Indiana University-Purdue University Indianapolis (IUPUI) / The over-accumulation of glycogen appears as a hallmark in various glycogen storage diseases (GSDs), including Pompe, Cori, Andersen, and Lafora disease. Glycogen synthase (GS) is the rate-limiting enzyme for glycogen synthesis. Recent evidence suggests that suppression of glycogen accumulation represents a potential therapeutic approach for treating these diseases. Herein, we describe the discovery, characterization, and development of small molecule inhibitors of GS through a multicomponent study including biochemical, biophysical, and cellular assays. Adopting an affinity-based fluorescence polarization assay, we identified a substituted imidazole molecule (H23), as a first-in-class inhibitor of yeast glycogen synthase 2 (yGsy2) from the 50,000 ChemBridge DIVERSet library. Structural data derived from X-ray crystallography at 2.85 Å, and enzyme kinetic data, revealed that H23 bound within the uridine diphosphate glucose binding pocket of yGsy2. Medicinal chemistry efforts examining over 500 H23 analogs produced structure-activity relationship (SAR) profiles that led to the identification of potent pyrazole and isoflavone compounds with low micromolar potency against human glycogen synthase 1 (hGYS1). Notably, several of the isoflavones demonstrated cellular efficacy toward suppressing glycogen accumulation. In an alternative effort to screen inhibitors directly against human GS, an activity-based assay was designed using a two-step colorimetric approach. This assay led to the identification of compounds with submicromolar potency to hGYS1 from a chemical library comprised of 10,000 compounds. One of the hit molecules, hexachlorophene, was crystallized bound to the active site of yGsy2. The structure was determined to 3.15 Å. Additional kinetic, mutagenic, and SAR studies validated the binding of hexachlorophene in the catalytic pocket and its non-competitive mode of inhibition. In summary, these two novel assays provided feasible biochemical platforms for large-scale screening of small molecule modulators of GS. The newly-developed, potent analogs possess diverse promising scaffolds for drug development efforts targeting GS activity in GSDs associated with excess glycogen accumulation. / 2021-07-01

No description available.

No description available.

Solar wind-magnetosphere-ionosphere (SW-M-I) coupling is investigated with three different computational models that characterize space plasma dynamics on distinct spatial/temporal scales. These models are used to explore three important aspects of SW-M-I coupling. A particle-in-cell (PIC) model has been developed to explore the kinetic scale dynamics associated with the magnetotail dipolarization front (DF), which is generated as a result of magnetotail reconnection. The PIC study demonstrates that the electron-ion hybrid (EIH) instability could relax the velocity shear within the DF via emitting lower hybrid waves. The velocity inhomogeneity driven instability is highlighted as an important mechanism for energy conversion and wave emission during the solar wind-magnetosphere coupling, which has been long neglected before. The Lyon-Fedder-Mobbary (LFM) global magnetohydrodynamic (MHD) model is used to explore the fluid scale electrodynamic response of the magnetosphere-ionosphere to the interplanetary electric field (IEF). It is found that the cross polar cap potential (CPCP) varies linearly with very large IEF if the solar wind density is high enough. With controlled experiments of global MHD modeling driven by observed parameters, the linearity was interpreted as a result of the magnetosheath force balance theory. This study highlights the role of solar wind density in the electrodynamic SW-M-I coupling under extreme driving conditions. The LFM-TIEGCM-RCM (LTR) model, which is the Coupled-Magnetosphere-Ionosphere-Thermosphere (CMIT) model with Ring Current extension, is used to explore the integrated SW-M-I system. The LTR simulation study focuses on the subauroral polarization streams (SAPS), which involve both MHD and non-MHD processes and three-way coupling in the SW-M-I system. The global structure and dynamic evolution of SAPS are illustrated with state-of-the-art first-principle models for the first time. This study has successfully utilized multiscale models to characterize the forefront issues in the space plasma dynamics, which is required by the facts that plasmas have both particle and fluid featured properties and those properties are vastly different across geospace regions. It is highlighted that SW-M-I coupling could be significantly influenced by both microscopic and macroscopic processes. In order for a comprehensive understanding of the SW-M-I coupling, multiscale models and integrated framework of their combinations are critical. / Doctor of Philosophy / Three numerical models are used to explore the processes occurring in the Earth’s space environment from an altitude of ∼ 100 km to 10s Earth radii (R_E). This environment is mainly filled with plasma, the gaseous state of charged particles that collectively behave like a fluid and are also subject to complex electromagnetic interactions. The intrinsic features of plasma determine that the physics on the scale of charged particles and that on the scale of fluids are both very important. On the other hand, considering the vast differences in the plasma properties throughout space, different regions need to be represented by different physically-based models. This dissertation study addresses the processes on three distinct spatial/temporal scales with different models. A particle model that treats plasma as a group of charged particles is used to explore wave generation in the magnetotail (10s R_E in the nightside). It is found that inhomogeneous plasma flow in the sharp boundary layer at the magnetotail (called “dipolarization front”) can excite plasma waves to dissipate the energy originating from the solar wind (high speed plasma ejected from the sun). A magnetohydrodynamic (MHD) model that treats the plasma as a magnetized fluid is used to explore the efficiency of electric field mapping from the solar wind (10s R_E) to the ionosphere (∼ 100 km altitude). The electric field in the ionosphere usually linearly increases with solar wind electric field until it is too strong. An observational event showed that their relationship remains linear for very large driving field. MHD modeling experiments demonstrate that the linearity at large driving field is due to the high solar wind density, which is explained with force balance theory. An integrated model framework is used to explore the system level response of geospace by investigating the enhanced plasma flow in the subauroral ionosphere (called the subauroral polarization streams, SAPS). The generation of SAPS involves driving and feedback processes in different regions (magnetosphere, ring current, ionosphere) that can not be simulated with any individual model. The global structures and dynamic evolution of SAPS have never been explored before with first-principle characterization of the effects from the solar wind to geospace. This integrated modeling represents a state-of-the-art model framework to explore processes in coupled geospace. These studies illustrate that different models are necessary to explore fundamental physics on small and large scales and the coupling processes between different space regions. It is also suggested that incorporating the different models into an integrated framework is necessary to get a comprehensive understanding of the dynamics in geospace.

The human brain is a complex organ that is responsible for regulating all the physiological processes in the body, ranging from memory to movement. As humans age, the brain goes through a variety of changes including a reduction in glymphatic waste clearance and increase in glial reactivity. Two neurological conditions that affect individuals over the age of 65 include glioblastoma (GBM) and Alzheimer's disease (AD). Interestingly, patients with GBM do not present with AD and vice versa. Both conditions are characterized by a disruption in interstitial fluid flow (IFF) and an increase in neuroinflammation. Throughout the following dissertation, we examined the role of IFF in AD and GBM progression using a three-sided approach (in vivo, in vitro, and in silico). Increased IFF underlies glioma invasion into the surrounding tumor microenvironment (TME) in GBM. We used a 3D hydrogel model of the GBM TME to examine potential pathways by which astrocytes and microglia contribute to glioma invasion. A reduction in IFF contributes to accumulation of the toxic protein amyloid beta (Aβ) in AD. We sought to create a novel, patient-inspired model of the AD hippocampus for examination of the relationship between IFF and Aβ clearance. Human AD and unaffected control hippocampal brain samples were stained for markers of neurons, astrocytes, microglia and Aβ. The percentage of each cell population in the CA1 region of the hippocampus was calculated. We also analyzed the amount and characteristics of the Aβ aggregates present in this hippocampal region. Pearson correlation analysis was completed to assess the relationships between the various cell populations, Aβ load, and patient descriptors. The cell ratios gleaned from the patient samples were incorporated into a novel, 3D hydrogel model of the AD hippocampus. This model features a hydrogel mixture like the native brain extracellular matrix (ECM) and allows for the application of IFF and Aβ. To our knowledge, we are the first group to create a patient-specific triculture model of the AD hippocampus, which is the main site of Aβ aggregation in the AD brain. We used this model to examine the relationship between IFF-mediated Aβ clearance and glial reactivity. The last aim of this dissertation was to create a computational model for examining Aβ binding within the ECM and the effects of IFF on Aβ clearance. In vitro experiments were conducted to generate 3D renderings of glial cells and to determine relevant parameters for our model. Throughout this work, we discuss the relationship between disruption in IFF and glial reactivity in the context of GBM and AD. / Doctor of Philosophy / The human brain is a complex organ that is responsible for regulating all the physiological processes in the body, ranging from memory to movement. As humans age, the brain goes through a variety of changes including a reduction in brain waste removal and an increase in inflammation. Two neurological conditions that affect individuals over the age of 65 include glioblastoma (GBM) and Alzheimer's disease (AD). Interestingly, patients with GBM do not present with AD and vice versa. Both conditions are characterized by a disruption in brain interstitial fluid flow (IFF) and an increase in neuroinflammation. Throughout the following dissertation, we examined the role of IFF in AD and GBM progression using a three-sided approach including analysis of mouse and human tissues, engineered cell models, and computational methods. Specific interactions between brain cell types and their relationships with glioma invasion were examined using a 3D cell model that mimics the brain. Through the work presented here, we also sought to create a novel cell model of the hippocampus region located in the AD brain. We quantified the various cell types in the hippocampus of AD patient samples and incorporated this information into our hydrogel model. The resulting model features three brain cell types (astrocytes, microglia, and neurons) that are added at patient relevant ratios, a matrix that mimics the native brain scaffold, and allows for the application of IFF. In the AD brain there is a reduction in brain waste removal that leads to accumulation of the toxic protein amyloid beta (Aβ). We were successfully able to incorporate this protein within our model so we could assess the relationship between IFF and Aβ removal from the brain. We further studied this relationship using a new computational model of Aβ accumulation in the brain. Throughout this work, we discuss the connection between disrupted IFF and neuroinflammation in the context of GBM and AD.

No description available.

Everyday an enormous amount of information is stored, processed and transmitted digitally around the world. Neurally-inspired compression models have been rapidly developed and researched as a solution to image processing tasks and channel error correction control. This dissertation presents a deep neural network (DNN) for gray high-resolution image compression and a fault-tolerant transmission system with channel error-correction capabilities. A feed-forward DNN implemented with the Levenberg-Marguardt learning algorithm is proposed and implemented for image compression. I demonstrate experimentally that the DNN not only provides better quality reconstructed images but also requires less computational capacity as compared to DCT Zonal coding, DCT Threshold coding, Set Partitioning in Hierarchical Trees (SPIHT) and Gaussian Pyramid. An artificial neural network (ANN) with improved channel error-correction rate is also proposed. The experimental results indicate that the implemented artificial neural network provides a superior error-correction ability by transmitting binary images over the noisy channel using Hamming and Repeat-Accumulate coding. Meanwhile, the network’s storage requirement is 64 times less than the Hamming coding and 62 times less than the Repeat-Accumulate coding. Thumbnail images contain higher frequencies and much less redundancy, which makes them more difficult to compress compared to high-resolution images. Bottleneck autoencoders have been actively researched as a solution to image compression tasks. However, I observed that thumbnail images compressed at a 2:1 ratio through bottleneck autoencoders often exhibit subjectively low visual quality. In this dissertation, I compared bottleneck autoencoders with two sparse coding approaches. Either 50\% of the pixels are randomly removed or every other pixel is removed, each achieving a 2:1 compression ratio. In the subsequent decompression step, a sparse inference algorithm is used to in-paint the missing the pixel values. Compared to bottleneck autoencoders, I observed that sparse coding with a random dropout mask yields decompressed images that are superior based on subjective human perception yet inferior according to pixel-wise metrics of reconstruction quality, such as PSNR and SSIM. With a regular checkerboard mask, decompressed images were superior as assessed by both subjective and pixel-wise measures. I hypothesized that alternative feature-based measures of reconstruction quality would better support my subjective observations. To test this hypothesis, I fed thumbnail images processed using either bottleneck autoencoder or sparse coding using either checkerboard or random masks into a Deep Convolutional Neural Network (DCNN) classifier. Consistent, with my subjective observations, I discovered that sparse coding with checkerboard and random masks support on average 2.7\% and 1.6\% higher classification accuracy and 18.06\% and 3.74\% lower feature perceptual loss compared to bottleneck autoencoders, implying that sparse coding preserves more feature-based information. The optic nerve transmits visual information to the brain as trains of discrete events, a low-power, low-bandwidth communication channel also exploited by silicon retina cameras. Extracting high-fidelity visual input from retinal event trains is thus a key challenge for both computational neuroscience and neuromorphic engineering. % Here, we investigate whether sparse coding can enable the reconstruction of high-fidelity images and video from retinal event trains. Our approach is analogous to compressive sensing, in which only a random subset of pixels are transmitted and the missing information is estimated via inference. We employed a variant of the Locally Competitive Algorithm to infer sparse representations from retinal event trains, using a dictionary of convolutional features optimized via stochastic gradient descent and trained in an unsupervised manner using a local Hebbian learning rule with momentum. Static images, drawn from the CIFAR10 dataset, were passed to the input layer of an anatomically realistic retinal model and encoded as arrays of output spike trains arising from separate layers of integrate-and-fire neurons representing ON and OFF retinal ganglion cells. The spikes from each model ganglion cell were summed over a 32 msec time window, yielding a noisy rate-coded image. Analogous to how the primary visual cortex is postulated to infer features from noisy spike trains in the optic nerve, we inferred a higher-fidelity sparse reconstruction from the noisy rate-coded image using a convolutional dictionary trained on the original CIFAR10 database. Using a similar approach, we analyzed the asynchronous event trains from a silicon retina camera produced by self-motion through a laboratory environment. By training a dictionary of convolutional spatiotemporal features for simultaneously reconstructing differences of video frames (recorded at 22HZ and 5.56Hz) as well as discrete events generated by the silicon retina (binned at 484Hz and 278Hz), we were able to estimate high frame rate video from a low-power, low-bandwidth silicon retina camera.

Current in vitro cellular methods utilised in drug metabolism and pharmacokinetic (DMPK) studies during drug development do not provide the 3D structure and functions of organs found in vivo, such that resulting in vitro-in vivo extrapolation (IVIVE) may not always accurately reflect clinical outcome. This highlights the need for the development of new dynamic in vitro cell models to aid improvement of IVIVE. The aim of this project was to incorporate characterised primary renal cells within a hollow fibre bioreactor for use in DMPK studies investigating renal clearance. Fluorescence based assays were developed to assess the functionality of three drug transporters involved in the renal transport of pharmaceutical compounds: P-gp, BCRP and OCT2. The developed assays were then applied alongside transporter visualisation and genetic expression assays to characterise primary human proximal tubule cells over a series of population doublings. Cells at a population doubling of 5 demonstrated the best transporter activity whilst allowing cells to be expanded in vitro. Polysulfone (PSF) based membranes, which are widely used in dialysis components were developed by blending additives to improve renal cell attachment and culture. The membranes exhibited a characteristic porous internal structure with smooth skin layers on the surface, and were able to be sterilised via autoclaving due to their high thermal stability. PSF blended with polyvinylpyrrolidone (PVP) was the most hydrophilic with cell metabolic activity similar to standard tissue culture plastic. The production of hollow fibres of varying thicknesses and properties from the PSF and PVP blend yielded a marked difference in renal cell attachment and long term viability. Fibres incorporated into glass casings to produce the single hollow fibre bioreactors (HFBs) were able to be sterilised by autoclaving whilst remaining intact. Due to the variation of fibre integrity within the batch, many fibres exhibited tears within the HFBs. This ultimately led to cell depletion within the fibre over the culture period; however, intact fibres demonstrated an increase in cell growth towards the end of the culture period under flow conditions. These results demonstrate the progress made towards a small scale in vitro renal model incorporating characterised primary renal cells to aid the improvement of IVIVE in DMPK research.

Charles University Faculty of Pharmacy in Hradec Králové Department of Pharmacology and Toxicology Student: Simona Hvolková Supervisor: PharmDr. Jana Ramos Mandíková, Ph.D. Title of diploma thesis: Cytotoxicity testing on 2D and 3D model of human liver cells. An inherent part of drug development are in vitro assays, which might be helpful in prediction of drug toxicity. Nowadays, the majority of assays use simple 2D structures for cell growth, but 3D structures with similar conditions to in vivo are becoming more popular. The goal of the study was to assess the cytotoxicity of selected xenobiotics in vitro by both 2D and 3D cell models. The research subjects were drugs from the group of antimycotics (amphotericin B, ketoconazole), NSAIDs (diclofenac, ibuprofen), antipyretics (paracetamol, fenacetine), sodium azide, tamoxifen, para-aminosalicylic acid, methanol and ethanol. For determination of cytotoxicity, the standard colorimetric method (CellTiter 96® ) based on reductive assessment of metabolic active cells was used. For drug testing it was used human standard line of liver cells HepG2. The cells were cultivated in monolayer or in 3D form with the Alvetex® Scaffold technology using high porous networked polystyrene. The parameter of inhibition concentration IC50 was chosen for toxicity assessment of...

Huntington’s disease (HD) is an autosomal dominant inherited neurodegenerative disease that specifically affects the striatum of the human brain. HD is characterized by a chorea-like movement disorder, cognitive decline, and psychiatric symptoms. In Europe, it has a relatively high prevalence of about 2.17-7.33 per 100,000 people compared with other continents. By far, there is no cure for HD. The mean survival time of patients after the diagnosis of HD is 15 to 20 years. Although the mutant form of the Huntingtin (HTT) as the cause of HD has been confirmed for decades, the exact pathogenesis of HD is still elusive. More recently, large global genome-wide association studies (GWAS) and several other studies provided new insights for HD mechanism, by highlighting several genes involved in DNA damage repair mechanisms as modifiers of age at onset and disease severity in HD. Thus, this project focused on the investigation of DNA damage response and repair in HD in vitro models. Fused in sarcoma (FUS) was the protein of our interest, as it has been confirmed to participate in DNA damage response and repair in multiple ways. Furthermore, FUS protein was implicated to have a relationship with neurodegenerative diseases, as it was found to play a role in the pathogenesis of subtypes of amyotrophic lateral sclerosis and frontotemporal dementia. FUS was also found to co-localize with mutant huntingtin protein in intracellular aggregates in HD mice models. In this project, donor/patient-specific induced pluripotent cells (iPSCs) and its derived striatal neurons were the main materials. By immunofluorescence staining approach of γH2AX and 53BP1, DNA double-strand breaks (DSBs) damage was investigated on iPSCs-derived in vitro striatal neurons. HD neurons showed an obvious and excessive accumulation of DNA DSBs damage. Then, in order to visualize FUS protein during DNA damage response procedure, eGFP tagged endogenous wild-type FUS iPSCs were generated, and later were differentiated into striatal neurons. UVA laser micro-irradiation was applied onto both hiPSCs and their differentiated striatal neurons in vitro models, simultaneously conducting with live-cell imaging approach. FUS was found to recruit to the DNA damage site induced by laser irradiation. For studying the kinetics of wild-type FUS protein during the response to laser irradiation, a novel and robust workflow was generated. By this workflow, the kinetics of FUS protein was characterized into four phases and a real-time scale of the kinetics was offered. After comparisons, a prominent change of FUS kinetics in HD at neuron-stage but not iPSC-stage was found. Furthermore, an intriguing different performance of FUS protein was found in different types of in vitro cellular models. In iPSCs, not all the laser-irradiated cells recruited FUS at the DNA damage site. The kinetics of the FUS protein also differed in different models. In conclusion, first, our in vitro striatal neuron model recapitulated the impaired DNA damage repair phenotype that published by other models. Second, new evidence was offered that wild-type FUS was involved in the pathogenesis of HD. Third, depending on cell-type, FUS performed differently during the response to the laser irradiation-induced DNA damage. Thus, these results suggest that the impaired DNA damage response and repair would be crucial to the mechanism of HD. Furthermore, the role of FUS protein playing especially the functional part in DNA damage response and repair might be a potential target for further investigation of neurodegenerative diseases including HD.