Compartmentalization allows for the spatial organization of cellular components and is crucial for numerous biological functions. One recently uncovered strategy for intracellular compartmentalization is phase separation via the de-mixing of biomacromolecules. Membraneless organelles, also referred to as biomolecular condensates, are compartments formed by phase separation and create distinct environments that are essential to cellular processes ranging from cell signaling to gene expression. Biomolecular condensates offer several advantages – for example, dynamic restructuring of internal constituents and diffusion of cellular components into/out of compartments – that make them suitable for applications in biocatalysis or pharmaceutical production. However, the ability to independently engineer the formation and disassembly of condensates in vivo remains a challenge. Here, concepts from polymer science have been used to understand parameters that govern intracellular phase separation. Many biomolecular condensates exhibit physical properties that are similar to complex coacervates as both are liquid-like phase separated mixtures formed via associative phase separation, frequently with oppositely charged polyelectrolytes. We utilize the physical phenomenon of complex coacervation and principles underlying the formation of liquid-like biological condensates to identify design parameters for engineering synthetic, phase separated organelles in E. coli. In this dissertation, we employed a library of cationic charge variants derived from superfolder green fluorescent protein (sfGFP) to elucidate the effects of overall cationic charge on intracellular phase separation. We first investigated the complex coacervation of engineered proteins with biological polyelectrolytes to determine predictive design rules for protein phase separation and translated these design rules in vivo to engineer bacterial condensates. Characterization of the coacervate-like properties and macromolecular composition revealed that these condensates can undergo dynamic restructuring and exhibit biomolecular specificity. To facilitate the engineering of active supercharged proteins, we also developed short, cationic peptide tags, ranging from 6-27 amino acids in length, that can be appended onto any protein of interest to promote intracellular phase separation. We find that overall charge generally determines protein phase behavior and observe the formation and disassembly of condensates near the physiological phase boundary. Interestingly, we find that small modifications in charge density can tune the interaction strength between associating biomacromolecules and thus tune condensate stability. We demonstrate the use of these protein design parameters and cationic peptide tags to sequester catalytic enzymes and manipulate the intracellular localization of multiple proteins. These studies pave the way to building synthetic, functional organelles.

No description available.

The mechanisms by which altered processing, distribution and secretion of proteolytic enzymes occur, facilitating degradation of the extracellular matrix in invasive and metastatic cells, are not fully understood. Studies on the MCF-10 A breast epithelial cell line and its premalignant, c-Ha-ras-transfected MCF-10AneoT counterpart have shown that the ras-transfected cell line has a more alkaline pH. The objective of this study was to determine which organelles of the endosome-lysosome route were alkalinized and shifted to the cell periphery after ras-transfection. Antibodies to the hapten 2,4-dinitrophenyl (DNP), required for pH studies, were raised in rabbits and chickens using DNP-ovalbumin (DNP-OVA) as immunogen. Cationised DNP-OVA (DNP-catOVA) was also inoculated to increase antibody titres. Anti-hapten and carrier antibody titres were assessed. In rabbits, cationisation seems useful to increase anti-DNP titres if a non-self carrier protein (OVA) is used. In chickens, cationisation of DNP-OVA seems necessary to produce a sustained anti-OVA (anti-self) response (implying a potential strategy for cancer immunotherapy). Oregon Green® 488 dextran pulse-chase uptake and fluorescent microscopy, and (2,4-dinitroanilino)-3'-amino-N-methyldipropylamine (DAMP) uptake, immunolabelling for DNP (a component of DAMP) and unique markers for the early endosome (early endosome antigen-I, EEAI), the late endosome (cation-independent mannose-6-phosphate receptor, CI-MPR) and the lysosome (small electron dense morphology and lysosome-associated membrane protein-2, LAMP-2) and electron mlcroscopy was performed. The pH of late endosomes and lysosomes in the ras-transfected MCF-10AneoT cell line were found to be relatively alkalinised and Iysosomes shifted toward the cell periphery. The acidic pH of late endosomes is required to release precursor cysteine and aspartic proteases from their receptors (e.g. CI-MPR), process the precursors to active proteases and to allow receptor recycling. The more alkaline pH observed potentially explains the altered processing of proteases in rastransfected cells. Alkalinisation ofthe cytosol may affect the cytoskeleton responsible for, among other things, the positioning and trafficking of various organelles, causing relocation of Iysosomes toward the cell periphery and actin depolymerisation. This may enable fusion of Iysosomes with the plasma membrane and the release of proteolytic enzymes, facilitating the observed invasive phenotype. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2005.