Article pubs.acs.org/ac
Universal Surface-Enhanced Raman Scattering Amplification Detector for Ultrasensitive Detection of Multiple Target Analytes Jing Zheng, Yaping Hu, Junhui Bai, Cheng Ma, Jishan Li, Yinhui Li, Muling Shi, Weihong Tan, and Ronghua Yang* State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, College of Biology, and Collaborative Innovation Center for Chemistry and Molecular Medicine, Hunan University, Changsha, Hunan 410082, China S Supporting Information *
ABSTRACT: Up to now, the successful fabrication of efficient hot-spot substrates for surface-enhanced Raman scattering (SERS) remains an unsolved problem. To address this issue, we describe herein a universal aptamer-based SERS biodetection approach that uses a single-stranded DNA as a universal trigger (UT) to induce SERS-active hot-spot formation, allowing, in turn, detection of a broad range of targets. More specifically, interaction between the aptamer probe and its target perturbs a triple-helix aptamer/UT structure in a manner that activates a hybridization chain reaction (HCR) among three short DNA building blocks that self-assemble into a long DNA polymer. The SERSactive hot-spots are formed by conjugating 4-aminobenzenethiol (4-ABT)encoded gold nanoparticles with the DNA polymer through a specific Au−S bond. As proof-of-principle, we used this approach to quantify multiple target analytes, including thrombin, adenosine, and CEM cancer cells, achieving lowest limit of detection values of 18 pM, 1.5 nM, and 10 cells/mL, respectively. As a universal SERS detector, this prototype can be applied to many other target analytes through the use of suitable DNA-functional partners, thus inspiring new designs and applications of SERS for bioanalysis.
S
depend on the fabrication of many metal nanoparticle probes labeled with different oligonucleotides or Raman-active dyes. Thus, a new strategy that allows efficient conversion of various target recognition events into one cascaded hot-spot production reaction would be highly desirable. However, as noted above, the SERS signal still depends on a specific target, and selectivity for every possible target may not be available. Recently, our group has developed a triple-helix DNA switch for fluorescent detection of a variety of target analytes and fabrication of reversible and regenerable Raman-active substrate.16,17 To address the mentioned challenge of SERS and in our continuous effort to design SERS sensors for various analytes, we describe herein a convenient and universal SERS detector that employs a single-stranded DNA as a universal trigger (UT) to activate SERS hot-spot formation and a triplehelix aptamer/UT structure as the recognition element, allowing detection of a broad range of targets. Aptamers are single-stranded RNA or DNA oligonucleotides isolated from large random-sequence nucleic acid libraries by a unique in vitro selection method known as SELEX (systematic evolution of ligands by exponential enrichment), in which ligands are isolated from highly diverse (1013−1015) starting pools via rounds of affinity capture and amplification. They can meet the
urface-enhanced Raman scattering (SERS) has become one of the most valuable tools in chemistry, biology, and materials science, mainly because of the ultrahigh sensitivity, rich molecular structural information, and tremendous multiplexing capabilities of Raman spectroscopy.1,2 In particular, by integrating the extremely strong Raman enhancement of metallic substrates and the highly specific recognition ability of biomolecules, SERS has been used to create many novel bioanalytical tools, such as DNA sequence detection,3,4 protein assay,5−7 and bacterial or pathogen identification.8−10 In spite of considerable progress, the successful fabrication of efficient hot-spot-active substrates remains to be the fundamental and unsolved problem. Consequently, SERS has fallen well behind some of the traditional detection models, such as fluorescence, colorimetry, and electrochemistry.2 In SERS-based assays, SERS signaling is determined by the specific target, but the formed hot-spots are irreversible and not regenerable, making a hot-spot-active substrate usable for detecting only a single target. As such, it is necessary to fabricate various specific hotspot substrates for each target, thus hampering its large-scale application as a common detection tool.2,11 These problems could be solved by a method that would allow the use of only one hot-spot-active substrate to detect multiple target analytes. Thus, pioneered by the seminal work of Mirkin and colleagues as early as 2002,12 various methods for SERS detection of multiplexed oligonucleotides or proteins have been developed.13−15 However, these approaches still © 2014 American Chemical Society
Received: November 27, 2013 Accepted: January 18, 2014 Published: January 18, 2014 2205
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 1. Schematic illustration of the structure and operation principle of a universal SERS detector with amplified signal for the detection of multiple target analytes. (a) The molecular recognition probe is hybridized with a universal trigger (UT) to form a triple-helix structure via Watson Crick (-) and Hoogsteen (•) base pairings. Following addition of the target, interaction between the recognition probe and its target perturbs the triple-helix structure and causes the release of the UT. (b) Upon the release of UT, HCR is triggered between M1 and M2 on the SiMB surface to form a long double-stranded DNA polymer and, finally, conjugation of the 4-ABT-functionalized AuNPs on the DNA polymer forms a SERS-active substrate.
structure and causes the release of UT. This event triggers the three short DNA building blocks to self-assemble and forms a long DNA polymer on the SiMB surface via HCR. After separation, the SERS-active hot-spots are formed by conjugating 4-aminobenzenethiol (4-ABT)-encoded gold nanoparticles (4-ABT/AuNPs) with the DNA polymer by forming a Au−S bond. Thus, ideal enhancement of Raman signals occurs when the Raman reporter molecule 4-ABT resides in the proper gap between neighboring AuNPs.26,27 Since the targets of functional aptamers can range from metal ions and small organic molecules to biomolecules, and even viruses or cells,28 this new approach offers a general platform to realize detection of various analytes with one SERS-active substrate by simply selecting the appropriate aptamer. Moreover, because only one initiator is required to trigger a cascade of hybridization events to yield a long DNA polymer, numerous 4-ABT/AuNPs can be conjugated to the formed long DNA polymer and aggregated to amplify SERS signals, a process which, in turn, achieves ultrahigh sensitivity for multianalyte detection. The development of universal sensors that can detect a broad range of targets remains to be an ongoing interest by researchers;29−31 as a universal SERS detector, this prototype can be applied to a versatile platform for many biologically important target analytes through the use of suitable DNA-functional partners, thus inspiring new designs and applications of SERS for bioanalysis.
stringent requirements for such uses as bioassay, drug delivery, signal transduction, and gene expression mediation.18,19 Established SERS-based sensors combine the molecular recognition element and SERS signal transduction moiety in one approach.1−10,12−15,17 However, the basic design of our target-responsive SERS detector differs conceptually, in which we have separated the molecular recognition function from the SERS signal transduction module to signal various aptamer binding events but using only one SERS-active substrate. The relationship between target recognition and signal transduction is established by using a universal trigger, or UT, as the bridge. Then, to achieve efficient sensitivity for targets at lower concentration, hybridization chain reaction (HCR) is introduced to amplify the SERS signal. In HCR, two stable species of DNA hairpins coexist in solution, and a single DNA strand initiates the cascade of hybridization to form a long concatamer structure.20,21 Each copy of the initiator can propagate a chain reaction of hybridization events between two alternating singlestranded DNAs to form a nicked double-helix. As illustrated in Figure 1, the molecular recognition element is designed as a DNA triple-helix structure that consists of a central, targetspecific aptamer sequence flanked by two arm segments and UT via Watson Crick (-) and Hoogsteen (•) base pairings.22,16 Although double helix DNA molecular switches have been successfully developed for a signaling aptamer/target binding event,23−25 it is not suitable for use in our design, since the signal transduction probe is aptamer sequence specific and, hence, limited by the aptamer-binding event. In our proposed approach, the SERS signal transduction moiety consists of a short DNA building block (P0) immobilized on a silicon microbead (SiMB) surface and sulfhydryl-labeled hairpin helpers, M1 and M2. The three DNA building blocks do not interact with each other without the deployment of UT. Following addition of a target, interaction between the aptamer probe and its target perturbs the triple-helix aptamer/UT
■
RESULTS AND DISCUSSION SERS-Active Substrate Fabrication and DNA Sensing Performance. To induce HCR between M1 and M2, these two DNA strands were designed as hairpin structures. Each was labeled with a sulfhydryl functional group at the 5′-end. Each hairpin has a stem of 18 base pairs enclosing a hexanucleotide loop, with an additional hexanucleotide sticky end at the 5′-end of M1 (complementary to the loop of M2) and at the 3′-end of 2206
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 2. (A) Representative SEM images of the formed SERS-active hot-spot substrate on the surface of SiMBs. Inset: SEM images of the surface of SiMB without HCR. (B) AFM topographic images of (A), and (C) is the corresponding height profiles of the formed SERS-active hot-spot substrate on the surface of SiMBs. Scale area: 1 × 1.2 μm.
Figure 3. (A) SERS spectra of (SiMB-P0 + M1 + M2) + 4-ABT/AuNPs (a, black), (SiMB-P0 + M1 + M2) + 100 nM UT + 4-ABT/AuNP (b, blue), and (SiMB-P0 + M1) + 100 nM UT + 4-ABT/AuNPs (c, red). Inset: SERS spectrum of the 4-ABT/AuNPs. (B) SERS intensity enhancements of the 1430 cm−1-band of 4-ABT/AuNPs, I/I0, plotted against the concentration of UT in PBS without (a) and with HCR (b). The concentrations of SiMB-P0, M1, and M2 were 500 nM, 1.0 μM, and 1.0 μM, respectively. The amount of 4-ABT/AuNPs was in 100-fold excess of M1 or M2. All error bars were obtained through the detection of six parallel samples.
M2 (complementary to the loop of M1) (sequences were shown in Table S1; see Supporting Information). The UT strand in solution can pair with the sticky end of M1, which undergoes an unbiased strand-displacement interaction to open the hairpin. The newly exposed sticky end of M1 nucleates at the sticky end of M2 and opens the hairpin to expose a sticky end on M2. This sticky end is identical in sequence to the initiator strands. In this way, each UT strand propagates a HCR event between alternating M1 and M2 to form a nicked doublehelix. In this case, the sulfhydryl group on one building block is brought into close proximity to other such moieties. Thus, numerous sulfhydryl functional groups are self-assembled together, providing potential binding sites for the Raman dyeencoded AuNPs via specific Au−S bond. As proof-of-principle, HCR between UT and M1 and M2 was first examined by ethidium bromide (EB)-stained agarose gel electrophoresis in solution (Figure S1; see Supporting Information). In the absence of UT, HCR between M1 and M2 was inhibited, and only a bright band of M1 or M2 could be observed, as shown in lane 2. However, emission bands of high-molecular weight structures could also be observed in lanes of 3−6, indicating the successful growth of the long polymer in the presence of UT. Under these circumstances, a low concentration of UT was able to trigger a proportionately small chain reaction of alternating
kinetic escapes by M1 and M2, resulting in the formation of a long DNA polymer. To fabricate SERS-active hot-spot substrates, AuNPs were prepared and characterized according to a reported method32 and then functionalized with Raman reporter 4-ABT.33 To ensure that AuNPs could still be conjugated to the thiolatedDNA long polymer, the AuNPs were not fully covered with 4ABT. In comparison with the SERS-nanoparticle tags reported previously,34 the surface coverage of 4-ABT molecules on AuNPs was evaluated to be 0.8 ± 0.12 pmol/cm2, thus reducing reporter dye surface coverage by over 10-fold.35 Transmission electron microscopy (TEM) shows a size distribution of the 4ABT/AuNPs from 20 to 30 nm, most being 30 nm (Figure S2; see Supporting Information). The UV−vis absorption spectrum shows a maximal absorption central at 520 nm (Figure S3; see Supporting Information), which can be attributed to the surface plasmon resonance of AuNPs.36 To form SERS-active hot-spots along the DNA polymer, the 4-ABT/AuNPs were conjugated to the DNA polymer using a Au−S bond. After centrifugation at 5000 rpm for 10 min, a process which has since been optimized to attain the best separation effect (data not shown), representative TEM images clearly demonstrate that the AuNPs were self-assembled along with the heterochains of the DNA polymer (Figure S4; see Supporting Information). These 2207
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 4. Performances of SERS detection of Tmb. (A) SERS spectra of 4-ABT/AuNPs in PBS(a), the SERS-active hot-spot substrate induced by 50 nM Tmb addition in PBS (b), and the spectrum induced by 50 nM Tmb addition in 50% human serum (c). (B) Dependence of SERS intensity enhancement of the 1430 cm−1 band, I/I0, on different concentrations of Tmb (a), BSA (c), and IgG (d) in PBS and Tmb in a 50% human serum sample (b). The measuring conditions as shown in Figure 3. All error bars were obtained through the detection of six parallel samples.
1305, and 1430 cm−1 can be assigned to b2-modes of the 4ABT molecule,38 the respective assignments of observed SERS bands were shown in Table S2 (see Supporting Information). The SERS enhancement, I/I0, of the most prominent Raman peak at 1430 cm−1 was estimated to be 187.2-fold by 100 nM UT trigging the HCR, where I0 and I are the SERS intensities at 1430 cm−1 in the absence and the presence of UT, respectively. According to methods in the literature,39 an enhancement factor of ∼4.14 × 106 was determined (for details, see Supporting Information). Meanwhile, to demonstrate SERS signal amplification by HCR, a control experiment was designed by directly employing SiMB-P0 and M1. A short DNA hybrid was formed among P0, UT, and M1 upon addition of UT, and only a few 4-ABT/AuNPs were loaded, resulting in a weak SERS signal state (curve c). In this case, the value of I/I0 dropped to 3.23 upon the addition of 100 nM UT, demonstrating the effective signal enhancement of our proposed platform. To assess the potential of this SERS detector for quantitative bioanalysis by HCR, the most prominent Raman peak at 1430 cm−1 of 4-ABT was compared to different concentrations of UT (Figure S6, see Supporting Information). Figure 3B reveals that the SERS signal dramatically increased upon increasing UT concentration from 0.01 to 150 nM. The value of I/I0 linearly increased with the UT concentration between 0.01 and 100 nM. This is mainly because, as the concentration of UT changed (from 0 to about 100 nM), the copies of M1 and M2 assembled together increased.Thus, numerous sulfhydryl functional groups are selfassembled together in our design, providing potential binding sites for the Raman dye-encoded AuNPs via specific Au−S bond, increasing SERS signal is observed. However, when the UT increased to about 100 nM, or even higher, one should also note that the SERS signal remained fairly constant due to the HCR system finally reaching the equilibrium state in terms of kinetics and thermodynamics. SERS Detection of Biomolecules. To demonstrate the feasability of our proposed approach as a universal SERS detector for ultrasensitive bioanalysis, human α-thrombin (Tmb) was chosen as a model by using anti-Tmb aptamer as the capture probe for Tmb. Tmb (activated Factor II) is a specific serine protease involved in the coagulation cascade, which converts soluble fibrinogen into insoluble strands of fibrin and catalyzes many other coagulation-related reactions.40 It is well established that aptamers “adaptively bind” target
collective results demonstrated that the UT had actually assisted in the self-assembly of nicked-DNA polymer structure and, subsequently, the 4-ABT/AuNPs conjugates. Finally, to fabricate the HCR-based SERS-active substrate on a large solid surface to attain preferable separating effect and eliminate the false positive signal produced by free 4-ABT/ AuNPs and the conjugation of 4-ABT/AuNPs and free M1/ M2. A capture probe, P0, for UT was employed (sequences were shown in Table S1; see Supporting Information) and immobilized on the surface of SiMBs through the streptavidin− biotin interaction. The surface density of streptavidin on SiMBs using biotin-labeled P0 was estimated to be 1.89 × 102/μm2. In the presence of UT, P0 hybridizes with one terminus of UT and served as a link to effectively initiate HCR that formed a long DNA polymer between M1 and M2 on the surface of SiMBs in the 0.1 M sodium phosphate buffer (PBS, 5 mM Mg2+, pH 7.4). 4-ABT/AuNPs were then added; after incubation at room temperature for 1.0 h, the mixtures were washed three times with PBS using centrifugation at 1000 rpm for 10 min to remove free 4-ABT/AuNPs, and the conjugates of 4-ABT/ AuNPs and M1 or M2. Scanning electron microscope (SEM) images showed that the loaded AuNPs were very closely aggregated on the surface of SiMBs, and a few bright spots appeared in the image (Figure 2A). However, the control experiment, in which UT was absent, showed no observable aggregates on the surface of SiMB (Figure 2A, inset). We further used atomic force microscopy (AFM) images to characterize the surface features of the formed AuNPs aggregates on SiMBs, which exhibit a mean height between 120 and 200 nm (Figure 2B,C), being largely higher than that of the free 4-ABT/AuNPs (Figure S5; see Supporting Information). Therefore, these special Raman reporter-encoded aggregates have the potential to provide a significant amplification of Raman signal intensity by several orders of magnitude through electromagnetic field enhancement. Figure 3A showed a set of SERS spectra of free 4-ABT/ AuNPs and UT-triggered formation of SERS-active hot-spot substrate on the surface of SiMBs. Either the free 4-ABT/ AuNPs or the mixture of SiMB-P0 + M1 + M2 + 4-ABT/ AuNPs shows very weak SERS signals by the weak interparticle plasmon coupling.37 However, upon addition of UT to the mixture, a stronger SERS signal was demonstrated by the aggregated 4-ABT/AuNPs on the surface of SiMB through HCR (curve b).The observed Raman bands centered at 1137, 2208
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 5. Performances of SERS detection of adenosine. (A) SERS intensity ratio of the 1430 cm−1-band, I/I0, plotted against the concentration of adenosine. The inset shows a linear relationship between I/I0 and the adenosine concentration. (B) Selectivity of this SERS amplification detector with 1.0 μM adenosine, guanosine, cytidine, and uridine. The measuring conditions as shown in Figure 3. All error bars were obtained through the detection of six parallel samples.
obvious SERS signals, even with a higher target concentration (Figure 4Bc,d). This SERS amplification detector even showed good sensitivity and selectivity in pure buffer; even to be more useful in the bioassay, the detection system should be able to tolerate any interference from real biological samples. To demonstrate such tolerance, pretreated blood serum was used to detect Tmb in a complex sample. Figure 4Ac was the signal enhancement of our SERS amplification detector upon 50 nM Tmb addition in 50% human serum, suggesting that the other components in blood serum provided little or no interference with the performance of our detector. To give a quantitative response to Tmb in a real biological sample, human serum sample spiked with different concentrations of Tmb was tested with our method. The SERS detector displayed a sensitive SERS response to Tmb in a 50% undiluted human blood sample (Figure 4Bb), which was close to that obtained in PBS buffer; however, even in serum contents >50%, this detector could also function well (data not shown). The result obtained in a human serum sample thereby demonstrates the wide applicability of this method in complex body fluids. To demonstrate the generality of this SERS detector for other small biomolecules, we also used an adenosine aptamer sequence flanked by two arm segments (P2−1−P2−4, Table S1; see Supporting Information) to detect adenosine, an important biological cofactor involved in many biological processes, such as those in the kidney and urine.41 Similar to the Tmb aptamerbased sensor described above, the extended adenosine aptamer (Ade-Apt) was used as the capture probe to hybridize with UT to form the triple-helix Ade-Apt/UT structure. The lengths of the two arms of Ade-Apt were optimized according to our previous report to achieve the best SERS response sensitivity, and P2−2 was chosen as a preferable capture probe for the subsequent experiment (Figure S11, Figure S12; see Supporting Information).16 As expected, in the absence of adenosine, UT could not be released from the triple-helix P2−2/UT structure, thus preventing HCR. However, upon the addition of adenosine to the solution of P2−2 and UT, the strong interaction between P2−2 and adenosine caused the disassembly of the triple-helix P2−2/UT structure. This allowed the released UT to initiate HCR, and an obvious SERS signal could be observed (Figure S13; see Supporting Information). Figure 5A showed the SERS signal was proportional to the concentrations of adenosine from 5.0 to 500 nM, establishing the quantitative detection capability of this SERS detector. A LOD of 1.5 nM
molecules, and a conformational change usually accompanies such binding events. Thus, they have become a key recognition element to detect various target analytes. To realize Tmb detection, the molecular recognition element was designed as a triple-helix aptamer/UT structure consisting of the UT and a central, Tmb-specific aptamer sequence flanked by two arm segments (P1−1−P1−4, Table S1; see Supporting Information). First, P1−2′ (sequence was shown in Supporting Information), which was dual-labeled with tetramethylrhodamine(TAMRA) and BHQ2 at each end, was employed to investigate the hybridization efficiency of the triple-helix, data shown in Figure S7 (see Supporting Information). To achieve the best response sensitivity, the DNA sequence and length of the arm segments of P1−2 were adapted from our recently reported work,16 which contained 7 bases in the stem as the optimal capture probe in subsequent experiments (Figures S8 and S9; see Supporting Information). We measured the SERS spectra of SERS-active hot-spot substrate on the SiMBs surface triggered by UT released from the triple-helix aptamer/UT structure. Figure 4Aa showed that, in the absence of Tmb, no SERS spectrum was observed, indicating that the triple-helix aptamer/UT structure remained stable, and no HCR occurred among the three DNA building blocks. However, upon addition of 50.0 nM Tmb to the solution, a strong SERS signal was observed (Figure 4Ab). To demonstrate the applicability of this SERS detector for the quantitative detection of Tmb, SERS singal enhancement was attained in the presence of different concentrations of Tmb in PBS (Figure S10; see Supporting Information). As shown in Figure 4Ba, depending on the Tmb concentration range, there were different sensitivities. The small but perceptible SERS change at low concentration of Tmb indicated that Tmb molecules were capable of interacting with the triple-helix aptamer/UT structure to release UT, even at very low concentration. In the concentration range of 0.05−100 nM Tmb, a dramatic SERS enhancement was observed. At >100 nM concentrations of Tmb, no further increase in SERS signal was observed and a plateau was reached with a maximum I/I0 value of a nearly 170. The limit of detection (LOD), based on 3σb/slope,16 where σb was the standard deviation of blank samples, was 18 pM. In addition, since the in vitro-selected antiTmb aptamer possesses high affinity for Tmb, we found that other proteins, such as IgG and BSA, which were commonly present in a complex environment, could not induce any 2209
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 6. Performance of SERS detection of cancer cells using the SERS detector. (A) SERS spectra corresponding to the addition of different concentrations (number of cells/mL) of CEM cells. The arrow indicates the signal changes as CEM concentrations increase (0, 1 × 102, 5 × 102, 1 × 103, and 1 × 104 cells/mL). (B) Plots demonstrating changes of Raman intensity at 1430 cm−1 upon the addition of different concentrations (number of cells/mL) of Ramos cells (a) and the target CEM cells (b). The measuring conditions as shown in Figure 3. All error bars were obtained through the detection of six parallel samples.
BHQ2 at the 5′-end, respectively. From Figure S15, Supporting Information, it can be seen that the fluorescence intensity of P3−2 was obviously decreased by the formation of the triplehelix aptamer/UT structure, while fluorescence recovery could be achieved upon target cell binding. Thus, we confirmed that the constructed triple-helix aptamer/UT structure can function well upon target cells addition and cause UT release. We next optimized the arm length of the different capture probes upon the addition of UT and subsequent target CEM, finally attaining amplified SERS signal readout through HCR. Figure S16, Supporting Information, showed that P3−2 would be a preferable capture probe sequence to attain higher I/I0. On the basis of this result, P3−2 was chosen as an optimal capture probe in subsequent experiments. For the sensitive detection of target cancer cells, different concentrations of CEM from one stock solution were attained, and we measured the SERS spectra of SiMB-TPS in PBS buffer. As shown in Figure 6A, the SERS intensity of 1430 cm−1 was highly sensitive to the concentration of CEM cells. When CEM cells were absent, only a weak peak was observed, indicating that the triple-helix aptamer/UT structure remained stable and that HCR could not be triggered without UT. This was mainly due to the two long hairpins M1 and M2, both have a stem of 18 base pairs enclosing a hexanucleotide loop. The UT can pair with the sticky end of M1, which undergoes an unbiased strand-displacement interaction to open the hairpin. The newly exposed sticky end of M2 nucleates at the sticky end of M2 and opens the hairpin to expose a sticky end on M2. This sticky end is identical in sequence to the initiator strands. In this way, each UT propagates a chain reaction of hybridization events between alternating M1 and M2 hairpins to form a nicked double-helix. On the contrary, in the absence of UT, M1 and M2 both keep stable hairpin structure in the solution and HCR could not be triggered. Therefore, in our experiment, when the target cell is absent, we could not attain the released UT from the triplehelix structure. Higher concentrations of CEM cells, however, led to a correspondingly stronger SERS signal. Figure 6B clearly demonstrated that the sensitivity of our platform was as low as 10 cells. To evaluate whether our assay is highly selective, we also studied how SERS intensity changed upon the addition of negative control Ramos cells. SERS intensity was negligible in the presence of 104 Ramos cells in the solution containing TPS. Since Ramos cells do not overexpress PTK7, only a weak interaction will be observed between TPS and Ramos cells.
adenosine was observed. Meanwhile, control experiments with adenosine analogues, such as guanosine, cytidine, and uridine, did not produce any enhanced SERS readings, indicating good selectivity for this SERS amplification detector (Figure 5B). The success of the Tmb and adenosine detections unequivocally supported the feasibility and versatility of our strategy. Moreover, based on SERS signal amplification via HCR, our SERS detector displayed ultrahigh sensitivity, exceeding that of a variety of previously reported aptamer-based sensors for Tmb and adenosine.42,43 SERS Detection of Human Cancer Cells. To further assess its biomedical and diagnostic applications, our SERS detector was investigated for cancer cell identification. It has been hypothesized that very sensitive monitoring of cancer cells could provide an easier and more effective way to monitor progression of the disease.44 From this perspective, identification and detection of cancer cells is fundamental to early diagnosis and therapy, as well as monitoring the relevant biological processes of cancers. For performance of human cancer cells detection, aptamer sgc8c,45 which binds the cell membrane protein tyrosine kinase-7 (PTK7) with high affinity and selectivity, was used in our study with different types of cell lines, including target cancer cell CCRF-CEM (acute lymphoblastic leukemia T-cells) and negative control Ramos cells (acute lymphoblastic leukemia B-cells).45 For optimization of the triple-helix aptamer/UT structure to achieve best detection sensitivity, we designed a series of capture probes (P3−1−P3−4, Table S1; see Supporting Information) comprising an aptamer sequence sgc8c and two arm segments of d(TC)n complementary to the sequence of UT. A poly-T chain links the 5′-end of capture probe to the two arm segments of d(TC) n. The binding of extended aptamer with CEM cells was studied by flow cytometry (Figure S14; see Supporting Information). After incubation with 5′-end-labeled carboxyfluorescein (FAM) of P3−1−P3−4, Figure S14, Supporting Information, showed that the larger fluorescence intensity changes of CEM cells corresponded to the stronger binding of capture probe to cells. This result showed the specific recognition of capture probe was preserved after the extension of aptamer sgc8. We then constructed an effective triplex switch consisting of UT and P3−1−P3−4. The formation of triplex was further confirmed by monitoring the fluorescence intensity of the capture probe P3−2 (P3−1−P3−4, Table S1; see Supporting Information) which was labeled with TAMRA at the 3′-end and 2210
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
three times with PBS using centrifugation at 1000 rpm to remove free 4-ABT/AuNPs and the conjugation of 4-ABT/ AuNPs and M1/M2. Characterization. For an extended aptamer binding assay, CCRF-CEM and negative control Ramos cells were washed with washing buffer and resuspended in binding buffer. Cells and extended aptamer, both labeled with FITC, were incubated on ice for 30 min. After washing with 1.0 mL of washing buffer, the cells were tested by a FACScan cytometer (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA) by counting 20 000 events. Raman spectra were obtained using a confocal microprobe Raman instrument (RamLab-010, Horiba Jobin Yvon, France), and the 632.8 nm He−Ne laser was used to excite the spectra, which were collected in a backscattering geometry. A 50× long working length objective (8 mm) was used in this work. The collection time for each spectrum was 20 s. The width of the slit and the size of the pinhole were set as 100 and 100 μm, respectively.
This lack of strong interaction prevented UT from being released to trigger HCR. Consequently, Raman enhancement was not observed, again demonstrating that our strategy is highly specific for CEM cells.
■
CONCLUSION In summary, an ultrasensitive SERS approach for detection of various target analytes with one Raman-active substrate has been developed for the first time. The design is based on the separation of molecular recognition element and SERS signal transduction moiety using a universal trigger to initiate cascaded Raman-active hot spots via HCR. The present approach can be engineered in ways that offer unique advantages and capabilities that are not available from conventional Raman detection systems. First of all, it is generalizable: when different analytes are present, we just need to change different label-free aptamer sequences in the triplehelix structure. For the signal transduction element, we have fabricated one hot active substrate using one UT, biotin-labeled P0, sulfhydryl-labeled M1/M2, which would solve the problem of resynthesis of various detection probes or even hot spotactive substrate. Second, by introducing HCR to amplify the SERS signal, ultrahigh sensitivity can be achieved, and thus, pM level of target biomolecules or single-cell level human cancer cells could be detected. Finally, and more importantly, although we used this approach to detect three target analytes, the strategy is versatile by the ability to select not only different types of aptamers for the capture probe, but also different types of signal transducers, such as fluorescence,46 colorimetry,47 or electrochemical,48 for detection assay, as well as drug delivery and multiple-stimulated release.49,50 Given the low cost, ultrahigh sensitivity, and excellent generality, we anticipate that this approach might open up new opportunities for the development of new bioanalytical and biomedical applications.
■
ASSOCIATED CONTENT
* Supporting Information S
More experimental details and additional spectroscopic data as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Fax: +86-731-8882 2523. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We are grateful for the financial support from the National Natural Science Foundation of China (21075032, 21005026, 21135001, 21305036, and J1103312), The Foundation for Innovative Research Groups of NSFC (21221003), and the “973”National Key Basic Research Program (2011CB91100-0). Zheng. J received financial support from Growth Program for Young Teachers in Hunan University.
■
MATERIALS AND METHODS Materials and Instruments. All oligonucleotides were synthesized by TaKaRa Biotechnology Co. Ltd. (Dalian, China). All chemicals were used as received, unless otherwise stated. Streptavidin-coated silicon beads (SiMBs, 5 μm) were purchased from Ocean Nanotech and dispersed at 0.1 mg/mL in 100 mM phosphate-buffered saline (PBS), pH 7.4. Cell lines CCRF-CEM (acute lymphoblastic leukemia T-cells) and Ramos (acute lymphoblastic leukemia B-cells) were obtained from the American Type Culture Collection (Manassas, VA). Construction of SERS-Active Substrate on the Surface of SiMBs. The biotin-labeled P0 was first incubated with streptavidin-coated SiMBs. The mixtures were vortexed at room temperature for 1 h, followed by washing three times with PBS using centrifugation at 1000 rpm to remove any DNA that did not conjugate to the SiMBs. The conjugates were dispersed in PBS and stored at 4 °C at a concentration of 0.1 mg/mL. Then, UT was added to the solution containing P0conjugated nanoparticles, and the mixture was vortexed at room temperature for another 12 h, followed by washing 3 times with PBS using centrifugation at 1000 rpm to remove triple-helix aptamer/UT and aptamer/target complex. Next, SH-labeled M1/M2 was added and reacted at room temperature for HCR 12 h, followed by washing three times with PBS using centrifugation at 1000 rpm to remove free SH-labeled M1/M2. Finally, 4-ABT/AuNPs were added to the above attained SiMBs and incubated for 2 h, followed by washing
■
REFERENCES
(1) Qian, X. M.; Nie, S. M. Chem. Soc. Rev. 2008, 37, 912−920. (2) Wang, Y. Q.; Yan, B.; Chen, L. X. Chem. Rev. 2013, 113, 1391− 1428. (3) Fabris, L.; Dante, M.; Braun, G.; Lee, S. J.; Reich, N. O.; Moskovits, M.; Nguyen, T. Q.; Bazan, G. C. J. Am. Chem. Soc. 2007, 129, 6086−6087. (4) Thompson, D. G.; Enright, A.; Faulds, K.; Smith, W. E.; Graham, D. Anal. Chem. 2008, 80, 2805−2810. (5) Grubisha, D. S.; Lipert, R. J.; Park, H. Y.; Driskell, J.; Porter, M. D. Anal. Chem. 2003, 75, 5936−5943. (6) Bonham, A. J.; Braun, G.; Pavel, L.; Moskovits, M.; Reich, N. O. J. Am. Chem. Soc. 2007, 129, 14572−14573. (7) Narayanan, R.; Lipert, R. J.; Porter, M. D. Anal. Chem. 2008, 80, 2265−2271. (8) Lee, A.; Kim, S.; Choo, J.; Shin, S. Y.; Lee, Y. H.; Choi, H. Y.; Ha, S.; Kang, K.; Oh, C. H. Anal. Chem. 2007, 79, 916−922. (9) Tang, H. W.; Yang, X. B.; Kirkham, J.; Smith, D. A. Anal. Chem. 2007, 79, 3646−3653. (10) Hu, Q.; Tay, L. L.; Noestheden, M.; Pezacki, J. P. J. Am. Chem. Soc. 2007, 129, 14−15. (11) Banholzer, M. J.; Millstone, J. E.; Qin, L. D.; Mirkin, C. A. Chem. Soc. Rev. 2008, 37, 885−897. 2211
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
(12) Cao, Y. W.; Jin, R. C.; Mirkin, C. A. Science 2002, 30, 1536− 1540. (13) Cao, Y. W.; Jin, R. C.; Nam, J. M.; Thaxton, C. M.; Mirkin, C. A. J. Am. Chem. Soc. 2003, 125, 14676−14677. (14) Su, X.; Zhang, J.; Sun, L.; Koo, T. W.; Chan, S.; Sundararajan, N.; Yamakawa, M.; Berlin, A. A. Nano Lett. 2005, 5, 49−54. (15) Wang, Z. Y.; Zong, S. F.; Li, W.; Wang, C. L.; Xu, S. H.; Chen, H.; Cui, Y. P. J. Am. Chem .Soc. 2012, 134, 2993−3000. (16) Zheng, J.; Li, J. S.; Jiang, Y.; Jin, J. R.; Wang, K. M.; Yang, R. H.; Tan, W. H. Anal. Chem. 2011, 83, 6586−6592. (17) Zheng, J.; Jiao, A. L.; Yang, R. H.; Li, H. M.; Li, J. S.; Shi, M. L.; Ma, C.; Tan, W. H. J. Am. Chem. Soc. 2012, 134, 19957−19960. (18) Ellington, A. D.; Szostak, J. W. Nature 1990, 346, 818−822. (19) Tuerk, C.; Gold, L. Science 1990, 249, 505−510. (20) Dirks, R. M.; Pierce, N. A. Proc. Nat. Acad. Sci. U.S.A. 2004, 101, 15275−15278. (21) Yin, P.; Choi, H. M.; Calvert, C. R.; Pierce, N. A. Nature 2008, 451, 318−322. (22) Salunkhe, M.; Wu, T. F.; Letsinger, R. L. J. Am. Chem. Soc. 1992, 114, 8768−8772. (23) Nutiu, R.; Li, Y. F. J. Am. Chem. Soc. 2003, 125, 4771−4778. (24) Li, D.; Shlyahovsky, B.; Elbaz, J.; Willner, I. J. Am. Chem. Soc. 2007, 129, 5804−5805. (25) Li, N.; Ho, C. M. J. Am. Chem. Soc. 2008, 130, 2380−2381. (26) Shegai, T.; Vaskevich, A.; Rubinstein, I.; Haran, G. J. Am. Chem. Soc. 2009, 131, 14390−14398. (27) Li, J. F.; Huang, Y. F.; Ding, Y.; Yang, Z. L.; Li, S. B.; Zhou, X. S.; Fan, F. R.; Zhang, W.; Zhou, Z. Y.; Wu, D. Y.; Ren, B.; Wang, Z. L.; Tian, Z. Q. Nature 2010, 464, 392−395. (28) Liu, J. W.; Cao, Z. H.; Lu, Y. Chem. Rev. 2009, 109, 1948−1998. (29) Giepmans, B. N. G.; Adams, S. R.; Ellisman, M. H.; Tsien, R. Y. Science 2006, 312, 217−224. (30) Xiang, Y.; Lu, Y. Nat. Chem. 2011, 3, 697−703. (31) Das, J.; Cederquist, K. B.; Zaragoza, A. A.; Lee, P. E.; Sargent, E. H.; Kelley, S. O. Nat. Chem. 2012, 4, 642−648. (32) Liu, J. W.; Lu, Y. Nat. Protoc. 2006, 1, 246−252. (33) Kim, K.; Lee, H. B.; Yoon, J. K.; Shin, D.; Shin, K. S. J. Phys. Chem. C 2010, 114, 13589−13585. (34) Qian, X. M.; Peng, X. H.; Ansari, D. O.; Chen, G. Z.; Shin, D. M.; Yang, L.; Young, A. N.; Wang, M. D.; Nie, S. M. Nat. Biotechnol. 2008, 26, 83−90. (35) Qian, X. M.; Zhou, X.; Nie, S. M. J. Am. Chem. Soc. 2008, 130, 14934−14935. (36) Jana, N. R.; Gearheart, L.; Murphy, C. J. Langmuir 2001, 17, 6782−6786. (37) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat. Mater. 2008, 7, 442−453. (38) Dendisova, M.; Havranek, L.; Oncak, M.; Matejka, P. J. Phys. Chem. C 2013, 117, 21245−21253. (39) Braun, G.; Lee, S. J.; Dante, M.; Nguyen, T. Q.; Moskovits, M.; Reich, N. J. Am. Chem. Soc. 2007, 129, 6378−6379. (40) Holland, C. A.; Henry, A. T.; Whinna, H. C.; Church, F. C. FEBS Lett. 2000, 484, 87−91. (41) Vallon, V.; Muhlbauer, B.; Osswald, H. Physiol. Rev. 2006, 86, 901−940. (42) Cho, H.; Baker, B. R.; Hogiu, S. W.; Lee, L. P.; Tok, J. B. Nano Lett. 2008, 8, 4386−4390. (43) Li, L. L.; Ge, P. H.; Selvin, P. R.; Lu, Y. Anal. Chem. 2012, 84, 7852−7856. (44) Lu, W.; Singh, A. K.; Khan, S. A.; Senapati, D.; Yu, H.; Ray, P. C. J. Am. Chem. Soc. 2010, 132, 18103−18114. (45) Shangguan, D. H.; Li, Y.; Tang, Z. W.; Cao, Z. H.; Chen, H. W.; Tan, W. H. Proc. Nat. Acad. Sci. U.S.A. 2006, 103, 11838−11843. (46) Huang, J.; Wu, Y. R.; Chen, Y.; Yang, C. Y.; Wang, K. M.; Tan, W. H. Angew. Chem., Int. Ed. 2011, 50, 401−404. (47) Shimron, S.; Wang, F.; Orbach, R.; Willner, I. Anal. Chem. 2012, 84, 1042−1048. (48) Zhang, B.; Liu, B. Q.; Tang, D. P.; Niessner, R.; Chen, G. N.; Knopp, D. Anal. Chem. 2012, 84, 5392−5399.
(49) Zhu, G. Z.; Zheng, J.; Song, E. Q.; Donovana, M.; Liu, C.; Tan, W. H. Proc. Nat. Acad. Sci. U.S.A. 2013, 110, 7998−8003. (50) Zheng, J.; Zhu, G. Z.; Li, Y. H.; Yang, R. H.; Tan, W. H. ACS Nano 2013, 7, 6545−6554.
2212
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Universal Surface-Enhanced Raman Scattering Amplification Detector for Ultrasensitive Detection of Multiple Target Analytes Jing Zheng, Yaping Hu, Junhui Bai, Cheng Ma, Jishan Li, Yinhui Li, Muling Shi, Weihong Tan, and Ronghua Yang* State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, College of Biology, and Collaborative Innovation Center for Chemistry and Molecular Medicine, Hunan University, Changsha, Hunan 410082, China S Supporting Information *
ABSTRACT: Up to now, the successful fabrication of efficient hot-spot substrates for surface-enhanced Raman scattering (SERS) remains an unsolved problem. To address this issue, we describe herein a universal aptamer-based SERS biodetection approach that uses a single-stranded DNA as a universal trigger (UT) to induce SERS-active hot-spot formation, allowing, in turn, detection of a broad range of targets. More specifically, interaction between the aptamer probe and its target perturbs a triple-helix aptamer/UT structure in a manner that activates a hybridization chain reaction (HCR) among three short DNA building blocks that self-assemble into a long DNA polymer. The SERSactive hot-spots are formed by conjugating 4-aminobenzenethiol (4-ABT)encoded gold nanoparticles with the DNA polymer through a specific Au−S bond. As proof-of-principle, we used this approach to quantify multiple target analytes, including thrombin, adenosine, and CEM cancer cells, achieving lowest limit of detection values of 18 pM, 1.5 nM, and 10 cells/mL, respectively. As a universal SERS detector, this prototype can be applied to many other target analytes through the use of suitable DNA-functional partners, thus inspiring new designs and applications of SERS for bioanalysis.
S
depend on the fabrication of many metal nanoparticle probes labeled with different oligonucleotides or Raman-active dyes. Thus, a new strategy that allows efficient conversion of various target recognition events into one cascaded hot-spot production reaction would be highly desirable. However, as noted above, the SERS signal still depends on a specific target, and selectivity for every possible target may not be available. Recently, our group has developed a triple-helix DNA switch for fluorescent detection of a variety of target analytes and fabrication of reversible and regenerable Raman-active substrate.16,17 To address the mentioned challenge of SERS and in our continuous effort to design SERS sensors for various analytes, we describe herein a convenient and universal SERS detector that employs a single-stranded DNA as a universal trigger (UT) to activate SERS hot-spot formation and a triplehelix aptamer/UT structure as the recognition element, allowing detection of a broad range of targets. Aptamers are single-stranded RNA or DNA oligonucleotides isolated from large random-sequence nucleic acid libraries by a unique in vitro selection method known as SELEX (systematic evolution of ligands by exponential enrichment), in which ligands are isolated from highly diverse (1013−1015) starting pools via rounds of affinity capture and amplification. They can meet the
urface-enhanced Raman scattering (SERS) has become one of the most valuable tools in chemistry, biology, and materials science, mainly because of the ultrahigh sensitivity, rich molecular structural information, and tremendous multiplexing capabilities of Raman spectroscopy.1,2 In particular, by integrating the extremely strong Raman enhancement of metallic substrates and the highly specific recognition ability of biomolecules, SERS has been used to create many novel bioanalytical tools, such as DNA sequence detection,3,4 protein assay,5−7 and bacterial or pathogen identification.8−10 In spite of considerable progress, the successful fabrication of efficient hot-spot-active substrates remains to be the fundamental and unsolved problem. Consequently, SERS has fallen well behind some of the traditional detection models, such as fluorescence, colorimetry, and electrochemistry.2 In SERS-based assays, SERS signaling is determined by the specific target, but the formed hot-spots are irreversible and not regenerable, making a hot-spot-active substrate usable for detecting only a single target. As such, it is necessary to fabricate various specific hotspot substrates for each target, thus hampering its large-scale application as a common detection tool.2,11 These problems could be solved by a method that would allow the use of only one hot-spot-active substrate to detect multiple target analytes. Thus, pioneered by the seminal work of Mirkin and colleagues as early as 2002,12 various methods for SERS detection of multiplexed oligonucleotides or proteins have been developed.13−15 However, these approaches still © 2014 American Chemical Society
Received: November 27, 2013 Accepted: January 18, 2014 Published: January 18, 2014 2205
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 1. Schematic illustration of the structure and operation principle of a universal SERS detector with amplified signal for the detection of multiple target analytes. (a) The molecular recognition probe is hybridized with a universal trigger (UT) to form a triple-helix structure via Watson Crick (-) and Hoogsteen (•) base pairings. Following addition of the target, interaction between the recognition probe and its target perturbs the triple-helix structure and causes the release of the UT. (b) Upon the release of UT, HCR is triggered between M1 and M2 on the SiMB surface to form a long double-stranded DNA polymer and, finally, conjugation of the 4-ABT-functionalized AuNPs on the DNA polymer forms a SERS-active substrate.
structure and causes the release of UT. This event triggers the three short DNA building blocks to self-assemble and forms a long DNA polymer on the SiMB surface via HCR. After separation, the SERS-active hot-spots are formed by conjugating 4-aminobenzenethiol (4-ABT)-encoded gold nanoparticles (4-ABT/AuNPs) with the DNA polymer by forming a Au−S bond. Thus, ideal enhancement of Raman signals occurs when the Raman reporter molecule 4-ABT resides in the proper gap between neighboring AuNPs.26,27 Since the targets of functional aptamers can range from metal ions and small organic molecules to biomolecules, and even viruses or cells,28 this new approach offers a general platform to realize detection of various analytes with one SERS-active substrate by simply selecting the appropriate aptamer. Moreover, because only one initiator is required to trigger a cascade of hybridization events to yield a long DNA polymer, numerous 4-ABT/AuNPs can be conjugated to the formed long DNA polymer and aggregated to amplify SERS signals, a process which, in turn, achieves ultrahigh sensitivity for multianalyte detection. The development of universal sensors that can detect a broad range of targets remains to be an ongoing interest by researchers;29−31 as a universal SERS detector, this prototype can be applied to a versatile platform for many biologically important target analytes through the use of suitable DNA-functional partners, thus inspiring new designs and applications of SERS for bioanalysis.
stringent requirements for such uses as bioassay, drug delivery, signal transduction, and gene expression mediation.18,19 Established SERS-based sensors combine the molecular recognition element and SERS signal transduction moiety in one approach.1−10,12−15,17 However, the basic design of our target-responsive SERS detector differs conceptually, in which we have separated the molecular recognition function from the SERS signal transduction module to signal various aptamer binding events but using only one SERS-active substrate. The relationship between target recognition and signal transduction is established by using a universal trigger, or UT, as the bridge. Then, to achieve efficient sensitivity for targets at lower concentration, hybridization chain reaction (HCR) is introduced to amplify the SERS signal. In HCR, two stable species of DNA hairpins coexist in solution, and a single DNA strand initiates the cascade of hybridization to form a long concatamer structure.20,21 Each copy of the initiator can propagate a chain reaction of hybridization events between two alternating singlestranded DNAs to form a nicked double-helix. As illustrated in Figure 1, the molecular recognition element is designed as a DNA triple-helix structure that consists of a central, targetspecific aptamer sequence flanked by two arm segments and UT via Watson Crick (-) and Hoogsteen (•) base pairings.22,16 Although double helix DNA molecular switches have been successfully developed for a signaling aptamer/target binding event,23−25 it is not suitable for use in our design, since the signal transduction probe is aptamer sequence specific and, hence, limited by the aptamer-binding event. In our proposed approach, the SERS signal transduction moiety consists of a short DNA building block (P0) immobilized on a silicon microbead (SiMB) surface and sulfhydryl-labeled hairpin helpers, M1 and M2. The three DNA building blocks do not interact with each other without the deployment of UT. Following addition of a target, interaction between the aptamer probe and its target perturbs the triple-helix aptamer/UT
■
RESULTS AND DISCUSSION SERS-Active Substrate Fabrication and DNA Sensing Performance. To induce HCR between M1 and M2, these two DNA strands were designed as hairpin structures. Each was labeled with a sulfhydryl functional group at the 5′-end. Each hairpin has a stem of 18 base pairs enclosing a hexanucleotide loop, with an additional hexanucleotide sticky end at the 5′-end of M1 (complementary to the loop of M2) and at the 3′-end of 2206
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 2. (A) Representative SEM images of the formed SERS-active hot-spot substrate on the surface of SiMBs. Inset: SEM images of the surface of SiMB without HCR. (B) AFM topographic images of (A), and (C) is the corresponding height profiles of the formed SERS-active hot-spot substrate on the surface of SiMBs. Scale area: 1 × 1.2 μm.
Figure 3. (A) SERS spectra of (SiMB-P0 + M1 + M2) + 4-ABT/AuNPs (a, black), (SiMB-P0 + M1 + M2) + 100 nM UT + 4-ABT/AuNP (b, blue), and (SiMB-P0 + M1) + 100 nM UT + 4-ABT/AuNPs (c, red). Inset: SERS spectrum of the 4-ABT/AuNPs. (B) SERS intensity enhancements of the 1430 cm−1-band of 4-ABT/AuNPs, I/I0, plotted against the concentration of UT in PBS without (a) and with HCR (b). The concentrations of SiMB-P0, M1, and M2 were 500 nM, 1.0 μM, and 1.0 μM, respectively. The amount of 4-ABT/AuNPs was in 100-fold excess of M1 or M2. All error bars were obtained through the detection of six parallel samples.
M2 (complementary to the loop of M1) (sequences were shown in Table S1; see Supporting Information). The UT strand in solution can pair with the sticky end of M1, which undergoes an unbiased strand-displacement interaction to open the hairpin. The newly exposed sticky end of M1 nucleates at the sticky end of M2 and opens the hairpin to expose a sticky end on M2. This sticky end is identical in sequence to the initiator strands. In this way, each UT strand propagates a HCR event between alternating M1 and M2 to form a nicked doublehelix. In this case, the sulfhydryl group on one building block is brought into close proximity to other such moieties. Thus, numerous sulfhydryl functional groups are self-assembled together, providing potential binding sites for the Raman dyeencoded AuNPs via specific Au−S bond. As proof-of-principle, HCR between UT and M1 and M2 was first examined by ethidium bromide (EB)-stained agarose gel electrophoresis in solution (Figure S1; see Supporting Information). In the absence of UT, HCR between M1 and M2 was inhibited, and only a bright band of M1 or M2 could be observed, as shown in lane 2. However, emission bands of high-molecular weight structures could also be observed in lanes of 3−6, indicating the successful growth of the long polymer in the presence of UT. Under these circumstances, a low concentration of UT was able to trigger a proportionately small chain reaction of alternating
kinetic escapes by M1 and M2, resulting in the formation of a long DNA polymer. To fabricate SERS-active hot-spot substrates, AuNPs were prepared and characterized according to a reported method32 and then functionalized with Raman reporter 4-ABT.33 To ensure that AuNPs could still be conjugated to the thiolatedDNA long polymer, the AuNPs were not fully covered with 4ABT. In comparison with the SERS-nanoparticle tags reported previously,34 the surface coverage of 4-ABT molecules on AuNPs was evaluated to be 0.8 ± 0.12 pmol/cm2, thus reducing reporter dye surface coverage by over 10-fold.35 Transmission electron microscopy (TEM) shows a size distribution of the 4ABT/AuNPs from 20 to 30 nm, most being 30 nm (Figure S2; see Supporting Information). The UV−vis absorption spectrum shows a maximal absorption central at 520 nm (Figure S3; see Supporting Information), which can be attributed to the surface plasmon resonance of AuNPs.36 To form SERS-active hot-spots along the DNA polymer, the 4-ABT/AuNPs were conjugated to the DNA polymer using a Au−S bond. After centrifugation at 5000 rpm for 10 min, a process which has since been optimized to attain the best separation effect (data not shown), representative TEM images clearly demonstrate that the AuNPs were self-assembled along with the heterochains of the DNA polymer (Figure S4; see Supporting Information). These 2207
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 4. Performances of SERS detection of Tmb. (A) SERS spectra of 4-ABT/AuNPs in PBS(a), the SERS-active hot-spot substrate induced by 50 nM Tmb addition in PBS (b), and the spectrum induced by 50 nM Tmb addition in 50% human serum (c). (B) Dependence of SERS intensity enhancement of the 1430 cm−1 band, I/I0, on different concentrations of Tmb (a), BSA (c), and IgG (d) in PBS and Tmb in a 50% human serum sample (b). The measuring conditions as shown in Figure 3. All error bars were obtained through the detection of six parallel samples.
1305, and 1430 cm−1 can be assigned to b2-modes of the 4ABT molecule,38 the respective assignments of observed SERS bands were shown in Table S2 (see Supporting Information). The SERS enhancement, I/I0, of the most prominent Raman peak at 1430 cm−1 was estimated to be 187.2-fold by 100 nM UT trigging the HCR, where I0 and I are the SERS intensities at 1430 cm−1 in the absence and the presence of UT, respectively. According to methods in the literature,39 an enhancement factor of ∼4.14 × 106 was determined (for details, see Supporting Information). Meanwhile, to demonstrate SERS signal amplification by HCR, a control experiment was designed by directly employing SiMB-P0 and M1. A short DNA hybrid was formed among P0, UT, and M1 upon addition of UT, and only a few 4-ABT/AuNPs were loaded, resulting in a weak SERS signal state (curve c). In this case, the value of I/I0 dropped to 3.23 upon the addition of 100 nM UT, demonstrating the effective signal enhancement of our proposed platform. To assess the potential of this SERS detector for quantitative bioanalysis by HCR, the most prominent Raman peak at 1430 cm−1 of 4-ABT was compared to different concentrations of UT (Figure S6, see Supporting Information). Figure 3B reveals that the SERS signal dramatically increased upon increasing UT concentration from 0.01 to 150 nM. The value of I/I0 linearly increased with the UT concentration between 0.01 and 100 nM. This is mainly because, as the concentration of UT changed (from 0 to about 100 nM), the copies of M1 and M2 assembled together increased.Thus, numerous sulfhydryl functional groups are selfassembled together in our design, providing potential binding sites for the Raman dye-encoded AuNPs via specific Au−S bond, increasing SERS signal is observed. However, when the UT increased to about 100 nM, or even higher, one should also note that the SERS signal remained fairly constant due to the HCR system finally reaching the equilibrium state in terms of kinetics and thermodynamics. SERS Detection of Biomolecules. To demonstrate the feasability of our proposed approach as a universal SERS detector for ultrasensitive bioanalysis, human α-thrombin (Tmb) was chosen as a model by using anti-Tmb aptamer as the capture probe for Tmb. Tmb (activated Factor II) is a specific serine protease involved in the coagulation cascade, which converts soluble fibrinogen into insoluble strands of fibrin and catalyzes many other coagulation-related reactions.40 It is well established that aptamers “adaptively bind” target
collective results demonstrated that the UT had actually assisted in the self-assembly of nicked-DNA polymer structure and, subsequently, the 4-ABT/AuNPs conjugates. Finally, to fabricate the HCR-based SERS-active substrate on a large solid surface to attain preferable separating effect and eliminate the false positive signal produced by free 4-ABT/ AuNPs and the conjugation of 4-ABT/AuNPs and free M1/ M2. A capture probe, P0, for UT was employed (sequences were shown in Table S1; see Supporting Information) and immobilized on the surface of SiMBs through the streptavidin− biotin interaction. The surface density of streptavidin on SiMBs using biotin-labeled P0 was estimated to be 1.89 × 102/μm2. In the presence of UT, P0 hybridizes with one terminus of UT and served as a link to effectively initiate HCR that formed a long DNA polymer between M1 and M2 on the surface of SiMBs in the 0.1 M sodium phosphate buffer (PBS, 5 mM Mg2+, pH 7.4). 4-ABT/AuNPs were then added; after incubation at room temperature for 1.0 h, the mixtures were washed three times with PBS using centrifugation at 1000 rpm for 10 min to remove free 4-ABT/AuNPs, and the conjugates of 4-ABT/ AuNPs and M1 or M2. Scanning electron microscope (SEM) images showed that the loaded AuNPs were very closely aggregated on the surface of SiMBs, and a few bright spots appeared in the image (Figure 2A). However, the control experiment, in which UT was absent, showed no observable aggregates on the surface of SiMB (Figure 2A, inset). We further used atomic force microscopy (AFM) images to characterize the surface features of the formed AuNPs aggregates on SiMBs, which exhibit a mean height between 120 and 200 nm (Figure 2B,C), being largely higher than that of the free 4-ABT/AuNPs (Figure S5; see Supporting Information). Therefore, these special Raman reporter-encoded aggregates have the potential to provide a significant amplification of Raman signal intensity by several orders of magnitude through electromagnetic field enhancement. Figure 3A showed a set of SERS spectra of free 4-ABT/ AuNPs and UT-triggered formation of SERS-active hot-spot substrate on the surface of SiMBs. Either the free 4-ABT/ AuNPs or the mixture of SiMB-P0 + M1 + M2 + 4-ABT/ AuNPs shows very weak SERS signals by the weak interparticle plasmon coupling.37 However, upon addition of UT to the mixture, a stronger SERS signal was demonstrated by the aggregated 4-ABT/AuNPs on the surface of SiMB through HCR (curve b).The observed Raman bands centered at 1137, 2208
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 5. Performances of SERS detection of adenosine. (A) SERS intensity ratio of the 1430 cm−1-band, I/I0, plotted against the concentration of adenosine. The inset shows a linear relationship between I/I0 and the adenosine concentration. (B) Selectivity of this SERS amplification detector with 1.0 μM adenosine, guanosine, cytidine, and uridine. The measuring conditions as shown in Figure 3. All error bars were obtained through the detection of six parallel samples.
obvious SERS signals, even with a higher target concentration (Figure 4Bc,d). This SERS amplification detector even showed good sensitivity and selectivity in pure buffer; even to be more useful in the bioassay, the detection system should be able to tolerate any interference from real biological samples. To demonstrate such tolerance, pretreated blood serum was used to detect Tmb in a complex sample. Figure 4Ac was the signal enhancement of our SERS amplification detector upon 50 nM Tmb addition in 50% human serum, suggesting that the other components in blood serum provided little or no interference with the performance of our detector. To give a quantitative response to Tmb in a real biological sample, human serum sample spiked with different concentrations of Tmb was tested with our method. The SERS detector displayed a sensitive SERS response to Tmb in a 50% undiluted human blood sample (Figure 4Bb), which was close to that obtained in PBS buffer; however, even in serum contents >50%, this detector could also function well (data not shown). The result obtained in a human serum sample thereby demonstrates the wide applicability of this method in complex body fluids. To demonstrate the generality of this SERS detector for other small biomolecules, we also used an adenosine aptamer sequence flanked by two arm segments (P2−1−P2−4, Table S1; see Supporting Information) to detect adenosine, an important biological cofactor involved in many biological processes, such as those in the kidney and urine.41 Similar to the Tmb aptamerbased sensor described above, the extended adenosine aptamer (Ade-Apt) was used as the capture probe to hybridize with UT to form the triple-helix Ade-Apt/UT structure. The lengths of the two arms of Ade-Apt were optimized according to our previous report to achieve the best SERS response sensitivity, and P2−2 was chosen as a preferable capture probe for the subsequent experiment (Figure S11, Figure S12; see Supporting Information).16 As expected, in the absence of adenosine, UT could not be released from the triple-helix P2−2/UT structure, thus preventing HCR. However, upon the addition of adenosine to the solution of P2−2 and UT, the strong interaction between P2−2 and adenosine caused the disassembly of the triple-helix P2−2/UT structure. This allowed the released UT to initiate HCR, and an obvious SERS signal could be observed (Figure S13; see Supporting Information). Figure 5A showed the SERS signal was proportional to the concentrations of adenosine from 5.0 to 500 nM, establishing the quantitative detection capability of this SERS detector. A LOD of 1.5 nM
molecules, and a conformational change usually accompanies such binding events. Thus, they have become a key recognition element to detect various target analytes. To realize Tmb detection, the molecular recognition element was designed as a triple-helix aptamer/UT structure consisting of the UT and a central, Tmb-specific aptamer sequence flanked by two arm segments (P1−1−P1−4, Table S1; see Supporting Information). First, P1−2′ (sequence was shown in Supporting Information), which was dual-labeled with tetramethylrhodamine(TAMRA) and BHQ2 at each end, was employed to investigate the hybridization efficiency of the triple-helix, data shown in Figure S7 (see Supporting Information). To achieve the best response sensitivity, the DNA sequence and length of the arm segments of P1−2 were adapted from our recently reported work,16 which contained 7 bases in the stem as the optimal capture probe in subsequent experiments (Figures S8 and S9; see Supporting Information). We measured the SERS spectra of SERS-active hot-spot substrate on the SiMBs surface triggered by UT released from the triple-helix aptamer/UT structure. Figure 4Aa showed that, in the absence of Tmb, no SERS spectrum was observed, indicating that the triple-helix aptamer/UT structure remained stable, and no HCR occurred among the three DNA building blocks. However, upon addition of 50.0 nM Tmb to the solution, a strong SERS signal was observed (Figure 4Ab). To demonstrate the applicability of this SERS detector for the quantitative detection of Tmb, SERS singal enhancement was attained in the presence of different concentrations of Tmb in PBS (Figure S10; see Supporting Information). As shown in Figure 4Ba, depending on the Tmb concentration range, there were different sensitivities. The small but perceptible SERS change at low concentration of Tmb indicated that Tmb molecules were capable of interacting with the triple-helix aptamer/UT structure to release UT, even at very low concentration. In the concentration range of 0.05−100 nM Tmb, a dramatic SERS enhancement was observed. At >100 nM concentrations of Tmb, no further increase in SERS signal was observed and a plateau was reached with a maximum I/I0 value of a nearly 170. The limit of detection (LOD), based on 3σb/slope,16 where σb was the standard deviation of blank samples, was 18 pM. In addition, since the in vitro-selected antiTmb aptamer possesses high affinity for Tmb, we found that other proteins, such as IgG and BSA, which were commonly present in a complex environment, could not induce any 2209
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
Figure 6. Performance of SERS detection of cancer cells using the SERS detector. (A) SERS spectra corresponding to the addition of different concentrations (number of cells/mL) of CEM cells. The arrow indicates the signal changes as CEM concentrations increase (0, 1 × 102, 5 × 102, 1 × 103, and 1 × 104 cells/mL). (B) Plots demonstrating changes of Raman intensity at 1430 cm−1 upon the addition of different concentrations (number of cells/mL) of Ramos cells (a) and the target CEM cells (b). The measuring conditions as shown in Figure 3. All error bars were obtained through the detection of six parallel samples.
BHQ2 at the 5′-end, respectively. From Figure S15, Supporting Information, it can be seen that the fluorescence intensity of P3−2 was obviously decreased by the formation of the triplehelix aptamer/UT structure, while fluorescence recovery could be achieved upon target cell binding. Thus, we confirmed that the constructed triple-helix aptamer/UT structure can function well upon target cells addition and cause UT release. We next optimized the arm length of the different capture probes upon the addition of UT and subsequent target CEM, finally attaining amplified SERS signal readout through HCR. Figure S16, Supporting Information, showed that P3−2 would be a preferable capture probe sequence to attain higher I/I0. On the basis of this result, P3−2 was chosen as an optimal capture probe in subsequent experiments. For the sensitive detection of target cancer cells, different concentrations of CEM from one stock solution were attained, and we measured the SERS spectra of SiMB-TPS in PBS buffer. As shown in Figure 6A, the SERS intensity of 1430 cm−1 was highly sensitive to the concentration of CEM cells. When CEM cells were absent, only a weak peak was observed, indicating that the triple-helix aptamer/UT structure remained stable and that HCR could not be triggered without UT. This was mainly due to the two long hairpins M1 and M2, both have a stem of 18 base pairs enclosing a hexanucleotide loop. The UT can pair with the sticky end of M1, which undergoes an unbiased strand-displacement interaction to open the hairpin. The newly exposed sticky end of M2 nucleates at the sticky end of M2 and opens the hairpin to expose a sticky end on M2. This sticky end is identical in sequence to the initiator strands. In this way, each UT propagates a chain reaction of hybridization events between alternating M1 and M2 hairpins to form a nicked double-helix. On the contrary, in the absence of UT, M1 and M2 both keep stable hairpin structure in the solution and HCR could not be triggered. Therefore, in our experiment, when the target cell is absent, we could not attain the released UT from the triplehelix structure. Higher concentrations of CEM cells, however, led to a correspondingly stronger SERS signal. Figure 6B clearly demonstrated that the sensitivity of our platform was as low as 10 cells. To evaluate whether our assay is highly selective, we also studied how SERS intensity changed upon the addition of negative control Ramos cells. SERS intensity was negligible in the presence of 104 Ramos cells in the solution containing TPS. Since Ramos cells do not overexpress PTK7, only a weak interaction will be observed between TPS and Ramos cells.
adenosine was observed. Meanwhile, control experiments with adenosine analogues, such as guanosine, cytidine, and uridine, did not produce any enhanced SERS readings, indicating good selectivity for this SERS amplification detector (Figure 5B). The success of the Tmb and adenosine detections unequivocally supported the feasibility and versatility of our strategy. Moreover, based on SERS signal amplification via HCR, our SERS detector displayed ultrahigh sensitivity, exceeding that of a variety of previously reported aptamer-based sensors for Tmb and adenosine.42,43 SERS Detection of Human Cancer Cells. To further assess its biomedical and diagnostic applications, our SERS detector was investigated for cancer cell identification. It has been hypothesized that very sensitive monitoring of cancer cells could provide an easier and more effective way to monitor progression of the disease.44 From this perspective, identification and detection of cancer cells is fundamental to early diagnosis and therapy, as well as monitoring the relevant biological processes of cancers. For performance of human cancer cells detection, aptamer sgc8c,45 which binds the cell membrane protein tyrosine kinase-7 (PTK7) with high affinity and selectivity, was used in our study with different types of cell lines, including target cancer cell CCRF-CEM (acute lymphoblastic leukemia T-cells) and negative control Ramos cells (acute lymphoblastic leukemia B-cells).45 For optimization of the triple-helix aptamer/UT structure to achieve best detection sensitivity, we designed a series of capture probes (P3−1−P3−4, Table S1; see Supporting Information) comprising an aptamer sequence sgc8c and two arm segments of d(TC)n complementary to the sequence of UT. A poly-T chain links the 5′-end of capture probe to the two arm segments of d(TC) n. The binding of extended aptamer with CEM cells was studied by flow cytometry (Figure S14; see Supporting Information). After incubation with 5′-end-labeled carboxyfluorescein (FAM) of P3−1−P3−4, Figure S14, Supporting Information, showed that the larger fluorescence intensity changes of CEM cells corresponded to the stronger binding of capture probe to cells. This result showed the specific recognition of capture probe was preserved after the extension of aptamer sgc8. We then constructed an effective triplex switch consisting of UT and P3−1−P3−4. The formation of triplex was further confirmed by monitoring the fluorescence intensity of the capture probe P3−2 (P3−1−P3−4, Table S1; see Supporting Information) which was labeled with TAMRA at the 3′-end and 2210
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
three times with PBS using centrifugation at 1000 rpm to remove free 4-ABT/AuNPs and the conjugation of 4-ABT/ AuNPs and M1/M2. Characterization. For an extended aptamer binding assay, CCRF-CEM and negative control Ramos cells were washed with washing buffer and resuspended in binding buffer. Cells and extended aptamer, both labeled with FITC, were incubated on ice for 30 min. After washing with 1.0 mL of washing buffer, the cells were tested by a FACScan cytometer (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA) by counting 20 000 events. Raman spectra were obtained using a confocal microprobe Raman instrument (RamLab-010, Horiba Jobin Yvon, France), and the 632.8 nm He−Ne laser was used to excite the spectra, which were collected in a backscattering geometry. A 50× long working length objective (8 mm) was used in this work. The collection time for each spectrum was 20 s. The width of the slit and the size of the pinhole were set as 100 and 100 μm, respectively.
This lack of strong interaction prevented UT from being released to trigger HCR. Consequently, Raman enhancement was not observed, again demonstrating that our strategy is highly specific for CEM cells.
■
CONCLUSION In summary, an ultrasensitive SERS approach for detection of various target analytes with one Raman-active substrate has been developed for the first time. The design is based on the separation of molecular recognition element and SERS signal transduction moiety using a universal trigger to initiate cascaded Raman-active hot spots via HCR. The present approach can be engineered in ways that offer unique advantages and capabilities that are not available from conventional Raman detection systems. First of all, it is generalizable: when different analytes are present, we just need to change different label-free aptamer sequences in the triplehelix structure. For the signal transduction element, we have fabricated one hot active substrate using one UT, biotin-labeled P0, sulfhydryl-labeled M1/M2, which would solve the problem of resynthesis of various detection probes or even hot spotactive substrate. Second, by introducing HCR to amplify the SERS signal, ultrahigh sensitivity can be achieved, and thus, pM level of target biomolecules or single-cell level human cancer cells could be detected. Finally, and more importantly, although we used this approach to detect three target analytes, the strategy is versatile by the ability to select not only different types of aptamers for the capture probe, but also different types of signal transducers, such as fluorescence,46 colorimetry,47 or electrochemical,48 for detection assay, as well as drug delivery and multiple-stimulated release.49,50 Given the low cost, ultrahigh sensitivity, and excellent generality, we anticipate that this approach might open up new opportunities for the development of new bioanalytical and biomedical applications.
■
ASSOCIATED CONTENT
* Supporting Information S
More experimental details and additional spectroscopic data as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Fax: +86-731-8882 2523. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We are grateful for the financial support from the National Natural Science Foundation of China (21075032, 21005026, 21135001, 21305036, and J1103312), The Foundation for Innovative Research Groups of NSFC (21221003), and the “973”National Key Basic Research Program (2011CB91100-0). Zheng. J received financial support from Growth Program for Young Teachers in Hunan University.
■
MATERIALS AND METHODS Materials and Instruments. All oligonucleotides were synthesized by TaKaRa Biotechnology Co. Ltd. (Dalian, China). All chemicals were used as received, unless otherwise stated. Streptavidin-coated silicon beads (SiMBs, 5 μm) were purchased from Ocean Nanotech and dispersed at 0.1 mg/mL in 100 mM phosphate-buffered saline (PBS), pH 7.4. Cell lines CCRF-CEM (acute lymphoblastic leukemia T-cells) and Ramos (acute lymphoblastic leukemia B-cells) were obtained from the American Type Culture Collection (Manassas, VA). Construction of SERS-Active Substrate on the Surface of SiMBs. The biotin-labeled P0 was first incubated with streptavidin-coated SiMBs. The mixtures were vortexed at room temperature for 1 h, followed by washing three times with PBS using centrifugation at 1000 rpm to remove any DNA that did not conjugate to the SiMBs. The conjugates were dispersed in PBS and stored at 4 °C at a concentration of 0.1 mg/mL. Then, UT was added to the solution containing P0conjugated nanoparticles, and the mixture was vortexed at room temperature for another 12 h, followed by washing 3 times with PBS using centrifugation at 1000 rpm to remove triple-helix aptamer/UT and aptamer/target complex. Next, SH-labeled M1/M2 was added and reacted at room temperature for HCR 12 h, followed by washing three times with PBS using centrifugation at 1000 rpm to remove free SH-labeled M1/M2. Finally, 4-ABT/AuNPs were added to the above attained SiMBs and incubated for 2 h, followed by washing
■
REFERENCES
(1) Qian, X. M.; Nie, S. M. Chem. Soc. Rev. 2008, 37, 912−920. (2) Wang, Y. Q.; Yan, B.; Chen, L. X. Chem. Rev. 2013, 113, 1391− 1428. (3) Fabris, L.; Dante, M.; Braun, G.; Lee, S. J.; Reich, N. O.; Moskovits, M.; Nguyen, T. Q.; Bazan, G. C. J. Am. Chem. Soc. 2007, 129, 6086−6087. (4) Thompson, D. G.; Enright, A.; Faulds, K.; Smith, W. E.; Graham, D. Anal. Chem. 2008, 80, 2805−2810. (5) Grubisha, D. S.; Lipert, R. J.; Park, H. Y.; Driskell, J.; Porter, M. D. Anal. Chem. 2003, 75, 5936−5943. (6) Bonham, A. J.; Braun, G.; Pavel, L.; Moskovits, M.; Reich, N. O. J. Am. Chem. Soc. 2007, 129, 14572−14573. (7) Narayanan, R.; Lipert, R. J.; Porter, M. D. Anal. Chem. 2008, 80, 2265−2271. (8) Lee, A.; Kim, S.; Choo, J.; Shin, S. Y.; Lee, Y. H.; Choi, H. Y.; Ha, S.; Kang, K.; Oh, C. H. Anal. Chem. 2007, 79, 916−922. (9) Tang, H. W.; Yang, X. B.; Kirkham, J.; Smith, D. A. Anal. Chem. 2007, 79, 3646−3653. (10) Hu, Q.; Tay, L. L.; Noestheden, M.; Pezacki, J. P. J. Am. Chem. Soc. 2007, 129, 14−15. (11) Banholzer, M. J.; Millstone, J. E.; Qin, L. D.; Mirkin, C. A. Chem. Soc. Rev. 2008, 37, 885−897. 2211
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
Analytical Chemistry
Article
(12) Cao, Y. W.; Jin, R. C.; Mirkin, C. A. Science 2002, 30, 1536− 1540. (13) Cao, Y. W.; Jin, R. C.; Nam, J. M.; Thaxton, C. M.; Mirkin, C. A. J. Am. Chem. Soc. 2003, 125, 14676−14677. (14) Su, X.; Zhang, J.; Sun, L.; Koo, T. W.; Chan, S.; Sundararajan, N.; Yamakawa, M.; Berlin, A. A. Nano Lett. 2005, 5, 49−54. (15) Wang, Z. Y.; Zong, S. F.; Li, W.; Wang, C. L.; Xu, S. H.; Chen, H.; Cui, Y. P. J. Am. Chem .Soc. 2012, 134, 2993−3000. (16) Zheng, J.; Li, J. S.; Jiang, Y.; Jin, J. R.; Wang, K. M.; Yang, R. H.; Tan, W. H. Anal. Chem. 2011, 83, 6586−6592. (17) Zheng, J.; Jiao, A. L.; Yang, R. H.; Li, H. M.; Li, J. S.; Shi, M. L.; Ma, C.; Tan, W. H. J. Am. Chem. Soc. 2012, 134, 19957−19960. (18) Ellington, A. D.; Szostak, J. W. Nature 1990, 346, 818−822. (19) Tuerk, C.; Gold, L. Science 1990, 249, 505−510. (20) Dirks, R. M.; Pierce, N. A. Proc. Nat. Acad. Sci. U.S.A. 2004, 101, 15275−15278. (21) Yin, P.; Choi, H. M.; Calvert, C. R.; Pierce, N. A. Nature 2008, 451, 318−322. (22) Salunkhe, M.; Wu, T. F.; Letsinger, R. L. J. Am. Chem. Soc. 1992, 114, 8768−8772. (23) Nutiu, R.; Li, Y. F. J. Am. Chem. Soc. 2003, 125, 4771−4778. (24) Li, D.; Shlyahovsky, B.; Elbaz, J.; Willner, I. J. Am. Chem. Soc. 2007, 129, 5804−5805. (25) Li, N.; Ho, C. M. J. Am. Chem. Soc. 2008, 130, 2380−2381. (26) Shegai, T.; Vaskevich, A.; Rubinstein, I.; Haran, G. J. Am. Chem. Soc. 2009, 131, 14390−14398. (27) Li, J. F.; Huang, Y. F.; Ding, Y.; Yang, Z. L.; Li, S. B.; Zhou, X. S.; Fan, F. R.; Zhang, W.; Zhou, Z. Y.; Wu, D. Y.; Ren, B.; Wang, Z. L.; Tian, Z. Q. Nature 2010, 464, 392−395. (28) Liu, J. W.; Cao, Z. H.; Lu, Y. Chem. Rev. 2009, 109, 1948−1998. (29) Giepmans, B. N. G.; Adams, S. R.; Ellisman, M. H.; Tsien, R. Y. Science 2006, 312, 217−224. (30) Xiang, Y.; Lu, Y. Nat. Chem. 2011, 3, 697−703. (31) Das, J.; Cederquist, K. B.; Zaragoza, A. A.; Lee, P. E.; Sargent, E. H.; Kelley, S. O. Nat. Chem. 2012, 4, 642−648. (32) Liu, J. W.; Lu, Y. Nat. Protoc. 2006, 1, 246−252. (33) Kim, K.; Lee, H. B.; Yoon, J. K.; Shin, D.; Shin, K. S. J. Phys. Chem. C 2010, 114, 13589−13585. (34) Qian, X. M.; Peng, X. H.; Ansari, D. O.; Chen, G. Z.; Shin, D. M.; Yang, L.; Young, A. N.; Wang, M. D.; Nie, S. M. Nat. Biotechnol. 2008, 26, 83−90. (35) Qian, X. M.; Zhou, X.; Nie, S. M. J. Am. Chem. Soc. 2008, 130, 14934−14935. (36) Jana, N. R.; Gearheart, L.; Murphy, C. J. Langmuir 2001, 17, 6782−6786. (37) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat. Mater. 2008, 7, 442−453. (38) Dendisova, M.; Havranek, L.; Oncak, M.; Matejka, P. J. Phys. Chem. C 2013, 117, 21245−21253. (39) Braun, G.; Lee, S. J.; Dante, M.; Nguyen, T. Q.; Moskovits, M.; Reich, N. J. Am. Chem. Soc. 2007, 129, 6378−6379. (40) Holland, C. A.; Henry, A. T.; Whinna, H. C.; Church, F. C. FEBS Lett. 2000, 484, 87−91. (41) Vallon, V.; Muhlbauer, B.; Osswald, H. Physiol. Rev. 2006, 86, 901−940. (42) Cho, H.; Baker, B. R.; Hogiu, S. W.; Lee, L. P.; Tok, J. B. Nano Lett. 2008, 8, 4386−4390. (43) Li, L. L.; Ge, P. H.; Selvin, P. R.; Lu, Y. Anal. Chem. 2012, 84, 7852−7856. (44) Lu, W.; Singh, A. K.; Khan, S. A.; Senapati, D.; Yu, H.; Ray, P. C. J. Am. Chem. Soc. 2010, 132, 18103−18114. (45) Shangguan, D. H.; Li, Y.; Tang, Z. W.; Cao, Z. H.; Chen, H. W.; Tan, W. H. Proc. Nat. Acad. Sci. U.S.A. 2006, 103, 11838−11843. (46) Huang, J.; Wu, Y. R.; Chen, Y.; Yang, C. Y.; Wang, K. M.; Tan, W. H. Angew. Chem., Int. Ed. 2011, 50, 401−404. (47) Shimron, S.; Wang, F.; Orbach, R.; Willner, I. Anal. Chem. 2012, 84, 1042−1048. (48) Zhang, B.; Liu, B. Q.; Tang, D. P.; Niessner, R.; Chen, G. N.; Knopp, D. Anal. Chem. 2012, 84, 5392−5399.
(49) Zhu, G. Z.; Zheng, J.; Song, E. Q.; Donovana, M.; Liu, C.; Tan, W. H. Proc. Nat. Acad. Sci. U.S.A. 2013, 110, 7998−8003. (50) Zheng, J.; Zhu, G. Z.; Li, Y. H.; Yang, R. H.; Tan, W. H. ACS Nano 2013, 7, 6545−6554.
2212
dx.doi.org/10.1021/ac404004m | Anal. Chem. 2014, 86, 2205−2212
