DNA-Catalytically Active Gold Nanoparticle ... - ACS Publications

Dec 14, 2016 - Colorimetric Multidimensional Sensor Array for Protein ... array sensing elements and the color-change time (CCT) of the solution as si...
5 downloads 0 Views 712KB Size
Subscriber access provided by EPFL | Scientific Information and Libraries

Letter

DNA-Catalytically Active Gold Nanoparticle Conjugates-Based Colorimetric Multidimensional Sensor Array for Proteins Discrimination Xiangcong Wei, Zhengbo Chen, Lulu Tan, Tianhong Lou, and Yan Zhao Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b04878 • Publication Date (Web): 14 Dec 2016 Downloaded from http://pubs.acs.org on December 15, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

DNA-Catalytically Active Gold Nanoparticle Conjugates-Based Colorimetric Multidimensional Sensor Array for Proteins Discrimination Xiangcong Wei, Zhengbo Chen,* Lulu Tan, Tianhong Lou, and Yan Zhao Department of Chemistry, Capital Normal University, Beijing, 100048, China * Corresponding author. Tel.: 010-68903047 E-mail: [email protected] ABSTRACT: A series of single-strand oligonucleotides functionalized atalytically active gold nanoparticle (AuNPs) as nonspecific receptors have been designed to build a protein sensing array. We take advantage of the correlation between the catalytic activity and the exposed surface area of AuNPs, i.e., DNA-proteins interactions mask the surface area of AuNPs, leading to poor catalytic performance of AuNPs. As the number of DNA-bound proteins increases, the surfaces of AuNPs become more masked, thus, the time of 4- nitrophenol/NaBH4 reaction for color change (yellow→colorless) of the solution increases. Taking advantage of three nonspecific SH-labeled DNA sequences (A15, C15, and T15) as array sensing elements, and the color-change time (CCT) of the solution as signal readout, colorimetric response patterns can be obtained on the array and identified via linear discriminant analysis (LDA). 11 proteins have been completely distinguished with 100% accuracy with the naked eye at the 30 nM level. Remarkably, two similar proteins (Bovine Serum Albumin and Human Serum Albumin) and two different proteins (Bovine Serum Albumin and Concanavalin) at the same concentration, and the mixtures of the two proteins with different molar ratios have been discriminated with 100%. The practicability of this sensor array is further validated by high accuracy (100%) identification of 11 proteins in human serum samples.

increase of the amount of sensor elements.10 Colorimetric sensor arrays, ie., the assemblies of cross-reactive colorimetric sensor elements, have proven to be a powerful analytical approach for the discrimination of a wide variety of analytes including explosives,11 biomolecules, toxic gases,12 beverages and foods,13 and pathogenic bacteria and fungi.14 Recently, synthetic enzyme mimics, such as iron oxide,15 cerium oxide,16 and gold nanoparticles (AuNPs) 17 have been developed. Due to higher reproducibility and stability than natural enzymes at various pH values and temperatures,18 these synthetic enzyme mimics have been utilized for highly sensitive detection of biological targets by amplifying the signal response. The combination of AuNPs with DNA19 may give a possible solution to the development of array-based protein sensors. Compared to the specific aptamers, random designed DNA could afford unlimited sensor elements for array sensing with different degree, as even a short DNA sequence (e.g., 15 bases) possesses up to billions of combinations.20 DNA strands are combined to AuNPs with multidimensional information, it is helpful to prepare a multidimensional sensor array with unlimited sensing elements. The DNA-AuNP conjugates as a nonspecific sensor array receptor has been reported recently, e.g., Zhang et al. reported aptamer-based plasmonic sensor array for discrimination of proteins with the naked eye.21 Liu

INTRODUCTION The detection of proteins in complex sample matrices plays an important role in allergy testing,1 disease diagnosis,2 and clinical treatment.3 In recent years, various sensing strategies for the detection of proteins based on the detection of color or fluorescence signal change are widely employed.4-7 These colorimetric/fluorescent protein detection systems, however, have the drawback that simultaneous multianalyte assays (particularly in biological molecules) could not be achieved directly owing to the limitation of binding interactions between the single lock and the corresponding key.8,9 Therefore, in consideration of the increasing demands on medical diagnosis, it is a need of the hour to develop a sensor array instead of a onesensor-per-analyte strategy for the detection and discrimination of various proteins. Compared with individual sensor elements, array-based sensors have two major advantages for detecting a series of analytes with common structures or properties as follows: (1) multiple measurements depending on several sensors and subsequent analysis of data or data patterns employing the suitable statistical models dramatically improves the accuracy and reproducibility of the target detection, and (2) sensor array improves the resolving power, which in turn increases with the

1

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 5

hydrogen bonding, electrostatic force, Van der Waals forces, and base accumulation affected the accessibility of reaction substrates (4-nitrophenol and NaBH4) to the catalytic AuNP surface, causing concomitant alteration of catalytic activity of AuNPs, which further led to change of the CCT of the solution. A distinct CCT response pattern had been obtained for each target protein by using such a multidimensional sensor.

et al. presented a dual-channel sensor (fluorometric and colorimetric) with dye-labeled DNA-AuNP conjugates as receptors.22 Our group presented a dual-channel sensor array for discrimination of heavy metal ions based on fluorescence and colorimetric response of the DNA-AuNP conjugates.23 While these DNA-AuNP conjugates are very promising for accurate detection and discrimination of targets, for colorimetric sensors, long-standing challenges remain associated with AuNP aggregation, which severely limit their application in real samples.24 First, the aggregation of cross-linked AuNPs is nondirective, resulting in the formation of large, wide range of aggregates. Owing to the increased size of AuNPs and decreased surface repelling force, these aggregates is unstable in solution, and thus, over time, the solution color diminishes, preventing accurate target determination; Second, relatively low sensitivity is obtained by a large-aggregate sensor; Finally, the dynamic range of detection based on AuNP aggregation is only 1-2 orders of magnitude, which is relatively narrower than that of fluorescence-based methods.25 To address these issues, we construct a single-channel (colorimetric) multidimensional sensor that renders amplified responses and the color-change time (CCT) of the solution as signal readout. Notably, the array’s most attractive advantages lie in that (1) three random designed, nonspecific DNA strands (A15, C15, and T15) instead of specific DNA were chosen to serve as array sensing elements; (2) catalytically active AuNP was used to amplify the visual signal, in the whole process of protein identification, AuNPs cannot form any aggregates, thus the sensor array offers long-term stable responses; (3) Although the CCT of the solution as a signal readout have been previously utilized for the sole protein detection,4,6 we have for the first time developed CCT sensor array for proteins discrimination. Differential interaction forces of analyte proteins with these DNA strands controlled the access of reaction substrates (4-nitrophenol here) to the surface of AuNPs, causing concomitant alteration in the AuNP surface coverage of DNA-bound proteins. Discern patterns were then generated with CCT readouts provided by the AuNP-catalyzed reduction of yellow 4-nitrophenol to colorless 4-aminophenol. The variation of the CCT (CCT/CCT0) obtained at different concentrations of proteins were analyzed by linear discriminant analysis (LDA) and principal component analysis (PCA). We found that with the increase of sensor elements added to the sensor array, the discrimination ability to proteins was significantly enhanced. The single-channel aptasensor with three sensor elements could not only discriminate 11 proteins as low as 30 nM with high accuracy (100%), but also could simultaneously discriminate the mixture of two proteins at the same concentration and their mixtures with different molar ratios.

Scheme 1. (A,B) Diagrams of Colorimetric Sensor Array and Detection Principle of Proteins Based on DNA-AuNP Conjugates as Sensing Elements and (C) CCT of the Solutions as Signal Readout.

Colorimetric Sensor Array Responses to Proteins. To verify the discrimination ability of the present colorimetric sensor array, 11 proteins (Table S1) with extensive usage were chosen as analytes for the test. To obtain the accurate limit of detection (LOD) of proteins, we generated the CCT response patterns with proteins at the concentration of 10, 30, and 50 nM, respectively. A15, C15, and 15T have different affinity to AuNPs, the as-prepared DNA-AuNPs conjugates have already had different gold exposure. It could be fair that we used CCT/CCT0 to eliminate the effect of the original difference. A representative figure of the array response against 30 nM of 11 proteins indicates that the CCT change profiles of the array were unique fingerprints for each protein, as shown in Figure 1A, the CCT signals were different for a given DNA-AuNP conjugates in the presence of different proteins, and different binding between the same protein with different oligonucleotides had different effects on CCTs of the solutions, confirming the cross-reactive property of the CCT signals from this sensor array. To expose the fingerprints of 11 proteins more explicitly, LDA was utilized to quantitatively distinguish the various proteins according to their linear combination of features (Figure 1B-D), the CCT patterns of the training matrix (1 channel×3 sensing elements×11 proteins×5 replicates) were subjected to LDA. 92.2% and 7.6% of the variation for proteins at 10 nM (Figure 1B), 96.7% and 3.0% of the variation for proteins at 30 nM (Figure 1C), 67.0% and 25.7% of the variation for proteins at 50 nM were obtained. In Figure

RESULTS AND DISCUSSION Sensing Mechanism. As a proof-of-concept, we designed a single-channel multidimensional sensor with three nonspecific SH-labeled DNA sequences (A15, C15, and T15) as sensing elements for discrimination of 11 proteins, taking advantage of DNA-AuNP conjugates as the recognition elements and the CCT of the solution as the signal readout. As shown in Scheme 1, three DNA strands (A15, T15, and C15) at one end with a thiol group were absorbed on the surface of AuNPs through Au-S bond, respectively. Differential interactions of analyte proteins with these oligonucleotides through

2

ACS Paragon Plus Environment

Page 3 of 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry 25% BSA + 75% Con, 50% BSA + 50% Con, 75% BSA + 25% Con, total protein concentration at 30 nM) were recorded. As shown in Figure 2B,D, it was found that 5 groups of protein samples were located distinctly in 5 independent clusters in the canonical score plot, and the cross-validation accuracy of identification was found to be 100%. Compared with their pure form, the mixtures of BSA and HSA, BSA and Con had different CCT responses. Training matrix of the response patterns against BSA and HSA, BSA and Con at the same concentration, respectively, as well as the mixtures of these two with various molar ratios was shown in Table S6-7. Furthermore, the HCA dendrogram (Figure S3) and the fingerprint (Figure 2A,C) of each protein and the mixtures of two proteins clearly showed five clusters, demonstrating a 100% correct classification (in quintuplicate trials). This clear discrimination means that the sensing technique could accurately identify proteins and that the proposed strategy can give a different CCT response fingerprint for individual target proteins and protein mixtures.

1B, among 11 proteins (all at 10 nM), only 4 proteins (Fib, Con, HRP, and Pap) can be successfully identified without any overlap, whereas the CCT/CCT0 responses of CytC and Try, TRF, BSA, and HSA, Hem and Pep show significant overlap in the 2D canonical score plot. In contrast, the 11 proteins (all at 30 and 50 nM, respectively) were successfully distinguished with identification accuracies of 100%, and basically have no overlap in the canonical score plot (Figure 1C,D). Of note, Figure 1C shows that two groups (BSA and HRP) are tangent, indicating that our sensor array could identify proteins at a concentration (i.e, limit of detection (LOD)) as low as 30 nM. To compare with other methods, Table S2 summarizes the sensitivity of protein sensor arrays recently reported. Obviously, our sensor array reveals a superior sensitivity. Moreover, the CCT profiles of the array (for example, 50 nM proteins) could be visualized by naked eyes (Figure 1E). The training matrix of the response patterns against proteins using this sensing assay at the three concentrations (10, 30, and 50 nM) was shown in Table S3-S5. For the accurate determination of the CCT, for example, for C15-AuNP conjugates, the absorbance change (Figure S1) of the solution (blank, Pep, Fib, and Con, each at 30 nM) vs reaction time at 400 nm was recorded. Without the addition of protein, a strong absorption peak at 400 nm was observed. As the reaction proceeds, however, no other absorbance peak at 400 nm was observed, indicating the complete reaction of 4-nitrophenol and NaBH4. In addition, to confirm that the colorimetric readout was indeed due to reduced 4-nitrophenol instead of aggregation of AuNPs, the stability of AuNPs was assessed using TEM, size distribution, and UV-vis of AuNPs (Figure S2). They confirm the stability of AuNPs before and after protein binding, suggesting the effective anti-aggregation of AuNPs.

Figure 2. (A,C) Fingerprints, (B,D) canonical score plot for the CCT response patterns as obtained from discrimination of the mixtures of the two proteins (BSA and HAS, BSA and Con) at different molar ratios (total concentration at 30 nM). Applicability. The performance of this sensing system was further evaluated for the discrimination of proteins in real human serum. For this purpose, the sensing array system was applied to human serum samples spiked with relevant 11 proteins. In virtue of the challenge for the identification of proteins brought by the very complex matrixes in human serum, higher concentrations of proteins were spiked in the serum samples (final protein concentration: each at 300 nM) to obtain the reliable discrimination ability of the sensor. However, in the present case, as shown in Figure 3, each of the relevant proteins involved in the serum sample matrix generated a distinct response, and a 100% identification accuracy was obtained for all the 11 proteins (Figure 3A,B). Training matrix of the response patterns against proteins (300 nM) in the human serum using the sensor array was shown in Table S8. Thus, the results indicated that this multidimensional sensor based on DNA-AuNP conjugates could be potentially applied for the discrimination of proteins in real biological samples.

Figure 1. (A) The fingerprints of 11 proteins (each at 30 nM) using the patterns of the corresponding values of CCT/CCT0 obtained from the colorimetric responses of three DNA-AuNP conjugates. Two-dimensional canonical score plot for discrimination of proteins at the same concentration: (B) 10 nM, (C) 30 nM, and (D) 50 nM. (E) The visual color changes of the detection solutions containing proteins with the same concentration (50 nM) based on three DNA-AuNP conjugates as the catalytic reaction progresses.

Discrimination of Protein Mixtures. After successful detection of the 11 proteins, owing to different molar ratios of different proteins, discrimination of protein mixtures was far more challenging than pure proteins. To testify the ability of the sensing strategy, the response of the sensor for BSA, HSA, and Con at the same concentration (30 nM) and the mixtures of two proteins with different molar ratios (25% BSA + 75% HSA, 50% BSA + 50% HSA, 75% BSA + 25% HSA, and

3

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 5

REFERENCES (1) Kingsmore, S. F.; Patel, D. D. Curr. Opin. Biotechnol. 2003, 14, 74-81.

Figure 3. (A) Fingerprints of 11 selected proteins (each at 300 nM) in human serum based on the patterns of the corresponding values of CCT/CCT0 obtained from the colorimetric responses of three DNA-AuNP conjugates (B) Canonical score plots for the discrimination of human serum spiked with different proteins analyzed by LDA.

(2) Larsson, A.; Johansson, M. E.; Wangefjord, S.; Gaber, A.; Nodin, B.; Kucharzewska, P.; Welinder, C.; Belting, M.; Eberhard, J.; Johnsson, A.; Uhlen, M.; Jirstrom, K. Br. J. Cancer 2011, 105, 666-672. (3) Okuno, J.; Maehashi, K.; Kerman, K.; Takamura, Y.; Matsumoto, K.; Tamiya, E. Biosens. Bioelectron. 2007, 22, 2377-2381. (4) Kim, B. H.; Yoon, I. S.; Lee, J. S. Anal. Chem. 2013, 85, 1054210548. (5) Ranallo, S.; Rossetti, M.; Plaxco, K. W.; Vallée-Bélisle, A.; Ricci, F. Angew. Chem. Int. Edit. 2015, 54, 13214-13218. (6) Chen, Z. B.; Tan, L. L.; Hu, L. Y.; Zhang, Y. M.; Wang, S. X.; Lv, F. Y. ACS Appl. Mater. Interfaces 2016, 8, 102-108.

CONCLUSIONS

(7) Huang, Y.; Liu, X. Q.; Huang, H. K.; Qin, J.; Zhang, L. L.; Zhao, S. L.; Chen, Z. F.; Liang, H. Anal. Chem. 2015, 87, 8107-8114. (8) Jr, P. A.; Liu, Y.; Palacios, M. A.; Minami, T.; Wang, Z.; Nishiyabu, R. Chem. -Eur. J. 2013, 19, 8497-8506.

In summary, we have proposed an extensible colorimetric sensor array for the discrimination of 11 proteins, which based on the amplification of the visual signal induced by catalytically active AuNPs and different binding between targets and DNA.The catalytic activity of AuNP surfaces was systematically controlled by masking the surfaces of AuNPs with DNAtarget binding. The CCTs of the solutions again proteins were completely different using only three DNA-AuNP elements. Using this sensor array, 11 proteins with a variety of molecular weights and isoelectric points were successfully discriminated at low concentration of 30 nM, lower concentration than previous colorimetric array-based protein sensors (50 nM-350 µM).21,22,26 Furthermore, the array could efficiently discriminate among individual proteins and their mixtures at various molar ratios. Finally, the 11 proteins spiked in human serum were well distinguished with a 100% discrimination accuracy and favorable reproducibility, which further validated the practical application of this sensor array. Taken together, the colorimetric sensing array can serve as a readily accessible, highly discriminative, and adaptive tool for high-precision discrimination of proteins, cells, tissues, and heavy metal ions.

(9) Chou, S. S.; De, M.; Luo, J.; Rotello, V. M.; Huang, J.; Dravid, V. P. J. Am. Chem. Soc. 2012, 134, 16725-16733. (10) Galpothdeniya, W. I. S.; Regmi, B. P.; McCarter, K. S.; de Rooy, S. L.; Siraj, N.; Warner, I. M. Anal. Chem. 2015, 87, 4464-4471 (11) Li, Z.; Bassett, W. P.; Askim, J. R.; Suslick, K. S. Chem. Commun. 2015, 51, 15312-15315. (12) Lin, H.; Jang, M.; Suslick, K. S. J. Am. Chem. Soc. 2011, 133, 16786-16789. (13) Zhang, C.; Suslick, K. S. J. Agr. Food Chem. 2007, 55, 237-242. (14) Minami, T.; Esipenko, N. A.; Zhang, B.; Isaacs, L.; Anzenbacher, P. Jr. Chem. Commun. 2014, 50, 61-63. (15) Li, X. N.; Wen, F.; Creran, B.; Jeong, Y. D.; Zhang, X. R.; Rotello, V. M. Small 2012, 8, 3589-3592. (16) Ornatska, M.; Sharpe, E.; Andreescu, D.; Andreescu, S. Anal. Chem. 2011, 83, 4273-4280. (17) Chen, Z. B.; Zhang, C. M.; Gao, Q. G.; Wang, G.; Tan, L. L.; Liao, Q. Anal. Chem. 2015, 87, 10963-10968. (18) Shivhare, A.; Ambrose, S. J.; Zhang, H. X.; Purves, R. W.; Scott, R. W. J. Chem. Commun. 2013, 49, 276-278. (19) Lukman, S.; Aung, K. M.; Liu, J.; Liu, B.; Su, X. ACS Appl Mater.Inter. 2014, 5, 12725-12734. (20) Clelland, C. T.; Risca, V.; Bancroft, C. Nature 1999, 399, 533534. (21) Lu, Y. X.; Liu, Y. Y.; Zhang, S. G.; Wang, S.; Zhang, S. C.; Zhang, X. R. Anal. Chem. 2013, 85, 6571-6574. (22) Sun, W. B.; Lu, Y. X.; Mao, J. P.; Chang, N.; Yang, J. E.; Liu, Y. Y. Anal. Chem. 2015, 87, 3354-3359.

AUTHOR INFORMATION Corresponding Author * Phone: +86-010-68903047. E-mail: [email protected]

Notes The authors declare no competing financial interest.

ASSOCIATED CONTENT

(23) Tan, L. L.; Chen, Z. B.; Zhao, Y.; Wei, X. C.; Li, Y. H.; Zhang, C.; Wei, X. L.; Hu, X. C. Biosen. Bioelectron. 2016, 85, 414-421.

Supporting Information Experimental section, TEM image, particle size distribution of AuNPs, spectral changes of the solution, physical properties of protein analytes, and the training matrix of the response patterns against proteins. These data are available free of charge via the Internet at http://pubs.acs.org.

(24) Guo, L. H.; Xu, Y.; Ferhan, A. R.; Chen, G. N.; Kim, D. H. J. Am. Chem. Soc. 2013, 135, 12338-12345. (25) Vallee-Belisle, A.; Ricci, F.; Plaxco, K. W. J. Am. Chem. Soc. 2012, 134, 2876-2879. (26) Wu, Z. S.; Lu, H. X.; Liu, X. P.; Hu, R.; Zhou, H.; Shen, G. L.; Yu, R. Q. Anal. Chem. 2010, 82, 3890-3898.

ACKNOWLEDGMENT All authors gratefully acknowledge the financial support of the Natural Science Foundation of China (No. 21371123) and Scientific Research Project of Beijing Educational Committee (KM201410028006).

4

ACS Paragon Plus Environment

Page 5 of 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

for TOC only

5

ACS Paragon Plus Environment