京都大学 / 新制・課程博士 / 博士(薬学) / 甲第24563号 / 薬博第861号 / 新制||薬||243(附属図書館) / 京都大学大学院薬学研究科薬学専攻 / (主査)教授 髙倉 喜信, 教授 山下 富義, 教授 小野 正博 / 学位規則第4条第1項該当 / Doctor of Pharmaceutical Sciences / Kyoto University / DFAM

京都大学 / 新制・課程博士 / 博士(薬学) / 甲第23845号 / 薬博第852号 / 新制||薬||242(附属図書館) / 京都大学大学院薬学研究科薬学専攻 / (主査)教授 髙倉 喜信, 教授 山下 富義, 教授 小野 正博 / 学位規則第4条第1項該当 / Doctor of Pharmaceutical Sciences / Kyoto University / DFAM

DNA serves as a mode of storage for genetic information and carries essential biophysical functions which continue to gain importance for new biomolecular applications. The functionalization of proteins with nucleic acids has found several different uses in a variety of fields such as DNA origami due to the highly programmable nature of DNA. The modulation of key biomolecular properties of proteins by the conjugation of DNA nanotools, mechanical forces and potential catalytic modulators is an under discovered area of protein engineering. In this doctoral thesis work, functionalization strategies of proteins with DNA oligonucleotides, the catalytic mechanism of AdhD from Pyrococcus furiosus and effects of mechanical forces on enzymatic catalysis have been investigated through the utilization of DNA tools. In the first part of the thesis (Chapter 2), several bioconjugation strategies which can be used to attach DNA oligonucleotides to proteins have been reviewed and assessed for emerging biomolecular applications. This work provides insights about commonly used bioconjugation reactions and investigates the performance of each conjugation reaction in terms of yield, site- specificity, flexibility in conjugation position, steric hinderance, cost, reagent availability and risk of altering native protein properties. In the next section of the thesis (Chapter 3), DNA spring attachment sites are strategically modeled and created on the model enzyme for assigned bioconjugation strategies and the DNA springs are assembled on this model enzyme. The effect of the forces generated by the DNA springs during bulk catalysis is investigated and the changes in binding pockets and substrate specificity properties are demonstrated. A novel magnetic bead-based purification strategy for the separation of DNA spring conjugated enzyme is established in this work to ensure homogenous catalysis conditions. This construct allows the tuning of catalytic properties by the usage of different lengths of DNA and is shown to be reversible. As enzymatic catalysis is investigated under bulk conditions and the amino acid sequence within the active site is not altered, this strategy provides a new platform for the modulation of enzymatic trajectories without isolation and altered microenvironment limitations as well as the irreversible effects of conventional techniques such as mutagenesis and directed evolution. In the last part of the thesis (Chapter 4), the effects of different DNA constructs on AdhD are investigated. Convex DNA springs are constructed on AdhD molecules with the goal of applying compressive forces over the active site. In addition, DNA tweezers are designed to apply comparable forces to the DNA spring studied in Chapter 3, assembled on AdhD and purified using the same magnetic bead-based strategy. This construct, as discussed in Chapter 1, allows the modulation of substrate specificity profiles consistent with its mechanical design. As a continuation of the last part of the thesis, the rare earth element (REE) binding capacity of AdhD and the effect of metal binding on its enzymatic trajectory are investigated in Chapter 5. The rare earth element (REE) binding affinity of AdhD and its effect on enzymatic catalysis on both the forward and reverse reactions are demonstrated with and without the presence of a metal chelator. This discovery sheds light on the potential allosteric regulation mechanisms of AdhD and the catalytic regulator effect of REEs. The work presented in this doctoral thesis demonstrates different biomolecular approaches towards the modulation of enzymatic properties through DNA oligonucleotide conjugation, application of mechanical forces and potential catalytic regulators. In the future, the results presented in this work can be utilized to initiate in depth studies about protein-DNA conjugation and modulation of enzymatic catalysis, and discover extended applications which can be used universally across different biomolecular platforms.

This thesis forms part of an ongoing project in the DNA Group to build and operate a synthetic ribosome. We present two synthetic ribosome designs that can be combined with DNA-templated chemistry to generate libraries of functional synthetic small molecules. In Chapter 2 we use the DNA strand displacement technique to construct a mechanism that is capable of moving along a DNA track. We explore ways to control the speed and the driving force of the mechanism, and present a mathematical model of the system. We discuss the ability of the design to incorporate chemically-functionalised DNA strands. In Chapter 3 we use a 2D DNA origami tile as the basis of the synthetic ribosome mechanism. Functionalised DNA strands are arranged on the surface of the tile, and we demonstrate the ability to template reactions between the strands, and discuss the possibility of creating a library of distinct chemical products from a single origami tile.

This thesis describes my investigations into the principles underlying self-assembly of DNA origami nanostructures and discusses how these principles may be applied. To study the origami folding process I designed, synthesized and characterized a polymorphic tile, which could adopt various shapes. The distribution of tile shapes provided new insights into assembly. The origami tiles I studied were based on scaffolds derived from customized plasmids, which I prepared using recombinant DNA technology. I developed a technique to monitor incorporation of individual staples in real time using fluorescence, measuring small differences in staple binding temperatures (~0.5-5 °C). I examined the tiles using Atomic Force Microscopy and I found that a remarkably high proportion of polymorphic tiles folded well, which suggests that there are assembly <b>pathways</b>, arising from strong cooperation between staples. In order to analyse the tile shapes quantitatively, I developed a specialized image processing technique. For validation of the method, I generated and analysed simulated data, and the results confirmed that I could measure individual tile parameters with sub-pixel resolution. I studied eleven variants of the polymorphic tile, and I proved that minor staple modifications can be used to change the folding pathway dramatically. The strength of cooperation between staples affects their behaviour, which is also influenced by their length and base sequences. Paired staples are particularly significant in assembly, and there are clear parallels with protein folding. I describe in an Appendix how I applied origami assembly principles in the development of my concept for an autonomous rotary nanomotor utilizing the sequential opening of DNA hairpins (already used for linear motors). This device represents an advance over non-autonomous rotary motors and I have simulated its performance. In this thesis I have answered important questions about DNA origami assembly, and my findings could enable the development of more sophisticated DNA nanostructures for specific purposes.

The uniquely predictable and controllable binding mechanism of DNA strands has been exploited to construct a vast range of synthetic nanodevices, capable of autonomous motion and computation. This thesis proposes novel ideas for the control and observation of such devices. The first of these proposals hinges on introducing mismatched base pairs into toehold-mediated strand displacement – a fundamental primitive in most dynamic DNA devices and reaction networks. Previous findings that such mismatches can impede strand displacement are extended insofar as it is shown that this impediment is highly dependent on mismatch position. This discovery is examined in detail, both experimentally and through simulations created with a coarse-grained model of DNA. It is shown that this effect allows for kinetic control of strand displacement decoupled from reaction thermodynamics. The second proposal improves upon a previously presented strand displacement scheme, in which two strands perform displacement cooperatively. This scheme is extended to be cascadable, so that the output of one such reaction serves as input to the next. This scheme is implemented in reaction networks capable of performing fundamental calculations on directed graphs. The third proposal is exclusively concerned with a novel observation methodology. This method is based on single-molecule fluorescence microscopy, and uses quantum dots, a fluorescent type of semiconductor nanocrystal, as a label. These quantum dots display a set of characteristics particularly promising for single-molecule studies on the time- and length scales most commonly encountered in DNA nanotechnology. This method is shown to allow for highly precise measurements on static DNA devices. Preliminary data for the observation of a complex dynamic device is also presented.

In the past decade, DNA origami self-assembly has been widely applied for creating customised nanostructures with base-pair precision. In this technique, the unique chemical addressability of DNA can be harnessed to create programmable architectures, using components ranging from dye or protein molecules to metallic nanoparticles. In this thesis, we apply DNA nanotechnology for developing novel light-harvesting and optical voltage sensing nano-devices. We use the programmable positioning of dye molecules on a DNA origami plate as a mimic of a light-harvesting antenna complex required for photosynthesis. Such a structure allows us to systematically analyse optimal design concepts using different dye arrangements. Complementary to this, we use the resistive-pulse sensing technique in a range of electrolytes to characterise the mechanical responses of DNA origami structures to the electric field applied. Based on this knowledge, we assemble voltage responsive DNA origami structures labelled with a FRET pair. These undergo controlled structural changes upon application of an electric field that can be detected through a change in FRET efficiency. Such a DNA-based device could ultimately be used as a sensitive voltage sensor for live-cell imaging of transmembrane potentials.

The profound simplicity and versatility of the molecule at the heart of all earth- bound life forms, DNA, continues to inspire new frontiers of scientific inquiry. Central to many of these, including the de novo design of novel DNA nanostructures and the use of DNA to probe the principles of biological self-assembly and the operation of cellular nanomachines, is the interaction of DNA with forces, both internal and external. This thesis comprises a survey of three key ways coarse-grained simulations using the oxDNA model can contribute to efforts to characterize these interactions. First, a non-equilibrium data analysis framework based on the Jarzynski equality from statistical physics is validated for use with oxDNA through the reconstruction of free energy landscapes for canonical DNA hairpin systems. We provide a framework for assessing errors in the method and apply it to study a system for which conventional equilibrium simulations would be impractical: DNA origami 'handles' proposed for use in force spectroscopy experiments. Next, we simulate the forcible unravelling of three DNA origami structures, the largest systems yet studied with simulated force spectroscopy. We combine these results with experimental AFM data to probe the mechanical response of origami in unprecedented detail, highlighting the effect of nanostructure design on unfolding behaviour. Lastly, we examine the validity of using widely-employed polymer elastic models to predict internal entropic forces in ssDNA. We develop a framework for measuring internal forces in the oxDNA coarse-grained model and apply it to analyze the pico-Newton range forces exerted by a recently proposed DNA origami force clamp, ultimately concluding that conventional means of estimating internal ssDNA forces are often inaccurate and should be supplemented with coarse-grained simulations. In addition to providing new insights about the DNA systems we present, our results highlight the significant fruits of complementing experimental studies with coarse-grained simulations.

DNA nanotechnology has revolutionised our capability to shape and control three-dimensional structures at sub-nanometre length scales. In this thesis, we use DNA to build synthetic membrane-inserting channels. Porphyrin and cholesterol tags serve as membrane anchors to facilitate insertion into the lipid membrane. With atomic force microscopy, confocal imaging and ionic current recordings we characterise our DNA nanochannels that mimic their natural protein-based counterparts in form and function. We find that they exhibit voltage-dependent conductance states. Amongst other architectures, we create the largest man-made pore in a lipid membrane to date approaching the electrical diameter of the nuclear pore complex. Pushing the boundaries on the other end of the spectrum, we demonstrate the ultimately smallest DNA membrane pore made from a single membrane-spanning DNA duplex. Thereby, we proof that ion conduction across lipid membranes does not always require a physical channel. With experiments and MD simulations we show that ions flow through a toroidal pore emerging at the DNA-lipid interface around the duplex. Our DNA pores spanning two orders of magnitude in conductance and molecular weight showcase the rational design of synthetic channels inspired by the diversity of nature - from ion channels to porins.

No description available.