Engineering antibody therapeutics and cell culture models for neurodegenerative diseases

Date
2016
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Neurodegenerative diseases are typically characterized by selective neuronal subtype vulnerability and associated incorrectly processed and misfolded amyloidogenic proteins. These disease pathologies differentially affect anatomically specialized areas made up of dozens of different subtypes of neurons interconnected in complicated pathways. Currently, there are no established treatments to halt or even slow the progression of these diseases. To deconvolute these interacting effectors of neuronal dysfunction, we can develop neurodegeneration cell culture models to break up different components into intrinsic and extrinsic effects. Intrinsic effects consist of age-related dysfunction and neuron subtype identity and how the type of neuron mediates selective vulnerability. Extrinsic effects include neuro-inflammation factors, apoptotic factors released from nearby dying cells, and misfolded disease proteins such as tau protein as they travel through the brain. By understanding some of these effects in isolation we can identify points of therapeutic intervention and quantify disease biomarkers to track disease progression and therapeutic efficacy. In this work, we develop a high-throughput platform for neuronal reprogramming to culture specific subtypes of patient-derived neurons for studying cell type-specific intrinsic properties, we establish an improved culture model of tau protein disease strain fibrillization to study extrinsically-added misfolded protein, and we demonstrate the feasibility of engineering therapeutic proteins in a heterologous host for improved stability and expression. ☐ Huntington’s disease (HD) is a dominantly inherited, progressive neurodegenerative disease characterized by specific and extensive loss of medium spiny neurons during striatal degeneration. It remains unclear whether this cell-type selectivity is primarily due to intrinsic cell-autonomous effects or neuronal dysfunction in surrounding neurons. The ability to obtain relevant cell-types in vitro would be an invaluable resource to study these cell-type specific effects. We assembled and screened a library of transcription factors to reprogram human fibroblasts or induced pluripotent stem cells from HD and non-HD patients into neuronal subtypes important in HD. We also developed a combinatorial computational model to assist in the selection of experimental parameters to optimize library screening efficiency. The resulting identified sets of transcription factors may be used to develop a fast, reproducible in vitro model that accurately captures the major phenotypic characteristics of HD in order to further understand the intrinsic cell-type specific pathogenesis of the disease. ☐ The microtubule-associated protein tau stabilizes microtubules in neurons for cytoskeletal support and cellular trafficking. Fibrillization and accumulation of tau is a defining characteristic of a group of neurodegenerative diseases called “tauopathies” in which tau forms distinct fibril strain conformations across different tauopathies which may contribute to their unique histological and clinical characteristics. Most current tauopathy cell models rely on the application of synthetic tau fibrils formed by induced fibrillization with the poly-anion heparin. While it has been previously demonstrated that heparin-induced fibrils are structurally distinct from brain-derived tauopathy fibrils, we have shown that synthetic fibrils are also significantly different when propagated in a cell culture model of extrinsically-applied tau strain propagation. Additionally, while truncated or mutant tau has been useful to study the mechanics of synthetic fibril propagation, we have shown that full length wild type tau is better suited for propagation of disease-associated strains in this model. The cell culture model presented in this study may be broadly applicable to many other neurodegenerative diseases. ☐ The protein-only hypothesis for prion-like diseases asserts that neurodegenerative disease-associated proteins such as the prion protein or tau protein can spread their misfolded disease conformation to natively folded protein by direct intermolecular interaction to template the misfolded structure. Understanding this interaction suggests a therapeutic approach in which a specifically-binding therapeutic may be able to disrupt the interaction of native and misfolded protein to halt the progression of the disease. We chose an anti-prion antibody ICSM18 that was previously engineered by our group to better understand the feasibility and fidelity of engineering therapeutic proteins in a heterologous host. CHO cells are the most common platform for protein secretion in the pharmaceutical industry, but Saccharomyces cerevisiae are an attractive engineering platform because of their eukaryotic expression machinery and rapid doubling time. We found that apparent affinity by yeast display may be significantly lowered by differences in expression machinery between the two expression systems but that improvements in stability in yeast translate directly to improved expression yields in CHO cells which indicates that yeast display may not provide exact affinity estimation but that complex protein therapeutics may still be efficiently engineered in a heterologous host.
Description
Keywords
Citation