Single-Molecule Kinetic Investigation of Cocaine-Dependent Split

Oct 3, 2018 - In this work, we measure assembly kinetics of cocaine-dependent split-aptamer molecules using single-molecule fluorescence imaging...
0 downloads 0 Views 547KB Size
Subscriber access provided by Kaohsiung Medical University

Article

Single-Molecule Kinetic Investigation of Cocaine-Dependent Split-Aptamer Assembly Frances D Morris, Eric M Peterson, Jennifer M. Heemstra, and Joel M. Harris Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b03637 • Publication Date (Web): 03 Oct 2018 Downloaded from http://pubs.acs.org on October 5, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Single-Molecule Kinetic Investigation of Cocaine-Dependent Split-Aptamer Assembly Frances D. Morris,# Eric M. Peterson,# Jennifer M. Heemstra,≠ and Joel M. Harris#* #

Department of Chemistry, University of Utah, 315 South 1400 East Salt Lake City, UT 84112 USA



Department of Chemistry, Emory University, 1515 Dickey Drive, Atlanta, GA 30322 USA

ABSTRACT Aptamers are short nucleic-acid biopolymers selected to have high affinity and specificity for protein or small-molecule target analytes. Aptamers can be engineered into split-aptamer biosensors comprising two nucleic acid strands that co-assemble as they bind to a target, resulting in a large signal change from attached molecular probes (e.g. molecular beacons). The kinetics of split-aptamer assembly and their dependence on target recognition are largely unknown; knowledge of these kinetics could help in design and optimization of split-aptamer biosensors. In this work, we measure assembly kinetics of cocaine-dependent split-aptamer molecules using single-molecule fluorescence imaging. Assembly is monitored between a DNA strand tethered to a glass substrate and solutions containing the other strand tagged with a fluorescent label, with varying concentrations of the cocaine analyte. Dissociation rates are measured by tracking individual molecules and measuring their bound lifetimes. Dissociationtime distributions are bi-exponential, possibly indicating different folded states of the aptamer. The dissociation rate of only the longer-lived complex decreases with cocaine concentration, suggesting that cocaine stabilizes the long-lived aptamer complex. The variation in the slow dissociation rate with cocaine concentration is well described with an equilibrium-binding model, where the dissociation rate approaches a saturation limit consistent with the dissociationequilibrium constant for cocaine-binding to the split-aptamer. This single-molecule methodology provides a sensitive readout of cocaine-binding based on the dissociation kinetics of the split aptamer, allowing one to distinguish target-dependent aptamer assembly from background assembly. This methodology could be used to study other systems where target association affects the stability of aptamer duplexes. * Corresponding author: [email protected]

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 21

2 INTRODUCTION Aptamers are short biopolymers that have been refined through evolutionary selection to bind and detect small molecules, proteins, and other analytes.1 Nucleic acid aptamers combine high target-binding specificity and selectivity that rival antibodies2-3 with other advantages including ease of synthesis, surface attachment and labeling, and stability in a wider range of conditions.3-5 Sensor surfaces with aptamers attached can be dried and rehydrated, and exposed to a wide range of pH and temperatures while retaining activity, yielding low cost, robust biosensors with a long shelf life.6 Aptamers have been utilized in numerous detection schemes including electrochemistry,7-10 fluorescence,11-15 colorimetry,16 chemiluminescence,7 and surfaceplasmon resonance,17-19 paired with various biorecognition strategies including structureswitching aptamers6, 9, 14 and split-aptamer assays.20-21 Structure-switching aptamers include selfcomplementary nucleic acid sequences, giving rise to changes in secondary structure upon binding of target molecules. However, these sequences are relatively short and the secondarystructure shift within a single aptamer sequence can be challenging to detect and/or can lead to high levels of background signal and false positives.20 To address this challenge, researchers have developed split aptamers, where the aptamer sequence has been separated into two strands at a region that is not essential to target-analyte binding.20, 22-23 These two strands must hybridize and fold into the correct secondary structure to bind their target, leading to a dramatic and easily detected change in secondary structure. Successful split-aptamer probes have been developed through systematic evolution of ligands by exponential enrichment (SELEX),24-25 with subsequent modification of sequences having promising binding properties. SELEX uses multiple cycles of mutation, amplification, selection, and purification to select oligonucleotides with high affinities for their target molecule.18 Following the development of an aptamer biosensor, the reported binding constants determined through equilibrium studies typically provide little information about the mechanism and kinetics of aptamer assembly or analyte binding. While the thermodynamics of split-aptamer assembly have been thoroughly explored, the kinetics for this process remain largely unknown. Single-molecule fluorescence imaging has been used to measure the kinetics of association of a

ACS Paragon Plus Environment

Page 3 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

3 fluorescently-labeled target analyte to an intact monolithic RNA aptamer,26 but split aptamer systems present additional challenges in monitoring the assembly of a ternary system in which the capture and probe strands interact with each other and with a target analyte. In this work, we explore the kinetics of the cocaine-dependent DNA split aptamer association at different concentrations of cocaine to understand formation and disassembly of the ternary complex. The cocaine-dependent aptamer has been widely studied,16,

27-29

and a split-

aptamer variant was recently developed from a structure-switching cocaine-dependent aptamer through removal of bases located in a hairpin loop.11 Target sensing with this split aptamer has been achieved using click chemistry30 to ligate the split strands prior to detection using fluorescence or an enzyme for conversion of a substrate to a detectable product. This split aptamer is capable of detecting cocaine with a probe strand immobilized at an interface,31 and a similar split aptamer has been used with a detection strand modified with a Cy3 fluorescent label.20 Using a combination of these approaches, it should be possible to monitor hybridization between an immobilized probe strand and a labeled detection strand with single-molecule totalinternal-reflection fluorescence (TIRF) imaging.32 In TIRF microscopy, the excitation light is internally reflected at the glass coverslip-solution interface, allowing selective detection of aptamer molecules immobilized at the glass interface with a thin ~150-nm evanescent wave. Single-molecule fluorescence imaging is capable of measuring the kinetics of DNA association and dissociation events at equilibrium,33-34 without the need for introducing a concentration step, the speed of which can be limited by slow mass transport at the liquid-solid interface. In the present study, monitoring fast kinetics is necessary because of the rapid (less than 1 s) association lifetime of split aptamers. In addition, observing single-molecule events can reveal heterogeneous kinetics that would be otherwise obscured in an ensemble measurement.35 Using this methodology, we have measured the influence of cocaine concentration on the dissociation rate of the cocaine-dependent split aptamer. We find dissociation kinetics are heterogeneous, with fast- and slow- components, where the slow component is sensitive to cocaine

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 21

4 concentration. Increasing cocaine concentration stabilizes the long-lived aptamer complex, slowing the dissociation according to an equilibrium model that suggests cocaine exchanges rapidly during the life of the assembled aptamer complex. This kinetics-based assay provides sensitive readout of the bound-state of the split aptamer based on its dissociation kinetics, allowing us to distinguish background assembly from target-induced assembly by their sensitivity to cocaine concentration. These data provide new insight into the molecular-level functioning of split-aptamer sensors and may be generally applicable to oligonucleotide split aptamers and structure switching aptamers for a variety of target analytes. EXPERIMENTAL SECTION Materials. Oligonucleotide sequences, both labeled and unlabeled were synthesized using solid-phase phosphoramidite chemistry (Glen Research) and cartridge-purified by the University of Utah HSC Core DNA Synthesis Facility. Fluorescently-labeled DNA sequences were further purified by HPLC. The capture probe ssDNA sequences were 3’-hexadecylamine modified with a poly-adenine tail to space them from surface. DNA sequences are as follows, in 5’ to 3’ order: capture probe: GTT CTT CAA TGA AGT GGG ACG ACA AAA AA-NH2, detection strand: Cy3-PEG6- GGG AGT CAA GAA C; and a labeled non-complementary strand: Cy3-PEG6- TAT CAT CGA GAG. Capture probe ssDNA Immobilization. DNA immobilization techniques were adapted from previous work.34,

36

Slides were all cleaned before DNA immobilization using an acid

piranha bath (1:3 30% H2O2: 96% H2SO4, 20 min). CAUTION: Piranha solution is highly corrosive and reacts explosively with organic solvents; face-shields, gloves and other PPE should be worn. Following Piranha cleaning, slides were rinsed with ultrapure water and cleaned with RCA base bath (1:1:5 30% H2O2: 57% NH4OH: water, 70 °C, 20 min).37 After being rinsed and dried, coverslips were placed in a jar containing 250 µL (3-glycidyloxypropyl)trimethoxysilane (GOPTS). Jars were subsequently placed into the oven at 70 °C for 24 h to allow the silane to be deposited on the slides from the vapor phase. Slides were then transferred to a humid jar at 85 °C for 1 h, followed by 1 h in a dry oven at 120 °C to anneal and promote cross-linking of the silane

ACS Paragon Plus Environment

Page 5 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

5 film. Following GOPTS deposition, slides were laid flat in a 50-mL beaker. A 40-µL droplet containing 250-µM amine-modified capture-probe DNA and 50 mM carbonate buffer (pH 10.01) was sandwiched between the GOPTS-modified slide and a clean glass coverslip. These beakers where then placed into a jar containing excess carbonate buffer, sealed, and put into an oven overnight at 40 °C to allow the amine-modified DNA to react with the epoxide functionality on the silane layer. After being rinsed with water, remaining reactive epoxide groups on the GOPTS layer were passivated by a reaction with 3-amino-1-proanesulfonic acid (40 mM) in carbonate buffer (50 mM) for 8 h at 40 °C. TIRF Microscopy. Association of the labeled detection strand with the immobilize capture probe was monitored as single-molecule events using a TIRF Olympus IX-71 inverted microscope with home-built through-the-objective TIRF illumination. Glass coverslips with immobilized DNA were assembled in a flow cell with acrylic polyester double-stick gasket (145µm) and a glass top-plate. A channel (10 x 2 mm) was cut in a gasket connecting the inlet and outlet ports of the top-plate. A 532-nm laser (B&W Tek BWN-532-50E) was focused into a single-mode, polarization-maintaining optical fiber. Upon emergence from the fiber, the beam was collimated with a 60mm achromatic doublet lens before passing through a quarter-wave plate. Laser-light is then directed into the back of the microscope, where it is reflected off a dichroic mirror and focused with a 150mm achromatic doublet lens onto the back focal plane of the microscope objective (60x PlanApo N TIRFM 1.45NA, Olympus). The excitation light is collimated by the objective lens, and reflected off the glass-solution interface at an angle greater than the critical angle which generates a ~150 nm 38 evanescent wave on the solution side of the interface. This light excites fluorescence emission from molecules at the glass-solution interface and emission is collected by the same objective lens, passes through emission band-pass filter (centered at 582 nm), a 1.6X magnifier assembly (magnification increased to 96X), and is focused onto the electron-multiplying charge-coupled device (EMCCD) camera (Andor iXon DU897) for detection. Aptamer kinetic imaging. A buffer solution (25-mM sodium phosphate buffer: pH 8.1, 150-mM NaCl) containing varying concentrations of cocaine and the detection strand were

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 21

6 injected into the imaging flow cell and allowed to equilibrate for 1 min. Experiments were conducted at 19.5°C ±0.5°C unless otherwise noted. Samples were excited with 3.5-mW laser radiation, and images were collected continuously with 100-200-ms exposures at 180x electronmultiplying gain with read-out rate of 16 MHz in 256 x 170 pixel regions (42.7 x 28.3 µm). Image analysis and single-molecule tracking. Techniques used for image analysis are derived and modified from previous work.34,

39

Briefly, single-molecule spots were located in

fluorescence images using an intensity threshold which requires 3 adjacent pixels with intensity 3.2-times the standard deviations above background intensity. The first-moment of the intensity of the spot is then calculated to determine the centroid position. Once located, individual molecule coordinates are tracked in subsequent frames (with a spatial tolerance of 2 pixels, or 333 nm) to calculate how long the detection strand remains hybridized at the surface. This analysis provides an absolute count of the population and the dissociation lifetime of every hybridized detection strand. Cumulative histograms40 of the time-to-dissociate are used to determine the dissociation rate constant for the split aptamer-target complex, as described below. To check whether photobleaching of the Cy3 label contributes to the apparent duplex dissociation rate, a test of the impact of laser power on the measured dissociation times was carried out at laser powers 1.5- and 2-times greater than used to acquire the reported results. No reductions in apparent dissociation times were observed as the laser power was increased (see Supporting Information). Data fitting and error analysis. Cumulative histograms of single molecule survival times were fit to a double-exponential decay model in Equation 1 to determine parameters for the individual decay lifetimes. Fitting was performed using nonlinear least squares with squared residuals weighted by the statistical error, represented by 1/N(t) where N(t) is the histogram bin count at time bin t. Although the magnitude of the residuals from a two-exponential fit exceed the expected statistical uncertainty in the histogram bin counts, a more sophisticated model with three exponential components showed evidence of over-fitting the data. The two-exponential model was chosen to avoid fitting noise in the histogram and provide more robust parameters for the dissociation lifetimes. The first histogram bin (0.1 s) was discarded, since these single-frame

ACS Paragon Plus Environment

Page 7 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

7 events include spurious pixel-noise false-positive events. Histograms plotted in Figure 2 were normalized by the sum of the pre-exponential constant parameters from a fit of Equation 1. The slow rate parameters from these fits to Equation 1 were fit to the equilibrium model for the influence of cocaine on rates (Equation 2) using unweighted nonlinear least squares to determine parameters for the cocaine association constant and the limiting rates. RESULTS AND DISCUSSION Split aptamer design and detection

In

selectivity.

previously work by Stojanovic and

coworkers,11

dependent derived

split from

cocaine-binding

a

cocaine-

aptamer the

was

monolithic

aptamer

by

cleaving a hairpin loop at Position 1 to form a “capture probe” and a Scheme 1. Split-aptamer assembly. Left: cocaine split aptamer sequence; “detection strand” (see Scheme 1).

numbers indicate aptamer stem positions. Right: cocaine analyte stabilizes the assembled aptamer, but exchanges faster than aptamer dissociation.

In the present work, we have immobilized the capture-probe onto a glass surface modified with a reactive epoxide silane which binds the DNA through a 3’-amine A5 linker. Split-aptamer assembly of 3’-Cy3-labeled detection strands with immobilized capture-probe sites was monitored using single-molecule fluorescence microscopy. Buffered solutions containing the labeled detection strand and cocaine were flowed into a microscopy flow cell, and the assembled aptamer population at the interface was measured by detecting and tracking individual bound detection-strand molecules as the concentration of cocaine was varied. Unlike previous work by Heemstra and coworkers, there are no DNA modifications to ligate the two assembled strands at Position 2.4 As a result, assembly of the split aptamer is reversible, with association and dissociation of the two strands occurring on the 0.1 to 5-s time scale.

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 21

8 Because the simple single-label fluorescence imaging (Scheme 1) does not rely on fluorescence-resonance-energy-transfer (FRET) between two dye labels,41-42 two-color colocalization,43-44 or labeled origami structures,33,

45-46

to ensure selectivity of detecting true

hybridization events, it is imperative to test whether detection-strand events are specific to hybridization interactions with the capture-probe strand and are not due to non-specific adsorption. This issue was tested by measuring surface populations of fluorescently-labeled DNA molecules in a series of control experiments with varying concentrations of detectionstrand DNA, labeled non-complementary DNA, and cocaine. Molecule populations were counted in fluorescence images collected at equilibrium using an intensity-threshold algorithm described in the Experimental Section. Example images and average populations of molecules at the interface are shown in Figure 1. Detected populations on a capture-probe surface exposed to 1-nM labeled DNA, having no sequence complementarity with the capture-probe, were indistinguishable from buffer blanks and independent of the presence of cocaine. Exposure to 1-nM detection-strand DNA results in a >20-fold increase in the population of labeled DNA at the interface due to the specific formation of the split-aptamer complex.

From the relative

surface populations of labeled non-complementary

and

detection-strand

we

DNA,

conclude that more than 95% of events observed with the DNA detection strand result from split-aptamer assembly. Addition of cocaine to the 1nM detection-strand solutions significantly

increases

Figure 1. Labeled-DNA densities detected on the capture-probe surface for: 1) 1-nM noncomplementary ssDNA; 2) 1-nM noncomplementary ssDNA with 100-µM cocaine; (3) 1-nM detection strand; and (4) 1-nM detection strand with 100-µM cocaine. Above: representative images from the same data. Uncertainties on detected molecule densities are ~1.5%.

the

ACS Paragon Plus Environment

Page 9 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

9 equilibrium surface population of bound detection-strands (Figure 1) with a corresponding increase the average bound lifetimes (see videos in Supporting Information). These results suggest that cocaine stabilizes the split-aptamer complex and increases its bound lifetime (see below). The aptamer sequence in this work was previously studied in an scheme where cocaine induces assembly and ligation of the assembled split aptamer using ‘click’ chemistry; following incubation, the ligated aptamer could be detected by means an enzyme-linked assay.31 The aptamer chosen for the present study exhibited background ligation in the absence of cocaine, so that cocaine was detected in the presence of a background response. Other aptamer sequences having less complementarity in the stem regions exhibited smaller background ligation, leading to a more selective cocaine-dependent assembly-ligation response. However, we found that these less stable and more selective aptamer sequences formed short-lived duplexes even in the presence of cocaine; these events were difficult to detect by single-molecule fluorescence imaging at reasonable imaging rates for this instrumentation (