Subscriber access provided by UNIV OF PITTSBURGH
Article
Highly Hybridizable Spherical Nucleic Acids by Tandem Glutathione Treatment and Polythymine Spacing Jing Sun, Dennis Everett Curry, Qipeng Yuan, Xu Zhang, and Hao Liang ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b00717 • Publication Date (Web): 29 Apr 2016 Downloaded from http://pubs.acs.org on May 3, 2016
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 free 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 accessible to all readers and 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.
ACS Applied Materials & Interfaces 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 16
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 Applied Materials & Interfaces
Highly Hybridizable Spherical Nucleic Acids by Tandem Glutathione Treatment and Polythymine Spacing Jing Sun,† Dennis Curry, ‡ Qipeng Yuan,† Xu Zhang,*,‡ Hao Liang*,† † State Key laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing, P. R. China. ‡ Verschuren Centre of for Sustainability in Energy and the Environment, Cape Breton University, 1250 Grand Lake Road, Sydney, Nova Scotia, Canada. * Corresponding authors. E-mail address:
[email protected].
[email protected].
Abstract Gold nanoparticle (AuNP)- templated spherical nucleic acids (SNAs) have been demonstrated as an important functional material in bionanotechnology. Fabrication of SNAs having high hybridization capacity to their complementary sequences is critical to ensure their applicability in areas such as antisense gene therapy and cellular sensing. The traditional salt-aging procedure is effective but tedious, requiring 1-3 days to complete. The rapid low-pH assisted protocol is efficient, but causes concerns related to nonspecific DNA adsorption to the AuNP core. To address these issue, we systematically compared the SNAs prepared by these two methods (salt-aging method and low-pH protocol). In terms of the number of complementary DNA that each SNA can bind and the average binding affinity of each thiolated DNA probe to its complementary strand, both methods yielded comparable hybridizability, although higher loading capacity was witnessed with SNAs made using the low-pH method. Additionally, it was found that nonspecific DNA binding could be eliminated almost completely by a simple glutathione (GSH) treatment of SNAs. Compared to conventional methods using toxic mercapto-hexanol or alkanethiols to remove nonspecific DNA adsorption, GSH is mild, cost-effective, and technically easy to use. In addition, GSH-passivated SNAs minimize the toxicity concerns related to AuNP-induced GSH depletion and therefore offer a more biocompatible alternative to previously reported SNAs. Moreover, rational design of probe sequences through inclusion of a poly-thymine spacer into the DNA sequences resulted in enhanced DNA loading capacity and stability against salt-induced aggregation. This work provides not only efficient and simple technical solutions to the issue of nonspecific DNA adsorption, but also new insights into the hybridizability of SNAs. KEYWORDS: Spherical nucleic acids, Gold nanoparticles, Poly-thymine spacer, Glutathione, hybridizability.
Introduction Gold nanoparticles (AuNPs) functionalized with single stranded oligo-DNAs of high density, named spherical nucleic acids (SNAs) by C. A. Mirkin and coworkers,1 have proven to be one of the most widely used materials in nanobiotechnology research since 1996,2, 3 where the physical and optical properties of AuNPs were coupled to the structural and chemical properties of DNA. Since then, an increased number of materials synthesis and biosensing applications have been demonstrated using SNAs, such as assembly of nanoparticle superlattices, gene knockdown and the monitoring of mRNA levels in live cells.4−10 Over the same time, many fundamental insights have been obtained using this hybrid nanomaterial.11-32 The merits of DNA−AuNPs include (i) high colloidal stability since the grafted DNAs provide charge and steric stabilization in high ionic strength solutions,23 (ii) excellent recognition capability due to the cooperative Watson−Crick base pairing mechanism,24, 25 and (iii) capability of coupling to other molecules for drug delivery or molecular imaging.26-29 Rational design of DNA sequences, including the incorporation of spacer elements and considerations related to charge, steric effects and ligand density have led to improvements in the efficiency of these nanoconjugates.33 To enable these applications, the first step is to fabricate SNAs by DNA attachment to the AuNP surface. Although nonthiolated oligo DNA strands can be attached to AuNPs to form functional SNAs via nitrogen-Au coordination,13-16 most work in this field is accomplished using AuNPs functionalized with thiolated DNAs
ACS Paragon Plus Environment
ACS Applied Materials & Interfaces
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
(SH-DNAs), where DNA strands are conjugated to AuNPs via thiol-Au bonds.2, 3 Since both DNAs and citrate-capped AuNPs are negatively charged under neutral pH,34, 35 electrostatic repulsion results in slow adsorption and low loading capacity by simple mixing of SH-DNAs with citrate-capped AuNPs. To increase DNA adsorption, the electrostatic repulsion between the DNA and nanoparticle surface must be reduced. Currently there are two approaches employed to achieve this goal. The first is to use electrolytes in solution, e.g., NaCl, to screen the negative charges between the DNAs and AuNPs, facilitating DNA adsorption.36, 37 However, salt induces irreversible aggregation of AuNPs in this process and therefore its addition must be delicately controlled to ensure DNA adsorption kinetics are faster than the aggregation kinetics of AuNPs. This strategy has been elegantly realized by the so-called “salt-aging” protocol , whereby salt is gradually introduced to a mixture of AuNPs and an excess amount of DNAs to form SNAs over 1-3 days.23 Although, sonication or heating may accelerate the DNA loading process, overnight incubation is still required.23 Later, Zu and Gao38 discovered that by adding a nonionic fluorosurfactant, AuNPs remained dispersed in the presence of 1 M NaCl and thus the DNA functionalization could be achieved within 2 h.39 Importantly, the orientation of the adsorbed DNAs were well controlled, significantly eliminating unwanted nonspecific adsorption. Similarly, we found that the commonly-used crowding polymer, polyethylene glycol (PEG), could stabilize AuNPs under high concentration of salt via depletion stabilization, which was exploited to fabricate SNAs in 2 h.40 Recently, J. Li et al improved this approach by using a combination of thiolated PEG and the surfactant Tween 20, allowing highly stable and specifc SNAs to be synthesized in 1.5 h.41 This method has been successfully applied to the functionalization of gold nanorods with SH-DNAs.30 In brief, all of the above-described methods are based on charge screening with salt while maintaining the colloidal stability using surfactants or polymers. Alternatively, without using surfactant, low pH buffer was found to accelerate the fabrication of SNAs.5 We found that by simply adjusting the pH of the solution of AuNPs to 3 with 10 mM citrate buffer,41 an ultrahigh density of SH-DNAs could be attached to AuNPs in a few minutes in a quantitative manner.42 Due to its simplicity and high efficiency, this method has been adopted to load DNA aptamers with secondary structures such as G-quadruplex for cancer treatment.31, 43 Admittedly, because of the extremely rapid adsorption kinetics, there is likely to be nonspecific adsorption44 to the AuNP core via the nitrogen atoms of the various nucleobases (Scheme 1),45 which can compromise the hybridization ability of the anchored DNA to its target. Therefore, it is desirable to quantitatively evaluate the nonspecific adsorption of thiolated DNA to gold and optimize the low-pH method to ensure the rapid production of highly hybridizable SNAs for target recognition applications.
Scheme 1. Nonspecific adsorption between nitrogenous ssDNA bases and AuNP core (left) and specific adsorption between the thiolated end of ssDNA and AuNP core (right).
ACS Paragon Plus Environment
Page 2 of 16
Page 3 of 16
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 Applied Materials & Interfaces
To address this issue, two effective approaches are proposed in the current contribution. First, rational sequence design, especially through introduction of a spacer sequence containing multiple thymine monophosphates (polyT), decrease the nonspecific DNA adsorption and increases the target binding kinetics. Second, treatment of as-prepared SNAs with glutathione (GSH) proves effective, similar to the use of 6-mercapto-1-hexanol (MCH). However, compared to the toxic and volatile MCH, nontoxic GSH showed comparable efficiency, was technically much easier to use and may be among the most bio-friendly passivation chemicals. Combining these two approaches, it was found that the hybridization efficiency of the SNAs could be significantly enhanced compared to controls. Therefore, this work provides a novel and facile means to fabricate highly hybridizable SNAs, and may find wide application potential in several areas of bionanotechnology and medicine, such as diagnostics, therapeutics, and imaging.
Results and Discussion Comparison of the SNAs prepared by salt-aging method and by low-pH method. Since thiolated DNA can be attached to AuNP surfaces either at pH 3 over several minutes to form SNAs (pH3-SNAs) or at neutral pH with NaCl over the course of 1-3 days (salt-aging-SNAs), a quantitative comparison of the SNAs prepared by these two methods may provide important mechanistic insights. First, we compared their loading capacity of the same set of DNA sequences with varied polyT spacer length (DNA 1-10, Table 1). For every SH-DNA sequence, AuNPs showed much higher loading capacity when using the pH 3 method compared to the salt-aging protocol (Fig. 1A). Second, despite the significantly lower DNA density of the salt-aging-SNAs, the number of fluorescent dye-labeled complementary DNA probes (FAM-cDNA) that hybridized on each salt-aging-SNA was slightly smaller than that on each pH3-SNA (Fig. 1B). As a result, the hybridization efficiency (h%, which is defined as the ratio of the number of hybridized SH-DNA on each AuNP to the total number of SH-DNA adsorbed on each AuNP) of pH3-SNA is significantly lower than that of salt-aging-SNAs: ~45% vs ~75% (p < 0.05, Fig. 1C). Nevertheless, it is crucial to note that in terms of the absolute number of cDNA hybridized to each individual SNA and average binding affinity (defined by the binding constant) of each AuNP-surface-immobilized DNA probe towards its target (Fig. S1), the pH3-SNAs demonstrated slightly greater performance than salt-aging-SNAs, which has been often underestimated in previous work. The low h% of pH3-SNAs may be explained in two ways. First, there may be more nonspecific DNA adsorption using pH3-SNAs. Second, the high DNA density of pH3-SNAs might lead to steric hindrance and a strong electrostatic barrier that prevents highly efficient DNA hybridization, as demonstrated previously.32, 37-39, 46 Based on P. S. Randeria’s work,32 such an electrostatic barrier (if any) could be overcome by increasing the concentration of NaCl during hybridization to screen the charge repulsion. Our results showed that for pH3-SNAs , further increase of NaCl significantly increased the hybridization efficiency (16-19% in Fig.2 and Table S1) and the number of hybridized cDNA (Fig. S2), but insignificant increases were witnessed for salt-aging SNAs (2-4%). This is due to the comparatively high DNA density of pH3-SNAs, which resulted in more pronounced electrostatic barriers for complementary strands than salt-aging SNAs. However, even with 400 mM NaCl, the greatest h% for pH3-SNAs was just ~70%. Based on the trend revealed by Fig. 2B, further increasing the NaCl concentration may not be able to increase h% significantly, suggesting the presence of a barrier due to either steric hindrance or nonspecific DNA adsorption. For salt-aging SNAs, the highest h% was ~85% for SH-DNA with a spacer of 7T (DNA 5); considering the relatively low DNA density of these SNAs, i.e., ~38 SH-DNAs/AuNP, the steric hindrance for cDNA strands was negligible (which was confirmed by data in Fig. 3, vide infra). Therefore, the nonspecific adsorption of the thiolated DNAs accounted for ≥15% loss of h%. It is worth noting that the insignificant increase of h% for salt-aging-SNAs in our work was different from the results in P. S. Randeria’s work, where a significant increase of h% was indeed observed, even for the complemtary DNA having the same length (9-mer). This discrepancy was reasonable considering that our 9-mer cDNA bound the adsorbed thiolated
ACS Paragon Plus Environment
ACS Applied Materials & Interfaces
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
DNAs from their outer end, while Randeria’s 9-mer cDNA bound to the middle of the thiolated DNAs within each SNA, where much more pronounced steric/electrostatic hindrance was expected. The data in Fig. 1A also revealed the effect of the length of the polyT spacer on DNA loading capacity. Regardless of which method was used, the loading capacity followed a similar pattern: DNAs with 3T spacer adsorbed most on AuNP surfaces, attributable to the combinatory effect of two opposite trends. On one hand, due to the weak interaction of citrate-capped AuNPs with thymine compared to other nucleobases,47, 48 the polyT spacer leads to the adoption of an upright conformation49 by each ssDNA on the AuNP surface thus decreasing the steric effect of adsorbed DNAs and increasing the loading capacity of DNAs relative to those DNAs lacking the polyT spacer. On the other hand, considering the large electronegativity of thymine compared to other nucleobases,47, 50, 51 longer polyT spacers increase the net negative charge of each DNA molecule and their intermolecular repulsion thereafter. In addition, the longer polyT spacers tend to form random coiled structures that may not only shed the thiol functionality of the DNA but also form larger footprints on the AuNPs surface,16, 52 collectively decreasing the loading capacity. Consistent with P. S. Randeria’s work, we observed that the SNAs with higher SH-DNA loading and hybridized cDNAs (i.e., 3T-spaced sequence) showed slightly lower thermostability in terms of the cDNA melting (Fig. S3). In addition, compared to polyT-spaced SH-DNAs, the widely used 5A-spaced SH-DNAs showed lower hybridization efficiency (Fig. 1C), attributed to even higher density packing DNA on AuNP surfaces resulting from high loading capacity and thus less accessibility by incoming FAM-cDNA probes. As illustrated later, in addition to the highest loading capacity of SH-DNAs with the 3T spacer, their binding affinity (Fig. S1) and hybridization kinetics to the complementrary sequence was also the highest, which is reasonable considering the favorable conformation of the 3T-spaced SH-DNA sequences on the AuNP surface.
Figure 1. Comparison of DNA loading and hybridization capacities on 13-nm AuNP surfaces using the salt-aging method and low-pH method, respectively, as a function of the length of polyT spacer sequences: (A) The loading capacity, (B) the number of hybridized cDNA, and (C) hybridization efficiency on each AuNP surface.
ACS Paragon Plus Environment
Page 4 of 16
Page 5 of 16
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 Applied Materials & Interfaces
Figure 2. Comparison of the hybridization efficiency of SNAs prepared via either the (A) salt-aging method or (B) low-pH (pH3) method towards the complementary strands under different concentrations of NaCl. Reducing Nonspecific DNA Adsorption with Glutathione Treatment Since hybridization efficiency and affinity of DNA probes immobilized on the surface are the critical parameters used to determine their functionality, much effort has been directed towards eliminating the nonspecific adsorption of DNAs on surfaces of both nanoparticles and plannar substrates. Common techniques include introducing short-chain alkylthiols to block the surface or replacing the nonspecific binding using thiol compounds,53, 54 which helps to orient the DNA radially, and facilitate hybridization. 6-Mercapto-1-hexanol (MCH) was used previously to passivate gold surfaces,53 where MCH displaced nonspecific nucleobase adsorption, thus allowing SH-DNAs to take an upright conformation and improve their hybridizability to complementary strands.49 However, due to its strong displacement capability towards SH-DNAs on the AuNP surface, MCH could completely remove SH-DNAs; therefore accurate control of MCH concentration and reaction time is required.49, 55 To terminate the MCH treatment before over-replacement of SH-DNA, organic solvent such as ethyl acetate must be used to extract MCH from the aqueous AuNP solution, resulting in extra effort, time, solvent usage and cost. Additonally, MCH is toxic and volatile and thus not convenient for routine use in applied biotechnology. Recently, (11-Mercaptoundecyl)hexa(ethylene glycol) (TOEG6) was demonstrated to facilitate synthesis of hybridizable SNAs by eliminating nonspecific DNA binding;54 however, the high cost of TOEG6 and relatively low DNA-capacity of the prepared SNAs (i.e., ~41 SH-DNAs on each 20 nm AuNP) motivated us to seek more cost-effective and less destructive passivating chemicals. We found that the nontoxic and biocompatible glutathione (GSH) showed comparable efficiency to MCH in removing the nonspecific DNA adsorption (Scheme 2). Firstly, we observed an increase in the hydrodynamic size of the SNAs upon GSH treatment. When the AuNP-templated SNAs prepared by the low-pH method were exposed to different concentrations of GSH (ranging from 0 to 100 mM) for 10 min, their size increase reached the maxium (~ 5 nm) at 10 mM GSH treatment, and then decreased gradually as higher concentrations of GSH were used (Fig. 4A and Fig. S4). The size increase was attributed to removal of nonspecific nucleobase adsorption, which led to the adsorbed SH-DNAs stretching out on the AuNP surface, resulting in a radial configuration of GSH-treated SNAs. Such conformational change was especially significant for the short 9-mer DNA (DNA 1) with the lowest loading density (~43 DNA/AuNP), which stretched out up to 100% of their theoretical fully-stretched length (herein, the elongation per base was estimated as 0.4 nm in ssDNAs based on established evidence56-58) upon GSH treatment. However, when GSH concentration was further increased, more SH-DNAs were desorbed from AuNP surfaces resulting in decreased hydrodynamic sizes due to loss of adsorbed SH-DNAs. For example, when GSH was ≥ 50 mM, the hydrodynamic size of SNAs decreased due to lower DNA density on
ACS Paragon Plus Environment
ACS Applied Materials & Interfaces
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
AuNP surfaces, as further elaborated in Fig. S5, where the effect of GSH concentrations on the number of adsorbed SH-DNA and hybridized cDNA, was illustrated. Since the Au-S bond is stronger than the nonspecific Au-nucleobase interaction, we reasoned that GSH should displace DNA’s nonspecific adsorption (Au – O or Au – N interactions) much easier than it replaces the SH-DNA attached to AuNP surface via the specific Au-S bond. To verify this hypothesis, we used GSH to treat SNAs in which each AuNP was functionalized with only ~30 SH-DNAs on the nanoparticle surface. Based on published work,32, 38 such a DNA density on the 13-nm AuNP surface would not result in steric effect or electrostatic barrier preventing DNA hybridization. After treating these SNAs with 10 µM GSH, we observed that
