Search

Glycogen Synthase Kinase 3 (GSK-3) functions as a consistently active serine/threonine kinase that is positioned at the intersection of numerous signaling pathways, effectively regulating the activity of numerous substrates. Its activity is regulated by multiple upstream signals, leading to alterations in the substrates it regulates. Given its pivotal role in significant pathways such as cellular proliferation, cell cycle progression, metabolism, and adhesion, GSK-3 kinase activity is tightly regulated and modulated. Dysregulation in its signaling and, as such, its downstream targets can contribute to disease. In several cancers, including colorectal cancer (CRC), aberrant signaling due to mutations in upstream proteins regulating GSK-3 leads to pathological inhibition of its enzymatic activity. Loss of GSK-3 kinase activity can also increase resistance to specific chemotherapeutic and targeted therapies. Its central role in signaling pathways, downregulation in cancer, and its role (or lack thereof) in therapeutic resistance make GSK-3 an ideal therapeutic target. However, there are currently no known compounds that can directly activate GSK-3 kinase activity. Studies on GSK-3 suggest the possibility of allosterically activating its enzymatic activity. Thus, we hypothesize that a small molecule can enhance GSK-3 kinase activity, restoring its downstream function in conditions with pathologically low enzymatic activity. This study aims to identify a novel GSK-3 small molecule activator and explore the impact of restoring GSK-3 kinase activity in both normal and CRC conditions.In this study, we conducted a high-throughput screening of 50,000 small molecule compounds and established a systematic approach to screen and identify a compound capable of activating or enhancing GSK-3 enzymatic activity. Following multiple screens, we narrowed down our initial pool of compounds to the top 35 hits. We then utilized in-vitro kinase assays alongside cell-based assays measuring

Post-translational modifications (PTMs) expand the functional proteome under physiological and pathophysiological conditions by altering and regulating protein structure, function, and localization. Greater than 700 PTMs have been identified in eukaryotic cells. Many of these PTMs are derived from endogenous metabolic intermediates, serving as sensors for feedback and feedforward processes aiding in the regulation of cell growth and homeostasis. The biochemical genesis of PTMs occur from either enzyme-catalyzed reactions or non-enzymatic adduction by electrophilic metabolites. Regulation of PTM abundance and site-specificity occurs through dedicated enzymes that that add (writers) or remove (erasers) these functional groups. We have identified a novel PTM derived from the reactive glycolytic intermediate, methylglyoxal (MGO). This reactive metabolite is detoxified by the glyoxalase cycle to prevent modification of nucleic acids and protein. MGO is rapidly conjugated to glutathione (GSH) by glyoxalase 1 (GLO1) to produce lactoylGSH (LGSH); glyoxalase 2 (GLO2), the rate limiting step of the glyoxalase cycle, then hydrolyzes LGSH to produce D-lactate and recycle GSH. We have shown that lactoylLys modifications are generated from a non-enzymatic acyl transfer of the lactate moiety on LGSH to free Lys residues. The following studies aimed to characterize lactoylLys by elucidating its generation, regulation, and role in metabolism. Collectively, we uncovered a novel glycolytic feedback mechanism mediated by the glyoxalase cycle, how these PTMs are enzymatically regulated, and the role of the glyoxalase cycle in inflammation. These results address major gaps in our understanding of how non-enzymatic PTMs are regulated and their role in metabolism.

Neurodegenerative diseases are incurable and debilitating conditions that result from gradu-al and progressive degeneration and death of neural cells, leading to nervous system dysfunction that causes problems with movement (called ataxias) and mental functions (called dementias). Neu-rodegenerative diseases already affect millions worldwide and cases are rapidly increasing with the accelerating population with no current efficacious therapies or cures. Mounting evidence supports toxic oligomerization from atypical protein folding and aggregation of hallmark proteins that are of therapeutic interest as means of intervention. TAR DNA-binding protein 43 (TDP-43) is one of these hallmark proteins indicated in several neurodegenerative diseases pathophysiology and has been largely implicated in both familial and sporadic cases of Amyotrophic Lateral Sclerosis (ALS). TDP-43 is an intrinsically disordered multidomain protein that consists of an N-terminal domain (NTD1-102), two tandem RNA recognition motifs (RRM1¬102-177 and RRM2191-260), and an intrinsically disordered glycine-rich C-terminal domain (CTD). Structural characterization of the prion-prone TDP-43 has been challenging since its intrin-sically disordered regions represent 15-30% of the total protein. In this study, we seek to define possible new targetable sites on the apo structure of TDP-43 RRM domains (TDP-43102-269). To do so, we utilized multi-dimensional NMR experiments to assign the backbone residues for TDP-43102-269 along with Molecular Dynamics (MD) simulations to investigate the structure and dynam-ics of TDP-43 RRM domains in the absence of RNA (apoTDP-43102-269). We compared apo TDP-43 structures obtained from MD, experimental NMR restraints, and de novo sequence to structure predictions using AlphaFold2 (AF2), RaptorX, and I-TASSER of TDP-43 RRM domains and found differences within structures and looped regions. A Sitemap analysis identified five drugga-ble sites for NMR structures both apo and R

I have examined the structure, mechanism, and functional relevance of two DNA-binding proteins in this study. The first of these proteins is SgrAI. It is a type II restriction endonuclease that recognizes and cleaves certain sequences, known as primary and secondary sites. Depending on the type of recognition site SgrAI is bound to, SgrAI has a dimeric form or a filamentous form. The dimeric form of SgrAI is an inactive structure which can cleave only primary site sequences slowly. On the other hand, when it is in the filamentous form, its cleavage rate increases dramatically for both primary and secondary sites. In the absence of primary site sequences, SgrAI will not form filaments. In addition, by substituting the first or second base pair of the primary recognition site, the secondary recognition site is formed. This suggests that interaction between SgrAI and these two base pairs plays an important role in filamentation regulation. The mechanism by which this regulation occurs, however, is as yet unknown. In previous studies, the structure of SgrAI bound to the primary site and one type of secondary site was solved. Cryo-EM was also used to determine the filamentous structure of SgrAI bound to primary site. According to structural and kinetic studies, substitution of the second base pair of the primary site can change the energy between the inactive (T state) and active (R state) states in a way that favors the T state and, as a result, prevents the formation of filamentous structures. However, these studies did not provide any information on the substitution of the first base pair. DNA is cleaved by SgrAI using a two-metal ion mechanism. The presence of both metal ions is required for optimal cleavage rates. In the previously solved dimeric structures, only one of these atoms was found at the correct position. The other one was located about 4A away from its predicted position. The displacement of this ion in dimeric form probably contributes to the slow DNA c

The blood brain barrier (BBB), regulated by the neurovascular unit (NVU), restricts movement of substances to and from the blood to the brain, posing a daunting challenge for drug design where effective uptake is critical to a drug’s efficacy and safety. Previous research has shown that cortical spreading depression (CSD) events may be associated with migraine and other neurological disorders. What has not been well studied is how CSD pathological conditions, including associated changes in potassium ion levels and the release of mediators such as calcitonin gene related peptide (CGRP), may affect BBB integrity at the level of the microvessels, and consequently, uptake of brain-targeting therapeutics such as Triptans and Gepants. Xenobiotic transport occurs by paracellular, transcellular and/or transport-mediated processes, including by the solute carrier organic anion transporter, OATP. OATP1A2 (human)/ Oatp1a4 (rodent) are widely expressed bidirectional sodium-independent solute carriers involved in transport of a wide range of amphipathic substrates. Previous studies have shown OATP expression and activity may be pathologically regulated during inflammatory and/or hypoxic/reoxygenation states. Previously, we showed CSD promotes paracellular leak as determined by sucrose uptake as well as transporter facilitated uptake of Sumatriptan. Herein, we first investigated how high abluminal potassium levels, in vitro, and potassium chloride (KCl)-induced CSD events, in vivo, impact Oatp1a4 functional expression and uptake of Sumatriptan following prophylactic or abortive administration. Oatp1a4 facilitated parenchymal uptake was observed when Sumatriptan was dosed prior, but not after, CSD induction suggesting subclinical insults can rapidly change Oatp1a4 functional expression and its role in Sumatriptan uptake relies on timing of each dosing paradigm. Next, to investigate how CSD events promote paracellular leak, tight junction reorganization and regulation of transport

The voltage-gated sodium NaV1.7 channel, critical for sensing pain, has been actively targeted by drug developers; however, there are currently no effective and safe therapies targeting NaV1.7. Here, we tested whether a different approach, indirect NaV1.7 regulation, could have antinociceptive effects in preclinical models. We found that preventing addition of small ubiquitin-like modifier (SUMO) on the NaV1.7-interacting cytosolic collapsin response mediator protein 2 (CRMP2) blocked NaV1.7 functions and had antinociceptive effects in rodents. In silico targeting of the SUMOylation site in CRMP2 (Lys374) identified >200 hits, of which compound 194 exhibited selective in vitro and ex vivo NaV1.7 engagement. Orally administered 194 was not only antinociceptive in preclinical models of acute and chronic pain but also demonstrated synergy alongside other analgesics-without eliciting addiction, rewarding properties, or neurotoxicity. Analgesia conferred by 194 was opioid receptor dependent. Our results demonstrate that 194 is a first-in-class protein-protein inhibitor that capitalizes on CRMP2-NaV1.7 regulation to deliver safe analgesia in rodents.

Post-transcriptional processes are crucial in the regulation of gene expression. Messenger ribonucleoprotein (mRNP) granules are dynamic, self-assembling membraneless organelles that harbor non-translating mRNAs implicated in multiple post-transcriptional processes like mRNA translation, repression, localization and turnover. Besides their involvement in cytoplasmic mRNA biology, mRNP granules are also associated with disease such as neurodegenerative diseases and cancer. In this thesis, we focus on P-bodies (PBs) and stress granules (SGs) studying their assembly, composition and relevance to disease. In Chapter 2, we studied PB assembly and found that a specific mRNA RPS28B is important for P-body assembly by acting as a scaffold that enhances the interaction between Edc3, an important PB assembly protein recruited to the RPS28B 3’UTR, with Rps28 protein being translated off of the mRNA. The Edc3-Rps28 interaction correlates with PB assembly. Our work suggests that PBs may be preferentially nucleated by specific mRNA scaffolds, possibly a common theme in mRNP granule assembly. Furthermore, this is the first description, in yeast, of a cis-translated protein interacting with a protein recruited to the 3’UTR of the same mRNA, which in turn has functional consequences for assembly of cellular structures. In Chapter 3, we studied the composition of SGs with the long term goal of uncovering additional functions of SGs. We developed novel SG purification protocols that are more accurate and sensitive in identifying SG components than previously published protocols. We have identified novel SG proteins that suggest that SGs could have roles in modulating translation during stress by altering the cellular tRNA charging status and cell cycle progression by sequestering Cdc28/CDK kinases, an area of immediate future interest. In Chapter 4, we studied the relevance of SGs to neurodegenerative disease. We examined the claims that SGs are important for TDP-43 aggregation and to

Drug preclinical test is a key step in evaluating the profile of drug treatment. Many drug tests have been designed for different diseases. For instance, researchers manually count the number of peristaltic waves of drosophila larvae to conduct the severity of amyotrophic lateral sclerosis (ALS). In other cases, pharmacologists have to count dead cells by visual scoring to assess the performance of chemotherapy treatment. Labeling the mitosis events is a time-consuming task, and thus are prohibitive for large scale drug screenings. Machine learning algorithms have allowed researchers to dramatically increase the throughput of analyzing a large amount of data. However, the current methods require massive ground truth annotations which is labor intensive in biomedical experiments. Approaches with few human interventions remain unexplored. This dissertation focuses on three tasks for drug evaluation and treatment assessment. First, we propose a machine learning method to evaluate the effectiveness of drug for ALS. This method leverages t-Distributed Stochastic Neighbor Embedding (tSNE) and statistical analysis to assess the locomotion behavior of drosophila larvae and compare the difference between groups with and without the testing drug. Second, we designed a first-of-the-kind weakly supervised deep neural network for dead cell detection and counting. Compared with many existing fully supervised approaches, our approach only requires image-level ground truth. We show classification performance compared to general purpose and cell classification networks, and report results for the image-level supervised counting task. Last but not least, we propose a sequence-level supervised neural networks model using convolutional long short-term memory (ConvLSTM) and convolutional layers to detect mitosis events at pixel-and-frame level. By using binary labels, the proposed network is able to localize the cell division spatially and temporally. We have evaluated our method with s

Hsp70s are among the most highly conserved proteins in all of biology. These molecular machines function through cofactor interaction and ATP binding and hydrolysis. Through iterative binding and release of exposed hydrophobic residues on client proteins, Hsp70s are able to prevent intermolecular aggregation and promote folding progression toward the native sate. The human proteome contains 8 canonical Hsp70 proteins that vary in tissue specificity, inducibility, and organellular localization. Because Hsp70s are relatively promiscuous with regard to client selection, Hsp70s play a significant role in folding a large proportion of the proteome. Hsp70s are implicated in numerous neurodegenerative diseases and cancer because of their ability to regulate protein homeostasis within cells. In recent years, many groups have attempted to develop selective inhibitors of Hsp70 isoforms to better understand the role of individual isoforms in biology. Through rational design, forced localization, and serendipity, inhibitors have been discovered, however discovery of a molecule that interacts with a single isoform remains elusive. Herein lies a report validating the substrate binding domain of Hsp70 as a targetable site for the development of selective inhibitors. Using a fluorescence polarization (FP) peptide binding assay, a pilot screen revealed hexachlorophene as a selective inhibitor of HspA5. While this molecule is not a lead, it serves as a proof a concept that small molecules can be selective towards Hsp70 isoforms through interaction with the substrate binding domain. From here, an expansion of the assay and scale up of the screen revealed 4.8 as a potent (IC50 = 162 nM) and selective HspA5 inhibitor (20-fold for HspA5 vs HspA9 and at least 300-fold for all other Hsp70s). Through the development of the assay used to discover these compounds it was discovered that Hsp70 isoforms appear to interact differently with substrate peptides. From here, the peptide binding patter