This thesis presents the design, fabrication, and testing of a micromixing device intended for use in investigating protein dynamics on a microsecond timescale by Fourier transform infrared (FTIR) spectroscopy. Numerical modeling of flow was implemented to predict the influence of flow rates and geometric variations on mixing performance in three passive mixers. The simulation models were validated by experimental measurements using optical and infrared detection. The optimum level of mixing was observed in a multi-lamination mixer that combined thin filaments of differing fluids in an alternating manner. The multi-laminates were transferred onto polished calcium fluoride infrared-transparent optical windows by lithographic processing of an Epon-based polymer, SU-8. A rigid seal between two microchannels was accomplished through thermal bonding of an unexposed resist layer, which acted as a thermal epoxy under the influence of temperature. The multi-lamination mixer was used to study the changes in the secondary structure of beta-Lactoglobulin in deuterated phosphate buffer under varying physicochemical conditions by time-resolved FTIR spectroscopy using focal plane array detection. Upon a pH jump from pH 2 to neutral pH, a gradual loss of alpha-helical content, accompanied by an increase in random coils and turns was observed within 2 ms of mixing. In a second kinetic experiment, mixing of a neutral-pH solution of beta-Lactoglobulin with a 60% trifluoroethanol solution resulted in the formation of an alpha-helical intermediate with an accompanying increase in intramolecular beta-sheet structure within 500 mus of mixing. These results indicate that the multi-lamination mixer designed and fabricated in this study is well suited for investigations of protein dynamics on the micro- to millisecond timescale by time-resolved FTIR spectroscopy.

This study seeks to design algorithms that may be used to determine if a given lattice is a good approximation to a given rigid protein structure. Ideal lattice models discovered using our techniques may then be used in algorithms for protein folding and inverse protein folding. In this study we develop methods based on dynamic programming and branch and bound in an effort to identify “ideal” lattice models. To further our understanding of the concepts behind the methods we have utilized a simple cubic lattice for our analysis. The algorithms may be adapted to work on any lattice. We describe two algorithms. One for aligning the protein backbone to the lattice as a walk. This algorithm runs in polynomial time. The second algorithm for aligning a protein backbone as a path to the lattice. Both the algorithms seek to minimize the CRMS deviation of the alignment. The second problem was recently shown to be NP-Complete, hence it is highly unlikely that an efficient algorithm exists. The first algorithm gives a lower bound on the optimal solution to the second problem, and can be used in a branch and bound procedure. Further, we perform an empirical evaluation of our algorithm on proteins from the Protein Data Bank (PDB). / ix, 47 leaves ; 29 cm.

Accumulation of misfolded proteins in the endoplasmic reticulum (ER) causes numerous human pathologies. Biochemical evidence suggests that soluble misfolded proteins are retrotranslocated out of the ER, via the endoplasmic reticulum associated degradation (ERAD) pathway, for proteosome-mediated cytoplasmic degradation. Excess, misfolded- or insoluble proteins, are suggested to cause induction of “overflow” degradation pathways. For soluble proteins, overflow to vacuole-mediated destruction is suggested to occur via two Golgi-to-vacuole (Gvt) routes, the alkaline phosphatase (ALP), direct route, or, a carboxypeptidase Y- (CPY-), prevacuolar compartmentvacuole, indirect route, though only the CPY route is thought to degrade soluble proteins. Insoluble aggregate-containing structures are suggested to be degraded by engulfment by membranes of unknown origin and trafficking to the vacuole for destruction, via an autophagic pathway. To confirm biochemical evidence, wild-type (BY4742), autophagosome- (W303/ATG14), CPY- and autophagy pathway- (W303/VPS30), and proteosome (WCG/2) mutants of S. cerevisiae yeasts were transformed with a high expression pYES plasmid and mutant (Z) human alpha-1- proteinase inhibitor (A1PiZ), giving rise to the derivatives cells BY4742/Z, W303/ATG14/Z, W303/VPS30/Z and WCG/2/Z, respectively. Electron microscopy using gold labeling for A1PiZ, markers for the ER, the ERAD ER channel protein, Sec61, or the chaperone, binding protein (BiP), ALP for the ALP pathway, and CPY for the CPY pathway, was used. Overexpression of A1PiZ seems to result in targeting to the vacuole via a prevacuolar, CPY-like compartment (PVC, 200-500 nm), though CPY and A1PiZ appears not to colocalise, unconvincingly confirming collaborative biochemical data. Large amounts of A1PiZ localise in the cytosol, possibly indicating a largely proteasome-mediated degradation. ER-resident A1PiZ targeting to the vacuole seems also to occur by the budding of the ER and peripheral plasma membrane or ER membrane only. This occurs in all cells, but especially in ATG14 gene (ΔATG14) mutants, possibly indicating autophagosome-mediated degradation independence, in the latter mutants. The ATG14 mutation gave rise to crescent-shaped, initiating membranelike (IM-like) structures of approximately Cvt vesicle-diameter, possibly indicating that ΔATG14 blocks autophagosome- (500-1000 nm) and Cvt vesicle (100-200 nm) enclosure, after core IM formation. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2008.

Recent studies of the relationship between protein sequence and protein structure are reviewed. A detailed discussion of past attempts to predict the structure of a protein from its amino acid sequence, the protein folding problem, is presented and the strengths and weaknesses of these methods are examined. The root-mean-square deviation is studied and a benchmark for structural comparisons is established. A combinatorial approach to the protein folding problem is outlined and its advantages over existing methods is discussed. Specific algorithms based on the combinatorial approach are developed and applied to a variety of proteins. The success of this approach in terms of the root-mean-square deviation benchmark as well as the drawbacks of this method are presented.

Membrane proteins are essential cellular components, responsible for a wide variety of biological functions. In order to better understand such aspects of cell activity, researchers have pursued detailed structural analysis of this class of proteins. Because of the complexities in isolating and studying membrane proteins in their native environment, detergents are often employed as a membrane mimetic media. This thesis examines several features of transmembrane (TM) protein structure and folding in detergents through which we are able to gain insights into membrane protein folding, as well as explore the suitability of detergents as membrane-mimetic environments. We first compare the helix-helix association of a series of model TM sequences in a native bilayer to the corresponding association in a detergent environment. We find that while various classes of helix-helix interaction motifs are preserved in detergents, alterations in detergent solvation may, in turn, lead to altered association affinity. We further explore this phenomenon through investigation of the consequences of the insertion of a strongly polar residue into a TM segment. In these studies we find a correlation between sequence-dependent alterations in detergent solvation and predicted in vivo folding. We also extend such analyses to a variety of detergents and native TM segments, finding that native secondary structure, as it occurs in the context of a full-length protein, is generally well preserved in a variety of detergents. Finally, we assess the determinants of membrane protein folding using two-transmembrane segment constructs, in the process optimizing expression, production and characterization techniques for a diverse range of transmembrane protein sequences. Overall this thesis finds that, detergents are capable of solubilizing membrane proteins in a form suitable for in-depth structural characterization that may not be feasible in other environments. Thus, as an approximation of a native membrane, detergents are able to preserve certain features of membrane proteins such as helix-helix association and native secondary structure.

Membrane proteins are essential cellular components, responsible for a wide variety of biological functions. In order to better understand such aspects of cell activity, researchers have pursued detailed structural analysis of this class of proteins. Because of the complexities in isolating and studying membrane proteins in their native environment, detergents are often employed as a membrane mimetic media. This thesis examines several features of transmembrane (TM) protein structure and folding in detergents through which we are able to gain insights into membrane protein folding, as well as explore the suitability of detergents as membrane-mimetic environments. We first compare the helix-helix association of a series of model TM sequences in a native bilayer to the corresponding association in a detergent environment. We find that while various classes of helix-helix interaction motifs are preserved in detergents, alterations in detergent solvation may, in turn, lead to altered association affinity. We further explore this phenomenon through investigation of the consequences of the insertion of a strongly polar residue into a TM segment. In these studies we find a correlation between sequence-dependent alterations in detergent solvation and predicted in vivo folding. We also extend such analyses to a variety of detergents and native TM segments, finding that native secondary structure, as it occurs in the context of a full-length protein, is generally well preserved in a variety of detergents. Finally, we assess the determinants of membrane protein folding using two-transmembrane segment constructs, in the process optimizing expression, production and characterization techniques for a diverse range of transmembrane protein sequences. Overall this thesis finds that, detergents are capable of solubilizing membrane proteins in a form suitable for in-depth structural characterization that may not be feasible in other environments. Thus, as an approximation of a native membrane, detergents are able to preserve certain features of membrane proteins such as helix-helix association and native secondary structure.

Protein disulfide isomerases (PDIs) catalyze the oxidation and isomerization of disulfide bonds in proteins passing through the endoplasmic reticulum (ER). Although as many as 20 enzymes are classified as PDI family members, their relative contributions to protein folding have remained an open question. Additionally, Ero1 has been characterized as the ER oxidase that transfers oxidizing equivalents from oxygen to PDI enzymes. However, knockout mice lacking the mammalian Ero1 isoforms, Ero1Lα and Ero1Lβ, are viable, and the role of other potential ER oxidases in maintaining an oxidative ER environment is now an important issue. By systematic depletion of ER PDI family members and potential ER oxidases and assessment of disulfide bond formation of secreted endogenous substrates, I have outlined the functional relationships among some of these enzymes. PDI family member depletion revealed that PDI, although not essential for complete disulfide bond formation in client proteins, is the most significant catalyst of oxidative folding. In comparison, ERp57 acts preferentially on glycosylated substrates, ERp72 functions in a more supplementary capacity, and P5 has no detectable role in formation of disulfide bonds for the substrates assayed. Initially, no impact of depletion of Ero1 was observed under steady state conditions, suggesting that other oxidase systems are working in parallel to support normal disulfide bond formation. Subsequent experiments incorporating a reductive challenge revealed that Ero1 depletion produces the strongest delay in re-oxidation of the ER and oxidation of substrate. Depletion of two other potential ER oxidases, peroxiredoxin 4 (PRDX4) and Vitamin K epoxide reductase (VKOR), showed more modest effects. Upon co-depletion of Ero1 and other oxidases, additive effects were observed, culminating in cell death following combined removal of Ero1, PRDX4, and VKOR activities. These studies affirm the predominant roles of Ero1 in ER oxidation processes and, for the first time, establish VKOR as a significant contributor to disulfide bond formation.

In this thesis we present computer modelling studies that were implemented to investigate protein behavior in various environments causing their folding, unfolding and aggregation. Applications related to two important proteins - insulin and apolipoprotein C-II (ApoC-II) are presented. The use of atomistic simulation methodologies based on empirical force fields has enhanced our understanding of many physical processes governing protein structure and dynamics. However, the force fields used in classical modelling studies are often designed for a particular class of proteins and rely on continuous improvement and validation by comparison of simulations with experimental data. In Chapter 4 we present a comprehensive comparison of five popular force fields for simulation of insulin. The effect of each force field on the conformational evolution and structural properties of the protein is analysed in detail and compared with available experimental data. A fundamental phenomenon in nature is the ability of proteins to fold ab initio to their functional native conformation, also known as their biologically active state. Due to the heterogeneity and dimensionality of the systems involved, it is necessary to employ methodologies capable of accelerating rare events, specifically, configurational changes that involve the crossing of large free energy barriers. In Chapter 5, using the recently developed method BE-META we were able to identify the structural transitions and possible folding pathways of insulin. Another interesting phenomenon is the misfolding of proteins causing their aggregation, that may lead to formation of either amorphous compounds or structures of elongated-unbranched morphology known as amyloid fibrils. The deposition of amyloid fibrils in the human body may cause many debilitating diseases such as Alzheimer's and variant Creutzfeldt-Jakob diseases, thus making this field of research important and urgent. The human plasma protein apoC-II serves important roles in lipid transport, and it has been shown to form amyloid-like aggregates in solution. We have performed computational studies to investigate the effect of mutations, such as Met oxidation and the residue substitutions to hydrophobic Val and hydrophilic Gln, on dynamics of apoC-II(60-70) peptide. The conformation features relevant to the amyloidogenic propensities of the peptide were identified and presented in Chapter 6. The involvement of lipids at the various stages of development of amyloid diseases is becoming more evident in recent research efforts. In particular, micellar and sub-micellar concentrations have showed to have different effect on fibril growth and kinetics of native apoC-II and derived peptides. In Chapter 7 we investigated the influences of phospholipids at various concentrations on the structure of apoC-II(60-70) using MD and umbrella sampling methods. The molecular mechanisms of lipid effects on the peptide conformation and dynamics were identified. In Chapter 8 preliminary results on the structural stability of pre-formed oligomeric composites of apoC-II(60-70) peptide of different sizes and arrangements were also presented. The effects of mutation (oxidised Met, Met60Val and Met60Gln) on the most stable cluster was also investigated. To conclude, several ideas for continuation of research in the protein folding and aggregation field are discussed in the Future Work section of this thesis.

Thesis (Ph.D.) -- McMaster University, 1997 / Includes bibliographical references (leaves 151-167) Also available via World Wide Web.

Diss. (sammanfattning) Stockholm : Karolinska institutet, 2007. / Härtill 3 uppsatser.