Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

FRET–based proteolytic activity assays on quantum dots Wu, Miao 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_november_wu_miao.pdf [ 14.88MB ]
Metadata
JSON: 24-1.0167013.json
JSON-LD: 24-1.0167013-ld.json
RDF/XML (Pretty): 24-1.0167013-rdf.xml
RDF/JSON: 24-1.0167013-rdf.json
Turtle: 24-1.0167013-turtle.txt
N-Triples: 24-1.0167013-rdf-ntriples.txt
Original Record: 24-1.0167013-source.json
Full Text
24-1.0167013-fulltext.txt
Citation
24-1.0167013.ris

Full Text

 FRET-BASED PROTEOLYTIC ACTIVITY ASSAYS ON QUANTUM DOTS  by  Miao Wu  B.Sc., Nanjing University, 2012  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in The Faculty of Graduate and Postdoctoral Studies (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) October 2014 © Miao Wu 2014   	   ii	  Abstract   Proteases play crucial roles in a multitude of biological processes. However, the behavior of proteases is different when the hydrolysis process occurs at the surface of nanoparticles when compared with that in bulk solution. Preliminary studies have reported an enhancement of hydrolase activity when multiple substrates are conjugated on a nanoparticle surface. The differences in activity and kinetic profiles were partly attributed to interactions between the hydrolase and the nanoparticle surface. Such phenomena have revealed the importance of studying the effect of nanoparticle surface properties on proteolysis. One of the most widely used nanoparticles in bioanalytical applications are quantum dots (QDs). In this work, QDs were used as a scaffold for the study of proteolytic activity on the surface of a nanoparticle where multiple copies of peptide substrate were co-localized, surface chemistry could be varied, and the progress of proteolysis tracked by Förster resonance energy transfer (FRET). The surface was modified with four different types of anionic, small-molecule ligand coatings that are commonly used in the literature: CYS (Cysteine), DHLA (Dihydrolipoic acid), GSH (Glutathione) and MPA (Mercaptopropionic acid). Difference in properties, such as the relative charge on the QDs, appeared to have a large effect on the rate of proteolysis, enhancing or inhibiting protease activity relative to bulk solution. Kinetic profiles were compared for two model proteases, trypsin and thrombin, that hydrolyze the same substrate. Of these two model proteases, thrombin was more sensitive to the QD coating and had a more varied response to different coatings. These results may provide a new way to adjust sensitivity and selectivity in proteolytic assays in vitro. Further, as a first step toward studying proteolysis in biological systems, the QD-FRET method used to track proteolysis in vitro has been adapted to fluorescence microscopy, which enables measurement of spatially heterogeneous protease activity, such as would be encountered with cultured cells. Optical parameters, such as exposure time and excitation intensity are optimized, and calibration samples and homogeneous proteolytic assays were compared between measurements with an epifluorescence microscope and a fluorescence plate reader. Proof-of-concept for heterogeneous proteolytic assays was also demonstrated. 	   iii	  Preface  The work presented in this thesis is based on currently unpublished work. I performed the experiments, simulations and data analysis, and wrote the thesis. My supervisor, Dr. Russ Algar, helped with the design of the research, interpretation and discussion of the results, and editing of the writing.   	   iv	  Table of contents Abstract .......................................................................................................................... ii Preface ........................................................................................................................... iii Table of contents .......................................................................................................... iv List of tables .................................................................................................................. vi List of figures ............................................................................................................... vii List of symbols and abbreviations ............................................................................ viii Acknowledgements ...................................................................................................... ix Chapter 1: Introduction ............................................................................................... 1 1.1 Introduction to quantum dots ............................................................................. 2 1.1.1 Optical properties ........................................................................................... 2 1.1.2 Materials ........................................................................................................ 3 1.1.3 Interfacial chemistry: coatings and bioconjugation ....................................... 4 1.1.4 Advantages of QDs for biosensing ................................................................ 7 1.2 Föster resonance energy transfer ........................................................................ 8 1.2.1 Fluorescence .................................................................................................. 8 1.2.2 Förster resonance energy transfer ................................................................ 10 1.2.3 QD and FRET .............................................................................................. 12 1.3 Proteolysis assays, bioprobes and biosensors .................................................. 14 1.3.1 Proteolysis ................................................................................................... 14 1.3.2 Proteolysis assays using QD and FRET ...................................................... 14 1.4 Contributions of this thesis .............................................................................. 16 Chapter 2: Enhancement and inhibition of proteolytic activity associated with selection of small molecule thiol ligand coatings on QD .......................................... 18 2.1 Introduction ...................................................................................................... 18 2.2 Materials and methods ..................................................................................... 20 2.2.1 Materials ...................................................................................................... 20 2.2.2 Preparation of ligand-coated QDs: CYS, DHLA, GSH and MPA .............. 21 2.2.3 Enzyme assays ............................................................................................. 22 2.2.4 Preparation of calibration curves ................................................................. 22 2.2.5 QD inhibition assays .................................................................................... 23 2.2.6 Homogeneous assays ................................................................................... 23 2.2.7 Evaluating the effect of free ligand on proteolytic activity ......................... 24 2.2.8 Evaluation of denaturation of protease on the QD surface .......................... 25 2.2.9 Data analysis ................................................................................................ 25 	   v	  2.3 Results and discussion ..................................................................................... 26 2.3.1 QD-A647 FRET pair ................................................................................... 26 2.3.2 Calibration curves ........................................................................................ 27 2.3.3 Proteolytic assays with QD probes .............................................................. 31 2.3.4 Adsorption of protease on QDs ................................................................... 33 2.3.5 Inhibition assays .......................................................................................... 35 2.3.6 Evaluating the effect of free ligand on proteolytic activity ......................... 38 2.3.7 Evaluation of denaturation of protease on the QD surface .......................... 38 2.3.8 Modeling ...................................................................................................... 39 2.3.9 Proteolysis in homogeneous solution and on QD surfaces .......................... 44 2.3.10 Discussion and conclusions ....................................................................... 45 Chapter 3: Proteolysis measurement using microscope .......................................... 47 3.1 Materials and methods ..................................................................................... 47 3.1.1 Materials and QD-peptide substrate conjugates .......................................... 47 3.1.2 Fluorescence microscopy ............................................................................. 47 3.1.3 Image acquisition and analysis software ..................................................... 48 3.1.4 Calibration of FRET pairs ........................................................................... 48 3.1.5 Homogeneous proteolytic assays ................................................................. 49 3.1.6 Heterogeneous proteolytic assays ................................................................ 49 3.2 Results and discussion ..................................................................................... 50 3.2.1 Crosstalk between different channels .......................................................... 50 3.2.2 Effect of excitation energy and exposure time ............................................ 51 3.2.3 Calibration using fluorescence microscopy: data collection and analysis ... 53 3.2.4 Homogeneous kinetic experiment and comparison with fluorescence plate reader ........................................................................................................... 54 3.2.5 Heterogeneous kinetic experiment using microscope ................................. 55 3.2.6 Discussion and conclusions ......................................................................... 57 Chapter 4: Conclusion and future prospects ............................................................ 59 Bibliography ................................................................................................................ 61    	   vi	  List of tables  Table 2.1 Crosstalk correction factors. ............................................................................. 25 Table 2.2 Quantum yields and Föster distances for the X-QD-A647 FRET pairs. .......... 27 Table 2.3 Apparent specificity constants, kcat/Km for X-QD624-Sub(A647)16 conjugates, and apparent dissociation constants for X-QD524, estimated from initial rates of proteolytic digestion. ................................................................................................. 37 Table 2.4 Kd (µM) and rate constant for X-QD-Sub(A647). ............................................ 41   	   vii	  List of figures  Figure 1.1 Introduction to the optical properties of quantum dots ..................................... 3 Figure 1.2 Illustration of essential elements of interfacial chemistry for QD .................... 7 Figure 1.3 Jablonski diagram .............................................................................................. 8 Figure 1.4 Methods for FRET efficiency measurement ................................................... 11 Figure 1.5 Schematic of the QD-based proteolysis assays ............................................... 13 Figure 1.6 QD-FRET proteolytic assay design ................................................................. 15 Figure 2.1 Schematic of proteolytic assays ....................................................................... 20 Figure 2.2 Normalized absorption and emission spectra for QD-A647 FRET pair ......... 27 Figure 2.3 Calibration curves ............................................................................................ 30 Figure 2.4 Progress curves and initial rates ...................................................................... 32 Figure 2.5 Gel electrophoresis for THR ........................................................................... 34 Figure 2.6 Gel electrophoresis for TRP ............................................................................ 34 Figure 2.7 Inhibition assays .............................................................................................. 36 Figure 2.8 Progress curves for THR proteolytic assays with small molecule substrates and initial rates ................................................................................................................. 38 Figure 2.9 Denaturation of enzyme on QD surface .......................................................... 39 Figure 2.10 Comparison of different models .................................................................... 42 Figure 2.11 Comparison of different models .................................................................... 43 Figure 2.12 Progress curves for homogeneous assays ...................................................... 45 Figure 3.1 PL spectrum and microscope filter channels ................................................... 50 Figure 3.2 FRET and crosstalk between different channels ............................................. 51 Figure 3.3 Calibration curves and images ......................................................................... 53 Figure 3.4 Comparison of calibration curves from plate reader and microscope ............. 54 Figure 3.5 Proteolytic assays in microcope and plate reader ............................................ 55 Figure 3.6 Progress images of proteolysis assays in capillary .......................................... 57 Figure 3.7 Ratio images of heterogeneous TRP assays in agarose gel ............................. 57   	   viii	  List of symbols and abbreviations  A555   Alexa Fluor 555    A647   Alexa Fluor 647    ACPPs   Activatable cell-penetrating peptides  BBS   Borate buffered saline          CYS    Cysteine                     DHLA   Dihydrolipoic acid                  DMSO   Dimethyl sulfoxide   FRET   Förster resonance energy transfer     GSH    Glutathione                     HEPES   4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid   MAA   Mercaptoacetic acid MMPs   Matrix metalloproteinases    MPA    Mercaptopropionic acid                  Ni-NTA   Ni2+-nitrilotriacetic acid    PEG    Polyethylene glycol. PL    Photoluminescence                              QD   Quantum dot     TEAA   Triethylamine acetate    THR    Thrombin      TLCK   Tosyl-L-lysine chloromethyl ketone hydrochloride  TMAH   Tetramethylammonium hydroxide      TRP    Trypsin         	   ix	  Acknowledgements   I would like to express my sincere gratitude to my supervisor Dr. Russ Algar for his support of my study and research, for his patience, inspiration, guidance and immense knowledge. I would also like to thank all of the members of the research group for their help and friendship: Daniel Bajj, Erin Conroy, David Kim, Kate Liu, Melissa Massey, Jennifer Moon, Eleonora Petryayeva, and John Sharp.  I would like to express gratitude to Dr. Straus, Dr. Bizzotto and Dr Mitchell my committee members, for their assistance. Finally, I would like to thank my parents for always being there for me. 	   1	  Chapter 1: Introduction Nanomaterials have had an important impact on the field of biosensing and bioanalysis. Of the many different nanomaterials that have been investigated for such applications, semiconductor nanocrystal quantum dots (QDs) are among the most promising applications.  QDs have many advantageous physical and optical properties, such as broad and strong light absorption, spectrally narrow and tunable luminescence, high quantum yield, and many others. In many QD applications, Föster resonance energy transfer (FRET) is involved as an energy transfer mechanism to link the photoluminescence (PL) of the QD to a chemical or biological process to generate a measurable signal. For example, FRET is often combined with QDs for biosensing applications where the activity of a protease is the target to be detected. Proteases play important roles in biological systems, including the initiation, termination and modification of many cellular functions. In this work, a QD-based probe is used to measure protease activity through changes in FRET efficiency. For such bioprobes, the surface of the QD acts as the platform where the biological process takes place. Thus, the chemical and physical properties of that surface may play a substantial role in the biosensing through interactions with the target biomolecule. This work focuses on the effect that different types of coatings on QDs have on proteolysis, and fluorescence methods for measuring these effects, including initial work on quantitative epifluorescence imaging.  This introductory chapter is organized into three sections that provide the background needed to appreciate the research in this thesis: Section 1.1 describes the properties and biofunctionalization of QDs; Section 1.2 summarizes the principles of fluorescence and FRET; and Section 1.3 reviews the proteolysis process. Finally, Section 1.4 summarizes the contribution of this thesis.Research methods and results are described in Chapters 2–3 of this thesis. Chapter 2 presents a detailed study of the effects of different QD ligand coatings on proteolysis. Chapter 3 demonstrates the use of epifluorescence microscopy to image and measure heterogeneous proteolytic activity with QD probes. Chapter 4 summarizes the work presented in this thesis, emphasizing its significance and directions for future work.  	   2	  1.1 Introduction to quantum dots 1.1.1 Optical properties QDs are colloidal semiconductor nanocrystals that are 2 to 10 nm in diameter.1 They comprise hundreds to tens of thousands of atoms arranged in a periodic lattice similar to the bulk crystalline material. The unique optical properties are caused by both their nanoscale size and the choice of semiconductor materials, which lead to quantum confinement and modification of band gap energies with changes in particle size when the diameter of the QD is comparable to its exciton Bohr radius .2  When bulk semiconductors absorb light, the photon energy is used to generate an exciton via excitation of an electron across the band gap, from the valence band to the conduction band. The same process occurs in QDs with the distinction that, as the particle size becomes comparable to the Bohr exciton radius for the material, the band gap energy becomes dependent on the size of the core nanocrystal.3 As the nanocrystal size decreases, the band gap energy increases (See Figure 1.1A). Changes in the band gap energy result in different colours of emission when the exciton electron and hole recombine and release the energy as PL (See Figure 1.1B).                  	   3	                    Figure 1.1 Introduction to the optical properties of quantum dots (A) Correlation between core size and emission colour (shown as both a photograph under UV light and PL emission spectra) for CdSe QDs. (B) Qualitative energy level diagram for CdSe QDs of different size, illustrating the effect of quantum confinement on transitions for absorption (Abs.) and emission (Em.) from recombination of the electron (e–) and hole (h+) across the band gap (Eg). (C) Core material and corresponding emission colour. Reprinted with permission from ref. 4. Copyright 2011 American Chemical Society.  The bright PL from QDs is a consequence of their strong, broad absorption spectra and quantum yields that are comparable to common fluorescent dyes. QDs usually have molar absorption coefficients in the range of 104–107 M–1 cm–1 (cf. 104–105 for organic dyes) with the lowest values at the absorption edge (i.e., the band gap energy) and increasing at shorter wavelengths of light. QD PL is spectrally narrow, with full-width-at-half-maxima typically in the range of 25–35 nm for relatively monodisperse samples, and its spectral peak position can be tuned through control of nanocrystal size, composition (see Figure 1.1C), shape, and structure. Tuning PL through the size and composition of the QD nanocrystal are most widely used for optical sensing and bioanalysis. 1.1.2 Materials The most widely utilized QD material for bioanalysis and bioimaging applications is the core/shell CdSe/ZnS QD. QDs with such structure and materials have several advantages, A BC 	   4	  including narrow PL across the visible spectrum as the core nanocrystal size is varied between approximately 2–6 nm.1b Methods for synthesizing high-quality CdSe cores are well-known, and the growth of a thin (typically ≤ 1 nm) ZnS shell around the CdSe core is applied to protect and enhance the optical properties of the core. ZnS is the most common shell material because of its structural compatibility with most core materials and relatively large band gap energy.5 The former property helps passivate surface trap states that decrease the quantum yield of the core nanocrytal, and the latter property helps confine a photogenerated exciton to that core nanocrystal, minimizing leakage of the exciton wavefunction into the surrounding environment.5 Common core materials include binary semiconductors such as CdS, CdSe, CdTe, PbS, PbSe, InAs, InP, ZnO, Si.1b, 1c, 4 Alloyed core materials such as CdSexS1–x and CdSexTe1–x are also utilized.6 The core nanocrystal composition determines the spectral range over which PL can be tuned through changes in nanocrystal size, and the PL from alloyed QDs can also be tuned through changes in composition. Shell materials are less diverse than core materials. CdS and ZnSe are the most common alternatives to ZnS.  1.1.3 Interfacial chemistry: coatings and bioconjugation Since high-quality semiconductor QDs are synthesized in organic solvent and are not compatible with aqueous solution, the unique properties of QDs did not attract widespread interest for biological applications until 1998, when Chan and Nie and Bruchez et al. reported methods to render high-quality core/shell QDs water soluble and biofunctional through interfacial chemistry.7 In order to develop energy transfer-based probes for bioanalysis, the ideal criteria for QD coatings include: (1) high affinity for the QD; (2) the ability to provide long-term colloidal stability under biologically relevant conditions; (3) capacity for bioconjugation; (4) minimal thickness; and (5) resistance to the adsorption of biomolecules.4 Other ideal criteria may be important in other applications with QDs, but are generally similar to the criteria above.  The two most established methods for coating QDs are encapsulation with amphiphilic polymers and ligand exchange with bifunctional molecules.8 The first example of ligand exchange was the replacement of hydrophobic phosphine (oxide) ligands on as-	   5	  synthesized CdSe/ZnS QDs with mercaptoacetic acid (MAA).7a This process was driven by mass action and the high affinity of the thiol group for the inorganic surface of the QDs. The carboxylate group enabled stable colloidal dispersion in aqueous solution and provided a functional group for bioconjugation with protein. Unfortunately, the stability of QDs coated with thioalkyl acids is limited to basic pH and low ionic strength. These ligands may also gradually desorb from the surface of the QD. This problem can be partly solved by using bidentate thiols ligands such as dihydrolipoic acid (DHLA).9 The limitations of electrostatic stabilization can be overcome by the use of poly(ethylene glycol) (PEG) or zwitterionic derivatives, which provide colloidal stability over a much broader range of ionic strength and pH, and better resist the non-specific adsorption of biomolecules.4, 8b A significant advantage of ligand coatings is their compact size, which permits QDs and fluorescent dyes or other optically active materials to be in close proximity, increasing the efficiency of FRET (see Section 1.2). A drawback is that ligand exchange and transfer to aqueous solution leads to the decline of QD quantum yield. In contrast, encapsulation of as-synthesized QDs within an amphiphilic polymer coating better retains the quantum yield from the initial synthesis but at the same time leads to a much larger size.10 Amphiphilic polymers are composed of a mixture of pendant alkyl chains and pendant hydrophilic groups such as carboxylic acids and PEG chains. The alkyl chains interdigitate with the hydrophobic ligands on the surface of the QD from synthesis.4, 11 Amphiphilic polymer coatings are less frequently used than ligand coatings for energy transfer-based bioanalyses with QDs, but are the underlying layer in commercially available streptavidin-coated QDs that are used in many other applications.  In addition to aqueous compatibility, bioconjugate chemistry is needed to render QDs biofunctional through the attachment of antibodies, enzymes, other proteins, peptides, oligonucleotides, aptamers, and many other biomolecular probes.12 The most common bioconjugate methods can generally be categorized as either covalent coupling or self-assembly. Covalent coupling involves the formation of new chemical bonds between the coating on the QD and the biomolecule(s) of interest. Such reactions generally happen at amine, carboxyl, and thiol groups on the target biomolecule and the coating of the QD via carbodiimide activation, succinimidyl esters, maleimides, and homo- or hetero-bifunctional crosslinkers based on these reactive groups.  	   6	  Non-covalent self-assembly such as the formation of dative bonds or biological complexes are also widely-used methods for bioconjugation. Examples of the former include the ability of thiol- and polyhistidine-terminated linkers to coordinate to the inorganic surface of QDs.13 For such methods, the linker must be able to penetrate the coating of the QD. Usually, peptides and oligonucleotides with these modifications can bind to QDs with a wide array of ligand coatings; however, proteins are not able to bind to PEG–appended DHLA ligands without the help of a long linker, which is specially designed for such coatings through protein engineering.14 Generally, polymer coatings do not work with direct self-assembly on the QD surface. Modification of polymers with Ni2+-nitrilotriacetic acid (Ni-NTA) is needed as a bridge for the self-assembly of polyhistidine-appended biomolecules. Carboxylated polymer coatings can also directly coordinate Ni2+ for the same purpose.15 Bulky ligands can also be terminated with Ni-NTA to permit self-assembly.16 Perhaps the most common bioconjugation strategy is the spontaneous binding of biotinylated biomolecules with commercial streptavidin-coated QDs. Biotin-streptavidin binding is one of the strongest known non-covalent interactions 17 and biotinylated biomolecules are very widely available.   Important considerations for bioconjugate chemistry include the degree of control over the number of conjugated biomolecules per QD (valence), the orientation of those biomolecules, the retention of their biological activity, the stability of the linkage, the mildness of the reaction conditions, the kinetics and yield of the coupling reaction, competing reactions (e.g., hydrolysis) and undesirable crosslinking, and compatibility with other bioconjugate chemistries.12 In general, QDs are a polyvalent interface and the final ensemble of conjugated QDs will typically exhibit a Poisson distribution in the number of biomolecules per QD.18 The bioconjugate chemistry selected for the preparation of QD probes can have a significant impact on final result.   	   7	                        Figure 1.2 Illustration of essential elements of interfacial chemistry for QD A core/shell QD (e.g., CdSe/ZnS) is either encapsulated in an amphiphilic polymer (i–ii) or modified with various hydrophilic, bifunctional ligands (iii–v). Biomolecules of interest (BOI) are then conjugated to the QD using affinity interactions (vi–vii) or covalent coupling (viii–xi). Representative examples of chemistries are shown, but many others have been utilized. Not drawn to scale. Reprinted with permission from 8a. Copyright 2013 Society for Applied Spectroscopy.  1.1.4 Advantages of QDs for biosensing The physical and optical properties of QDs offer several advantages for optical sensing and bioanalysis. The excellent photostability of high-quality QDs permits measurements over extended periods of time, and their high brightness can provide a sensitivity advantage. The broad absorption of QDs permits efficient excitation with a wide variety sources and wavelengths. QDs also afford several technical advantages for multiplexed bioanalysis: multiple colours of QD can be excited using a common excitation wavelength, and the narrow, symmetric, and tunable emissions help with maximizing resolution and deconvolution of more signals in a given spectral range than is typically possible with the broader, red-tailed emission of fluorescent dyes. Another potential advantage of QDs is their high surface area-to-volume ratio. QDs are small enough to enter cells and tissues but still have large surface areas for chemical modification and bioconjugation. On the contrary, modification of fluorescent dyes is usually limited to one reactive functional group for conjugation to a biomolecule of interest. QDs can be 	   8	  modified with diverse functional coatings and can be conjugated with multiple copies of that biomolecule or even multiple copies of different biomolecules. 1.2  Föster resonance energy transfer Photoluminescence (PL) is a phenomenon where ultraviolet, visible or infrared radiation is emitted by a material and this emission is induced by photon absorption. Fluorescence is a form of photoluminescence, where the transition of an electron from a singlet excited state to the ground state is involved.  1.2.1 Fluorescence                     Figure 1.3 Jablonski diagram S0, S1 and S2 are the ground, first excited, and second excited electronic singlet states. In each electronic state superimposed several vibrational states are shown.   𝒉𝒗𝒂𝟐  and 𝒉𝒗𝒂𝟏  indicates the energy absorption of the excitation from ground state to the fisrt and second excited states.  𝒉𝒗𝒇 indicates the emission of energy through fluorescence.  A chromophore can be excited to higher electronic states through the absorption of photons and then release the energy and relax to lower state through several ways, such as (1) vibrational relaxation occurring on timescale of 10-10–10-13 s, and (2) internal conversion, both of which transfer heat to the microenvironment through molecular vibrational transfer. Another possibility, (3) fluorescence, is a radiative transition with the emission of photon. A Jablonski Diagram is widely used to depict the absorption and emission of light, and to illustrate molecular process at the excited electronic states. A 	   9	  typical Jablonski diagram is shown in Figure 1.3. The ground state and first and second excited singlet states are S0, S1 and S2, respectively. The radiative transitions between different states are depicted with vertical lines and the non-radiative relaxations are depicted with dashed lines. The sublevels within a single electronic state are vibrational energy levels, which are closely spaced and responsible for the broad absorption and emission bands of fluorophores. Molecules in the ground states can transition to excited states by absorbing photons. Based on the amount of energy absorbed, molecules might be excited to the first excited state or higher excited states. In each excited electronic state, molecules occupying excited vibrational levels relax quickly (10-12 s) to the ground vibrational states, in a process referred to as vibrational relaxation. Except from the first excited states, molecules occupying the lowest vibrational state of higher excited electronic states will transition to an excited vibrational level of a lower electronic state, in a process referred to as internal conversion. Internal conversion is a non-radiative process, on the timescale of 10–9–10–7 s. The internal conversion between first excited state and the ground state is slower and thus fluorescence always happens from the lowest vibrational state of the first excited electronic state to the ground electronic state, a principle called Kasha’s rule, which also explains why the emission is not dependent on the excitation wavelength. Excess energy is dissipated as heat. The quantum yield is a measure of the efficiency of fluorescence compared with other relaxation pathways, defined by equation. 1.1, where kf is the rate of fluorescence and knr is the rate of non-radiative relaxation. In practice, the quantum yield represents the ratio of the number of emitted photons over the number of absorbed photons. Φ? = ?????? ™                                                         (1.1)  After returning to the ground electronic state, S0, the molecule relaxes from higher vibrational states to reach thermal equilibrium. Since the ground and excited electronic states have similar vibrational states, the absorption and emission spectrum is usually a mirror image. The rapid decay to the ground vibrational energy level of S1 consumes some energy of excitation and results in a less emission energy and longer emission wavelength compare to absorption. This difference between the absorption peak and the emission peak is defined as Stokes Shift.  	   10	  1.2.2 Förster resonance energy transfer Förster (or fluorescence) resonance energy transfer (FRET) is one of the most powerful fluorescence techniques in biosensing. It is a non-radiative, through-space energy transfer mechanism, mediated by dipole-dipole interactions between an excited state donor fluorophore and a ground state acceptor chromophore. The rate of FRET between a donor fluorophore and an acceptor chromophore, kFRET, is given by equation 1.2, where τD = (kr + knr)–1 is the inverse decay rate of the donor, kr is the radiative decay rate, knr is the non-radiative decay rate, R0 is the Förster distance, r is the distance separating the donor and acceptor. ΦD is the quantum yield of the donor, κ2 is the orientation factor, J(λ) is the spectral overlap integral, NA is Avogadro’s number, and n is the refractive index of the medium between the donor and the acceptor.  (1.2)  The spectral overlap integral is defined by equation 1.3, where ID(λ) is the wavelength-dependent emission intensity of the donor, εA(λ) is the wavelength-dependent absorption coefficient of the acceptor, and λ is the wavelength. It measures the degree of resonance between the donor and acceptor.    (1.3)  The efficiency of energy transfer, E, is the ratio of the rate of FRET to all processes including both radiative and non-radiative processes, given by equation 1.4, where it is seen that R0 is the distance between donor and acceptor that corresponds to kFRET = τD–1. By definition, the FRET efficiency is 50% when r = R0.   (1.4) kFRET = 1τD R0r!"# $%&6 = 1τD 9(ln10)ΦDκ 2J(λ)128π 5NAn4r6J(λ) = ID (λ)εA (λ)λ 4 dλ∫ ID (λ)dλ∫E = kFRETkr + knr + kFRET = R06r6 + R06	   11	   Since the rate of FRET decreases as the inverse-sixth power of the distance between the donor and the acceptor, energy transfer is very sensitive to donor-acceptor distance in the range of 0.5R0< r <1.5R0. Since Förster distances for typical donor-acceptor pairs are in the range of ca. 3–6 nm, FRET makes an excellent molecular ruler to accurately measure distances at typical biomolecule length scales. This scaling, combined with the fast, sensitive, and non-invasive nature of fluorescence measurements, makes FRET ideal for monitoring dynamic biological processes at the ensemble and single-molecule levels.19          Figure 1.4 Methods for FRET efficiency measurement Model data for the change in the emission spectra for a Cy3 donor paired with a Cy5 acceptor (1:1 ratio) at different FRET efficiencies (0 %, 33 %, 67 %). Note the small direct excitation of Cy5 in the absence of FRET. Increases in FRET efficiency are reflected by progressive quenching of the Cy3 fluorescence and sensitization of Cy5 fluorescence. (b) Model data for the change in Cy3 donor lifetime with increasing FRET efficiency.  Experimentally, the most common method to measure FRET efficiency is either from quenching of the donor emission intensity or the decrease in the lifetime (Figure 1.4 B), equation 1.5, where I is an emission intensity, τ is an emission lifetime, D is a subscript denoting a donor-only reference state, and DA is subscript denoting a donor quantity measured in the presence of acceptor. 	   12	    (1.5) Otherwise, FRET efficiency can be measured from the relative amounts of quenched donor emission, IDA, and FRET-induced acceptor emission, IAD (Figure 1.4 A), as shown in equation 1.6, where IA is acceptor emission from direct optical excitation, and Φ is a quantum yield. This method is only applicable when the acceptor is fluorescent too; however, it can provide a more reliable measurement since it is no longer dependent on only the donor emission. The ratio of donor and acceptor emission minimizes the effects of variations in concentration and variations in excitation intensity or detector sensitivity, improving reliability when these parameters are not easily controlled. In many studies, the ratio IAD/IDA is used as a semi-quantitative measure of the amount of FRET without explicit calculation of the FRET efficiency.    (1.6)   1.2.3 QD and FRET QDs are highly advantageous as both donors and acceptors in FRET pairs for many reasons.20 Since QDs are more photostable than many other fluorophores, they extend the measurement time and can provide more stable and reliable emission. The high surface area-to-volume ratio enables the conjugation of multiple copies of all kinds of biomolecules on the surface and thus extends the versatility of FRET-based assays. As donors, the broad absorption spectrum of QDs affords more choices for excitation wavelength. From a wider range of possible excitation wavelengths, selection of an optimal wavelength that provides minimal direct excitation of acceptor dyes, a large effective Stokes shift, and simultaneous excitation of multiple QDs is possible. The advantageous properties of QDs also enable a higher FRET efficiency. The large interface (surface area) can be modified chemically or physically, and associated with multiple acceptors, which dramatically increases in the FRET efficiency according to equation 1.7. Furthermore, the narrow, tunable PL emission from QDs permits E = 1−IDAID= 1−τDAτDE = (IAD − IA )(ΦA ΦD )IDA + (IAD − IA )	   13	  maximization of the spectral overlap integral, J(λ), without introducing problematic emission crosstalk.    (1.7) As acceptors, QDs are advantageous because of their broad absorption spectra and large molar absorption coefficients, which afford large spectral overlap integrals and large Förster distances, with potentially large spectral separation between the donor emission and QD emission. However, the broad absorption will lead to unavoidable direct excitation of the QD. The special circumstances required to use QDs as acceptors are beyond the scope of this work.               Figure 1.5 Schematic of the QD-based proteolysis assays  E = NR06r6 + NR06EXC.FRETPL.PROTEOLYSISAcceptorFRET 'OFF'QDEXC.PL.QDQD	   14	  1.3 Proteolysis assays, bioprobes and biosensors 1.3.1 Proteolysis  A protease is a type of hydrolase enzyme that is able to catalyze the hydrolysis of the peptide bonds that link amino acids together in the polypeptide chains that form proteins. Protease targets include many kinds of bioactive molecules that play critical roles in biological systems, including other proteases, inhibitors, clotting factors, growth factors, growth factor-binding proteins, cell surface receptors, cell-cell adhesion molecules, and structural extracellular matrix proteins.21 The wide range of targets allows proteases to control the initiation, termination and modification of numerous biological functions in living organisms.  Since proteases have the ability to degrade or activate proteins that play key roles in biological systems, their activity that is deficient, absent, excessive, or misdirected could cause problems and lead to disease. Cancer, arthritis, and cardiovascular diseases are all related to protease activity in some capacity. Many infections also rely on the involvement of a protease to effect tissue destruction.21-22  Proteolysis takes place in the active site of the protease, which catalyzes the chemical transformation. Serine proteases are a large family of proteases that coordinate various functions;23 for example, thrombin is important in the blood coagulation and trypsin is important in digestion. The mechanism of hydrolysis in the active site of a protease includes a nucleophile and a base. In serine proteases, the nucleophile is serine and the base is histidine. The tetrahedral intermediate has an oxyanion within the structure, which is generally stabilized by hydrogen bond donors, such as the amide backbone NH groups of other amino acids in the active site. The position and properties of the oxyanion hole in the active site determines, in part, the substrate specificity of different proteases. And the interaction of the oxyanion and the peptide bond is essential for substrate hydrolysis.   1.3.2 Proteolysis assays using QD and FRET FRET has been widely used as a method for the detection of protease activity with dye labeled peptides. For example, Whitney et al. reported a FRET-based method to detect spatially localized thrombin activity in vivo.24 Activatable cell-penetrating peptides 	   15	  (ACPPs) with a Cy7-labeled polyanionic domain and a Cy5-labeled polycationic domain formed a hairpin to generate efficient FRET. Thrombin activity cleaved the ACPP at a recognition site, disengaging FRET and restoring Cy5 PL. With QDs, the most widely used configuration is where the QD and an optically active molecule are bridged by a peptide substrate of the analyte protease. The presence of the analyte protease hydrolyzes the peptide substrate, increases the distance between the FRET donor and acceptor, and thus decreases the FRET efficiency with a corresponding change in the measurable PL. This configuration is shown in Figure 1.5 and Figure 1.6.              Figure 1.6 QD-FRET proteolytic assay design Design of a QD-FRET assay for sensing proteolytic activity. Adapted with permission from Macmillan Publishers Ltd: Nature Materials.25 Copyright 2006.                                       It has been suggested that the peptide substrate should include (1) a linker that has affinity for the surface or coating of QDs (e.g., a polyhistidine tag or thiol group, both of which have affinity for the ZnS shell of a QD; or biotin, which has affinity for streptavidin-coated of QDs) or has a functional group for covalent coupling; (2) a spacer (e.g., a helix-forming motif such as a polyproline or an alanine/α-aminoisobutyric acid-rich sequence) following the linker to help a protease access its recognition site in the substrate; (3) a recognition site for hydrolysis of the peptide by the target protease; and (4) a terminal modification with a dye or quencher.25-26 Chang et al., Medintz et al., and Shi et al. reported the first examples of the above designs.25-27 In each case, enzyme-induced changes in QD or dye PL were used to track the proteolytic activity. For example, in the report by Chang et al., collagenase activity was detected through loss of QD PL quenching associated with energy transfer from CdSe/CdS QD620 to proximal 1.4 nm Au NPs that were terminal labels on peptide substrates covalently coupled to the QDs.27 Shi et al. dispersed QD545 in aqueous solution using short, thiol-terminated and Rhodamine Red-X dye-labeled peptides as ligands.26 These QD-peptide conjugates were 	   16	  used for qualitative detection of the activity of matrix metalloproteinases (MMPs) secreted into the culture medium of HTB 126 breast cancer cells. The in vitro assay showed a clear difference in extracellular proteolytic activity between cancerous cells and non-cancerous cells. In the work by Medintz et al., DHLA-coated QDs were conjugated with a peptide substrate for one of four different proteases.25 The peptides were labeled with either Cy3 or QXL, a dark quencher, and paired with QD510, QD520, or QD540. The QD-peptide FRET-probes were useful for measuring initial proteolytic rates and estimating apparent Michaelis-Menten parameters, and model protease inhibitor screening assays. Algar et al. later developed a method for measuring proteolysis kinetics associated with QD-peptide FRET-probes in real-time, including application of an integrated Michaelis-Menten kinetic model.28 These results suggested that the classical Michaelis-Menten model was not directly applicable with QD-peptide conjugates and that the interactions between the protease and the surface of the QD played an important role in the rate of proteolytic digestion of the peptides. A “hopping” mechanism was proposed where the protease molecules associate with one QD, then digest all the substrates on that specific QD before dissociating and adsorbing to a fresh QD-peptide substrate conjugate. Overall, some of the proteases targeted for QD-FRET sensing have included caspase-1, caspase-3, chymotrypsin, collagenase, matrix metalloproteinase-7, thrombin, trypsin, and others.20a  1.4 Contributions of this thesis The attractive properties of QD have resulted in the widespread development and application of QD-based bioassays and biosensors. Protease analysis is one of the most widely used QD-based bioassays with potential clinical applications, whether for the detection of proteolytic activity or the effects of protease inhibitors. One of the most important but least understood factors in these assays is the role of the QD surface, which is the site where proteolysis takes place. The selection of QD ligand coatings has always been one of the first considerations for the design of QD-based bioassays, but usually from the standpoint of colloidal stability and biocompatibility, without consideration for how it could alter the process being measured.  	   17	  In Chapter 2, this thesis investigates how QDs with four frequently used anionic, small molecule thiol ligand coatings affect that activity of thrombin and trypsin. Both enhancement and inhibition of proteolytic activity are caused by the selection of ligand coatings, and experiments indicate that interactions between the ligand-coated QDs and the target protease are the cause of this behaviour. Analysis of progress curves and kinetic parameters, characterization of physisorption, and kinetic modeling reveal that the four different QDs exhibit different behaviours towards trypsin and thrombin. In addition, a comparison between QD-based assays and homogeneous assays in bulk solution reveals a substantial enhancement in protease activity for the QD-based assays. Overall, this work provides new insight into the role of QD surface ligand coatings in mediating proteolytic activity, and how the selection of ligand coatings can help optimize analytical figures of merit for QD-based probes. In Chapter 3, this thesis evaluates how to extend the methodology of quantitative, QD-based proteolytic assays to fluorescence microscopy. Measurement of heterogeneous proteolytic activity is realized through the acquisition of QD donor and dye acceptor PL images using an epifluorescence microscope and different filter channels. Effects of excitation intensity, exposure time, and crosstalk are studied for their effects on quantitation. Homogeneous assays in the microtiter plate wells, and heterogeneous assays in a glass capillary and a hydrogel, are demonstrated. These initial studies lay a critical foundation for the future work aimed at quantitative protease assays with cells.   	   18	  Chapter 2: Enhancement and inhibition of proteolytic activity associated with selection of small molecule thiol ligand coatings on QD 2.1 Introduction Many probes based on semiconductor quantum dots (QDs) and other nanoparticle materials have been developed for biosensing and other biological applications.11, 29 Several of these bioprobes have been designed to detect the activity of enzymes. Among studies on hydrolase enzymes, a common format is to conjugate a central optically active nanoparticle such as a QD,30 gold nanoparticle (AuNP),31 or graphene oxide sheet,32 with a biomolecular substrate for the target enzyme. This substrate is often labeled with a dye or another nanoparticle at the opposite terminus such that energy transfer can occur between the central nanoparticle and the acceptor molecule.33 Hydrolytic activity then cleaves the substrate, which disconnects the central nanoparticle and the label resulting in a loss of energy transfer that can be tracked optically. In the case of QDs, Medintz et al.,25 Shi et al.,34 and Chang et al.27 reported the first QD-based probes for detecting the activity of proteases, and Gill et al. reported QD-based probes for detecting nuclease activity.35 Many other iterations and variations of this design have been reported since these reports (for example, refs.36 and many others). Although many biophysical studies on the activity of nanoparticle-enzyme conjugates toward substrates in bulk solution have been published (for a recent review, see ref.37), biophysical studies of the converse configuration, with substrate bound to the nanoparticle, have been relatively limited. To date, the interest in such probes has almost exclusively been their analytical utility; however, some recent studies on the kinetics of hydrolase activity toward nanoparticle-substrate conjugates have provided some interesting new insight. Prigodich et al. found that the activity of ribonuclease H (RNase H) toward multivalent AuNP-DNA/RNA heteroduplex conjugates was enhanced ~2-fold relative to the same substrate in bulk solution.38 The association of the RNaseH with the high-density of DNA oligonucleotides around the Au NP resulted in co-localization of the enzyme with its RNA substrate, leading to enhanced activity. Similarly, Algar et al. reported that the net proteolytic activity associated with multivalent QD-peptide substrate conjugates was 3-fold higher and that the classic Michaelis-Menten model was no longer 	   19	  applicable.28 A new model was put forward, where localization of multiple copies of peptide substrates and weak interactions between the protease and the coatings on the QD were responsible for the deviation of the proteolysis behavior, such as the enhancement of the proteolysis activity.  This chapter focuses on the kinetics of proteolysis associated with multivalent QD-peptide substrate conjugates that have been modified with four different anionic, small-molecule thiol ligand coatings. The results show that the selection of ligand coating can lead to both dramatic enhancements and inhibition of proteolytic activity, and the observations were linked to interactions between the enzyme and the QD interface. Two well-known serine proteases, trypsin (TRP) and thrombin (THR), were used as models in this study, with a Förster resonance energy transfer (FRET)-based configuration for tracking proteolysis. As illustrated in Figure 2.1, CdSeS/ZnS QDs with red emission (QD624) were coated with either cysteine (CYS), dihydrolipoic acid (DHLA), glutathione (GSH), or 3-mercaptoprionic acid (MPA), then self-assembled with multiple copies of a peptide substrate. The peptide, which contained cleavage sites for both TRP and THR, was labeled with a terminal Alexa Fluor 647 dye that functioned as an acceptor for Förster resonance energy transfer (FRET) with the QD donor. Proteolysis processes were recorded through the changes in the FRET-induced PL changes of both donor and acceptor.28 Depending on the choice of ligand coating on the QD and its adsorptive interactions with the protease, activity enhancements or inhibition relative to substrate in bulk solution are studied. Besides, the comparison between TRP and THR shows that the effects of the coating also depend on the physical properties of the protease. This work provides new insights into the mechanism of proteolysis associated with QD-peptide substrate conjugates, and more generally, nanoparticle-enzyme substrate conjugates. Because of the importance of proteases and other enzymes in numerous biological processes and as drug targets,39 such insights are important for the design of nanoparticle-based bioanalytical probes that exhibit improved sensitivity and selectivity toward their target enzymes.   	   20	   Figure 2.1 Schematic of proteolytic assays (A) Schematic of the QD-based proteolytic systems studied. CdSeS/ZnS QDs (peak PL at 624 nm) were coated with cysteine (CYS), dihydrolipoic acid (DHLA), glutathione (GSH), or 3-mercaptopropionic acid ligands (MPA), and self-assembled with a fluorescent dye (A647)-labeled peptide substrate. The dye served as an acceptor for FRET with the QD donor. (i) Interactions between thrombin (THR) or trypsin (TRP) and the QD-peptide conjugate led to (ii) protease-catalyzed hydrolysis of the peptide with loss of FRET. (B) Amino acid sequence of the peptide substrate. (C) Structures of the different small molecule thiol ligands used to coat the QDs.   2.2 Materials and methods 2.2.1 Materials  CdSeS/ZnS core/shell QDs (CytoDiagnostics, Burlington, ON, Canada) with red emission (peak PL at 624 nm; QD624) or green emission (peak PL at 524 nm; QD524) were made water-soluble by coating with one of four different small molecule thiol ligands: glutathione (GSH), cysteine (CYS), dihydrolipoic acid (DHLA), or 3-mercaptopropionic acid (MPA). The ligand exchange procedures are described in detail in Section 2.2.2 A synthetic peptide (Bio-Synthesis Inc., Lewisville, TX) with a polyhistidine tag was labeled at a terminal cysteine residue with Alexa Fluor 647 (A647)-maleimide (Life Technologies, Carlsbad, CA) as described previously.40 The peptide sequence was H6SP6GSDGNESGLVPRGSGC (written N-terminal to C-terminal) and is abbreviated as Sub(A647). Bovine thrombin (THR), bovine trypsin (TRP), and Nα-tosyl-L-lysine chloromethyl ketone hydrochloride (TLCK) were obtained from Sigma-Aldrich 	   21	  (Oakville, ON, Canada). Borate buffered saline (BBS; pH 8.5, 50 mM, 13.7 mM NaCl, 0.27 mM KCl) was prepared in-house and sterilized by autoclaving. 2.2.2 Preparation of ligand-coated QDs: CYS, DHLA, GSH and MPA Glutathione (GSH) and cysteine (CYS). As-received QDs in toluene (100 µL, ~0.8 nmol) were diluted with 800 µL of chloroform and then mixed with either 80 mg of GSH or 40 mg of CYS in 300 µL TMAH (25% w/w in methanol) in a glass vessel. The mixture was vortexed and let stand overnight at room temperature so that the aqueous and organic phases would separate. Borate buffered saline (BB2: 200 µL, 50 mM, 250 mM NaCl, pH 9.5) was added to extract the GSH/CYS-QDs. The organic layer was discarded and the QDs were purified from excess ligand with four rounds of precipitation with ethanol and centrifugation (4800 rcf, 4 min), followed by redispersion in BB2. The final pellet of QDs was separated from the supernatant and redispersed in borate buffered saline (BB1: 50 mM, 13.7 mM NaCl, 0.27 mM KCl, pH 8.5) and quantified via UV-visible spectrophotometry. Dihydrolipoic acid (DHLA). As-received QDs in toluene (100 µL) and neat DHLA (100 µL) were added to a glass vessel, mixed, and purged with argon. The reaction was brought to 70 °C for 4 h and then cooled to room temperature. Next, 200 µL of TMAH in methanol, 200 µL of borate buffered saline (BB3: 50 mM, 250 mM NaCl, pH 8.5), and 1 ml of chloroform was added in succession. The DHLA-QDs were dispersed in the aqueous layer and purified from excess ligand via four times of precipitation, as described for the GSH/CYS-QDs.  3-Mercaptopropionic acid (MPA). Neat MPA (200 µL) was added to as-received QDs in toluene (100 µL) in a glass vessel, followed by 800 µL chloroform and 300 µL of N, N-diisopropylethylamine (DIPEA). The reagents were mixed and let stand for 6 h at 60 °C under argon. The mixture was cooled to room temperature and then 200 µL of TMAH in methanol and 300 µL of BB3 was added to extract the MPA-QDs. The MPA-QDs were purified from excess ligand via four rounds of precipitation, as described for the GSH/CYS-QDs.  	   22	  2.2.3  Enzyme assays Thrombin was tested against four different ligand-coated QD624 seperately. Stock solutions of 0.4 µM X-QD624-[Sub(A647)]16 conjugates (where X = CYS, DHLA, GSH, or MPA) were prepared by first diluting 320N pmol of Sub(A647) (20 µM, 16N µL) with BBS (35N µL), then adding 20N pmol of X-QD624 (5.0 µM, 4N µL), where N was the number of samples to be assayed. The stock solutions were let stand for 30 min, analogous to calibration experiments, to enable assembly prior to the addition of protease. A series of solutions of either THR (0.04–20 µM, scaling by factors of two) or TRP (0.5–240 nM, scaling by factors of two) were prepared in the same buffer BBS. Aliquots (50 µL) of X-QD624-[Sub(A647)]16 conjugate solution were added to an equal volume of the series of THR/TRP solutions in a 96-well microtiter plate (100 µL final volume, 0.20 µM final concentration of QD624) followed by immediate initiation of the PL emission measurements. The PL intensity at 624 nm and 668 nm were collected at 2 min intervals over 4 h, corresponding to the QD624 and the A647 emission, respectively, using 400 nm excitation. Experiments with TRP were done similarly except that the highest concentration of the final TRP solution was 240 nM. 2.2.4 Preparation of calibration curves Calibration samples were prepared by mixing X-QD624 (X = GSH, CYS, DHLA, or MPA, in reference to the ligand coating on the QD) with N equivalents of native Sub(A647) and (16–N) equivalents of predigested Sub(A647) in BBS to a final volume of 100 µL and a QD624 concentration of 0.20 µM (20 pmol). The mixtures were let stand for 0.5 h prior to measuring their PL emission spectra between 500–750 nm.  Pre-digested peptide was prepared by incubating labeled peptides with THR or TRP (25 µM) overnight, at room temperature, followed by irreversible inhibition of the proteases with three separate 5 µL additions of 25 mM TLCK (in DMSO) at 60 min intervals. Separate calibration curves were generated for GSH-, CYS-, DHLA-, and MPA-QD624.  	   23	  2.2.5 QD inhibition assays Inhibition experiments used X-QD524 as an inhibitor and used GSH-QD624-[Sub(A647)]16 conjugates as a probe for THR activity. The final conjugate include three seperate aliquots, 33.3 µL thrombin solution (5 µM), 33.3 µL X-QD524 (6.0 nM–3.0 µM, where X = GSH, CYS, DHLA, or MPA), and 33.3 µL GSH-QD624-[Sub(A647)]16 conjugates (0.60 µM). The THR and X-QD524 were mixed first in the 96-well microtiter plate, followed by the addtion of the QD624-[Sub(A647)]16 conjugates. PL emission from the GSH-QD624 and A647 was measured at 1 min intervals for 4 h. The final concentrations of THR and GSH-QD624-[Sub(A647)]16 conjugate were 1.7 µM and 0.20 µM, respectively. Experiments with TRP were done similarly except that the concentration of the added TRP solution was 72 nM (final concentration of 24 nM). In both cases, two control samples without protease were run: (i) GSH-QD624-[Sub(A647)]16 conjugates, to correct for any non-proteolytic changes in PL over time, and (ii) a combination of  GSH-QD624-[Sub(A647)]16 and X-QD524, to ensure that the peptide substrates did not desorb from the original QD624 and reabsorb on the QD524, creating the false appearance of proteolysis. 2.2.6  Homogeneous assays Two different assays were done to measure the hydrolysis of Sub(A647) by THR and TRP in the absence of QDs. Method 1 added QDs for readout after a predefined incubation period for Sub(A647) and THR/TRP. Method 2 did not involve QDs and rather isolated native Sub(A647) from dye-labeled product fragments using affinity chromatography. For both methods, the effect that TLCK had on the PL ratio was corrected. Method 1. A series of Sub(A647) samples (23 µL, 6.4 µM each) were hydrolyzed by mixing with THR (23 µL, 20 µM) or TRP (23 µL, 1.9 µM) in BBS for a certain time followed by the termination of the proteolytic reaction by the addition of TLCK (5 µL, 25 mM in DMSO). TLCK was added to each sample in the series at a different time point: 2, 5, 8, 15, 20, 40, 60, or 80 min. The samples with TLCK were let stand for > 3 h, and GSH-QD624 were added (4 µL, 2.5 µM) to form QD-[Sub(A647)]x FRET pairs and 	   24	  generate FRET-induced readout, the A647/QD PL ratio. PL ratios were then converted to the number of peptides based on the calibration curves for the kinetic experiment. Method 2. A sample of Sub(A647) (420 µL, 6.4 µM) was mixed with THR (420 µL, 20 µM) or TRP (420 µL, 1.9 µM) in BBS. Aliquots (100 µL) were withdrawn from the sample at intervals of 2, 5, 8, 15, 20, 40, 60, and 80 min, and added to tubes with TLCK (5 µL, 25 mM in DMSO) to terminate the proteolytic reaction. Rhodamine B (5 µL, 0.46 µM) was then added to these aliquots as an internal standard. Next, Ni2+-nitrilotriacetic acid (Ni-NTA) agarose (~0.10 g) was mixed with the aliquots for 1 h. The aliquots with Ni-NTA were then centrifuged (1 min, 2 rcf), the supernatant was collected to measure their absorption spectra. The ratio of the absorbance values at 652 nm (Alexa Fluor 647) and 554 nm (Rhodamine B) was then compared to a calibration curve to determine the number of Sub(A647). The calibration curve was generated by diluting Rhodamine B (5 µL, 0.46 µM) and Sub(A647) (40 µM, 1–8 µL in 1 µL increments) with BBS (final volume of 100 µL).   2.2.7 Evaluating the effect of free ligand on proteolytic activity CYS, DHLA, GSH and MPA molecules were added to small molecule substrate solutions and tested with THR to determine if free ligand molecules that can potentially desorb from the QD surface can affect proteolysis. Aliquots (25 µL) of THR solution (10 µM) was mixed a series of dilutions of each of the four ligand molecules, with a high concentration of 160 µM, and dilutions scaling by a factor of 2. Aliquots (50 µL) small molecule substrate N-Benzoyl-Phe-Val-Arg 4-methoxy-β-naphthylamide hydrochlorid solution were added to the THR-X solution (X denotes CYS, DHLA, GSH and MPA) and the solutions were excited with 290 nm light and the emission was measured at 418 nm every 1.5 min for 100 min. Calibration samples were prepared by predigesting the small molecule substrate in 250 µM THR solution overnight, then diluting the predigested solution to make a series of different concentrations. In parallel, the native substrate molecules were prepared the same way except for the addition of THR. The two series of solutions were excited at 290 nm and the emission intensities at 418 nm were plotted versus the concentration. The kinetic PL data was converted to the product 	   25	  concentration according to the calibration curve and then fitted with exponential function. The first derivative of the progress curves was calculated at the starting point as the initial rate.  2.2.8 Evaluation of denaturation of protease on the QD surface X-QD525 was mixed with THR or TRP for 0, 20, 40 and 60 min at room temperature before adding GSH-QD624-[Sub(A647)]16 to initiate enzyme assays. The X-QD524 was used an inhibitor and potential cause of protease denaturation, and GSH-QD624-[Sub(A647)]16 conjugates were used as a probe for protease activity. The final concentration of X-QD525 was 50 nM. The final concentration of THR and TRP was 1.3 µM and 30 nM respectively.  2.2.9  Data analysis For each time point, t, in an enzyme assay or inhibition assay, the A647/QD624 PL intensity ratio, ρ, was calculated using equation 2.1, where IQD624 and IA647 are the PL intensities of the QD624 and A647 respectively, and I668 and I624 are the measured PL intensities at 668 nm and 624 nm. The PL intensity measured at 624 nm was exclusively from the QD624; the PL intensity measured at 668 nm, however, was composed of two contributions. One was from the A647 and the other was from the QD624. A correction factor, 𝜎668 (= 0.04–0.05; refer to Table 2.1) was used to account for the contribution from the QD624. Table 2.2 tabulates the correction factors.   (2.1)  Table 2.1 Crosstalk correction factors. Quantum Dot 𝝈668 CYS-QD624 0.049  DHLA-QD624 0.035 GSH-QD624 0.038 MPA-QD624 0.042 ρ = IA647IQD624 = I668 −σ 688I624I624	   26	  The PL ratios measured at time, t, and enzyme concentration, c, during an assay are denoted by ρ(t, c). For analysis, these PL ratios were normalized to control samples (no THR or TRP) according to equation 2.2, where  is the normalized PL ratio.  (2.2) The values of  were converted to the average number of substrate peptides per QD, N(t, c), by comparison to the calibration curves described above. The progress curves were fit with equation 2.3 using non-linear regression, where t0 adjusts the starting position of the curve to account for the delay between adding enzyme and starting PL measurements, k1 and k2 are empirical rates, A1 and A2 are empirical amplitudes, and N∞ is the residual peptide substrate per QD at long digestion times. The regressions were constrained so that (i) A1 + A2 + N∞ = N(0, 0) and (ii) all progress curves in a data set shared the same value of t0 (~2 min in most assays). Initial rates, v0, were extracted from progress curves by taking the derivative of eqn. 3 at time t = t0. 𝑁(𝑡, 𝑐) = 𝐴?𝑒–??(????) + 𝐴?𝑒–??(????) + 𝑁?  (2.3) 2.3 Results and discussion 2.3.1 QD-A647 FRET pair Quantum yield values for the X-QD624 (X = CYS, DHLA, GSH, MPA) and the corresponding Förster distances for energy transfer with A647 are listed in Table 2.2. Figure 2.2 shows the normalized absorption and emission spectra for the QD624 and the A647. QDs with peak emission at 624 nm were paired as donors with A647 as an acceptor. The spectral overlap integral for this FRET pair is 1.96×10–9 cm6 mol–1, which enables efficient FRET. A647 has its peak absorption at 652 nm with an extinction coefficient of ε = 250 000 M–1 cm–1. The emission peaks for the QD624 and A647 are 624 nm and 668 nm respectively, which well separated from each other and easy to resolve. With 400 nm excitation light, directly excited fluorescence from the A647 was ρ(t,c)ρ(t,c) = ρ(t,c)ρ(t,0) ρ(0,0)ρ(t,c)	   27	  negligible, which permits more sensitive tracking of proteolysis through changes in the ratio of FRET-sensitized A647 emission (peak at 668 nm) and FRET-quenched QD PL. Table 2.2 Quantum yields and Föster distances for the X-QD-A647 FRET pairs. Quantum Dot Quantum Yield Förster Distance with A647, R0 CYS-QD624 <0.1% 3.3 nm DHLA-QD624 1.6% 5.4 nm GSH-QD624 6.1% 6.7 nm MPA-QD624 0.3% 4.2 nm   Figure 2.2 Normalized absorption and emission spectra for QD-A647 FRET pair  2.3.2 Calibration curves Calibration samples were prepared by mixing X-QD624 (X = GSH, CYS, DHLA, or MPA, indicating the ligand coating on the QD) with N equivalents of native Sub(A647) and (16–N) equivalents of predigested Sub(A647) in BBS to a final volume of 100 µL and QD624 concentration of 0.20 µM (20 pmol). The mixtures were let stand for 0.5 h prior to measuring their PL emission spectra between 500–750 nm during which time, Sub(A647) and the polyhistidine-terminated fragment of its digest self-assembled to the surface of the QDs ZnS shell with high-affinity (Kd ~ 1 nM).41 The pre-digested peptides were included in the calibration in order to account for the non-specific adsorption of the dye-labeled fragment on the QD surface.28 Due to the different quantum yields for QDs 	   28	  with different ligand coatings, they exhibited different calibration curves, which are generated separately and shown in Figure 2.3 C.  Sub(A647) is hydrolyzed during proteolytic assays to yield unlabeled peptide fragments that remain bound to the QD, and A647-labeled peptide fragments that either diffuse away from the QD or non-specifically adsorb. Such adsorption can yield non-trivial background FRET. If FRET ratio data is to be converted into progress curves in terms of the number of Sub(A647), then it is necessary to account for any adsorption in calibration experiments. In a previous study, this accounting was done by first pre-digesting substrate peptide, then mixing together both the pre-digested and native substrates at N:(N0–N) ratios to generate calibration samples, where N0 is the initial number of substrate peptides per QD in a proteolytic assays and 0 ≤ N ≤ N0. Here, this method produces a mixture native and digested peptide fragments on each QD. The assumption inherent to this procedure is that, during assays, individual substrate peptides are hydrolyzed in separate encounters between the QD-substrate conjugates and protease. Figure 2.3 A(i) illustrates the conjugates expected from this mechanism of proteolysis at the beginning, midpoint, and end of an assay, and the corresponding points in a calibration curve. Figure 2.3 B(i) shows an example of a calibration curve for GSH-QD624 with Sub(A647) and N0 = 16. Analogous calibration curves for X-QD624 with X = CYS, DHLA, and MPA are shown in Figure 2.3 C.  The study that pioneered the above method of calibration ultimately suggested a so-called “hopping” model of proteolysis. In this model, a protease hydrolyzes all of the substrate peptide conjugated to a QD in a single encounter. Thus, at the midpoint of an assay, it would be expected that half of the QD-substrate conjugates are undigested and half are fully digested, as illustrated in Figure 2.3 A(ii). Previous studies have not proposed a calibration method that parallels this model. To this end, we mixed undigested QD-substrate conjugates with fully digested conjugates at ratios of 𝜒:(1–𝜒), where 𝜒 is the mole fraction undigested conjugates. Figure 2.3 B(ii) shows an example of a calibration curve for GSH-QD624 with Sub(A647) prepared in this manner. It is notable that the calibration data are nearly identical. Note that this agreement is expected theoretically when FRET ratios scale linearly with the number of acceptors per donor, as was the case 	   29	  for Sub(A647) with X-QD624. More importantly, this result indicates that two different mechanisms of proteolysis will yield the same apparent progress curves when measured via QD-FRET. For this reason, the progress curves shown here have two y-axis scales: a left-side scale that corresponds to N, the average number of Sub(A647) per QD624, for the model in Figure 2.3 A(i); and a right-side scale that corresponds to the concentration of X-QD624-[Sub(A647)]16 remaining, assuming the model in Figure 2.3 B(ii). The latter was a function of the 𝜒 value determined from the A647/QD624 PL ratio. As shown in Figure 2.3 C, the calibration curves of A647/QD624 PL ratio, ρ versus the number of Sub(A647) for all QDs with different coatings exhibit linear trends as expected. CYS-QDs are much different from the other three in terms of the slope of the plot. This result can be explained by theoretical expectation that  ρ = N(ΦA647/ΦX-QD624)(R0/r)6                                               (2.4) where Φ is a quantum yield value and N is the number of equivalent acceptors per QD. Since CYS-QD624 have the smallest quantum yield, the PL ratios are largest. The calibration curves are used to convert PL ratios into the number of Sub(A647) that have not been cleaved and are still attached on the QD, enabling the quantitative analysis of proteolysis rates. 	   30	                                  Figure 2.3 Calibration curves  (A) Two possible models of proteolysis and corresponding methods of calibration. (i) Digestion of Sub(A647) conjugated to X-QD624 through multiple encounters between each X-QD624 and protease, with hydrolysis of only a single Sub(A647) in each interaction. (ii) Digestion of Sub(A647) conjugated to X-QD624 through a series of interactions where protease hydrolyzes all of Sub(A647) conjugated to an X-QD624 in a single interaction. This model is referred to as “hopping.” (B) Calibration data for GSH-QD624-[Sub(A647)]16. For model (i), FRET ratios, ρ, were calibrated by mixing N and N0–N equivalents of Sub(A647) and pre-digested peptide fragments, respectively. For model (ii), FRET ratios were calibrated by mixing 𝜒 and 1–𝜒 equivalents of Sub(A647) and pre-digested peptide fragments, respectively. The same concentration of X-QD624 was used for both calibrations. (C) Calibration plots for the A647/X-QD624 PL ratio, ρ, as a function of the number of Sub(A647) per QD. Note that for samples with N×Sub(A647) per QD, there is also Nmax–N pre-digested Sub(A647) mixed into the sample, where Nmax = 16. The legend indicates the ligand coating on the QD (X = CYS, DHLA, GSH, or MPA). C 	   31	  2.3.3 Proteolytic assays with QD probes  X-QD624-Sub(A647)8 conjugates were used to detect THR or TRP proteolytic activity. The hydrolysis was tracked in real-time through the FRET-induced changes in QD donor and A647 acceptor PL. Sub(A647) contained a LVPRGS amino acid sequence, which is a well-known cleavage site for THR42 and, simultaneously, a cleavage site for TRP, with hydrolysis occurring at C-terminal to the arginine residue in both cases (see Figure 2.1).  Progress curves plotting the average number of peptide substrates per QD (and the concentration of X-QD624-[Sub(A647)]16 remaining) versus time are shown in Figure 2.4A as a function of enzyme concentration. Conversion from PL ratios to the number of peptides per QD (and the concentration of X-QD624-[Sub(A647)]16 remaining) were done through calibration experiments as described in Section 2.3.2. In agreement with expectations, more peptide substrates were digested as incubation time increased and, for each ligand coating, the rate of digestion scaled in proportion to the concentration of enzyme added. However, QDs with different coatings exhibited significant differences in hydrolytic rate, even for equal amounts of protease. Initial rates of hydrolysis of Sub(A647) by TRP and THR were calculated from progress curves (see Section 2.2.9 for details) and are plotted in Figure 2.4C. Both of these proteases had the same general trend in proteolytic activity as a function of ligand coating, with activity increasing in the order MPA < DHLA < CYS < GSH.  Despite the similar trend in relative activity between ligand coatings for TRP and THR, progress curves of these two proteases had several important differences. First, TRP activity was more robust than the THR activity. TRP had significant activity with all four ligand coatings, whereas THR exhibited much less activity when working with DHLA and MPA compared with CYS and GSH. Several smaller concentrations of THR that exhibited significant activity with the GSH and CYS coatings were completely inactive with the DHLA and MPA coatings. Further, for each ligand, the progress curves for TRP exhibited a much greater degree of digestion. All progress curves tended to converge if digestion times were long enough (i.e., approximately zero Sub(A647) per QD at long incubation times). In contrast, the progress curves for THR proteolysis often failed to 	   32	  converge or reach complete turnover. The general shape of the progress curves also differed between THR and TRP. The progress curves for THR activity often had a rapid initial phase of activity followed by a second phase that was almost a plateau with very little activity, with an obvious change in the digestion rate. In the contrast, the TRP had a more classic progress curve. These results suggested that TRP and THR had significantly different behaviours with the X-QD624-Sub(A647)16 conjugates. Our initial hypothesis was that THR had significant non-specific interactions with the QD, in addition to its selective interaction with the conjugated Sub(A647).    Figure 2.4 Progress curves and initial rates  (A) Progress curves for the digestion of X-QD-[Sub(A647)]16 by different concentrations of THR: 0 µM (control), (i) 0.040 µM, (ii) 0.080 µM, (iii) 0.16 µM, (iv) 0.31 µM, (v) 0.63 µM, (vi) 1.3 µM, (vii) 2.5 µM, (viii) 5.0 µM, (ix) 10 µM, and (x) 20 µM. (B) Plots of initial rates as a function of THR concentration, determined from mathematical fitting of the progress curves in panel A (see Section 2.2.9 and Equation 2.3). (C) Progress curves for the digestion of X-QD-Sub(A647) by different concentrations of TRP: 0 µM (control), (i) 0.5 nM, (ii) 0.9 nM, (iii) 1.9 nM, (iv) 3.8 nM, (v) 7.5 nM, (vi) 15 nM, (vii) 30 nM, (viii) 60 nM, (ix) 120 nM, and (x) 240 nM. (D) Plots of initial rates as a function of TRP concentration, determined from mathematical fitting of the progress curves in panel C.   	   33	  2.3.4 Adsorption of protease on QDs Electrophoretic mobility shift assays were done to investigate the possibility of non-specific interactions between X-QD624 and THR or TRP. Adsorption of protease on X-QD624 was expected to result in decreased mobility versus a QD-only reference sample because of the increase in size upon protein adsorption.43 Since the running buffer used for gel electrophoresis was TBE buffer pH 8.5, the X-QDs were all negatively charged, consistent with the anionic character of the deprotonated carboxylate groups of the ligands.  Tests were done where electrophoretic mobility shifts were measured as a function of THR and TRP concentration. Figure 2.5 and 2.6 show the results when GSH- and MPA-QD624 which were pre-incubated with either THR or TRP respectively at concentrations in the range 0.1–33 µM. Mobility shifts from THR adsorption on X-QDs became evident at enzyme concentrations of 0.1 µM for MPA, ~0.75 µM for DHLA and CYS, and 2.0 µM for GSH (Figure 2.5). This trend was largely in agreement with the THR digestion rates in Figure 2.4, where MPA-QD624 exhibited the slowest rates, DHLA-QD624 and CYS-QD624 yielded intermediate rates, and GSH-QD624 had the fastest rates. In the case of TRP, protease started to adsorb on MPA-QD624 and GSH-QD624 at higher concentrations than for THR, with changes in the mobility becoming evident at ~1 µM and ~6 µM, respectively (Figure 2.5). In contrast to THR, the adsorption of TRP appeared to result in aggregates of QD624 rather than a discrete shift of the band on the gel. Importantly, the result that TRP adsorbed more readily on MPA-QD624 than GSH-QD624 was in agreement with the slower rates of digestion with the former, consistent with the results obtained with THR. The overall lower tendency of TRP to adsorb to X-QDs was also consistent with the smaller effect that QD ligand selection had on the rates of TRP digestion.   	   34	                           Figure 2.5 Gel electrophoresis for THR Pseudocolour PL images of agarose gels (1.0%) showing changes in the electrophoretic mobility of CYS-, DHLA-, GSH-,  and MPA-QD624 (0.2 µM) with different concentrations of THR. The dashed line indicates the position of the sample wells; the arrow at the left indicates the polarity of the electric field (7.3V/cm). The gels clearly show that adsorption is more favored on MPA-QD624 than GSH-QD624.                    Figure 2.6 Gel electrophoresis for TRP Pseudocolour PL images of agarose gels (1.0%) showing changes in the electrophoretic mobility of GSH-,  and MPA-QD624 (0.2 µM) with different concentrations of TRP. The dashed line indicates the position of the sample wells; the arrow at the left indicates the polarity of the electric field (7.3V/cm). The gels clearly show that the gels clearly show that THR has a greater tendency to adsorb than TRP and adsorption is more favored on MPA-QD624 than GSH-QD624.  	   35	  Because of the apparent adsorption of THR on X-QD624, an important consideration was whether the adsorbed THR retained its structure or denatured. Attempts to characterize the adsorption of THR and TRP on QDs using circular dichroism spectroscopy were unsuccessful because of the overwhelming background UV absorption of the QDs. Measurements of tryptophan fluorescence, which can sometimes serve as a probe of protein structure,44 were also inconclusive. 2.3.5 Inhibition assays To better characterize the effect of QD ligand coatings and interfacial adsorption on protease activity, green-emitting X-QD524 bearing GSH, CYS, DHLA and MPA ligands were prepared without conjugated peptide and added to samples for FRET-based proteolytic assays with GSH-QD624-[Sub(A647)]16 as “inhibitors.” As illustrated in Figure 2.7A, the hypothesis was that if ligands on the QD promoted the adsorption of THR or TRP, then these non-substrate X-QD524 should act as inhibitors. The inhibition level would depend on the selection of ligand and its interaction with the target protease. Since QD524 has a PL peak at 524 nm and the spectra was narrow, it does not interfere with the energy transfer process and measurement of the QD624-A647 FRET pair. Because the QDs used in these experiments have an alloyed CdSeS core, QD624 and QD524 had similar sizes and surface areas for adsorption of protease despite of the difference in their emission spectrum. Among the four ligand coatings, GSH showed the least interaction with THR and TRP, and thus GSH-QD624 was selected as the substrate scaffold for the inhibition assays.  To realize the inhibition assays described above, the GSH-QD624-[Sub(A647)]16 conjugates and THR were used at a fixed concentration and increasing concentrations of X-QD524 were added as inhibitors. Figure 2.7B-C shows that, for THR, the rate of proteolytic digestion of GSH-QD624-[Sub(A647)]16 decreased as the concentration of each inhibitor X-QD524 increased. According to our hypothesis, the formation of adsorption complexes between QD524 and THR increased as the concentration of X-QD524 increased, resulting in a reduction of the activity of THR. Qualitatively, the observed degree of inhibition in Fig. 2.7B-C is in agreement with the observed rate of 	   36	  proteolysis in Fig. 2.4A, with GSH and CYS coatings on QD524 having the smallest inhibitory effect. As shown in Fig. 2.7D-E, analogous inhibition experiments were done with TRP and, as expected from the results in Figure. 2.4B, the various X-QD524 had a much smaller inhibitory effect on TRP activity. Even the highest concentrations of GSH- and CYS-QD524 had little effect on the TRP activity toward GSH-QD624-Sub(A647) conjugates. Although DHLA- and MPA-QD524 had some inhibitory effect, it was much less than with THR.   Figure 2.7 Inhibition assays (A) Schematic of the QD inhibition experiment. Protease (THR or TRP) can adsorb on GSH-QD624-[Sub(A647)]16 where it hydrolyze substrate, or protease can adsorb on X-QD524 where it cannot hydrolyze substrate. (B) Progress curves for the digestion of GSH-QD624-[Sub(A647)]16 by THR in the presence of different concentrations of X-QD524. (C) Plots of initial rates as a function of X-QD524 concentration, determined from mathematical fitting of the progress curves in panel B (see Section 2.2.9 and Equation 2.3). (D) Progress curves for the digestion of GSH-QD624-[Sub(A647)]16 by TRP in the presence of different concentrations of X-QD524. (E) Plots of initial rates as a function of TRP concentration, determined from mathematical fitting of the progress curves in panel D.   	   37	  To quantitatively analyze THR and TRP inhibition by the various QD coatings, equation 2.5 was applied. This equation was reported by Gray et al. for the analysis of acetylcholinesterase inhibition.45 In this model, kcat and Km are the turnover number and Michaelis-Menten constant, respectively, for the enzyme-substrate interaction; Vmax = kcat[E]0 is the limiting reaction velocity; [E]0 is the added enzyme concentration; [S]0 = 0.2 µM, it is the concentration of the QD624-[Sub(A647)]16 remaining; [I] is the concentration of X-QD524; and Kd is the dissociation constant of the inhibitory interaction between the target protease THR and the inhibitor X-QD524. We further made the assumption the QD624-[Sub(A647)]16 concentration [S]0 << the Michaelis-Menten constant Km, in which case equation 2.5 reduced to equation 2.6. The data were fit to equation 2.6. Apparent Kd values for THR were on the order of 101–102 nM, with Kd, GSH > Kd, CYS > Kd, MPA > Kd, DHLA. For TRP, Kd, GSH ≈ Kd, CYS > Kd, MPA ≈ Kd, DHLA, with the later two values on the order of 1 µM (see Table 2.3).                                              (2.5)  (2.6)   Table 2.3 Apparent specificity constants, kcat/Km for X-QD624-Sub(A647)16 conjugates, and apparent dissociation constants for X-QD524, estimated from initial rates of proteolytic digestion.   THR TRP  kcat/Km (min–1 µM–1) a Kd (µM) b kcat/Km (min–1 µM–1) a Kd (µM) b CYS-QD 0.057 ± 0.0054 0.21 ± 0.033 1.7 ± 0.031 -- DHLA-QD 0.035 ± 0.0044 0.034 ± 0.0070 1.4 ± 0.021 0.84 ± 0.093 GSH-QD 0.16 ± 0.028 0.51 ± 0.0000 2.2 ± 0.024 5.0 ± 1.6 MPA-QD 0.0033 ± 0.00067 0.099 ± 0.0000 0.39 ± 0.0020 1.1 ± 0.29 MM fit 0.0017 -- 0.033 -- a Derived from the data in Fig. 2 for X-QD624 with [S] = [*] = 0.20 µM, the concentration of X-QD624-[Sub(A647)]16. These values can be divided by a factor of 16 for the case that [S] = 3.2 µM, the equivalent concentration of  Sub(A647). b Derived from the data in Fig. 4 for X-QD524. The last raw derived from the bulk solution assay fitted with standard Michaelis Menten mechanism. v0 = Vmax[S]0Km 1+ [S]0Km−1 + [I ]Kd−1( )v0≈kcatKm( )[E]0[S]01+ [I ]Kd−1	   38	  2.3.6 Evaluating the effect of free ligand on proteolytic activity Thiol ligands on QD surfaces may have an adsorption–desorption equilibrium over time46. It was necessary to determine if the desorption of ligands from the QD surface caused the inhibitory effects observed in Figures 2.4 and 2.7. N-Benzoyl-Phe-Val-Arg 4-methoxy-β-naphthylamide hydrochloride is a small molecule substrate for THR that is only weakly fluorescent but upon digestion becomes brightly fluorescent at 410 nm with excitation at 290 nm. Various concentrations of free ligand molecules (CYS, DHLA, GSH, and MPA) were added to the THR and small molecule substrate assays to investigate the effect of free ligands on the proteolytic activity. Figure. 2.8A shows progress curves for the hydrolysis of different concentrations of the substrate. A high concentration of 40 µM for each ligand was tested, which is 200-fold larger than the QD concentration in kinetic experiments (Section 2.3.3). Progress curves for different concentrations of ligand have similar similar shapes, as do the curves for different ligands. Figure 2.8B shows the initial rate, 𝜐?, as a function of the concentration of ligand molecules added, indicating no significant inhibitory effects from free ligands. Rather, the data may indicate a small increase in activity with free ligand present.   Figure 2.8 Progress curves for THR proteolytic assays with small molecule substrates and initial rates (A)Progress curves for the digestion of small molecule substrate by THR in the presence of different concentrations of free ligand X (X=CYS, DHLA, GSH or MPA). (B) Plots of initial rates as a function of ligand molecule X concentration, determined from mathematical fitting of the progress curves in panel A.   2.3.7 Evaluation of denaturation of protease on the QD surface The X-QD525 was incubated with THR or TRP for 0, 20, 40 and 60 min before initiation of enzyme assays. The hypothesis was that, if THR or TRP were denaturing on the QD surface, part of the THR and TRP molecules became inactive during the incubation 	   39	  period and decrease the proteolytic activity observed afterwards. Figure 2.9A shows progress curves for the four different GSH-QD624-Sub(A647)12 conjugates after the incubating the enzyme with X-QD525 for different periods of time at room temperature. Figure 2.9B shows the corresponding initial rates. The initial rates dropped after a 20 min incubation, but remained relatively constant afterwards, indicating that the decrease in activity happened in the short time after the mixing  X-QD525 and enzyme.  Figure 2.9 Denaturation of enzyme on QD surface (A) Progress curves for the activity of THR and TRP with X-QD-Sub(A647) after incubation with green QDs for 0, 20, 40 and 60 min. (B) Summary of initial rates as a function of time and X-QD525.   2.3.8 Modeling In an attempt to better understand the TRP data in Figure 2.4, the progress curves were fit with five different kinetic models (A)–(E), as defined below. These fits were global fits 	   40	  with all parameters other than the enzyme concentration being shared between the individual progress curves for a given ligand coating. The fits are shown in Figure 2.11.  (A) Standard Michaelis-Menten model, where S represents X-QD624-[Sub(A647)]16 conjugates:  E + S k1k−1! ⇀!!↽ !!! ES  (2.7) ES k2! →! E +P  (2.8)   (B) Michaelis-Menten model with adsorption on the QD, where P represents X-QD624 with digested substrates (i.e., product). E + S k1k−1! ⇀!!↽ !!! ES  (2.7) ESk2⎯ →⎯ EP  (2.9) EP k−1k1! ⇀!!↽ !!! E +P  (2.10)  (C) The model in (B), with denaturation of the protease, D, induced by the QD surface.   E + S k1k−1! ⇀!!↽ !!! ES  (2.7) ESk2⎯ →⎯ EP  (2.9) EP k3! →! P +D  (2.11) EP k−1k1! ⇀!!↽ !!! E +P  (2.10) (D) The model in (B) with spontaneous denaturation of the enzyme, independent of the X-QD624. E + S k1k−1! ⇀!!↽ !!! ES  (2.7) ESk2⎯ →⎯ EP  (2.9) 	   41	  E k4! →! D  (2.12) (E) The model in (A) with denaturation of the protease, D, induced by the QD surface E + S k1k−1! ⇀!!↽ !!! ES  (2.7) ES k5! →! DS  (2.13) ES k2! →! E +P  (2.8)  For the TRP kinetic assays, the standard MM model was unable to fit the progress curves for all four X-QD524-[Sub(A647)]16 conjugates. The addition of adsorption on QD surface into kinetic model made the fitted curves match the data points more closely. With the addition of protease denaturation to the kinetic model, the fitted curves matched the data points well. The quality of fits is similar for model C and model D, suggesting that denaturation may be important to the measured progress curves, but that the progress curve data cannot distinguish between different mechanisms. The improvement from model A to D was indicated by the chi values of the fitting (Table 2.4). If adsorption was excluded from the model and only denaturation was considered, the fitting was not as good as with both denaturation and adsorption, suggesting that both denaturation and adsorption are important processes in the observed proteolytic activity. Table 2.4 and figure 2.10 compares the Kd values and rate constants from experiments and from models A, B and D. Table 2.4 Kd (µM) and rate constant for X-QD-Sub(A647). TRP  k1 k-1 k2 k4 Kd (µM) χ2 GSH A 0.360 7.06×10-9   3.92×10-9 1.54 GSH B 0.372 15.6 1590  8.39 1.54 GSH D 0.485 13.4 907 0.124 5.52 0.249 CYS A 0.198 6.82×10-9   6.88×10-9 5.61 CYS B 1.58 0.194 135  0.0246 3.32 CYS D 2.41 1.45 0.621 0.0221 0.120 0.989 DHLA A 0.196 5.79×10-9   5.90×10-6 3.19 DHLA B 1.25 0.215 6.16  0.0344 1.23 DHLA D 1.28 0.295 6.16 0.00549 0.0460 0.328 MPA A 0.694 0.11   0.648 3.24 MPA B 0.368 0.0374 83.8  0.235 1.17 MPA D 0.333 0.114 0.299 0.014 0.0321 0.161 	   42	    Figure 2.10 Comparison of different models  Trypsin progress curves and the fitting results using Model A, Model B and Model D, representing standard Michaelis-Menten mechanism, Michaelis-Menten mechanism with adsorption of protease on QD surface and Michaelis-Menten mechanism with both adsorption on QD surface and denaturation of protease by itself.  The THR data in Figure 2.4 were also fit with the same models (A)–(E). Represented fits are presented in Figure 2.11. A proper quantitative analysis was not achieved based on these models. But obvious improvements were observed when adsorption and denaturation were taken into account. The standard Michaelis-Menten were poorly fit. The mechanism with either adsorption or denaturation yielded a little improvement. The ones with both of these two showed the best fit, indicating that both denaturation and adsorption made a difference in the proteolysis activity. A B DTime (Min)Scaled Sub(A647)0 50 100 150 200 2500.00.30.60.9Time (Min)0 50 100 150 200 2500.00.30.60.9Time (Min)0 50 100 150 200 2500.00.30.60.9Time (Min)Scaled Sub(A647)0 50 100 150 200 2500.00.30.60.9CYSDHLAGSHMPA	   43	   Figure 2.11 Comparison of different models   Thrombin progress curves and the fitting results using Model A, Model B, Model C and Model D, representing standard Michaelis-Menten mechanism, Michaelis-Menten mechanism with adsorption of protease on QD surface, Michaelis-Menten mechanism with both adsorption and denaturation of protease on QD surface, and Michaelis-Menten mechanism with both adsorption and denaturation of protease on QD surface.  As noted in Section 2.3.7, an initial decrease in protease activity within 20 min of exposure to QDs has been observed experimentally with TRP and THR. However, continued denaturation of protease, as included in the kinetic models above, was not observed experimentally. This discrepancy may suggest that other factors contribute to the observed progress curves.   Time (Min)Scaled Sub(A647)0 50 100 150 200 2500.00.30.60.9Time (Min)0 50 100 150 200 2500.00.30.60.9Time (Min)0 50 100 150 200 2500.00.30.60.9Time (Min)Scaled Sub(A647)0 50 100 150 200 2500.00.30.60.9CYSDHLAGSHMPATime (Min) Time (Min)....A B C D	   44	  2.3.9 Proteolysis in homogeneous solution and on QD surfaces The results presented have demonstrated that the adsorption of proteases on QDs can strongly affect the kinetics of proteolysis. In particular, ligand coatings that promoted adsorption inhibited proteolysis relative to coatings that had weaker adsorption; however, these results were only discussed within the scope of proteolysis taking place around a QD conjugated with co-localized substrates. A more general baseline was needed for a thorough comparison. To this end, the rate of proteolysis associated with GSH-QD-Sub(A647)16 conjugates was compared with an equivalent amount of only Sub(A647) (no QDs) in bulk solution. To analyze proteolytic rates in bulk solution, substrate peptides were first hydrolyzed by the addition of THR or TRP for a certain time interval (2, 5, 8, 15, 20, 40, 60, and 80 min) followed by termination of the reaction with the addition of TLCK, an irreversible protease inhibitor. QDs were then added and any non-hydrolyzed substrate assembled to the QDs would form QD-Sub(A647) FRET pairs and provide a readout signal. The signal was the PL ratio, which was then analyzed similarly as in the kinetic experiments, where calibration curves with TLCK added were used to convert PL ratios to the number of peptides to generate progress curves. Each of these assays was done in triplicate. Compared to GSH-QD-Sub(A647) model, a much slower rate of proteolysis was observed when THR and TRP were mixed with Sub(A647) in bulk solution without QDs, as shown in Figure 2.12. The results of these experiments were confirmed using a second measurement method that did not involve QDs at any step. These results are also shown in Figure 2.12 and were in good agreement with the results of the first analysis. Overall, the rate of proteolysis of Sub(A647) by THR is faster with GSH-QD-Sub(A647) conjugates than with only peptide in bulk solution. The initial rate was enhanced by a factor of 8 (𝜈? in the bulk solution is 0.53 min–1 versus 4.11 min–1 on the QDs). Similarly, the rate of proteolysis of Sub(A647) by TRP was faster with GSH-QD-Sub(A647) conjugates than with only peptide in bulk solution. Assuming 𝜈? scales proportionally with TRP concentration, the QD enhances proteolytic activity by a factor of more than 26 (𝜈? was 0.7 min–1 in the bulk solution with 960 nM TRP and 4.5 min–1 on the QDs with 240 nM TRP).  Table 2.3 shows the apparent specificity constants, kcat/Km, which agrees with the protease assays. 	   45	   Figure 2.12 Progress curves for homogeneous assays Progress curves for the activity of (A) THR and (B) TRP with GSH-QD-Sub(A647) conjugates (blue circles) and with an equivalent amount of Sub(A647) in bulk solution (red symbols). The activity in the later case was determined by FRET-measurements with QDs after the enzyme reaction had been quenched (closed red circles) and by using a pull-down assay with an affinity resin and absorption spectrophotometry (open red circles). Protease concentrations are indicated in the figure. The progress curves for the GSH-QD-Sub(A647) conjugates are from the kinetic curves. The black line is the fit with standard Michaelis Menten mechanism.  2.3.10 Discussion and conclusions The results of this section have clearly shown the profound effect of the ligand coating on QD-based proteolytic activity. There appears to be an obvious affinity between the protease and the QD surface, which can cause both enhancement and inhibition of proteolytic activity. When compared to the bulk solution, where all substrate molecules distribute homogeneously in solution, the interaction between ligand coatings and protease provides a significant enhancement effect of the proteolysis, consistent with the “hopping” mechanism that has been suggested previously. When a protease molecule that is reversibly adsorbed to a QD rapidly digests all of its conjugated peptide substrates and then diffuses to another QD to repeat the process, an overall enhancement of activity is observed compared to discrete encounters with individual peptides in bulk solution.47 Here, enhancements were up to 10-times larger than those previously reported. However, 0 20 40 60 800510150123Time (min)Sub(A647) per QD 960 nM TRP + Sub(A647)240 nM TRP + GSH–QD–Sub(A647)0 20 40 60 80051015012310 µM THR + Sub(A647)10 µM TΗR + GSH–QD–Sub(A647)Time (min)Sub(A647) per QDΑΒSub(A647) (μΜ)Sub(A647) (μΜ)	   46	  when adsorption is too strong, smaller enhancements or even a decrease in activity can be observed. THR was more sensitive to ligand coatings than TRP, and this chapter showed that careful selection of the QD ligand coating can permit more sensitive detection of TRP and THR activity, and generate selectivity between TRP and THR (using MPA-QD624) even though these two enzymes hydrolyze the same substrate. Experiments suggest that adsorption of protease on QDs is partly responsible for the observed trends; however, kinetic modeling indicates that it is not the only factor. Denaturation of protease may be another contributing factor. Further studies will be necessary to completely understand proteolysis at the surface of QDs and other nanoparticles.   	   47	  Chapter 3: Proteolysis measurement using microscope Microscopy is widely used in biological applications; for example, to study protein interactions, Ca2+ signaling, and gene expression.48 Here, epifluorescence microscopy was used with FRET to image proteolytic activity. This basic feature is important for many prospective biological applications where different reactions may be happening in parallel and spatially heterogeneous activity needs to be recorded. Proof-of-concept for quantitative analysis using a method analogous to Chapter 2 is demonstrated in this chapter. 3.1 Materials and methods 3.1.1 Materials and QD-peptide substrate conjugates CdSeS/ZnS core/shell QDs (CytoDiagnostics, Burlington, ON, Canada) with green emission (peak PL at 525 nm; QD525) were made water-soluble by coating with glutathione (GSH), a small molecule thiol ligand. The ligand exchange procedure is described in detail in Section 2.1.2. Synthetic peptides (Bio-Synthesis Inc., Lewisville, TX) with a polyhistidine tag were labeled at a terminal cysteine residue with Alexa Fluor 647 (A647)-maleimide or Alexa Fluor 555 (A555)-maleimide (Life Technologies, Carlsbad, CA).40 The A647-labeled peptide sequence was H6SP6GSDGNESGLVPRGSGC-(A647) (N- to C-terminal) and is abbreviated as Sub(A647). The A555-labeled peptide sequence was H6SP6SGQGEGGN SDDDDKSGNGC-(A555) and is abbreviated as Sub(A555). Bovine trypsin (TRP) and tosyl-L-lysine chloromethyl ketone hydrochloride (TLCK) was obtained from Sigma-Aldrich (Oakville, ON, Canada). Borate buffered saline (BBS; pH 8.5, 50 mM, 13.7 mM NaCl, 0.27 mM KCl) was prepared in-house and sterilized by autoclaving. 3.1.2 Fluorescence microscopy Images were acquired using Olympus IX83 inverted epifluorescence microscope. Three fluorescence filter combinations were used to acquire fluorescence emission from QD525, A555, and A647. A 4×/0.16 (magnification/numerical aperture) objective lens, a120XL X-Cite metal-halide light source, and a 405/20x excitation filter were used for all 	   48	  measurements. A 520/40M band-pass filter (Chroma, Bellows Falls, Vermont, USA) was used for measurements of QD525 PL. The two values in the filter notation refer to the center wavelength and the approximate full-width-at-half-maximum, respectively. Thus, when this filter was being used during image capture, only light with wavelength in the range of ca. 500–540 nm was transmitted and detected by the sCMOS camera (C11440, Hamamatsu Photonics, Hamamatsu, Japan.).  Similarly, 565/30M and 665LP filters were selected to image A555 and A647 fluorescence respectively, transmitting emission of wavelengths from ca. 520–580 nm, and >665 nm, respectively (note: the 665LP filter is a long-pass filter). With such filters, emissions from the QD donor and dye acceptors can be resolved to some extent and measured quantitatively with the help of correction factors determined in calibration experiments (see Section 3.2.3).  3.1.3 Image acquisition and analysis software MetaMorph and ImageJ were used for acquisition and analysis of fluorescence images, respectively. MetaMorph Microscopy Automation & Image Analysis Software is designed for automated microscope image acquisition, device control, and image analysis. It provides the tools to control the switching between optical filters, excitation power, magnification, and all the other necessary configurations associated with the acquisition of fluorescence images using microscope. After the image acquisition, the images were exported from MetaMorph and analyzed in ImageJ. In ImageJ, all of the images were stacked together and thus the same area was able to be selected for all images. Fluorescence intensities were measured in ImageJ and used for further analysis. 3.1.4 Calibration of FRET pairs A series of QD-(A555)N conjugates and a series of QD-(A647)N conjugates were prepared (where N = 0, 1, 2, 4, 6, 8).  After initial measurement of the samples in the fluorescence plate reader (Infinite M1000, Tecan Ltd., Morrisville, NC, USA), the samples were transferred to a half-area clear bottom 96-well plate. The microscope was focused at the bottom of the well. For each conjugate, three images were taken using different filters corresponding to QD525, A555, and A647 emission, respectively. The images were acquired in MetaMorph and exported as raw .tiff files, followed by 	   49	  quantitative analysis using ImageJ to extract fluorescence intensities. Calibration curves of fluorescence intensity versus the number of acceptors were compared between the fluorescence microscope and the fluorescence plate reader. Crosstalk in each filter channel was also evaluated. 3.1.5 Homogeneous proteolytic assays Trypsin proteolytic assays were done by taking a series of microscope images with the A647 channel and the QD525 channel after mixing QD525-[Sub(A647)]8 conjugate with different concentrations of trypsin in the 96-well microtiter plate. To assay N (N=3) samples, 20N pmol of QD525-[Sub(A647)]8 was prepared in 55N µL of BBS by mixing each component at the desired stoichiometry and allowing self-assembly to proceed for 0.5 h. A series of dilutions of TRP (8 nM, 16 nM and 32 nM, 64 nM) was prepared in BBS and 50 µL of each was transferred to a 96-well plate in parallel with a BBS control sample. Aliquots of QD525-[Sub(A647)]8 (50 µL) were added and mixed with the TRP solutions. The final concentrations of TRP were half of the initial concentrations added to the wells. The plate was shaken briefly before imaging. All the parameters were set up beforehand so that microscope measurements were started immediately after the mixing. The figures were then exported and analyzed in ImageJ to obtain the fluorescence intensity in each channel and to calculate the A647/QD fluorescence intensity ratios after mixing QD525-[Sub(A647)]8 conjugates with different concentrations of protease.  Analogous assays were monitored using the fluorescence plate reader. After mixing the substrate and protease, the black bottom 96-well microtiter plate with analyte were placed in the fluorescence plate reader. PL were measured at 524 nm and 668 nm every 2 min.  3.1.6 Heterogeneous proteolytic assays Heterogeneous proteolytic assays were done by repeating the homogeneous experiments in a different setup where distribution of TRP concentration is applicable. A glass capillary and an agarose gel were used to realize the heterogeneous distribution. In the capillary model system, the TRP solution was dropped at one terminus of the capillary filled with QD-Sub(A647)16. Substrate hydrolysis took place when TRP mixed with the 	   50	  conjugates. QD525-[Sub(A647)]8 conjugates were injected into glass capillaries using capillary action and fixed on top of a microscope slide. A 2 µL aliquot of 20 µM TRP was dropped at the end of the capillary after all parameters were set up in MetaMorph, followed by an immediate initiation of image collection. Images were collected every 2 min for a duration of 30 min in each channel.  Analogous assays were done in an agarose gel matrix. The QD525-[Sub(A647)]8 conjugates were placed in the agarose gel using electrophoresis. The gel was poked with a glass capillary at the center of the QD525-[Sub(A647)]8 band. A 5 µL aliquot of 20 µM trypsin was placed in the resulting hole and gradually diffused from the center of the QD525-[Sub(A647)]8 band in all directions. Images were taken in the QD525 channel and A647 channel every 4 min for a duration of 30 min.                              Figure 3.1 PL spectrum and microscope filter channels The shaded regions represent the light allowed to go throught the three filter channels.  	  	  3.2 Results and discussion 3.2.1 Crosstalk between different channels Crosstalk is a fundamental challenge for multicolour imaging. It is the contribution from emission generated by other fluorophores that do not belong to the active channel, usually caused by the broad emission profiles of fluorescent dyes. Crosstalk must be identified and corrected for reliable quantitative interpretation of fluorescence intensity images. Figure 3.1 illustrates the PL spectrum of QD525, A555 and A647 as well as the microscope filter channels used in this work. 	   51	  Calibration samples were used to evaluate FRET and crosstalk. Images of QD-(A555)N and QD-(A647)N conjugates (N = 0, 1, 2, 4, 6, 8) were taken and analyzed in QD525, A555, and A647 channels. If no crosstalk from QD emission existed, the fluorescence intensity should have been 0 a.u. when N = 0 for A555 and A647; however, as shown in Figure 3.2 for the A555 channel, the initial fluorescence intensity is approximately 1300 a.u. for N = 0, and for QD-(A555)8 only increased to 2000 a.u., suggesting a significant contribution from QD525 emission in the A555 filter channel, such that corrections have to be made for this contribution from QD525. QD-(A647)N conjugates also show a similar contribution in the A555 channel. The fluorescence decreased as more A647 assembled per QD525, consistent with quenching of the QD due to FRET, indicating that this crosstalk mainly comes from QD emission instead of A647 emission.   The A647 channel started at a relatively low point, indicating a negligible contribution from QD525. Moreover, the QD-(A555)N PL intensity in the A647 channel was very low compared with that of QD-(A647)N, suggesting a small contribution from A555 in the A647 channel. In the QD525 channel, the intensity decreased with incremental increases in the number of acceptors due to FRET-based quenching. The QD525 channel was not expected to be disturbed by either A555 or A647 emission because the wavelength of the PL peak of A555 and A647 were longer emission (>550 nm).   Figure 3.2 FRET and crosstalk between different channels The fluorescence intensity in different channels as a function of the number of acceptors, A555 or A647, per QD. The red curves represent QD-A555N conjugates. The blue curves represent the QD-A647N conjugates. For each data set N = 0, 1, 2, 4, 6, or 8. 3.2.2 Effect of excitation energy and exposure time The image acquisition parameters of the microscope need to be optimized and understood for reliable data collection. For a FRET-based protease assay with several fluorescent 0 2 4 6 8500.01000.01500.02000.02500.0Sub(A647/A555) per QDFluorescence Intensity (a.u.) (A) A555 channel0 2 4 6 80.0500.01000.01500.02000.02500.0Sub(A647/A555) per QDQD–A555QD–A647(B) A647 channel0 2 4 6 80.0.1005.0.1031.0.1041.5.1042.0.104Sub(A647/A555) per QD(C) QD525 channelQD–A647QD–A647QD–A555QD–A555	   52	  dyes and QDs, the effect of excitation intensity and exposure time on measured PL ratios was investigated. Images of QD-(A555)N (N = 0, 1, 2, 4, 6, 8) and QD-(A647)N (N = 0, 1, 2, 4, 6, 8) conjugates were again acquired with each specific fluorescence filter channel. For each channel, a series of increasing excitation intensities (13%, 25%, 50% and 100% of the total lamp power) and a series of increasing camera integration times (25 ms, 50 ms, 100 ms, and 200 ms) were tested.  When the ratios of excitation intensities for two different channels were the same, for example, 25% for QD525 and 25% for A647, or 50% for QD525 and 50% for A647, the observed A647/QD525 PL ratios were nearly identical. In this chapter, the notation of A647/QD525 PL ratio represents the ratio of the mean values of fluorescence intensity of the images taken in A647 channel and in QD525 channel, respectively. When the excitation intensity was doubled for imaging in only the A647 channel, A647/QD525 PL ratio was doubled too, scaling proportionally to the excitation intensity ratio, as shown in Figure 3.3A. For the QD525 and A647 emission channels, the same excitation filters were used (ca. 420 nm), meaning that, for both channels, the QD was the only emitter that was excited by the incident light directly. The reason for using different excitation intensity in different channels is that QD emission is brighter than the A647 emission in most cases. In order to get comparable signal intensity between the channels, or to improve the signal/noise ratio for the A647 emission, a higher excitation intensity may be useful. Similarly, the ratio of integration times for these two channels can determine their relative sensitivity. With longer integration times, higher PL intensities were observed in both channels, which accounts for the A647/QD525 PL ratio being constant when the integration times for both channels scaled together by the same factor (see Figure 3.3B). The close agreement in the PL intensity ratios with different integration times but a common ratio of integration times between donor and acceptor channels suggests that integration times can be optimized without affecting quantitation in proteolytic assays as long as these parameters remain consistent throughout both the kinetic assays and calibration experiments, or are corrected with appropriate scaling factors.    	   53	    Figure 3.3 Calibration curves and images (A) Calibration plots for the A647/QD525 fluorescence intensity ratio (acquired using microscope and analyzed in ImageJ), as a function of the number of Sub(A647) per QD. All the figures were acquired with 50 ms integration time. The curves with the same incident light energy ratios almost overlap with each other. The rectangles at the right of the figure indicates the ratio of incident light in QD channel versus that in A647 channel and the time in the square brackets indicates the incident light energy in the QD channel. The colour matches the curves in the figure.  (B) Change of PL ratio A647/QD525 as a function of number of A647 per QD (acquired using microscope and analyzed in ImageJ) , as a function of the number of Sub(A647) per QD. All the results were acquired with 100% incident light. The curves with the same exposure time ratios overlap with each other. The rectangles at the right of the figure indicates the ratio of exposure time in QD channel versus exposure time in A647 channel and the time in the square brackets indicates the exposure time in the QD channel. The colour matches the curves in the figure. (C) Images taken using microscope in the QD525 and A647 channels, respectively. The numbers at the right top indicate the number of A647 per QD 100% excitation intensity and 200 ms integration time were used. Brightness was adjusted to better present the difference in intensity. The same brightness setup was used for the same pictures with the same channel.   3.2.3 Calibration using fluorescence microscopy: data collection and analysis As shown in Figure 3.3C, fluorescence images get brighter in the A647 channel and darker in the QD525 channel as more Sub(A647) assembled on the QD. With more acceptors, the donor fluorescence was quenched to a larger level, showing darker intensity in the images. Conversely, the increase in the number of acceptors leads to more efficient FRET-induced A647 fluorescence. Calibration curves acquired using the microscope and the plate reader presented an identical shape (see Figure 3.4). The close 	   54	  agreement of the calibration curves acquired with the fluorescence microscope and with the fluorescence plate reader validates the imaging method. This method also suggests that homogeneous kinetic experiments with the fluorescence plate reader can be replicated with the microscope and, presumably, the microscopy method extended to spatial resolution and quantitative analysis of heterogeneous proteolysis.              Figure 3.4 Comparison of calibration curves from plate reader and microscope The microscope images were collected with 50% excitation intensity and 100 ms exposure time.    3.2.4 Homogeneous kinetic experiment and comparison with fluorescence plate reader  TRP proteolytic assays were done by taking a series of microscope images with the A647 channel and the QD525 channel after mixing QD525-[Sub(A647)]8 conjugates with different concentrations of TRP in a 96-well microtiter plate. According to Figure 3.2, the crosstalk between measurements of QD525 and A647 PL is negligible, permitting tracking of proteolysis through changes of fluorescence intensities of the images in QD525 channel and A647 channel without correction. When Sub(A647) was hydrolyzed by TRP, FRET-induced A647 emission would be lost and the images taken in the A647 channel were expected to become darker and darker. At the same time, the FRET-quenched QD525 emission was expected to be recovered and brighter images were expected in the QD525 channel with longer incubation time. Progress curves in Figure 3.4A plot the A647/QD525 PL ratio versus time, as a function TRP concentration, derived from fluorescence imaging with the microscope (i.e., the ratio of the average intensity in the A647 channel to the average in the QD525 channel). Progress curves in Figure 3.4B plot changes in the A647/QD525 PL ratio measured with the fluorescence 2 4 6 80.000.020.040.060.080.100.000.060.120.180.24Sub(A647) per QDPL ratio A647/QD(Fluorescence plate reader)fluorescence plate readermicroscopyPL ratio A647/QD(Microscopy)	   55	  plate reader (i.e., the ratio of PL intensities measured at 668 nm and 524 nm) for analogous samples. In both cases, the TRP activity increased as its concentration was increased. The highest concentration of TRP (32 nM) resulted in a complete digestion in about 30 min in the plate reader and in about 35 min with the microscope. The intermediate concentration of TRP (16 nM and 8 nM ) digested approximately half of the Sub(A647) in 80 min in both cases. The lowest concentration of TRP (4 nM) digested 20–25% of Sub(A647) in both the plate reader and microscope measurements. The general trends in digestion rate correlated well with each other.     Figure 3.5 Proteolytic assays in microcope and plate reader Response of the FRET probe to different concentrations of TRP: (A) changes in the A647/QD525 fluorescence intensity ratios as a function of time acquired using microscope; and (B) changes in the A647/QD525 fluorescence intensity ratios as a function of time acquired using the plate reader.   3.2.5 Heterogeneous kinetic experiment using microscope The most important advantage of the fluorescence microscope versus the fluorescence plate reader is the ability to provide spatial information as well as temporal information. When a heterogeneous reaction occurs, the microscope is able to acquire the various processes taking place simultaneously at different positions, and thus tell the whole story of a heterogeneous biological process while a plate reader homogenizes all of those processes. Microscopy is therefore an essential tool for cellular studies.  An agarose gel and a glass capillary were used with TRP to test imaging of heterogeneous proteolysis. In the capillary model system, the TRP solution was dropped at one terminus of the capillary filled with QD-Sub(A647)8. Hydrolysis takes place when 0 20 40 60 800.100.150.200.250.304.07.010.0Time (min)PL Ratio A647/QD0 20 40 60 800.100.150.200.254.006.008.0010.00Time (min)(A) Microscope (B) Plate readerPL Ratio A647/QDSub(A647) per QDSub(A647) per QD8 nM4 nM32 nM16 nM	   56	  TRP is mixed with the conjugates. Since the TRP started from one terminus of the capillary and diffused to the other direction, the concentrations of TRP at different positions along the length of the capillary were heterogeneous, and changed with time because of diffusion. Hydrolysis happened from the end of the capillary and slowly moved to the other way. The hydrolysis at different positions of the capillary started at different times with different activities based on the concentration of TRP at the specific spot. Similarly, when a TRP solution was dropped on the top of an agarose gel with embedded QD-Sub(A647)8 the TRP diffused from the center out in all other directions. The hydrolysis also started from the middle and diffused outward.  Given that the TRP solution would diffuse along the capillary, hydrolysis of QD525-[Sub(A647)]8 conjugates  by TRP was expected to start at that end of the glass capillary and gradually move towards the other end. QD525-[Sub(A647)]8 substrate at different positions along the capillary gets digested at different times with a different rate. A series of PL ratio images were generated in MetaMorph to present the gradual process (See Figure 3.6A). The ratio images were scaled between 0 and 0.5. The capillary color starts to change from one end and spreads in the other direction indicating that the PL ratio gets smaller as trypsin moves from one end to the other. Three regions of interest (ROI) were selected and labeled in the ratio image. The progress curves of these three ROI are presented in Figure 3.6D. ROI1 was close to the end of the capillary where TRP was applied, so the proteolysis process starts from 0 min. However, there was a 4 min delay time before ROI2 started to show the proteolysis phenomenon, and the delay time increased to 12 min in ROI3. Note that the distance between ROI1 and ROI2 was larger than that between ROI2 and ROI3. However, the difference in delay time between ROI2 and ROI1 was smaller than the difference in delay time between ROI3 and ROI2. The reason is that the diffusion of TRP solution was slower as it moved further along the capillary. 	   57	   Figure 3.6 Progress images of proteolysis assays in capillary (A) Ratio images, (B) Images in the A647 channel, and (C) images in the QD525 channel of capillary. The stack of ratio images are arranged by time. Images were taken every 2 min. The first image was taken immediately after the mixing of the TRP and substrate. The arrow in panel A indicates the position where TRP was added. PL ratios were scaled between 0 and 0.5. Green indicates lower PL ratio and red indicates the higher PL ratio. The green portion diffuses along the capillary and moves toward the other direction, indicating the diffusion of the trypsin. (D) Progress curves for the three regions of interest (ROI) highlighted in the ratio image in panel A.   An aliquot of TRP solution was placed at the center of a QD525-[Sub(A647)]8 conjugate band in the agarose. TRP started to come in contact with the conjugate from the center hole and gradually diffused from the center of the QD525-[Sub(A647)]8 band in all directions. The same image collection and analysis were done as the capillary setup. A bright spot diffusing to all directions were observed demonstrating the spread of protease and proteolysis in the hydrophilic gel matrix (see Figure 3.7).          Figure 3.7 Ratio images of heterogeneous TRP assays in agarose gel Ratio images of gel. The stack of ratio images are arranged by time. The arrow in in the first image indicates the position where TRP was added. PL ratios were scaled between 0 and 0.5. Blue indicates lower PL ratio and red indicates the higher PL ratio.    3.2.6 Discussion and conclusions In this chapter, fluorescence microscopy was evaluated for a quantitative measurement of QD-based proteolytic assays, through the acquisition of QD and dye PL images using 	   58	  epifluorescence microscope and different filter channels. Homogeneous assays in the microtiter plate well were performed and compared with fluorescence plate reader. Heterogeneous formats were also tested to confirm the ability of epifluorescence microscopy to record FRET-based, spatially heterogeneous biological processes in real-time, and do so quantitatively in different matrices. Since quantitative analyses were done and agreed with the plate reader method, this microscope-based method of analysis can be implemented in future cell-based experiments, hopefully to realize quantitative measurements of proteolysis within cells. Microscope parameters such as the exposure time and excitation intensity can be optimized and calibrated, such that PL ratios from fluorescence images are robust and reliable, with minimal effect on data collection when compared to a fluorescence plate reader.   	   59	  Chapter 4: Conclusion and future prospects QDs are widely used for bioanalysis because of their unique physical and optical properties. These desirable properties include, but not restricted to, wide absorption spectra, narrow and symmetric PL spectra, better resistance to photobleaching, a large surface area-to-volume ratio, and chemically modifiable surface. The work in this thesis has focused on using the optical properties of QDs to investigate lesser-studied interfacial properties, to better understand the effect of the QD surface in biosensing. In chapter 2, proteolytic activity associated with the choice of small molecule ligand coatings on QD-substrate conjugates was investigated. These ligands exhibit different contributions to the proteolytic activity on the QD surface. Four common anionic ligands—GSH, CYS, DHLA and MPA—and two well-known proteases, TRP and THR, were used as model systems. FRET was used to track proteolysis. GSH-QD had the weakest interactions with the two proteases and supported a faster, more complete, and robust proteolysis process. In contrast, MPA-QDs had the strongest interactions with the two protease and had the slowest proteolysis and greatest inhibitory effects. THR was also found to be sensitive to the QD ligand coating than TRP, suggesting that QDs can exploit differences in the physical properties of the protease to induce selectivity where none existed previously. Experiments such as gel electrophoresis and inhibition assays suggest that the adsorption of proteases on QDs is important in determining their activity. A comparison of proteolysis on QDs and in homogeneous bulk solution showed that protease activity could be enhanced by more than order of magnitude on QDs. This chapter is a contribution to both biophysical and bioanalytical chemistry in that it helps elucidate the effect of the QD interface on enzyme activity and shows how these effects can alter sensitivity and selectivity in assays. In chapter 3, a fluorescence microscope was used for the FRET-based proteolysis assays and the effects of various parameters image acquisition were investigated. Results show that the fluorescence intensity of the images approximately scaled in proportion to the exposure time and excitation intensity, suggesting that optimization of imaging parameters and quantitative analysis were possible in parallel. FRET-based proteolysis assays were done using a fluorescence microscope and compared with the results 	   60	  achieved from a fluorescence plate reader. Both calibration data and the progress curves for proteolytic assays mirrored each other, validating the reliability of the microscope-based measurements. Emission crosstalk was evaluated and corrected for the analysis to resolve signals exclusively from QD525, A555 and A647 in QD-dye FRET pairs. The main advantage of a fluorescence microscopy assay over a fluorescence plate reader assay is the ability to resolve spatially heterogeneous processes. The preliminary work in this chapter has shown some very promising results. Future work will be focused on fluorescence microscopy-based proteolysis studies and assays, including adapting these microscopy methods to the new concept of concentric FRET, as well as the measurement of proteolytic activity associated with cultured cells.    	   61	  Bibliography   1. (a) Chan, W. C. W.; Maxwell, D. J.; Gao, X.; Bailey, R. E.; Han, M.; Nie, S., Luminescent quantum dots for multiplexed biological detection and imaging. Cur. Opin. Biotechnol. 2002, 13, 40-46; (b) Medintz, I. L.; Uyeda, H. T.; Goldman, E. R.; Mattoussi, H., Quantum dot bioconjugates for imaging, labelling and sensing. Nat. Mater. 2005, 4, 435-445; (c) Michalet, X.; Pinaud, F. F.; Bentolila, L. A.; Tsay, J. M.; Doose, S.; Li, J. J.; Sundaresan, G.; Wu, A. M.; Gambhir, S. S.; Weiss, S., Quantum Dots for Live Cells, in Vivo Imaging, and Diagnostics. Science 2005, 307, 538-544. 2. Alivisatos, A. P., Perspectives on the physical chemistry of semiconductor nanocrystals. J. Phys. Chem. 1996, 100, 13226-13239. 3. Nirmal, M.; Brus, L., Luminescence photophysics in semiconductor nanocrystals Acc. Chem. Res. 1999, 32, 407-414. 4. Algar, W. R.; Susumu, K.; Delehanty, J. B.; Medintz, I. L., Semiconductor Quantum Dots in Bioanalysis: Crossing the Valley of Death. Analytical Chemistry 2011, 83 (23), 8826-8837. 5. Smith, A. M.; Nie, S., Semiconductor Nanocrystals: Structure, Properties, and Band Gap Engineering. Acc. Chem. Res. 2010, 43, 190-200. 6. (a) Swafford, L. A.; Weigand, L. A.; Bowers, M. J.; McBride, J. R.; Rapaport, J. L.; Watt, T. L.; Dixit, S. K.; Feldman, L. C.; Rosenthal, S. J., Homogeneously alloyed CdSxSe1- x nanocrystals: synthesis, characterization, and composition/size- dependent band gap. J. Am. Chem. Soc. 2006, 128, 12299-12306; (b) Norris, D. J.; Yao, N.; Charnock, F. T.; Kennedy, T. A., High-quality manganese-doped ZnSe nanocrystals. Nano Lett. 2001, 1, 3-7. 7. (a) Chan, W. C. W.; Nie, S. M., Quantum dot bioconjugates for ultrasensitive nonisotopic detection. Science 1998, 281 (5385), 2016-2018; (b) Bruchez, M.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P., Semiconductor nanocrystals as fluorescent biological labels. Science 1998, 281 (5385), 2013-2016. 8. (a) Petryayeva, E.; Algar, W. R.; Medintz, I. L., Quantum Dots in Bioanalysis: A Review of Applications Across Various Platforms for Fluorescence Spectroscopy and Imaging. Applied Spectroscopy 2013, 67 (3), 215-252; (b) Zhang, Y. J.; Clapp, A., Overview of Stabilizing Ligands for Biocompatible Quantum Dot Nanocrystals. Sensors 2011, 11, 11036-11055; (c) Resch-Genger, U.; Grabolle, M.; Cavaliere-Jaricot, S.; Nitschke, R.; Nann, T., Quantum dots versus organic dyes as fluorescent labels. Nat. Methods 2008, 5, 763-775. 9. Mattoussi, H.; Mauro, J. M.; Goldman, E. R.; Anderson, G. P.; Sundar, V. C.; Mikulec, F. V.; Bawendi, M. G., Self-assembly of CdSe-ZnS quantum dot bioconjugates using an engineered recombinant protein. Journal of the American Chemical Society 2000, 122 (49), 12142-12150. 10. Smith, A. M.; Duan, H.; Rhyner, M. N.; Ruan, G.; Nie, S., A systematic examination of surface coatings on the optical and chemical properties of semiconductor quantum dots. Phys. Chem. Chem. Phys. 2006, 8, 3895-3903. 11. Mattoussi, H.; Palui, G.; Na, H. B., Luminescent quantum dots as platforms for probing in vitro and in vivo biological processes. Adv. Drug Delivery Rev. 2012, 64, 138-166. 	   62	  12. Algar, W. R.; Prasuhn, D. E.; Stewart, M. H.; Jennings, T. L.; Blanco-Canosa, J. B.; Dawson, P. E.; Medintz, I. L., The Controlled Display of Biomolecules on Nanoparticles: A Challenge Suited to Bioorthogonal Chemistry Bioconj. Chem. 2011, 22, 825-858. 13. Blanco-Canosa, J. B.; Wu, M.; Susumu, K.; Petryayeva, E.; Jennings, T. L.; Dawson, P. E.; Algar, W. R.; Medintz, I. L., Recent progress in the bioconjugation of quantum dots. Coord. Chem. Rev. 2013, DOI 10.1016/j.ccr.2013.08.030. 14. Boeneman, K.; Deschamps, J. R.; Delehanty, J. B.; Susumu, K.; Stewart, M. H.; Glaven, R. H.; Anderson, G. P.; Goldman, E. R.; Huston, A. L.; Medintz, I. L., Optimizing Protein Coordination to Quantum Dots with Designer Peptidyl Linkers. Bioconjugate Chem. 2013, 24, 269-281. 15. Yao, H.; Zhang, Y.; Xiao, F.; Xia, Z.; Rao, J., Quantum dot/bioluminescence resonance energy transfer based highly sensitive detection of proteases. Angew. Chem. Int. Ed. 2007, 46, 4346-4349. 16. Susumu, K.; Medintz, I. L.; Delehanty, J. B.; Boeneman, K.; Mattoussi, H., Modification of Poly(ethylene glycol)-Capped Quantum Dots with Nickel Nitrilotriacetic Acid and Self-Assembly with Histidine-Tagged Proteins. J. Phys. Chem. C 2010, 114, 13526-13531. 17. Green, N. M., Avidin and Streptavidin. Methods in Enzymology 1990, 184, 51-67. 18. Medintz, I. L.; Clapp, A. R.; Brunel, F. M.; Tiefenbrunn, T.; Uyeda, H. T.; Chang, E. L.; Deschamps, J. R.; Dawson, P. E.; Mattoussi, H., Proteolytic activity monitored by fluorescence resonance energy transfer through quantum-dot-peptide conjugates. Nature Materials 2006, 5 (7), 581-589. 19. Sahoo, H., Forster resonance energy transfer - A spectroscopic nanoruler: Principle and applications. Journal of Photochemistry and Photobiology C-Photochemistry Reviews 2011, 12 (1), 20-30. 20. (a) Algar, W. R.; Tavares, A. J.; Krull, U. J., Beyond labels: A review of the application of quantum dots as integrated components of assays, bioprobes, and biosensors utilizing optical transduction. Analytica Chimica Acta 2010, 673 (1), 1-25; (b) Medintz, I. L.; Mattoussi, H., Quatum dot-based resonance energy  transfer and its growing application in biology. Phys. Chem. Chem. Phys. 2009, 11 (17-45); (c) Algar, W. R.; Kim, H.; Medintz, I. L.; Hildebrandt, N., Emerging non-traditional Förster resonance energy transfer configurations with semiconductor quantum dots; Investigations and applications. Coord. Chem. Rev. 2013, DOI 10.1016/j.ccr.2013.07.015. 21. Sternlicht, M. D.; Werb, Z., How matrix metalloproteinases regulate cell behavior. Annual Review of Cell and Developmental Biology 2001, 17, 463-516. 22. Lopez-Otin, C.; Overall, C. M., Protease degradomics: A new challenge for proteomics. Nature Reviews Molecular Cell Biology 2002, 3 (7), 509-519. 23. Madala, P. K.; Tyndall, J. D. A.; Nall, T.; Fairlie, D. P., Update 1 of: Proteases universally recognize beta strands in their active sites. Chemical reviews 2010, 110 (6), PR1-31. 24. Whitney, M.; Svariar, E. N.; Friedman, B.; Levin, R. A.; Crisp, J. L.; Glasgow, H. L.; Lefkowitz, R.; Adams, S. R.; Steinbach, P.; Nashi, N.; Nguyen, Q. T.; Tsien, R. Y., Ratiometric Activatable Cell-Penetrating Peptides Provide Rapid In Viov Readout of Thrombin Activation. Angew. Chem. Int. Ed. 2013, 52, 325-330. 	   63	  25. Medintz, I. L.; Clapp, A. R.; Brunel, F. M.; Tiefenbrunn, T.; Uyeda, H. T.; Chang, E. L.; Deschamps, J. R.; Dawson, P. E.; Mattoussi, H., Proteolytic activity monitored by fluorescence resonance energy transfer through quantum-dot-peptide conjugates. Nat. Mater. 2006, 5, 581-589. 26. Shi, L. F.; De Paoli, V.; Rosenzweig, N.; Rosenzweig, Z., Synthesis and application of quantum dots FRET-based protease sensors. Journal of the American Chemical Society 2006, 128 (32), 10378-10379. 27. Chang, E.; Miller, J. S.; Sun, J.; Yu, W. W.; Colvin, V. L.; Drezek, R.; West, J. L., Protease-activated quantum dot probes. Biochem. Biophys. Res. Commun. 2005, 334, 1317-1321. 28. Algar, W. R.; Malanoski, A.; Deschamps, J. R.; Blanco-Canosa, J. B.; Susumu, K.; Stewart, M. H.; Johnson, B. J.; Dawson, P. E.; Medintz, I. L., Proteolytic Activity at Quantum Dot-Conjugates: Kinetic Analysis Reveals Enhanced Enzyme Activity and Localized Interfacial "Hopping". Nano Lett. 2012, 12, 3793-3802. 29. (a) Petryayeva, E.; Algar, W. R.; Medintz, I. L., Quantum Dots in Bioanalysis: A Review of Applications Across Various Platforms for Fluorescence Spectroscopy and Imaging. Appl. Spectrosc. 2013, 67 (3), 215-252; (b) Zrazhevskiy, P.; Sena, M.; Gao, X. H., Designing multifunctional quantum dots for bioimaging, detection, and drug delivery. Chem. Soc. Rev. 2010, 39, 4326-4354. 30. Kim, G. B.; Kim, Y. P., Analysis of Protease Activity Using Quantum Dots and Resonance Energy Transfer. Theranostics 2012, 2, 127-138. 31. Lee, S.; Cha, E. J.; Park, K.; Lee, S. Y.; Hong, J. K.; Sun, I. C.; Kim, S. Y.; Choi, K.; Kwon, I. C.; Kim, K.; Ahn, C. H., A near-infrared-fluorescence-quenched gold-nanoparticle imaging probe for in vivo drug screening and protease activity determination. Angew. Chem. Int. Ed. 2008, 47, 2804-2807. 32. Wang, H. B.; Zhang, Q.; Chu, X.; Chen, T. T.; Ge, J.; Yu, R. Q., Graphene Oxide-Peptide Conjugate as an Intracellular Protease Sensor for Caspase-3 Activation Imaging in Live Cells. Angew. Chem. Int. Ed. 2011, 50, 7065-7069. 33. Algar, W. R.; Kim, H.; Medintz, I. L.; Hildebrandt, N., Emerging non-traditional Förster resonance energy transfer configurations with semiconductor quantum dots: Investigations and applications. Coord. Chem. Rev. 2014, 263-264, 65-85. 34. Shi, L. F.; De Paoli, V.; Rosenzweig, N.; Rosenzweig, Z., Synthesis and application of quantum dots FRET-based protease sensors. J. Am. Chem. Soc. 2006, 128 (32), 10378-10379. 35. Gill, R.; Willner, I.; Shweky, I.; Banin, U., Fluorescence resonance energy transfer in CdSe/ZnS-DNA conjugates: Probing hybridization and DNA cleavage. J. Phys. Chem. B 2005, 109, 23715-23719. 36. (a) Huang, S.; Xiao, Q.; He, Z. K.; Liu, Y.; Tinnefeld, P.; Su, X. R., A high sensitive and specific QDs FRET bioprobe for MNase. Chem. Commun. 2008, 5990-5992; (b) Lowe, S. B.; Dick, J. A. G.; Cohen, B. E.; Stevens, M. M., Multiplex Sensing of Protease and Kitnase Enzyme Activity via Orthogonal Coupling of Quantum Dot Peptide Conjugates. ACS Nano 2012, 6, 851-857; (c) Kim, Y. P.; Oh, Y. H.; Oh, E.; Ko, S.; Han, M. K.; Kim, H. S., Energy Transfer-Based Multiplexed Assay of Proteases by Using Gold Nanoparticle and Quantum Dot Conjugates on a Surface. Anal. Chem. 2008, 80, 4634-4641; (d) Boeneman, K.; Mei, B. C.; Dennis, A. M.; Bao, G.; Deschamps, J. R.; 	   64	  Mattoussi, H.; Medintz, I. L., Sensing Caspase 3 Activity with Quantum Dot-Fluorescent Protein Assemblies. J. Am. Chem. Soc. 2009, 131, 3828-3829. 37. Johnson, B. J.; Algar, W. R.; Malanoski, A. P.; Ancona, M. G.; Medintz, I. L., Understanding enzymatic acceleration at nanoparticle interfaces: Approaches and challenges. Nano Today 2014, 9, 102-131. 38. Prigodich, A. E.; Alhasan, A. H.; Mirkin, C. A., Selective enhancement of nucleases by polyvalent DNA-functionalized gold nanoparticles. J. Am. Chem. Soc. 2011, 133, 2120-2123. 39. (a) Chapman, H. A.; Riese, R. J.; Shi, G. P., Emerging roles for cysteine proteases in human biology. Annu. Rev. Physiol. 1997, 59, 63-88; (b) Hedstrom, L., Serin Protease Mechanism and Specificity. Chem. Rev. 2002, 102, 4501-4523; (c) Nelson, A. R.; Fingleton, B.; Rothenbergy, M. L.; Matrisian, L. M., Matrix metalloproteinases: Biologic activity and clinical implications. J. Clin. Oncol. 2000, 18, 1135-1149; (d) Earnshaw, W. C.; Martins, L. M.; Kaufmann, S. H., Mammalian caspases: Structure, activation, substrates, and functions during apoptosis. Annu. Rev. Biochem. 1999, 68, 383-424; (e) Overall, C. M.; Kleifeld, O., Validating Matrix Metalloproteinases as Drug Targets and Anti-Targets for Cancer Therapy. Nat. Rev. Cancer 2006, 6 (3), 227-239. 40. Algar, W. R.; Blanco-Canosa, J. B.; Manthe, R. L.; Susumu, K.; Stewart, M. H.; Dawson, P. E.; Medintz, I. L., Synthesizing and modifying peptides for chemoselective ligation and assembly into quantum dot-peptide bioconjugates. Meth. Mol. Biol. 2013, 1025, 47-73. 41. (a) Aldeek, F.; Safi, M.; Zhan, N. Q.; Palui, G.; Mattoussi, H., Understanding the Self-Assembly of Proteins onto Gold Nanoparticles and Quantum Dots Driven by Metal-Histidine Coordination. ACS Nano 2013, 7, 10197-10210; (b) Sapsford, K. E.; Pons, T.; Medintz, I. L.; Higashiya, S.; Brunel, F. M.; Dawson, P. E.; Mattoussi, H., Kinetics of Metal-Affinity Driven Self-Assembly Between Proteins or Peptides and CdSe-ZnS Quantum Dots. J. Phys. Chem. C 2007, 111, 11528-11538. 42. Jenny, R. J.; Mann, K. G.; Lundblad, R. L., A critical review of the methods for cleavage of fusion proteins with thrombin and factor Xa. Protein Express. Purif. 2003, 31, 1-11. 43. Prasuhn, D. E.; Deschamps, J. R.; Susumu, K.; Stewart, M. H.; Boeneman, K.; Blanco-Canosa, J. B.; Dawson, P. E.; Medintz, I. L., Polyvalent Display and Packing of Peptides and Proteins on Semiconductor Quantum Dots: Predicted Versus Experimental Results. Small 2010, 6, 555-564. 44. Vivian, J. T.; Callis, P. R., Mechanisms of tryptophan fluorescence shifts in proteins. Biophys. J. 2001, 80, 2093-2109. 45. Gray, P. J.; Duggleby, R. G., Analysis of kinetic data for irreversible enzyme inhibition. Biochem. J. 1989, 257, 419-424. 46. Resch-Genger, U.; Grabolle, M.; Cavaliere-Jaricot, S.; Nitschke, R.; Nann, T., Quantum dots versus organic dyes as fluorescent labels. Nature Methods 2008, 5 (9), 763-775. 47. Algar, W. R.; Malonoski, A.; Deschamps, J. R.; Banco-Canosa, J. B.; Susumu, K.; Stewart, M. H.; Johnson, B. J.; Dawson, P. E.; Medintz, I. L., Proteolytic Activity at Quantum Dot-Conjugates: Kinetic Analysis Reveals Enhanced Enzyme Activity and Localized Interfacial "Hopping". Nano Letters 2012, 12 (7), 3793-3802. 	   65	  48. Sekar, R. B.; Periasamy, A., Fluorescence resonance energy transfer (FRET) microscopy imaging of live cell protein localizations. Journal of Cell Biology 2003, 160 (5), 629-633.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0167013/manifest

Comment

Related Items