This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.
Communication pubs.acs.org/JACS
Freezing Directed Construction of Bio/Nano Interfaces: Reagentless Conjugation, Denser Spherical Nucleic Acids, and Better Nanoflares Biwu Liu and Juewen Liu* Department of Chemistry, Waterloo Institute for Nanotechnology, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada S Supporting Information *
DNA sequences in a fast time scale (e.g., a few minutes).5f A key of most aforementioned strategies is to increase salt concentration, and this might be achieved by freezing without adding more salt. Herein, we communicate an extremely simple yet effective freezing method to functionalize AuNPs with DNA, for which no other chemicals are needed. More importantly, the DNA loading density is even higher than the conjugates prepared using various other methods, reflecting unique fundamental surface science during freezing. Dispersed 13 nm citrate-capped AuNPs (cit-AuNPs) are red. As expected, the red color turned blue upon freezing and remained blue after thawing (Figure 1B), indicating irreversible aggregation of the AuNPs. We then added a thiolated DNA (DNA1, sequence in Table S1) before freezing. Interestingly, the red color was retained after a freeze−thaw cycle (Figure 1B). After thawing, we centrifuged the AuNPs and removed the supernatant. The redispersed AuNPs remained stable even with 1 M NaCl, while the nonfrozen sample (simply mixed with DNA without salt aging) was aggregated by 0.3 M NaCl (Figure 1C). Such a high stability strongly indicated that the DNA strands were densely attached to the AuNPs during the freeze−thaw process. To examine the quality of the DNA/AuNP conjugates, we compared freezing with two other commonly used methods: saltaging2b,6 and low-pH.7 All these methods retained the stability of AuNPs as confirmed by UV−vis spectroscopy (Figure 1D). The number of DNA loaded on each 13 nm AuNP was determined to be 110 for the freezing method (Figure 1E), even higher than that by the salt-aging (final 700 mM NaCl for 12 h) and the low-pH method.10 Such a high DNA loading explained its excellent colloidal stability. From dynamic light scattering (DLS), the cit-AuNPs had a hydrodynamic size of 15 nm (Figure 1F and Table S2). After DNA modification, the size increased to ca. 21 nm due to the DNA layer, and no large aggregates were detected, consistent with our UV−vis data. The surface charge of the sample by freezing was slightly more negative than that prepared by salt-aging (Figure S1). To confirm the function of the conjugates prepared by freezing, a complementary linker DNA was added to two types of DNA/AuNP conjugates, and the color of the system turned purple, indicating DNA-directed assembly (Figure S2). We were intrigued by the very high DNA density, and a dense layer of charged DNA crowded on AuNPs can lead to interesting properties.2b Therefore, we further tested the functional and mechanistic aspects of the spherical nucleic acids.
ABSTRACT: While nanoparticle solutions cannot freeze in general, they may remain stable in the presence of polymer stabilizers. We herein communicate that gold nanoparticles (AuNPs) are stable in the presence of thiolated DNA after a freeze−thaw cycle. The DNA is conjugated to AuNPs during freezing without additional reagents and the conjugation can be completed in a few minutes. More importantly, the DNA density is 20−30% higher than that prepared by the typical salt-aging method. By lowering temperature, DNA hybridization is also promoted, allowing the construction of better nanoflares with doubled probe density and signaling sensitivity. This freezing method works for AuNPs from 5 to 100 nm and all tested DNA sequences. The mechanism was studied by separating the effect of temperature, freezing and thawing, where the exclusion of salt and AuNPs by the growing ice crystals is deemed critical. In addition to developing a simple method, this study articulates unique physical processes during freezing with important fundamental surface science implications, and it could be extended to other systems.
I
n colloid science, it is well-known that nanoparticle solutions cannot be frozen, or they aggregate irreversibly. This is attributable to the exclusion of salt and nanoparticles by the growing ice crystals, leading to increased local salt concentration screening the electrostatic repulsion between the nanoparticles.1 However, if the nanoparticles are capped by polymers, aggregation might be reversible due to steric stabilization. Because of the adverse effect of freezing, most nanoparticles research was performed in solution. For biochemists, however, freezing is quite common since biopolymers often retain activity upon thawing. Given the unique concentrating effect during freezing and stability of biopolymers, we reason that freezing might be used to our advantage to functionalize nanoparticles. For example, attaching DNA to nanoparticles is critical for DNA-directed assembly,2 biosensor development,3 and drug and gene delivery.4 The most researched material in this regard is gold nanoparticles (AuNPs).2b,5 To attach negatively charged DNA to negatively charged AuNPs, a salt-aging method was developed by Mirkin and co-workers to screen charge repulsion (Figure 1A, reaction 1).2b,6 Salt needs to be added very slowly to avoid aggregation of AuNPs, and the whole process takes a full day. While a few other methods have been developed such as the addition of acids,7 surfactants,8 or using modified DNA,9 none of them can be generalized to all © 2017 American Chemical Society
Received: May 11, 2017 Published: June 29, 2017 9471
DOI: 10.1021/jacs.7b04885 J. Am. Chem. Soc. 2017, 139, 9471−9474
Communication
Journal of the American Chemical Society
Figure 2. (A) Schemes showing two methods of preparing nanoflares: (1) coadsorption and (2) posthybridization. The probe DNA captured by each NP using (B) the coadsorption and (C) the posthybridization method. (D) Kinetics of detecting target DNA (10 nM) using the prepared nanoflares.
where the probe DNA was added to preformed DNA/AuNP conjugates. However, very few probe DNAs were hybridized regardless of the preparation method (Figure 2C), consistent with previous observations.12 This was attributable to the steric hindrance from the high DNA density on the AuNPs. To test the functionality of our nanoflares, we added 10 nM of target DNA to induce fluorescence recovery (Figure 2D). A faster and stronger response (e.g., doubled signal) was observed with the flares prepared by the freezing method, consistent with the above probe density measurement. This improvement is attributable to the favored DNA duplex formation upon cooling. After loading a high density of capture DNA, hybridization with the probe DNA becomes more and more difficult as hybridization proceeds. Our coadsorption and freezing method has provided a good solution to this problem.13 Given the merits of reagentless conjugation, higher DNA density, and facilitated DNA hybridization of the freezing method, we then tested its generality. The traditional salt-aging method does not work well with larger AuNPs unless surfactants are added,8a while the low pH method prefers poly-A containing DNA.7,14 We used the freezing method for AuNPs from 5 to 100 nm. All of the samples retained colloidal stability (Figures S4 and S5). A set of UV−vis spectra of the 100 nm AuNPs are shown in Figure 3A as an example, indicating the stability of the conjugates. For all the AuNP sizes, the freezing method produced a higher DNA density by 20−30% than the salt-aging method (Figure 3B). Such a high DNA density may further improve their stability against salt-induced aggregation. Without DNA, AuNPs of 50 nm aggregate quickly after adding 0.5 M NaCl (Figure S6). Using the salt-aging method, the 50 nm AuNP conjugates were stable in 0.75 M NaCl. We then quantitatively followed the color by plotting the extinction ratio at 700 nm over 530 nm upon adding 1 M NaCl (Figure 3C). The 50 nm AuNP conjugate prepared by freezing retained its stability, but the salt-aged sample quickly changed color. Therefore, for all these AuNPs, freezing resulted in a higher DNA density and better colloidal stability. We then tested if freezing could be applied to different DNA sequences. Most of our above work used a DNA containing a poly-A spacer after the thiol group. For a DNA with a poly-T spacer, the DNA loading capacity was still the highest using the
Figure 1. (A) Scheme of attaching DNA to AuNPs by (1) salt-aging; (2) lowering pH; and (3) freezing. (B) Photographs of AuNPs before and after a freeze−thaw cycle. While cit-AuNPs aggregated, adding a thiolated DNA (DNA1, 3 μM) protected the AuNPs. (C) Photographs showing the stability of AuNPs mixed with DNA1 with or without freezing after adding NaCl. Comparison of different conjugation methods by (D) normalized UV−vis absorption spectra; (E) the number of DNAs on each NP (* stands for p value < 0.05, n = 3); and (F) the hydrodynamic size of AuNPs and DNA/AuNP conjugates (3 measurements, 20 runs each).
An interesting application of such conjugates is “nanoflares”, where a fluorophore-labeled probe DNA is hybridized to the DNA on AuNPs (capture DNA) resulting in quenched fluorescence.11 Addition of a target nucleic acid (e.g., cDNA or mRNA) can release the fluorescent probe and produce fluorescence. To prepare nanoflares, the probe DNA needs to hybridize to the capture DNA, and freezing might promote DNA hybridization. To test this, we mixed the probe DNA and the capture DNA simultaneously with AuNPs (Figure 2A, reaction 1). After salt aging or freezing, the attached probe DNA was quantified. Interestingly, the freezing method nearly doubled the probe density compared to the salt-aging method at each probe concentration (Figure 2B). At the same time, less capture DNA was immobilized by the freezing method (Figure S3), likely due to the limited total surface area. For comparison, we also tried the traditional posthybridization method (Figure 2A, reaction 2), 9472
DOI: 10.1021/jacs.7b04885 J. Am. Chem. Soc. 2017, 139, 9471−9474
Communication
Journal of the American Chemical Society
Figure 3. (A) UV−vis absorption spectra of 100 nm AuNPs before and after DNA modification. (B) The number of DNA loaded on each AuNP of different sizes using different methods. (C) Kinetics of color change indicating aggregation of DNA/AuNPs (50 nm) induced by 1 M NaCl. (D) Adsorption capacity of DNA with a poly-T block (DNA2, Table S1) onto 13 nm AuNPs (insets are the photographs of AuNPs after attaching DNA); *** stands for p < 0.001, and **** for p < 0.0001.
freeze method (Figure 3D). A G-rich DNA (DNA3, Table S1) was also successfully attached onto AuNPs by freezing (Figure S7). Salt-aging is known to be effective in pushing for a high DNA density on AuNPs. It is quite interesting that DNA can be loaded at even higher densities by simply freezing. To understand the mechanism, we divided this reaction into three stages: cooling, freezing, and thawing. By adding 25% (w/v) glycerol we achieved a −20 °C solution without freezing, but this sample did not achieve a high DNA loading (Figure 4A). From 37 to −20 °C, the DNA loading density was similar, only ca. 45% density of that of the frozen sample (no salt aging performed). After ruling out the cooling effect, we reason that DNA attachment occurred during freezing (instead of thawing) since the red color remained in the frozen state (Figure S8). Furthermore, had DNA adsorption occurred mainly during thawing, adding a hot NaCl solution to quickly thaw the sample might cause aggregation. However, the AuNPs remained red by adding 1 M NaCl of 80 °C (Figure S9). We reason that crystallization of water molecules may repel AuNPs, DNA, and salt out of the growing ice crystals, resulting in a high local concentration for all of them, enhancing DNA attachment kinetics. To this end, we have established that formation of ice is important for successful loading of DNA on AuNPs. By putting a sample (e.g., 2 mL) in a −20 °C freezer, it takes around 1 h to freeze. We then tried faster freezing by immersing the DNA/ AuNP mixture into dry ice for 2 min, which was sufficient for freezing. In this case, the DNA containing sample still showed the typical red color in 1 M NaCl (Figure 4B). Therefore, as long as freezing takes place, even 2 min is sufficient for DNA conjugation. However, keeping the sample in the frozen state for longer did not further increase the DNA loading density (Figures S10 and S11). We then studied the effect of the initial DNA concentration. With more DNA added, more DNA attached onto the AuNPs after freezing, and adsorption profile followed the Langmuir isotherm (Figure 4C). The surface of AuNPs approached saturation when the DNA/AuNP ratio was higher than 300. Good colloidal stability of the AuNPs was achieved only when the initial DNA/AuNP ratio exceeded 200 (insets of Figure 4C).
Figure 4. Understanding DNA loading in the freeze−thaw process. (A) DNA loading capacity as a function of temperature. (B) Photographs showing the stability of DNA/AuNPs prepared by fast freezing with dry ice. (C) Adsorption capacity at different DNA concentrations. Insets: photographs at those DNA concentrations. (D) Effect of preloaded DNA on further attachment of DNA using the freezing method. The error bars represent standard deviation from three independent samples. (E) Proposed mechanism of attaching DNA onto AuNPs by freezing.
Since each 13 nm AuNP can only adsorb ∼110 DNA strands, excess DNA was required to retain the stability of AuNPs during freezing. Similarly, excess DNA was required for the salt-aging method. Therefore, freezing is similar to the salt-aging process in this regard, and freezing can be considered as a faster way of raising the salt concentration.5f To further understand the freezing method, we performed DNA loading in two steps. First, AuNPs were modified with a low density of DNA using the low pH method (Figure 4D, red bars). The reason to use the low pH method was to maintain the stability of AuNPs during DNA attachment.7 Second, additional DNA was added to make the total DNA concentration identical for all samples. Finally, freezing was performed. Interestingly, with increasing preloaded DNA, further DNA loading by freezing decreased (Figure 4D, blue bars), and the total DNA capacity also decreased. With a low density of preadsorbed DNA, some DNA bases can also adsorb, and newly added DNA needs to use their thiol groups to displace the bases. Such decrease in DNA loading suggested that the thiol group in the DNA may not have enough time to displace the adsorbed DNA bases during freezing. It also suggests that the thiol group reacted with AuNPs more quickly relative to the DNA bases for the freezing method than the typical salt-aging method. Along this line of thinking, the attachment of nonthiolated DNA was further tested. Interestingly, our freezing method failed to attach a high density of such DNA or provide high colloidal stability (Figure S12), which was different from the salt-aging method5d or the low pH method.15 This observation also 9473
DOI: 10.1021/jacs.7b04885 J. Am. Chem. Soc. 2017, 139, 9471−9474
Communication
Journal of the American Chemical Society Table 1. Comparison of Three Methods in Preparing DNA/AuNPs Conjugates
a
method
reagent
DNA hybridization
nonspecific DNA base adsorption
DNA sequence requirement
AuNP size
DNA concentration
reaction time
salt-aging low pH freezing
salt (Na+) acid (H+) none
allowed no allowed
salt dependent sequence dependent low
none prefers A-rich DNA none
24 h 3−30 min 2−60 minb
Surfactants needed for AuNPs > 40 nm. b∼2 min in dry ice, ∼60 min in −20 °C freezer; the time also depends on sample volume. S.; Schatz, G. C.; Mirkin, C. A. Nature 2008, 451, 553. (d) Nykypanchuk, D.; Maye, M. M.; van der Lelie, D.; Gang, O. Nature 2008, 451, 549. (e) Pinheiro, A. V.; Han, D.; Shih, W. M.; Yan, H.; Nat. Nat. Nanotechnol. 2011, 6, 763. (f) Tan, S. J.; Campolongo, M. J.; Luo, D.; Cheng, W.; Nat. Nat. Nanotechnol. 2011, 6, 268. (g) Sun, W.; Boulais, E.; Hakobyan, Y.; Wang, W. L.; Guan, A.; Bathe, M.; Yin, P. Science 2014, 346, 1258361. (h) Ohta, S.; Glancy, D.; Chan, W. C. W. Science 2016, 351, 841. (i) Tan, L. H.; Xing, H.; Lu, Y. Acc. Chem. Res. 2014, 47, 1881. (3) (a) Rosi, N. L.; Mirkin, C. A. Chem. Rev. 2005, 105, 1547. (b) Wang, H.; Yang, R.; Yang, L.; Tan, W. ACS Nano 2009, 3, 2451. (c) Liu, J.; Cao, Z.; Lu, Y. Chem. Rev. 2009, 109, 1948. (d) Zhao, W.; Brook, M. A.; Li, Y. ChemBioChem 2008, 9, 2363. (e) Pei, H.; Zuo, X.; Zhu, D.; Huang, Q.; Fan, C. Acc. Chem. Res. 2014, 47, 550. (f) Sun, H.; Ren, J.; Qu, X. Acc. Chem. Res. 2016, 49, 461. (g) Tao, Y.; Lin, Y.; Ren, J.; Qu, X. Biomaterials 2013, 34, 4810. (h) Lu, C.; Huang, Z.; Liu, B.; Liu, Y.; Ying, Y.; Liu, J. Angew. Chem., Int. Ed. 2017, 56, 6208. (i) Zhang, H.; Lai, M.; Zuehlke, A.; Peng, H.; Li, X.-F.; Le, X. C. Angew. Chem., Int. Ed. 2015, 54, 14326. (j) Zhou, W.; Saran, R.; Liu. Chem. Rev. 2017, 117, 8272. (4) (a) Giljohann, D. A.; Seferos, D. S.; Daniel, W. L.; Massich, M. D.; Patel, P. C.; Mirkin, C. A. Angew. Chem., Int. Ed. 2010, 49, 3280. (b) Li, J.; Mo, L.; Lu, C.-H.; Fu, T.; Yang, H.-H.; Tan, W. Chem. Soc. Rev. 2016, 45, 1410. (c) Raeesi, V.; Chou, L. Y. T.; Chan, W. C. W. Adv. Mater. 2016, 28, 8511. (d) Torabi, S. F.; Lu, Y. Curr. Opin. Biotechnol. 2014, 28, 88. (5) (a) Herne, T. M.; Tarlov, M. J. J. J. Am. Chem. Soc. 1997, 119, 8916. (b) Zhang, X.; Servos, M. R.; Liu, J. Langmuir 2012, 28, 3896. (c) Liu, J. Phys. Chem. Chem. Phys. 2012, 14, 10485. (d) Pei, H.; Li, F.; Wan, Y.; Wei, M.; Liu, H.; Su, Y.; Chen, N.; Huang, Q.; Fan, C. J. J. Am. Chem. Soc. 2012, 134, 11876. (e) Carnerero, J. M.; Jimenez-Ruiz, A.; Castillo, P. M.; PradoGotor, R. ChemPhysChem 2017, 18, 17. (f) Liu, B.; Liu, J. Anal. Methods 2017, 9, 2633. (g) Li, F.; Zhang, H.; Dever, B.; Li, X.-F.; Le, X. C. Bioconjugate Chem. 2013, 24, 1790. (6) Storhoff, J. J.; Elghanian, R.; Mucic, R. C.; Mirkin, C. A.; Letsinger, R. L. J. J. Am. Chem. Soc. 1998, 120, 1959. (7) Zhang, X.; Servos, M. R.; Liu, J. J. J. Am. Chem. Soc. 2012, 134, 7266. (8) (a) Hurst, S. J.; Lytton-Jean, A. K. R.; Mirkin, C. A. Anal. Chem. 2006, 78, 8313. (b) Zu, Y.; Gao, Z. Anal. Chem. 2009, 81, 8523. (9) Xu, Q.; Lou, X.; Wang, L.; Ding, X.; Yu, H.; Xiao, Y. ACS Appl. Mater. Interfaces 2016, 8, 27298. (10) Demers, L. M.; Mirkin, C. A.; Mucic, R. C.; Reynolds, R. A.; Letsinger, R. L.; Elghanian, R.; Viswanadham, G. Anal. Chem. 2000, 72, 5535. (11) (a) Rosi, N. L.; Giljohann, D. A.; Thaxton, C. S.; Lytton-Jean, A. K. R.; Han, M. S.; Mirkin, C. A. Science 2006, 312, 1027. (b) Zheng, D.; Seferos, D. S.; Giljohann, D. A.; Patel, P. C.; Mirkin, C. A. Nano Lett. 2009, 9, 3258. (c) Prigodich, A. E.; Seferos, D. S.; Massich, M. D.; Giljohann, D. A.; Lane, B. C.; Mirkin, C. A. ACS Nano 2009, 3, 2147. (d) Wang, W.; Satyavolu, N. S. R.; Wu, Z.; Zhang, J.-R.; Zhu, J.-J.; Lu, Y. Angew. Chem., Int. Ed. 2017, 56, 6798. (e) Wu, P.; Hwang, K.; Lan, T.; Lu, Y. J. Am. Chem. Soc. 2013, 135, 5254. (12) (a) Lytton-Jean, A. K. R.; Mirkin, C. A. J. Am. Chem. Soc. 2005, 127, 12754. (b) Randeria, P. S.; Jones, M. R.; Kohlstedt, K. L.; Banga, R. J.; Olvera de la Cruz, M.; Schatz, G. C.; Mirkin, C. A. J. Am. Chem. Soc. 2015, 137, 3486. (13) Sedighi, A.; Li, P. C. H.; Pekcevik, I. C.; Gates, B. D. ACS Nano 2014, 8, 6765. (14) Huang, Z.; Liu, B.; Liu, J. Langmuir 2016, 32, 11986. (15) Zhang, X.; Liu, B.; Dave, N.; Servos, M. R.; Liu, J. Langmuir 2012, 28, 17053.
supports the need for a thiol group for the freezing method. Taken together, a reaction mechanism was proposed to explain the freezing method (Figure 4E). At room temperature, negatively charged cit-AuNPs and DNA repel each other. Upon decreasing the temperature, ice crystals are gradually formed, pushing AuNPs, DNA, and salt (Na+, citrate ions) out to the “micropockets” to be concentrated, which significantly enhances the reaction rate. The excess amount of DNAs likely contribute to the separation of AuNPs, which is important for colloidal stability. In Table 1, the three methods of attaching DNA to AuNPs are compared. In summary, we demonstrated freezing-directed construction of bio/nano interfaces using the DNA/AuNP system. We discovered a general, facile, and cost-effective method to functionalize AuNPs of different sizes with various DNAs. Compared to all the previous methods, freezing is fundamentally different in the following three aspects. (1) It is reagentless and no extra salts, acids, or surfactants are needed. (2) The resulting conjugate has a higher DNA density with higher colloidal stability. (3) Hybridization is also promoted. Finally, this study can be extended to other bio/nano systems taking advantage of the unique physical processes occurred during freezing, and it has revealed important surface science upon freezing.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b04885. Characterization of DNA/AuNPs by DLS, ζ-potential, UV−vis, effect of AuNP size and DNA sequence, adsorption of nonthiolated DNA, and control experiments (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*
[email protected] ORCID
Biwu Liu: 0000-0001-7357-9875 Juewen Liu: 0000-0001-5918-9336 Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS Funding for work is from the Natural Sciences and Engineering Research Council of Canada (NSERC). REFERENCES
(1) Deville, S. Freezing Colloids: Observations, Principles, Control, and Use: Applications in Materials Science, Life Science, Earth Science, Food Science, and Engineering, 1st ed.; Springer International Publishing: 2017. (2) (a) Jones, M. R.; Seeman, N. C.; Mirkin, C. A. Science 2015, 347, 1260901. (b) Cutler, J. I.; Auyeung, E.; Mirkin, C. A. J. J. Am. Chem. Soc. 2012, 134, 1376. (c) Park, S. Y.; Lytton-Jean, A. K. R.; Lee, B.; Weigand, 9474
DOI: 10.1021/jacs.7b04885 J. Am. Chem. Soc. 2017, 139, 9471−9474