Anal. Chem. 2006, 78, 3738-3744
Silver Nanoparticle-Based Ultrasensitive Chemiluminescent Detection of DNA Hybridization and Single-Nucleotide Polymorphisms Cheng-Hui Liu, Zheng-Ping Li,* Bao-An Du, Xin-Rui Duan, and Yu-Cong Wang
College of Chemistry and Environment Science, Hebei University, Baoding 071002, China
A new nanoparticle-based chemiluminescent (CL) method has been developed for the ultrasensitive detection of DNA hybridization. The assay relies on a sandwich-type DNA hybridization in which the DNA targets are first hybridized to the captured oligonucleotide probes immobilized on polystyrene microwells and then the silver nanoparticles modified with alkylthiol-capped oligonucleotides are used as probes to monitor the presence of the specific target DNA. After being anchored on the hybrids, silver nanoparticles are dissolved to Ag+ in HNO3 solution and sensitively determined by a coupling CL reaction system (Ag+-Mn2+-K2S2O8-H3PO4-luminol). The combination of the remarkable sensitivity of the CL method with the large number of Ag+ released from each hybrid allows the detection of specific sequence DNA targets at levels as low as 5 fM. The sensitivity increases 6 orders of magnitude greater than that of the gold nanoparticle-based colorimetric method and is comparable to that of surfaceenhanced Raman spectroscopy, which is one of the most sensitive detection approaches available to the nanoparticle-based detection for DNA hybridization. Moreover, the perfectly complementary DNA targets and the single-base mismatched DNA strands can be evidently differentiated through controlling the temperature, which indicates that the proposed CL assay offers great promise for singlenucleotide polymorphism analysis. Detection of specific DNA sequences is extremely important in clinical diagnosis, gene therapy, and a variety of biomedical studies. Pathogens responsible for disease states, bacteria and viruses, are also detectable via their unique nucleic acid sequences. So there has been increasing demand to develop simple and sensitive methods for the detection of specific sequence DNA. Thus, DNA hybridization biosensors have become a very attractive topic in the past several years, and they hold enormous promise for the clinical diagnosis of inherited diseases and the rapid detection of infectious microorganisms. Various techniques have been developed for detection of DNA hybridization, and their sensitivities depend mainly on the specific activity of the labels linked to the oligonucleotide probes. The * To whom correspondence should be addressed. E-mail: mail.hbu.edu.cn. Tel: +86 312 5079357. Fax: +86 312 5079525.
3738 Analytical Chemistry, Vol. 78, No. 11, June 1, 2006
lzpbd@
labels that provide a radioactivity1 or fluorescence2,3 are the most popular. Still other detection techniques employ indirect methods that rely on enzymes to generate colorimetric, fluorescent, or chemiluminescent signals.4 The radioactive labels have their inherent safety problems, e.g., health hazard and waste disposal problems. The colorimetric methods were restricted by their poor sensitivity. The fluorescence labels have been most widely used in standard applications of DNA analysis. But the quantitative measurements remain challenging because of low fluorescence intensities and susceptibility to photobleaching. They are also hampered by the need of sophisticated fluorescence microscopy/ scanners as well as strongly environment-depending quantum yields. In recent years, a new way for the detection of DNA hybridization is the employment of metal nanoparticle labels. Metal nanoparticles offer excellent prospects for biological sensing because of their unique optical and electrical properties; thus, great attention is focused on this field in order to overcome the problems associated with the radioisotopic, fluorescent, and enzyme labels. Mirkin and co-workers first developed an entirely new colorimetric detection scheme for DNA hybridization based on aggregation of oligonucleotide-functionalized gold nanoparticles directed by target DNA.5 This technique offered several advantages over conventional fluorophore-based assays with regard to quick and easy readout and no requirement for expensive instrumentation. Moreover, it exhibited a high degree of discrimination between perfectly matched target oligonucleotides and targets with single base-pair mismatch. However, the main disadvantage of this approach is its low sensitivity, which is in the 1-10 nM range and is not as good as the results from fluorophore-based assay (typically in picomolar range, best reported ∼600 fM).6 To improve the sensitivity, many detection techniques, such as surface plasmon resonance (SPR),7 laser (1) Woff, S. F.; Haines, L.; Fisch, J.; Kremsky, J. N.; Dougherty, J. P.; Jacobs, K. Nucleic Acids Res. 1987, 15, 2911-2916. (2) Weiss, S. Science 1999, 283, 1676-1683. (3) Fang. X. H.; Tan, W. Anal. Chem. 1999, 71, 3101-3105. (4) Mansfield, E. S.; Worley, J. M.; Mckenzie, S. E.; Surrey, S.; Rappaport, E.; Fortina, P. Mol. Cell. Probes 1995, 9, 145-156. (5) Elghnian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078-1081. (6) Rosi, N. L.; Mirkin, C. A. Chem. Rev. 2005, 105, 1547-1562. (7) He, L.; Musick, M. D.; Nicewarner, S. R.; Salinas, F. G.; Benkovic, S. J.; Natan, M. J.; Keating, C. D. J. Am. Chem. Soc. 2000, 122, 9071-9077. 10.1021/ac0522409 CCC: $33.50
© 2006 American Chemical Society Published on Web 04/21/2006
Table 1. DNA Sequences Used in the Proposed Assay
a
oligonucleotides
sequencesa
probe 1 probe 2 perfectly complementary target single-base mismatched strand three-base mismatched strand noncomplementary strand
5′-biotin-TTG TGC CTG TCC TGG-3′ 5′-GAG AGA CCG GCG CAC-T10-(CH2)3 -SH-3′ 5′-GTG CGC CGG TCT CTC CCA GGA CAG GCA CAA-3′ 5′-GTG CGC CAG TCT CTC CCA GGA CAG GCA CAA-3′ 5′-GTG GGC CAG TCT GTC CCA GGA CAG GCA CAA-3′ 5′-AGT TGT AAG GGA AGA TGC AAT AGT AAT CAG-3′
The mismatched bases in DNA strands are underlined.
diffraction,8 surface-enhanced Raman spectroscopy (SERS),9,10 and array-based electrical detection,11 have been explored to detect gold nanoparticle probes for DNA hybridization. These approaches push the detection limit of nanoparticle-based methods to picomolar, and even femtomolar concentrations of target DNA. However, these detection methods generally need a sophisticated process of silver amplification and expensive instruments. Another strategy of metal nanoparticle-based detection of DNA hybridization is to detect the metal ions released from the metal nanoparticle probes anchored on the DNA hybrids by oxidative metal dissolution with electrochemical detection techniques,12-14 which enable detection of DNA targets at the picomolar level by using relatively simple instrumentation. Chemiluminescence (CL) was one of the most sensitive techniques for trace analysis of metal ions with a very simple instrumentation,15 and a large number of metal ions can be released from one metal nanoparticle (e.g., ∼2.9 × 105 silver atoms are theoretically contained in a 21-nm spherical silver particle). Therefore, it should be believed that the CL technique possesses great potential for the highly sensitive detection of DNA hybridization using metal nanoparticles as labels of oligonucleotide probes. However, to the best of our knowledge, CL detection of metal nanoparticle-based DNA hybridization has not been reported. Recently, Lu’s group and our group have separately developed gold nanoparticle-based CL immunoassay.15,16 However, the gold nanoparticle is not considered to be an ideal tag for CL detection because the dissolution of a gold nanoparticle needs extremely severe conditions (high concentrated HNO3-HCl or poisonous HBr-Br2), which result in high CL background and further restrict the detection sensitivity. The sensitivity of gold nanoparticle-based CL immunoassay is only about equal to that of the standard enzyme-linked immunosorbent assay (∼pM detection limits).6 In this contribution, a new CL scheme for the detection of DNA hybridization based on a silver nanoparticle label has been developed. First, we found that silver nanoparticle probes can be easily dissolved in dilute HNO3 solution. Second, a coupling CL (8) Bailey, R. C.; Nam, J.-M.; Mirkin, C. A.; Hupp, J. T. J. Am. Chem. Soc. 2003, 125, 13541-13547. (9) Cao, Y. W. C.; Jin, R.; Mirkin, C. A. Science 2002, 297, 1536-1540. (10) Ginger, D. S.; Cao, Y. C.; Mirkin, C. A. Biophotonics Int. 2003, 10, 48-51. (11) Park, S.-J.; Taton, T. A.; Mirkin, C. A. Science 2002, 295, 1503-1506. (12) Wang, J.; Xu, D.; Kawde, A.-N.; Polsky, R. Anal. Chem. 2001, 73, 55765581. (13) Cai, H.; Xu, Y.; Zhu, N.; He, P.; Fang, Y. Analyst 2002, 127, 803-808. (14) Authier, L.; Grossiord, C.; Brossier, P. Anal. Chem. 2001, 73, 4450-4456. (15) Fan, A. P.; Lau, C.; Lu, J. Z. Anal. Chem. 2005, 77, 3238-3242. (16) Li, Z. P.; Wang, Y. C.; Liu, C. H.; Li, Y. K. Anal. Chim. Acta 2005, 551, 85-91.
reaction system (Ag+-Mn2+-K2S2O8-H3PO4-luminol)17 was employed to sensitively measure Ag+ released from dissolution of silver nanoparticle probes. The detection limit of Ag+ with the coupling CL reaction is 2.8 × 10-11 M, which is ∼10-fold lower than that of Au3+ with the luminol-H2O2 CL reaction (∼2 × 10-10 M).15 For the proposed CL assay of DNA hybridization, the detection limit of target DNA (3σ) is estimated to be 5 fM. The sensitivity increases 6 orders of magnitude greater than that of the gold nanoparticle-based colorimetric method 5 and is comparable to that of the SERS method,9,10 which is one of the most sensitive detection approaches available to the nanoparticle-based detection of DNA hybridization. Additionally, the CL assay offers the obvious advantages of being low cost, simple, and rapid compared with other techniques. Moreover, the perfectly complementary DNA targets and the single-base mismatched DNA strands can be clearly distinguished by controlling the temperature using the proposed CL assay. Therefore, the CL method may open a new way for the highly sensitive detection of DNA hybridization and the single-nucleotide polymorphism (SNP) analysis. EXPERIMENTAL SECTION Apparatus. The CL measurements were performed with a BPCL ultraweak luminescence analyzer (Institute of Biophysics Academic Sinica, Beijing, China). A KYC 100 C rocking incubator with thermostatic controller (Shanghai Fuma Laboratory Instrument Co., Ltd.) was used to control the temperature of the hybridization reaction. A thermostatic water bath (Liaota) that can control the temperature at 0.1 °C intervals was used to carry out the SNP analysis. A pHS-3C digital pH meter (Shanghai Weiye Instrument Plant, Shanghai, China) was used to measure the pH values of the solutions. The transmission electron microscope (TEM) image of the silver nanoparticles was acquired on a JEM1200EX II TEM (Jeol). A TU 1901 UV-visible spectrophotometer (Purkinje General Instrument Co., Ltd., Beijing, China) was used to determine the surface coverage of oligonucleotides on the silver nanoparticles. Materials and Reagents. Polystyrene 96-well microtiter plates (Folcon) were used to perform the hybridization reactions. Streptavidin and bovine serum albumin (BSA) were purchased from Sigma Co. Ltd. All the synthetic oligonucleotides (the sequences are listed in Table 1) used in this study were obtained from TaKaRa Biotechnology Co. Ltd. (Dalian, China). The combined sequences of probe 1 and probe 2 are complementary to that of the perfectly complementary target, and all of them are (17) Geng, Z.; Wu, J. C.; Chen, Z. G.; Liu, Q. G. Chin. J. Anal. Chem. 1995, 23, 401-403.
Analytical Chemistry, Vol. 78, No. 11, June 1, 2006
3739
Figure 1. TEM image (100000×) of the silver nanoparticles.
fragments of human p53 gene (exon8). Colloidal silver nanoparticles and colloidal silver-labeled oligonucleotides were synthesized in our laboratory. The following buffers were used: 0.1 M PBS (0.1 M NaCl + 10 mM sodium phosphate buffer, pH 7.0); 0.3 M PBS (0.3 M NaCl + 10 mM sodium phosphate buffer, pH 7.0); 0.3 M PBN (0.3 M NaNO3 + 10 mM sodium phosphate buffer, pH 7.0); 0.05 M carbonate/bicarbonate buffer (pH 9.6). All other reagents were of analytical reagent grade and used as purchased without further purification. Doubly distilled and deionized water was used throughout. Preparation of Silver Nanoparticles. Silver nanoparticles were prepared by potassium borohydride reduction of AgNO3 according to the literature18 with some modifications. Briefly, icecold AgNO3 (1 × 10-3 M) and an equal volume of KBH4 (3.0 × 10-3 M) were mixed dropwise with stirring in an ice bath, and the mixture was continuously stirred until the temperature naturally increased to the room temperature. The solution was subsequently filtered through a polycarbonate membrane (0.2 µm). The final silver nanoparticles prepared by this method have an average diameter of 21 nm as measured by TEM (Figure 1). Preparation of Silver Nanoparticle-Labeled Oligonucleotides. Silver nanoparticle-oligonucleotide conjugates were synthesized by derivatizing 1 mL of silver nanoparticle solution with 3′-alkanethiol-capped oligonucleotides (probe 2, final concentration, 1.08 µM). Meanwhile, the same amount of probe 2 was added in 1 mL of water, which was treated in the same way as the silver nanoparticle solution in order to determine the total amount of probe 2 through measuring the absorbance at 260 nm. After standing for 24 h at room temperature, the solution was adjusted to the pH value and ionic strength of the 10 mM sodium phosphate buffer, pH 7.0, and allowed to stand for another 6 h. Then, aqueous 2 M NaCl was added to the solution to bring the total NaCl concentration to 0.075 M. This procedure was repeated after 2 h to adjust the NaCl concentration to 0.1 M. After an additional 84 h, the nanoparticles were isolated by centrifugation for 30 min at 12 000 rpm. The absorbance of the supernatant was measured at 260 nm to obtain the amount of the nonbound probe 2. The difference between the total amount of probe 2 and the amount of the nonbound probe 2 was used to estimate the amount of the binding probe 2 on the silver nanoparticles. The average coverage of a silver nanoparticle was calculated (∼73 oligonucle(18) Creighton, J. A.; Blatchford, C. G.; Albrecht, M. G. J. Chem. Soc., Faraday Trans. 2 1979, 75, 790-798.
3740
Analytical Chemistry, Vol. 78, No. 11, June 1, 2006
otide units/nanoparticle). The resulting DNA nanoparticle precipitate was washed with 0.1 M PBS, recentrifuged, and redispersed in a fresh 0.3 M PBS. A 10-fold dilution of the solution in 0.3 M PBS was used as the work solution to perform the hybridization in this study. Hybridization Procedures. Briefly, the polystyrene 96-well microtiter plate was coated with 100 µL/well of streptavidin (5 mg L-1 in 0.05 M carbonate/bicarbonate buffer, pH 9.6) overnight at 4 °C and then blocked by 3% BSA at 37 °C for 1 h. After washing three times with phosphate buffer containing 0.05% Tween 20 (PBS-T, pH 7.0), the biotinylated probe 1 (100 µL/well, 96 nM) was transferred to combine with the immobilized streptavidin at room temperature for 1 h with gentle shaking and was used as the captured probe. Unbound probe 1 was removed by washing three times with PBS-T. The 100 µL/well series dilutions of target DNA in 0.3 M PBS were then transferred and incubated for 2 h at 37 °C with shaking. Afterward, the wells were washed three times with PBS-T and treated with 100 µL/well silver nanoparticlelabeled oligonucleotides (probe 2) in 0.3 M PBS solution containing 0.1% BSA for another 2 h at 37 °C with shaking to perform the sandwich-type hybridization with the target DNA. Finally, the wells were thoroughly washed three times with PBS-T and two times with 0.3 M PBN to remove residual Cl-. Coating with BSA for both the polystyrene surface and the probe 2-labeled silver nanoparticles was needed to avoid nonspecific interaction. SNP Analysis. First, the sandwich-type hybridization was performed with the same amount (140 fM) of perfectly complementary targets in a series of wells and one-base mismatched DNA strands in another series of wells according to the hybridization procedures mentioned above. After the nonhybridized probe 2 labeled with silver nanoparticle tags was washed away, 100 µL/ well 0.3 M PBS was added in. Second, one well with perfectly complementary targets and another one with one-base mismatched DNA strands were simultaneously brought to the thermostatic water bath at a certain temperature for at least 5 min and then washed thoroughly with 0.3 M PBN at the same temperature. Finally, the CL detection was performed according to the standard procedures below. Standard Procedures for CL Detection. The silver nanoparticles, which had been anchored on the hybrids in the polystyrene wells, were dissolved by 100 µL/well 1:3 (v/v) HNO3 for 30 min to ensure that the silver dissolution was completed. Afterward, the pH value of the resultant Ag+ solutions was adjusted by adding 60 µL of 5 M NaOH, and the resultant mixture was then transferred into a 1-mL centrifuge tube containing 200 µL of 2% (m/v) K2S2O8, 40 µL of 6 × 10-3 M MnSO4, and 36 µL of 1:1 (v/v) H3PO4 and immediately incubated in a 90 °C water bath for 7 min; thus, the resultant KMnO4 was produced. The reaction was stopped with flowing cold water, and the solution was adjusted to the pH 10.0 using 5 M NaOH. A 50-µL aliquot of the solution was transferred to a 40 × 14 mm quartz tube, then 200 µL of luminol (10-3 M, pH 13.5) was injected, and the CL signal was measured with the BPCL luminescence analyzer. RESULTS AND DISCUSSION The principle of silver nanoparticle-based CL detection of DNA hybridization is depicted in Figure 2. The polystyrene 96-well microtiter plate was first coated with streptavidin. The biotinylated DNA probes (probe 1) were arrayed on the wells through the
Figure 2. Schematic representation of the CL detection of DNA hybridization based on a silver nanoparticle label.
Figure 3. Effect of the dissolving time of silver nanoparticles in 1:3 HNO3. Experimental conditions: Silver nanoparticles (final concentration, 5 × 10-8 g mL-1) were dissolved in 100 µL of 1:3 HNO3 solution for different times and then 60 µL5 M NaOH was added in to adjust the acidity. The solution was immediately transferred into the mixed solution of 200 µL of 2% (m/v) K2S2O8, 40 µL of 6 × 10-3 M MnSO4, and 36 µL of 1:1 (v/v) H3PO4 and incubated in the 90 °C water bath for 7 min. The reaction was stopped with flowing cold water, and the pH value of the resultant KMnO4 was adjusted to 10.0 with 5 M NaOH. A 50-µL sample of the solution and 200 µL of luminol (10-3 M, pH 13.5) were added into the BPCL analyzer for CL measurement.
interaction between streptavidin and biotin. Then the target DNA, the second DNA probe (probe 2) labeled with silver nanoparticle tags, which were used to monitor the presence of the specific target strands, were respectively added to perform a sandwichtype hybridization. Finally, the nonhybridized probe 2 labeled with silver nanoparticle tags was discarded. The silver nanoparticles anchored on the hybrids were dissolved in HNO3 solution, and the released Ag+ was sensitively determined by the coupling CL reaction system (Ag+-Mn2+-K2S2O8-H3PO4-luminol). This coupling CL reaction was based on the strong catalytic effect of Ag+ on the reaction of Mn2+-K2S2O8-H3PO4. The catalytically produced KMnO4 can oxidize luminol in an alkaline medium, which would generate a strong CL signal, and the CL intensities are proportional to the amount of target DNA. Detailed optimizations of the CL assay, and characterization of the performance of the resulting protocol, are presented below. Dissolution of Silver Nanoparticles. It was found that silver nanoparticles could be easily dissolved in 1:3 HNO3 (v/v) solutions. Therefore, 100 µL/well 1:3 HNO3 (v/v) was used to dissolve the silver nanoparticles anchored on the hybrids in this
Figure 4. Signal/background ratio versus the volume of 5 M NaOH added to the completely released Ag+ in HNO3 solution. Experimental conditions: silver nanoparticles (final concentration, 5 × 10-9 g mL-1) were dissolved in 100 µL of 1:3 HNO3 solution for 30 min, and then different volumes of 5 M NaOH were added to adjust the acidity. The blanks were treated in the same way as the samples without silver nanoparticles. Other conditions were the same as in Figure 3.
study. It can be seen from Figure 3 that the silver nanoparticles can be completely dissolved within 30 min. The acidity of the released Ag+ solution can also affect its catalytic activity on the K2S2O8-Mn2+-H3PO4 reaction. After the silver was dissolved completely, a different volume of 5 M NaOH was used to test the optimal acidity. As shown in Figure 4, when 60 µL of 5 M NaOH was added, the signal/background ratio of the CL intensity reached its highest value. Thus, 60 µL of 5 M NaOH was used to adjust the acidity of the oxidatively released Ag+ solution in the further experiments. Optimization of the CL Reaction Conditions. Since the quantitative determination of Ag+ was based on the Ag+-K2S2O8Mn2+-H3PO4-luminol coupling CL reaction system, several parameters were investigated systematically to establish the optimal conditions for the CL reaction. The optimal amount of the Mn2+, K2S2O8, and H3PO4 was first investigated. Considering both the CL intensity and the signal/ background ratio, 200 µL of 2% (m/v) K2S2O8, 40 µL of 6 × 10-3 M MnSO4, and 36 µL of 1:1 (v/v) H3PO4 was selected for this study (see Supporting Information). Since the CL originated from the reaction of KMnO4-luminol, the conditions for the catalytic resultant KMnO4 and luminol were also studied. It can be seen from Figure 5 that the optimal pH value of the resultant KMnO4 Analytical Chemistry, Vol. 78, No. 11, June 1, 2006
3741
Figure 5. Effect of pH value of the resultant KMnO4 solutions on the CL intensities. (Inset) signal/background ratio vs pH value of resultant KMnO4. (a) Samples (silver nanoparticles, final concentration 5 × 10-9 g mL-1); (b) blanks. The blanks were treated in the same way as the samples without silver nanoparticles. Other conditions were the same as in Figure 3.
Figure 6. Signal/background ratio versus the incubation time of the catalytic reaction of Ag+-K2S2O8-Mn2+-H3PO4 in the 90 °C water bath. Silver nanoparticles, final concentration of 2.5 × 10-8 g mL-1. The blanks were treated in the same way as the samples without silver nanoparticles. Other conditions were the same as in Figure 3.
was found to be 10.0 (adjusted by 5 M NaOH), where both the CL intensity and the signal/background ratio were the optimum. And 1 × 10-3 M luminol (pH 13.5) was found to be the optimum for the CL reaction (Supporting Information). The incubation time for the catalytic reaction of Ag+-K2S2O8Mn2+-H3PO4 in the 90 °C water bath was also studied. It was found that both CL signal and background were increased with increasing incubation time. However, the signal/background ratio reaches its highest value when the incubation time was 7 min (Figure 6). So, a 7-min incubation time was selected for the following experiments. Analytical Performance. Under the optimized experimental conditions mentioned above, the relationship between the CL intensity and the concentration of target strands was investigated. It was found that the CL intensity depended linearly on the concentration of target strands in the range of 14 fM-1.4 pM. The correlation equation was ICL ) 32.4 + 25.7Ctarget (fM) and the correlation coefficient R ) 0.9990. In the targets’ concentration range from 1.4 to 14 pM, the CL response increased more slowly and afterward started to level off. The detection limit (3σ, n ) 11) was estimated to be 5 fM (i.e., ∼0.5 amol in a 100-µL sample). 3742 Analytical Chemistry, Vol. 78, No. 11, June 1, 2006
A series of seven repetitive measurements of 14 fM target DNA were used for estimating the precision, and the relative standard deviation was 2.1%. The comparison between the proposed silver nanoparticle-based CL assay and other reported methods, especially the metal nanoparticle-based methods, are listed in Table 2. Most of the detected single-strand DNA (ssDNA) in these listed methods were all ∼30 bases long. So although the detection limits of ssDNA can vary when the target length and sequences were different, it also can be found that the sensitivity of the proposed CL assay for DNA hybridization was higher than that of the colorimetric method, SPR, and electrochemical method and was comparable to that of the SERS method. It can also be seen from Table 2 that some amplified techniques, such as liposome,26 magnetic particles,27 bio bar code,28 and Au nanoparticle-based Ag amplification,9-11,20,21 can greatly improve the sensitivity of the DNA assays. Among them, Au nanoparticle-based Ag amplification, in which silver ions are catalytically reduced to silver metal and precipitated on the surface of the gold nanoparticle labels to produce larger particles, is widely used to amplify the signals of gold nanoparticles. Mirkin and co-workers have demonstrated that the Au nanoparticle-based Ag amplification is initiated by gold nanoparticle catalysis and the deposited silver further catalyzes the silver ion reduction.11 Therefore, the sensitivity of the proposed Ag nanoparticle-based CL DNA assay may also be greatly improved by further Ag amplification. The specificity of the new silver nanoparticle-based CL hybridization assay was examined by detecting the CL response of perfectly complementary targets, one-base mismatched strands, three-base mismatched strands, and noncomplementary strands (the sequences were shown in Table 1) at the same concentration of 280 fM. The ratio of the CL intensities of the four sequences was 100:76.3:12.6:3.2, respectively. So the well-defined CL response was observed for the perfectly complementary targets and the CL response arising from the noncomplementary strands was negligible (less than 5.0%). The three-base mismatched strands yield a CL response that was significantly smaller (only 12.6%) than that of the perfectly complementary targets. The results indicate that the CL assay for DNA hybridization has a high specificity and the absence of nonspecific binding. It is worth noting that the single-base mismatched DNA strands can also be differentiated from the perfectly complementary targets with the proposed CL assay without controlling the hybridization temperature, reflecting the effective discrimination and great potential for SNP analysis. (19) Sato, K.; Hosokawa, K.; Maeda, M. J. Am. Chem. Soc. 2003, 125, 81028103. (20) Taton, T. A.; Mirkin, C. A.; Letsinger, R. L. Science 2000, 289, 1757-1760. (21) Storhoff, J. J.; Marla, S. S.; Bao, P.; Hagenow, S.; Mehta, H.; Lucas, A.; Garimella, V.; Patno, T.; Buckingham, W.; Cork, W.; Muller, U. R. Biosens. Bioelectron. 2004, 19, 875-883. (22) Wang, J.; Polsky, R.; Xu, D. Langmuir 2001, 17, 5739-5741. (23) Perez, J. M.; Josephson, L.; O′Loughlin, T.; Hogemann, D.; Weissleder, R. Nat. Biotechnol. 2002, 20, 816-820. (24) Wang. J.; Liu, G.; Mercoci, A. J. Am. Chem. Soc. 2003, 125, 3214-3215. (25) Gerion, D.; Chen, F. Q.; Kannan, B.; Fu, A. H.; Parak, W. J.; Chen, D. J.; Majumdar, A.; Alivisatos, A. P. Anal. Chem. 2003, 75, 4766-4772. (26) Patolsky, F.; Lichtenstein, A.; Willner, I. Angew. Chem., Int. Ed. 2000, 39, 940-943. (27) Patolsky, F.; Weizmann, Y.; Katz, E.; Willner, I. Angew. Chem., Int. Ed. 2003, 42, 2372-2376. (28) Nam, J. M.; Stoeva, S. I.; Mirkin, C. A. J. Am. Chem. Soc. 2004, 126, 59325933.
Table 2. Comparison between the Proposed CL Assay and Other Reported Techniques for the Detection of DNA Hybridizationa format nanoparticle and nanostructure-based methods for DNA hybridization
techniquesc
label
detection limit. of ssDNA
no. of steps
Au nanoparticles (cross-linked)5
colorimetric
∼10 nM
2
Au nanoparticles (non-cross-linked)19 Au nanoparticles8 Au nanoparticles7 Au nanoparticles12 Au nanoparticles with Ag amplification20, 21 Au nanoparticles with Ag amplification9,10 Au nanoparticles with Ag amplification11 Au nanoparticles with Ag amplification22 silver nanoparticles13 silver nanoparticles b iron oxide nanoparticles23 ZnS, CdS, PbS nanoparticles24 ZnS and CdSe quantum dots25 liposome26
colorimetric laser diffraction SPR PSA Scanometric Raman spectroscopy electrical PSA ASV CL magnetic relaxation stripping voltammetry fluorescence liposome-amplified electrochemical magnetically amplified electrogenerated CL bio-bar-code amplified scanometric
60 nM ∼50 fM 10 pM 15 nM 50 fM ∼20 fM 500 fM 32 pM 0.5 pM 5 fM 20 pM 270 pM 2 nM 50 fM
2 4 4 6 5 5 5 7 4 7 2 5 5 6
8.3 aM
5
500 aM
8
magnetic particles27 Au nanoparticles with Ag amplification
28
a Some were adapted from ref 6. b This method. c PSA, potentiometric stripping analysis; ASV, anodic stripping voltammetry; SPR, surface plasmon resonance.
SNP Analysis. SNPs are the most frequent form of DNA sequence variation in the human genome and, as such, are becoming increasingly popular genetic markers for genome mapping studies and medical diagnostics.29 The high sensitivity and effective discrimination of the silver nanoparticle-based CL hybridization assay enable us to differentiate the one-base mismatched DNA strands from the perfectly complementary targets at very low concentration by controlling the hybridization temperature, which has important significance for the early detection of genetic point mutation and early diagnostics of corresponding diseases. The thermal dissociation curves for the sandwich-type DNA duplexes with the perfectly complementary targets and one-base mismatched strands are shown in Figure 7. It can be seen that the silver nanoparticle-associated DNA duplexes exhibit sharp melting transitions, which were similar to the earlier reported gold nanoparticle-based methods.20,30 As shown in Figure 7, the “melting” temperature (Tm) of perfectly complementary targets and the single-base mismatched strands were 61.5 (Tm1) and 57.0 °C (Tm2), respectively. The CL response of the perfectly complementary targets can be evidently differentiated from the singlebase mismatched strands at an appropriate temperature between Tm1 and Tm2. At such a given temperature, the silver nanoparticlebased DNA duplexes with one mismatched base can be selectively dehybridized, which will cause the release of the silver nanoparticles from the hybrids, and thus, the CL intensity decreases. But the hybrids formed from the perfectly complementary targets will not be destroyed at that temperature, so its CL intensity will not change. Thus, the CL intensities of the targets and the single(29) Venter, J. C.; Adams, M. D.; et al. Science 2001, 291, 1304-1351. (30) Jin, R.; Wu, G.; Li, Z.; Mirkin, C. A.; Schatz, G. C. J. Am. Chem. Soc. 2003, 125, 1643-1654.
Figure 7. Comparison of the thermal dissociation curves for the silver nanoparticle-based sandwich-type hybrids with perfectly complementary targets (a) and one-base mismatched DNA strands (b) in 0.3 M PBS. The concentrations of the two DNA strands are all 140 fM. The intercepts at the vertical dashed line (58.5 °C) in the graph can be used to estimate the selectivity of sequence identification.
base mismatched strands were selectively distinguished. Experimental results showed that when the temperature was controlled at 58.5 °C, the CL intensities of the single-base mismatched strands (140 fM) were all below 5% of that of the same amount of perfectly complementary targets. The results indicate that the proposed CL assay shows great promise for the SNP analysis. CONCLUSIONS We have demonstrated for the first time the use of chemiluminescent analysis of metal nanoparticle label for monitoring DNA hybridization. This assay combines the large number of silver ions released from the silver nanoparticles and the inherent high sensitivity of CL method in metal analysis, which offers great potential for the ultrasensitive detection of DNA hybridization and the SNP analysis. Compared with pioneering works that also Analytical Chemistry, Vol. 78, No. 11, June 1, 2006
3743
utilized metal nanoparticle labels for the detection of DNA hybridization, the proposed assay can obtain superior sensitivity for detection of DNA hybridization by using simple CL instrument, inexpensive reagents, and rapid procedures. The new approach described in this work is promising for several reasons: (1) the successful sensitive and simple quantification of target oligonucleotides on the microwells establishes a general detection methodology that can be extended to a variety of DNA diagnostic formats including various immobilization supports. The proposed CL approach can also be extended to a large variety of bioaffinity assays of analytes of environmental or clinical significance, such as the immunoassay; (2) this CL method was particularly attractive for large-scale DNA testing in very small volumes with inexpensive materials and simple instrumentations; (3) metal nanoparticle labels had the advantage over radioisotopic, fluorescent, or enzyme labels of being stable, and the labeling procedure is very simple. The main drawback of the present CL method is the requirement of many steps (see Table 2) because the dissolution of silver nanoparticles and the coupling CL reaction are employed. Fortunately, CL method is a sensitive technique for analysis of various trace metal ions. Therefore, we have a wide
3744
Analytical Chemistry, Vol. 78, No. 11, June 1, 2006
choice of nanoparticle labels and CL reaction systems. We now envision finding more suitable nanoparticle tags and CL reactions to make the CL DNA assay more sensitive and simple. Moreover, the use of different metal nanoparticle tags may open the door to the simultaneous detection of multiple targets. ACKNOWLEDGMENT The project is supported by the National Natural Science Foundation of China (NSFC, 20375011), Program for New Century Excellent Talents in University (NCET), and the National Science Foundation of Hebei Province (203111). SUPPORTING INFORMATION AVAILABLE Figures for the optimization of the conditions of luminol and the amount of K2S2O8, MnSO4, and H3PO4. This material is available free of charge via the Internet at http://pubs.acs.org.
Received for review December 19, 2005. Accepted March 14, 2006. AC0522409
