Chinese Chemical Letters  2017, Vol. 28 Issue (9): 1881-1884   PDF    
Immune fluorescence test strips based on quantum dots for rapid and quantitative detection of carcino-embryonic antigen
Yudong Wua,1, Weipan Penga,1, Qian Zhaoa, Jiafang Piaoa, Bo Zhanga, Xiaoli Wua, Hanjie Wanga, Zhihong Shib, Xiaoqun Gonga, Jin Changa    
a School of Life Sciences, School of Materials Science and Engineering, Tianjin University and Tianjin Engineering Center of Micro-Nano Biomaterials and Detection-Treatment Technology(Tianjin), Tianjin 300072, China;
b Department of Neurology, and Tianjin Neurosurgery Institute, Tianjin Huanhu Hospital, Tianjin 300060, China
Abstract: At present, many researchers focused on the point-of-care testing (POCT), a method of disease markers detection without large-scale instruments and specialized persons. However, most POCT diagnostic methods were suffered from poor detection sensitivity or inefficiency in quantitative detection. Herein, we developed a newly QD-immune fluorescence test strips (QD-IFTS) based on quantum dots (QDs) as the fluorescence nanocarrier to prepare the immune fluorescence probes in the classical immunochromatography detection system for sensing carcino-embryonic antigen (CEA), a kind of glycoprotein produced by intestinal tissue and a broad spectrum of tumor marker for cancer diagnosis. And we designed a homemade strips fluorescence reader for detection of fluorescence intensity of QDs on the QD-IFTS. Under the optimized reaction conditions, chromatographic time of the newly QD-IFTS was only 25 min, sample volume of the newly QD-IFTS was only 40 μL and the LOD of the newly QD-IFTS was 0.72 ng/mL. In addition, the efficiency and robustness of the newly QD-IFTS were confirmed by successfully application in 300 clinical serum samples, and the results revealed great potential in clinical POCT of other biomarkers.
Key words: Point-of-care testing     Immunochromatography assay     Quantum dots     Carcinoembryonic antigen     Quantitative detection    

Recently, quantitative detection of tumor markers without the help of large-scale instruments and specialized persons was one of the research hotspots in the field of disease diagnosis, known as POCT [1-4]. The main advantages of POCT were easy to operate, cost-effective and time-saving, which were contributed to the broad spectrum screening of tumor markers with an on-site detection method. At present, commonly used POCT diagnostic formats mainly included immunochromatography assay (ICA) [5, 6], dry-chemical assay (DCA) [7], immunosensor assay (ISA) [8, 9], immune biochip assay (IBCA) [10], immune microfluidic-chip assay (IMCA) [11], and among them, ICA was the most commonly used.

ICA was a kind of rapid diagnostic technique, whose immune reaction was triggered on the antibodies-coated nitrocellulose membrane based on capillary siphon to generate the detection signal. However, it was necessary to develop a new kind of ICA for rapid and quantitative detection of tumor markers. QDs, as a new kind of semiconductor nanomaterials, showed unique optical performance due to its quantum confinement effect, which was widely used in the field of fluorescent labeling [12-15]. The marriage of ICA and QDs should be a good choice for diseases POCT diagnostic due to the complementary advantages, known as QDsICA [16-18].

Herein, we reported a newly QD-IFTS for tumor markers detection to realize cancers POCT diagnosis. CEA, a kind of glycoprotein produced by intestinal tissue, which has been recognized as a broad spectrum of tumor marker for cancer diagnosis, was used to be the model analyte in this study [19-21]. On the basis of optimizing the conditions of nitrocellulose membrane, pretreatment buffer, chromatographic time and sample volume, we further evaluated the performance of the newly QD-IFTS, including sensitivity, specificity and stability, and achieved excellent results. In addition, the newly QD-IFTS was also applied to patient serums to confirm the robustness and efficiency of the immunoassay, showing great potential with an on-site detection method.

Zinc oxide (powder, 99.999%), sulfur (powder, 99.98%), selenium (powder, 99.99%), cadmium oxide (99.98%), oleic acid (technical grade, 90%), 1-octadecene (technical grade, 90%), octadecylamine (technical grade, 90%), trioctylphosphine oxide (technical grade, 90%), trioctylphosphine (97%) and poly-maleic anhydride-alt-1-octadecene in this study were purchased from Sigma-Aldrich. Mouse monoclonal CEA antibody and goat antimouse antibody were supplied by Bioscience Diagnostic Technology Co. (Tianjin, China). EDC was purchased from Aladdin (Shanghai, China). Fetal bovine serum and bovine serum albumin were obtained from Dingguo Biotechnology Co. (Beijing, China). The serum samples were supplied by General Hospital of Tianjin Medical University and deionized water (18.2 MV cm, Milli-Q grade) was chosen for this study.

Core/shell CdSe/CdS/CdxZn1-xS/ZnS QDs were prepared according to our previous work with a successive ion-layer adsorption and reaction (SILAR) method [22]. And the hydrophilic QDs were prepared by an ultrasonic emulsification method with amphiphilic polymer poly-maleic anhydride-alt-1-octadecene (PMAO) [23, 24]. The immune fluorescence probes were prepared by activating the carboxyl group of the surface of the quantum dots to couple with the amino groups of detection antibodies (Ab2) [25, 26]. EDC (5μmol) solution in PBS buffer (200μL, pH 7.4) was added to the above hydrophilic QDs (5 nmol), and the solution was stirred at room temperature for 15 min to activate the carboxyl groups on the surface of QDs. Subsequently, Ab2 (50 nmol) were added into the above mixture at room temperature for 2 h, the obtained solution was purified by centrifugation to obtain the immune fluorescence probes QDs-Ab2, the probes were stored at 4 ℃ in a block buffer (5 mL, pH 7.4, containing 0.01 mol/L PBS and 1% BSA).

Assemble of QD-IFTS was according to the previous work [16, 22]. As shown in Scleme 1, there were four core parts in our QDIFTS: absorbent pad, nitrocellulose membrane, conjugate pad and sample pad. Firstly, capture antibody (Ab1, 2 mg/mL) and goat antimouse antibody (Ab3, 2 mg/mL) were immobilized onto the surface of the nitrocellulose membrane with a BioJet Dispenser (XYZ-3050) to form T line and C line, respectively. Then, the conjugate pad was immersed in a pretreatment buffer (pH 7.4, containing 0.01 mol/L PBS, 3% BSA, 5% sugar, 5% PVA10000 and 2% Tween-20) for 5 min, dried in an oven at 37 ℃, and the treated pad was immersed in another buffer (pH 7.4, containing 0.01 mol/L PBS, 0.1μmol/L immune fluorescence probes QDs-Ab2, 5% BSA, 5% sugar, 1% PEG4000 and 0.1% Tween-20) for 5 min, dried again in an oven at 37 ℃. Subsequently, the sample pad was immersed in a pretreatment buffer (pH 7.4, containing 0.01 mol/L PBS and 0.5% Tween-20), and dried in an oven at 37 ℃. Lastly, the absorbent pad, the treated nitrocellulose membrane, the treated conjugate pad and the treated sample pad were assembled on a backing card in turn with an overlapping of 2 mm, the obtained backing card was cut into 4 mm strips.

Construction of strips fluorescence reader was according to our previous work [16, 22]. There were four core parts in our strips fluorescence reader: a tablet personal computer, a 405 nm laser diode, an optic fiber spectrometer and a stepper motor controller. The QD-IFTS was driven by the stepper motor controller at a constant speed, QDs on the QD-IFTS were excited by the 405 nm laser diode to emit fluorescence, and fluorescence signal was converted into electric signal with the optic fiber spectrometer, the electric signal was collected by the tablet personal computer.

Subsequently, the prepared immune fluorescence probes QDsAb2 were characterized. It is generally known that QDs have unique optical properties, which were widely used in the field of disease diagnosis, especially in fluorescence labeling. In this work, QDs were prepared by a SILAR method to obtain the alloy core/shell structure CdSe/CdS/CdxZn1-xS/ZnS QDs. The TEM photographs (Fig. S1 in Supporting information) and the XRD spectrum (Fig. S2a in Supporting information) showed that the as prepared alloy core/ shell structure QDs revealed an excellent lattice structure with uniform size distribution, indicated the successful preparation of core-shell QDs. As shown in Fig. 1a, the hydrophilic QDs were prepared by an ultrasonic emulsification method with amphiphilic polymer PMAO to wrap up hydrophobic core/shell QDs. TEM images (Fig. 1b) showed that the hydrophilic PMAO@QDs with narrow size distribution had been successfully prepared, which was of great significance for the performance of our QD-IFTS. The hydrophilic PMAO@QDs were water soluble with 18.2 nm of hydrodynamic particle size (Fig. 1c) and -39.2 mV of surface Zeta potential (Fig. 1d) by the dynamic light scattering (DLS), which provided an excellent stability in PBS buffer with slight change of the hydrodynamic particle size in 90 days (Fig. S2b in Supporting information). The immune fluorescence probes QDs-Ab2 were prepared by activating the carboxyl group of the surface of the QDs to couple with the amino groups of Ab2. The fluorescence intensity of QD solution and three QD probes solution, as shown in Fig. S3a (Supporting information), suggested the excellent fluorescence stability of the QDs. And the immobilization of Ab2 was confirmed by an immunity test, and the results showed that the prepared immune fluorescence probes possessed a strong ability to resist nonspecific adsorption and an excellent immune activity (Fig. S3b in Supporting information).

Download:
Fig. 1. Characterization of the as-prepared hydrophilic quantum dots PMAO@QDs. (a) Schematic diagram of QDs modified by amphiphilic polymer PMAO; (b) TEM photographs of PMAO@QDs; (c) Dynamic light scattering particle size of PMAO@QDs; (d) Zeta potential of PMAO@QDs.

We optimized the QD-IFTS, as the performance of QD-IFTS was influenced by many factors, such as types of the pads, pretreatment buffer, chromatographic time, sample volume, and so on [27]. Ideally, the QD-IFTS should have higher signal value and lower noise value at the same time. In general, the release of the immune fluorescence probes QDs-Ab2 was mainly affected by the conjugate pad and sample pad, which was contributed to enhance the detection signal. S/N value was calculated to evaluate the performance of the conjugate pad and sample pad (S/N = (FT1/ FC1)/(FT0/FC0), where FT1, FC1 were the fluorescence intensity of T line and C line with 200 ng/mL CEA, and FT0, FC0 were the fluorescence intensity of T line and C line with FBS), and S/N value reached the highest with VL98 conjugate pad and sample pad, as shown in Fig. S4 (Supporting information).

Research showed that the release of immune fluorescence probes and the ability to resist non-specific adsorption were enhanced obviously, only after the pretreatment of the conjugate pad and sample pad with pretreatment buffer, which containing BSA, sugar, Tween-20 and hydrophilic polymer. As we can see in Fig. S5 (Supporting information), S/N value reached the highest when the conjugate pad pretreatment PBS buffer containing 3% BSA, 5% sugar, 5% PVP10000 and 2% Tween-20, the sample pad pretreatment PBS buffer containing 0.5% Tween-20 and 2% PVP10000, and the dilution PBS buffer of immune fluorescence probes containing 5% BSA, 5% sugar, 1% PEG4000 and 0.1% Tween-20.

Chromatographic time was another important factor that affects the performance of our QD-IFTS. As previously mentioned, the QD-IFTS should have higher signal value and lower noise value. In other words, the best chromatographic time should reach the highest S/N. However, taking into account the actual situation, we will choose a shorter time with better performance as the best chromatographic time. As shown in Fig. S6a (Supporting information), we chose 25 min as the chromatographic time throughout the entire study with the relatively high signal value and low noise value. The sample volume also has an effect on the chromatographic process of test strip, and we discussed the effect of sample volume on the performance of our QD-IFTS. In theory, patients were more willing to provide a small amount of serum samples, but the immune reaction could not be completed when the sample volume was less than 30 mL. As shown in Fig. S6b (Supporting information), by comparing the S/N values of different sample volume, we chose 40 mL as the sample volume used in our QD-IFTS.

CEA was a kind of broad spectrum tumor marker used for cancer diagnosis, especially for the cancerous persons suffered colon cancer. Therefore, it was of great significance to detect CEA by a relatively simple detection method, such as immunochromatographic based POCT assays. In this work, CEA was chosen as the model analyte for our newly QD-IFTS. Firstly, we investigated the specificity of the newlyQD-IFTS with eight of different analyte.Fig. 2b showed that the fluorescence intensity of the non-target group was far below that of the CEA group, suggesting that the newly QD-IFTS possessed a strong ability to resist external interference. Subsequently, the detection sensitivity of the newly QD-IFTS was further investigated with concentration of CEA standard analytes at 0-1000 ng/mL. As expected, the fluorescence intensity of T line and the FT/FC value were increased with the concentration of CEA (Fig. 2a). The detection limit of the newly QD-IFTS for CEA, which was defined as the concentration of CEA corresponding with 3 times of FT(FBS)/FC (FBS), was 0.72 ng/mL and far below that of the conventional gold immunochromatographic test strips (AuNP-ICTS). As we all know, the detection performance of most immunoassays mainly depended on the background fluorescence signal, labeling efficiency and affinity of antibody. And the excellent sensitivity of the QDIFTS could be attributed to the excellent optical properties of QDs, the high loading efficiency of antibody to QDs and the maintenance of antibody activity.

Download:
Fig. 2. Sensitivity and specificity test of the newly QD-IFTS. (a) Concentrations of CEA were from 0 to 1000 ng/mL; (b) Antigens were AFP, HCG, FBS, CA125, CA153, CA199, PSA and CEA.

In addition, the robustness and efficiency of our newly QD-IFTS were further investigated with 300 serum samples from patients suffered colon cancer at different period of disease. As shown in Fig. 3a, we calculated FT/FC value of five CEA-positive serum samples and one CEA-negative serum sample, and built up the calibration curve for CEA quantification, the calibration curve for CEA was: y = 0.0995x0.333, (R2 = 0.998). Generally, the serum sample was regarded as CEA-positive, when the concentration of CEA in serum was higher than 5.9 ng/mL. In other words, 5.9 ng/mL was the cutoff value of immunoassays for CEA. We studied 300 serum samples from General Hospital of Tianjin Medical University with the cutoff value of 5.9 ng/mL, and the results were used to compare with that of electrochemiluminescence immunoassay. As we can see from Fig. 3b, the concentration of all the serum samples was well correspondent to the FT/FC value of our newly QD-IFTS. Fig. 6c revealed that there were 13 false positive case and 2 false negative case in 300 serum samples, which was reliable and accurate in serum samples detection. Therefore, the newly QD-IFTS possessed the possibility to detect clinical serum samples and showed great potential in clinical diagnosis of diseases (Scheme 1).

Download:
Fig. 3. Clinical serum samples detection of the newly QD-IFTS. (a) Standard curve of CEA serum; (b) 300 Serum samples of CEA; (c) False negative and False positive of the clinical test.

Download:
Scheme 1. Schematic diagram of quantum dots immune fluorescence test strip for tumor markers detection. (a) Quantum dots immune fluorescence test strips; (b) Detection of positive and negative samples; (c) Qualitative and quantitative detection of tumor markers.

Finally, we came to the conclusion that a newly QD-IFTS was developed, which was based on QDs as the fluorescence nanocarrier to prepare the immune fluorescence probes in the classical immunochromatography detection system, fulfilled the requirements of POCT. And we designed a homemade strips fluorescence reader for detection of fluorescence intensity of QDs on the QDIFTS.

Given the excellent optical properties of QDs, the high loading efficiency of antibody to QDs and the maintenance of antibody activity, the detection limit of the newly QD-IFTS was far below than conventional AuNP-ICTS, and the newly QD-IFTS was easy to operate, convenient and rapid. Under the optimized reaction conditions, chromatographic time of the newly QD-IFTS was only 25 min, sample volume of the newly QD-IFTS was only 40 μL and the LOD of the newly QD-IFTS was 0.72 ng/mL. The clinical serum test revealed that the results of the newly QD-IFTS were consistent with that of electrochemiluminescence immunoassay. Therefore, we have reason to believe that the newly QD-IFTS will certainly have great potential in clinical diagnosis of diseases.

Acknowledgments

The authors thank Dr. Jiumin Yang (School of Basic Medical Sciene, Tianjin Medical University) and Dr. Shuangnan Zhang (School of Materials Science and Engineering, Tianjin University) for designing the homemade strips fluorescence reader in this study. This study was financially supported by the National Natural Science Foundation of China (Nos. 51373117, 51303126 and 31600800), Tianjin Natural Science and Technology Foundation (No. 16ZXMJSY00010).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cclet.2017.07.026.

References
[1]
P. Craw, W. Balachandran, Lab Chip 12(2012) 2469-2486. DOI:10.1039/c2lc40100b
[2]
J. Wang, Biosens. Bioelectron. 21(2006) 1887-1892. DOI:10.1016/j.bios.2005.10.027
[3]
M. Davenport, K.E. Mach, L.D. Shortliffe, et al., Nat. Rev. Urol. 14(2017) 296-310. DOI:10.1038/nrurol.2017.20
[4]
G.J. Kost, Am. J. Clin. Pathol. 104(1995) S111-127. DOI:10.1093/ajcp/104.1.111
[5]
B.M. Lingerfelt, H. Mattoussi, E.R. Goldman, J.M. Mauro, G.P. Anderson, Anal. Chem. 75(2003) 4043-4049. DOI:10.1021/ac034139e
[6]
B.H. Liu, Z.J. Tsao, J.J. Wang, F.Y. Yu, Anal. Chem. 80(2008) 7029-7035. DOI:10.1021/ac800951p
[7]
C.A. Bessel, G.M. Denison, J.M. DeSimone, et al., J. Am. Chem. Soc. 125(2003) 4980-4981. DOI:10.1021/ja034091m
[8]
F. Ricci, G. Volpe, L. Micheli, G. Palleschi, Anal. Chim. Acta. 605(2007) 111-129. DOI:10.1016/j.aca.2007.10.046
[9]
G. Tsekenis, G.Z. Garifallou, F. Davis, et al., Anal. Chem. 80(2008) 2058-2062. DOI:10.1021/ac702070e
[10]
F. Rusmini, Z. Zhong, J. Feijen, Biomacromolecules 8(2007) 1775-1789. DOI:10.1021/bm061197b
[11]
J.S. Marcus, W.F. Anderson, S.R. Quake, Anal. Chem. 78(2006) 3084-3089. DOI:10.1021/ac0519460
[12]
X.Y. Wu, H.J. Liu, J.Q. Liu, et al., Nat. Biotechnol. 21(2003) 41-46.
[13]
X. Michalet, F.F. Pinaud, L.A. Bentolila, et al., Science 307(2005) 538-544. DOI:10.1126/science.1104274
[14]
I.L. Medintz, H.T. Uyeda, E.R. Goldman, H. Mattoussi, Nat. Mater. 4(2005) 435-446. DOI:10.1038/nmat1390
[15]
D.K. Yi, S.T. Selvan, S.S. Lee, et al., J. Am. Chem. Soc. 127(2005) 4990-4991. DOI:10.1021/ja0428863
[16]
Q.H. Yang, X.Q. Gong, T. Song, et al., Biosens. Bioelectron. 30(2011) 145-150. DOI:10.1016/j.bios.2011.09.002
[17]
C. Wang, F. Hou, Y. Ma, Biosens. Bioelectron. 68(2015) 156-162. DOI:10.1016/j.bios.2014.12.051
[18]
X.Y. Chen, M. Gan, H. Xu, et al., Food Control 46(2014) 225-232. DOI:10.1016/j.foodcont.2014.04.044
[19]
N.L. Berinstein, J. Clin. Oncol. 20(2002) 2197-2207. DOI:10.1200/JCO.2002.08.017
[20]
D.B. Beard, C.M. Haskell, Am. J. Med. 80(1986) 241-245. DOI:10.1016/0002-9343(86)90015-X
[21]
D.F. Wang, Y.Y. Li, Z.Y. Lin, B. Qiu, L.H. Guo, Anal. Chem. 87(2015) 5966-5972. DOI:10.1021/acs.analchem.5b01038
[22]
X. Li, W.B. Li, Q.H. Yang, et al., ACS Appl. Mater. Inter. 6(2014) 6406-6414. DOI:10.1021/am5012782
[23]
E.S. Speranskaya, N.V. Beloglazova, P. Lenain, et al., Biosens. Bioelectron. 53(2014) 225-231. DOI:10.1016/j.bios.2013.09.045
[24]
E.E. Lees, T.L. Nguyen, A.H. Clayton, P. Mulvaney, ACS Nano 3(2009) 1121-1128. DOI:10.1021/nn900144n
[25]
T. Jamieson, R. Bakhshi, D. Petrova, et al., Biomaterials 28(2007) 4717-4732. DOI:10.1016/j.biomaterials.2007.07.014
[26]
J.M. Slocik, E.R. Beckel, H. Jiang, et al., Adv. Mater. 18(2006) 2095-2100. DOI:10.1002/(ISSN)1521-4095
[27]
R. Tanaka, T. Yuhi, N. Nagatani, et al., Anal. Bioanal. Chem. 385(2006) 1414-1420. DOI:10.1007/s00216-006-0549-4