Letter Cite This: Anal. Chem. 2017, 89, 10639-10643
pubs.acs.org/ac
Colorimetric Sensor Array Based on Gold Nanoparticles with Diverse Surface Charges for Microorganisms Identification Bingyu Li, Xizhe Li, Yanhua Dong, Bing Wang, Dongyang Li, Youmin Shi,* and Yayan Wu* Key Laboratory of Biomedical Information Engineering of Ministry of Education, School of Life Science and Technology, Xi’an Jiaotong University, 710049, Xi’an, PR China S Supporting Information *
ABSTRACT: We report a simple and novel colorimetric sensor array for rapid identification of microorganisms. In this study, four gold nanoparticles (AuNPs) with diverse surface charges were used as sensing elements. The interactions between AuNPs and microorganisms led to obvious color shifts, which could be observed by the naked eye. A total of 15 microorganisms had their own response patterns and were differentiated by linear discriminant analysis (LDA) successfully. Moreover, microorganism mixtures could also be well discerned. The method is simple, fast (within 5 s), effective, and visual, showing the potential applications in pathogen diagnosis and environmental monitoring.
M
phenyleneethynylene) conjugates as sensing elements for the identification of 12 bacteria.15 Fan Chunhai et al. have made use of nanographene oxide-different DNA labeled with 6carboxyfluorescein (6-FAM) as sensing elements for proteins and bacteria identification.16 Although the fluorescence sensor arrays have high sensitivity and strong discriminatory power, complicated synthesis of various fluorescent materials as sensing elements as well as photobleaching are still problematic. Compared with fluorescence sensor arrays, colorimetric sensor arrays have the advantages of simpleness, fast response, and visualization detection. Although they have been applied to differentiate a large number of chemicals and bioanalytes,18−22 there are few reports about the colorimetric sensor arrays for the identification of bacteria. Gold nanoparticles (AuNPs) are easy to bind to biosystems such as proteins, bacteria, and cells. A few studies have been reported on the interactions of proteins and AuNPs with different surface properties (size, shape, surface charge, and coating material).23,24 AuNPs have been regarded as an ideal sensitive material to build colorimetric sensors for biosystems sensing. In a previous study, we used unmodified gold and silver nanoparticles with different sizes as sensing elements to fabricate a colorimetric sensor array for proteins and bacteria identification.25 AuNPs with different sizes have different curvatures and differential binding interactions with protein, resulting in different color shifts due to the aggregation of AuNPs.25 Microorganisms, no matter bacteria or fungi, have different surface electronic properties and can be suitable for
icroorganisms such as bacteria and fungi are ubiquitous in the world, whether they are pathogenetic or not. The identification of microorganisms is very important in many fields including clinical diagnostics, biology, and food safety. Several methods are available to detect bacteria. Plate culturing is a standard method with high accuracy. However, it is timeconsuming and laborious. Polymerase chain reaction (PCR) is considered to be another standard method for bacteria detection, which suffers from complexity and high cost. Enzyme-linked immunosorbent assay (ELISA) is one of the most popular immunological assays. It may not require a longer time compared with plate culturing and PCR, while high falsepositive results limit its application. Mass spectrometry1−3 are also used in bacteria detection, but it requires a large instrument. Furthermore, all the methods mentioned above are unable to in situ test. Chemical sensors based on electrochemical,4,5 fluorescence,6,7 and colorimetric detection8,9 have the advantages of being fast, simple, sensitive, and able to online monitor. For sensing many bacteria simultaneously or bacteria mixtures, chemical sensors are challenging because a kind of sensor can be employed to detect a specific bacteria. Therefore, fast, simple, and sensitive methods for simultaneous identification of microorganisms will be highly demanded in clinical as well as public health areas. In recent years, chemical sensor arrays, especially fluorescence and colorimetric sensor arrays, have received more and more attention due to their ability to differentiate a variety of bacteria with high classification accuracy.10−17 Jiang Xingyu et al. have synthesized five fluorescent probes as sensing elements to fabricate a fluorescence sensor array for the identification of eight bacteria.14 Rotello et al. have developed a fluorescence sensor array by using gold nanoparticle-poly(para© 2017 American Chemical Society
Received: July 4, 2017 Accepted: September 21, 2017 Published: September 21, 2017 10639
DOI: 10.1021/acs.analchem.7b02594 Anal. Chem. 2017, 89, 10639−10643
Letter
Analytical Chemistry interaction with charged AuNPs.26 We hypothesize that microorganisms have differential bindings to surface charged AuNPs, allowing their distinction using a sensor array. Therefore, we have made an attempt to synthesize a group of AuNPs with diverse charged surfaces as sensing elements to develop a colorimetric sensor array for microorganisms sensing. In the present study, a colorimetric sensor array for rapid and effective identification of microorganisms has been developed using four AuNPs as sensing elements, which are AuNPs coated with mercaptopropionic acid (AuNPs@MPA), mercaptosuccinic acid (AuNPs@MSA), cysteamine (AuNPs@Cys), and cetyltrimethylammonium bromide (AuNPs@CTAB), as illustrated in Figure 1. In the presence of microorganisms, the fast
characterized by TEM, UV−vis spectrometry, and zeta potentials measurements, as shown in Figure S1 (Supporting Information). Zeta potentials of four AuNPs are −12.37 mV, −17.43 mV, 31.27 mV, and 4.37 mV, respectively, showing that four AuNPs have different surface electronic properties from each other. AuNPs@MPA and AuNPs@MSA both have negatively charged property. AuNPs@Cys has a strong positively charged property and AuNPs@CTAB has a weak positively charged property. As we know, most microorganisms have negatively charged surfaces. Their interactions with diverse surface charge will result in distinct color profiles due to the aggregation of AuNPs. Therefore, a colorimetric sensor array was fabricated for microorganisms identification. The discrimination power of the sensor array was investigated for the identification of 15 microorganisms including 12 bacteria and 3 fungi, which were Staphylococcus aureus (S. aureus), Staphylococcus epidermidis (S. epidermidis), Listeria monocytogenes (L. monocytogenes), Bacillus aceticus (B. aceticus), Pseudomonas aeruginosa (P. aeruginosa), Escherichia coli (E. coli), Bacillus subtilis (B. subtilis), Salmonella paratyphi (S. paratyphi), Enterobacter sakazakii (E. sakazakii), Shigella f lexneri (Sh. f lexneri), Vibrio parahemolyticus (V. parahemolyticus), Clostridium putrefaciens (C. putrefaciens), Candida albicans (C. albicans), Aspergillus f lavus (A. flavus), and Penicillium. Regarding the detailed operation procedures, please see the Supporting Information. When 100 μL of microorganism suspensions (OD600 = 0.05) were mixed with 100 μL of AuNPs, very fast color changes could be observed within 5 s, indicating that the sensor array had fast response for microorganisms. The interactions of microorganisms and AuNPs caused aggregation of four sensing elements at different degrees, leading to unique color shift patterns (Figure 2). From Figure 2, it can be seen that almost all microorganisms give rise to color changes of positively charged AuNPs due to the electrostatic interaction with negatively charged microorganisms, indicating that surface charge plays an important role in microorganisms identification. However, C. putrefaciens only made a very slight color change on AuNPs@CTAB. Moreover, a small number of microorganisms could lead to color changes of negatively charged AuNPs@MPA and AuNPs@MSA, while most microorganisms could not. In fact, the interactions of AuNPs and microorganisms are not only electrostatic interaction, but also other interactions such as hydrophobic interaction. Even if the microorganisms had the similar surface zeta potentials, they gave different color changes of positively charged AuNPs, indicating that microorganisms surface or structural differences may also play a role in the interactions of AuNPs and microorganisms. For example, E. coli and S. epidermidis had similar surface zeta potentials, while they led to different color changes of AuNPs@Cys and AuNPs@CTAB because of the different levels of interactions caused by structural differences. Similarity, as a consequence of multifactorial interactions, the negatively charged AuNPs can also be attached to microorganisms surfaces in different degrees, leading to various color shifts. Therefore, 15 microorganisms have their own color shifts patterns which can differentiate from each other. The color patterns were also transformed into RGB files (Figure S2, Supporting Information), which could be much clearer to observe. The color change could also be recorded by UV−vis absorption spectra. The absorption spectra of AuNPs with 15 microorganisms were scanned using a micro-plate reader (Figure S3, Supporting Information). The colorimetric
Figure 1. Schematic diagram of the colorimetric sensor array based on four AuNPs with different surface charges. The interactions of AuNPs and microorganisms result in color shifts. In the diagram, each row from A to D represents AuNPs@MPA, AuNPs@MSA, AuNPs@Cys, and AuNPs@CTAB, respectively. Column 1 represents blank control, while other columns represent different microorganism samples.
aggregation of gold nanoparticles results in clearly distinguished color shifts within 5 s due to the interactions of microorganisms and AuNPs, which can be observed by the naked eye. Each microorganism has its unique color shift pattern, which can differentiate from each other.
■
RESULTS AND DISCUSSION In order to design a colorimetric sensor array for microorganisms identification based on the interactions between microorganisms and AuNPs, we synthesized four AuNPs with diverse charged surfaces, including AuNPs@MPA, AuNPs@ MSA, AuNPs@Cys, and AuNPs@CTAB, which were described in the Supporting Information. These four AuNPs were
Figure 2. Photograph of the color change upon addition of 15 microorganisms (OD600 = 0.05) based on array-based sensing. 10640
DOI: 10.1021/acs.analchem.7b02594 Anal. Chem. 2017, 89, 10639−10643
Letter
Analytical Chemistry
Figure 3. Array-based sensing of 15 microorganisms. (a) Response (k/k0) patterns in the presence of microorganisms (OD600 = 0.05) (responses are an average of six measurements and the error bars are the standard deviation). (b) Canonical score plot for the response patterns as obtained from LDA.
response was obtained by the value of k/k0, where k and k0 represent the absorbance ratio between 620 and 526 nm (k = OD620 nm/OD526 nm) in the presence and absence of microorganism, respectively. The value of k/k0 was used to assess the degree of AuNPs aggregation. As shown in Figure 3a, the k/k0 values of S. aureus, S. paratyphi, Sh. f lexneri, C. albicans on AuNPs@CTAB are high, so color changes are very obvious (from red to blue) which is consistent with the result of Figure 2. While the k/k0 values of S. epidermidis and C. putrefaciens on
AuNPs@CTAB are low, color changes are very slight. Therefore, different microorganisms have different absorption responses on the different sensing elements, resulting in their own response (k/k0) patterns (Figure 3a). The response (k/k0) pattern was analyzed by linear discriminant analysis (LDA). This analysis reduced the size of the training matrix (4 sensing elements × 15 microorganisms × 6 replicates) and transformed them into canonical factors. First three canonical factors are visualized as a well-clustered three-dimensional (3D) plot, as 10641
DOI: 10.1021/acs.analchem.7b02594 Anal. Chem. 2017, 89, 10639−10643
Analytical Chemistry
Letter
■
CONCLUSIONS In summary, we found that diverse surface charged AuNPs could give a rapid and differential response to microorganisms with different surface electronic properties. Based on this, a sensor array composing of four AuNPs was developed to differentiate 15 microorganisms. Each microorganism has obtained a distinct color shift pattern as well as response pattern. The method is very fast, simple, label-free, and visual. The colorimetric sensor array has a great potential in quickly detecting a large number microorganisms involved in clinical diagnostics and environmental monitoring. Also, it is easy to make a paper-based colorimetric sensor array for online identification of microorganisms.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.7b02594. Chemicals and materials, AuNPs synthesis and characterization, fabrication of sensor array and microorganisms sensing, RGB image, UV−vis spectra, Jackknifed classification matrix, and quantitative analysis (PDF)
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. Fax: +86 29 82663941. *E-mail:
[email protected]. Fax: +86 29 82663941. ORCID
Yayan Wu: 0000-0002-5127-2597 Notes
Figure 4. Array-based sensing of four sets of microorganism mixtures. (a) Response (k/k0) patterns in the presence of the microorganism mixtures (responses are an average of six measurements and the error bars are the standard deviation). Mixtures 1−4 represent (L. monocytogenes + C. albicans), (L. monocytogenes + E. coli), (L. monocytogenes + V. parahemolyticus), and (L. monocytogenes + S. epidermidis) suspensions, respectively. (b) Canonical score plot for the response patterns as obtained from LDA.
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS This project was supported by the National Natural Science Foundation of China (Grant Nos. 21175103 and 21125525). REFERENCES
(1) Kailasa, S. K.; Wu, H. F. Analyst 2012, 137, 4490−4496. (2) Kailasa, S. K.; Wu, H. F. Talanta 2013, 114, 283−290. (3) Kailasa, S. K.; Wu, H. F. TrAC, Trends Anal. Chem. 2015, 65, 54− 72. (4) Liu, X.; Marrakchi, M.; Xu, D.; Dong, H.; Andreescu, S. Biosens. Bioelectron. 2016, 80, 9−16. (5) Webster, M. S.; Cooper, J. S.; Chow, E.; Hubble, L. J.; SosaPintos, A.; Wieczorek, L.; Raguse, B. Sens. Actuators, B 2015, 220, 895−902. (6) Akram, A. R.; Avlonitis, N.; Lilienkampf, A.; Perez-Lopez, A. M.; McDonald, N.; Chankeshwara, S. V.; Scholefield, E.; Haslett, C.; Bradley, M.; Dhaliwal, K. Chem. Sci. 2015, 6, 6971−6979. (7) Feng, G.; Yuan, Y.; Fang, H.; Zhang, R.; Xing, B.; Zhang, G.; Zhang, D.; Liu, B. Chem. Commun. 2015, 51, 12490−12493. (8) Miranda, O. R.; Li, X.; Garcia-Gonzalez, L.; Zhu, Z.-J.; Yan, B.; Bunz, U. H. F.; Rotello, V. M. J. Am. Chem. Soc. 2011, 133, 9650− 9653. (9) Sung, Y. J.; Suk, H.-J.; Sung, H. Y.; Li, T.; Poo, H.; Kim, M.-G. Biosens. Bioelectron. 2013, 43, 432−439. (10) Chen, S.; Wei, L.; Chen, X.-W.; Wang, J.-H. Anal. Chem. 2015, 87, 10902−10909. (11) Tao, Y.; Ran, X.; Ren, J.; Qu, X. Small 2014, 10, 3667−3671. (12) Li, X.; Kong, H.; Mout, R.; Saha, K.; Moyano, D. F.; Robinson, S. M.; Rana, S.; Zhang, X.; Riley, M. A.; Rotello, V. M. ACS Nano 2014, 8, 12014−12019.
shown in Figure 3b. A total of 15 microorganisms were successfully differentiated in LDA with a classification accuracy of 100% according to Jackknifed Classification Matrix (Table S1, Supporting Information). Furthermore, the quantitative detection power of this sensing platform was also investigated by detecting L. monocytogenes water suspensions of different concentrations against AuNPs@MPA (Figure S4, Supporting Information). The change of solution color was approximately positively correlated with the concentration of microorganisms. To assess the potential ability of the proposed sensor array to differentiate the complex samples, the array was employed to differentiate the microorganisms mixtures. We randomly explored four sets of mixtures: L. monocytogenes suspension solution was mixed with same volume of (1) C. albicans, (2) E. coli, (3) V. parahemolyticus, (4) S. epidermidis suspensions, respectively. When the four sets of artificial mixture samples were analyzed by the proposed sensor array, their response (k/ k0) patterns are shown in Figure 4a and the differential results are presented in Figure 4b. Each mixture sample had its own response patterns and could basically differentiate from each other. 10642
DOI: 10.1021/acs.analchem.7b02594 Anal. Chem. 2017, 89, 10639−10643
Letter
Analytical Chemistry (13) Bueno, L.; Cottell, A.; Reddy, S. M.; Paixão, T. R. L. C. RSC Adv. 2015, 5, 97962−97965. (14) Chen, W.; Li, Q.; Zheng, W.; Hu, F.; Zhang, G.; Wang, Z.; Zhang, D.; Jiang, X. Angew. Chem., Int. Ed. 2014, 53, 13734−13739. (15) Phillips, R. L.; Miranda, O. R.; You, C.-C.; Rotello, V. M.; Bunz, U. H. F. Angew. Chem., Int. Ed. 2008, 47, 2590−2594. (16) Pei, H.; Li, J.; Lv, M.; Wang, J.; Gao, J.; Lu, J.; Li, Y.; Huang, Q.; Hu, J.; Fan, C. J. Am. Chem. Soc. 2012, 134, 13843−13849. (17) Chu, Y. W.; Wang, B. Y.; Engebretson, D. A.; Carey, J. R. Analyst 2013, 138, 5879−5885. (18) Lin, H.; Suslick, K. S. J. Am. Chem. Soc. 2010, 132, 15519− 15521. (19) Lu, Y.; Liu, Y.; Zhang, S.; Wang, S.; Zhang, S.; Zhang, X. Anal. Chem. 2013, 85, 6571−6574. (20) Ghasemi, F.; Hormozi-Nezhad, M. R.; Mahmoudi, M. Anal. Chim. Acta 2015, 882, 58−67. (21) Sun, W.; Lu, Y.; Mao, J.; Chang, N.; Yang, J.; Liu, Y. Anal. Chem. 2015, 87, 3354−3359. (22) Fahimi-Kashani, N.; Hormozi-Nezhad, M. R. Anal. Chem. 2016, 88, 8099−8106. (23) Daniel, M.-C.; Astruc, D. Chem. Rev. 2004, 104, 293−346. (24) Aubin-Tam, M. E.; Hamad-Schifferli, K. Biomed. Mater. 2008, 3, 034001. (25) Li, D.; Dong, Y.; Li, B.; Wu, Y.; Wang, K.; Zhang, S. Analyst 2015, 140, 7672−7677. (26) Berry, V.; Gole, A.; Kundu, S.; Murphy, C. J.; Saraf, R. F. J. Am. Chem. Soc. 2005, 127, 17600−17601.
10643
DOI: 10.1021/acs.analchem.7b02594 Anal. Chem. 2017, 89, 10639−10643
