Simultaneous and Label-Free Determination of Wild-Type and Mutant

Sep 22, 2009 - Cancer Research Institute, Central South University. , §. California State University, Los Angeles. Cite this:Anal. Chem. 2009, 81, 20...
0 downloads 0 Views 544KB Size
Download PDF
Anal. Chem. 2009, 81, 8441–8446

Simultaneous and Label-Free Determination of Wild-Type and Mutant p53 at a Single Surface Plasmon Resonance Chip Preimmobilized with Consensus DNA and Monoclonal Antibody Yongcan Wang,† Xu Zhu,† Minghua Wu,‡ Ning Xia,† Jianxiu Wang,*,† and Feimeng Zhou§ College of Chemistry and Chemical Engineering, Central South University, Changsha, Hunan, People’s Republic of China 410083, Cancer Research Institute, Central South University, Changsha, Hunan, People’s Republic of China 410013, and Department of Chemistry and Biochemistry, California State University, Los Angeles, Los Angeles, California 90032 Simultaneous determination of wild-type and total p53 proteins (wild-type and mutant combined) present in cancer cell lysates has been performed with a dualchannel surface plasmon resonance (SPR) instrument. To achieve specificity, each channel of the SPR chip was modified with a consensus double-stranded (ds-) DNA and a monoclonal antibody. The high affinity of the consensus ds-DNA to the wild-type p53 and the antibody to total p53 results in remarkably low detection levels (10.6 and 1.06 pM for the wild-type and total p53, respectively). The difference between the SPR signals reveals the extent of p53 mutation, which is indicative of cancer development. The SPR signals increase with the p53 concentration across a wide range (from low picomolar to nanomolar levels) that amply encompasses the typical cellular p53 concentrations. The applicability of the method to real sample analysis has been demonstrated with the comparative analyses of normal and cancer cell lysates. The normal cell samples all displayed significantly higher levels of wild-type p53. In contrast, elevated levels of mutant p53 were observed from the cancer cell lysates. In comparison with enzyme-linked immunosorbant assay (ELISA), SPR obviates the need of a second antibody labeled with an enzyme in the “sandwich enzyme immunoassay” format and is capable of realtime monitoring of the binding events. Thus, SPR could potentially serve as an attractive technique for rapid, sensitive, reliable, and label-free cancer diagnoses. p53, a well-known tumor suppressor and a transcription factor, plays an important role in inducing cell cycle arrest for DNA repair or apoptosis to eliminate severely damaged cells.1-3 In more than 50% of human cancers, p53 has been found to be mutated, raising * To whom correspondence should be addressed. E-mail: jxiuwang@ mail.csu.edu.cn. † College of Chemistry and Chemical Engineering, Central South University. ‡ Cancer Research Institute, Central South University. § California State University, Los Angeles. (1) Balint, E.; Vousden, K. H. Br. J. Cancer 2002, 85, 1813–1823. (2) Evan, G. I.; Vousden, K. H. Nature 2001, 411, 342–348. (3) Polyak, K.; Xia, Y.; Zweier, J. L.; Kinzler, K. W.; Vogelstein, B. Nature 1997, 389, 300–305. 10.1021/ac9014269 CCC: $40.75  2009 American Chemical Society Published on Web 09/22/2009

clinical possibilities for both diagnosis and treatment.4 Human p53 comprises 393 amino acids with three distinct functional domains, viz., the transactivation domain at the N-terminus, the DNAbinding domain, and the oligomerization domain near the C-terminus.5-9 Mutations that deactivate p53 in cancer cells usually occur in the DNA-binding domain. The DNA consensus site contains two or more copies of the 10 base pair half-site with a sequence of PuPuPuC(A/T)(T/A)GPyPyPy (where Pu and Py represent purines and pyrimidines, respectively).10-14 Once mutated, the p53 protein loses its ability to bind the consensus double-stranded (ds-) DNA. Mutations of p53 also lead to change or even loss of p53 binding activity to its downstream genes, such as the mdm2 gene, resulting in aberrant cell proliferation and malignant cellular transformation.15 An abnormal expression of mutant p53 protein could lead to carcinogenesis of several major types of human cells.16 Thus, determination of the extent of p53 mutation is of great significance in clinical research and early diagnosis of cancers. Thus far, a variety of methods for measuring endogenous p53 have been reported. These methods include enzyme-linked immunosorbant assay (ELISA),17-20 electrophoretic mobility shift (4) Koshland, D. E. J. Science 1993, 262, 1953. (5) Clore, G. M.; Omichinski, J. G.; Sakaguchi, K.; Zambrano, N.; Sakamoto, H.; Appella, E.; Gronenborn, A. M. Science 1994, 265, 386–391. (6) Kato, S.; Han, S.-Y.; Liu, W.; Otsuka, K.; Shibata, H.; Kanamaru, R.; Ishioka, C. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 8424–8429. (7) May, P.; May, E. Oncogene 1999, 18, 7621–7636. (8) Rose, S. L.; Robertson, A. D.; Goodheart, M. J.; Smith, B. J.; DeYoung, B. R.; Buller, R. E. Clin. Cancer Res. 2003, 9, 4139–4144. (9) Stavridi, E. S.; Huyen, Y.; Sheston, E. A.; Halazonetis, T. D. Protein Rev. 2005, 2, 25–52. (10) El-Deiry, W. S.; Kern, S. E.; Pietenpol, J. A.; Kinzler, K. W.; Vogelstein, B. Nat. Genet. 1992, 1, 45–49. (11) Funk, W. D.; Pak, D. T.; Karas, R. H.; Wright, W. E.; Shay, J. W. Mol. Cell. Biol. 1992, 12, 2866–2871. (12) Miyashita, T.; Reed, J. C. Cell 1995, 80, 293–299. (13) Zambetti, G. P.; Bargonetti, J.; Walker, K.; Prives, C.; Levine, A. J. Genes Dev. 1992, 6, 1143–1152. (14) Wang, J.; Zhu, X.; Tu, Q.; Guo, Q.; Zarui, C. S.; Momand, J.; Sun, X.; Zhou, F. Anal. Chem. 2008, 80, 769–774. (15) Bai, L.; Zhu, W. J. Cancer Mol. 2006, 2, 141–153. (16) Xiang, C.; Shen, C.; Wu, Z.; Qin, Y.; Zhang, Y.; Liu, C.; Chen, J.; Zhang, S. J. Occup. Health 2007, 49, 279–284. (17) Thomas, M. D.; McIntosh, G. G.; Anderson, J. J.; McKenna, D. M.; Parr, A. H.; Johnstone, R.; Lennard, T. W.; Horne, C. H.; Angus, B. J. Clin. Pathol. 1997, 50, 143–147.

Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

8441

assay,21,22 immunostaining,23-25 electrochemistry,14,26 and immunohistochemistry.27,28 ELISA, based on the catalytic reaction by an enzyme conjugated to a second antibody that recognizes p53, is the most popular and best developed method. In general, the primary antibody used to capture p53 interacts with a conserved region of the p53 molecule, and the four conserved regions of p53 in the DNA-binding domain have been found to be the mutation hotspots identified in tumors.29 Thus, several commercially available detection kits are suitable only for the determination of the total p53 protein (i.e., wild-type and mutant combined). Jagelska´ et al. replaced the primary antibody with a consensus ds-DNA for binding p53 in solution to the ELISA plate, allowing sequence-specific binding of wild-type p53 to DNA to be analyzed.20 We recently developed a nanoparticle-amplified voltammetric method for sensitive detection of wild-type p53.14 Our method also utilizes a consensus ds-DNA as the capture ligand but obviates the employment of the secondary (detection) antibody. Instead, the method resorts to the ferrocene-capped gold nanoparticle/streptavidin conjugates for signal transduction. Although we carried out ELISA which determines both the wildtype and mutant p53 proteins and demonstrated that the nanoparticle-amplified voltammetric method can probe the cellular status (i.e., cancer vs normal cells) as reliably as ELISA, the voltammetric detection is quite different than ELISA (i.e., different sensing platforms and different detection techniques). Such a difference does not allow an accurate determination of the extent of p53 mutation (i.e., the ratio between the wild-type and mutant p53 molecules), which is an important criterion for cancer diagnosis.30 Moreover, the nanoparticle-amplified voltammetric detection scheme is relatively complicated since multiple crosslinking and binding steps are involved. Thus, it is highly desirable to develop a simple, sensitive, and label-free technique that can simultaneously determine both wild-type and mutant p53 proteins at the cellular level. Surface plasmon resonance (SPR) is an optical technique that is sensitive to extremely small changes in the thickness or refractive index of adsorbate at the sensor surface.31-34 Biosensors based on SPR have been widely used for monitoring biological (18) Weisinger, G.; Tendler, Y.; Zinder, O. Brain Res. Protoc. 2000, 6, 71–79. (19) Gould, K. A.; Nixon, C.; Tilby, M. J. Mol. Pharmacol. 2004, 66, 1301– 1309. (20) Jagelska´, E.; Bra´zda, V.; Pospisilova´, S.; Vojtesek, B.; Palecek, E. J. Immunol. Methods 2002, 267, 227–235. (21) Adachi, Y.; Chen, W.; Shang, W. H.; Kamata, T. Anal. Biochem. 2005, 342, 348–351. (22) Thornborrow, E. C.; Maurer, M.; Manfredi, J. J. Methods Mol. Biol. 2003, 223, 87–100. (23) Noguchi, S.; Koyama, H.; Kasugai, T.; Tsuji, N.; Tsuda, H.; Akiyama, F.; Motomura, K.; Inaji, H. Oncology 1998, 55, 450–455. (24) Pardo, F. S.; Hsu, D. W.; Zeheb, R.; Efird, J. T.; Okunieff, P. G.; Malkin, D. M. Br. J. Cancer 2004, 91, 1678–1686. (25) Suto, T.; Sugai, T.; Nakamura, S.; Uesugi, N.; Sasaki, R.; Kanno, S.; Saito, K. Oncology 1997, 54, 407–413. (26) Potesil, D.; Mikelova, R.; Adam, V.; Kizek, R.; Prusa, R. Protein J. 2006, 25, 23–32. (27) Barnes, D. M.; Dublin, E. A.; Fisher, C. J.; Levison, D. A.; Millis, R. R. Hum. Pathol. 1993, 24, 469–476. (28) Henke, R. P.; Kru ¨ ger, E.; Ayhan, N.; Hu ¨ bner, D.; Hammerer, P.; Huland, H. J. Urol. 1994, 152, 1297–1301. (29) Cho, Y.; Gorina, S.; Jeffrey, P. D.; Pavietich, N. P. Science 1994, 265, 346– 355. (30) Dowell, S. P.; Wilson, P. O. G.; Derias, N. W.; Lane, D. P.; Hall, P. A. Cancer Res. 1994, 54, 2914–2918. (31) Tao, N.; Boussaad, S.; Huang, W.; Arechabaleta, R. A.; D’Agnese, J. Rev. Sci. Instrum. 1999, 70, 4656–4660.

8442

Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

processes as diverse as DNA hybridization, DNA/protein binding, and protein/protein interaction.35-41 Moreover, rapid and labelfree analysis of specific biomolecular interactions can be conducted in real time.42-46 In comparison with ELISA, SPR is more advantageous in that the use of an enzyme conjugated to a second antibody for ELISA is not required. To our knowledge, only one paper has shown the qualitative measurement of the interaction between p53 and antibodies in serum.47 In that work, histidinetagged p53 (instead of antibody) was first bound to the SPR chip, and as a result the method is not applicable for the determination of p53 in biological specimen. For cancer diagnosis, it is the knowledge about the p53 in tissue samples that is critical.30 In this study, determination of wild-type and mutant p53 proteins has been carried out using a dual-channel SPR instrument. Wild-type p53 was measured via its binding to a consensus ds-DNA preimmobilized onto a carboxymethylated dextran film. Relying on the specific recognition of a monoclonal antibody (PAb421) toward wild-type p53 and its mutant, the total p53 concentration can also be determined. Thus, the mutated p53 level can be deduced from the difference between the total and wildtype p53 concentrations. Samples from several cancer cell lines were analyzed to demonstrate the viability of this methodology for clinical diagnosis. EXPERIMENTAL SECTION Chemicals and Materials. N-(3-Dimethylaminopropyl)-N′ethylcarbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), tris(hydroxymethyl) aminomethane (Tris), cystamine hydrochloride, ethylenediaminetetraacetic acid disodium salt (EDTA), dithiothreitol (DTT), ethanolamine hydrochloride (EA), KH2PO4, and K2HPO4 were acquired from Sigma (St. Louis, MO). Carboxymethylated dextran was synthesized from the dextran precursor (Acros Organics, Belgium) according to a procedure reported previously.48 DNA samples were purchased from Shanghai Sangon Co., Ltd. (Shanghai, China). To im(32) Economou, E. N. Phys. Rev. 1969, 182, 539–554. (33) Pockrand, I. Surf. Sci. 1978, 72, 577–588. (34) Swalen, J. D.; Gordon, J. G.; Philpott, M. R.; Brillante, A.; Pockrand, I.; Santo, R. Am. J. Phys. 1980, 48, 669–672. (35) 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. (36) Komolov, K. E.; Senin, I. I.; Philippov, P. P.; Koch, K.-W. Anal. Chem. 2006, 78, 1228–1234. (37) Liang, X.; Nazarenus, T. J.; Stone, J. M. Biochemistry 2008, 47, 3645–3653. (38) Liu, J.; Tian, S.; Tiefenauer, L.; Nielsen, P. E.; Knoll, W. Anal. Chem. 2005, 77, 2756–2761. (39) Lyon, L. A.; Musick, M. D.; Natan, M. J. Anal. Chem. 1998, 70, 5177– 5183. (40) Tsoi, P. Y.; Yang, M. Biosens. Bioelectron. 2004, 19, 1209–1218. (41) Yu, X.; Xu, D.; Cheng, Q. Proteomics 2006, 6, 5493–5503. (42) Song, F.; Zhou, F.; Wang, J.; Tao, N.; Lin, J.; Vellanoweth, R. L.; Morquecho, Y.; Wheeler-Laidman, J. Nucleic Acids Res. 2002, 30, e72. (43) Zeng, D.; Wang, J.; Yin, L.; Zhang, Y.; Zhang, Y.; Zhou, F. Front. Biosci. 2007, 12, 5117–5123. (44) Zhang, Y.; Xu, M.; Wang, Y.; Toledo, F.; Zhou, F. Sens. Actuators, B 2007, 123, 784–792. (45) Yao, X.; Li, X.; Toledo, F.; Zurita-Lopez, C.; Gutova, M.; Momand, J.; Zhou, F. Anal. Biochem. 2006, 354, 220–228. (46) Battaglia, T. M.; Masson, J.-F.; Beaudoin, S.; Sierks, M.; Booksh, K. S. Anal. Chem. 2005, 77, 7016–7023. (47) Campagnolo, C.; Meyers, K. J.; Ryan, T.; Atkinson, R. C.; Chen, Y.-T.; Scanlan, M. J.; Ritter, G.; Old, L. J.; Batt, C. A. J. Biochem. Biophys. Methods 2004, 61, 283–298. (48) Stevens, M. M.; Allen, S.; Davies, M. C.; Roberts, C. J.; Schacht, E.; Tendler, S. J. B.; VanSteenkiste, S.; Williams, P. M. Langmuir 2002, 18, 6659–6665.

mobilize the consensus ds-DNA onto the dextran surface, an aminated oligonucleotide probe with a sequence of 5′H2N-(CH2)6-TTT TTA GAC ATG CCC AGA CAT GCC C-3′ and its complementary target with a sequence of 5′-GGG CAT GTC TGG GCA TGT CT-3′ were used. The sequences for producing the nonconsensus ds-DNA are 5′-H2N-(CH2)6-TTT TTG TCG GCC GAG GTC GGC CGA G-3′ and 5′-CTC GGC CGA CCT CGG CCG AC-3′. Recombinant p53 sample was purchased from BD Biosciences Pharmingen (San Diego, CA). Mouse monoclonal antibodies PAb421 and PAb246 were obtained from EMD Chemicals Inc. (Darmstadt, Germany). p53 from normal and cancer cell lysates was extracted according to our published procedure.14 Other reagents were all from commercial sources with analytical purity and used as received. All stock solutions were prepared daily with deionized water treated with a water purification system (Simplicity 185, Millipore Corp., Billerica, MA). Instruments. The SPR measurements were conducted on a BI-SPR 1000 system (Biosensing Instrument Inc., Tempe, AZ) equipped with a dual-channel flow cell and two through-the-handle six-port injection valves.49 Thoroughly degassed phosphatebuffered saline (PBS buffer, 10 mM phosphate/10 mM NaCl, pH 7.0) was used as the carrier solution. The recombinant p53 solution or p53 from normal or cancer cell lysate was preloaded into 50 µL sample loops on the valves and then delivered into the flow cells by Genie Plus syringe pumps (Kent Scientific, Torrington, CT) at a flow rate of 10 µL/min. Procedures. Solution Preparation. TE buffer (10 mM Tris/ 1.0 mM EDTA, pH 7.0) was used to prepare the DNA probe solutions. DNA targets were dissolved in TNE buffer (TE buffer/ 0.1 M NaCl). Samples of p53 were diluted with PBS buffer comprising 20 mM DTT. EDC/NHS solution was prepared by mixing 75 mM EDC with 15 mM NHS in water right before the dextran film activation step (vide infra). Cystamine hydrochloride, carboxymethylated dextran, and EA were all dissolved in water. SPR Sensor Surface Modifications. Au films with a 50 nm thickness and a 2 nm Cr underlayer were either coated onto BK7 glass slides using a sputter coater (model 208, Kurt J. Lesker, Clairton, PA) or purchased from Biosensing Instrument. Prior to surface modification, the Au films were annealed in a hydrogen flame. The preparation of carboxymethylated dextran-covered SPR chips has been described in our previous papers.45,49 Such chips are similar to the CM5 chip commercially available from Biacore (GE Healthcare, Piscataway, NJ) in terms of eliminating nonspecific adsorption and offering higher binding capacity. The main difference is the preparative chemistry: whereas the CM5 chip is formed via epoxy derivatization of the terminal hydroxyl selfassembled monolayers (SAMs) and the subsequent nucleophilic reaction of the dextran under alkaline conditions, the hydrogellike dextran layer is prepared in our method via cross-linking carboxymethylated dextran onto the cystamine SAMs. The antibody and ds-DNA molecules can be attached to the dextran chips through either off-line immobilization or online derivatization. In the off-line procedure, DNA probe was first attached to the dextran surface by covering the surface with 50 µL of EDC/ NHS solution containing 2 µM of aminated DNA for 3 h. After washing thoroughly with TE buffer and then water, the surface (49) Du, M.; Zhou, F. Anal. Chem. 2008, 80, 4225–4230.

was soaked in 0.1 M EA solution for 20 min to eliminate the unreacted NHS ester groups. The hybridization reaction between the DNA probe and target was allowed to proceed by casting 2 µM DNA target onto the chip for 2 h. Similarly, the PAb421 antibody was immobilized onto the dextran film by casting a mixture of EDC, NHS, and 5 µg/mL PAb421 onto the chip surface. The surface was deactivated with 0.1 M EA. For online immobilization of consensus ds-DNA or monoclonal antibody, a dextran-modified chip was first mounted onto the SPR instrument. Injections of 50 µL of DNA probe and 2 µM DNA target solution were sequentially made into one channel. In the other channel, 50 µL of 5 µg/mL PAb421 was injected through a different injection valve into the second fluidic channel. The performance of the SPR chips via online derivatization is highly comparable to that via the off-line immobilization but is more advantageous because both consensus ds-DNA and antibody species can be attached to the same chip for simultaneous detection of the wildtype and total p53 proteins. We therefore used online derivatization for most of this work. SPR Detection of p53 in Normal and Cancer Cell Lysates. Samples of p53 were injected into the SPR cell to react with the respective consensus ds-DNA and antibody preimmobilized into the individual fluidic channels. To achieve low detection level, the detection was carried out at a relatively slow flow rate (10 µL/ min) to provide sufficient time for the reaction. For determining the kinetic parameters the flow rate was increased to 30 µL/min to enhance the mass transfer rate.50 RESULTS AND DISCUSSION Figure 1 illustrates the scheme behind the simultaneous SPR detection of wild-type and mutant p53 using consensus ds-DNA (panel a) and PAb421 (panel b). In Figure 1a, consensus ds-DNA molecules are tethered onto the carboxymethylated dextran surface in one fluidic channel. In the assay step, wild-type p53 in solution forms conjugates with consensus ds-DNA at the DNAbinding domain.51,52 The most stable binding of p53 tetramers is known to occur at the full binding sites of individual ds-DNA molecules (Figure 1a).14,53 In Figure 1b, p53 interacts with the PAb421 antibody, which is preattached to the dextran surface via the amine coupling chemistry. The use of a dextran film was to limit nonspecific adsorption of proteins or other species present in cell lysates.45,54,55 Curve a in Figure 2A is a representative SPR sensogram corresponding to the formation of the conjugate between wildtype p53 and consensus ds-DNA. The SPR dip shift, measured as the change in the SPR baseline before and after the sample injection, is about 0.0501°. The exposure of a single-stranded (ss)DNA-covered sensor chip to p53 solution caused a much smaller change (∼0.0059° in curve b). Such a small change can be ascribed to the nonspecific binding of the C-terminus of p53 to (50) Myszka, D. G. Curr. Opin. Biotechnol. 1997, 8, 50–57. (51) Wang, Y.; Schwedes, J. F.; Parks, D.; Mann, K.; Tegtmeyer, P. Mol. Cell. Biol. 1995, 15, 2157–2165. (52) McLure, K. G.; Lee, P. W. K. EMBO J. 1998, 17, 3342–3350. (53) Mclure, K. G.; Lee, P. W. K. EMBO J. 1999, 18, 763–770. (54) Jung, L. S.; Nelson, K. E.; Campbell, C. T.; Stayton, P. S.; Yee, S. S.; PerezLuna, V.; Lopez, G. P. Sens. Actuators, B 1999, 54, 137–144. (55) Lahiri, J.; Isaacs, L.; Grzybowski, B.; Carbeck, J. D.; Whitesides, G. M. Langmuir 1999, 15, 7186–7198.

Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

8443

Figure 1. Schematic representation of the simultaneous SPR detection of wild-type and total p53 proteins by consensus DNA duplexes (a) and monoclonal antibody (b) in two separate fluidic channels covering a dextran-modified Au sensor chip.

Figure 2. SPR sensograms correspond to (A) injections of 50 µL of 0.266 nM wild-type p53 solution into fluidic channels covered with consensus ds-DNA (curve a), ss-DNA (curve b), and nonconsensus ds-DNA (curve c) and (B) injections of 50 µL of 0.106 nM wild-type p53 solution (curve a) and 0.106 nM IgG solution (curve b) into fluidic channels covered with PAb421. The arrows indicate the time when the injections were made.

ss-DNA.14,56 The SPR angle remained essentially unchanged at the chip covered with nonconsensus ds-DNA (curve c). Thus, the binding of p53 to the consensus sequence is highly specific and is in accordance with many previous studies.10-14 Notice that the injection peaks of the three curves all contain changes caused by the difference in the refractive indices between the sample and the carrier solutions. In the p53 solution, the addition of a small amount of dithiothreitol is necessary to prevent oxidation of the cysteine residues at the surface of the p53 molecules, which can reduce the binding affinity of p53 to ds-DNA.57,58 As can be seen from Figure 2A, the efficacy of the dextran film in eliminating nonspecific adsorption of p53 is evident, as there is little change in the baseline of curve c after the p53 sample had been replenished out of the fluidic channel. Figure 2B is an overlay of two SPR sensograms acquired from injecting 50 µL of 0.106 nM wild-type p53 solution (curve a) and 0.106 nM IgG solution (curve b) into channels precovered with the PAb421 antibody. The overall SPR dip shift of 0.0527° was observed from curve a, indicative of binding between wild-type p53 and PAb421. This is in contrast to (56) Bakalkin, G.; Selivanova, G.; Yakovleva, T.; Kiseleva, E.; Kashuba, E.; Magnusson, K. P.; Szekely, L.; Klein, G.; Terenius, L.; Wimam, K. G. Nucleic Acids Res. 1995, 23, 362–369. (57) Sun, X.; Vinci, C.; Makmura, L.; Han, S.; Tran, D.; Nguyen, J.; Hamann, M.; Grazziani, S.; Sheppard, S.; Gutova, M.; Zhou, F.; Thomas, J.; Momand, J. Antioxid. Redox Signaling 2003, 5, 655–665. (58) Makmura, L.; Hamann, M.; Areopagita, A.; Furuta, S.; Mun ˜oz, A.; Momand, J. Antioxid. Redox Signaling 2001, 3, 1105–1118.

8444

Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

curve b, which shows a much smaller SPR dip shift (0.0064°). Note that such a change even at lower p53 concentration is slightly higher than that depicted by curve a in Figure 2A, suggesting that the binding affinity between p53 and its antibody is higher than that between p53 and the consensus ds-DNA. This point is also discussed below. It is worth noting that the channel preimmobilized with the PAb421 antibody will yield the total p53 concentration when mutant p53 is also present, as this antibody interacts specifically with the N-terminal domain of p53.59 The efficacy of the consensus ds-DNA for selective detection of wild-type p53 is further confirmed by comparing the response of a consensus ds-DNA to that of an antibody known to be specific to wild-type p53 (PAb24660,61). In the first experiment, a solution comprising 0.106 nM IgG and 0.106 nM p53 was delivered to two separate fluidic channels that had been, respectively, preimmobilized with PAb246 and the same consensus ds-DNA used for curve a in Figure 2A. Interestingly, the overall SPR dip shift observed from the channel covered with the consensus ds-DNA upon injecting the mixed solution is almost the same as that acquired with the injection of 0.106 nM p53 alone (∼0.0282°). However, in the case of the fluidic channels immobilized with the PAb246 antibody, the SPR dip shift upon injection of the mixed (59) Me´plan, C.; Richard, M. J.; Hainaut, P. Biochem. Pharmacol. 2000, 59, 25–33. (60) Milner, J.; Medcalf, E. A.; Cook, A. C. Mol. Cell. Biol. 1991, 11, 12–19. (61) Martinez, J.; Georgoff, I.; Martinez, J.; Levine, A. J. Genes Dev. 1991, 5, 151–159.

Figure 3. SPR sensograms of the binding between PAb246 (A) and the consensus ds-DNA (B) covered sensor chips with the increasing concentrations of wild-type p53. Curves a, b, and c correspond to p53 concentrations of 2.5, 1.2, and 0.5 nM, respectively. The solid lines are the original data, and the dashed lines show the simulated ones. The arrows indicate the time when the injections were made.

Figure 4. Dependences of SPR dip shift on p53 concentration at fluidic channels covered with consensus ds-DNA (A) and the monoclonal antibody PAb421 (B). The absolute errors for p53 are shown as the error bars, and the relative standard deviations (RSDs) in panels A and B range from 14.3% to 2.4% and from 13.4% to 1.5%, respectively. Concentrations of p53 determined are 0.00106, 0.00266, 0.0106, 0.0532, 0.106, 0.266, 0.532, 1.06, 5.32, 26.6, and 53.2 nM. Each concentration was repeated for at least three times. The insets in panels A and B are enlarged displays of the linear portions of the responses at the low p53 concentrations. Table 1. Wild-Type and Total p53 Concentrations in Normal and Cancer Cell Lysates cell lysates

wild-type 53 (nM)

total p53 (nM)

mutant p53 (nM)

mutation percentage (%)

liver cell L-02 endothelial cell ECV-304 liver cancer cell HepG-2 colorectal cancer cell SW620

2.87 ± 0.043 1.03 ± 0.013 0.24 ± 0.015 0.64 ± 0.015

3.14 ± 0.13 1.18 ± 0.047 1.26 ± 0.058 1.98 ± 0.023

0.27 ± 0.17 0.15 ± 0.060 1.02 ± 0.073 1.34 ± 0.038

8.6 13 81 68

solution (∼0.0380°) is greater than that obtained from injecting 0.106 nM p53 alone (∼0.0337°). This suggests that the selectivity of consensus ds-DNA to p53 is better than the monoclonal antibody PAb246. In the second experiment, the affinity constants between ds-DNA and p53 and between PAb246 and p53 were determined by simulating a series of binding curves recorded at different p53 concentrations (Figure 3). In performing the simulation, the bulk refractive index changes were subtracted out. The binding constant (KA ) 1.1 × 108 M-1) between the consensus DNA to wild-type p53 is only slightly smaller than that between PAb246 and p53 (3.7 × 108 M-1). This KA value is also in good agreement with that reported previously.62 However, PAb246 is substantially more expensive than DNA; thus, determination of wild-type p53 has been performed at an SPR chip modified with consensus ds-DNA. The dependence of SPR dip shift at the consensus ds-DNAcovered channel on the wild-type p53 concentration is shown in Figure 4A. As can be seen, the SPR dip shift increases sharply (62) Wo ¨lcke, J.; Reimann, M.; Klumpp, M.; Go ¨hler, T.; Kim, E.; Deppert, W. J. Biol. Chem. 2003, 278, 32587–32595.

with the wild-type p53 concentration and the slope begins to level off at around 1.06 nM. The much steeper slope (∆θ/deg ) 0.0193 + 0.0849C/nM, r ) 0.97, shown in the inset of panel A) in the lower region of the plot (from 0.0106 to 1.06 nM) reflects the significant increase in the formation of the ds-DNA/p53 conjugates, whereas the smaller slope between 1.06 and 53.2 nM (∆θ/ deg ) 0.107 + 0.00131C/nM, r ) 0.99) is resulted from the decrease of consensus ds-DNA that remains available for the binding reaction. The relative standard deviations (RSDs) range from 14.3% to 2.4% within the range of p53 concentrations determined. These RSDs are quite reasonable since the replicate measurements were carried out either in different fluidic channels or at different chips. It is well-known that film uniformity and surface conditions could affect the SPR signals.45 The lowest point of the calibration curve corresponds to 0.0106 nM of wild-type p53 protein, which compares well with that obtained by adsorptive stripping voltammetry of p5326 and is only 4.8 times higher than that by the amplified voltammetric detection of wild-type p53 using gold nanoparticle/streptavidin conjugates capped with multiple ferrocene tags.14 However, the obviation of labeling, signal Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

8445

the p53 gene had been severely mutated in these cancer cells. The decline in the wild-type p53 concentration in HepG-2 and SW620 is accompanied by a concomitant elevation of the mutant p53 concentration. It is also worth mentioning that voltammetric detection of wild-type p53 from normal and cancer cell lines brings about variability inherent in multiple cross-linking and binding steps.14 Obviation of the labeling step using the nanoparticles also simplifies the experimental procedure and improves the fidelity of the measured results. The applicability of the dual-channel SPR thus affords a simple, selective, and accurate approach to the simultaneous determination of wild-type and mutant p53 at the cellular levels.

Figure 5. SPR sensograms showing the simultaneous detection of wild-type and total p53 present in a colorectal cancer cell lysate at fluidic channels covered with the consensus ds-DNA (a) and PAb421 (b). The arrow indicates the time when injections of the sample were made.

amplification, and sample accumulation steps makes the present method simple and rapid. The variation of the SPR dip shift recorded at a PAb421-covered channel with the wild-type p53 concentration is shown in Figure 4B. The plot is in resemblance to that in Figure 4A in that the slope begins to change at ∼0.106 nM. However, a closer examination of panels A and B reveals some differences. First, the slope is steeper in Figure 4B than its counterpart in Figure 4A, suggesting that the binding affinity between p53 and PAb421 is greater. In fact, the KA value of 6.9 × 108 M-1 measured by us is greater than both values for the consensus ds-DNA and the PAb246. Such a higher binding affinity is responsible for the lower detection level of p53 (0.00106 nM). We should note that, for both consensus ds-DNA and antibody PAb421, binding of p53 was found to obey the Langmuir isotherm and the abovementioned KA values were obtained from the best fits based on the Langmuir adsorption model. Finally, simultaneous determination of wild-type and total p53 in cancer cell lysates is illustrated by sensograms shown in Figure 5. The overall SPR dip shifts of 0.0245° and 0.0560° were observed from curves a and b, respectively. The use of a single chip for detections of both wild-type and mutant p53 provides higher fidelity to assessment of the level of p53 mutation because the variability in surface property from chip to chip is avoided. To demonstrate further the clinical prospect of this method, we performed assays of two normal cell lysates and two cancer cell lysates. The mutant p53 concentration can be deduced from the difference between wild-type and total p53 (Table 1). HepG-2 and SW620 (cancer cell lines) both exhibited lower levels of wild-type p53 than L-02 and ECV-304 (normal cell lines), a trend in excellent agreement with a separate ELISA. Our observation suggests that

8446

Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

CONCLUSIONS Simultaneous determination of wild-type and total p53 concentrations in cell lysates has been accomplished with a dual-channel SPR instrument. Total p53 (i.e., wild-type and mutant p53 combined) concentration can be measured through the binding of p53 to a monoclonal antibody, while wild-type p53 in the same sample is detected via its specific interaction with a consensus ds-DNA. The difference between wild-type and total p53 provides an accurate measure of the level of p53 mutation. Our results demonstrate that wild-type and total p53 concentrations as low as 0.0106 and 0.00106 nM can be measured, respectively. Such detection levels are remarkably low considering that no sample labeling and signal amplification were attempted. The binding affinity constants between p53 and consensus ds-DNA and between p53 and monoclonal antibodies were also measured, and the high affinities of both are apparently responsible for the high sensitivity of the method. In comparison with ELISA, SPR is more advantageous in that no enzyme conjugated to a second antibody is needed and the binding reactions can be studied in real time. The amenability of the method to real sample analysis has been demonstrated by determining the variation of wild-type p53 from normal to cancer cell lysates. The decline of the wild-type p53 concentration in cancer cell lysates due to the severe mutation of the p53 gene is accompanied by a considerable elevation of the mutant p53. The method described herein is sensitive, selective, and rapid, and the compact instrument and simple procedure could make the method suitable for in-point cancer diagnosis. ACKNOWLEDGMENT Partial support of this work by the National Natural Science Foundation of China (Nos. 20975114 and 20775093 to J.W. and No. 30770825 to M.W.) and an NSF-RUI Grant (No. 0555224 to F.Z.) is gratefully acknowledged. Received for review June 29, 2009. Accepted September 4, 2009. AC9014269