Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2009. / Includes bibliographical references. / Our DNA is constantly under the assault of DNA damaging agents that are ubiquitous in nature and unavoidable. Fortunately, our cells have evolved DNA repair mechanisms to maintain genomic integrity against this constant attack. An important type of DNA damage is alkylation damage, which has been the focus of this thesis, the major goal of which is to explore the biological role of a set of alkylation repair proteins, the E. coli AlkB and two human AlkB homologs (ABH2 and ABH3), and how they relate to the 3methyladenine DNA glycosylase (AAG). AAG is a base excision repair (BER) protein that has been well-studied and is known to be involved in the repair of a wide variety of substrates. On the other hand, the direct reversal protein AlkB, and its human homologs, have not been so extensively characterized, but it is known that they can repair not only DNA, but also RNA. Although there are eight human AlkB homologs, attention was focused on ABH2 and ABH3 since they are the more well-characterized homologs and recently implicated in DNA repair.In order to investigate the role of the AlkB proteins, particularly in mammalian cells, I expressed ABH2 and ABH3 in established human cell lines and investigated whether their expression would enhance cell survival after alkylation treatment. However, no detectable phenotype was observed in the cell lines upon treatment with the alkylating agent methyl methanesulfonate (MMS). This is possibly due to endogenous ABH levels being sufficient for repair. We therefore turned to characterization of the Abh2 and Abh3 null mice, as compared to wildtype and to another alkylation repair deficient animal, Aag null mice. In addition to the primary substrates 1methyladenine and 3-methylcytosine, AlkB, ABH2, and ABH3 can also repair an important class of damage, the etheno base DNA lesions, which can also be repaired by AAG. / (cont.) Here we have shown in a mouse model that Abh2 and Abh3 overlap with Aag in protecting mice from sensitivity in response to chemically induced chronic inflammation, in which etheno base lesions are readily generated. In addition, we also employed a biochemical approach using a comprehensive library of lesion-containing DNA oligonucleotides to study the redundancy in repair activity between AAG and AlkB. In doing so, we have found new substrates for AAG and in particular, 1-methylguanine, is a new substrate shared between AAG and AlkB. Thus, although these two proteins employ different mechanisms for repair, our studies established further evidence of the interplay between these proteins and the different repair pathways they represent, underscoring the importance of alkylation damage repair for proper cell homeostasis. / by Chun-Yue Lee. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2005. / Includes bibliographical references. / This thesis focused on the development and analysis of a new class of magnetic fluids for recovery of recombinant proteins from fermentation broth. Magnetic fluids are colloidally stable dispersions of magnetic nanoclusters in water that do not settle gravitational and moderate magnetic fields due to their small size. The magnetic nanoclusters possess large surface area for protein adsorption without any porous structure, resulting in much faster mass transfer than in traditional separations. The magnetic nanoclusters consist of 25-200 nm clusters of 8 nm magnetite (Fe₃0₄) cores coated with poly(acrylic acid-co-styrenesulfonic acid-co-vinylsulfonic acid). For use in separation, clusters must be recoverable from solution. Individual nanoparticles are too small to be recovered efficiently, while 50nm or larger clusters of primary particles are easily recovered. Cluster size depends on polymer molecular weight and hydrophobicity as well as the amount of polymer present at nucleation. When a polymer coating with optimal molecular weight is used in limited amounts, clusters are formed. When the clusters are subsequently coated with additional polymer, the clusters are stable in high ionic strength environments (>5M NaCl), while retaining the necessary cluster size for efficient magnetic recovery. / (cont.) Models have been developed to predict the optimal molecular weight, and the cluster size obtained with limited amounts of polymer or polymers other than the optimal molecular weight. The models and methods have been verified with other polymer coatings, indicating that the methods can be used to synthesize a wide range of stable nanoclusters. Due to rapid mass transfer, the rate-limiting step of the purification scheme is recovery of the nanoclusters from solution with high gradient magnetic separation (HGMS). The nanoclusters can be recovered extremely efficiently, up to 99.9% at high flow rates, up to 10,000 cm/hr. A detailed model of HGMS has been developed to quantitatively predict capture, and simpler methods have been developed to predict the maximum capture and capacity of the column without computationally expensive simulations. The use of the nanoclusters for protein purification was studied both with model proteins the recombinant protein drosomycin from Pichia pastoris fermentation broth. The nanoclusters have high adsorptive capacities of up to 900 mg protein/mL adsorbent, nearly an order of magnitude higher than the best commercially available porous adsorbents. Adsorption can be performed both by ion exchange and hydrophobic interactions, allowing nearly pure drosomycin to be recovered from clarified fermentation broth in a single step. / (cont.) When used in whole cell broth, the nanoclusters bind to proteins on the surface of the Pichia pastoris cells at conditions where drosomycin is bound, limiting the effectiveness of the separation. When proteins are bound at conditions where nanoclusters do not bind to cells, cell clarification and protein purification can be performed in one fast step. A simple model of the cell binding has been developed, providing guidelines for use of magnetic nanoparticles in the presence of cells. / by Andre Ditsch. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, February 2008. / Includes bibliographical references. / The extracellular matrix (ECM) has been shown to be a significant source of hindrance to the transport of macromolecules in solid tumors. This thesis shows that by limiting their interstitial transport, the tumor ECM can reduce the efficacy of cancer therapeutics. Furthermore, techniques for overcoming this transport barrier and improving the effectiveness of existing cancer therapetuics are developed. Mathematical modeling was utilized to characterize the distribution of a therapeutic herpes simplex virus (HSV) vector in solid tumors. The model showed that the spread of virus following intratumoral injection is severely limited by rapid binding and limited diffusion. Importantly, the model demonstrates that an improvement in virus diffusion can enhance its distribution significantly. In vivo multiphoton imaging of fibrillar collagen type I and injected HSV vectors largely supported the model predictions. Injected viral particles could not penetrate dense networks of fibrillar collagen. Both the initial distribution and subsequent propagation of these replication-competent vectors were limited by collagen. Degradation of tumor collagen with bacterial collagenase enhanced the distribution and efficacy of the virus. To develop a technique that is clinically applicable, human collagenases were screened for similar activity. Matrix metalloproteinase (MMP) -1 and -8 were identified as viable candidates and tested for their ability to degrade collagen and enhance diffusion in tumors. When overexpressed in tumors, neither MMP significantly altered collagen content or diffusive transport. However, these MMPs have multiple matrix substrates and were found to deplete the tumor of sulfated glycosaminoglycans (GAGs). This, in turn, increased the hydraulic conductivity of these tumors, enhancing the distribution and efficacy of infused oncolytic HSV. Genetic mutations were employed to enhance the activity of these enzymes, but impaired intracellular processing and inactivating autoproteolytic degradation reduced overall activity. / (cont.) Thus, our work demonstrates the importance of the tumor extracellular matrix in regulating the distribution and efficacy of cancer therapeutics. Methods to modulate both tumor collagen and sulfated GAGs are developed to enhance interstitial transport and improve the treatment of solid tumors. / by Wilson Mok. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2010. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from student submitted PDF version of thesis. / Includes bibliographical references (p. 177-192). / Microfluidic systems provide valuable tools for exploring, studying, and optimizing organic syntheses. The small scales and fast transport rates allow for faster experiments and lower amounts of chemicals to be used, reducing costs and increasing safety. Additionally, continuous flow processes allow for a large number of experiments to be performed after a single setup. These advantages were exploited to enable continuous-flow study of chemical syntheses that are hazardous or difficult to perform by conventional methods and of applying the acquired knowledge toward improvement of industrial processes at large scales. Using silicon semiconductor microfabrication techniques, microdevices have designed and produced to address various challenges in continuous-flow reaction study and synthesis, enabling operation at reaction conditions not easily obtained in batch setups or on macroscopic scale. Several model reactions and systems were selected for study and/or augmentation. Silicon micromixers were designed and microfabricated to ensure low-pressure-drop millisecond-scale mixing of liquid streams. The micromixers were used to perform a quantitative kinetics and scale-up study of the direct two-step synthesis of sodium nitrotetrazolate by a Sandmeyer type reaction via a reactive diazonium intermediate. The use of continuous-flow microsystems significantly reduced the typically high explosion hazard associated with the energetic product and intermediate. An epoxide ring opening reaction was augmented and kinetics of the reaction were rapidly obtained using a silicon microreactor at high temperatures and pressures, demonstrating microreactor utility for rapid reaction space profiling, as well as the use of continuous flow to easily study and sample reaction conditions not readily accessible in batch. Scale-up was demonstrated using obtained kinetics. Synthesis steps of two pharmaceutical APIs were thus studied and greatly accelerated, which may be useful for considerations of continuous manufacturing. Finally, a system has been designed and studied to enable microfluidic study of solids forming reactions such as an organic coupling reaction with inorganic salt byproduct precipitate. Conventionally, these solids render such reactions difficult to study in microreactors, which limits the types of chemistries that could be investigated and improved using microfluidic technology. To minimize these constraints, the formation of solids in flow was systematically studied, and a combination of reactor design and application of acoustic forces to effect solid agglomerate disruption was used to allow slurries with relatively large amounts of solids to flow through microchannels. / by Nikolay Zaborenko. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2011. / Cataloged from student submitted PDF version of thesis. / Includes bibliographical references. / Electrospinning is a straight forward method to produce fibers with diameter on the order of a few tens of nanometers to the size approaching commercial fibers (on the order of 10 prm or larger). Recently, the length scale effect on physical properties has attracted great attention because of the potential to produce new materials with unique behavior. In general, the behavior of commercial fibers can be investigated by traditional experiments, and that of nanofibers can be studied by molecular dynamics simulation or Monte Carlo technique. However, the transition of their properties from the bulk to the nanoscale materials is not well understood. Electrospinning provides us a bridge to understand the properties of fibers transiting from the behavior of the bulk material to that of the nanofibers. Among these areas, I am interested in the possible remarkable changes in mechanical properties that may occur in electrospun fibers due to the size effect, where the comprehensive understanding is still lacking. My research objectives are to understand mechanical properties of electrospun polymeric fibers as a function of their size, structure and morphology. The first part of my research is to study internal structures and external topographies of electrospun fibers, and to understand their effect on mechanical properties. Amorphous polystyrene (PS) and semicrystalline polyacrylonitrile (PAN) were dissolved in a high boiling point solvent, dimethylformamide (DMF), for electrospinning. When electrospun in a high-humidity environment, the interior of these fibers was found to be highly porous rather than consolidated, despite the smooth and nonporous appearance of the fiber surfaces. The formation of interior porosity is attributed to the miscibility of water, a nonsolvent for the polymers in solution, with DMF. The resulting morphology is a consequence of the relatively rapid diffusion of water into the jet, leading to a liquid-liquid phase separation that precedes solidification due to evaporation of DMF from the jet. When electrospun in a low humidity environment, the fibers exhibit a wrinkled morphology that can be explained by a buckling instability. Understanding which structures and morphology form under a given set of conditions is achieved through the comparison of three characteristic times: the drying time, the buckling time and the phase separation time. The structures and morphology have important consequences for the properties of the fibers such as their mechanical strength and stiffness. / (cont.) Secondly, we studied the size effects of single electrospun fibers on their stiffness and strength. The Young's modulus and yield strength of individual electrospun fibers of amorphous poly(trimethyl hexamethylene terephthalamide) (PA 6(3)T) have been obtained in uniaxial extension. The Young's modulus is found to exhibit values in excess of the isotropic bulk value, and to increase with decreasing fiber diameter for fibers with diameter less than roughly 500 nm. The yield stress is also found to increase with decreasing fiber diameter. These trends are shown to correlate with increasing molecular level orientation within the fibers with decreasing fiber diameter. Using Ward's aggregate model, the correlation between molecular orientation and fiber modulus can be explained, and reasonable determinations of the elastic constants of the molecular unit are obtained. Finally, we identified a relation of stiffness between single electrospun fibers and their nonwoven fabrics. This is of interest because adequate mechanical integrity of nonwoven fabrics is generally a prerequisite for their practical usage. The Young's modulus of electrospun PA 6(3)T nonwoven fabrics were investigated as a function of the diameter of fibers that constitute the fabric. Two quantitative microstructure-based models that relate the Young's modulus of these fabrics to that of the fibers are considered, one assuming straight fibers and the other allowing for sinuous fibers. This study is particularly important for meshes comprising fibers because of our recent discovery of an enhanced size effect on their Young's modulus as well as the tendency towards a curved fiber topology between fiber junctions. The governing factors that affect the mechanical properties of nonwoven mats are the fiber network, fiber curvature, intrinsic fiber properties, and fiber-fiber junctions. Especially for small fibers, both the intrinsic fiber properties and fiber curvature dominate the mechanical behavior of their nonwoven fabrics. This thesis helps us to understand the mechanism behind the enhanced mechanical behavior of small fibers, and to identify determining parameters that can be used to tailor their mechanical performance. / by Chia-Ling Pai. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2001. / Includes bibliographical references. / Organosilicon thin films deposited by pulsed plasma-enhanced chemical vapor deposition (PPECVD) and hot-filament chemical vapor deposition (HFCVD) were investigated as potential biopassivation coatings for neural probes. It was found that organosilicon films from identical precursors differ in structure according to the method of deposition. For films produced from the cyclic siloxane precursor, hexamethylcyclotrisiloxane, pulsed plasma excitation reduced crosslink density over continuous excitation and produced flexible films resistant to prolonged saline soak testing. Deposition via a thermal process, HFCVD, allowed films of novel organosilicon structure to be formed from both hexamethylcyclotrisiloxane and its eight-membered analog, octamethylcyclotetrasiloxane. Characterization of these films was accomplished, and the effect of filament temperature on the chemical structure was elucidated. Silicon-silicon bonding and the retention of ring structures from the precursor was observed in HFCVD organosilicon films using Micro-Raman spectroscopy. A direct dielectric patterning process was proposed for semiconductor manufacturing. In this process, a dielectric material is patterned directly and developed without the need for a sacrificial photoresist layer. HFCVD fluorocarbon films are under consideration as low-dielectric constant interlayer dielectrics, and direct patterning of these materials was demonstrated using e-beam irradiation and supercritical CO2 development. The use of a gas-phase initiator species for HFCVD of fluorocarbon thin films was also demonstrated. Initiation enhanced deposition rates significantly and provided a means of selectively end-capping polymer chains present in the film structure. / by Hilton G. Pryce Lewis. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, February 2007. / Includes bibliographical references (p. 136-148). / The lipid bilayer forms the basic structure of the cell membrane, which is a heterogeneous matrix of proteins and lipids that provides a barrier between the interior of a cell and its outside environment. Protein and lipid domains in cell membranes can facilitate receptor localization, stabilize membranes, and influence membrane fluidity. In this thesis, we study how ordered protein and lipid domains influence the physical properties of lipid bilayers to better understand the roles of membrane domains in biological mechanisms. Model cellular membranes that mimic the behavior of biological membranes offer a controllable environment for systematically studying the isolated effects of protein and lipid ordering on membrane organization. Using fluid and solid-supported lipid bilayers, we study ordered peripheral membrane proteins and lateral lipid phase separation with fluorescence microscopy and X-ray reflectivity. To model cellular protein coatings and peripheral proteins, we prepare biotin-functionalized membranes that bind the proteins streptavidin and avidin. Fluorescence microscopy studies demonstrate that proteins crystallized in a single layer on lipid bilayer surfaces can change the lipid curvature and stabilize lipid vesicles against osmotic collapse. / (cont.) At solid interfaces, we characterize the electron density profiles of protein-coated bilayers to determine how a water layer separates an immobile protein layer from the fluid lipid bilayer. Liquid-ordered lipid phases enriched in cholesterol and sphingomyelin can localize molecules in cell membranes and this lipid phase separation behavior may be influenced by proteins and molecules in the membrane. Caveolae are specialized liquid-ordered domains in the plasma membrane that are enriched in the protein caveolin-1. We demonstrate that caveolin-1 peptides influence the onset of lipid phase separation and bind phase-separated lipid bilayers in solution. On solid surfaces, the formation of liquid-ordered lipid phases is influenced by surface roughness; with reflectivity, we determine that lipid bilayers containing cholesterol and sphingomyelin thicken with increasing cholesterol content. The membrane receptor GM1 also thickens the lipid bilayer when it is incorporated into the bilayer upper leaflet. The diverse experimental platforms that we present are applicable to studying additional and more complex biological systems to elucidate the influence of lipid and protein domains on cell membrane structure, organization and fluidity. / by Margaret R. Horton. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2009. / Vita. Cataloged from PDF version of thesis. / Includes bibliographical references. / Molecular recognition reagents are a critical component of targeted therapeutics, in vivo and in vitro diagnostics, and biotechnology applications such as purification, detection, and crystallization. Antibodies have served as the gold standard binding molecule because of their high affinity and specificity and, historically, because of their ability to be generated by immunization. However, antibodies suffer from several shortcomings that hinder their production and reduce their efficacy in a breadth of applications. The tenth type III domain of human fibronectin provides a small, stable, single-domain, cysteine-free protein scaffold upon which molecular recognition capability can be engineered. In the current work, we provide substantial improvements in each phase of protein engineering through directed evolution and develop a complete platform for engineering high affinity binders based on the fibronectin domain. Synthetic combinatorial library design is substantially enhanced through extension of diversity to include three peptide loops with inclusion of loop length diversity. The efficiency of sequence space search is improved by library focusing with tailored diversity for structural bias and binding capacity. Evolution of lead clones was substantially improved through development of recursive dual mutagenesis in which each fibronectin gene is subtly mutated or the binding loops are aggressively mutated and shuffled. This engineering platform enables robust generation of high affinity binders to a multitude of targets. Moreover, the development of this technology is directly applicable to other protein engineering campaigns and advances the scientific understanding of molecular recognition. Binders were engineered to tumor targets carcinoembryonic antigen, CD276, and epidermal growth factor receptor as well as biotechnology targets human serum albumin and goat, mouse, and rabbit immunoglobulin G. Binders have demonstrated utility in affinity purification, laboratory detection, and cellular labeling and delivery. Of particular interest, a panel of domains was engineered that bind multiple epitopes of epidermal growth factor receptor. Select non-competitive heterobivalent combinations of binders effectively downregulate receptor in a non-agonistic manner in multiple cell types. These agents inhibit proliferation and migration and provide a novel potential cancer therapy. / by Benjamin Joseph Hackel. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2011. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 317-322). / Engineering design work relies on the ability to predict system performance. A great deal of effort is spent producing models that incorporate knowledge of the underlying physics and chemistry in order to understand the relationship between system inputs and responses. Although models can provide great insight into the behavior of the system, actual design decisions cannot be made based on predictions alone. In order to make properly informed decisions, it is critical to understand uncertainty. Otherwise, there cannot be a quantitative assessment of which predictions are reliable and which inputs are most significant. To address this issue, a new design method is required that can quantify the complex sources of uncertainty that influence model predictions and the corresponding engineering decisions. Design of experiments is traditionally defined as a structured procedure to gather information. This thesis reframes design of experiments as a problem of quantifying and managing uncertainties. The process of designing experimental studies is treated as a statistical decision problem using Bayesian methods. This perspective follows from the realization that the primary role of engineering experiments is not only to gain knowledge but to gather the necessary information to make future design decisions. To do this, experiments must be designed to reduce the uncertainties relevant to the future decision. The necessary components are: a model of the system, a model of the observations taken from the system, and an understanding of the sources of uncertainty that impact the system. While the Bayesian approach has previously been attempted in various fields including Chemical Engineering the true benefit has been obscured by the use of linear system models, simplified descriptions of uncertainty, and the lack of emphasis on the decision theory framework. With the recent development of techniques for Bayesian statistics and uncertainty quantification, including Markov Chain Monte Carlo, Polynomial Chaos Expansions, and a prior sampling formulation for computing utility functions, such simplifications are no longer necessary. In this work, these methods have been integrated into the decision theory framework to allow the application of Bayesian Designs to more complex systems. The benefits of the Bayesian approach to design of experiments are demonstrated on three systems: an air mill classifier, a network of chemical reactions, and a process simulation based on unit operations. These case studies quantify the impact of rigorous modeling of uncertainty in terms of reduced number of experiments as compared to the currently used Classical Design methods. Fewer experiments translate to less time and resources spent, while reducing the important uncertainties relevant to decision makers. In an industrial setting, this represents real world benefits for large research projects in reducing development costs and time-to-market. Besides identifying the best experiments, the Bayesian approach also allows a prediction of the value of experimental data which is crucial in the decision making process. Finally, this work demonstrates the flexibility of the decision theory framework and the feasibility of Bayesian Design of Experiments for the complex process models commonly found in the field of Chemical Engineering. / by Kenneth T. Hu. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 135-151). / For a better understanding of how biology carries information within cells, it is not sufficient to look at individual protein or gene interactions, but to understand these networks of interactions as a whole. The goal of this thesis is to understand various aspects of how cells in general and T-cells in particular function, using models built from basic principles in chemical engineering, statistical physics and network theory, together with experiments performed by our collaborators. The ultimate objectives are to gain an insight into the mechanisms of certain key biological processes, understand the cause of certain diseases and to generate new ideas for methods of treating these diseases. First, we look at an example of a specific network built from previously published experiments and data collected by our collaborators, which governs the mechanism of activation of the T-cell receptor (TCR) by its kinase Lck and a negative regulator of Lck called Csk. We show that the mechanism by which the cell regulates TCR levels, together with the manner in which Lck activates the TCR produces interesting behavior, such as a "perfectly adaptive" system and a high-pass filter. Second, we look at heterogeneity in cancer cells at the level of protein signaling networks. Many common cancers are not treatable at the "source" or initial mutation, so one has to target downstream effector molecules. However, different cell lines bearing the same initial cancerous mutation exhibit varying signaling patterns due to differing secondary mutations which makes this difficult. The objective of this project is to try to characterize this heterogeneity and be able to identify molecules in the cell which would be the most effective drug targets. A general model for signaling in networks has been developed, analogous to models of neural networks, with mutations modeled as changes in the topology of this network. Keeping in mind that cancer cells are trying to maximize their growth, we are looking for patterns in secondary mutation during the directed evolution of these networks. A method for looking at free energy landscapes in topology space has also been developed. We find that lowest degree nodes along the shortest paths from the driver mutation to effector nodes tend to be the most conserved, and the frequency of multiple optima depends on the number of feedback loops. Finally, we look at the problem of constant activation thresholds for activation of various types of T-Cells. Despite having different TCRs, T-Cells of a certain type have a fixed activation threshold in terms of a peptide-MHC interaction strength (and a corresponding time, earlier than which they do not activate). We built a reaction-diffusion model for the network involved in the search process by which a pMHC-TCR finds a coreceptor-Lck, which enables us to understand how the threshold for activation is determined by the parameters of a particular cell type. We also developed an analytical solution for a simplified Markov Chain form of the model, which predicts how the activation rate scales with the parameters of interest in the system. We find that this rate is proportional to the fraction of coreceptors with Lck, increases (slowly) with diffusion and is independent of the number of coreceptors on the surface of the cell. / by Arvind Shankar Prabhakar. / Ph.D.