Detection of single-stranded nucleic acids by hybridization of probe oligonucleotides on polystyrene nanospheres and subsequent release and recovery of fluorescence

Abstract

In this contribution, we demonstrate that polystyrene (PS) nanospheres can serve as an effective sensing platform for fluorescence-enhanced DNA detection. This kind of assay can be completed by the following two steps: (1) PS quenches the fluorescence of dye-labeled single-stranded DNA (ssDNA) probes very effectively when they are brought into close proximity as a result of the adsorption of ssDNA on PS. The adsorption is ascribed to the strong π–π stacking between unpaired DNA bases and PS. (2) Upon presence of target ssDNA, the specific hybridization of the probe with its target produces a double-stranded DNA (dsDNA). The duplex detaches from PS due to its rigid conformation, the absence of unpaired DNA bases, and electrostatic repulsion between negatively charged dsDNA backbone and PS, leading to recovery of dye fluorescence. This assay system exhibits high selectivity and sensitivity with a detection limit as low as 5 nM. It suggests that this sensing platform can differentiate between perfectly complementary and mismatched targets. The fluorescence enhancement in response to single-base mismatched target T₂, two-base mismatched target T₃, and three-base mismatched target T₄ is about 71%, 63%, and 58% of that from complementary target T₁, respectively. The suggested method can also discriminate complementary and single-base mismatched sequences embedded in rather larger strands with short oligonucleotide probes. The fluorescence enhancement in response to single-base mismatched sequence embedded in large strand is 83% of that from the one embedded with complementary sequence. In further experiments, it is demonstrated that the present system can be used for multiple DNA detection. Finally, efforts are made toward its application in human blood serum system to evaluate the ability to withstand the interference arising from real sample.

---

Introduction

Rapid, cost-effective, sensitive, and specific methods for the detection of DNA possesses important applications in the fields, such as gene expression profiling, clinical disease diagnostics, clinical treatment, and so on.¹ The increasing availability of nanostructures has created widespread interest in their use in biotechnological systems for diagnostic applications,² and the employment of various nanostructures for this purpose has been well documented.³ Recently, nanostructures have also proven of particular utility as “nanoquenchers” in fluorescense-enhanced nucleic acid detection.^4–8 Because the same nanostructure is capable of quenching dyes with different emission wavelengths, the selection issue of a fluorophore-quencher pair is eliminated from the nanostructure-based system.⁴ Dubertret et al. have pioneered the use of dye fluorescence quenching ability of small gold nanoparticles (AuNPs) for DNA detection.⁵ In their study, a DNA moiety is decorated to a 1.4-nm AuNP surface and the DNA strand is curved to a hairpin structure by Watson–Crick hydrogen bonding. This hairpin conformation brings fluorescent dye into close proximity of the nanoparticle, leading to quenching of dye fluorescence. The subsequent specific hybridization of the moiety with target opens the hairpin and thus separates the fluorophore from the AuNP at a sufficient distance to allow fluorescence recovery. Maxwell et al. have also developed a similar AuNP-based nanobiosensor to detect nucleic acid.⁶ Although both of them are able to differentiate single-base mismatch in nucleic acid, they require tedious and laborious surface attachment chemistry for probe immobilization and suffer from slow response. To solve these problems, Liet al. have designed a novel fluorescent assay for DNA hybridization,⁷ which is based on that single-stranded DNA (ssDNA) adsorbs on negatively charged AuNP while double-stranded DNA (dsDNA) does not. As a result, dye-labeled probes have their fluorescence efficiently quenched when they are mixed with AuNPs unless they hybridize with components of the analyte.⁷ Up to now, considerable efforts have been made to develop fluorescent methods for DNA detection based on AuNPs.^5–8 Recently, both single-walled carbon nanotubes⁹ and graphene oxide¹⁰ have found their significant application for fluorescence-enhanced DNA detection. Till now, many other structures have been used in this assay, including silver nanoparticles,¹¹multi-walled carbon nanotubes and single-walled carbon nanohorns,¹²carbon nanoparticles,¹³carbon nanospheres,¹⁴nano-C₆₀,¹⁵ mesoporous carbon microparticles,¹⁶ poly(p-phenylenediamine) nanobelts,¹⁷polyaniline nanofibres,¹⁸ poly(o-phenylenediamine) colloids,¹⁹ poly(m-phenylenediamine) nanorods,²⁰ poly(2,3-diaminonaphthalene) microspheres,²¹ coordination polymer colloids, nanobelts and microdendrites,^22–25Ag@poly(m-phenylenediamine) core-shell nanoparticles,²⁶tetracyanoquinodimethane nanoparticles,²⁷ supramolecular microparticles,²⁸ and porphyrin nanoparticles.²⁹

Polystyrene (PS) colloids are important in many areas of technology, such as paints, coatings, ceramics processing, and the construction of photonic bandgap crystals.³⁰ More recently, Socholet al. have constructed a microfluidic system for parallel detection of multiple DNA via molecular beacon probes immobilized on polystyrene microbeads. However, the proposed procedure is somewhat complicated, involving dual-labeled molecular beacon and its immobilization on PS microbeads.³¹ In this paper, we demonstrate a simpler fluorescence-enhanced DNA detection using PS nanospheres as an effective sensing platform. DNA detection is accomplished by the following two steps. First, PS adsorbs and quenches a dye-labeled single-stranded DNA (ssDNA) probe. In the second step, the specific hybridization of the probe with its target produces a double-stranded DNA (dsDNA) which detaches from PS, leading to recovery of dye fluorescence. We further demonstrate that this sensing platform can differentiate between perfectly complementary and mismatched targets. It can also discriminate complementary and single-base mismatched sequences embedded in rather larger strands with short oligonucleotide probes. In further experiments, it is demonstrated that the present system can be used for multiple DNA detection. Finally, efforts are made toward its application in human blood serum system to evaluate its ability to withstand the interference arising from real sample.

Experimental

All chemically synthesized oligonucleotides were purchased from TaKaRa Biotechnology (Dalian). DNA concentration was estimated by measuring the absorbance at 260 nm.³² All the other chemicals and the aqueous dispersion of COOH-functionalized PS nanoparticles (2.5% w/v) with a mean diameter of about 100 nm were purchased from Aladin Ltd. (Shanghai, China) and used as received without further purification. The water used throughout all experiments was purified through a Millipore system.

Transmission electron microscopy (TEM) measurements were made on a HITACHI H-8100 EM (Hitachi, Tokyo, Japan) with an accelerating voltage of 200 kV. Fluorescent emission spectra were recorded on a RF-5301PC spectrofluorometer (Shimadzu, Japan) using a 300-μL quartz cuvette. Zeta potential measurements were performed on a Nano-ZS Zetasizer ZEN3600 (Malvern Instruments Ltd., U.K.).

The desired DNA concentration was achieved by diluting the concentrated DNA stock solutions with 20 mM Tris-HCl buffer containing 100 mM NaCl, 5 mM KCl, and 5 mM MgCl₂ (pH: 7.4). The volume of each sample for fluorescence measurement is 300 μL. All the experiments were carried out at room temperature (about 25 °C) if not specified.

The precise PS concentration was achieved following this protocol: Drawing the corresponding amount of 2.5% (w/v) PS, then the solvent was removed by centrifugation and the nanospheres were suspended in 300-μL Tris-HCl buffer.

Oligonucleotide sequences are listed as below (mismatch underlined):

(1) P_HIV (FAM dye-labeled ssDNA):

5’-FAM-AGT CAG TGT GGA AAA TCT CTA GC-3’

(2) T₁ (complementary target to P_HIV):

5’-GCT AGA GAT TTT CCA CAC TGA CT-3’

(3) T₂ (single-base mismatched target to P_HIV):

5’-GCT AGA GAT TG̲T CCA CAC TGA CT-3’

(4) T₃ (two-base mismatched target to P_HIV):

5’-GCT AGA GAT TG̲T A̲CA CAC TGA CT-3’

(5) T₄ (three-base mismatched target to P_HIV):

5’-GCT A̲TA GAT TG̲T A̲CA CAC TGA CT-3’

(6) T₅ (non-complementary target to P_HIV):

5’-TTT TTT TTT TTT TTT TTT TTT TT-3’

(7) T_L1 (The middle part of a long strand as a target complementary to P_HIV):

5’-TTT TTT TTT TTT TTT TTT TTT TGC TAG AGA TTT TCC ACA CTG ACT TTT TTT TTT TTT TTT TTT TTT T-3’

(8) T_L2 (The middle part of a long strand as a single-base mismatched target to P_HIV):

5’-TTT TTT TTT TTT TTT TTT TTT TGC TAG AGA TTG̲ TCC ACA CTG ACT TTT TTT TTT TTT TTT TTT TTT T-3’

(9) P_K167 (Cy5 dye-labeled ssDNA):

5’-Cy5-TCT GCA CAC CTC TTG ACA CTC CG-3’

(10) T₆ (complementary target to P_K167):

5’-CGG AGT GTC AAG AGG TGT GCA GA-3’

Results and discussion

Fig. 1 shows representative TEM images of the PS nanospheres used in this study. The low magnification image indicates that the nanospheres range in diameter from 80 to 120 nm, as shown in Fig. 1a. The high magnification image further reveals that these nanospheres have a smooth surface, as shown in Fig. 1b.

Fig. 1 (a) Low and (b) high magnification TEM images of PS nanospheres with a mean diameter of about 100 nm.

To demonstrate that PS nanospheres can be used as an effective fluorescent sensing platform for DNA detection, an oligonucleotide sequence associated with human immunodeficiency virus (HIV) was chosen as a model system according to related literature,¹⁰ labeled at the 5’ end with the dye carboxyfluorescein (FAM). In the absence of PS, these species (P_HIV) exhibit strong fluorescence emission in Tris-HCl buffer due to the presence of the fluorescein-based dye (Fig. 2, curve a). In contrast, the presence of 0.83% PS (w/v) leads to about 59% quenching of the fluorescence emission (curve c), demonstrating that the nanosphere can adsorb ssDNA and quench the dye effectively. It should be noted that the addition of the complementary sequence T₁ in the absence of PS has a small influence on the fluorescence of the free P_HIV (curve b), whereas the P_HIV-PS complex exhibits significant fluorescence enhancement upon its incubation with T₁ over a period of 15 min, as is evidenced by a 84% fluorescence recovery (curve d). It should be noted that this hybridization between complementary target and probe is competitive with ssDNA probe binding to PS, which means that the complementary strands can’t hybridize with 100% efficiency. Curve e shows that PS itself exhibits no fluorescence emission and thus makes no contribution to the fluorescence intensity of each PS-involved sample measured. Fig. 2 inset illustrates the fluorescence intensity changes (F/F₀–1) of P_HIV-PS complex upon addition of different concentrations of T₁, where F₀ and F are FAM fluorescence intensities at 517 nm in the absence and the presence of T₁, respectively. In the DNA concentration range of 5–2000 nM, a dramatic increase of FAM fluorescence intensity is observed, suggesting that the PS/DNA assembly approach is effective in probing biomolecular interactions. The detection limit is estimated to be 5 nM (three times the standard deviation of blank sample).

Fig. 2 Fluorescence emission spectra of P_HIV (100 nM) at different conditions: (a) P_HIV; (b) P_HIV + 600 nM T₁; (c) P_HIV + PS; (d) P_HIV + PS + 600 nM T₁. Curve e is the emission spectra of the PS sample. Inset: fluorescence intensity changes (F/F₀–1) of P_HIV−PS complex (F₀ and F are the fluorescence intensities without and with the presence of T₁, respectively) plotted against the logarithm of the concentration of T₁. The concentration of PS used in each sample is 0.83% (w/v). All measurements were done in Tris-HCl buffer in the presence of 5 mM Mg²⁺ (pH: 7.4). Excitation was at 480 nm, and the emission was monitored at 517 nm .

The zeta potential of the PS nanospheres in Tris-HCl buffer (pH: 7.4) was measured to be about −47.5 mV. Electrostatic repulsion between PS and the negatively charged backbone of ssDNA exists in this system, but is greatly weakened in the presence of a large amount of salt in buffer.²⁶ The stronger π–π stacking between unpaired DNA bases and the styrene polymer dominates the mutual interactions between ssDNA and PS.³³ In contrast, dsDNA does not bind to the surface because of its rigid conformation, the absence of unpaired DNA bases, and electrostatic repulsion between the negatively charged dsDNA backbone and PS. The adsorption of dye-labeled ssDNA on PS is accompanied by fluorescence quenching. This quenching may be ascribed to unimolecular collision between excited fluorophore and aromatic group in PS when they are brought into close proximity.³⁴ In the presence of target, the specific hybridization generates a dsDNA, which detaches from PS and leads to the dye’s fluorescence recovery. Since PS effectively quenches the fluorescence of adsorbed dye-labeled ssDNA, the ssDNA-selective adsorption provides a mechanism for optical sensing of target oligonucleotide (Scheme 1).

Scheme 1 A schematic (not to scale) illustrating the PS-based fluorescence-enhanced DNA detection.

To get further insight into the kinetics of the fluorescence quenching and recovery, we collected the time-dependent fluorescence spectra of P_HIV and PS, as well as of the P_HIV-PS complex with T₁, as shown in Fig. 3. Plot a shows the fluorescence quenching of P_HIV by PS as a function of incubation time. In the absence of the target, the curve exhibits a rapid reduction in the first 1 min and reaches equilibrium over a period of 10 min, indicating that the adsorption process of ssDNA on PS is very fast. Plot b shows the fluorescence recovery of P_HIV-PS by T₁ as a function of incubation time. In the presence of the target T₁, the curve shows a fast increase in the first 1 min, and the best fluorescence response is obtained over a period of 15 min in terms of the tendency as there is almost no obvious change in fluorescence intensity from 2 to 15 min. All the above observations indicate that PS is superior to SWCNT⁹ and GO¹⁰ in detection speed and the present method is time-saving.

Fig. 3 (a) Fluorescence quenching of P_HIV (100 nM) by PS and (b) fluorescence recovery of P_HIV-PS by T₁ (600 nM) in Tris-HCl buffer as a function of incubation time. The concentration of PS used in this sample is 0.83% (w/v). All measurements were done in Tris-HCl buffer in the presence of 5 mM Mg²⁺ (pH: 7.4). Excitation was at 480 nm, and the emission was monitored at 517 nm.

The sensing platform described herein can differentiate complementary and mismatched sequences. Fig. 4 shows the fluorescence responses of P_HIV-PS complex toward perfect complementary target T₁, single-base mismatched target T₂, two-base mismatched target T₃, and three-base mismatched target T₄. The F/F₀ value (F₀ and F are the fluorescence intensities without and with the presence of target, respectively) obtained upon addition of 600 nM T₂, T₃, and T₄ is about 71%, 63%, and 58% of the value obtained upon addition of 600 nM T₁ into P_HIV-PS complex, respectively. Compared to the complementary target, the mismatched target should have lower hybridization ability toward the adsorbed dye-labeled ssDNA probe, leading to decreased hybridization and thus decreased fluorescence recovery efficiency. However, only very small fluorescence enhancement was observed for the P_HIV-PS complex upon addition of 600 nM non-complementary target T₅, indicating that the observed fluorescence enhancement in our present system is indeed due to the base pairing between probe and its target other than competitive displacement. The inset presents the corresponding fluorescence intensity histograms with error bar. All the above observations suggest that the results exhibit good reproducibility and the PS-based system is likely to be capable of practically useful mismatch detection upon further development.

Fig. 4 Fluorescence emission spectra of P_HIV (100 nM) at different conditions: (a) P_HIV-PS complex; (b) P_HIV-PS complex + 600 nM T₁; (c) P_HIV-PS complex + 600 nM T₂; (d) P_HIV-PS complex + 600 nM T₃; (e) P_HIV-PS complex + 600 nM T₄; (f) P_HIV-PS complex + 600 nM T₅. Inset: the corresponding fluorescence intensity histograms with error bar. The concentration of PS used in each sample is 0.83% (w/v). All measurements were done in Tris-HCl buffer in the presence of 5 mM Mg²⁺ (pH: 7.4). Excitation was at 480 nm, and the emission was monitored at 517 nm.

For genomic analysis, it is desirable to detect specific target sequence which is only a small fraction of much longer DNA strand. To do this, two 67-mer DNA strands were tested: the middle part of T_L1 is complementary target sequence to P_HIV and the middle part of T_L2 is single-base-mismatched target sequence to P_HIV. Fig. 5 shows the results of the experiments for distinguishing complementary and single-base matched sequences embedded in longer strands. The addition of 600 nM T_L1 to P_HIV-PS complex leads to about 67% fluorescence recovery, which is much lower than 84% observed when 600 nM T₁ was used as the target. Such observation can be explained as follows: P_HIV-T_L1 is a complex with a duplex DNA in the middle and two single strands on both ends and thus there are unpaired DNA bases for binding to PS. The F/F₀ value obtained upon addition of 600 nM T_L2 is about 83% of the value obtained upon addition of 600 nM T_L1 into P_HIV-PS complex. All the above observations indicate that the present assay system is able to discriminate complementary and mismatched target sequences embedded in longer strands with the use of short probe. Thus, it is promising for practical detection of single-base mismatch in target sequences.

Fig. 5 Fluorescence emission spectra of P_HIV (100 nM) in the presence of PS at different conditions: (a) P_HIV-PS complex; (b) P_HIV-PS complex + 600 nM T_L1; (c) P_HIV-PS complex + 600 nM T_L2. Inset: fluorescence intensity histograms with error bar. Excitation was at 480 nm, and the emission was monitored at 517 nm. The concentration of PS used in each sample is 0.83% (w/v). All measurements were done in Tris-HCl buffer in the presence of 5 mM Mg²⁺ (pH: 7.4).

In further experiments, we evaluated the application potential of this sensing system for real sample analysis by challenging it with human blood serum samples. Fig. 6 shows the fluorescence responses of P_HIV-PS complex toward T₁ and T₂ in the presence of 50% (volume ratio) human blood serum, respectively. The F/F₀ value obtained upon addition of T₂ is about 89% of the value obtained upon addition of T₁ into P_HIV-PS complex. Fig. 6 inset is the corresponding fluorescence intensity histograms with error bar. These observations demonstrate that the present system holds great promise for real sample analysis upon further development.

Fig. 6 Fluorescence emission spectra of P_HIV (100 nM) at different conditions: (a) P_HIV-PS complex; (b) P_HIV-PS complex + 600 nM T₁; (c) P_HIV-PS complex + 600 nM T₂. Inset: the corresponding fluorescence intensity histograms with error bar. The concentration of PS used in each sample is 0.83% (w/v). All measurements were done in Tris-HCl buffer (pH: 7.4) containing 5 mM Mg²⁺ in the presence of 50% (volume ratio) human blood serum. Excitation was at 480 nm, and the emission was monitored at 525 nm.

We also explored the feasibility of using the platform described herein to detect multiple DNA targets simultaneously. To this end, we chose FAM-labeled P_HIV and another probe P_K167 labeled with Cy5 (cyanine 5) as model systems. Because these two dyes are individually excited at 480 and 643 nm to emit at 518 and 660 nm, respectively, significant dye-to-dye energy transfer is avoided. In the presence of PS, the fluorescence of these two dyes in the probe mixture was quenched to a significant extent, suggesting that PS can be used to quench dyes of different emission frequencies. Fig. 7 shows the fluorescence intensity histograms of the probe mixture toward different target combinations in the presence of PS under excitation/emission wavelengths of 480/517 and 643/660 nm/nm. As expected, the addition of T₁ only lead to fluorescence enhancement at 517 nm when excited at 480 nm while the addition of T₆ only results in fluorescence enhancement at 660 nm emission peaks with the excitation at 643 nm. In contrast, the combination of T₁+T₆ results in fluorescence enhancement of both emission peaks. These observations suggest that this PS-based sensing platform can be used for multiple DNA detection.

Fig. 7 Fluorescence intensity histograms of the probe mixture toward different target combinations in the presence of PS under excitation/emission wavelengths of 480/518 and 643/660 nm/nm. ([P_HIV]=[P_K167]=100 nM; [T₁]=[T₆]=600 nM). The concentration of PS used in each sample is 0.83% (w/v). All measurements were done in Tris-HCl buffer in the presence of 5 mM Mg²⁺ (pH: 7.4).

A survey of the influence of the concentration of PS on the efficiency of fluorescence quenching and recovery is shown in Fig. 8, revealing that larger amounts of PS provide increased quenching efficiency but decreased recovery efficiency. This is consistent with the concept that excess PS can adsorb target complementary sequences rather than allow them to hybridize with the adsorbed probe strands. An optimal balance between the amount of PS and fluorescence quenching-recovery was achieved when 0.83% (w/v) PS was used. So an optimal PS concentration 0.83% (w/v) was used throughout all experiments if not specified.

Fig. 8 Fluorescence intensity histograms of P_HIV + PS and P_HIV + PS + T₁ with the use of PS at the concentration of 0, 0.50%, 0.67%, 0.83%, and 1.00% (w/v), respectively. ([P_HIV]=100 nM; [T₁]=600 nM). All measurements were done in Tris-HCl buffer in the presence of 5 mM Mg²⁺ (pH: 7.4). Excitation was at 480 nm, and the emission was monitored at 517 nm.

Conclusions

In conclusion, we demonstrate that inexpensive PS nanospheres can be used as a sensing platform for fluorescence-enhanced DNA detection. This sensing platform can differentiate perfectly complementary and mismatched targets, it can also discriminate complementary and single-base mismatched sequences embedded in rather larger strands with short oligonucleotide probes. Furthermore, this sensing platform is suitable for multiple DNA detection. Most importantly, it holds great potential for practical application in terms of its performance in human blood serum system. This is extended application of PS, and the method holds great potential as a platform for a variety of other target molecule analysis, such as proteins and heavy metal ions.

Acknowledgements

This work was supported by National Basic Research Program of China (No. 2011CB935800).

References

1. D. Gresham, D. M. Ruderfer, S. C. Pratt, J. Schacherer, M. J. Dunham, D. Botstein and L. Kruglyak, Science, 2006, 311, 1932.
2. R. Brayner, Nano Today, 2008, 3, 48.
3. N. L. Rosi and C. A. Mirkin, Chem. Rev., 2005, 105, 1547.
4. P. C. Ray, G. K. Darbha, A. Ray, J. Walker and W. Hardy, Plasmonics, 2007, 2, 173.
5. B. Dubertret, M. Calame and A. J. Libchaber, Nat. Biotechnol., 2001, 19, 365.
6. D. J. Maxwell, J. R. Taylor and S. Nie, J. Am. Chem. Soc., 2002, 124, 9606.
7. H. Li and L. J. Rothberg, Anal. Chem., 2004, 76, 5414.
8. (a) S. Song, Z. Liang, J. Zhang, L. Wang, G. Li and C. Fan, Angew. Chem., Int. Ed., 2009, 48, 8670; (b) D. Li, S. Song and C. Fan, Acc. Chem. Res., 2010, 43, 631.
9. (a) R. Yang, Z. Tang, J. Yan, H. Kang, Y. Kim, Z. Zhu and W. Tan, Anal. Chem., 2008, 80, 7408; (b) R. Yang, J. Jin, Y. Chen, N. Shao, H. Kang, Z. Xiao, Z. Tang, Y. Wu, Z. Zhu and W. Tan, J. Am. Chem. Soc., 2008, 130, 8351.
10. (a) C. Lu, H. Yang, C. Zhu, X. Chen and G. Chen, Angew. Chem., Int. Ed., 2009, 48, 4785; (b) S. He, B. Song, D. Li, C. Zhu, W. Qi, W. Wen, L. Wang, S. Song, H. Fang and C. Fan, Adv. Funct. Mater., 2010, 20, 453.
11. (a) Y. Zhang, H. Li and X. Sun, Chinese J. Anal. Chem., 2011, 39, 998; (b) C. Liu, X. Yang, H. Yuan, Z. Zhou and D. Xiao, Sensors, 2007, 7, 708; (c) Y. Cao, X. Wu and M. Wang, Talanta, 2011, 84, 1188; (d) J. Geng, J. Liang, Y. Wang, G. G. Gurzadyan and B. Liu, J. Phys. Chem. B, 2011, 115, 3281; (e) H. I. Peng and B. L. Miller, Analyst, 2011, 136, 436; (f) H. I. Peng, T. D. Krauss and B. L. Miller, Anal. Chem., 2010, 82, 8664; (g) Y. Wang, B. Liu, A. Mikhailovsky and G. C. Bazan, Adv. Mater., 2010, 22, 656.
12. (a) S. Zhu, Z. Liu, W. Zhang, S. Han, L. Hu and G. Zhu, Chem. Commun., 2011, 47, 6099; (b) H. Li, J. Tian, L. Wang, Y. Zhang and X. Sun, J. Mater. Chem., 2011, 21, 824.
13. H. Li, Y. Zhang, L. Wang, J. Tian and X. Sun, Chem. Commun., 2011, 47, 961.
14. H. Li, Y. Zhang, T. Wu, S. Liu, L. Wang and X. Sun, J. Mater. Chem., 2011, 21, 4663.
15. H. Li, Y. Zhang, Y. Luo and X. Sun, Small, 2011, 7, 1562.
16. S. Liu, H. Li, L. Wang, J. Tian and X. Sun, J. Mater. Chem., 2011, 21, 339.
17. L. Wang, Y. Zhang, J. Tian, H. Li and X. Sun, Nucleic Acids Res., 2011, 39, e37.
18. S. Liu, L. Wang, Y. Luo, J. Tian, H. Li and X. Sun, Nanoscale, 2011, 3, 967.
19. J. Tian, H. Li, Y. Luo, L. Wang, Y. Zhang and X. Sun, Langmuir, 2011, 27, 874.
20. Y. Zhang, H. Li, Y. Luo, X. Shi, J. Tian and X. Sun, PLoS One, 2011, 6, e20569.
21. J. Tian, Y. Zhang, Y. Luo, H. Li, J. Zhai and X. Sun, Analyst, 2011, 136, 2221.
22. H. Li and X. Sun, Chem. Commun., 2011, 47, 2625.
23. H. Li, L. Wang, Y. Zhang, J. Tian and X. Sun, Macromol. Rapid Commun., 2011, 11, 899.
24. Y. Luo, F. Liao, W. Lu, G. Chang and X. Sun, Nanotechnology, 2011, 22, 195502.
25. H. Li, J. Zhai and X. Sun, RSC Advances, 2011 10.1039/c1ra00359c.
26. Y. Zhang, L. Wang, J. Tian, H. Li, Y. Luo and X. Sun, Langmuir, 2011, 27, 2170.
27. H. Li, L. Wang, J. Zhai, Y. Luo, Y. Zhang, J. Tian and X. Sun, Anal. Methods, 2011, 3, 1051.
28. H. Li, J. Zhai and X. Sun, PLoS One, 2011, 6, e18958.
29. J. Zhai, H. Li and X. Sun, RSC Advances, 2011, 1, 36.
30. (a) Polymer Colloids , ed. R. M. Fitch, Academic Press, San Diego, 1997; (b) Y. Xia, B. Gates, Y. Yin and Y. Lu, Adv. Mater., 2000, 12, 693.
31. R. D. Sochol, B. P. Casavant, M. E. Dueck, L. P. Lee and L. Lin, J. Micromech. Microeng., 2011, 21, 054019.
32. O. Warberg and W. Christian, Biochem., 1942, 310, 384.
33. N. Varghese, U. Mogera, A. Govindaraj, A. Das, P. K. Maiti, A. K. Sood and C. N. R. Rao, ChemPhysChem, 2009, 10, 206.
34. R. Matsushima and S. Sakuraba, J. Am. Chem. Soc., 1971, 93, 7143.

---

Footnote

† L. Wang and H. Li made equal contribution to this work.

---