One-Step Label-Free Optical Genosensing System for Sequence

Oct 21, 2008 - developed in this contribution by taking short DNA target .... genosensing system with advantages of real-time, label-free, simple, and...
0 downloads 0 Views 1MB Size
Download PDF
Anal. Chem. 2008, 80, 8424–8430

One-Step Label-Free Optical Genosensing System for Sequence-Specific DNA Related to the Human Immunodeficiency Virus Based on the Measurements of Light Scattering Signals of Gold Nanorods Wei He,† Cheng Zhi Huang,*,‡ Yuan Fang Li,‡ Jian Ping Xie,§ Rong Ge Yang,| Pei Fu Zhou,§ and Jian Wang† College of Chemistry and Chemical Engineering, College of Pharmaceutical Sciences, and Institute of Modern Biopharmaceuticals, College of Life Sciences, Education Ministry Key Laboratory on Luminescence and Real-Time Analysis, Southwest University, Chongqing 400715, China, and AIDS and HIV Research Group, State Key Laboratory of Virology, Wuhan Institute of Virology, Chinese Academy of Sciences, Wuhan 430071, China A one-step label-free optical genosensing method has been developed in this contribution by taking short DNA target with its sequence related to the human immunodeficiency virus type 1 (HIV-1) as an example. By employing anisotropic nonspherical and positively charged gold nanorods (Au-NRs) as the recognition platform, which show high stability against aggregation under high ionic strength conditions without any additional stable reagent, we found that the addition of target DNA to the mixture of nonmodified Au-NRs suspension and label-free probe DNA in high ionic strength buffer leads to a color change from red to light purple in less than 5 min, displaying strong plasmon resonance light scattering (PRLS) signals. Mechanism investigations showed that the strong PRLS signals should be ascribed to the aggregation of Au-NRs induced by the formed double-stranded oligonucleotides (dsDNA) from the hybridization of target DNA with probe DNA. With the PRLS signals, we monitored the hybridization process of a 21-mer single-stranded oligonucleotide (ssDNA) from the HIV-1 U5 long terminal repeat (LTR) sequence with its complementary oligonucleotide and detected the effect of single-base-pair mismatches. Two polymerase chain reaction (PCR) amplicon artificial samples derived from Mycobacterium tuberculosis glmS and genes encoding for Bacillus glucanase and an HIV-1 LTR sample isolated from HIV-1-positive blood were detected with satisfactory results, showing that the present method has simplicity, sensitivity, specificity, and reliability for sequence-specific DNA detection related to the HIV gene. Increasing interest has been focused on the development of simple, rapid, and highly selective methods for DNA sequence * To whom correspondence should be addressed. Phone: +86-23-68254659. Fax: +86-23-68866796. E-mail: [email protected]. † College of Chemistry and Chemical Engineering, Southwest University. ‡ College of Pharmaceutical Sciences, Southwest University. § College of Life Sciences, Southwest University. | Chinese Academy of Sciences.

8424

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

and mutant gene analysis in recent decades since that allows early and precise diagnoses of diseases such as worldwide-concerned acquired immure deficiency syndrome (AIDS) caused by the human immunodeficiency virus (HIV). Serologic tests as the standard diagnostic test for HIV infection, including the enzymelinked immunosorbent assay (ELISA) and Western blot (WB) assay for the detection of HIV antibodies, are very sensitive, but at the cost of relatively complex instrumentation and elaborate sample preparation1 and at the risk of bringing about false positive and false negative results in certain stages of infection.2,3 On the other hand, colorimetry,4,5 fluorescence resonance energy-transfer (FRET)-based assays,6 DNA electrochemical biosensor,7 surfaceenhanced Raman gene probe technique,3,8 and surface-enhanced Raman scattering molecular sentinel technology (SERS-MS)1 have been developed for highly sensitive and specific HIV-related DNA detection. Nevertheless, most of these methods still involve labeling protocols and require complex experimental procedures. Therefore, it is desirable to develop a much simpler, direct labelfree nucleic acid-based test method for the sequences related to HIV, which is also continuing to encourage researchers to explore other DNA genosensing systems for diagnostic applications with unique and complementary advantages. Colloidal gold is a type of promising nanomaterial playing important roles in the design of genosensing systems for simple and rapid DNA sequence detection.9,10 For example, Rothberg and co-workers made use of the different adsorbing properties of (1) Wabuyele, M. B.; Vo-Dinh, T. Anal. Chem. 2005, 77, 7810–7815. (2) Proffitt, M. Anal. Chem. 1993, 65, 396–400. (3) Liang, Y.; Gong, J.-L.; Huang, Y.; Zheng, Y.; Jiang, J.-H.; Shen, G.-L.; Yu, R.-Q. Talanta 2007, 72, 443–449. (4) Mulder, J.; McKinney, N.; Christopherson, C.; Sninsky, J.; Greenfield, L.; Kwok, S. J. Clin. Microbiol. 1994, 32, 292–300. (5) Livache, T.; Fouque, B.; Teoule, R. Anal. Biochem. 1994, 217, 248–254. (6) Thelwell, N.; Millington, S.; Solinas, A.; Booth, J.; Brown, T. Nucleic Acids Res. 2000, 28, 3752–3761. (7) Wang, J.; Cai, X.; Rivas, G.; Shiraishi, H.; Farias, P. A. M.; Dontha, N. Anal. Chem. 1996, 68, 2629–2634. (8) Isola, N. R.; Stokes, D. L.; Vo-Dinh, T. Anal. Chem. 1998, 70, 1352–1356. (9) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078–1081. (10) Ray, P. C. Angew. Chem. 2006, 118, 1169–1172. 10.1021/ac801005d CCC: $40.75  2008 American Chemical Society Published on Web 10/21/2008

single-stranded and double-stranded DNA (ssDNA and dsDNA) on negatively charged gold nanoparticles (Au-NPs) and proposed a label-free simple colorimetric sensing method for oligonucleotide hybridization.11-13 Their method, although not requiring covalent functionalization of the Au-NPs, the probe, or the target DNA, and having made great advances in respect of the simplicity of the hybridization assay, should have to experience multisteps and cannot monitor the hybridization process in real time since the hybridization is completely separated from the detection step. In addition, ordinary colorimetric detection could not be achieved for absolute quantitative analysis. In order to find an ideal optical genosensing system with advantages of real-time, label-free, simple, and fast quantification, we herein try to develop an optical genosensing method with gold nanorods (Au-NRs) by taking the detection of HIV-related DNA as an example. It has known that Au-NRs are characteristic as anisotropic configuration and unique optical properties,14-17 which can be widely exploited for chemical and biochemical sensing such as metal ions,18,19 amino acids,20 antibodies,21 and for cancer cell imaging and photothermal therapy.22 With regard to hybridization detection, Dujardin et al. obtained specific organization of short Au-NRs into anisotropic three-dimensional (3-D) aggregates by DNA hybridization.23 Subsequently, Li et al. took advantage of fluorescence properties of long Au-NRs for DNA hybridization studies.24 However, whichever method is used, the oligonucleotides must be decorated with thiol so that they could be employed for the functionalization of Au-NRs through the Au-S bond, which should have to undergo the expense of long time (about 73 h) and complex procedures.23,24 Cetyltrimethylammonium bromide (CTAB), as a commonly used cationic surfactant playing the role of a rod-inducing and stabilizing agent in the process of preparing Au-NRs, could be strongly adsorbed on the surface of Au-NRs by forming a bilayer,25 making Au-NRs net positively charged so that both ssDNA and dsDNA could interact with Au-NRs through electrostatic attraction.26 This case is totally different from that of citrate-coated negatively charged sphere Au-NPs.11-13 On the other hand, AuNRs have a notable merit of high stability in virtue of abundant (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26)

Li, H.; Rothberg, L. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 14036–14039. Li, H.; Rothberg, L. Anal. Chem. 2005, 77, 6229–6233. Li, H.; Rothberg, L. J. J. Am. Chem. Soc. 2004, 126, 10958–10961. Yu, Y.-Y.; Chang, S.-S.; Lee, C.-L.; Wang, C. R. C. J. Phys. Chem. B 1997, 101, 6661–6664. Brioude, A.; Jiang, X. C.; Pileni, M. P. J. Phys. Chem. B 2005, 109, 13138– 13142. Link, S.; Mohamed, M. B.; El-Sayed, M. A. J. Phys. Chem. B 1999, 103, 3073–3077. Murphy, C. J.; Sau, T. K.; Gole, A. M.; Orendorff, C. J.; Gao, J.; Gou, L.; Hunyadi, S. E.; Li, T. J. Phys. Chem. B 2005, 109, 13857–13870. Rex, M.; Hernandez, F. E.; Campiglia, A. D. Anal. Chem. 2006, 78, 445– 451. Nakashima, H.; Furukawa, K.; Kashimura, Y.; Torimitsu, K. Chem. Commun. 2007, 1080–1082. Sudeep, P. K.; Joseph, S. T. S.; Thomas, K. G. J. Am. Chem. Soc. 2005, 127, 6516–6517. Yu, C.; Irudayaraj, J. Anal. Chem. 2007, 79, 572–579. Huang, X.; El-Sayed, I. H.; Qian, W.; El-Sayed, M. A. J. Am. Chem. Soc. 2006, 128, 2115–2120. Dujardin, E.; Hsin, L.-B.; Wang, C. R. C.; Mann, S. Chem. Commun. 2001, 1264–1265. Li, C.-Z.; Male, K. B.; Hrapovic, S.; Luong, J. H. T. Chem. Commun. 2005, 3924–3926. Nikoobakht, B.; El-Sayed, M. A. Langmuir 2001, 17, 6368–6374. Pan, B.; Cui, D.; Ozkan, C.; Xu, P.; Huang, T.; Li, Q.; Chen, H.; Liu, F.; Gao, F.; He, R. J. Phys. Chem. C 2007, 111, 12572–12576.

CTAB, which is either adsorbed on the surface of Au-NRs or is free in the solution, and can withstand high salt concentration (about 0.5 M) without obvious salt-induced aggregation even if without any complicated and expensive decoration or disposal, providing promising applications in realistically complicated sample detection. This merit is impossibly possessed for citratecoated Au-NPs or Au-NRs purified by centrifugation.24 In this contribution, we take advantage of Au-NRs in CTAB medium as a genosensing platform for DNA hybridization toward simple, fast, label-free, and real-time detection by taking 21-mer single-stranded oligonucleotide from the HIV-1 U5 long terminal repeat (LTR) sequence7 as a model sample. The experimental procedure for detecting hybridization only experiences one step and is very simple. It only needs the addition of target DNA to the mixture of the Au-NRs without any modification and the labelfree probe DNA in certain buffer salt solutions. On the basis of the aggregation of Au-NRs and corresponding color change owing to the plasmon resonance absorption (PRA), which derives from electron oscillations in the metallic particles induced by the incident light field,27 a visual detection for hybridization of target DNA could reach as low as 1.67 nM. Corresponding to the PRA features, the aggregation of Au-NRs displays particular enhanced plasmon resonance light scattering (PRLS) features through the excitation of the surface plasmon transverse mode.28 Moreover, PRLS signals of metal nanoparticles have been successfully measured with a common spectrofluorometer.29-31 Thus, on the basis of the enhanced PRLS signals that originate from the aggregation of Au-NRs directed by the target DNA, we propose an assay of short DNA sequences related to HIV. Different from that of Au-NPs-based linear light scattering detection of DNA hybridization,32 wherein Au-NPs are functionalized with either 3′or 5′-terminal of the thiol-capped oligonucleotide probes, and a consecutive process involving heating and cooling the solution was compulsory, our present PRLS signal based method is onestep and label-free with a detection limit of 80 pM HIV-1 U5 LTR segment. Moreover, single-base-pair mismatches are easily detected without temperature control, providing promising applications in the detection of single-nucleotide polymorphisms (SNPs) and disease diagnosis. EXPERIMENTAL SECTION Apparatus. The PRLS and PRA measurements were made with an F-4500 fluorescence spectrophotometer and a U-3010 spectrophotometer (both were from Hitachi, Tokyo, Japan), respectively. A TecNai-10 electron microscope (FEA) was utilized to measure the transmission electron microscopy (TEM) images of Au-NRs. A QL-901 vortex mixer (Haimen, China) was used to blend the solution. A constant-temperature water-base boiler was employed to control the temperature when preparing Au-NRs. Polymerase chain reaction (PCR) amplification reactions were carried out in a DNA-Engine thermal cycler (Bio-Rad). (27) Roll, D.; Malicka, J.; Gryczynski, I.; Gryczynski, Z.; Lakowicz, J. R. Anal. Chem. 2003, 75, 3440–3445. (28) Nappa, J.; Revillod, G.; Abid, J.-P.; Russier-Antoine, I.; Jonin, C.; Benichou, E.; Giraultb, H. H.; Brevet, P. F. Faraday Discuss. 2004, 125, 145–156. (29) Pasternack, R. F.; Collings, P. J. Science 1995, 269, 935–939. (30) Zhu, J.; Huang, L.; Zhao, J.; Wang, Y.; Zhao, Y.; Hao, L.; Lub, Y. Mater. Sci. Eng., B 2005, 121, 199–203. (31) Wu, L. P.; Li, Y. F.; Huang, C. Z.; Zhang, Q. Anal. Chem. 2006, 78, 5570– 5577. (32) Du, B.-A.; Li, Z.-P.; Liu, C.-H. Angew. Chem., Int. Ed. 2006, 45, 1–5.

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

8425

Table 1. DNA Sequences Used in This Contributiona

a

oligonucleotides

sequences

P, probe T, perfectly complementary target MT1, one-base-mismatched MT2, three-base-mismatched NT, noncomplimentary oligomer

5′-ATG TGG AAA ATC TCT AGC AGT-3′ 5′-ACT GCT AGA GAT TTT CCA CAT-3′ 5′-ACT GCT AGA TAT TTT CCA CAT-3′ 5′-ACT TCT AGA TAT TTT TCA CAT-3′ 5′-ATG TGG AAA ATC TCT AGC AGT-3′

The mismatched base in DNA strands is underlined.

Reagents. Tetrachloroauric acid trihydrate (HAuCl4 · 3H2O) and CTAB were commercially available from Shanghai Chemical Reagent Co. (Shanghai, China), whereas L-ascorbic acid was from Chuandong Chemical Co., Ltd. (Chongqing, China), sodium borohydride (NaBH4) was from Huanwei Fine Chemical Co., Ltd. (Tianjin, China), and silver nitrite (AgNO3) was from Chongqing Chemical Reagent Co., Ltd., (Chongqing, China). All chemicals were used as such without further purification. Milli-Q purified water (18.2 MΩ) was used for all the experiments. A 21-mer synthetic oligonucleotide, corresponding to portions of the HIV-1 U5 LTR DNA segment, was purchased from Sunbiotech Co., Ltd. (Beijing, China). The sequences include ssDNA probe (P), its complementary target oligomer (T), its onebase-mismatched oligomer (MT1), its three-base-mismatched oligomer (MT2), and its noncomplimentary oligomer (NT, all the sequences are listed in Table 1). PCR sequences were derived from genes coding for Bacillus glucanase and Mycobacterium tuberculosis glmS. Preparation of Gold Nanorods. The Au-NRs were synthesized in aqueous solution with reference to the seed-mediated method as reported by Murphy and others.33,34 Briefly, gold seeds were prepared at first by reducing HAuCl4 · 3H2O (2.89 × 10-4 M) with freshly prepared ice-cold NaBH4 (6.0 × 10-4 M) in the presence of CTAB (9.0 × 10-2 M). With vigorous shaking for 20 s, the mixture solution displays as light brown. After being left undisturbed for 2 h at the temperature of 25 °C, the prepared gold seeds with the size less than 5 nm could be used for the synthesis of Au-NRs. Into a clean test tube was placed 25 mL of growth solution containing 4.0 × 10-4 M HAuCl4 and 0.095 M CTAB. To this solution were added 150 µL of 0.01 M AgNO3 and 160 µL of 0.1 M freshly prepared ascorbic acid, and then the mixture was shaken vigorously for about 2 min. During the shaking, the color of the mixture changed from orange to colorless. Finally, 108 µL of above-prepared gold-seed suspension was added in. After being shaken vigorously for 20 s, the color of the mixture began to change and even became reddish brown within 10 min. If further left undisturbed for 24 h at 30 °C, we could obtain Au-NRs with average aspect ratio about 3 (10 nm in diameter and 30 nm in length) as inferred from TEM images. The concentration of AuNRs synthesized was estimated by the measured absorbance and corresponding molar extinction coefficient (3.9 ± 0.5 × 109 M-1 cm-1) at wavelength maxima of the longitudinal absorption band.35 Different from previous reports,23,24 it is not necessary for us to remove the free CTAB in the solution by centrifugation and (33) Sau, T. K.; Murphy, C. J. Langmuir 2004, 20, 6414–6420. (34) Jiang, X. C.; Brioude, A.; Pileni, M. P. Colloids Surf., A 2006, 277, 201– 206. (35) Orendorff, C. J.; Murphy, C. J. J. Phys. Chem. B 2006, 110, 3990–3994.

8426

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

Scheme 1. Schematic Representation of the DNA Hybridization Detection with Au-NRs Genosensing Systema

a

The different colors of Au-NRs represent the dispersed and aggregated states, respectively.

redispersion in order to purify the Au-NRs for hybridization detection. General Procedures. The protocol of our method is displayed in Scheme 1, and the hybridization reaction was carried out at room temperature. Typically, 120 µL of as-prepared Au-NRs (1.92 nM) suspension was added to 60 µL of Tris-HCl (pH 7.4, 50 mM) buffer solution, followed by the addition of 60 µL of probe DNA (1.0 × 10-7 M) and 60 µL of NaCl (1.2 M). After being entirely mixed, suitable target DNA with complementary sequence was added to the mixture. The solution is 360 µL in summation and is vortexed thoroughly. About 2 min, a clear colorimetric change from red into light purple could be visually observed. The PRLS spectra were obtained by scanning simultaneously the excitation and emission monochromators of the F-4500 spectrofluorometer (Hitachi, Tokyo, Japan) from 300 to 700 nm (namely, ∆λ ) 0 nm), and all PRLS measurements were made with a 5.0 nm slit width of the excitation and the emission of the spectrofluorometer. The PRLS intensity was measured at the maximum PRLS peak of 555 nm. Meanwhile, the PRA spectra were recorded by keeping a 5 mm path length with a U-3010 UV-vis spectrophotometer (Hitachi, Tokyo, Japan). Besides that, the samples for TEM were prepared by dropping 5 µL of the solution onto a copper grid and the TEM images were then acquired with A TecNai-10 electron microscope (FEA). Polymerase Chain Reaction. The PCR reaction was routinely performed as follows (with minor changes such as temperature and cycles, if required): 95 °C soak for 5 min, followed by 30 cycles of 95 °C denature for 1 min, 58 °C anneal for 1 min, 72 °C extension for 3 min; after 30 cycles, 72 °C soak for 10 min. The PCR product was confirmed by sequencing (Tiangen Company).

Figure 1. Plasmon resonance light scattering (PRLS) spectra of AuNRs in the presence of probe ssDNA (curve 2, red, 16.67nM), probe ssDNA/complementary target complex (curves 3-7, green, blue, cyan, magenta, and yellow, corresponding to the target DNA concentration of 3.33, 5.00, 6.67, 8.33, and 11.67 nM) as well as in salt buffer solution (curve 1, black). The DNA hybridization is performed in Tris-HCl buffer solution (8.3 mM; pH 7.4) containing 0.2 M NaCl for 5 min at room temperature. The inset shows the plot of the intensity of PRLS signals measured at 555 nm vs the concentration ratio of target DNA to probe DNA.

Preparation of HIV-1 LTR Real Samples. The HIV-1 LTR real sample was isolated from HIV-1-positive blood. PCR reaction and restriction endonuclease reaction were made in order to make the final real sample of DNA sequence for detection (details for sample preparation are listed in the Supporting Information). By this method, we obtained an HIV PCR product of HB-hp3-LTR1.8, which consists of 1840 bp, and the base at the site from 615 to 635 is perfectly matched with the probe ssDNA. With restriction endonuclease reaction, we got a 314 bp sequence fragment which contains the target ssDNA at the base site from the 615 to 635 and adjacent sequence. Then, these sequence fragments were denatured by boiling to obtain a single-stranded molecule. RESULTS AND DISCUSSION Spectral Features of the Hybridization Reactions. The light scattering (LS) signals of Au-NRs are weak in buffer salt solution (Figure 1, curve 1), as is the case in the presence of ssDNA probe (Figure 1, curve 2), with a characteristic LS peak at 555 nm corresponding to the transversal band of the PRA band of AuNRs located around 520 nm (Figure 2). Thus, the LS signals of Au-NRs, which act as plasmon scatterers, detected with the common spectrofluorometer should indeed be ascribed to the PRLS.36 When target DNA is added, however, Au-NRs get aggregated, resulting in strong PRLS signals (Figure 1, curves 3-7) that are linearly enhanced and increased with increasing target DNA concentration. When the concentration ratio of target DNA to probe DNA is about 0.8:1, the intensity of PRLS at 555 nm reaches maximum (Figure 1, inset). It is known that the PRA spectra of Au-NRs displays two bands assigned to the transversal and longitudinal modes of electronic oscillations, and the locations depends on the aspect ratio.37 As is often the case, the term “absorption” is used for convenience instead of the term “extinction” for preciseness, because the extinction spectra and visible colors of Au-NRs are due to both (36) Aslan, K.; Holley, P.; Davies, L.; Lakowicz, J. R.; Geddes, C. D. J. Am. Chem. Soc. 2005, 127, 12115–12121. (37) Liz-Marza´n, L. M. Langmuir 2006, 22, 32–41.

Figure 2. Plasmon resonance absorption (PRA) bands of the AuNRs in the presence of probe ssDNA (curve 2, red, 16.67nM), probe ssDNA/complementary target complex (curves 3-7, green, blue, cyan, magenta, and yellow, corresponding to the target DNA concentration of 3.33, 5.00, 6.67, 8.33, and 11.67 nM) as well as in salt buffer solution (curve 1, control). All the experimental conditions are the same as in Figure 1. The inset shows visual detection of hybridization based on the color change of Au-NRs. Medium: water (tube 1), buffer salt solution (tube 2), ssDNA probe (tube 3), and ssDNA probe/complementary target complex (tube 4). The concentrations of both probe DNA and target DNA were kept at 8.33 nM for tubes 3 and 4.

absorption and scattering.27,38 As Figure 2 shows, the two PRA bands for our prepared Au-NRs with the aspect ratio of 3 are located around 520 and 732 nm, respectively, correspondingly assigned to the transversal and longitudinal modes of electronic oscillations, and the characteristic band of PRLS signals at 555 nm in Figure 1 is correspondent to the transversal absorption band of Au-NRs located around 520 nm (Figure 2). When probe DNA is added to the suspension of Au-NRs, it is hard to observe obvious variations concerning the PRA features including wavelength shift or absorbance fluctuations, whereas if addition of target DNA is made to the mixture of probe DNA and Au-NRs suspension, the transverse PRA band displays a weak hyperchromic effect and gets slightly red-shifted, while the longitudinal PRA band shows obvious hypochromic and hypsochromic effects. The blue-shift of the longitudinal band and the red-shift of the transverse band indicate the aggregation of Au-NRs in a side-by-side orientation in solution.39 These variations of PRA features resulting from the hybridization between the probe and target DNA could be visually observed even if 1.67 nM target DNA is added, and the color of the mixture changes from red to purple (Figure 2, inset). The TEM images clearly show that Au-NRs in both the presence and absence of probe DNA in the corresponding Tris-HCl buffer solution containing 0.2 M NaCl are dispersed in the aqueous medium (Figure 3, parts A and B). Once target DNA is added into the suspension and the hybridization starts off, the aggregation of Au-NRs occurs (Figure 3C), and the main aggregation orientation indeed follows the mode of side-by-side (Figure 3C, inset), identifying the blue-shift of the longitudinal band and the red-shift of the transverse band.39 Dependence of the Hybridization Reactions on Ionic Strength. We found that CTAB-coated Au-NRs without any modification are stable even if in a medium of high ionic strength (38) Jain, P. K.; Lee, K. S.; El-Sayed, I. H.; El-Sayed, M. A. J. Phys. Chem. B 2006, 110, 7238–7248. (39) Jain, P. K.; Eustis, S.; El-Sayed, M. A. J. Phys. Chem. B 2006, 110, 18243– 18253.

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

8427

Figure 3. TEM images of Au-NRs (A) in the presence of probe ssDNA (B) and the mixture of probe and target DNA (C) in buffer salt solution. The concentrations of both probe DNA and target DNA are kept at 8.33 nM. The DNA hybridization is performed in Tris-HCl buffer solution (8.3 mM; pH 7.4) containing 0.2 M NaCl for 5 min at room temperature.

Figure 4. Effect of ionic strength on the PRLS signals of Au-NRs (black solid squares), Au-NRs with probe DNA (red solid rotundities), and the mixture of Au-NRs with probe DNA and target DNA (green solid triangles). Both probe DNA and target DNA were kept at 16.67 nM; Tris-HCl buffer solution (8.3 mM; pH 7.4).

up to 0.5 M NaCl, displaying weak PRLS signals (Figure 4, the curve with solid squares). This stability is greatly different from that of citrate-coated Au-NPs which have strong aggregation tendency in salt buffer solution.11 When ssDNA is added into the suspension of Au-NRs in a buffer with high ionic strength, there is no obvious change of the PRLS signals (Figure 4, the curve with solid rotundities). On the other hand, if target DNA is added soon after the addition of probe DNA into the suspension of AuNRs, the PRLS signals get enhanced. Therefore, the ionic strength should be an important factor for the hybridization reaction. As the curve with solid triangles in Figure 4 shows, the PRLS signals get increased with increasing ionic strength if the ionic strength is lower than 0.175 mM, and then get up to the maximum if the ionic strength is controlled in the range of 0.175-0.225 mM. After that, the enhanced PRLS signals get decreased with further increase of ionic strength, indicating that the salt effect can reduce the repulsive electrostatic interactions between probe DNA and target DNA and improve the efficiency of DNA hybridization, whereas too high ionic strength may restrain the attractive electrostatic interaction between dsDNA and Au-NRs.40,41 Real-Time Detection of DNA Hybridization. We monitored the kinetic process of the hybridization reaction with 21-mer singlestranded oligonucleotide related to HIV-1 to perfectly comple(40) Leal, C.; Moniri, E.; Pegado, L.; Wennerstrom, H. J. Phys. Chem. B 2007, 111, 5999–6005. (41) Ganachaud, F.; Elaıssari, A.; Pichot, C. Langmuir 1997, 13, 7021–7029.

8428

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

Figure 5. Real-time detection of the hybridization of probe DNA with target DNA: λ, 555 nm; Tris-HCl buffer solution (8.3 mM; pH 7.4); NaCl, 0.2 M. Probe DNA was kept at 16.67 nM.

mentary sequence by investigating the influence of the incubation time on PRLS intensity in a period of 10 min immediately after mixing the different amounts of target DNA with the solution containing probe DNA, Au-NRs, Tris-HCl buffer solution, and NaCl. As shown in Figure 5, the PRLS intensity at 555 nm for the mixture of Au-NRs with probe DNA alone keeps the same all the time. On the contrary, with the addition of different concentrations of target DNA, distinct changes of the PRLS intensity at 555 nm are immediately visible. It can be seen that the PRLS intensity at 555 nm gets increased rapidly and reaches a plateau in less than 5 min, indicating the hybridization has finished. Quantitive Analysis of Hybridization and Detection of Base Mismatches. It was found that the enhanced PRLS signals linearly get increased with increasing target DNA concentration. As Figure 6 shows, a linear equation of ∆I ) 136.56 + 36.67c is followed in the range of 0.17-11.67 nM with the correlation coefficients of 0.996, and the limit of determination is around 80 pM (3σ, n ) 11). In order to identify whether the system is useful or not in the point-of-care tests, where DNA genosensing system should discriminate oligonucleotides with one or more mismatches, we evaluated the specificity of the proposed assay by detecting the PRLS signals of Au-NRs arising from perfectly complementary target and base-mismatched DNA strands, by keeping ssDNA concentration at 8.33 nM. It could be seen that Au-NRs in the presence of probe/perfectly complementary target DNA show the strongest PRLS signals (column 3 in Figure 7), followed by the probe/one-base-mismatched DNA (column 4 in Figure 7) and

Figure 6. Linear graph of the PRLS signals on the concentration of target DNA: probe DNA, 16.67 nM; λ, 555 nm; Tris-HCl buffer solution (8.3 mM; pH 7.4); NaCl, 0.2 M.

Figure 7. PRLS signal of Au-NRs suspension in the presence of ssDNA probe (black column), ssDNA probe/noncomplementary oligomer complex (red column), ssDNA probe/complementary target complex (green column), ssDNA probe/one-base-mismatched target complex (blue column), ssDNA probe/three-base-mismatched target complex (cyan column). All the concentrations of ssDNA are kept at 8.33 nM. λ, 555 nm. Tris-HCl buffer solution (8.3 mM; pH 7.4); NaCl, 0.2 M; n ) 5.

probe/three-base-mismatched DNA (column 5 in Figure 7). Furthermore, it was found that the mixture of ssDNA probe and noncomplementary target could not induce obvious aggregation of Au-NRs and the PRLS signals of the solution keep the same as ssDNA probe alone (columns 1 and 2 in Figure 7). The high specificity and selectivity of this genosensing system indicates the potential application in detecting point mutations for the diagnosis of diseases. Detection of Target DNA in PCR Amplicon Artificial Samples and HIV-1 LTR Real Samples. We employed two PCR amplicons derived from Mycobacterium tuberculosis glmS and genes encoding for Bacillus glucanase as artificial samples, in which the target DNA were artificially added since the two PCR amplicons do not contain the complementary sequence of the probe DNA, and an HIV-1 LTR real sample, which was isolated from HIV-1-positive blood and pretreated as stated in the Supporting Information, to demonstrate the capability of our genosensing system for sequence-specific detections. These PCR amplicons were exposed to thermal denaturation to obtain single-stranded forms. Figure 8 shows the results of the detection for two PCR amplicons derived from Mycobacterium tuberculosis glmS and genes encoding for Bacillus glucanase. It seems that the addition of PCR amplicons cannot exert any effect on the PRLS signals of Au-NRs with probe DNA (columns 1, 3, 4 and columns 1, 6, 7 in Figure 8), showing that PCR amplicons could not induce obvious aggregation of Au-NRs. However, the addition of PCR amplicon artificial samples, in which the target of the DNA probe had been

Figure 8. Detection of target DNA in PCR amplicon artificial samples: P, probe DNA; T, target DNA; BG, Bacillus glucanase PCR products; MTG, Mycobacterium tuberculosis glmS PCR products. Conditions: probe DNA, 8.33 nM; target DNA, 8.33 nM; MTG, 2.1 × 10-7 M; BG, 2.1 × 10-7 M; n ) 5. The concentration of PCR products was calculated according to the absorbance at 260 nm with the molar absorptivity of 6600 M-1 cm-1.

Figure 9. Detection of target DNA in HIV-1 LTR samples: Au, AuNRs; P, probe DNA; T, target DNA; HB, HB-hp3-LTR1.8; HB′, sequence fragments of HB-hp3-LTR1.8 after restriction endonuclease reaction; enzymes, the mixture of HindIII and EcoRI. Conditions: probe DNA, 83.3 nM; target DNA, 83.3 nM; HB-hp3-LTR1.8, 4.1 × 10-6 M; HB-hp3-LTR1.8′, 4.1 × 10-6 M; Tris-HCl buffer solution (8.3 mM; pH 7.4); NaCl, 0.2 M; n ) 3. The concentration of HIV-1 LTR real sample was calculated according to the absorbance at 260 nm with the molar absorptivity of 6600 M-1 cm-1.

artificially added, could make the PRLS signals get increased and are almost the same as hybridization solution without PCR amplicons (columns 2, 5, and 8 in Figure 8), indicating that the detection of the target artificially added to the PCR amplicons is successful. It should be noted that the detection for both PCR amplicon artificial samples could be visually observed since the color of the solutions changes from red to light purple. In order to test the applicability of the present method, we applied it to the detection of target DNA sequence in HIV-1 LTR real sample of HB-hp3-LTR1.8. As Figure 9 shows, although the HB-hp3-LTR1.8 alone induces some aggregation of Au-NRs (column 4 in Figure 9) because the long DNA sequences of 1840 bp might have the tendency of self-folding into a secondary structure, the mixture of HB-hp3-LTR1.8 and probe DNA brings about much stronger degree of aggregation of Au-NRs (column 5 in Figure 9), showing that the probe DNA has hybridized with the target DNA sequence in the long sequence real sample. Besides that, we take advantage of restriction endonuclease reaction to treat with HB-hp3-LTR1.8 to get a sequence fragment with 314 bp, which contains the target ssDNA, in order to avoid the aggregation of Au-NRs induced by the real sample itself in the hybridization test (as stated above and shown in column 4 in Figure 9). It was found that these sequence fragments resulted from the restriction endonuclease reaction could not induce aggregation of Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

8429

Au-NRs any more (column 7 in Figure 9), and when probe DNA is added into the Au-NRs solution in the presence of these short sequence fragments, strong enhanced PRLS signals were measured (column 8 in Figure 9), indicating the successful detection of the hybridization between the target DNA with probe DNA in the real sample. By the way, a control experiment of enzymes indicates that these enzymes do not interfere with the detection (column 6 in Figure 9). Therefore, the present method for detecting target DNA in real samples related to the HIV gene can be applied to practice. Mechanism Investigations. Our experiment on ionic strength effect has proved that the aggregation of Au-NRs here is not saltinduced as that of citrate-coated Au-NPs in the presence of dsDNA formed during the hybridization. In the case of citrate-coated AuNPs, ssDNA is adsorbed on negatively charged Au-NPs to prevent salt-induced aggregation, whereas dsDNA cannot be adsorbed, and thus salt-induced aggregation of Au-NPs occurs,11 whereas when it comes to Au-NRs, a different mechanism of aggregation should be considered. When added to Au-NRs suspension, ssDNA is also adsorbed onto the surface of the positively charged Au-NRs. The adsorption process is mainly governed by the electrostatic interaction between the anionic backbone phosphates of oligonucleotides and the cationic surfactant bilayer around the nanorods.42-44 However, the charge of ssDNA itself is negligible compared with the net charge of CTAB-coated Au-NRs, so Au-NRs remain highly positively charged after reacting with ssDNA. That is, ssDNA could not change the suspension state of Au-NRs, and there is no obvious change in the PRLS spectra. Besides that, the sequence length of 21-mer is about 2-4 nm, which is much shorter than the longitude of Au-NRs, so the ssDNA can uncoil and the bases are exposed in the solution for further hybridization. Once target DNA is added in, however, there are two possibilities for the interaction in the system. One possibility is that both target DNA and probe DNA are adsorbed onto different Au-NRs, then the basepairing effect causes the ssDNA to form double-helical DNA, making the Au-NRs close, and eventually the aggregation of large amounts of Au-NRs occurs. That is to say, the base pairing is the driving force for the aggregation of Au-NRs as reported by Dujardin et al.23 The second possibility is that the hybridization between target DNA and probe DNA is first completed as the consequence of the formation of dsDNA, and the electrostatic interaction of the dsDNA with adjacent Au-NRs becomes stronger than that of ssDNA, ultimately, the strong electrostatic interaction induces the aggregation of Au-NRs. As for the second possibility, there are several reasons to prompt the interaction. One reason is that the charge density of dsDNA is much larger than that of ssDNA,45 and once dsDNA is formed, the dsDNA molecules might be enough “glue” to stick the nanorods together at the sides by the bilayer of CTAB one by one through strong electrostatic interaction. Thus, Au-NRs get together with the increase of target DNA content, inducing enhanced PRLS signals. The other reason might be related to the difference of chain flexibility between ssDNA and dsDNA. Namely, the rigid structure of dsDNA is in favor of its interaction with the Au-NRs on facets along the long axis. This conclusion can be (42) Zhu, D.-M.; Evans, R. K. Langmuir 2006, 22, 3735–3743. (43) Ganachaud, F.; Elaissari, A.; Pichot, C.; Laayoun, A.; Cros, P. Langmuir 1997, 13, 701–707. (44) Wei, G.; Wang, L.; Liu, Z.; Song, Y.; Sun, L.; Yang, T.; Li, Z. J. Phys. Chem. B 2005, 109, 23941–23947. (45) Rosa, M.; Dias, R.; Miguel, M. d. G.; Lindman, B. Biomacromolecules 2005, 6, 2164–2171.

8430

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

Table 2. Control Experiments of Different Reagents (n ) 5)a IPRLS reagent

λmax/nm

CTAB CTAB and Au+, Ag+ gold nanospheres (Au-NSs) gold nanorods (Au-NRs)

340 340 560 555

ssDNA 136.4 ± 1.2 135.1 ± 2.7 97.8 ± 0.4 138.6 ± 2.1

dsDNA 141.7 ± 3.2 135.6 ± 3.1 111.1 ± 1.4 504.1 ± 5.7

a Concentrations: Tris-HCl buffer solution, 8.3 mM, pH 7.4; NaCl, 0.2 M; probe DNA, 8.33 nM; target DNA, 8.33 nM; CTAB, 31.67 mM; Au+, 3.4 × 10-4 M; Ag+, 5.7 × 10-5 M; Au-NSs, 1.05 nM; Au-NRs, 0.64 nM. The concentration of Au-NSs was estimated according to the absorbance at 524 nm with the molar extinction coefficient of 7.8 × 108 M-1 cm-1 (ref 47).

confirmed by the control experiments of CTAB-coated gold nanospheres (Au-NSs) in the same experimental conditions. The CTAB-coated AuNSs with the size of 20 nm, which were prepared from the continuous growth of the remaining Au seeds about 1 month without any future operation, can also withdraw high ionic strength and without any aggregation tendency. When target DNA is added into the suspension of the CTAB-coated Au-NSs, there is no obvious change of color and PRLS signals (see Table 2), indicating that no obvious aggregation happens, perhaps owing to that the large curvature of Au-NSs is not suitable for its interaction with the stiff chain of dsDNA.46 On the other hand, whether the solutions contain the unreduced gold and silver ions or not,35 the interaction between the free CTAB micelle with DNA could be negligible. That is to say, the byproduct Au-NSs, free CTAB micelle, and other unreduced metal ions will not interfere in the detection of hybridization. CONCLUSION In this contribution, we report an optical genosensing system by taking Au-NRs as a recognition platform that has combined advantages such as fast, label-free, and simple one-step measurements. It has potential application to monitor the process of hybridization in real time. As demonstrated in this work, we can perform the detection under physiological conditions, which is crucial for direct detection of biological samples. Moreover, the genosensing system has better stability under high ionic strength conditions than citrate-coated Au-NPs. In a word, we are only required to add the samples containing the target DNA to our genosensing system and detection can be completed in less than 5 min, much faster and easier than recent reports.32 Thus, the success of the present report for HIV-related DNA detection indicates the potential of the Au-NRs as a genosensing platform to be put into practice toward simple, fast, and exact disease diagnoses. ACKNOWLEDGMENT This research was supported by the Ministry of Science and Technology of the People’s Republic of China (2006CB 933100) and the National Natural Science Foundation of China (NSFC, No: 20425517, 20675065). SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review May 17, 2008. Accepted September 15, 2008. AC801005D (46) Wang, Y.; Dubin, P. L.; Zhang, H. Langmuir 2001, 17, 1670–1673. (47) Jana, N. R.; Gearheart, L.; Murphy, C. J. Langmuir 2001, 17, 6782–6786.