Thesis (Ph. D.)--University of Washington, 2001. / Vita. Includes bibliographical references (leaves 202-216).

Studies of biomolecular interactions are of interest for several reasons. Beside basic research, the knowledge gained from such studies is also very valuable in for example drug target identification. Medical care is another area where biomolecules may be used as biomarkers to aid physicians in making correct diagnosis. In addition, the highly specific interactions between antibodies and almost any substance opens up the possibilities to design systems for detection of trace amounts of both biological and non-biological substances within environmental restoration, law enforcement, correctional care, customs service and national security. A biochip, which contains a biologically active material, offers a means of monitoring the molecular interactions in the above applications in a sensitive and specific manner. The biochip is a key component of a biosensor, which also includes components for transforming the interaction events into a human-readable signal. This thesis describes the use of poly(ethylene glycol) (PEG) in biochip design. Two different approaches are presented, the first based on ethylene glycol (EG)-containing alkyl thiol self-assembled monolayers (SAMs) on flat gold and the second on photo-induced graft copolymerisation of PEG-containing methacrylate monomers onto various substrates. The former is a two dimensional system where EG-terminated thiols are mixed with similar thiols presenting tail groups that mimic the explosive substance 2,4,6-trinitrotoluene (TNT). In an immunoassay, the detection limit for TNT was determined to fall in the range 1-10 µg/L. In the second approach, a branched three dimensional biosensor matrix (hydrogel) is proposed. The carboxymethylated (CM) dextran matrix, which is commonly used within the biosensing community, is not always ideal for studies of biointeractions, due to the non-specific binding frequently encountered in work with complex biological solutions and various proteins. To employ PEG, which displays a low non-specific binding of such species, is therefore an interesting option worth investigating. The use of a branched graft polymerised PEG matrix in biosensor applications is novel as compared to previous reports which have focused on linear PEG chains. The latter approach provides, at maximum, one functional group, per surface anchoring point, for immobilisation of sensor elements. Thus, it has the inherited disadvantage that it limits the number of available immobilisation sites. The present PEG matrix contains a large number of functional groups, for immobilisation of sensor elements, per grafting site and offers the potential of improved response upon binding to the analyte as demonstrated in a series of successful sensor experiments. Furthermore, the nature of the process enables easy preparation of matrix patterns and gradients. In a PEG matrix gradient, protein permeability is studied and the capabilities of immobilising proteins are demonstrated. By combining the patterning technique with different monomers in a two-step process, an inert platform, lacking chemical attachment sites, is provided with arrays of spots (with immobilisation capabilities), which are conveniently addressed via microdispensing and used for biosensor purposes. The EG-terminated thiols present another means of generating such inert platforms, a route which is also investigated. To further explore the sensor quality of these spots, the concepts of patterning and gradient formation are combined and studied. / Det är intressant att studera biomolekylära interaktioner av många anledningar. För att kunna bedriva framgångsrik läkemedelsutveckling är det oerhört viktigt att känna till hur olika molekyler samverkar i människokroppen. Inom sjukvården kan biomolekyler användas som biomarkörer, då närvaro av dem eller förändringar av deras koncentrationer är kopplade till sjukdomstillstånd, och därmed hjälper läkaren att ställa rätt diagnos. Dessutom kan de mycket specifika interaktionerna mellan antikroppar och (i princip) valfri substans användas för detektion av spårämnen vid miljösaneringsarbete, gränskontroller, polisarbete, fängelser och arbete med nationell säkerhet. Den här avhandlingen beskriver hur polymeren polyetylenglykol (PEG) kan användas vid design av biochip. Ett biochip är en liten anordning, som kan användas för att detektera specifika molekyler med hjälp av en biologisk interaktion. Traditionellt har PEG använts inom biomaterialsektorn, men återfinns även i hygienartiklar som tvål och tandkräm. Ett annat användningsområde är konservering av bärgade träskepp och i en del litiumjonbatterier ingår PEG som en komponent. Dessutom pågår utveckling av PEG-innehållande skyddsvästar. I det här arbetet används PEG framför allt på grund av sin förmåga att minimera ospecifik inbindning av proteiner, som utgör en stor del av gruppen biomolekyler, till ytor på biochip. Två olika typer av ytbeläggningar, som innehåller den här polymeren, har använts. Den första typen ger mycket tunna (~0.000003 mm), tvådimensionella filmer medan den andra ger en något tjockare (~0.00005 mm), tredimensionell struktur (matris). De tvådimensionella filmerna har använts för att utveckla en sprängämnesdetektor med mycket hög känslighet (detektionsgräns mellan 1-10 ppb). En viktig beståndsdel i detta system är antikroppar riktade mot sprängämnet trinitrotoluen (TNT). Den tredimensionella matrisen är mer generell och kan användas för att studera många olika molekylära interaktioner. Tillverkningsmetoden av matrisen är baserad på belysning med ultraviolett ljus och är därmed lämpad för att skapa mönstrade ytor. Genom att blockera delar av ljusflödet begränsas tillväxten av matrisen till de belysta delarna. På så sätt har bland annat så kallade mikro-arrayer, bestående av mikrometerstora (tusendels millimeter) strukturer i ett regelbundet mönster, tillverkats. Tekniken tillåter även tillverkning av gradienter, där matrisens tjocklek varierar längs med provet, genom att belysa olika delar av provytan olika länge. Genom att undersöka dessa gradienter har information om matrisens genomsläpplighet för proteiner kunnat extraheras. Gradientkonceptet har även kombinerats med mikro-arraytillverkningen och gett möjlighet att studera interaktioner mellan flera olika modellproteiner och deras motsvarande antikroppar i olika tjocka matriser på en och samma yta. Det finns ett stort antal sätt att utnyttja interaktionerna mellan olika molekyler på ett biochip. Ett tilltalande tillvägagångssätt är exempelvis att i en mikro-array binda in olika molekyler som kan fånga kliniskt intressanta biomolekyler, i syfte att skapa en hälsoprofil. Ett sådant biochip skulle ge möjlighet att parallellt detektera eller bestämma koncentrationen av ett stort antal biomolekyler i till exempel en droppe blod. På så sätt kan en diagnos snabbt ställas, kanske till och med utan att patienten behöver uppsöka sjukvården. Den utvecklade PEG-matrisen har god potential att fungera i en sådan applikation.

Mesoporous silica materials are versatile platforms for biological catalysis, isolation of small molecules for detection and separation applications. The design of mesoporous silica supports for tailored protein and biomolecule interactions has been limited by the techniques to demonstrate biomolecule location and functionality as a function of pore size. This work examines the interaction of proteins and lipid bilayers with engineered porous silica surfaces using spherical silica particles with tunable pore diameters (3 – 12 nm) in the range relevant to biomolecule uptake in the pores, and large particle sizes (5 - 15 µm) amenable to microscopy imaging The differentiation of protein location between the external surface and within the pore, important to applications requiring protein protection or catalytic activity in pores, is demonstrated. A protease / fluorescent protein system is used to investigate protein location and protection as a function of pore size, indicating a narrow pore size range capable of protein protection, slightly larger than the protein of interest and approaching the protease dimensions. Selective functionalization, in this case exterior-only surface functionalization of mesoporous particles with amines, is extended to larger pore silica materials. A reaction time dependent functionalization approach is demonstrated as the first visually confirmed, selective amine functionalization method in protein accessible supports. Mesoporous silica nanoparticles are effective supports for lipid bilayer membranes and membrane associated proteins for separations and therapeutic delivery, although the role of support porosity on membrane fluidity is unknown. Transport properties of bilayers in lipid filled nanoparticles as a function of pore diameter and location in the particle are measured for the first time. Bilayer diffusivity increases with increasing pore size and is independent of bilayer location within the core, mid or cap of the particle, suggesting uniform long range bilayer mobility in lipid filled pores. Application of lipid bilayers on mesoporous silica was examined for membrane associated proteins A unique method to adhere functional proteins in lipid bilayers on mesoporous silica particles is established using vesicles derived from cell plasma membranes and their associated proteins. This method of membrane protein investigation retains proteins within native lipid membranes, stabilizing proteins for investigation on supports.

The cost of deconstructing cellulose into soluble sugars is a key impediment to the commercial production of lignocellulosic biofuels. The use of the quartz crystal microbalance (QCM) to investigate reaction variables critical to enzymatic cellulose hydrolysis is investigated here, extending previous studies of fungal cellulase activity for the first time to whole cell cellulases. Specifically, the activity of the cellulases of Clostridium thermocellum, which are in the form of cellulosomes, was investigated. To clearly differentiate the activity of free cellulosome and cell-bound cellulosome, the distribution of free cellulosome and cell-bound cellulosome in crude cell broth at different growth stages of C. thermocellum (ATCC 27405) was quantified. Throughout growth, greater than 70% of the cellulosome in the crude cell broth was unattached to the cell. The frequency response of the QCM was shown to capture adsorption and hydrolysis of amorphous cellulose films by the whole-cell cellulases. Further, both crude cell broth and free cellulosomes were found to have similar inhibition pattern (within 0 - 10 g/L cellobiose). Thus, kinetic models developed for the cell-free cellulosomes, which allow for more accurate interfacial adsorption analysis by QCM than their cell-attached counterparts, may provide insight into hydrolysis events in both systems.

This thesis describes the development of sensitive and high-resolution fluorescence spectroscopic and microscopic techniques and their application to probe biomolecules and their interactions in solution, lipid membrane model systems and in cells. Paper I-IV are largely focused on methodological developments. In paper I, a new fluorescence method based on fluorescence correlation spectroscopy (FCS) for detecting single particles was realized, requiring no fluorescent labeling of the particles. The method can yield information both about the diffusion properties of the particles as well as about their volumes. In paper II, a modified fluorescence cross correlation spectroscopy procedure with well characterized instrumental calibration was developed and applied to study cis interactions between an inhibitory receptor and its Major Histocompatibility Complex class I ligand molecule, both within the same cellular membranes. The quantitative analysis brought new insights into the Nature killer cell’s self-regulating of tolerance and aggressiveness for immune responses. Paper III describes a multi-color STED (STimulated Emission Depletion) microscopy procedure, capable of imaging four different targets in the same cells at 40nm optical resolution, which was developed and successfully demonstrated on platelets. In paper IV, a modified co-localization algorithm for fluorescence images analysis was proposed, which is essentially insensitive to resolutions and molecule densities. Further, the performance of this algorithm and of using STED microscopy for co-localization analysis was evaluated using both simulated and experimentally acquired images. Papers V-VII have their main emphasis on the application side. In paper V, transient state imaging was demonstrated on live cells to image intracellular oxygen concentration and successfully differentiated different breast cancer cell lines and the different metabolic pathways they adopted to under different culturing conditions. Paper VI describes a FCS-based study of proton exchange at biological membranes, the size-dependence of the membrane proton collecting antenna effect as well as effects of external buffer solutions on the proton exchange, in a nanodisc lipid membrane model system. These findings provide insights for understanding proton transport at and across membranes of live cells, which has a central biological relevance. In paper VII, STED imaging and co-localization analysis was applied to analyze cell adhesion related protein interactions, which are believed to have an important modulating role for the proliferation, differentiation, survival and motility of the cells. The outcome of efforts taken to develop means for early cancer diagnosis are also presented. It is based on single cells extracted by fine needle aspiration and the use of multi-parameter fluorescence detection and STED imaging to detect protein interactions in the clinical samples. Taken together, detailed studies at a molecular level are critical to understand complex systems such as living organisms. It is the hope that the methodologies developed and applied in this thesis can contribute not only to the development of fundamental science, but also that they can be of benefit to mankind in the field of biomedicine, especially with an ultimate goal of developing novel techniques for cancer diagnosis. / <p>QC 20140609</p>

With advances in high-throughput measurement techniques, large-scale biological data, such as protein-protein interaction (PPI) data, gene expression data, gene-disease association data, cellular pathway data, and so on, have been and will continue to be produced. Those data contain insightful information for understanding the mechanisms of biological systems and have been proved useful for developing new methods in disease diagnosis, disease treatment and drug design. This study focuses on two main research topics: (1) identifying protein complexes and (2) identifying disease genes from biomolecular networks. Firstly, protein complexes are groups of proteins that interact with each other at the same time and place within living cells. They are molecular entities that carry out cellular processes. The identification of protein complexes plays a primary role for understanding the organization of proteins and the mechanisms of biological systems. Many previous algorithms are designed based on the assumption that protein complexes are densely connected sub-graphs in PPI networks. In this research, a dense sub-graph detection algorithm is first developed following this assumption by using clique seeds and graph entropy. Although the proposed algorithm generates a large number of reasonable predictions and its f-score is better than many previous algorithms, it still cannot identify many known protein complexes. After that, we analyze characteristics of known yeast protein complexes and find that not all of the complexes exhibit dense structures in PPI networks. Many of them have a star-like structure, which is a very special case of the core-attachment structure and it cannot be identified by many previous core-attachment-structure-based algorithms. To increase the prediction accuracy of protein complex identification, a multiple-topological-structure-based algorithm is proposed to identify protein complexes from PPI networks. Four single-topological-structure-based algorithms are first employed to detect raw predictions with clique, dense, core-attachment and star-like structures, respectively. A merging and trimming step is then adopted to generate final predictions based on topological information or GO annotations of predictions. A comprehensive review about the identification of protein complexes from static PPI networks to dynamic PPI networks is also given in this study. Secondly, genetic diseases often involve the dysfunction of multiple genes. Various types of evidence have shown that similar disease genes tend to lie close to one another in various biomolecular networks. The identification of disease genes via multiple data integration is indispensable towards the understanding of the genetic mechanisms of many genetic diseases. However, the number of known disease genes related to similar genetic diseases is often small. It is not easy to capture the intricate gene-disease associations from such a small number of known samples. Moreover, different kinds of biological data are heterogeneous and no widely acceptable criterion is available to standardize them to the same scale. In this study, a flexible and reliable multiple data integration algorithm is first proposed to identify disease genes based on the theory of Markov random fields (MRF) and the method of Bayesian analysis. A novel global-characteristic-based parameter estimation method and an improved Gibbs sampling strategy are introduced, such that the proposed algorithm has the capability to tune parameters of different data sources automatically. However, the Markovianity characteristic of the proposed algorithm means it only considers information of direct neighbors to formulate the relationship among genes, ignoring the contribution of indirect neighbors in biomolecular networks. To overcome this drawback, a kernel-based MRF algorithm is further proposed to take advantage of the global characteristics of biological data via graph kernels. The kernel-based MRF algorithm generates predictions better than many previous disease gene identification algorithms in terms of the area under the receiver operating characteristic curve (AUC score). However, it is very time-consuming, since the Gibbs sampling process of the algorithm has to maintain a long Markov chain for every single gene. Finally, to reduce the computational time of the MRF-based algorithm, a fast and high performance logistic-regression-based algorithm is developed for identifying disease genes from biomolecular networks. Numerical experiments show that the proposed algorithm outperforms many existing methods in terms of the AUC score and running time. To summarize, this study has developed several computational algorithms for identifying protein complexes and disease genes from biomolecular networks, respectively. These proposed algorithms are better than many other existing algorithms in the literature.

Thesis (Ph.D. in Molecular Biology) -- University of Colorado Denver, 2008. / Typescript. Includes bibliographical references (leaves 155-167). Free to UCD affiliates. Online version available via ProQuest Digital Dissertations;

As interações moleculares, em especial as de caráter não-covalente, são processos-chave em vários aspectos da biologia celular e molecular, desde a comunicação entre as células ou da velocidade e especificidade das reações enzimáticas. Portanto, há a necessidade de estudar e criar métodos preditivos para calcular a afinidade entre moléculas nos processos de interação, os quais encontram uma gama de aplicações, incluindo a descoberta de novos fármacos. No geral, entre esses valores de afinidade, o mais importante é a energia livre de ligação, que normalmente é determinada por modos computacionalmente rápidos, porém sem uma forte base teórica, ou por cálculos muito complexos, utilizando dinâmica molecular, onde mesmo com um grande poder de determinação da afinidade, é muito custoso computacionalmente. O objetivo deste trabalho é avaliar um modelo menos custoso computacionalmente e que promova um aprofundamento na avaliação de resultados obtidos a partir de simulações de docking molecular. Para esta finalidade, o método de Monte Carlo é empregado para a amostragem de orientações e conformações do ligante do sítio ativo macromolecular. A avaliação desta metodologia demonstrou que é possível calcular grandezas entrópicas e entálpicas e analisar a capacidade interativa entre complexos proteína-ligante de forma satisfatória para o complexo lisozima do bacteriófago T4. / The molecular interactions, especially the ones with a non-covalent nature, are key processes in general aspects of cellular and molecular biology, including cellular communication and velocity and specificity of enzymatic reactions. So, there is a strong need for studies and development of methods for the calculation of the affinity on interaction processes, since these have a wide range of applications like rational drug design. The free energy of binding is the most important measure among the affinity measurements. It can be calculated by quick computational means, but lacking on strong theoretical basis or by complex calculations using molecular dynamics, where one can compute accurate results but at the price of an increased computer power. The aim of this project is to evaluate a computationally inexpensive model which can improve the results from molecular docking simulations. For this end, the Monte Carlo method is implemented to sample different ligand configurations inside the macromolecular binding site. The evaluation of this methodology showed that is possible to calculate entropy and enthalpy, along analyzing the interactive capacity between receptor-ligands complexes in a satisfactory way for the bacteriophage T4.

Microfluidics deals with the behavior, control and manipulation of fluids which are confined at micrometer length scale. It has important application in lab-on-a chip technology, micro-propulsion, additive manufacturing, and micro-thermal technologies. Microfluidics has been widely used in detection, separation, transportation, and mixing of fluids and particles. The work carried out for the thesis to study the fluid-structure interaction in micro-channel involves an experimental part and a simulation part. In the experimental part the characterization of biofluid (RBC in BSA) is carried out based on the power law of fluid and flow behavior is studied. Also the dependence of fluid concentration on the viscosity in the channel is studied. The results are analyzed. Transition of fluid behavior from non-Newtonian shear thickening to non-Newtonian shear thinning is observed when the RBC concentration varies from 5.5×106 to 5.5×107 cells/ml in the channel. From the viscosity measurements of the biofluid it is observed that the average viscosity in the channel increases on increasing concentration of the fluid for shear thickening fluids. In the simulation part, interaction behavior of biomolecule DNA is studied in the channel containing biofluid which is characterized in the experimental part. Cell free DNAs are common problem in infectious disease detection. Based on the assumptions of the WLC model, DNA strand is assumed as a one dimensional elastic member with its one end fixed at the channel wall and the other end free to move in the fluid. Bent and straight DNAs are considered for the study. Multiple scales are involved in this problem which is not fully understood. DNA strands in the channel are exposed to different forces in the channel which are mainly due to the pressure and viscous effects. Numerical simulations are carried out for the multiphysics problem of DNA in the fluid using a coupled multiphysics finite element scheme and the results are obtained. Same procedure is carried out considering smaller channels and also for PBS solution as background fluid to obtain consistent results. It is found that when the channel width increases the tip displacement of DNA decreases. It was observed that DNA tip displacement is more in the channel when its end-to-end length is approximately half the width of the channel. Potential application of these modeling and simulation are in molecular screening processes to improve the performance of microfluidic DNA chips, and in design of micro-channel structures of microfluidic devices.

Recent pharmaceutical developments have investigated using supramolecular nanoparticles in order to increase the bioavailability and solubility of drugs delivered in various methods. Modification of the carbohydrate cyclodextrin increases the ability to encapsulate hydrophobic pharmaceutical molecules by forming a carrier with a hydrophobic core and hydrophilic exterior. Guest molecules are commonly added to these inclusion complexes in order to add stability and further increase targeting abilities of the carriers. One such guest molecule is adamantine combined with a poly(ethylene glycol) chain. Vesicles are formed by hydrating a thin film of amphiphilic cyclodextrin and guest molecules in buffer solution that mimics physiological conditions. The solution is subject to freeze-thaw cycles and extrusion, and the complexes are separated out via size exclusion chromatography. Dynamic Light Scattering instrumentation is used to observe the particle size distribution. Cargo release can be observed in fluorescent dye-loaded vesicles by addition of a membrane-cleaving agent under a fluorimeter instrument. Future work involving this drug delivery technology includes synthesizing a chemically sensitive guest that will cleave in the presence of an intra-cellular anti-oxidant, and finally observing the uptake of these vesicles into live cells and testing the delivery of cargo in vitro under physiological conditions.