- Email: [email protected]
Colorimetric detection of protein microarrays based on nanogold probe coupled with silver enhancement
Journal of Immunological Methods 285 (2004) 157 – 163 www.elsevier.com/locate/jim
Colorimetric detection of protein microarrays based on nanogold probe coupled with silver enhancement Ru-Qiang Liang, Cui-Yan Tan, Kang-Cheng Ruan * Key Laboratory of Proteomics, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 320 Yue-Yang Road, Shanghai 200031, China Received 28 July 2003; received in revised form 10 November 2003; accepted 11 November 2003
Abstract This work presents a method for analyzing protein microarrays using a colorimetric nanogold probe coupled with silver enhancement (gold – silver detection). In this method, the gold nanoparticles were introduced to the microarray by the specific binding of the gold-conjugated antibodies or streptavidins and then coupled with silver enhancement to produce black image of microarray spots, which can be easily detected with a commercial CCD camera. The method showed high detection sensitivity (1 pg of IgG immobilized on slides or 2.75 ng/ml IgG in solution) and a good linear correlation between the signal intensity and the logarithm of the sample concentration. The examination of this method in analyzing a demonstrational ToRCH antigen microarray developed in our lab showed an identical result as in the fluorescent method. These results suggest the colorimetric gold – silver detection method has potential applications in proteomics research and clinical diagnosis. D 2004 Elsevier B.V. All rights reserved. Keywords: Gold – silver detection; Protein microarrays; CCD detector
1. Introduction As the post-genomic era develops, an increase for knowledge proposes a challenge as biological questions become more complex and need to be addressed on the functional genomic and proteomic level (Vidal, Abbreviations: BSA, bovine serum albumin; CCD, chargecoupled device; CMV, Cytomaglovirus; CVB6, Coxsackie Virus B6; GOPTS, 3-glycidoxypropyltrimethoxysilane; HSV, Herpse Simplex Virus; IgG, immunoglobulin G; PBS, phosphate-buffered saline; RBV, Rubella Virus; SAM, self-assembled monolayer; Tox, Toxplasma gondii. * Corresponding author. Tel.: +86-21-54921168; fax: +86-2154921011. E-mail address: [email protected] (K.-C. Ruan). 0022-1759/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2003.11.008
2001). DNA microarray technology has played an important role in functional genomics research. However, cell homeostasis, tissue or an organism can be reflected much more accurately, directly and informatively by the dynamics of the proteome. To explore the territory of the uncharacterized proteome, novel strategies in research technology are needed and their development is very challenging (Weinberger et al., 2000). Protein microarray technology is an excellent tool for these applications. In most of the existing DNA or protein microarray analysis, fluorescence detection is used (MacBeath and Schreiber, 2000; Haab et al., 2001; Zhu et al., 2001; Mezzasoma et al., 2002; Robinson et al., 2002; Tam et al., 2002). However, expensive instrumenta-
158
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
tions, such as a laser confocal microarray scanner, have limited the wide applications of DNA and protein microarray technologies, especially in clinical diagnosis. Many efforts have been made to solve this problem. Joos et al. (2000) successfully used an enzyme-linked immuno-assay to detect autoimmune antigen microarrays, with a 40-fg detection limit. The gold-conjugated antibodies coupled with silver enhancement have been employed in histochemistry and later in dot and blot assay (Holgate et al., 1983; Moeremans et al., 1984; Lackie, 1996). In this method the signal is largely amplified by the reduction of silver promoted by the gold particles. Based on the same principle, oligonucleotide-modified gold nanoparitcles were used to detect DNA microarrays (Taton et al., 2000), in which a simple commercial flatbed scanner was used as a detector, yielding a better or similar sensitivity in compared with fluorescence detection. Alexandre et al. (2001) has also implemented this colorimetric method in DNA microarray detection, as 1 fmol of biotinylated DNA attached on an array and 0.1 fmol of target DNA in a sample solution could be detected with a colorimetric-based workstation containing a charge coupled device (CCD) camera. Therefore, we thought, if the gold – silver detection method can be applied in protein microarray analysis, it should have greater potential applications in proteomics research and clinical diagnosis. Here we report the results of the study on the application of the gold – silver detection method in analyzing protein microarrays. These studies reveal that for this method, the detection limit for protein immobilized on microarray slides (human immuno-
globulin G, IgG) can be as low as 1 pg and for protein in aqueous solution (anti-human IgG, anti-IgG) 2.75 ng/ml. The detection dynamic ranges are from 2 pg to 1 ng for IgG immobilized on slides and from 5.5 ng/ ml to 1.4 Ag/ml for anti-IgG in solution. This detection sensitivity is comparable to fluorescence methods that require expensive equipment.
2. Materials and methods The schematic principle of the colorimetric gold – silver method is illustrated in Fig. 1. The human IgGs in the spots are covalently bound on the activated glass slides, the biotin-conjugated anti-IgGs are then bound to the human IgG. Excess unbound biotinconjugated anti-IgGs are removed by the washing with doubly distilled water (ddH2O). Upon the addition of the gold-conjugated streptavidin to the slides, they will bind to the biotin-conjugated anti-IgGs which are already bound to the IgGs, resulting in formation of a sandwich-like complex and introduction of the gold particles to the spot where the human IgGs are immobilized. In the silver enhancement, the Ag+ in the silver enhancement solution is reduced by the promotion of the gold particles, a large amount of silver is precipitated on the spots where the gold particles are introduced, giving the significant colorimetric images (Alexandre et al., 2001). The effect of the silver enhancement is very powerful, so that the black spots formed on slides can be detected with a commercial CCD camera (Olympus C-4000Z digital camera) mounted on the microscope. Furthermore, the
Fig. 1. Schematic illustration of gold-conjugated antibody recognition and signal amplification with silver enhancement.
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
silver precipitation on each spot is proportional to the amount of the gold nanoparticles deposited on the spot. The latter quantity is proportional to the quantity of the human IgG immobilized in the spot. Therefore, the gray level of each spot in the microarray can be used to analyze the quantity of the assayed component. Parenthetically, the gold particle can be introduced into the microarray spots in various manners depending on the type of the microarray and/or the experimental design. For instance, in Fig. 1, the gold particle can be also introduced by the labeling the gold to anti-IgG, or by the gold-conjugated protein G which binds to the anti-IgG, etc. 2.1. Materials Human IgG, rabbit myosin, gold-conjugated streptavidin, gold-conjugated anti-human IgG, 3glycidoxypropyltrimethoxysilane (GOPTS), silver enhancement solutions A and B, microscope glass slides, and cover glass were all purchased from Sigma (St. Louis, MO). Bovine serum albumin V(BSA) was purchased from Roche (Mannheim, Germany). Biotin-conjugated goat anti-human IgG (biotin-conjugated anti-IgG) and Cy5k-conjugated goat anti-human IgG were from Jackson ImmunoResearch Laboratories (West Grove, PA). Human serum samples were from Ruijin Hospital (Shanghai, China). All other chemicals were of analytical grade. 2.2. Preparation of self-assembled monolayer (SAM) The SAM formed on microscope glass slides was prepared in the manner previously described with some modifications (Bhatia et al., 1989; Eisen and Brown, 1999). Briefly, microscope glass slides were soaked in 10% NaOH and then 0.1 M HCl for 2 h, respectively. After thorough rinsing with doubly distilled water (ddH2O) and boiling for 1 h, the glass slides were set in 1% GOPTS (in 95% ethanol) at 37 jC for 6 h. The slides were then dried and incubated at 135 jC for another 1 h, and finally stored at 4 jC for further use. 2.3. Preparation of protein microarray Protein samples of human IgG or antigens were dissolved in phosphate-buffered saline (PBS, 137 mM
159
NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.4 mM KH2PO4, pH 7.4) with 40% glycerol. These solutions were transferred to 384-well plate respectively. The printing of protein microarrays on GOPTS-activated glass slides was performed with PixSys 5500 spotting robot (Cartesian Technology, Irvine, CA) mounted with ArrayIt SMP3 micro spotting pin from TeleChem (Sunnyvale, CA). On the resulting protein microarrays each spot has about 1 nl sample solution. The diameter of the spots and the space between spots were 100 and 300 Am, respectively. Immediately before use, the protein microarray slides were blocked with 1% BSA/PBS for 1 h at room temperature. After rinsing with PBS the slides were dried by centrifugation at 200 g for 5 min. 2.4. The detection of demonstrational ToRCH antigen microarrays To examine the application of the gold-silver detection in practice, the demonstrational ToRCH antigen microarray developed in our lab was used. The microarray is used for detecting the antibody of Cytomaglovirus (CMV), Toxoplasma gondii (Tox), Rubella Virus (RBV), Herpse Simplex Virus (HSV), and Coxsackie Virus B6 (CVB6). The ToRCH microarrays were first incubated with 10 Al of 100-fold dilution of patient’s serum and then washed with PBS to remove all unbound molecules. After that, in the colorimetric detection, the slides were probed with the gold-conjugated goat anti-human IgG and then silverenhanced, in which the silver-enhancement time was 15 min. The detection was carried out on the CCD camera as described above. By contrast in the fluorescent detection method, the slides was probed with Cyk 5-conjugated goat anti-human IgG, then the fluorescent detection was carried out on the ScanArray 5000 from PerkinElmer (Boston, MA).
3. Results and discussions The image of a microarray spotted with a series of different human IgG concentrations shown in Fig. 2 indicates that the spots printed with concentration as low as 1 Ag/ml IgG can be detected (line 12). The actual amount of the IgG contained in these spots is equal to or less than about 1 pg as the volume of the
160
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
Fig. 2. The gray level of the spot at various concentrations of human IgG immobilized on the microarray. Inset: the gold – silver detection image of different concentration human IgG immobilized on microarray. The concentrations of spotted human IgG from columns 1 to 12: 2.0 mg/ml, 1.0 mg/ml, 0.5 mg/ml, 0.25 mg/ml, 0.13 mg/ml, 62.5 Ag/ml, 31.25 Ag/ml, 15.625 Ag/ml, 7.8 Ag/ml, 3.9 Ag/ml, 1.95 Ag/ml and 0.97 Ag/ml. The concentration of biotin-conjugated anti-IgG: 11 Ag/ml; the silver-enhancement time: 15 min. The data in the plot were obtained from the image in the inset by averaging the mean gray levels of seven replicate spots.
sample printed to each spot is about 1 nl. Therefore, the lowest detectable amount for the IgG on the slides is about 1 pg, which is similar with fluorescence detection. The plot in Fig. 2 shows that the gray level of the spots is linear to the logarithm of the human IgG amounts immobilized on the slides in a range from 2 pg to 1 ng (corresponding IgG concentrations of sample printed on the slides are from 2 Ag/ml to 1 mg/ml). This indicates that the colorimetric detection method has a large linear detection dynamic range (about three orders of magnitude) for the components immobilized on the slides, which is very important for its practical application in quantitative analysis. The dynamic detection range also provides important information to determine the optimum concentration of protein immobilized on slides. Fig. 3A shows a set of images of model microarrays used to determine the concentration of biotinconjugated anti-IgG with the colorimetric gold –silver method. In the model microarrays, all the spots were printed onto the activated glass slides with human IgG solution of 250 Ag/ml. As shown in Fig. 3A the biotinconjugated anti-IgG as low as 2.75 ng/ml (17.2 pM)
can be detected, such a sensitivity is close to that in the fluorescent detection method (1 ng/ml) (Haab et al., 2001). Fig. 3B shows that the gray level of the spots is linearly dependent on the logarithm of concentration of biotin-conjugated anti-IgG from 5.5 ng/ ml to 1.4 Ag/ml, suggesting that the linear detection dynamic range spans 300 fold. Such high detection sensitivity and linear detection dynamic range are more valuable for the application of the gold – silver detection method because protein microarrays are mainly used to probe the components in aqueous solution. Fig. 3B reveals when the biotin-conjugated anti-IgG is higher than 1.4 Ag/ml, the gray level levels off, forming a plateau. This plateau might be caused by the limited capacity of IgG immobilized on the spots to capture biotin-conjugated anti-IgG. When the biotin-conjugated anti-IgG concentration in the sample is in excess of the capture capacity, more biotinconjugated anti-IgG can no longer be captured, so the gray level will not increase. According to the results obtained in Fig. 2, it is reasonable to expect that the detection sensitivity and the dynamic range of the gold –silver detection method can be improved with
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
161
Fig. 3. Detection of the biotin-conjugated anti-IgG at different concentration. (A) The detection images of biotin-conjugated anti-IgG using the gold – silver detection method. All the spots in the microarrays were printed with 250 Ag/ml human IgG. The biotin-conjugated anti-IgG concentration to be detected was shown in the image. The silver-enhancement time was 25 min. (B) Dynamic detection range of the gold – silver detection method. The data were obtained from the image shown in (A) by the same method mentioned in Fig. 2.
Fig. 4. The effects of various silver enhancement time on the image gray level. The concentration of human IgG spotted on the slides was 250 Ag/ml. The concentration of biotin-conjugated anti-IgG: 110 ng/ml (E), 1.1 Ag/ml (n) and 11 Ag/ml ( ).The silver enhancement time varies from 5 to 30 min.
.
162
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
an increase in the concentration of the immobilized IgG. Our results also indicate that the detection sensitivity is critically related to the silver enhancement reaction time, ten (see Fig. 4). For instance, almost no spots can be detected for 110 ng/ml biotin-conjugated anti-IgG when the ten is less than 10 min. Meanwhile, if the ten is extended to 15, 20 or 25 min, the gray levels of spots become obvious. Therefore, the detection sensitivity can be increased by the extension of ten. However, when the ten is more than about 25 min, the silver enhancement reaction seems to be saturated, a plateau in gray level change can be found. The time needed for the appearance of the plateau is termed as the saturated silver enhancement time (tsat). The facts related to tsat are complicated. At least, we thought, the concentration of probed component in the sample is one of them. As seen in Fig. 4, the lower the concentration of biotin-conjugated antiIgG, the longer the tsat is. This interpretation is consistent with the autocatalytic mechanism of the gold – silver enhancement reaction. It is certain that the ten used in the gold –silver method must be shorter than tsat, otherwise the gray level will not reflect the quantitative information. On the other hand, we also noticed that the longer ten lead to an increased background, in agreement with that reported by (Alexandre et al., 2001). To obtain the highest detectable sensitivity and a good signal/noise ratio the enhancement time must be optimized. In our experiment, it is found that the optimum ten ranges from 15 to 25 min. The specific IgG concentration yielded by a typical natural immune response is in a range from 10 ng/ml (Anthony et al., 1994) to over 3 Ag/ml (Granoff et al., 1986). This range is within the detection limit and linear detection dynamic range of our method described above. Fig. 5 shows the result of a demonstrational ToRCH antigen microarrays detected by the gold – silver and fluorescent method. The ToRCH antigen microarrays were developed in our lab (unpublished data). In the microarrays, CMV, Tox, RVB, HSV, and CVB6 antigens were printed onto the activated glass slides as introduced in materials and methods. After reaction with diluted serums of the patient, the microarrays were analyzed by the gold – silver and fluorescence detection methods. As seen in Fig. 5, the gold – silver detection method presented in a pattern similar to the fluores-
Fig. 5. The detection of the demonstrational ToRCH microarrays using the gold – silver and fluorescent method. The silver enhancement time was 15 min. The fluorescent image was scanned with 80% PMT Gain and 80% Laser Power.
cence method, suggesting that the colorimetric method have potential application in diagnostic antigen microarrays. Taken together, our results indicate the gold – silver colorimetric method can provide a highly accurate and linear dynamic range in practical applications to detect protein components present in both aqueous solution and immobilized on slides. In addition, a conventional digital CCD camera as a detector and well-managed conjugation of colloidal gold to biomolecules make the method easy to perform with lower cost. Therefore the method is of potential application in protein microarray.
Acknowledgements We thank Dr. Michelle Digman for making useful suggestion on the manuscript. This work was funded by the Innovation Project of Chinese Academy of Sciences and partially supported by the National 973 Project of China (No. 2002CB713802).
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
References Alexandre, I., Hamels, S., Dufour, S., Collet, J., Zammatteo, N., Longueville, F.D., Gala, J.-L., Remacle, J., 2001. Colorimetric silver detection of DNA microarrays. Anal. Biochem. 295, 1 – 8. Anthony, B.J., Concepcion, I.E., Concepcion, N.F., Vadheim, C.M., Tiwari, J., 1994. Relation between maternal age and serum concentration of IgG antibody to type III group B streptococci. J. Infect. Dis. 170, 717 – 720. Bhatia, S.K., Shriver-Lake, L.C., Prior, K.J., Georger, J.H., Calvert, J.M., Bredehorst, R., Ligler, F.S., 1989. Use of thiol-terminal silanes and heterobifunctional crosslinkers for immobilization of antibodies on silica surfaces. Anal. Biochem. 178, 408 – 413. Eisen, M.B., Brown, P.O., 1999. DNA arrays for analysis of gene expression. Methods Enzymol. 303, 179 – 205. Granoff, M., Shackelford, P.G., Suarez, B.K., Nahm, M.H., Cates, K.L., Murphy, T.V., Karasic, R., Osterholm, M.T., Pandey, J.P., Daum, R.S., 1986. Hemophilus influenzae type B disease in children vaccinated with type B polysaccharide vaccine. New Engl. J. Med. 315, 1584 – 1590. Haab, B.B., Dunham, M.J., Brown, P.O., 2001. Protein microarrays for highly parallel detection and quantitation of specific proteins and antibodies in complex solutions. Genome Biol. 2 research 0004.1 – 0004.13. Holgate, C.S., Jackson, P.I., Cowen, P.N., Bird, C.C., 1983. Immunogold – silver staining: new method of immunostaining with enhanced sensitivity. J. Histochem. Cytochem. 31, 938 – 944. Joos, T.O., Schrenk, M., Hopfl, P., Kroger, K., Chowdhury, U., Stoll, D., Schorner, D., Durr, M., Herick, K., Rupp, S., Sohn, K., Hammerle, H., 2000. A microarray enzyme-linked immunosorbent assay for autoimmune diagnostics. Electrophoresis 21, 2641 – 2650. Lackie, P.M., 1996. Immunogold silver staining for light microscopy. Histochem. Cell Biol. 106, 9 – 17.
163
MacBeath, G., Schreiber, S.L., 2000. Printing proteins as microarrays for high-throughput function determination. Science 289, 1760 – 1763. Mezzasoma, L., Bacarese-Hamilton, T., Di Cristina, M., Rossi, R., Bistoni, F., Crisanti, A., 2002. Antigen microarrays for serodiagnosis of infectious diseases. Clin. Chem. 48, 121 – 130. Moeremans, M., Daneels, G., Dijck, D.V., Langanger, G., Mey, J.D., 1984. Sensitive visualization of antigen – antibody reactions in dot and blot immune overlay assays with immunogold and immunogold/silver staining. J. Immunol. Methods 74, 353 – 360. Robinson, W.H., DiGennaro, C., Hueber, W., Haab, B.B., Kamachi, M., Dean, E.J., Fournel, S., Fong, D., Genovese, M.C., de Vegvar, H.E., Skriner, K., Hirschberg, D.L., Morris, R.I., Muller, S., Pruijn, G.J., van Venrooij, W.J., Smolen, J.S., Brown, P.O., Steinman, L., Utz, P.J., 2002. Autoantigen microarrays for multiplex characterization of autoantibody responses. Nat. Med. 8, 295 – 301. Tam, S.W., Wiese, R., Lee, S., Gilmore, J., Kumble, K.D., 2002. Simultaneous analysis of eight human Th1/Th2 cytokines using microarrays. J. Immunol. Methods 261, 157 – 165. Taton, T.A., Mirkin, C.A., Letsinger, R.L., 2000. Scanometric DNA array detection with nanoparticle probes. Science 289, 1757 – 1760. Vidal, M., 2001. A biological atlas of functional maps. Cell 104, 333 – 339. Weinberger, S.R., Morris, T.S., Pawlak, M., 2000. Recent trens in potein biochip technology. Pharmacogenomics 1, 395 – 416. Zhu, H., Bilgin, M., Bangham, R., Hall, D., Casamayor, A., Bertone, P., Lan, N., Jansen, R., Bidlingmaier, S., Houfek, T., Mitchell, T., Miller, P., Dean, R.A., Gerstein, M., Snyder, M., 2001. Global analysis of protein activities using proteome chips. Science 293, 2101 – 2105.
Colorimetric detection of protein microarrays based on nanogold probe coupled with silver enhancement Ru-Qiang Liang, Cui-Yan Tan, Kang-Cheng Ruan * Key Laboratory of Proteomics, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 320 Yue-Yang Road, Shanghai 200031, China Received 28 July 2003; received in revised form 10 November 2003; accepted 11 November 2003
Abstract This work presents a method for analyzing protein microarrays using a colorimetric nanogold probe coupled with silver enhancement (gold – silver detection). In this method, the gold nanoparticles were introduced to the microarray by the specific binding of the gold-conjugated antibodies or streptavidins and then coupled with silver enhancement to produce black image of microarray spots, which can be easily detected with a commercial CCD camera. The method showed high detection sensitivity (1 pg of IgG immobilized on slides or 2.75 ng/ml IgG in solution) and a good linear correlation between the signal intensity and the logarithm of the sample concentration. The examination of this method in analyzing a demonstrational ToRCH antigen microarray developed in our lab showed an identical result as in the fluorescent method. These results suggest the colorimetric gold – silver detection method has potential applications in proteomics research and clinical diagnosis. D 2004 Elsevier B.V. All rights reserved. Keywords: Gold – silver detection; Protein microarrays; CCD detector
1. Introduction As the post-genomic era develops, an increase for knowledge proposes a challenge as biological questions become more complex and need to be addressed on the functional genomic and proteomic level (Vidal, Abbreviations: BSA, bovine serum albumin; CCD, chargecoupled device; CMV, Cytomaglovirus; CVB6, Coxsackie Virus B6; GOPTS, 3-glycidoxypropyltrimethoxysilane; HSV, Herpse Simplex Virus; IgG, immunoglobulin G; PBS, phosphate-buffered saline; RBV, Rubella Virus; SAM, self-assembled monolayer; Tox, Toxplasma gondii. * Corresponding author. Tel.: +86-21-54921168; fax: +86-2154921011. E-mail address: [email protected] (K.-C. Ruan). 0022-1759/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2003.11.008
2001). DNA microarray technology has played an important role in functional genomics research. However, cell homeostasis, tissue or an organism can be reflected much more accurately, directly and informatively by the dynamics of the proteome. To explore the territory of the uncharacterized proteome, novel strategies in research technology are needed and their development is very challenging (Weinberger et al., 2000). Protein microarray technology is an excellent tool for these applications. In most of the existing DNA or protein microarray analysis, fluorescence detection is used (MacBeath and Schreiber, 2000; Haab et al., 2001; Zhu et al., 2001; Mezzasoma et al., 2002; Robinson et al., 2002; Tam et al., 2002). However, expensive instrumenta-
158
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
tions, such as a laser confocal microarray scanner, have limited the wide applications of DNA and protein microarray technologies, especially in clinical diagnosis. Many efforts have been made to solve this problem. Joos et al. (2000) successfully used an enzyme-linked immuno-assay to detect autoimmune antigen microarrays, with a 40-fg detection limit. The gold-conjugated antibodies coupled with silver enhancement have been employed in histochemistry and later in dot and blot assay (Holgate et al., 1983; Moeremans et al., 1984; Lackie, 1996). In this method the signal is largely amplified by the reduction of silver promoted by the gold particles. Based on the same principle, oligonucleotide-modified gold nanoparitcles were used to detect DNA microarrays (Taton et al., 2000), in which a simple commercial flatbed scanner was used as a detector, yielding a better or similar sensitivity in compared with fluorescence detection. Alexandre et al. (2001) has also implemented this colorimetric method in DNA microarray detection, as 1 fmol of biotinylated DNA attached on an array and 0.1 fmol of target DNA in a sample solution could be detected with a colorimetric-based workstation containing a charge coupled device (CCD) camera. Therefore, we thought, if the gold – silver detection method can be applied in protein microarray analysis, it should have greater potential applications in proteomics research and clinical diagnosis. Here we report the results of the study on the application of the gold – silver detection method in analyzing protein microarrays. These studies reveal that for this method, the detection limit for protein immobilized on microarray slides (human immuno-
globulin G, IgG) can be as low as 1 pg and for protein in aqueous solution (anti-human IgG, anti-IgG) 2.75 ng/ml. The detection dynamic ranges are from 2 pg to 1 ng for IgG immobilized on slides and from 5.5 ng/ ml to 1.4 Ag/ml for anti-IgG in solution. This detection sensitivity is comparable to fluorescence methods that require expensive equipment.
2. Materials and methods The schematic principle of the colorimetric gold – silver method is illustrated in Fig. 1. The human IgGs in the spots are covalently bound on the activated glass slides, the biotin-conjugated anti-IgGs are then bound to the human IgG. Excess unbound biotinconjugated anti-IgGs are removed by the washing with doubly distilled water (ddH2O). Upon the addition of the gold-conjugated streptavidin to the slides, they will bind to the biotin-conjugated anti-IgGs which are already bound to the IgGs, resulting in formation of a sandwich-like complex and introduction of the gold particles to the spot where the human IgGs are immobilized. In the silver enhancement, the Ag+ in the silver enhancement solution is reduced by the promotion of the gold particles, a large amount of silver is precipitated on the spots where the gold particles are introduced, giving the significant colorimetric images (Alexandre et al., 2001). The effect of the silver enhancement is very powerful, so that the black spots formed on slides can be detected with a commercial CCD camera (Olympus C-4000Z digital camera) mounted on the microscope. Furthermore, the
Fig. 1. Schematic illustration of gold-conjugated antibody recognition and signal amplification with silver enhancement.
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
silver precipitation on each spot is proportional to the amount of the gold nanoparticles deposited on the spot. The latter quantity is proportional to the quantity of the human IgG immobilized in the spot. Therefore, the gray level of each spot in the microarray can be used to analyze the quantity of the assayed component. Parenthetically, the gold particle can be introduced into the microarray spots in various manners depending on the type of the microarray and/or the experimental design. For instance, in Fig. 1, the gold particle can be also introduced by the labeling the gold to anti-IgG, or by the gold-conjugated protein G which binds to the anti-IgG, etc. 2.1. Materials Human IgG, rabbit myosin, gold-conjugated streptavidin, gold-conjugated anti-human IgG, 3glycidoxypropyltrimethoxysilane (GOPTS), silver enhancement solutions A and B, microscope glass slides, and cover glass were all purchased from Sigma (St. Louis, MO). Bovine serum albumin V(BSA) was purchased from Roche (Mannheim, Germany). Biotin-conjugated goat anti-human IgG (biotin-conjugated anti-IgG) and Cy5k-conjugated goat anti-human IgG were from Jackson ImmunoResearch Laboratories (West Grove, PA). Human serum samples were from Ruijin Hospital (Shanghai, China). All other chemicals were of analytical grade. 2.2. Preparation of self-assembled monolayer (SAM) The SAM formed on microscope glass slides was prepared in the manner previously described with some modifications (Bhatia et al., 1989; Eisen and Brown, 1999). Briefly, microscope glass slides were soaked in 10% NaOH and then 0.1 M HCl for 2 h, respectively. After thorough rinsing with doubly distilled water (ddH2O) and boiling for 1 h, the glass slides were set in 1% GOPTS (in 95% ethanol) at 37 jC for 6 h. The slides were then dried and incubated at 135 jC for another 1 h, and finally stored at 4 jC for further use. 2.3. Preparation of protein microarray Protein samples of human IgG or antigens were dissolved in phosphate-buffered saline (PBS, 137 mM
159
NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.4 mM KH2PO4, pH 7.4) with 40% glycerol. These solutions were transferred to 384-well plate respectively. The printing of protein microarrays on GOPTS-activated glass slides was performed with PixSys 5500 spotting robot (Cartesian Technology, Irvine, CA) mounted with ArrayIt SMP3 micro spotting pin from TeleChem (Sunnyvale, CA). On the resulting protein microarrays each spot has about 1 nl sample solution. The diameter of the spots and the space between spots were 100 and 300 Am, respectively. Immediately before use, the protein microarray slides were blocked with 1% BSA/PBS for 1 h at room temperature. After rinsing with PBS the slides were dried by centrifugation at 200 g for 5 min. 2.4. The detection of demonstrational ToRCH antigen microarrays To examine the application of the gold-silver detection in practice, the demonstrational ToRCH antigen microarray developed in our lab was used. The microarray is used for detecting the antibody of Cytomaglovirus (CMV), Toxoplasma gondii (Tox), Rubella Virus (RBV), Herpse Simplex Virus (HSV), and Coxsackie Virus B6 (CVB6). The ToRCH microarrays were first incubated with 10 Al of 100-fold dilution of patient’s serum and then washed with PBS to remove all unbound molecules. After that, in the colorimetric detection, the slides were probed with the gold-conjugated goat anti-human IgG and then silverenhanced, in which the silver-enhancement time was 15 min. The detection was carried out on the CCD camera as described above. By contrast in the fluorescent detection method, the slides was probed with Cyk 5-conjugated goat anti-human IgG, then the fluorescent detection was carried out on the ScanArray 5000 from PerkinElmer (Boston, MA).
3. Results and discussions The image of a microarray spotted with a series of different human IgG concentrations shown in Fig. 2 indicates that the spots printed with concentration as low as 1 Ag/ml IgG can be detected (line 12). The actual amount of the IgG contained in these spots is equal to or less than about 1 pg as the volume of the
160
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
Fig. 2. The gray level of the spot at various concentrations of human IgG immobilized on the microarray. Inset: the gold – silver detection image of different concentration human IgG immobilized on microarray. The concentrations of spotted human IgG from columns 1 to 12: 2.0 mg/ml, 1.0 mg/ml, 0.5 mg/ml, 0.25 mg/ml, 0.13 mg/ml, 62.5 Ag/ml, 31.25 Ag/ml, 15.625 Ag/ml, 7.8 Ag/ml, 3.9 Ag/ml, 1.95 Ag/ml and 0.97 Ag/ml. The concentration of biotin-conjugated anti-IgG: 11 Ag/ml; the silver-enhancement time: 15 min. The data in the plot were obtained from the image in the inset by averaging the mean gray levels of seven replicate spots.
sample printed to each spot is about 1 nl. Therefore, the lowest detectable amount for the IgG on the slides is about 1 pg, which is similar with fluorescence detection. The plot in Fig. 2 shows that the gray level of the spots is linear to the logarithm of the human IgG amounts immobilized on the slides in a range from 2 pg to 1 ng (corresponding IgG concentrations of sample printed on the slides are from 2 Ag/ml to 1 mg/ml). This indicates that the colorimetric detection method has a large linear detection dynamic range (about three orders of magnitude) for the components immobilized on the slides, which is very important for its practical application in quantitative analysis. The dynamic detection range also provides important information to determine the optimum concentration of protein immobilized on slides. Fig. 3A shows a set of images of model microarrays used to determine the concentration of biotinconjugated anti-IgG with the colorimetric gold –silver method. In the model microarrays, all the spots were printed onto the activated glass slides with human IgG solution of 250 Ag/ml. As shown in Fig. 3A the biotinconjugated anti-IgG as low as 2.75 ng/ml (17.2 pM)
can be detected, such a sensitivity is close to that in the fluorescent detection method (1 ng/ml) (Haab et al., 2001). Fig. 3B shows that the gray level of the spots is linearly dependent on the logarithm of concentration of biotin-conjugated anti-IgG from 5.5 ng/ ml to 1.4 Ag/ml, suggesting that the linear detection dynamic range spans 300 fold. Such high detection sensitivity and linear detection dynamic range are more valuable for the application of the gold – silver detection method because protein microarrays are mainly used to probe the components in aqueous solution. Fig. 3B reveals when the biotin-conjugated anti-IgG is higher than 1.4 Ag/ml, the gray level levels off, forming a plateau. This plateau might be caused by the limited capacity of IgG immobilized on the spots to capture biotin-conjugated anti-IgG. When the biotin-conjugated anti-IgG concentration in the sample is in excess of the capture capacity, more biotinconjugated anti-IgG can no longer be captured, so the gray level will not increase. According to the results obtained in Fig. 2, it is reasonable to expect that the detection sensitivity and the dynamic range of the gold –silver detection method can be improved with
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
161
Fig. 3. Detection of the biotin-conjugated anti-IgG at different concentration. (A) The detection images of biotin-conjugated anti-IgG using the gold – silver detection method. All the spots in the microarrays were printed with 250 Ag/ml human IgG. The biotin-conjugated anti-IgG concentration to be detected was shown in the image. The silver-enhancement time was 25 min. (B) Dynamic detection range of the gold – silver detection method. The data were obtained from the image shown in (A) by the same method mentioned in Fig. 2.
Fig. 4. The effects of various silver enhancement time on the image gray level. The concentration of human IgG spotted on the slides was 250 Ag/ml. The concentration of biotin-conjugated anti-IgG: 110 ng/ml (E), 1.1 Ag/ml (n) and 11 Ag/ml ( ).The silver enhancement time varies from 5 to 30 min.
.
162
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
an increase in the concentration of the immobilized IgG. Our results also indicate that the detection sensitivity is critically related to the silver enhancement reaction time, ten (see Fig. 4). For instance, almost no spots can be detected for 110 ng/ml biotin-conjugated anti-IgG when the ten is less than 10 min. Meanwhile, if the ten is extended to 15, 20 or 25 min, the gray levels of spots become obvious. Therefore, the detection sensitivity can be increased by the extension of ten. However, when the ten is more than about 25 min, the silver enhancement reaction seems to be saturated, a plateau in gray level change can be found. The time needed for the appearance of the plateau is termed as the saturated silver enhancement time (tsat). The facts related to tsat are complicated. At least, we thought, the concentration of probed component in the sample is one of them. As seen in Fig. 4, the lower the concentration of biotin-conjugated antiIgG, the longer the tsat is. This interpretation is consistent with the autocatalytic mechanism of the gold – silver enhancement reaction. It is certain that the ten used in the gold –silver method must be shorter than tsat, otherwise the gray level will not reflect the quantitative information. On the other hand, we also noticed that the longer ten lead to an increased background, in agreement with that reported by (Alexandre et al., 2001). To obtain the highest detectable sensitivity and a good signal/noise ratio the enhancement time must be optimized. In our experiment, it is found that the optimum ten ranges from 15 to 25 min. The specific IgG concentration yielded by a typical natural immune response is in a range from 10 ng/ml (Anthony et al., 1994) to over 3 Ag/ml (Granoff et al., 1986). This range is within the detection limit and linear detection dynamic range of our method described above. Fig. 5 shows the result of a demonstrational ToRCH antigen microarrays detected by the gold – silver and fluorescent method. The ToRCH antigen microarrays were developed in our lab (unpublished data). In the microarrays, CMV, Tox, RVB, HSV, and CVB6 antigens were printed onto the activated glass slides as introduced in materials and methods. After reaction with diluted serums of the patient, the microarrays were analyzed by the gold – silver and fluorescence detection methods. As seen in Fig. 5, the gold – silver detection method presented in a pattern similar to the fluores-
Fig. 5. The detection of the demonstrational ToRCH microarrays using the gold – silver and fluorescent method. The silver enhancement time was 15 min. The fluorescent image was scanned with 80% PMT Gain and 80% Laser Power.
cence method, suggesting that the colorimetric method have potential application in diagnostic antigen microarrays. Taken together, our results indicate the gold – silver colorimetric method can provide a highly accurate and linear dynamic range in practical applications to detect protein components present in both aqueous solution and immobilized on slides. In addition, a conventional digital CCD camera as a detector and well-managed conjugation of colloidal gold to biomolecules make the method easy to perform with lower cost. Therefore the method is of potential application in protein microarray.
Acknowledgements We thank Dr. Michelle Digman for making useful suggestion on the manuscript. This work was funded by the Innovation Project of Chinese Academy of Sciences and partially supported by the National 973 Project of China (No. 2002CB713802).
R.-Q. Liang et al. / Journal of Immunological Methods 285 (2004) 157–163
References Alexandre, I., Hamels, S., Dufour, S., Collet, J., Zammatteo, N., Longueville, F.D., Gala, J.-L., Remacle, J., 2001. Colorimetric silver detection of DNA microarrays. Anal. Biochem. 295, 1 – 8. Anthony, B.J., Concepcion, I.E., Concepcion, N.F., Vadheim, C.M., Tiwari, J., 1994. Relation between maternal age and serum concentration of IgG antibody to type III group B streptococci. J. Infect. Dis. 170, 717 – 720. Bhatia, S.K., Shriver-Lake, L.C., Prior, K.J., Georger, J.H., Calvert, J.M., Bredehorst, R., Ligler, F.S., 1989. Use of thiol-terminal silanes and heterobifunctional crosslinkers for immobilization of antibodies on silica surfaces. Anal. Biochem. 178, 408 – 413. Eisen, M.B., Brown, P.O., 1999. DNA arrays for analysis of gene expression. Methods Enzymol. 303, 179 – 205. Granoff, M., Shackelford, P.G., Suarez, B.K., Nahm, M.H., Cates, K.L., Murphy, T.V., Karasic, R., Osterholm, M.T., Pandey, J.P., Daum, R.S., 1986. Hemophilus influenzae type B disease in children vaccinated with type B polysaccharide vaccine. New Engl. J. Med. 315, 1584 – 1590. Haab, B.B., Dunham, M.J., Brown, P.O., 2001. Protein microarrays for highly parallel detection and quantitation of specific proteins and antibodies in complex solutions. Genome Biol. 2 research 0004.1 – 0004.13. Holgate, C.S., Jackson, P.I., Cowen, P.N., Bird, C.C., 1983. Immunogold – silver staining: new method of immunostaining with enhanced sensitivity. J. Histochem. Cytochem. 31, 938 – 944. Joos, T.O., Schrenk, M., Hopfl, P., Kroger, K., Chowdhury, U., Stoll, D., Schorner, D., Durr, M., Herick, K., Rupp, S., Sohn, K., Hammerle, H., 2000. A microarray enzyme-linked immunosorbent assay for autoimmune diagnostics. Electrophoresis 21, 2641 – 2650. Lackie, P.M., 1996. Immunogold silver staining for light microscopy. Histochem. Cell Biol. 106, 9 – 17.
163
MacBeath, G., Schreiber, S.L., 2000. Printing proteins as microarrays for high-throughput function determination. Science 289, 1760 – 1763. Mezzasoma, L., Bacarese-Hamilton, T., Di Cristina, M., Rossi, R., Bistoni, F., Crisanti, A., 2002. Antigen microarrays for serodiagnosis of infectious diseases. Clin. Chem. 48, 121 – 130. Moeremans, M., Daneels, G., Dijck, D.V., Langanger, G., Mey, J.D., 1984. Sensitive visualization of antigen – antibody reactions in dot and blot immune overlay assays with immunogold and immunogold/silver staining. J. Immunol. Methods 74, 353 – 360. Robinson, W.H., DiGennaro, C., Hueber, W., Haab, B.B., Kamachi, M., Dean, E.J., Fournel, S., Fong, D., Genovese, M.C., de Vegvar, H.E., Skriner, K., Hirschberg, D.L., Morris, R.I., Muller, S., Pruijn, G.J., van Venrooij, W.J., Smolen, J.S., Brown, P.O., Steinman, L., Utz, P.J., 2002. Autoantigen microarrays for multiplex characterization of autoantibody responses. Nat. Med. 8, 295 – 301. Tam, S.W., Wiese, R., Lee, S., Gilmore, J., Kumble, K.D., 2002. Simultaneous analysis of eight human Th1/Th2 cytokines using microarrays. J. Immunol. Methods 261, 157 – 165. Taton, T.A., Mirkin, C.A., Letsinger, R.L., 2000. Scanometric DNA array detection with nanoparticle probes. Science 289, 1757 – 1760. Vidal, M., 2001. A biological atlas of functional maps. Cell 104, 333 – 339. Weinberger, S.R., Morris, T.S., Pawlak, M., 2000. Recent trens in potein biochip technology. Pharmacogenomics 1, 395 – 416. Zhu, H., Bilgin, M., Bangham, R., Hall, D., Casamayor, A., Bertone, P., Lan, N., Jansen, R., Bidlingmaier, S., Houfek, T., Mitchell, T., Miller, P., Dean, R.A., Gerstein, M., Snyder, M., 2001. Global analysis of protein activities using proteome chips. Science 293, 2101 – 2105.