Super-Resolution Optical Lithography with DNA

We scanned the surface with a 355 nm UV beam through a preset series of target spots that collectively form a three-letter pattern “PSU” (Fig. 3C)...
0 downloads 0 Views 966KB Size
Subscriber access provided by ECU Libraries

Communication

Super-Resolution Optical Lithography with DNA Shi Ho Kim, Yu Liu, Conner Hoelzel, Xin Zhang, and Tae-Hee Lee Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.9b01873 • Publication Date (Web): 19 Aug 2019 Downloaded from pubs.acs.org on August 20, 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 17 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

Nano Letters

Super-Resolution Optical Lithography with DNA

Shi Ho Kim, Yu Liu, Conner Hoelzel, Xin Zhang* and Tae-Hee Lee* Department of Chemistry, The Pennsylvania State University, University Park 16802

*Corresponding authors ([email protected] and [email protected])

ACS Paragon Plus Environment

Nano Letters 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 17

Abstract We developed an efficient, versatile, and accessible super-resolution microscopy method to construct a nanoparticle assembly at a spatial resolution below the optical diffraction limit. The method utilizes DNA and a photo-activated DNA crosslinker. Super-resolution optical techniques have been used only as a means to make measurements below the light diffraction limit. Furthermore, no optical technique is currently available to construct nanoparticle assemblies with a precisely designed shape and internal structure at a resolution of a few tens of nanometers (nm). Here we demonstrate that we can fulfill this deficiency by utilizing spontaneous structural dynamics of DNA hairpins combined with single-molecule fluorescence resonance energy transfer (smFRET) microscopy and a photo-activated DNA crosslinker. The stochastic fluorescence blinking due to the spontaneous folding and unfolding motions of DNA hairpins enables us to precisely localize a folded hairpin and solidify it only when it is within a pre-designed target area whose size is below the diffraction limit. As the method is based on an optical microscope and an easily clickable DNA crosslinking reagent, it will provide an efficient means to create large nanoparticle assemblies with a shape and internal structure at an optical super-resolution, opening a wide window of opportunities toward investigating their photophysical and optoelectronic properties and developing novel devices.

Keywords nanoparticle assembly, super-resolution microscopy, DNA-directed assembly, optical lithography, single-molecule FRET, smFRET

ACS Paragon Plus Environment

1

Page 3 of 17 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

Nano Letters

DNA has been utilized as a building material for macromolecular assemblies based on the programmability and reversibility of double-strand hybridization. More recently, a photolithographic technique utilizing UV-crosslinkable DNA has also been reported to chemically pattern a surface at a micrometer resolution1. The high efficiency of DNA crosslinking for this application, however, requires the DNA to be custom-synthesized so that a photo-activated crosslinker can be incorporated during synthesis2-3. Moreover, the spatial resolution of the patterning is subject to the optical diffraction limit. To break through the diffraction limit of the conventional optical microscopy, several super-resolution microscopy (SRM) techniques at a resolution of a few to few tens of nanometers have been reported. In particular, STochastic Optical Reconstruction Microscopy (STORM) and Photoactivated Localization Microscopy (PALM) are single-molecule localization methods that require stochastic blinking of fluorophores4-9. Unlike these methods based on photophysical blinking, DNA Points Accumulation for Imaging in Nanoscale Topography (DNA-PAINT) depends on fluorescence blinking induced by reversible DNA hybridization between immobilized DNA and the diffusing complementary DNA in solution10-13. We will refer to this blinking as “physical” blinking because the fluorescence signal is turned on by the physical process of DNA hybridization. The diffusing fluorophore-labeled single-stranded (ss-) DNA (imaging strand) transiently binds onto the immobilized ss-DNA that is fixed on a substrate to achieve physical and stochastic blinking which enables single-molecule localization for super resolution imaging. One drawback of the DNA-PAINT method is that the way to control the frequency of DNA hybridization events needs to rely on modulating the fluorescent DNA concentration. This approach may results in an elevated fluorescence background level and many non-specific binding events under some imaging conditions, consequently compromising the imaging resolution. Despite the revolutionary contributions of the above methods to various fields of science and engineering, no super-resolution technique has been reported to construct or pattern materials. To this end, we developed an efficient, versatile, and accessible optical method to construct and pattern materials below the diffraction limit and successfully demonstrated nanoparticle assembly at a super-resolution. The method is based on photo-crosslinkable DNA and single fluorophore localization, which we referred to as Super-resolution Optical Lithography with DNA (SOLiD). Our method employs a very simple way to conjugate a high efficiency photo-crosslinker to DNA and physical blinking of fluorophores due to DNA hairpin folding and unfolding14. By avoiding custom synthesis of photo-crosslinkable DNA and diffuse background signals from fluorescent particles in solution, SOLiD provides an efficient and versatile way to assemble nanoparticles in any desired shape and internal structure at an optical super-resolution. The DNA hairpin for SOLiD consists of three ss-DNA fragments (Fig. 1). The anchor strand is the stem part of the DNA hairpin labeled with biotin and a FRET donor Cy3 during synthesis (Integrated DNA Technologies Inc., Coralville IA). The anchor strand is immobilized on a microscope slide via the streptavidin-biotin conjugation (Fig. 1A)15. The pin strand is the loop of the DNA hairpin labeled with a FRET acceptor Cy5 during synthesis (Integrated DNA Technologies Inc., Coralville IA) and the DNA inter-strand crosslinker CNV (Fig. 2). The anchor (stem) – pin (loop) is bonded via hybridizing the bond strand (Fig. 1A) and will form a hairpin-like structure when folded. A folded hairpin undergoes unfolding whose frequency is determined mainly by the DNA hybridization energy or the base-pairing energy in the stem region. The melting temperature (Tm) of the double-stranded (ds-) DNA stem region is a measure

ACS Paragon Plus Environment

2

Nano Letters 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 17

of this energy. An unfolded hairpin can fold into a hairpin, frequency of which is determined mainly by the entropy of folding, a function of the length and flexibility of the loop region. Therefore, the folding and unfolding rates and the lifetimes of folded and unfolded states of the hairpin can be controlled with the lengths and structures of the stem and loop regions16-19. Representative time trajectories of single molecule fluorescence resonance energy transfer (smFRET)20 signals from two stem-loop Tm’s are shown in figure 1B. When the crosslinker is activated while the hairpin is in the folded state, the anchor and pin strands will be crosslinked to each other. The ds-DNA region where the bond strand connects the anchor and pin strands can be melted above its melting temperature so that the bond strand will be washed out while the pin strand will still remain if it is already crosslinked to the anchor strand (see the lower line of events in figure 1C). The bond and pin strands will be washed out together if the pin is not crosslinked to the anchor, leaving only the anchor strand on the site (upper line in figure 1C). The effector strand is a short DNA fragment labeled with biotin whose sequence is complementary to a part of the pin sequence. The effector is conjugated to nanoparticles that will be used to construct the designed nanoparticle assembly (see the last step of the lower line in figure 1C). We used fluorescent polystyrene nanoparticles (40 nm diameter, Thermofisher Scientific, Watham MA) coated with neutravidin for effector conjugation. An NHS-ester functionalized Cyanovinylcarbazole (CNV-NHS-ester) photo-activated DNA crosslinker shown in figure 2A was synthesized in a high purity (>90 %, see figure S1 for characterizations). This crosslinker was conjugated to a primary amine group (Fig. 2B) that is labeled at the pin DNA during synthesis (Integrated DNA Technologies, Coralville IA). Upon photo-activation, the CNV crosslinker undergoes a 2+2 cyloaddition reaction with a pyrimidine base (Fig. 2C) on the complementary DNA, crosslinking the pin to the anchor. As is shown in the UV-Vis spectrum of the CNV molecule and the CNV-DNA pin molecule in figure 2D, the conjugation reaction is nearly complete. According to the spectrum, 20 sec which can be easily achieved with strong commercial fluorophores, assuming 15 % photon detection efficiency at 720 nm. Photobleaching after 3 x 107 photon emission under an oxygen scavenging environment results in 2000 sec lifetime. With this lifetime, we have a large room to expand the focus to balance between precision and efficiency. For example, in order to avoid any simultaneous “on” events of two or three fluorophores, we may need to tune the unfolded lifetime to be three times longer, which will require three times longer photobleaching lifetime (= 60 sec). This is still far below the average photobleaching lifetimes of the fluorophores under the required FRET illumination intensity. Secondly, the surface coverage with hairpins is not uniform because we depended on random surface immobilization. We still have some control over the average surface density of hairpins by varying the surface density of biotin-PEG (Laysan Bio, Arab AL). We can mix PEG and biotin-PEG stoichiometrically according to a target biotin-PEG density. In order to have at least one hairpin within a target area of 20 x 20 nm2 for a target resolution of 20 nm, we will need to use an average surface density higher than one per 20 x 20 nm2. The above resolution estimation will still work at a higher surface density. Assume that we have a two-fold higher surface density of hairpins than one per 20 x 20 nm2. Under the same conditions described above, the chance of having only one hairpin turn on at one 100 ms moment would be halved. However, the doubled density means that we will have two hairpins within a target spot of 20 x 20 nm2, and proportionally, we will have the same chance of detecting one folded hairpin within a 100 ms moment. Therefore, we can use hairpins with the unfolded lifetime accordingly elongated to a given surface density in order to achieve the target resolution regardless of the absolute density. Cross-folding between two hairpins should not be a problem because we will still have a crosslinker DNA strand at the target site regardless whether it was formed by intra- or intermolecular folding. Lastly, hairpin folding and unfolding rates should be adjusted as discussed above. In order to control the folded state lifetime, one can vary the length and sequence of the hairpin stem region. With 5 ~ 10 base-pairs of ds-DNA (~1.7 ~ 3.4 nm), one can control the Tm between 10 ~ 40 oC. As for the unfolded state lifetime, one can control the sequence and the length of the hairpin loop region as well as the salt concentration of the buffer. It is impossible to predict precisely how these three parameters (sequence and length of DNA and the salt condition) will control the unfolded state lifetime. One needs to explore these parameters by trial and error based

ACS Paragon Plus Environment

5

Page 7 of 17 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

Nano Letters

on the general principles of DNA mechanics as a function of the length, sequence, and the salt condition. A more quantitative description on the resolution limit is given in Supporting Information assuming that the hairpin density and its folding kinetics are controlled not to be the limiting factor. According to the formula (Eq. S1), the current theoretical limit of the resolution with commercially available fluorophores is 5.4 nm with 1.4 numerical aperture for imaging, 100 ms signal integration, and 23 kHz photon detection rate. This photon detection rate is set to result in the fluorophore localization precision of 5.4 nm, thereby not limiting the SOLiD resolution. A potential application of SOLiD is to construct a noble assembly of optoelectronically active nanoparticles (e.g. 5 gold nanoparticles with a 50 nm diameter in a row with a