Files in this item



application/pdfSMITH-DISSERTATION-2018.pdf (8MB)Restricted Access
(no description provided)PDF


Title:Single molecule methods for clinical biomarker quantification
Author(s):Smith, Lucas David
Director of Research:Smith, Andrew M
Doctoral Committee Chair(s):Smith, Andrew M
Doctoral Committee Member(s):Cunningham, Brian T; Nie, Shuming; Kohli, Manish
Department / Program:Bioengineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Single Molecule, Biomarker, miRNA, Rolling Circle Amplification, RCA, Imaging, Reverse Transcription Polymerase Chain Reaction, RT-PCR, Flow Cytometry, Multiplexed, Molecular Barcode
Abstract:In recent years there has been an immense expansion of efforts focused on the identification of sensitive and specific molecular indicators of disease. As our understanding of these biomarkers continues to advance, the ability to quantify low abundance biomarkers from small sample sizes has become increasingly important for characterizing new targets and expanding the range of available disease indicators. In addition to contributing to an ever-growing catalogue of robust disease biomarkers, the potential to measure these targets using multiplexed panels and combinatorial analysis shows immense promise for deriving even greater improvements in test efficacy. While these tests show immense potential when utilized in clinical laboratories, an optimal testing platform would also enable point of care testing to improve patient outcomes by facilitating rapid adjustments to treatment strategies. Despite extensive demand for improved bioanalysis platforms, state of the art methods remain dependent on incredibly complex and time consuming techniques which exhibit insufficient sensitivity and minimal capacity for multiplexing. This thesis aims to address these limitations through the design of improved molecular diagnostic technologies. In particular, we focus on the quantification of miRNA, an emerging class of nucleic acid biomarkers that are among the most promising diagnostic and prognostic indicators of cancer. To address the main drawbacks of existing diagnostic methods, we develop a series of tools for directly labeling miRNAs with fluorescence probes to enable detection at the single molecule level. These techniques are then used to show that fluorescently labeled miRNAs can be differentiated from background signal with 100% sensitivity and 100% specificity in all samples analyzed, effectively eliminating the contribution of background signal to assay sensitivity. We go on to develop an algorithm which utilizes fluorophore single molecule properties to convert averaged image intensities into precise single molecule counts over a 4.01×107 dynamic range. As a second stage of assay development, a real time rolling circle amplification (RT-RCA) technique is developed to provide a convenient method for monitoring miRNA labeling reaction kinetics. Enzymatic reaction parameters are then sequentially analyzed, and optimized parameters are demonstrated to improve assay limit of detection >455-fold relative to standard RCA methods. RCA amplicons are then shown to be sufficiently bright to enable the first known demonstration of single molecule detection of small nucleic acids using flow cytometry (SM-Flow), a platform typically utilized for measuring eukaryotic cells quadrillions (81×1015) of times larger than a single miRNA. Single molecule flow cytometry is then demonstrated to be capable of quantifying miRNA in the presence of serum RNA with a limit of detection of 47 fM. In comparison, qRT-PCR, the gold standard assay for miRNA, yields a limit of detection of 3.71 pM. In addition to improving assay limit of detection, SM-Flow is shown to achieve a superior dynamic range of 2.12×104. While dramatically improving miRNA quantification limit of detection and dynamic range represents a substantial advancement, an ideal detection platform would also be capable of measuring multiple targets in tandem. To achieve this goal, a method for molecular barcoding miRNA amplicons is developed. Molecular barcoding is then demonstrated to be capable of quantifying miRNA over 8 intensity levels and using 5 distinct fluorophores, a combination which is has the potential to enable the detection of 8008 unique targets. The capacity for large scale multiplexing is then further validated using empirical measurements of nonspecific signal, and a multiplex capacity of up to 126 targets is shown to be feasible without any detectable nonspecific binding. Additional data indicates that as many as 495 targets can be measured with a false positive rate of <6.6×10-5. Together with the immense benefits to assay sensitivity relative to the gold standard qRT-PCR, these findings indicate that the platforms developed herein have the potential to provide a user friendly, ultrasensitive, and highly multiplexed method for the detection of miRNA disease biomarkers.
Issue Date:2018-12-05
Rights Information:© 2018 Lucas David Smith
Date Available in IDEALS:2020-10-07
Date Deposited:2018-12

This item appears in the following Collection(s)

Item Statistics