Marine phytoplankton account for 50% of global primary productivity, regulate nutrient cycles, and form the base of the oceanic food web. Recent studies on freshwater species have shown that cross-feeding between phytoplankton and bacteria in their microbiome influences phytoplankton thermal tolerance. The phytoplankton supplies bacteria with photosynthate and gets cobalamin in return, which allows it to function at high temperatures by shifting from METE to the METH pathway for methionine production. However, few studies have investigated this mutualism in marine species. In this project, we study how cross-feeding affects marine phytoplankton and its response to global warming. We consider a model where the two mutualists exchange the nutrients they synthesize (cobalamin and carbon) and also compete for nitrogen. The model explores tradeoffs between allocating resources to growth VS. substrate synthesis, temperature effects on algal growth rate, conditions required for coexistence, and sensitivity of the mutualism to warming. This study provides a quantitative tool for understanding future algal blooms under climate change. Additionally, it expands fundamental understanding of the role of species interactions in adaptation to thermal gradients.

90% of the body’s serotonin is synthesized in the gut. The gut microbe Turicibacter sanguinis (T. sanguinis) imports serotonin through a novel transporter (TuriSERT) structurally and functionally similar to the mammalian serotonin transporter. T. sanguinis and TuriSERT are both inhibited by the antidepressant fluoxetine. T. sanguinis treated with serotonin downregulated sporulation-related genes and enhanced its colonization of the intestine. T. sanguinis also is inhibited by bile acids (BA) and modulates host lipid metabolism. Certain strains of T. sanguinis modify BA in vitro via deconjugation. This study aimed to identify the genetic determinants of T. sanguinis TuriSERT import of serotonin and BA modifications. This was done through the comparison of both serotonin import and BA modification abilities of different T. sanguinis natural isolate strains. The approach explored was to express TuriSERT from different strains in Bacillus subtilis and compare uptake rates of the fluorescent serotonin-like molecule and radiolabeled serotonin. To determine distinctions in BA modification abilities, different T. sanguinis strains and BA hydrolase genes within T. sanguinis expressed in Escherichia coli were exposed to a panel of BAs and subsequent BA derivatives were identified via LC-MS. Strains which indicate worse or better serotonin import or BA modification abilities can be identified and indicate a relationship between the gene and the respective T. sanguinis phenotype. Insights on the molecular mechanisms which govern T. sanguinis’ interactions with host serotonin levels and lipid metabolism could have implications on the study of major mood disorders and metabolic diseases.

Cellular senescence is a rapidly developing field in biomedical research because of its relevance to a variety of chronic diseases and its potentially detrimental role in the aging process. A primary focus in the field is identifying and characterizing which cell types and tissues are driving and undergoing senescence. To answer this question, our lab developed an in vitro system to test whether macrophages can become senescent. We found that when mouse bone marrow derived macrophages were exposed to DNA damaging reagents such as irradiation or doxorubicin, a chemotherapeutic, they expressed senescent phenotypes, as characterized by a senescence-associated secretory phenotype (SASP), permanent cell cycle arrest, and failure to activate apoptosis signaling pathways. Unbiased RNA sequencing of these senescent macrophages, coupled with confirmatory western blot analyses, have also begun to reveal new markers of senescence in macrophages such as the upregulation of Cyclin D2, a cyclin dependent kinase involved in cell cycle regulation. Continuing to explore other possible markers, such as Lamin B1, p21 and p27, can potentially help identify senescent macrophages in vivo. To investigate the molecular mechanisms underlying these senescent macrophages, we conducted metabolomic analyses and seahorse assays. Interestingly, these experiments revealed significant differences in metabolite production, oxygen consumption rates, and extracellular acidification rate between senescent macrophages and healthy pro-inflammatory M1 and anti-inflammatory M2 macrophages. Further examination of these differences through knocking out different metabolic and signaling pathways in senescent macrophages may better illuminate senescent cell biology and potential therapeutic targets to help treat aging-related diseases in humans.

Reactive oxygen species (ROS) is a broad term to describe various reactive and free radical forms of molecular oxygen that are generated during aerobic metabolism. Oxidative stress is marked by elevated ROS which damage various proteins, lipids, and DNA. To restore ROS homeostasis, the cells then activate its antioxidant responses via the Nrf2-mediated signaling (i.e., catalase, glutathione, superoxide dismutase). Essentially, Nrf2, a key transcription factor of the antioxidant genes, binds to the antioxidant response element (ARE) in the promoter region of several antioxidant enzyme genes. However, with pulmonary fibrosis (PF), a progressive scarring disease of the lungs, the oxidative stress level is highly elevated. By persistent microinjuries, the cells in PF-diseased lungs continue to express elevated levels of oxidative stress biomarkers despite having antioxidant responses, indicating excessive oxidative stress may cause impairment of antioxidative mechanisms. Recently, the induced fibroblast activation (iFA) model, a novel in vitro model exhibiting spontaneous and progressive fibrogenesis by using induced pluripotent stem cells, has identified ROS levels are increased during generating fibrotic foci; however, it has not been shown how fibrotic cells regulate antioxidative mechanisms. Hence, we hypothesize that impairment of antioxidant mechanisms in fibrotic cells could establish progressive fibrosis. This study aims to investigate the cause of the impairment of the antioxidant responses using the induced fibroblast activation (iFA) model. Exploring the defects in antioxidant mechanisms of fibrotic cells will aid our understanding of the underlying mechanisms of IPF and help to improve therapeutic strategies against IPF.

Finding the optimal network topology for wireless sensor networks is an important problem that has a wide variety of applications in the field of environmental and earth sensing. This is because issues with connectivity and routing can arise if the environment a topology is in contains obstacles that impede the communication between nodes – examples of such environments being towns or cities. As such, our research focuses on synthesizing wireless sensor networks that satisfy certain sensing and communications requirements in these challenging environments. Towards this end, we created occupancy grids based on satellite images that display the locations of obstacles and free space. For different placements of sensor nodes on the occupancy grids, we assessed the performance of the resulting networks based on our sensing requirements and the maximum flow that can be sent from a source to a destination. The data that we collected will be the basis for a new algorithm that can determine effective sensor network topologies for high-obstacle environments.

Starting in the integer plane, placing two vertical lines as boundaries that create a region of width one gives rise to the ladder graph. Starting on the left boundary at the point of height 0, we look at the random walk that moves up, down and to the right with equal probability, and it ends when the walk first reaches the right boundary, at height k, where k is an integer. We say the walk exits at k, and the probability of that happening is p(k). In order to find p(k) for any integer k, we give a bijective proof of the equality of log-concavity of the sequence of exit probabilities {p(k)} and use that to find a formula for p(k) for any integer k.

The transcriptome represents all the RNA in an organism and directs the identity and function of a cell. The specialization of cells is achieved by differential expression from the genome, which is reflected in distinctive transcriptome profiles. A main concept when studying the idea of differential gene expression is specificity. Specificity of gene expression is a measure of the degree to which a gene is restricted to a specific tissue versus ubiquitous to most cells and tissues of the body. Current methods, such as, Z-scores and Tau explore tissue specificity by comparing how much the gene is expressed to the overall expression value of that gene in all tissues. However, they are limited since they do not account for the similarity between tissues. The main aim is to see how we can use a new method that will take into account the similarity of transcriptomes when calculating similarity scores. We will use a robust method of reweighting the samples that will incorporate the similarity information obtained and use it to quantify the similarity between sample transcriptomes. We will be using a GTEx dataset to compare the different similarity measures and our robust method to a developmental lineage tree in order to be able to make the justification for this more robust reweighting method. It will allow us to see how accurate our reweighting method is compared to others.

Polyelectrolyte complexation between oppositely charged macromolecules is primarily driven by non-covalent (electrostatic) interactions. These self-assembled materials find applications as bio-adhesives, encapsulants, delivery and purification agents. Analytical techniques such as small angle x-ray scattering (SAXS) and dynamic light scattering (DLS) offer insights about the size, structure and behavior of these materials, all of which are influenced by the polymer backbone, intermolecular interactions and solution conditions. In this project, we focus specifically on obtaining size and structure estimates of these materials via light scattering measurements, to explain structure-property relationships. We propose a Bayesian optimization model that will best describe and fit the size and structure estimates found. We are working to fit our equation for a spherical model to our data. We are looking at the relationship between their form factor (P(q)) and length scale (q), where the form factor is the dependent variable, and the length scale is the independent variable. Our model will be constructed using the Statistics and Machine Learning toolbox in MATLAB, as well as the Parallel Computing toolbox. The effectiveness of toolboxes offered by MATLAB, to perform Bayesian Optimization, was tested with preexisting generic examples. The terms involved in the optimization process have been learnt and understood.

Despite the existence and efficacy of SARS-CoV-2 vaccines, there is an additional need for novel therapeutics due to the emergence of variants and inability for some populations to receive the vaccine. Previous data has suggested that 1) specific bioactive and oxidized lipids drive SARS-CoV-2 pathogenesis and 2) mitochondrial antioxidants possess properties that may make them favorable antivirals against SARS. In order to better characterize the in vivo effects of novel antioxidants on SARS-CoV-2- associated inflammation and oxidative stress, we utilized lung tissue protein lysates from infected mice and ran ELISAs to measure key mediators of inflammation. We quantified protein levels of i) enzymes responsible for making key bioactive lipids [prostaglandin D2 synthase (PGDS) and 5-lipoxygenase (5-LO)]; ii) Nrf2, a major antioxidant and anti-inflammatory protein; iii) IL-6 and C reactive protein (CRP), 2 important inflammatory mediators. There were no differences in protein levels of PGDS and LO between uninfected and SARS-CoV-2 infected mice. However, infected mice that received antioxidant therapy had increased levels of anti-inflammatory Nrf2, reduced levels of IL-6 and similar levels of CRP compared to infected mice given no treatment. Our data suggests that the usage of mitochondrial antioxidants can have favorable anti-inflammatory effects in COVID-19-related lung damage. Additional protein studies of lung tissues from established preclinical models of SARS-CoV-2 infection will further advance our understanding of how to therapeutically target lung inflammation in COVID-19.

Direct reprogramming has been effectively utilized to induce functional neurons from murine fibroblasts in vitro. Yet, the effect of the immune microenvironment on direct reprogramming is not well understood. To work towards translating induced neuronal (iN) reprogramming to in vivo studies, pilot experiments were performed to shed light on the relationship between neuronal reprogramming and immunomodulatory factors. We investigated the effects of inflammatory and anti-inflammatory cytokines used to induce non-polarized (M0), proinflammatory (M1), and anti-inflammatory (M2) macrophage phenotypes, macrophage conditioned mediums and several immunomodulatory drugs during the reprogramming of fibroblasts to iN cells. Preliminary results indicate that M1, or proinflammatory cytokines, may inhibit the reprogramming process. Future studies will further examine these findings to determine the mechanism by which specific immunomodulatory factors affect reprogramming efficiency, providing some insight into how the immune microenvironment influences cellular reprogramming, and eventually paving the way for future clinical applications.