Subscriber access provided by University of Glasgow Library
Article
Multiplexed nucleic acid sensing with single-molecule FRET Anisa Kaur, Kumar Sapkota, and Soma Dhakal ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.8b01373 • Publication Date (Web): 20 Feb 2019 Downloaded from http://pubs.acs.org on February 21, 2019
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 27 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
ACS Sensors
Multiplexed nucleic acid sensing with single-molecule FRET
Anisa Kaur, Kumar Sapkota & Soma Dhakal* Department of Chemistry, Virginia Commonwealth University, 1001 West Main Street, Richmond, VA 23284, USA *Correspondence and requests for materials should be addressed to S.D. (email:
[email protected])
Abstract Multiplex detection of biomolecules is important in bionanotechnology and clinical diagnostics. Multiplexing using engineered solutions such as microarrays, synthetic nanopores, and DNA barcodes are promising but they require sophisticated design/engineering and typically yield semi-quantitative information. Single-molecule FRET (smFRET) is an attractive tool in this regard as it enables both sensitive and quantitative detection. However, multiplexing with smFRET remains a great challenge as it requires either multiple excitation sources, an antenna system created by multiple FRET pairs, or multiple acceptors of the donor fluorophore, which not only complicates the labeling schemes but also data analysis due to overlapping of FRET efficiencies (EFRET). Here, we address these currently outstanding issues by designing interconvertible hairpin-based sensors (iHabSs) with non-overlapping EFRET utilizing a single donor/acceptor pair and demonstrate a high-confidence multiplex-detection of unlabeled nucleic acid sequences. We validated the reliability of our approach by systematically omitting one target at a time. Further, we demonstrate that these iHabSs are fully recyclable, sensitive with a limit of detection of ~200 pM, and are able to discriminate against single base mismatches. The multiplexed approach developed here has the potential to benefit the fields of biosensing and diagnostics by allowing a simultaneous and quantitative detection of unlabeled nucleic acid biomarkers. Keywords: Multiplexing, Single-Molecule FRET, Recyclable Sensors, Single-Nucleotide Polymorphism (SNP), TIRF Microscopy 1 ACS Paragon Plus Environment
ACS Sensors 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
Multiplex assays allow for the simultaneous detection of multiple analytes, help improve the diagnostic capacity of testing, save time, expense, and other resources associated with the analysis.1-3 For this reason, multiplexing is an attractive technique, especially in the field of diagnostics and biotechnology where achieving multiplex detection of biomarkers can facilitate diagnosis.4 For example, it has been shown that the accuracy of diagnosis increases from as low as 65% to over 94% by measuring the level of 3 to 5 different types of biomarkers instead of just one.5-8 Although microarrays are the first technologies capable of parallel analysis of hundreds of analytes simultaneously from one sample, they are only semi-quantitative. For example, in fluorescence-based microarrays, an absolute intensity observed on a particular spot of a microchip does not provide quantitative information. This is mainly because, with the current stage of array technologies, it is not feasible to create calibration curves for hundreds of targets.9,10 Additionally, these parallel-array techniques including synthetic nanopores, barcodes, and force-based approaches are limited by the need for precise and sophisticated design/engineering.9,11-14 While there have been many strides in producing ultrasensitive, target-specific, and lowcost multiplex assays such as fluorescent microbeads and multiplex quantitative polymerase chain reaction (qPCR), these methods utilize ensemble measurements and are relatively simpler to operate, however, they typically suffer from significant false positives.15 In addition, the vast majority of these methods require that targets must be labeled, modified, or amplified to enable detection.16-18 Although other popular multiplexing methods such as surface enhanced Raman spectroscopy (SERS), electrochemical biosensors, and fluorescence-based techniques involving quantum dots can overcome this drawback by using labeled probes to detect unmodified targets,19-27 there is a limit to how many non-overlapping redox/fluorescent labels can be used to resolve the multiplexed data. Unlike commonly used ensemble techniques, single-molecule FRET (smFRET) provides a slew of information about the behavior of individual molecules, allowing a sensitive detection.28-32 However, the emerging paradigm of multiplexed sensing based on smFRET requires multiple excitation sources, complicated labeling-schemes such as “antenna” or “surplus” systems,18,33-38 and sophisticated numerical algorithms to analyze and interpret the data.39,40
2 ACS Paragon Plus Environment
Page 2 of 27
Page 3 of 27 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
ACS Sensors
Here, we developed a novel approach for multiplex detection of nucleic acids by rational designing of DNA-based smFRET sensors called interconvertible hairpin-based sensors (iHabSs), which allow multiplex detection by filling the FRET efficiency gaps (unutilized space) present in the conventional approach (Fig. 1a). The key advantage of this approach is that, regardless of the position of the sensors on the slide surface, the FRET efficiency (EFRET) determines the identity of the targets. Our iHabS is comprised of a DNA hairpin forming sequence sandwiched between two short pieces of double-stranded DNA (dsDNA), each labeled with either a donor or acceptor fluorophore (Figs 1b, 1c, and Supplementary Fig. 1). The unique design of the hairpin allows hybridization of a DNA probe to a portion of the hairpin forming sequence and toehold-mediated displacement of the probe by the target enables detection by increase in the EFRET value upon successful formation of a hairpin (Fig. 1b). We achieved 3fold multiplexing by using a combination of iHabSs with different inter-dye distances tuned by the length of flanking thymine-spacers and internal DNA-labeling at a different position of the DNA (Fig. 1b and c). In this study, we demonstrated simultaneous detection of three different targets (DNA sequences), producing three spectrally resolvable FRET efficiencies while still using a single FRET pair. Additionally, these sensors discriminate single-base mismatch sequences and exhibit many desirable features, for example they are fully-recyclable via the alternate addition of probe and target sequences (simple one-step conversion) allowing multiple rounds of detection and are highly sensitive with a detection limit down to picomolar (pM) concentration. Further, unlike expensive enzymes or antibody-based sensors, our iHabSs can be readily prepared from short synthetic DNA strands. With fine tuning of the buffer conditions, singlestranded spacer lengths, and utilizing both halves of the hairpins to recruit probes, these iHabSs have the potential for the simultaneous detection of more than three targets. Additionally, these iHabSs may find applications in biotechnology, diagnostics, and for the analysis of singlenucleotide polymorphism (SNP).23,41-43 Results Design, assembly, and characterization of sensors In this study, we designed and demonstrated a simple, sensitive, and fully-recyclable FRETbased multiplex detection platform to overcome current requirements of complex 3 ACS Paragon Plus Environment
ACS Sensors 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
Figure 1. Experimental design and single-molecule characterization of interconvertible hairpin-based sensors (iHabSs). (a) Left: conventional approach that allows the detection of only one target. Right: Our approach for simultaneous detection of multiple targets by filling the unutilized spaces (FRET efficiency gaps). P = probe, T = target. (b) The experimental setup for the smFRET analysis of iHabSs using prism-based total internal reflection fluorescence (pTIRF) microscopy (fig. not to scale) and working principle of iHabS. The probe bound iHabS with a low-EFRET state (open conformation) switches to a high-EFRET state (closed conformation) in the presence of target DNA forming a dsDNA by-product. iHabSs are designed to be recyclable upon alternate addition of probe and target. An alternate labeling scheme for Cy3 is highlighted and referred to as “INT”. Numbers 1-8 correspond to strands listed in 4 ACS Paragon Plus Environment
Page 4 of 27
Page 5 of 27 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
ACS Sensors
Supplementary Table 1 (1: BioStrand1 Comp, 2: Cy3-Rdm1, 3: Bio5’ Comp, 4: Strand1, 5: Hairpin (HPXX), 6: Probe (PX), 7: Cy5-Rdm2 Bottom Comp, 8: Cy5-Rdm2). (c) Hairpins (HP) with various flanking thymine spacers (represented as the number of thymine nucleotides “nt”) used to tune the EFRET values. The length of the thymine spacers (2nt to 6nt) are directly identified in the Figure (see Supplementary Fig. 1 for sequence detail of an iHabS).
labeling schemes and complicated data analysis algorithms which employs smFRET microscopy in multiplexing. While conventional smFRET detection techniques allow for the analysis of one target at a time, our approach utilizes the gaps between high and low-EFRET histograms (Fig. 1a) to allow simultaneous detection of multiple targets. For this, we used a combination of DNA hairpins with various lengths of flanking single-stranded sequences that are designed to be interconvertible between the open and closed conformations to enable detection via change in EFRET value. The sequence design, construction, and working principle of our iHabSs are shown in Figure 1. All iHabSs were prepared by thermal annealing of eight single-stranded DNA (ssDNA) oligonucleotides (Supplementary Table 1) in 1x TAE-Mg buffer (40 mM Tris, 20 mM acetic acid, 1 mM EDTA, 12.5 mM Mg2+, pH 7.4). Of these eight oligonucleotides only the hairpin and probe sequences vary depending on the target of interest. The hairpin in all iHabSs is comprised of a 6 base-pair (bp) stem and a 20-nucleotide loop (See Supplementary Fig. 1 for mFold-predicted structures). The 6-bp stem was specifically chosen after the analysis of 5, 6, and 7-bp stems as it allows; (i) the formation of a stable hairpin in the absence of probe and (ii) an efficient opening of the hairpin in the presence of probe (Supplementary Fig. 2). The assembly was confirmed by a slower migration of iHabSs compared to an ssDNA control in a 2% native agarose gel (Supplementary Fig. 3). To allow detection of targets by monitoring the EFRET in the absence and presence of targets, the donor (Cy3) and acceptor (Cy5) fluorophores were incorporated into the constructs using fluorophorelabeled complementary oligonucleotides (Fig. 1b). This was achieved by designing DNA probes that are complementary to a portion of the hairpin sequences so that a low-EFRET is maintained in the absence of targets. However, in the presence of targets, the probes are removed from the iHabSs by toehold-mediated displacement, leading to the formation of hairpins thereby switching the iHabSs from the low to high EFRET (detection). While all of the iHabSs share a 5 ACS Paragon Plus Environment
ACS Sensors 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
common low EFRET, to achieve an iHabS-specific high EFRET in multiplex detection, the hairpins were designed to have flanking thymine spacers of various lengths represented as the number of thymine nucleotides (Fig. 1c). First, we determined the performance of the iHabSs by smFRET experiments on our prismbased TIRF (pTIRF) microscope.44,45 Briefly, one of the oligos in each iHabS was biotinmodified (Supplementary Table 1) to enable surface-immobilization on the microscope slide which is coated with biotin-BSA and streptavidin (see Methods). The design of our flow cell is shown in Supplementary Fig. 5. Upon binding of the iHabS(s), the unbound molecules were washed off with an imaging buffer (1x TAE-Mg, pH 7.4) containing an oxygen-scavenging system (OSS) (see Methods for details).46,47 The OSS helps to retard photobleaching of the fluorophores upon laser-illumination. The fluorescence intensity traces were recorded at 10 frames per second (≈100 ms camera integration time) for both Cy3 and Cy5 while the microscope slide was illuminated only with the green laser (532 nm). The presence of fully assembled iHabSs were confirmed by direct excitation of Cy5 (red laser, 639 nm) towards the end of data acquisition. Only the molecules showing evidence for both Cy3 and Cy5 were picked manually for further FRET efficiency analysis. Our initial FRET efficiency analyses were carried out in a 1x TAE buffer containing 10 mM Mg2+ (pH 7.4). The FRET movies were processed using IDL and MATLAB codes (See Methods) and the FRET efficiencies were calculated using an established method as IA/(ID + IA), where IA and ID represent the background-corrected fluorescence intensities of the acceptor (Cy5) and donor (Cy3), respectively.28,48 As expected, all iHabSs show a low EFRET value (~0.2) in their open conformations regardless of the length of thymine-spacers and the fluorophore labeling schemes – terminal or internal (“INT”) (Figs. 1, 2 and Supplementary Fig. 6). It was expected that the EFRET values of iHabSs in their closed conformations rely on the spacer-length (the longer the spacer, the lower the EFRET), to our surprise, all of the terminally labeled iHabSs (HP22, HP34, HP45, and HP66) show no difference in the EFRET values (~0.80, Fig. 2a). This observation indicated that the inter-dye distance is apparently the same in these four iHabSs. We attributed this observation to compaction of single-stranded spacers at 10 mM Mg2+ due to electrostatic shielding of the negative charge on the DNA
6 ACS Paragon Plus Environment
Page 6 of 27
Page 7 of 27 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
ACS Sensors
Figure 2. smFRET characterization of iHabSs under different concentrations of Mg2+. (a) FRET efficiency histograms for all iHabSs at 10 mM Mg2+. Since the mean EFRET value of the open conformation was similar for all iHabSs, the FRET data were combined into a single histogram (gray, negative control). Each histogram at closed conformation is separately plotted and fitted with a single-peak Gaussian function before combining them. The shaded area highlights the unresolved EFRET peaks for HP22, HP34, HP45 and HP66 (“unresolved”). (b) FRET analysis of the iHabSs under the same buffer condition as in Fig. 2a except at 2 mM Mg2+. The shaded area highlights the resolved EFRET peaks (“resolved”) that were not resolved 7 ACS Paragon Plus Environment
ACS Sensors 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
at 10 mM Mg2+ in Fig. 2a. (c) The mean EFRET value for all iHabSs at their closed (circles: 10 mM Mg2+, squares: 2 mM Mg2+) conformations derived from Figs. 2a and 2b. The histograms were prepared in Origin by binning the mean FRET efficiency of each molecule as described in Methods section. The error bars in Fig. 2c represent the standard deviation in the mean EFRET values obtained after randomly assigning the molecules of given iHabSs into three groups. HP = hairpin and INT = internal labeling of the Cy3 fluorophore.
backbone, thereby nullifying the effect of spacers.49 The same experiments performed at a lower, more biologically relevant, concentration of Mg2+ (2 mM) show spacer-dependent EFRET of iHabSs (Fig. 2b), confirming that the 2 mM Mg2+ provides resolvable EFRET values for all six iHabSs. This may be due to the fact that although the thymine spacers exhibit an inherent flexibility due to their single stranded nature, ssDNA tethered to dsDNA regions have been shown to have a higher stiffness due to strong electrostatic repulsion from the duplex, particularly at low ion concentrations.50,51 It is important to note that, due to the presence of 1 mM EDTA in the 1x TAE buffer, the effective concentration of Mg2+ in these experiments can be
