Protein Sensing and Cell Discrimination Using a Sensor Array Based

Feb 16, 2011 - Protein Sensing and Cell Discrimination Using a Sensor Array Based on ... Phone: (+86) 10-6277-6888. ... For a more comprehensive list ...
0 downloads 0 Views 3MB Size
Download PDF
LETTER pubs.acs.org/ac

Protein Sensing and Cell Discrimination Using a Sensor Array Based on Nanomaterial-Assisted Chemiluminescence Hao Kong,† Da Liu,†,‡ Sichun Zhang,† and Xingrong Zhang*,† †

Department of Chemistry, Key Laboratory for Atomic and Molecular Nanosciences of the Education Ministry, Tsinghua University, Beijing 100084, P. R. China ‡ Beijing National Laboratory for Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing, 100190, P. R. China

bS Supporting Information ABSTRACT:

Cross-reactive sensor arrays, known as “chemical noses”, offer an alternative to time-consuming analytical methods. Here, we report a sensor array based on nanomaterial-assisted chemiluminescence (CL) for protein sensing and cell discrimination. We have found that the CL efficiencies are improved to varied degrees for a given protein or cell line on catalytic nanomaterials. Distinct CL response patterns as “fingerprints” can be obtained on the array and then identified through linear discriminant analysis (LDA). The sensing of 12 kinds of proteins at three concentrations, as well as 12 types of human cell lines among normal, cancerous, and metastatic, has been performed. Compared with most fluorescent or colorimetric approaches which rely on the strong interaction between analytes and sensing elements, our array offers the advantage of both sensitivity and reversibility.

T

he early diagnostics of diseases through olfactory systems has attracted increasing interest.1 In spite of several successes achieved by animals like dogs,2 the field is open when it comes to the mimic of diagnostic procedures in the artificial systems. Cross-reactive sensor arrays, known as “chemical/ artificial noses”, have the potential of “smelling diseases” conveniently.3 Strategically, nonspecific or semiselective sensing elements, which can generate a differential response pattern for each analyte, play the most important role in sensor arrays.4 Various approaches which depend essentially on strong molecular interactions have been employed in the construction of sensing elements,5 including colorimetric pigments,6 fluorescence quenching assays,7 and indicator displacement assays (IDAs).8 Although these methods are highly sensitive and differentiable, the strong interaction between analytes and sensing elements generally feature irreversible sensing processes. It is a fundamental flaw against reversible and reusable sensor arrays like mammalian noses. Therefore, the creation of sensor arrays both sensitive and reversible is still a great challenge. The phenomenon of chemiluminescence (CL) observed when organics or biological substances are thermally oxidized is named as thermochemiluminescence (TCL).9 Despite of several successes of traditional TCL assays in bioanalysis and medical diagnostics, the sensitivity is quite low and one sample can offer only a single TCL response, making it difficult to design r 2011 American Chemical Society

a sensor array. In this study, we found that TCL of proteins and cell lines can be performed on the surface of catalytic nanomaterials. Also these nanosized catalysts have diverse catalytic activity for a specific analyte, generating unique TCL response “fingerprints” for discrimination. Moreover, catalysis-amplified reactions significantly increase TCL intensity, and the reversibility is improved as well since analytes can be wiped out through catalytic degradation at a higher temperature when the detection is finished. Hence, we created a sensor array (Figure 1) based on nanomaterial-assisted TCL, which holds the promise of achieving both satisfactory sensitivity and good reversibility. In the initial study, six catalytic nanomaterials, Pt/Ba/Al, MgO, ZrO2, γ-Al2O3, MgCO3, and SrCO3, were chosen to serve as sensing elements. For example, the catalytic assistance of appropriate amounts of MgO can significantly amplify the TCL response of BSA (Figure 2). Notably, the array’s most attractive advantage is the good reversibility and long-term stability because nanomaterials are solid catalysts and not consumed while analytes are wiped out during a temperature increase after detection. As shown in Figure 3 and Figure S1 in the Supporting Received: October 13, 2010 Accepted: February 10, 2011 Published: February 16, 2011 1867

dx.doi.org/10.1021/ac200076c | Anal. Chem. 2011, 83, 1867–1870

Analytical Chemistry

LETTER

Figure 1. The schematic of the sensing process. The protein solutions or cell suspensions (10 μL) are directly added on ceramic heaters sintered with catalytic nanomaterials. Then the proteins or cells are trapped onto the surface of nanomaterials after the volatilization of water. When the temperature is raised to 205 °C, analytes are thermally oxidized with the generation of distinct TCL. After TCL decays, remains of analytes are wiped out from the surface of nanomaterials at 500 °C.

Figure 2. The enhancement of TCL responses of BSA (10 μg/mL) with the assistance of different amounts (0-0.5 mg) of the nanosized MgO layer.

Figure 3. The stable TCL signals in the actual experiment when two cell suspensions (∼100 000 cells/mL) are switched several times. Six peaks of each sample represent the TCL responses on six array elements in the order of Pt/Ba/Al, MgO, ZrO2, γ-Al2O3, MgCO3, and SrCO3, respectively. A is A549, and B is MCF-7.

Information, the sensor array offers reversible and long-term stable responses. For the differentiability test, 12 common proteins (Table S1 in the Supporting Information) including glycoproteins and metalloproteins were chosen as sensing targets. As illustrated in

Figure 4. Array-based sensing of protein solutions at 10 μg/mL. (a) TCL peak intensity patterns of the 12 proteins on the nanomaterials array as an average of 5 parallel measurements. (b) Canonical score plots for the first three factors of TCL response patterns analyzed by LDA.

Figure 4a, direct addition of 10 μL of aqueous solutions of proteins (10 μg/mL) on the surface of nanomaterials resulted in a variety of unique TCL response patterns due to their thermal catalytic oxidations and diverse light emission, which depends on the characteristics of both proteins and nanomaterials. Such an outcome is reasonable because we have shown that nanomaterials possess tremendous diversity for CL of organic compounds generated during catalytic oxidations.10 By the same token, libraries of nanomaterials may generate unique TCL patterns for proteins, which have more distinct features. Then the TCL peak intensity patterns were subjected to linear discriminant analysis (LDA). These patterns were transformed to canonical scores which were visualized as a well-clustered three-dimensional plot (Figure 4b) with a classification accuracy of 100%. Moreover, we successfully discriminated protein solutions at three concentrations (5, 10, and 20 μg/mL) utilizing the training matrix generated from TCL patterns at only one concentration (10 μg/mL). Our system was sensitive enough to differentiate proteins at 5 μg/mL (70-400 nM for the 12 proteins with varying molecular weights), more sensitive than most previously proposed methods (>1 μM). 11 Additionally, we achieved 1868

dx.doi.org/10.1021/ac200076c |Anal. Chem. 2011, 83, 1867–1870

Analytical Chemistry

LETTER

To further examine the differentiability of the array, we tested three types of human hepatocellular carcinoma cell lines (MHCC97-L, LM3, and HepG2) and one normal liver cell line (L02). As shown in Figure S9 in the Supporting Information, the array presented the differential TCL responses of four types of human liver cell lines and LDA classified the cell lines into 4 distinct clusters with 100% accuracy. Finally, we tested 30 samples of cell suspensions controlled at the same cell density of 100 000 cells/mL but with unknown identity, which were randomly chosen from the 12 types of cell lines mentioned above. All of them were assigned to the accurate groups generated using the training matrix (6 nanomaterials  12 cell lines  5 replicates) during LDA (Tables S8 and S9 in the Supporting Information). In conclusion, we have developed a sensitive and reusable sensor array for the discrimination of proteins and cell types on the basis of distinct TCL responses patterns generated by the catalytic nanomaterials array. With the benefit from the reversible response and long-term stability as well as the simple sensing elements and instrumentation, this “nose” takes an important step for real-world “disease smelling”, though there is still a long way to go from our present study. We are currently expanding the libraries of catalytic nanomaterials to further improve the differentiability and sensitivity and exploiting new data analysis strategies to apply this methodology to complex matrixes in real-world medical diagnostics. Figure 5. Array-based sensing of eight human cell lines. The density of each cell suspension is controlled at 100 000 cells/mL. (a) TCL response patterns of the eight cells on the nanomaterials array as an average of five parallel measurements. (b) Canonical score plots for the first three factors of TCL response patterns analyzed by LDA.

successful discrimination of similar mixtures of BSA, HSA, and ovalbumin. (Supporting Information). To study the relationship between TCL signal and MW of proteins, we analyzed the TCL patterns of two groups of proteins on the sensor array. The MW of group 1 is much higher than group 2, and proteins from the same group have close MW values. As shown in Figure S6 in the Supporting Information, proteins with similar MW do not cluster in the LDA plot, indicating that other factors of proteins may play a key role in their TCL diversities. On the other hand, two metal-containing proteins, Hem (64.5 kDa) and CytC (12.3 kDa), have similar TCL patterns, resulting in the cluster of LDA plots. A plausible explanation is that iron contained in proteins might be oxidized to Fe2O3, which may have both the abilities to accelerate the decomposition of the intermediate hydroperoxide compounds and to reduce the produced radical species into nonradical products.12 In addition, we tested BSA samples of different purities purchased from four manufacturers (Figure S7 in the Supporting Information). The failure of distinguishing these four samples suggested that impurities may not dominate the discrimination process because of the very low contents. Then, we studied whether this array can discriminate different types of cell lines including normal, cancerous, and metastatic ones. The results showed that diverse catalytic oxidations of cells on nanomaterials generated distinct TCL intensity patterns (Figure S8 in the Supporting Information), which could distinguish cell types through pattern recognition. The eight different cells we chose for preliminary examination displayed excellent separation when the TCL patterns (Figure 5a) were subjected to LDA and the canonical scores were plotted in a well-clustered graph (Figure 5b), validating our array’s ability of cell discrimination.

’ ASSOCIATED CONTENT

bS

Supporting Information. Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Phone: (þ86) 10-62776888. Fax: (þ86) 10-6278-2485.

’ ACKNOWLEDGMENT The authors gratefully thank the financial support from the National Natural Science Foundation of China (Grant No. 21027013), the Innovation Method Fund of China (Grant No. 2008IM040600), the National High Technology Research and Development Program of China (Grant No. 2009AA03Z321), and the Tsinghua University Initiative Scientific Research Program. ’ REFERENCES (1) Whittle, C. L.; Fakharzadeh, S.; Eades, J.; Preti, G. Ann. N.Y. Acad. Sci. 2007, 1098, 252. (2) Horvath, G.; Jarverud, G. A.; Jarverud, S.; Horvath, I. Integr. Cancer Ther. 2008, 7, 76. (3) (a) Peng, G.; Tisch, U.; Adams, O.; Hakim, M.; Shehada, N.; Broza, Y. Y.; Billan, S.; Abdah-Bortnyak, R.; Kuten, A.; Haick, H. Nat. Nanotechnol. 2009, 4, 669. (b) Bajaj, A.; Miranda, O. R.; Kim, I. B.; Phillips, R. L.; Jerry, D. J.; Bunz, U. H. F.; Rotello, V. M. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 10912. (4) Albert, K. J.; Lewis, N. S.; Schauer, C. L.; Sotzing, G. A.; Stitzel, S. E.; Vaid, T. P.; Walt, D. R. Chem. Rev. 2000, 100, 2595. (5) Wright, A. T.; Anslyn, E. V. Chem. Soc. Rev. 2006, 35, 14. (6) Rakow, N. A.; Suslick, K. S. Nature 2000, 406, 710. (7) (a) Dickinson, T. A.; White, J.; Kauer, J. S.; Walt, D. R. Nature 1996, 382, 697. (b) Baldini, L.; Wilson, A. J.; Hong, J.; Hamilton, A. D. 1869

dx.doi.org/10.1021/ac200076c |Anal. Chem. 2011, 83, 1867–1870

Analytical Chemistry

LETTER

J. Am. Chem. Soc. 2004, 126, 5656. (c) Miranda, O. R.; You, C. C.; Phillips, R.; Kim, I. B.; Ghosh, P. S.; Bunz, U. H. F.; Rotello, V. M. J. Am. Chem. Soc. 2007, 129, 9856. (8) (a) Buryak, A.; Severin, K. J. Am. Chem. Soc. 2005, 127, 3700. (b) Nguyen, B. T.; Anslyn, E. V. Coord. Chem. Rev. 2006, 250, 3118. (c) Schiller, A.; Wessling, R. A.; Singaram, B. Angew. Chem., Int. Ed. 2007, 46, 6457. (d) You, C. C.; Miranda, O. R.; Gider, B.; Ghosh, P. S.; Kim, I. B.; Erdogan, B.; Krovi, S. A.; Bunz, U. H. F.; Rotello, V. M. Nat. Nanotechnol. 2007, 2, 318. (e) Adams, M. M.; Anslyn, E. V. J. Am. Chem. Soc. 2009, 131, 17068. (9) (a) Bellamy, W. D.; Tweet, A. G. Nature 1963, 197, 482. (b) Schumann, A.; Bauer, A.; Hermes, M.; Gilbert, M.; Hengstler, J. G.; Wilhelm, C. Arch. Toxicol. 2009, 83, 709. (c) Millington, K. R.; Ishii, H.; Maurdev, G. Amino Acids 2010, 38, 1395. (10) (a) Na, N.; Zhang, S. C.; Wang, S. A.; Zhang, X. R. J. Am. Chem. Soc. 2006, 128, 14420. (b) Kong, H.; Zhang, S. C.; Na, N.; Liu, D.; Zhang, X. R. Analyst 2009, 134, 2441. (c) Wu, Y. Y.; Na, N.; Zhang, S. C.; Wang, X.; Liu, D.; Zhang, X. R. Anal. Chem. 2009, 81, 961. (11) Collins, B. E.; Anslyn, E. V. Chem.—Eur. J. 2007, 13, 4700. (12) Nakatani, H.; Shibata, H.; Miyazaki, K.; Yonezawa, T.; Takeda, H.; Azuma, Y.; Watanabe, S. J. Appl. Polym. Sci. 2010, 115, 167.

1870

dx.doi.org/10.1021/ac200076c |Anal. Chem. 2011, 83, 1867–1870