12023 research outputs found
Sort by
Vibrational Imaging for Chemical Biology: from Label-Free to Molecular probes
No full textSince the invention of stimulated Raman scattering (SRS) microscopy in 2008, vibrational imaging is increasingly recognized as a powerful tool for biological investigation. As the most suitable far field vibrational imaging modality for live biological studies, SRS microscopy is taking the lead role within its vibrational counterparts with desired sensitivity and image quality. The totally different mechanism of generating vibration signals from fluorescence signals determines the special features of vibrational imaging. Bond vibration originating signals provide inherent optical contrast for every molecule and the quantitative manner allows straightforward quantification. Since the inception, SRS microscopy has achieved large success in label-free imaging. Label-free imaging avoids tedious labeling step and has the least perturbation to the biological samples but with limited sensitivity and specificity. The introducing of labeling starting about 10 years ago opens up a new avenue for SRS microscopy to tackle the fundamental limitations of label-free approaches. Whether to use label-free or molecular probes for SRS microscopy depends on the specific studies. This thesis aims to utilize SRS microscopy (both label-free and minimally labeling) for metabolic study and develop new molecular probes for SRS microscopy.
We start from comparing different vibrational imaging modality and fluorescence imaging and conclude that SRS is the best vibrational imaging technique for biological samples. Then we discuss the features of label-free, bioorthogonal labeling and super-multiplexed SRS imaging. The minimally perturbative triple bond tagging and isotope labeling makes SRS especially suitable for tracking metabolites and accessing metabolic pathways. Furthermore, we also summarize the design principles for functional Raman imaging probe development based on their spectroscopic signatures. (Chapter 1).
Non-invasively probing metabolites within single live cells is highly desired but challenging. We explored Raman spectro-microscopy towards spatially-resolved single cell metabolomics, with the specific goal of identifying druggable metabolic susceptibilities from a series of patient-derived melanoma cell lines. The chemical composition analysis of single cell and single organelle lipid droplets identified the fatty acid synthesis pathway and lipid mono-unsaturation as druggable susceptibility. More importantly we revealed that inhibiting lipid mono-unsaturation leads to cellular apoptosis accompanied by the formation of phase-separated intracellular membrane domains. (Chapter 2).
Next, we established a first-in-class design of multi-color photoactivatable Raman probes for subcellular imaging and tracking. The fast photochemically generated alkynes from cyclopropenones enable background-free Raman imaging with desired photocontrollable features. After necessary molecule engineering to improve the biocompatibility and sensitivity, we generated organelle-specific probes for targeting mitochondria, lipid droplets, endoplasmic reticulum, and lysosomes. Multiplexed photoactivated imaging and tracking at both subcellular and single-cell levels was also demonstrated to monitor the dynamic migration and interactions of the cellular contents. (Chapter 3).
Further improvement of the Raman signal with molecular probes is a central topic for Raman imaging. Recently developed electronic preresonance (epr) probes boost Raman signals and pushed SRS sensitivity close to that offered by confocal fluorescence microscopy. To guide the development of even stronger Raman probes and fill the final gap between epr-SRS probes and single molecule imaging, the structure-function relationship of epr-SRS probes is indispensable. We therefore used ab initio approach employing the displaced harmonic oscillator (DHO) model for calculating the epr-SRS signals, which proves to provide a consistent agreement between simulated and experimental SRS intensities of various triple-bond bearing epr-SRS probes. The theory also allows us to illustrate how the observed intensity differences between molecular scaffolds stem from the coupling strength between the electronic excitation and the targeted vibrational mode. Utilizing the discovered structure-function relationship of epr-SRS probes, we engineered MARS palette for higher sensitivity. With chemical modification to improve Raman mode displacement or enhance transition dipole moment or adjust detuning, we enhance the signal of alkynyl pyronins and nitrile pyronins, setting the current sensitivity records for small molecule far-field Raman probes. (Chapter 4 and 5).</p
Development of Microcrystal Electron Diffraction Techniques for the Characterization of Small Molecules and Novel Materials
Get PDFTraditional techniques for structural analysis, such as X-ray crystallography and Nuclear Magnetic Resonance (NMR), have been invaluable in understanding the composition of various substances. However, these methods often encounter challenges when applied to the analysis of small molecules and certain novel materials, particularly those that cannot form large, high-quality crystals. The research presented here focuses on the evolution and applications of Microcrystal Electron Diffraction (MicroED), a transformative technique that has expanded the boundaries of structural analysis. We trace the developmental trajectory of MicroED, exploring its underlying principles, technological advancements, and comparative advantages over conventional methods. A variety of data from several key studies was collected through a series of experiments utilizing MicroED to analyze a range of substances, from small organic molecules to complex novel materials and innovative inorganic complexes. MicroED offers unprecedented resolution and sensitivity, capable of structural elucidation where other methods fail. In particular, MicroED has been successful in determining the structures of several novel materials and small molecules with applications in areas such as renewable energy, advanced manufacturing, and pharmaceuticals. Furthermore, this technique is highly amenable to integration with other analytical and computational methods, including machine learning algorithms for data interpretation, enhancing its applicability and efficiency. This research contends that MicroED is not merely an alternative but a substantial upgrade to existing methodologies, holding the potential to revolutionize fields as diverse as materials science, chemistry, and medicine
Non-Equilibrium Quantum Dynamics in a Disordered Ising Magnet
Get PDFThe quantum two-level system, or “qubit,” is a simple platform that nonetheless displays fundamentally non-trivial quantum behavior. The rare-earth magnet LiHoF₄ is a natural physical representation of a system of coupled qubits. With its uncommonly high crystal anisotropy, LiHoF₄ can be mapped to the problem of the Ising model in a transverse field. However, while this Ising approximation can quantitatively predict much of the equilibrium behavior, quantum corrections, originating from off-diagonal terms in the dipolar interaction that generate quantum fluctuations, are crucial in driving non-equilibrium dynamics when subject to an external drive. Furthermore, quenched disorder can be introduced through chemical substitution, which, through the dipolar interaction, generates spatially random pinning fields, as well as internal transverse fields, which drive quantum fluctuations. Noise measurements on the disordered ferromagnet LiHo0.65Y0.35F4 show critical behavior, whose statistics are driven from the underlying pinning distribution, while measurements on LiHo0.40Y0.60F4 display non-critical behavior that can only be attributed to quantum co-tunneling processes. This is the first demonstration of crackling noise in a ferromagnet in the purely quantum regime. Furthermore, pump-probe susceptibility measurements on the decoupled cluster glass show the system being driven out of equilibrium with astonishingly weak drives, due to resonant transitions arising from off-diagonal dipolar terms σiz σjx. Non-linear sample response is observable in inelastic Raman scattering measurements, and these spin clusters also exhibit asymmetric Fano resonances with high Q-factors of ~10⁵. Quantum interference effects can be tuned to fully decouple one of the dressed states from the others, rendering the sample transparent to the drive. This is analogous to optical systems that display electromagnetically-induced transparency, but at 100 Hz frequencies
Acquiring Enzyme Sequence-Fitness Data at Scale Toward Predictive Methods for Enzyme Engineering
Get PDFThe emergence of machine learning methods for expediting directed evolution via protein fitness prediction has recently shed light on the need for more, high quality sequence-fitness data from which to learn the mapping from sequence to fitness. Enzymes specifically are highly selective catalysts and engineered enzymes are becoming increasingly important for human applications such as pharmaceutical synthesis. This thesis thus focuses on the collection of enzymatic sequence-fitness data to enable both development and validation of emerging approaches. Chapter 1 describes the process of traditional directed evolution as well as ways that machine learning methods have been used to accelerate it. It also discusses the experimental considerations for applying machine learning to the various steps of protein engineering campaigns, as the experimental constraints are not always obvious to the machine learning community. One of the major constraints for the application of machine learning methods is the requirement to sequence all variants required for model training, a step that is often skipped by traditional, lab-only directed evolution due to it not being worth the time and cost. Chapter 2 introduces a solution to this problem with “every variant sequencing” (evSeq), which enables higher throughput collection of sequencing data for a similar time and cost as commonly used Sanger sequencing methods. This method not only enables implementation of ML methods such as machine learning-assisted directed evolution (MLDE) and focused training MLDE (ftMLDE) by sequencing variants during an evolution campaign, but also offers promise to fill existing protein sequence-fitness databases with protein engineering datasets. This type of data collection can enable the development of newer, more accurate ML methods, and was an inspiration for the work presented in Chapter 3, which details the collection of a combinatorially complete, epistatic sequence-fitness landscape in an enzyme active site. Oftentimes, the effects of mutations on protein fitness can be considered largely independent and laboratory recombination of them can find an optimal variant. This general principle breaks down when the effects of mutations are not independent, termed epistasis, and sequence-fitness landscapes with these interactions are difficult to traverse. Thus, collection of this dataset provides a challenging task for the development of both ML and physics-based models and pushes the boundary of predictive methods for protein engineering
Organic Films at the Electrode-Electrolyte Interface in CO₂ Reduction
Get PDFThis thesis focuses on the use of use high-throughput experimentation and analytical electrochemistry techniques to understand how organic films on (photo)electrode surfaces alter catalyst selectivity. Specifically, the objective has been to deconvolute effects associated with the organic film from the atomic identity of the catalyst, reactant and intermediate concentration polarization effects, and temperature in the context of electrochemical CO₂ reduction. The first chapter provides the motivations behind the transformation of CO₂ into value-added materials using electricity and the challenges that the field faces. The second chapter introduces the data-driven identification of a scaling relationship between the partial current densities of methane and C₂₊ products among 14 bulk copper bimetallic alloys. This strict dependence represents an intrinsic limitation of selectivity tuning through alloying. However, it can be disrupted to favor C₂₊ products by the presence of an organic additive, highlighting the potential of hybrid organic–inorganic catalysts to tune branching ratios in the CO₂R reaction network. The third chapter highlights that with the wide band gap CuGa₃Se₅ chalcopyrite absorber, organic coatings can not only provide dramatic increases in selectivity toward CO₂R products compared to the unmodified system, but also and significantly moderate catalyst corrosion. The fourth chapter unveils a new class of molecular films on polycrystalline copper, derived from aryl diazonium and iodonium salts, that are corrosion resistant even at pH 1 and have the potential for many future electrochemical applications. In the fifth chapter, we demonstrate that increased mass transport at the electrode surface directly resulted in changes to the ethylene and methane Tafel slope values on copper electrodes. These findings emphasize that the apparent Tafel slope reported for any copper system is not necessarily representative of the catalyst’s intrinsic kinetics alone, but also contains information about the cell geometry and electrolyte convective transport. The final chapter investigates the combined effect of organic films, mass transport, and electrode heating on electrocatalysis. We find that we can use surface heating to replace bulk heating, but that the complexity of CO₂R prevents predictable behavior. However, the addition of additive films to the electrode surface enables idealized electrochemical CO₂ reduction kinetics, and therefore the calculation of important parameters such as the activation energy for C₂₊ product formation
Fluid-Rock Interactions from the Lithosphere to Earth’s Surface
Get PDFFluids can cycle and migrate through planetary bodies, transporting soluble ions and influencing physical properties of the surrounding rock or magma, such as fracture toughness, seismic wave velocity, melting point, viscosity, and more. Precipitated minerals, fluids trapped in inclusions, and free pore fluids can be used to constrain fluid provenance, mixing relationships, and paleoenvironmental information such as temperature, pressure, redox conditions, salinity, and pH. In my thesis, I discuss my research on topics pertaining to the geochemistry associated with fluid-rock interactions that occur from the depths of the lithospheric mantle to Earth’s surface. Broadly, these chapters address open questions pertaining to 1) the retention timescales and metasomatic overprinting of fluids sourced from the mantle in obducted peridotites, 2) the capacity for pedogenic Mg-carbonates to preserve palaeohydrological information with implications for Martian carbonates, and 3) the influences hydrous fluids have on lithospheric magmas and minerals.
Helium isotopes are arguably the best tracer for fluid sources in Earth materials at the planetary scale. ³He/⁴He ratios of the Earth’s 1) continental crust, 2) atmosphere, 3) upper mantle, and 4) core or deep isolated mantle (mantle plume source) vary by over two orders of magnitude, offering considerable dynamic range compared to measurement precision. While helium isotope signatures in Earth’s mantle have been determined almost exclusively by the analysis of helium retained in mantle xenoliths, phenocrysts, erupted glasses, and vent gases, this selection introduces a sampling bias towards fluids that have been transported to Earth’s surface by eruptive processes. In contrast, residual mantle peridotites take much longer to arrive at Earth’s surface and are therefore more susceptible to metasomatic processes that can overprint primary helium isotopic signatures. In Chapter 1, I use concentrations and isotopes of helium and argon along with concentrations of U and Th to place constraints on the sources and siting of helium retained in exhumed mantle peridotites collected from Twin Sisters Mountain of the Northern Cascades in Washington State, USA. Helium isotope ratios of peridotites from the Twin Sisters Mountain span from 0.8 to 6 times the atmospheric ratio (1RA=1.4*10⁻⁶ ³He/⁴He). Fluid inclusions in these peridotites capture a two-component mixture that included a mantle-like endmember (~6 RA) and a serpentinizing endmember (1.0 ± 0.5 RA) that is consistent with a mixture of surface-derived groundwater, leached crustal radiogenic helium and reworked mantle helium. While these components are not effectively isolated by extraction using vacuum crushing and powder fusion, step-heating analysis reveals that the serpentinizing endmember is released at lower temperatures (<1000°C) and the mantle-like endmember is released at higher temperatures. Results demonstrate that helium signatures can be retained in lithospheric peridotites against both diffusive loss and radiogenic ingrowth over at least 10⁸-year timescales but can be greatly modified by cryptic metasomatic processes during emplacement.
Mg-carbonates have become increasingly relevant in the scientific community due to their orbital and in situ detection on the Martian surface. Like Ca-carbonate on Earth, Martian Mg-carbonates may preserve paleoenvironmental information associated with their formation on Mars billions of years ago, shedding light on habitability. Yet, unlike Ca-carbonates, the capacity for Mg-carbonates to preserve paleoenvironmental information through trace element signatures associated with their source fluids has not been well established for surficial magnesite samples on Earth. In Chapter 2, I 1) develop a digestion protocol to selectively digest Mg-carbonates (magnesite ± dolomite) while obviating influences of contaminant phases and ions adsorbed to mineral surfaces, 2) validate a method to analyze trace elements with Mg-matrix by solution ICP-MS, and 3) apply these procedures to determine trace element concentrations of pedogenic Mg-carbonates sampled along a depth profile in the Kunwarara open pit magnesite mine in Queensland, Australia. Results from this study confirm that the method we implemented selectively digests magnesite ± dolomite. A relationship between negative Ce anomaly in the carbonates and Fe/Mn-oxides/hydroxides in corresponding host sediment collected along the depth profile demonstrates that pedogenic magnesites can capture redox gradients in the soil column. This finding implies that Ce anomaly in carbonates can potentially be used to place constraints on the paleo-redox conditions associated with Mg-carbonate formation on ancient Mars.
Numerous questions in Earth science depend on quantitative understanding of how elements fractionate during melting and crystallization. To name a few: assessment of how lithospheric fluids influence geodynamical processes, constraining mechanisms that led to the formation of the Earth’s continental crust, evaluation of elemental fluxes from the mantle to Earth's surface, calibration of a reliable crustal barometer, and gauging how magmatism and plate tectonics differed with the higher geothermal gradients of a younger Earth. MELTS thermodynamic software is a widely available free tool utilized by geoscientists to both test hypotheses and model the geochemistry of magmatic processes. However, minerals of the amphibole supergroup, although common in magmatic systems, rarely crystallize in MELTS simulations, even when well controlled experiments demonstrate that they should. The decrease in the Gibbs energy needed to stabilize amphibole in MELTS is often on the order of the configurational entropy contribution to the Gibbs energy associated with minor elements that are not present in any of the current amphibole solution models used in MELTS but are frequently incorporated in the amphibole crystal lattice. In Chapter 3, I outline a framework for a volume model for monoclinic amphiboles that can be used in an expanded amphibole solution model to be incorporated in MELTS software. A volume model is prerequisite to calibrating the other model terms because it accounts for differences in pressure among experimental constraints. The framework I develop extends the model to include minor components that are not present in existing versions of the MELTS amphibole models. I calibrate a preliminary model using a dataset composed of x-ray refinements that supply amphibole volume and site occupancy data. Results reveal regions in parameter space where data is limited and the sensitivity that model coefficients have to uncertainties in the data, suggesting that filtering the dataset to remove outliers may be necessary.</p
Modeling Frameworks for Modular and Scalable Biological Circuit Design
Get PDFSynthetic biology is a rapidly evolving interdisciplinary field that combines principles from biology, bioengineering, biochemistry, and computational sciences to design and engineer new biological systems for various applications. This thesis focuses on addressing the challenges in engineering large and complex biological circuits. We develop modular modeling frameworks, formal theory, and computer-assisted design (CAD) tools for design and analysis of biological systems at a larger scale.
This thesis introduces a new problem of robustness in structured model reduction of dynamical systems and provides bounds on a robustness distance metric for linear and nonlinear systems. With this theory, we show the discrimination and quantification of different mathematical models, considering resource loading effects in biological circuits.
Using our proposed model reduction robustness theory and its associated software development, we build a modeling, analysis, and parameter identification pipeline. This pipeline is demonstrated through the characterization of DNA recombination enzymes in a cell-free protein expression system. This pipeline is a general approach to systematically develop mathematical models, infer parameters from experimental data, and guide experimental design choices.
Identification of parameters in detailed mathematical models is a major challenge in synthetic biology where only sparse data is available. This prevents the application of our detail-driven modeling approach to larger biological systems. Hence, to address this limitation, we present a formal methods-based approach for specifying and synthesizing implementations for the design of biological circuits. We present a contract-based design framework for synthetic biology. We write formal description of design objectives at a higher level of abstraction without modeling the details of each component. This design framework facilitates the design and prediction of complex synthetic biological circuits at scale.
Overall, this thesis contributes to the advancement of synthetic biology by providing novel modeling frameworks, analysis methods, and design approaches. These contributions aim to enable the design and analysis of complex biological systems and foster the systematic engineering of biological circuits.</p
Distributed Control Theory for Biological and Cyberphysical Systems
Get PDFIn engineering, control theory plays a crucial role in the design and analysis of robust and efficient systems --- including robots, spacecraft, and power grids. In biology, control theory underlies sensorimotor and locomotion models of organisms. Distributed control is particularly useful for large-scale cyber-physical systems and also in biological systems, where communication is more limited than in engineered counterparts. In this thesis, I provide a number of theoretical advances in distributed control theory on the relationship between communication within controllers vs. closed-loop behavior in both the online and offline settings, on the application of distributed methods to robust control, and on necessarily information flow within controllers subject to communication constraints. I then discuss the applications of these theoretical advances to the primate cortex, as well as to sensorimotor models of drosophila locomotion. Overall, the contributions outlined in this thesis facilitate modeling techniques and insights that were previously unavailable
Exploring How Entangled Photon Correlations Can Enhance Spectroscopy
Get PDFQuantum light sources consisting of highly correlated or "entangled" photon pairs are increasingly becoming popular alternatives to classical light sources to perform microscopy and spectroscopy. Entangled photon pairs can replicate and enhance spectroscopic signals and have practical advantages compared to the pulsed laser systems that are typically utilized to perform these measurements. For instance, entangled photons are inherently low-flux, enabling measurements to be performed without undesired photoeffects, such as sample heating and degredation or nonideal photoinduced sample behavior. In addition, entangled photon sources can be generated and manipulated on much smaller physical footprints than state-of-the-art pulsed laser systems with comparable frequency bandwidths and time resolutions. Together, these capabilities could allow for the development of spectroscopic instruments that do not rely on bulky, expensive pulsed laser systems that necessitate teams of specialists to maintain. In turn, this instrument development could enable more widespread access to exotic forms of atomic and material characterization.
Despite a growing body of theoretical work, the field of experimental entangled photon spectroscopy is still nascent and entangled light-matter interactions have yet to be fully characterized in laboratory settings. Here, we investigate entangled photon light-matter interactions towards the goal of developing entangled spectroscopic techniques. A broadband entangled photon source with femtosecond coherence times is designed and characterized to perform these measurements. Using this source and an entangled photon spectrometer, characterization of the entangled photon enhancement to two-photon absorption are attempted by in studies of two different molecular dyes, Rhodamine 6G and zinc tetraphenylporphyrin. The entangled photon two-photon absorption enhancement is determined to be below previously reported values due to the presence of single photon scattering signals. Finally, entangled photons are utilized to replicate fluorescence lifetime measurements using a continuous wave pump laser and the temporal correlations inherent to entangled photon pairs. As the first experimental demonstration of this technique, the fluorescence lifetimes of indocyanine green in three solvent systems are measured.</p
Stimulated Raman Scattering: a Biophysical Perspective for Imaging Cells and Tissues
Get PDFThis thesis explores the utilization of Stimulated Raman Scattering (SRS) microscopy as a novel imaging method in the biomedical field, aiming to overcome the limitations associated with traditional fluorescence-based techniques. Given the drawbacks of fluorescence imaging, such as photobleaching, auto-fluorescence, and the complexity of fluorophore labeling, SRS microscopy emerges as a promising solution. The optical imaging contrast in this method originates from bond vibrations of endogenous biomolecules. Grounded in the principle of Raman scattering, SRS amplifies weak spontaneous Raman transitions through stimulated emission, offering a target-specific, high-speed, and label-free imaging modality that can overcome the challenges of traditional bio-imaging techniques.
To tackle the interference from fluorescent proteins when imaging small proteins of interest, we demonstrated a combination of SRS with selective deuterium labeling for visualizing polyQ aggregates in Huntington's disease. We targeted the C-D vibration on deuterated glutamines, which are metabolically enriched in the polyQ sequence. This allowed us to image Huntingtin aggregates without using fluorescent labels. Our method enables, for the first time, the quantification of protein concentrations and compositional analyses of polyQ and non-polyQ proteins within native Huntingtin aggregates. This novel perspective suggests that aggregates have distinct biophysical roles at different stages of aggregation.
In addition to fluorescent proteins, immunofluorescence is the gold standard for visualizing the location and distribution of proteins within cells or tissues. However, the proper delivery of antibodies is slow and labor-intensive. To overcome this issue, we developed a novel method, Vibrational Imaging of Swelled Tissue and Analysis (VISTA), that combines SRS microscopy with sample expansion to enable label-free super-resolution volumetric imaging in tissues. We developed a unique fixation hydrogel chemistry to maximize protein retention, delipidation, and isotropic expansion in tissue samples. By targeting the bond vibrations from endogenous proteins, VISTA bypasses the limitations of antibody labeling and provides an efficient tool for high-throughput imaging that can be scaled to large-volume clinical samples. The addition of image segmentation methods to VISTA equips it with protein-level specificity similar to immunofluorescence. We further used this technique to study protein aggregates, such as amyloid-β plaques in Alzheimer's disease, revealing intricate aggregate structures and polymorphisms absent in conventional fluorescence methods.
Finally, as fluorescent biosensors are indispensable tools for studying intracellular dynamics, we worked on extending the utility of SRS microscopy into the realm of sensing. We employed hydrogen-deuterium exchange on alkyne substrates to develop a Raman-based sensing strategy sensitive to subtle variations in local microenvironments. The rate of hydrogen-deuterium exchange changes under different conditions, and the resulting frequency shift from alkyne to deuterated alkyne is captured by SRS microscopy. This new platform enhances the study of chemical environments in various biological structures, marking a pivotal step in integrating imaging and sensing in biophysical research.</p