRBs to thrive and consequently methylate inorganic mercury is essential for addressing the ongoing problems associated with Hg toxicity. In the initial phase of this study, ion chromatography was used to quantify sulfate (SO42-) concentrations along an urban river in an industrial region of Seattle (the Duwamish/Green River Watershed). The measurements revealed that SO42- concentrations as high as 5300 ppm were present at several sites along the lower portion of the Duwamish River. A number of previous studies also showed significant amounts of mercury in this region’s sediments and fish tissues. The secondary phase of this study involved an investigation determining whether the measured high sulfate concentrations were related to the production of Me-Hg and additionally examined if SRBs likewise played a role. River sediment was analyzed for the presence of the hgcAB gene cluster. Ongoing studies are focused on quantifying Me-Hg at sites where SRBs were found and on the identification of key indicators for mercury methylation conditions along a watershed. This study provides further insight into the relationship between mercury, sulfates, and SRBs when found in combination in an aqueous environment.

Lecture demonstrations have a long history of use in science learning spaces. While extant research in STEM education show that demonstrations can be used to rouse student interest, some studies suggest that demonstrations have little to no effect on improving students’ understanding of the concepts taught in class. In chemistry, demonstrations are commonly used in lectures with the belief that student understanding of chemical concepts improves from observations of chemical phenomena. The purpose of this mixed methods study was to investigate students’ perceptions of lecture demonstrations and their usefulness in preparing for exams. To measure students’ perception of lecture demonstrations, a survey was administered to first year students (n <1500) in nine introductory chemistry classes over two quarters. Initial findings reveal that while students do in fact remember and enjoy lecture demonstrations, they found it difficult to apply demonstration concepts to problems on exams. Results also suggest a strong link between student engagement and exam performance. Overall this study highlights the importance of soliciting student feedback for improving and designing course materials and tools. Various strategies and implications for practice are also addressed.

 In vivo animal models are an advanced tool for testing complex biological phenomena and functions in humans, yet are difficult to adapt due to high cost, biological and anatomical differences, and ethical concerns. As an alternative to in vivo models, in vitro organotypic models have been created that can accurately and efficiently reproduce biological processes, allowing researchers to probe specific factors without the difficulties associated with in vivo models. We present further development of a novel microfluidic platform, called Stacks, which uses open and suspended microfluidic principles to enable spatial and temporal manipulation of 3D cultures in suspended collagen plugs. We are developing the Stacks platform to accurately reflect morphology exhibited in the air-liquid interface (ALI) of the epithelial layer of the lung and allow analysis of lung pathophysiology, cell signaling, and immune responses of the lung in both monoculture and coculture. Furthermore, using the flat faces of the device that allow for stacking, we have connected a layer of human lung endothelial cells, representing a model vascular system, with a second layer of immune cells and mycobacterium, representing a model infection system, to study soluble factor signaling in granulomatous infections. Through the modification of the Stacks device, we can achieve granuloma formation and utilize soluble factor analyses (e.g., immunoassay) to observe and characterize infection-dependent changes in vasculature resulting from the exchange of chemical signals generated by the infection model. Future development of the platform will build on these newly developed lung models to combine them for studies involving signaling and disease mechanisms between the lung ALI and the underlying tissue layers.

There has been growing concern regarding rising carbon dioxide concentration in the atmosphere because CO₂ traps excessive heat and warms the planet through the greenhouse effect. One strategy to mitigate this problem is through the capture and conversion of CO₂ via hydrogenation to fuels and chemical feedstocks that are currently produced from fossil fuels. Given the complexity of these multi-electron, multi-proton transformations and the multitude of products that can result, catalysts are required to lower the energy barrier and direct the selectivity of CO₂ conversion reactions. One catalyst currently under development is a Ru(II) bis-(protic N-heterocyclic carbene) phosphine catalyst, which incorporates protic N-H wingtips adjacent to the metal center. The N-H wingtips are an interesting feature due to their ability to activate CO₂ through metal-ligand cooperation, their accessibility as a proton source near a metal active center, and the likelihood that they aid in splitting H₂ between the metal center and the nitrogen. Preliminary results have shown moderate turnover numbers (TONs) for both formate (130) and methanol (7), the latter of which is a rare transformation in a single catalyst system. In order to improve catalytic TONs and understand the role of protic N-H wingtips, a library of catalysts with varying ancillary ligands, including 2,2’-bipyridine, 4,4’-dimethoxy-2,2’-bipyridine (electron donating), 4,4’-dibromo-2,2’-bipyridine (electron withdrawing), and 1,2 bis(diphenylphosphino)ethane (sterics), is synthesized and screened under high pressure and temperature conditions using THF solvent and varying additives (e.g. Li₃PO, K₃PO₄, KPF₆) and additive concentration. It is expected that the ancillary ligands, bound trans to the bis-carbenes, will influence the proton donor ability of the N-H wingtips and catalytic turnover.

Social determinants of health (SDOH) are non-medical factors that contribute to a person’s health and well-being. The Centers for Medicare/Medicaid Services recently released a standardized 26-question survey designed to capture SDOH in the following 13 domains: living situation, food insecurity, transportation, utility help, personal safety, financial strain, employment, family and community support, education, physical activity, substance use, mental health, and disabilities. However, there is limited study using this survey tool in a safety-net institution. Harborview Medical Center (HMC) is the public hospital serving King County, WA and provides care to a wide variety of patients. This project was a prospective, observational study which aimed to investigate implementation of this survey tool and to measure relevant SDOH and other medical factors in this setting. My role included administering the survey to patients admitted to the HMC Hospital Medicine service from October to December, 2018. I collected data related to the duration of the survey and any challenges patients encountered responding to the survey, as well as review of patient charts for demographic information, medical system utilization, and medical history. In total, only 24 of 90 patients (27%) completed the survey. Major challenges included patient lack of interest, language barrier, ability to consent, and competing time demands. These results have implications for the generalizability of this tool and whether it comprehensively identifies SDOH needs.

Harmful radiation, such as UV photons from the sun, can damage DNA in a manner that results in cation-radical formation which leads to nucleobase loss and strand breakage. In the Frantisek Tureček research group, DNA cation-radicals were generated in the gas phase, isolated and subjected to incremental wavelengths between 210-700 nm with intent to observe their absorption action spectra that can be used to assign structures and intrinsic properties. The difficulty with these experiments is the ability to effectively communicate the experimental set-up of mass spectrometers, tunable lasers, and various other pieces of equipment in a supporting manner that delivers concise information to be understood regardless of prior knowledge in this specific field. I had the unique opportunity to take two-dimensional, over-simplified schematics and transform them into three-dimensional, professional and photorealistic images of our experimental arrangements using Siemens NX CAD modeling software.

All-trans-retinoic acid (RA) serves an important role in maintaining tissue health, either deficient or excessive levels can lead to health issues. Aldehyde dehydrogenase 1A1 (ALDH1A1), is generally believed to be the main enzyme responsible for the conversion of retinaldehyde (RAL) to RA in the liver, requiring the cofactor nicotinamide adenine dinucleotide (NAD+). However, previous studies indicated that after WIN18,446, a potent inhibitor of ALDH1A enzymes, was administered to mice; liver RA concentrations were not significantly altered, and in vitro the RA formation in mouse liver was only inhibited by about 50% suggesting other enzymes except ALDH1A1 synthesis RA in mouse liver. Mouse aldehyde oxidase has previously been proposed to synthesize RA. Hence, in the current study we tested the hypothesis that aldehyde oxidase (AOX) also contributes to the formation of RA in human liver. Our data shows that purified human recombinant AOX catalyzes the oxidation of RAL to RA. The K_m (indicating substrate binding affinity to the enzyme) and k_cat (indicating maximum enzyme velocity) values were determined by enzyme kinetic assays as 1.4 μM and 3.5 min-1. In the absence of NAD+(AOX mediated activity), RA formation was observed in human liver S9 fractions (HLS9) and the RA formation rate was, on average, 60% lower than that measured in the presence of NAD+(n=4). In addition, hydralazine, a selective AOX inhibitor, inhibited about 55% of RA formation in HLS9 in the presence of NAD+ while combining hydralazine and WIN18,446, more than 85% of RA formation in HLS9 was inhibited when compared to the control. In conclusion, this data shows that AOX and ALDH1A1 each contribute about 50% of RA biosynthesis in human liver. This research helps us better understand the regulation of retinoid homeostasis in humans.

 A method of negative-pulse partial modulation for comprehensive two-dimensional gas chromatography (GC×GC) using a pulse flow is studied. The capability of a pulse flow valve modulator has been shown to enhance many aspects of GC performance such as chemical selectivity and reduced analysis time, allowing complex GC×GC analysis in a shorter time scale. In our study, we explored a particular configuration of a pulse flow valve that periodically (specified by modulation period, P_M) disrupts the carrier gas flow to achieve partial modulation. This disruption of flow creates a "plug" of higher concentration on top of the primary GC signal, which is further separated in a secondary column. The size of the plug determined by pulse width, P_W, has a direct relationship to chemical selectivity and sensitivity. For our experiment, we determined the ideal P_W to be around 8 ms. Therefore, a 20-component mixture was evaluated using P_W of 8 ms and P_M of 100 ms, producing an average peak width, ²W_b, of 12 ms and approximated ideal peak capacity, ²n_c, of 8 on the second dimension. All 20 compounds were separated in a 12 second separation window. This powerful partial modulation method was achieved by the unique concept of the flow disruption with short modulation period, demonstrating high peak capacity and improved selectivity between compounds and chemical sensitivity. Due to the significant increase in GC performance, this form of partial modulation using a pulse flow valve allows efficient and faster GC×GC analysis that can be implemented in a variety of GC×GC applications.