Microarray-Based Detection of Protein Binding and Functionality by

Anal. Chem. 2005, 77, 5770-5774

Microarray-Based Detection of Protein Binding and Functionality by Gold Nanoparticle Probes Zhenxin Wang,† Jason Lee,‡ Andrew R. Cossins,‡ and Mathias Brust*,†

Centre for Nanoscale Science, Department of Chemistry and School of Biological Sciences, The University of Liverpool, Liverpool, L69 7ZD U.K.

We report a microarray format for the detection of proteins and protein functionality (kinase activity) based on marking either specific antibody-protein binding or peptide phosphorylation events by attachment of gold nanoparticles followed by silver deposition for signal enhancement. The attachment of the gold nanoparticles is achieved by standard avidin-biotin chemistry. The detection principle is resonance light scattering. Highly selective recognition of standard proteins (proteins A and G) down to 1 pg/mL for proteins in solution and 10 fg for proteins on the microarray spots is demonstrated. Enzyme activity of the kinase (PKA) is detected with high specificity down to a limit of 1 fg for an established peptide substrate (kemptide) on the microarray spots. Kinase inhibition by the inhibitor (H89) is shown, demonstrating the potential for high-throughput screening for inhibitors. The development of microarray-based, high-throughput screening techniques is currently revolutionizing many aspects of biological detection and monitoring technologies.1 The strongest impact, to date, has been made in the area of DNA diagnostics, allowing for the massively parallel detection of a large number of different sequences on a small scale and in an automated fashion. Entering the postgenomic era has created a demand for similarly successful approaches to the discovery of proteins, their functions, and interactions.1,2 Currently, microarray techniques mainly rely on the use of fluorescent molecular dye labels,1,2 which have several potential drawbacks, such as the need of amplification, i.e., the use of PCR in DNA diagnostics due to lack of sensitivity, and photoinstability of the dyes employed.1,3-5 Recently, it has been recognized that metal, semiconductor, and magnetic nanoparticles can offer a unique set of physical properties that may be exploited in biological detection assays as an alternative to the use of fluorescent dyes.6-18 In particular, the detection of resonance light scattering (RLS) by metal particles represents a great step forward, toward higher sensitivity, with the eventual goal of * Corresponding author. E-mail: [email protected]. Fax:(+44) 151-794-3588. † Centre for Nanoscale Science, Department of Chemistry. ‡ School of Biological Sciences. (1) Schena, M. Microarray analysis; Wiley-Liss: Hoboken, NJ, 2003. (2) MacBeath, G.; Schreiber, S. L. Science 2000, 289, 1760-1763. (3) Kodadek, T. Chem. Biol. 2001, 8, 105-115. (4) Sano, T.; Smith, C. L.; Cantor, C. R. Science 1992, 258, 120-122. (5) Schweitzer, B.; Wiltshire, S.; Lambert, J.; O’Malley, S.; Kukanskis, K.; Zhu, Z.; Kingsmore, S. F.; Lizardi, P. M.; Ward, D. C. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 10113-10119.

5770 Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

detecting single biomolecular binding events. This is, in principle, possible since the optical detection of individual metal particle probes can be achieved.8 Specific recognition of proteins by conjugates of gold nanoparticles and antibodies followed by an enhancement step based on the electroless deposition of silver onto the gold particles has been employed for protein identification in histochemistry using dot and blot assays.19,20 Using the same detection principle, Mirkin and co-workers developed a DNAbased bio-bar-code for the specific detection of proteins on microarrays.7,12,13,17 This technology uses DNA hybridization as the specific surface recognition process. An alternative binding scheme based on antibody-antigen recognition and labeling with DNA-stabilized gold nanoparticles via avidin-biotin interactions on a 96-well microplate has been developed by Niemeyer and Ceyhan.9 Here we report a highly sensitive and simple microarray method, which can be employed both for protein detection and for assaying enzyme functionality. The method is based on labeling specific recognition or phosphorylation events on a microarray with gold nanoparticles using avidin-biotin chemistry followed by silver enhancement and RLS detection. In two different proofof-principle experiments, we demonstrate that it is possible to (i) clearly discriminate between the standard analytes protein A and (6) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382, 607-609. (b) Alivisatos, A. P.; Johnsson, K. P.; Peng, X.; Wilson, T. E.; Loweth, C. J.; Bruchez, M. P.; Schultz, P. G. Nature 1996, 382, 609611. (7) Taton, T. A.; Mirkin, C. A.; Letsinger, R. L. Science 2000, 289, 17571760. (8) Yguerabide, J.; Yguerabide, E. E. J. Cell. Biochem. 2001, 37 (Suppl.), 7181. (9) Niemeyer, C. M.; Ceyhan, B. Angew. Chem., Int. Ed. 2001, 40, 36853688. (10) Niemeyer, C. M. Trends Biotechnol. 2002, 20, 395-401. (11) Oldenburg, S. J.; Genick, C. C.; Clark, K. A.; Schultz, D. A. Anal. Biochem. 2002, 309, 109-116. (12) Nam, J. M.; Thaxton, C. S.; Mirkin, C. A. Science 2003, 301, 1884-1886. (13) Cao, Y. C.; Jin, R.; Nam, J.; Thaxton, C. S.; Mirkin, C. A. J. Am. Chem. Soc. 2003, 125, 14676-14677. (14) Alivisatos, A. P. Nat. Biotechnol. 2004, 22, 47-52. (15) Brakmann, S. Angew. Chem., Int. Ed. 2004, 43, 5730-5734. (16) Katz, E.; Willner, I. Angew. Chem., Int. Ed. 2004, 43, 6042-6108. (17) Georganopoulou, D. G., Chang, L.; Nam, J.-M.; Thaxton, C. S.; Mufson, E. J.; Klein, W.; Mirkin, C. A. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 22732276. (18) Wang, Z.; Lee, J.; Cossins, A. R.; Brust, M. IEE Proc.-Nanobiotechnol. 2005, 152, 85-88. (19) Holgate, C. S.; Jackson, P. I.; Cowen, P. N.; Bird, C. C. J. Histochem. Cytochem. 1983, 31, 938-944. (20) Moeremans, M.; Daneels, G.; Van Dijck, A.; Langanger, G.; Mey, J. J. Immunol. Methods 1984, 74, 353-360. 10.1021/ac050679v CCC: $30.25

© 2005 American Chemical Society Published on Web 07/23/2005

protein G using biotinylated antibodies as specific recognition agents, and (ii) identify peptides that can serve as a substrate for phosphorylation by the protein kinase PKA. The latter is achieved by coupling the phosphorylation reaction with biotinylation of the kinase substrate using a biotin-modified ATP as a cosubstrate.21-24 The identification and characterization of substrates and inhibitors of kinases is of interest for the biomedical sciences since phosphorylation of proteins by kinases plays a key role in regulating cellular processes and is believed to be involved in many diseases such as cancer, diabetes, and inflammations.21-35 Importantly, our approach circumvents the need for radioactive labeling usually required for assaying kinase activities.2,21-25 Also, our method is capable of multiplexing since, in principle, the screening of many different protein-protein or peptide-kinase interactions can be carried out simultaneously on the same microarray. EXPERIMENTAL SECTION Materials and Reagents. Tetrachloroaurate (HAuCl4), bovine serum albumin (BSA), avidin, protein A, protein G, biotin-modified antibodies, human IgM, horse IgG, and mouse IgG were purchased from Sigma-Aldrich Co. Cyclic adenosine 5′-monohosphatedependent protein kinase (PKA, catalytic subunit) was purchased from New England Biolabs (Beverly, MA). Kemptide (LRRASLG) and control peptide (LRRAGLG) were purchased from Pepsyn Ltd. (Liverpool, U.K.). Biotin-ATP (adenosine 5′-triphosphate [γ]biotinyl-3,6,9-trioxaundecanediamine (ATP [γ]Biotin-LC-PEOamine)) was purchased from Alt. Inc. (Lexington, KY). Inhibitor H89 was purchased from Calbiochem-Novabiochem Corp. (La Jolla, CA). Aldehyde-glass microscope slides (Nexterion Slide AL) and NHS-ester glass microscope slides (Nexterion Slide H) were purchased from Schott AG. Preparation of Gold Nanoparticle Probes. Avidin-stabilized gold nanoparticles were prepared by stirring an aqueous mixture of succinylated avidin (2 mL, 1 mg/mL, pH 7.5, 50 mM PB, 0.15 M NaCl) and citrate-stabilized 13-nm gold nanoparticles (20 mL, (21) Davies, S. P.; Reddy, H.; Caivano, M.; Cohen, P. Biochem. J. 2000, 351, 95-105. (22) Falsey, J. R.; Renil, M.; Park, S.; Li, S.; Lam, K. S. Bioconjugate Chem. 2001, 12, 346-353. (23) Houseman, B. T.; Huh, J. H.; Kron, S. J.; Mrksich, M. Nat. Biotechnol. 2002, 20, 270-274. (24) Min, D.-H.; Mrksich, M. Curr. Opin. Chem. Biol. 2004, 8, 554-558. (25) Schutkowski, M.; Reineke, U.; Reimer, U. ChemBioChem. 2005, 6, 513521. (26) Hunter, T. Cell 2000, 100, 113-127. (27) Beveridge, M.; Park, Y. W.; Hermes, J.; Marenghi, A.; Brophy, G.; Santos, A. J. Biomol. Screening 2000, 5, 205-212. (28) Lesaicherre, M.-L.; Uttamchandani, M.; Chen, G. Y. J.; Yao, S. Q. Bioorg. Med. Chem. Lett. 2002, 12, 2079-2083. (b) Lesaicherre, M.-L.; Uttamchandani, M.; Chen, G. Y. J.; Yao, S. Q. Bioorg. Med. Chem. Lett. 2002, 12, 2085-2088. (29) Manning, G.; Whyte, D. B.; Martinez, R.; Hunter, T.; Sudarsanam, S. Science 2002, 298, 1912-1934. (30) Cohen, P. Nat. Rev. Drug Discovery 2002, 1, 309-315. (31) Espanel, X.; Walchli, S.; Ruckle, T.; Harrenga, A.; Huguenin-Reggiani, M.; van Huijsduijnen, R. H. J. Biol. Chem. 2003, 278, 15162-15167. (32) Noble, M. E. M.; Endicott, J. A.; Johnson, L. N. Science 2004, 303, 18001805. (33) Fabian, M. A.; et al. Nat. Biotechnol. 2005, 23, 329-336. (34) Shults, M. D.; Janes, K. A.; Lauffenburger, D. A.; Imperiali, B. Nat. Methods 2005, 2, 277-283. (35) Sun, H.; Low, K. E.; Woo, S.; Noble, R. L.; Graham, R. J.; Connaughton, S. S.; Gee, M. A.; Lee, L. G. Anal. Chem. 2005, 77, 2043-2049.

3.8 × 10-9 M) for 30 min at room temperature.36,37 Excess protein was removed by repeated centrifugation at 13 000 rpm (11340g, 3×) using a Sigma 1-13 centrifuge (Sigma), and resuspension in PBS buffer. The material was then stored at 4 °C. Protein Microarrays. Protein microarrays were manufactured using a BioRobotics MicroGrid II Arrayer (Genomic Solutions Inc, Ann Arbor, MI). Proteins A and G were spotted on aldehydetreated glass microscope slides in a PBS spotting buffer (∼1 nL/ spot, pH 7.5, 50 mM PB, 0.15 M NaCl) with 40% glycerol (v/v) included to prevent evaporation of the nanodroplets.2,38 After an overnight incubation under 75% humidity at 15 °C, each slide was quickly rinsed with 20 mL of phosphate buffer (pH 7.5) containing 1% (w/v) BSA and then immersed in the blocking buffer (pH 7.5 0.05 M PB, 0.15 M NaCl containing 1% (w/v) BSA and 0.1 M ethanolamine) for 1 h to remove remaining free aldehyde groups. The arrays were incubated with biotin-modified proteins, human IgM, horse IgG, and mouse IgG, which were diluted to the desired concentration with 60 µL of probe buffer, (pH 7.5, 50 mM PB, 0.15 M NaCl, supplemented with 0.1% Tween-20 (v/v) and 1% BSA (w/v)). Following 1-h incubation at room temperature, the slides were rinsed with probe buffer and then washed with 50 mL of washing buffer (pH 7.5, 50 mM PB, 0.15 M NaCl supplemented with 0.1% Tween-20) for 3 min (3 times). The slides were rinsed with PBS (3 times) and Milli-Q water (18.2MΩ, 3 times) and dried by centrifugation (200g for 5 min). Then the microarrays were treated with 200 µL of a solution of gold nanoparticle probes (6.1 × 10-9 M) in PBS (pH 7.5, 50 mM PB, 0.15 M NaCl) for 2 h at room temperature. The slides were subjected to a series of rinsing steps: (1) PBS buffer with 0.1% Tween-20 for 3 min (three times); (2) PBS buffer for 3 min (three times); (3) H2O for 3 min (three times). The rinsed slides were dried by centrifugation. Peptide Microarrays. Kemptide and control peptide were spotted on NHS-ester functionalized slides in 0.3 M PBS (pH 8.5, 0.2M NaCl and 20 µg/mL BSA) and processed as described for the protein microarrays. Before phosphorylation, the slides were washed with washing buffer (pH 7.5, 20 mM tris, 150 mM NaCl, 10 mM EDTA, 1 mM EGTA, 0.1% Triton X-100) for 10 min (3 times). The slides were subsequently washed for 10 min with kinase buffer (pH 7.5, 50 mM tris, 10 mM MgCl2, 1 mM DTT), incubated with kinase buffer containing 100 µM ATP for 10 min, and washed for an additional 10 min with kinase buffer. The slides were then incubated for 1 h at 30 °C with 200 µL of kinase solution. This was applied to the slides under a frame-seal incubation chamber (MJ Research INC., Waltham, MA). The kinase solution was composed of the recommended buffer for PKA containing 50 µM biotin-ATP and 100 units of PKA. For an inhibititor test, 50 µM H89 was applied to the kinase solution. Following the 1-h incubation, the slides were washed with washing buffer for 5 min (3 times), with washing buffer without Triton X-100 for 5 min (3 times), and with H2O for 3 min (3 times). The slides were then centrifuged at 200g for 1 min to remove excess water. The dry microarrays were treated with 200 µL of a probe solution of gold nanoparticles and washed as described for the protein microarrays. Silver Enhancement and Detection. After being labeled with gold nanoparticles, silver enhancer (1:1 mixed (1 mL total volume) (36) Turkevich, J.; Stevenson, P. C.; Hillier, J. Discuss. Faraday Soc. 1951, 11, 55-75. (37) Frens, G. Nat. Phys. Sci. 1973, 241, 20-22. (38) http://cmgm.stanford.edu/pbrown/.

Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

5771

Figure 1. Schematic representation of one spot on the microarray showing from left to right, (a) the binding of a biotin-modification antibody to the spotted protein and (b) the phosphorylation and biotinylation of the spotted peptide followed by the attachment of avidin-stabilized gold nanoparticles, the silver enhancement step, and reading by detection of resonant light scattering.

solutions A (AgNO3) and B (hydroquinone) (Sigma-Aldrich)) was applied to each microarray for 8 min and washed with water (3 times). After signal amplification by silver deposition, the slides were imaged with a Qiagen Hight-light system collecting light scattered at a wavelength of 565 nm (Qiagen Instruments) or, alternatively, with a standard office use flatbed scanner.7 AFM images of the surfaces of individual spots on the microarray before and after the enhancement step were obtained with an Explorer atomic force microscope (Thermo Microscopes) operating at noncontact mode with a silicon tip and a tripod scanner. RESULTS AND DISCUSSION Protein Microarrays. Proteins were spotted and immobilized on commercial aldehyde-functionalized glass microscope slides by a standard robotic procedure.2,38 The aldehyde groups on the glass surface react readily with the primary amines of proteins to form a Schiff base linkage. The binding scheme and detection rationale employed is schematically shown in Figure 1a. After recognition of the immobilized protein by a biotinylated antibody, avidin-stabilized gold nanoparticles were used to label these recognition events. Subsequently, a silver enhancement step was applied to the microarrays for signal amplification since the light scattering properties of gold nanoparticles by themselves are relatively poor, if the particles are smaller than ∼40 nm.8 The effect of the deposition of silver on the surface morphology of the spots has been investigated by AFM. Figure 2 shows AFM images before and after the enhancement procedure. While our experiment on the scale used does not resolve the lateral dimensions of the metal islands on the spot, it is clearly demonstrated that their average height increases due to silver deposition on the gold 5772

Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

Figure 2. AFM images showing the surface of a spot on the microarray (protein G probed with horse IgG and labeled with gold nanoparticles) before (a) and after (b) the silver enhancement step. The x-y scale is 5 × 5 µm.

seeds from ∼15 to 100 nm. Gold and silver features of this final size are very strong light scatterers and are readily detected even by a common office use flatbed scanner.7 It was found that the detection sensitivity and dynamic range was critically related to the amount of silver deposited. While short exposure times (2-3 min) to the enhancement reaction led to poor sensitivity, long exposure times (15 min) generated significant background noise and signal saturation, which also reduced the sensitivity and dynamic range. Under optimum conditions (8-min exposure time), silver features of the size shown in Figure 2 were achieved. In this study, all results shown were obtained under these conditions. To detect specific protein-protein (here: antibody-antigen) interactions on the microarray, two different proteins (protein A. protein G) were spotted on the array to be probed with three different antibodies (human IgM, horse IgG, mouse IgG), which were chosen for their well-known specificities. Human IgM binds only to protein A and horse IgG binds only to protein G, while mouse IgG binds to both protein A and protein G. All antibodies

Figure 3. Light scattering images of protein microarrays after probing with antibodies, labeling positives with gold nanoparticles and enhancement by silver deposition. Protein G was spotted on columns 1 and 2, and protein A was spotted at columns 3 and 4. The microarrays were probed with horse IgG (a), human IgM (b), and mouse IgG (c). The concentrations of spotted and probe proteins were 10 µg/mL and 100 ng/mL, respectively.

used were modified with biotin. The results of this array-based binding assay are shown in Figure 3. As anticipated, only the spots probed with specifically binding antibodies give a positive signal; i.e., human IgM only recognizes protein A, horse IgG only recognizes protein G, and mouse IgG, as expected, does not discriminate between proteins A and G. These results demonstrate that the immobilized proteins retain their recognition function on the glass surface and, more importantly, that any nonspecific binding of antibodies is below the detection limit. Quantitative Protein Detection. A set of experiments was designed to determine the detection limits of the method described. By varying the concentration of protein G in the spotting solution while keeping the horse IgG concentration of the probing solution constant, the optimal signal-to-noise ratio was obtained at spotting solution concentrations above 10 ng/mL. Given a spotting solution volume of ∼1 nL, this means that the actual amount of protein in each spot is on the order of 10 fg. This signal increased linearly with the logarithm of the protein concentration from 10 ng/mL to 10 µg/mL and began to saturate above 10 µg/m, indicating a dynamic range of 3 orders of magnitude as shown in Figure 4a. Blank samples give readings of light scattering intensity of 400 ( 100 units. The detection limit can thus be regarded as the concentration of analyte that gives a scattering signal of at least 700 units (blank reading plus three times the standard deviation of the blank). This corresponds in Figure 4a to protein concentrations of below 20 ng/mL in the spotting solution. The run-to-run and sample-to-sample reproducibilities are high with standard deviations that are expressed in the error bars given in each figure. Furthermore, these findings are not affected by the use of nanoparticles from different preparation batches. For the antibody in solution, much lower concentrations are needed. For example, as shown in Figure 4b, the detection limit for specific binding using IgG is below 20 pg/mL in the probe solution. For comparison, conventional fluorescence techniques for detecting the same target on a microarray have a detection limit of 1 ng/mL,39 i.e., 2 orders of magnitude less sensitive, whereas Mirkin’s bio-bar-code approach (without PCR) has a reported detection limit of 1 fg/mL (30 aM) for the detection of prostate-specific antigen (PSA).12 Saturation was obtained at a concentration of 500 ng/mL giving a dynamic range of 5 orders (39) Haab, B. B.; Dunham, M. J.; Brown, P. O. Genome Biol. 2001, 2, research 0004.1.-0004.13.

Figure 4. Images of microarrays and corresponding logarithmic plots of the integrated light scattering intensity as a function of the concentration of protein G in the spotting solution (a) and of horse IgG in the probe solution (b). In (a), the concentration of horse IgG was 500 ng/mL, and in (b), the concentration of the protein G in the spotting solution was 10 µg/mL.

Figure 5. Light scattering images of peptide microarrays after phosphorylation, labeling positives with gold nanoparticles and enhancement by silver deposition (a) without and (b) in the presence of 50 µM inhibitor H89 during the phosphorylation reaction. Kemptide was spotted on columns 1 and 2, and control peptide was spotted at columns 3 and 4. The concentrations of spotted peptides were 10 µg/mL.

of magnitude. As only a few microliters of each protein is sufficient to fabricate thousands of spots, purified proteins without target amplification may be readily detected. Peptide Microarrays for Enzyme Functionality Assays. The reaction scheme for the determination of kinase functionality is shown in Figure 1b. The peptide, kemptide (LRRASLG), is phosphorylated on the microarray spot by the kinase, PKA, using biotinylated ATP (biotin-ATP) as a cosubstrate.2,21,28,33 Through this reaction, together with the phosphate, a biotin site is transferred to the kemptide. In a subsequent step, the biotinylated peptide is specifically labeled by attachment of avidin-stabilized gold nanoparticles followed by the conventional silver enhancement step. Multiple copies of two different peptides, kemptide and the control peptide (LRRAGLG), which is not a substrate of PKA, were spotted on commercial NHS-ester containing microscopy slides. Figure 5a shows the selective biotinylation of the spots that contain the kemptide, while no detectable reaction occurs on the control spots. This demonstrates the high selectivity of this detection method, i.e., the absence of detectable nonspecific attachment of gold nanoparticles. Figure 5b shows the same array in the presence of the specific PKA inhibitor H89. As expected, the Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

5773

rescence-based arrays are, in turn, more sensitive than the conventional radiolabeling methods.28 The signal scaled linearly with the logarithm of the kemptide concentration in the spotting solution over 4 orders of magnitude between 1 ng/mL and 10 µg/mL. In all experiments where low signals were detected, it was observed that the remaining signal originated only from the center of the spot rather than from its entire surface. This is a consequence of typical inhomogeneities during the spotting and drying process, which result in more material being deposited in the spot center. As a consequence, the central region of the spot is more sensitive than other regions and therefore still leads to a detectable signal when no scattering can be detected from other regions.

Figure 6. Images of microarray and corresponding logarithmic plot of the integrated light scattering intensity as a function of the concentration of peptide in the spotting solution.

enzyme activity is in this case very low, close to the detection limit. Importantly, this experiment demonstrates the potential for high-throughput screening for kinase inhibitors, which is of great current interest for drug discovery.21,30-34 Detection Limits and Dynamic Range. The sensitivity of this method was examined by varying the concentration of kemptide in the spotting solution while keeping the concentrations of biotin-ATP and PKA constant. The results are shown in Figure 6. Here the detection limit is 1 ng/mL of peptide in the spotting solution with the total amount of kemptide on each spot (1-nL spotting volume) equal to or less than 1 fg (1.5 amol). This outperforms currently reported fluorescence-based kinase functionality arrays, which typically operate with amounts of phosphorylated substrate in the order of several picograms.28 Fluo-

5774

Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

CONCLUSIONS A versatile gold nanoparticle-based microarray format with high selectivity and sensitivity for both protein binding and kinase functionality has been developed. The results indicate that the method has reasonably low detection limits for protein binding combined with a large dynamic range. For kinase functionality arrays, our method is significantly more sensitive than previously reported approaches. Well-known biomolecular recognition systems were chosen here to establish this new microarray format by proof-of-principle experiments. However, our approach is readily transferable to real analytical problems such as the quantitative determination of protein abundance and modification in cell or tissue extracts and, in addition, shows great promise for high-throughput screening of kinase functionality and inhibitor activity on a microarray format. ACKNOWLEDGMENT The authors thank Professor David G. Fernig (Liverpool University) and Professor Richard L. McCreery (Ohio State University) for helpful comments and the BBSRC (Centre for Bioarray Innovation at the University of Liverpool) for financial support. Received for review April 20, 2005. Accepted June 27, 2005. AC050679V