- Email: [email protected]
Label-free and homogeneous aptamer proximity binding assay for fluorescent detection of protein biomarkers in human serum
Talanta 141 (2015) 230–234
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Label-free and homogeneous aptamer proximity binding assay for fluorescent detection of protein biomarkers in human serum Yulian Wei a, Wenjiao Zhou a, Jun Liu b, Yaqin Chai a, Yun Xiang a,n, Ruo Yuan a a Key Laboratory of Luminescent and Real-Time Analytical Chemistry, Ministry of Education, School of Chemistry and Chemical Engineering, Southwest University, Chongqing 400715, PR China b Department of Biomedical Sciences, Texas Tech University Health Sciences Center, Amarillo, TX 79106, USA
art ic l e i nf o
a b s t r a c t
Article history: Received 21 January 2015 Received in revised form 25 March 2015 Accepted 2 April 2015 Available online 9 April 2015
By using the aptamer proximity binding assay strategy, the development of a label-free and homogeneous approach for fluorescent detection of human platelet-derived growth factor BB (PDGF-BB) is described. Two G-quadruplex forming sequence-linked aptamers bind to the PDGF-BB proteins, which leads to the increase in local concentration of the aptamers and promotes the formation of the G-quadruplex structures. Subsequently, the fluorescent dye, N-methylmesoporphyrin IX, binds to these G-quadruplex structures and generates enhanced fluorescence emission signal for sensitive detection of PDGF-BB. The association of the aptamers to the PDGF-BB proteins is characterized by using native polyacrylamide gel electrophoresis. The experimental conditions are optimized to reach an estimated detection limit of 3.2 nM for PDGF-BB. The developed method is also selective and can be applied for monitoring PDGF-BB in human serum samples. With the advantages of label-free and homogeneous detection, the demonstrated approach can be potentially employed to detect other biomarkers in a relatively simple way. & 2015 Elsevier B.V. All rights reserved.
Keywords: Aptamer Label-free Human platelet-derived growth factor Proximity binding Fluorescent detection
1. Introduction With the fast development of proteomics, it has been demonstrated that the changes in protein concentrations and altered protein expressions and distributions are associated with the occurrence and progression of different cancers [1–3]. For instance, elevated concentration of prostate specific antigen (PSA, 44.0 ng mL 1) is an indication of prostate cancer [4] while increasing concentration of cancer antigen 15-3 (CA15-3, 440.0 U mL 1) is related to breast cancer [5]. The identification and detection of these protein biomarkers is therefore of great importance in the early diagnosis, progression and prognosis of various cancers. The enzyme-linked immunosorbent assays (ELISAs) represent the most commonly used methods for protein biomarker detection. Although ELISAs can realize sensitive detection of different types of protein biomarkers, these methods are inherently complicated and time-consuming with the requirements of multiple washing, probe immobilization and highly trained personnel, which limit their wide application in routine clinical diagnosis [6,7]. The development of protein assays without involving complex assay protocols, to achieve homogeneous assay of
n
Corresponding author. Tel.: þ 86 23 68252277; fax: þ 86 23 68254000. E-mail address: [email protected] (Y. Xiang).
http://dx.doi.org/10.1016/j.talanta.2015.04.005 0039-9140/& 2015 Elsevier B.V. All rights reserved.
proteins for example, will potentially facilitate the detection of protein biomarkers. The proximity ligation protein assays first developed by the Landegren group have significantly simplified the monitoring of proteins [8]. In this type of assays, oligonucleotides were respectively conjugated to a pair of antibody recognition probes. When the antibodies associated with the target proteins, the conjugated oligonucleotides were brought into close proximity and hybridized with a connector oligonucleotide, which served as the ligation template to join the two termini of the conjugated oligonucleotides to form a new DNA strand with the assistance of the DNA ligase. The DNA was then subjected to real-time PCR amplification to achieve indirect and sensitive detection of the protein targets. Following this mechanism, proximity binding-induced DNA annealing [9–11] and assembly [12,13] have also been suggested as alternatives for protein detection. Indeed, the proximity bindingbased assays have enabled the detection of proteins in homogenous solutions. Despite this advantage, the proximity binding assays using antibodies as the recognition probes encountered the stability and cost issues. This has intrigued the employment of aptamers as recognition probes in proximity binding assays. Aptamers with high binding specificity and selectivity to the corresponding targets [14,15] are synthetic single stranded oligonucleotides (DNA or RNA) selected from random nucleic acid libraries by the SELEX approach. Aptamers, unlike antibodies, can be
Y. Wei et al. / Talanta 141 (2015) 230–234
generated in vitro, which is independent on the animal immune system. Besides, aptamers can be easily manipulated and have longer shelf-life. Due to these obvious advantages, aptamers have been increasingly used as recognition probes for the detection of proteins [16], cells [17], and ions [18,19]. Considering the nucleic acid nature, aptamers are particularly suitable to be used as recognition probes in proximity binding assays [20,21]. Recently, several aptamer proximity binding assays have been reported for the detection of thrombin [22–25]. These methods show some improvements, yet, they still require the conjugation of the probes with fluorescent tags, and label-free proximity binding assays have been rarely reported. Herein, we report on the development of an aptamer proximity binding assay for label-free and homogenous fluorescent detection of human platelet-derived growth factor BB (PDGF-BB), a critical growth factor protein dimer found in human platelets that plays important roles in regulating cell growth and division [26,27]. It can stimulate autocrine growth of different types of tumor cells and is often overexpressed in human malignant tumors [28], making PDGF-BB a potential protein marker for cancer diagnosis [29]. In our sensing strategy, the association of PDGF-BB with two binding aptamers facilitates the formation of G-quadruplex structures, which bind to the organic dye, N-methylmesoporphyrin IX (NMM) [30,31], and results in significantly enhanced fluorescent emission for sensitive detection of PDGF-BB. This approach integrates aptamer probes and G-quadruplex/NMM complexes into the proximity binding assays, thus leading to simple, convenient and sensitive detection of protein biomarkers.
2. Experimental 2.1. Materials and reagents PDGF-BB was purchased from ExCell Biology Inc. (Shanghai, China). Tris–HCl, thrombin, mouse immunoglobin G (IgG) and lysozyme were purchased from Sigma (St. Louis, MO). The extended PDGF-BB binding aptamers (PBA1 and PBA2) and the blocking ssDNA (B-DNA) with the sequences listed in Table 1 were ordered from Shanghai Sangon Biological Engineering Technology and Services Co., Ltd. (Shanghai, China). NMM was obtained from J&K Scientific Ltd. (Beijing, China). Other reagents were obtained from Kelong Chemical Company (Chengdu, China). All reagents were of analytical grade and ultrapure water (specific resistance of 18.25 MΩ cm) was used to prepare all solutions during the experimental process.
231
The mixture of B-DNA/PBA1 (50 nM) and PBA2 (50 nM) was then incubated with various concentrations of PDGF-BB in Tris–HCl buffer for 45 min at 25 °C. This was followed by the addition of NMM (2 mM) and further incubation of the mixture for 30 min. Finally, the fluorescence intensities of the mixtures were measured on a RF-5301-PC spectrophotometer (Shimadzu, Tokyo, Japan) with the excitation at 399 nm. The fluorescence emission spectra were collected from 580 nm to 650 nm with the slit widths for excitation and emission both at 10 nm. 2.3. Native polyacrylamide gel electrophoresis (PAGE) The sample solutions were subjected to electrophoresis analysis on the 16% native polyacrylamide gel. Electrophoresis was carried out in 1 TBE (pH 8.3) at a 100 V constant voltage for 90 min. The gels were then stained with ethidium bromide for 10 min, followed by photographing with a digital camera under UV irradiation.
3. Results and discussion 3.1. Principle for fluorescent detection of PDGF-BB Our aptamer proximity binding assay approach for fluorescent detection of PDGF-BB involves three oligonucleotide sequences: two extended PDGF-BB binding aptamers (denoted as PBA1 and PBA2) and one blocking ssDNA (B-DNA). Extra sequences are linked to the PDGF-BB binding aptamers to validate the assay protocol while not affecting the binding affinity of the aptamers to PDGF-BB. The G-quadruplex forming sequence employed to generate the signal output is split into two segments (the red regions in Scheme 1), which are separately integrated into PBA1 and PBA2. The probes for PDGF-BB detection are thus composed of four regions (Scheme 1): the G-quadruplex forming sequences (the red regions), the complementary sequences (the green regions), the T25 spacer sequences (the black regions) that eliminate the steric hindrance to the hybridization between the complementary sequences when binding to PDGF-BB, and the original PDGF-BB binding sequences (the blue regions). The B-DNA is hybridized to the green region and part of the G-quadruplex forming sequences in PBA1 to prevent hybridization between the complementary
2.2. Proximity binding assay of PDGF-BB The B-DNA/PBA1 was first prepared by mixing B-DNA (1.2 mM) with PBA1 (1.0 mM) in Tris–HCl buffer (10 mM Tris–HCl, 100 mM NaCl, 10 mM KCl, 1 mM MgCl2, pH 7.4), followed by heating to 90 °C for 5 min and cooling down to 25 °C slowly in a period of 3 h. Table 1 The sequences of the oligonucleotides used in this work. Name
Sequence
PBA1
5′-TAC TCA GGG CAC TGC AAG CAA TTG TGG TCC CAA TGG GCT GAG TAT TTT TTT TTT TTT TTT TTT TTT TTTCTG CTA GGT CTG GGT AGG G-3′ PBA2 5′-TTG GGC GGG TGA CCT AGC AAT TTT TTT TTT TTT TTT TTT TTT TTT ACT CAG GGC ACT GCA AGC AAT TGT GGT CCC AAT GGG CTG AGT A-3′ B-DNA 5′-CAT AGC GAG ATC CAG ACC TAG CAG-3′
*The italic, bold and underlined sequences, respectively, indicated the PDGF-BB binding aptamers, G-quadruplex forming sequence and the complementary sequences.
Scheme 1. Illustration of the label-free and homogeneous aptamer proximity binding assay for fluorescent detection of PDGF-BB. (For interpretation of the references to color in this scheme, the reader is referred to the web version of this article.)
232
Y. Wei et al. / Talanta 141 (2015) 230–234
sequences of PBA1 and PBA2 and to inhibit the formation of the G-quadruplex structures in the absence of the target PDGF-BB, leading to low fluorescent emission upon the addition of the fluorescent dye NMM. On the contrary, when the target PDGF-BB is added to the probe solution, PBA1 and PBA2 simultaneously bind to PDGF-BB, which results in the increase in the local concentration of PBA1 and PBA2 and facilitates the hybridizations between the complementary sequences of PBA1 and PBA2 through strand exchange of the B-DNA. As a result, the B-DNA is released from PBA1 and the two segments of the G-quadruplex forming sequences are brought into close proximity, which promotes the formation of the G-quadruplex structures. Subsequently, NMM binds to the G-quadruplex structures and generates amplified fluorescent emission for sensitive detection of PDGF-BB. 3.2. Assay validation The binding of the probes to the PDGF-BB target proteins was first examined by using native PAGE. As shown in Fig. 1A, the two probes PBA1/B-DNA and PBA2 exhibit two clear bands (Lane a and b), respectively. The mixture of PBA1/B-DNA and PBA2 shows no obvious mobility shift (Lane c) compared to that of PBA1/B-DNA or PBA2, indicating insignificant binding between the complementary regions of PBA1/B-DNA and PBA2 through strand exchange of the B-DNA. However, when PDGF-BB is added to the mixture of PBA1/B-DNA and PBA2, a clear band with significantly low mobility (top band of Lane d) is observed with the reduction
in band intensity in Lane c, suggesting the successful binding of both PBA1 and PBA2 to PDGF-BB. In addition, a band with high mobility (the bottom band), is also observed in Lane d. This band corresponds to the B-DNA released through strand exchange between the complementary regions of PBA1/B-DNA and PBA2 with increased local concentrations upon binding to PDGF-BB. The proof-of-demonstration of the assay protocol for fluorescent detection of PDGF-BB was also verified by fluorescence microscopic spectra. Considering the fact that high concentration of probes could result in a high background noise [15], the concentration of PBA1/B-DNA and PBA2 was fixed at 50 nM. Fluorescence measurements were performed after incubation of the solutions with NMM (2 mM) for 20 min. As displayed in Fig. 1B, the solution of PBA1/B-DNA or PBA2 alone shows minimal fluorescent emission (curves a and b) upon incubation with the NMM dye. Despite that the mixture of PBA1/B-DNA and PBA2 exhibits slight increase in fluorescence emission (curve c), the addition of PDGFBB (100 nM) results in substantial increase in fluorescence intensity (curve d). The comparison between curve c and d clearly indicates that the B-DNA effectively inhibits the hybridization between the complementary regions of PBA1/B-DNA and PBA2 and the formation of the G-quadruplex structures with the absence of PDGF-BB (curve c) while the presence of PDGF-BB facilitates the strand exchange and the formation of the G-quadruplex structures (curve d). These results thus suggest great potential of the method for fluorescent detection of PDGF-BB. 3.3. Optimization of the assay conditions
Fig. 1. (A) Native PAGE analysis of different reaction mixtures: (a) PBA1/B-DNA, (b) PBA2, (c) PBA1/B-DNA and PBA2, (d) PBA1/B-DNA, PBA2 and PDGF-BB. The concentrations of PBA1/B-DNA, PBA2 and PDGF-BB were 1 mM, 1 mM and 500 nM, respectively. (B) Typical fluorescence emission spectra of different solutions: (a) PBA1/B-DNA, (b) PBA2, (c) PBA1/B-DNA and PBA2, (d) PBA1/B-DNA, PBA2 and PDGF-BB. The concentrations of PBA1/B-DNA, PBA2 and PDGF-BB were 50 nM, 50 nM and 100 nM, respectively. Fluorescence spectra were recorded by incubating the mixtures at 25 °C for 45 min, followed by further incubation with NMM (2 mM) for 20 min.
To achieve optimal conditions for PDGF-BB detection, experimental parameters including the concentration and incubation time of NMM that affect the assay performance were optimized. According to Fig. 2A, although elevated concentration of NMM from 1 mM to 10 mM leads to the increase in fluorescence intensity (F) with the presence of 100 nM PDGF-BB (with incubation time of 20 min), the background noise (F0, with the absence of PDGF-BB) is also increased due to the increase in the concentration of the fluorescent dye, NMM. The best value of signal-to-noise ratio in the investigated concentration of NMM is determined to be 2 mM, and this concentration of NMM is fixed for subsequent experiments. The effect of the incubation time of NMM was also investigated from 5 to 45 min with a time interval of 5 min. As shown in Fig. 2B, the fluorescence intensity increases with prolonged incubation of NMM (2 mM) with PBA1/B-DNA and PBA1 in the presence of PDGF-BB (100 nM) from 5 to 30 min and levels off thereafter, indicating 30 min of NMM incubation time is suitable for the assay protocol.
Fig. 2. Effects of (A) the concentration (with incubation time of 20 min) and (B) incubation time (with concentration of NMM at 2 mM) of NMM for fluorescent detection of PDGF-BB. F and F0 corresponded to the incubation of the mixture of PBA1/B-DNA (50 nM) and PBA2 (50 nM) with and without, respectively, the presence of PDGF-BB (100 nM). Error bars, SD, n¼ 3.
Y. Wei et al. / Talanta 141 (2015) 230–234
233
Fig. 3. (A) Typical fluorescence emission spectra of the developed method for the detection of PDGF-BB at various concentrations. From a to g: 0 nM, 10 nM, 20 nM, 50 nM, 100 nM, 150 nM, and 200 nM. (B) The corresponding calibration plot of the concentration of PDGF-BB (from 10 nM to 200 nM) vs. the fluorescence intensity. Error bars, SD, n¼3.
Table 2 Recovery tests for PDGF-BB spiked into 10% diluted human serum samples (n¼ 6). Sample
Added (nM)
Found (nM)
Rate of recovery (%)
RSD (%)
1 2 3
10.0 15.0 20.0
9.9 15.1 20.3
96.2–102.5 98.3–103.4 97.9–102.9
3.3 1.9 2.0
into 10% diluted human serum samples, and recovery tests for the spiked PDGF-BB were performed by using the developed method. As listed in Table 2, the recoveries for the spiked PDGF-BB are in the range of 96.2–103.4% with relative standard deviations from 1.9% to 3.3%, indicating the applicability of the sensing strategy for PDGF-BB in serum samples.
Fig. 4. Selectivity investigation of the method for the detection of the target PDGFBB (100 nM) against other control proteins of lysozyme (1 mM), mouse IgG (1 mM), thrombin (1 mM) and the mixture of PDGF-BB with the control proteins.
3.4. Sensitivity and selectivity of the proposed method for the detection of PDGF-BB The quantitative behavior of the developed method for the detection of PDGF-BB was evaluated with different concentrations of PDGF-BB under the optimized conditions. From Fig. 3A, it can be seen that increasing concentration of PDGF-BB from 10 nM to 200 nM (curves b–g) leads to gradual increase in fluorescence intensity. By plotting fluorescence intensity vs. the concentration of PDGF-BB, a linear correlation (R2 ¼0.9980) was obtained over the range from 10 nM to 200 nM (Fig. 3B). Based on the 3s rule, the detection limit for PDGF-BB is estimated to be 3.2 nM. To evaluate the selectivity of the proposed method for PDGF-BB detection, other control proteins, including lysozyme, mouse IgG and thrombin were tested. As shown in Fig. 4, despite the presence of the control proteins (1 mM for the corresponding protein) at 10-fold excess over the target PDGF-BB, the fluorescence intensities are similar to that of the blank test (background noise, F0). However, the presence of even lower concentration of PDGF-BB (100 nM) causes significant increase in fluorescence intensity, and the mixture of PDGF-BB and the control proteins exhibits only slight change on the fluorescence intensity, indicating high selectivity of the method that is basically associated with the high binding specificity of the aptamers to the PDGF-BB target proteins. 3.5. Detection of PDGF-BB in serum samples To evaluate the potential application of the proposed method for real samples, various concentrations of PDGF-BB were spiked
4. Conclusion In summary, we have demonstrated a label-free and homogenous approach for fluorescent detection of PDGF-BB protein biomarkers by using the aptamer-based proximity binding strategy. The developed method here takes the advantages of the aptamer recognition probes, avoids the conjugation of the fluorescent tags to the probes and achieves single-step, homogeneous detection of protein biomarkers. This protein detection method is also sensitive and selective and can be applied for the monitoring of PDGF-BB in diluted human serum samples. With the selection of appropriate dual aptamer binding targets, the developed method can be a convenient sensing platform for the detection of other small molecules such as adenosine triphosphate and cocaine with split binding aptamers [32,33].
Acknowledgments This work was supported by the National Natural Science Foundation of China (Nos. 21275004 and 21275119), the New Century Excellent Talent Program of MOE (NCET-12-0932) and the Fundamental Research Funds for the Central Universities (XDJK2014A012).
References [1] M.S. Pepe, R. Etzioni, Z.D. Feng, J.D. Potter, M.L. Thompson, M. Thornquist, M. Winget, Y. Yasui, J. Natl. Cancer Inst. 93 (2001) 1054–1061. [2] S. Hanash, Nature 422 (2003) 226–232. [3] V. Kulasingam, E.P. Diamandis, Nat. Clin. Pract. Oncol. 5 (2008) 588–599. [4] H. Lilja, D. Ulmert, A.J. Vickers, Nat. Rev. Cancer 8 (2008) 268–278. [5] F. Safi, I. Kohler, H.G. Beger, E. Röttinger, Cancer 68 (1991) 574–582.
234
Y. Wei et al. / Talanta 141 (2015) 230–234
[6] S.A. Bhakta, E. Evans, T.E. Benavidez, C.D. Garcia, Anal. Chim. Acta 836 (2014) 1–17. [7] B. Deng, Y.W. Lin, C. Wang, F. Li, Z.X. Wang, H.Q. Zhang, X.F. Li, X.C. Le, Anal. Chim. Acta 837 (2014) 1–15. [8] S. Fredriksson, M. Gullberg, J. Jarvius, C. Olsson, K. Pietras, S.M. Gustafsdottir, A. Ostman, U. Landegren, Nat. Biotechnol. 20 (2002) 473–477. [9] T. Heyduk, Biophys. Chem. 151 (2010) 91–95. [10] E. Heyduk, T. Heyduk, Anal. Biochem. 396 (2010) 298–303. [11] L.M. McGregor, D.J. Gorin, C.E. Dumelin, D.R. Liu, J. Am. Chem. Soc. 132 (2010) 15522–15524. [12] H. Zhang, X.F. Li, X.C. Le, Anal. Chem. 84 (2012) 877–884. [13] F. Li, H. Zhang, C. Lai, X.F. Li, X.C. Le, Angew. Chem. Int. Ed. 51 (2012) 9317–9320. [14] H. Li, M. Wang, C.Z. Wang, W. Li, W.B. Qiang, D.K. Xu, Anal. Chem. 85 (2013) 4492–4499. [15] Y.Y. Xu, W.J. Zhou, M. Zhou, Y. Xiang, R. Yuan, Y.Q. Chai, Biosens. Bioelectron. 64 (2015) 306–310. [16] Y.Y. Xu, J. Xu, Y. Xiang, R. Yuan, Y.Q. Chai, Biosens. Bioelectron. 51 (2014) 293–296. [17] X.H. Fang, W.H. Tan, Acc. Chem. Res. 43 (2010) 48–57. [18] Y.L. Wei, W.J. Zhou, Y.Y. Xu, Y. Xiang, R. Yuan, Y.Q. Chai, RSC Adv. 4 (2014) 39082–39086. [19] L.H. Wang, X.F. Liu, X.F. Hu, S.P. Song, C.H. Fan, Chem. Commun. (2006) 3780–3782. [20] Y.L. Zhang, Y. Huang, J.H. Jiang, G.L. Shen, R.Q. Yu, J. Am. Chem. Soc. 129 (2007) 15448–15449.
[21] B.Q. Liu, B. Zhang, G.N. Chen, H.H. Yang, D.P. Tang, Anal. Chem. 86 (2014) 7773–7781. [22] Y.S. Ang, L.Y. Lanry Yung, Nucleic Acids Res. 42 (2014) 9523–9530. [23] H. Wu, K. Zhang, Y.L. Liu, H.Y. Wang, J. Wu, F.F. Zhu, P. Zou, Biosens. Bioelectron. 64 (2015) 572–578. [24] J.Y. Zhuang, Y. He, G.N. Chen, D.P. Tang, Electrochem. Commun. 47 (2014) 25–28. [25] X.Y. Wang, J.M. Zhou, W. Yun, S.S. Xiao, Z. Chang, P.G. He, Y.Z. Fang, Anal. Chim. Acta 598 (2007) 242–248. [26] R.F. Doolittle, M.W. Hunkapiller, L.E. Hood, S.G. Devare, K.C. Robbins, S.A. Aaronson, H.N. Antoniades, Science 221 (1983) 275. [27] M.D. Waterfield, G.T. Scrace, N. Whittle, P. Stroobant, A. Johnsson, A. Wasteson, B. Westermark, C.H. Heldin, J.S. Huang, T.F. Deuel, Nature 304 (1983) 7. [28] Y. Nakamura, F. Tanaka, Y. Yoshikawa, K. Mimori, H. Inoue, K. Yanaga, M. Mori, Ann. Surg. Oncol. 15 (2008) 2129. [29] A.H. Shih, C. Dai, X. Hu, M.K. Rosenblum, J.A. Koutcher, E.C. Holland, Cancer Res. 64 (2004) 4783. [30] C. Lin, J.K. Nangreave, Z. Li, Y. Liu, H. Yan, Nanomedicine 3 (2008) 521–528. [31] T. Li, E. Wang, S. Dong, Anal. Chem. 82 (2010) 1515–1520. [32] X.L. Zuo, Y. Xiao, Kevin W. Plaxco, J. Am. Chem. Soc. 131 (2009) 6944–6945. [33] B.Y. Jiang, M. Wang, Y. Chen, J.Q. Xie, Y. Xiang, Biosens. Bioelectron. 32 (2012) 305–308.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Label-free and homogeneous aptamer proximity binding assay for fluorescent detection of protein biomarkers in human serum Yulian Wei a, Wenjiao Zhou a, Jun Liu b, Yaqin Chai a, Yun Xiang a,n, Ruo Yuan a a Key Laboratory of Luminescent and Real-Time Analytical Chemistry, Ministry of Education, School of Chemistry and Chemical Engineering, Southwest University, Chongqing 400715, PR China b Department of Biomedical Sciences, Texas Tech University Health Sciences Center, Amarillo, TX 79106, USA
art ic l e i nf o
a b s t r a c t
Article history: Received 21 January 2015 Received in revised form 25 March 2015 Accepted 2 April 2015 Available online 9 April 2015
By using the aptamer proximity binding assay strategy, the development of a label-free and homogeneous approach for fluorescent detection of human platelet-derived growth factor BB (PDGF-BB) is described. Two G-quadruplex forming sequence-linked aptamers bind to the PDGF-BB proteins, which leads to the increase in local concentration of the aptamers and promotes the formation of the G-quadruplex structures. Subsequently, the fluorescent dye, N-methylmesoporphyrin IX, binds to these G-quadruplex structures and generates enhanced fluorescence emission signal for sensitive detection of PDGF-BB. The association of the aptamers to the PDGF-BB proteins is characterized by using native polyacrylamide gel electrophoresis. The experimental conditions are optimized to reach an estimated detection limit of 3.2 nM for PDGF-BB. The developed method is also selective and can be applied for monitoring PDGF-BB in human serum samples. With the advantages of label-free and homogeneous detection, the demonstrated approach can be potentially employed to detect other biomarkers in a relatively simple way. & 2015 Elsevier B.V. All rights reserved.
Keywords: Aptamer Label-free Human platelet-derived growth factor Proximity binding Fluorescent detection
1. Introduction With the fast development of proteomics, it has been demonstrated that the changes in protein concentrations and altered protein expressions and distributions are associated with the occurrence and progression of different cancers [1–3]. For instance, elevated concentration of prostate specific antigen (PSA, 44.0 ng mL 1) is an indication of prostate cancer [4] while increasing concentration of cancer antigen 15-3 (CA15-3, 440.0 U mL 1) is related to breast cancer [5]. The identification and detection of these protein biomarkers is therefore of great importance in the early diagnosis, progression and prognosis of various cancers. The enzyme-linked immunosorbent assays (ELISAs) represent the most commonly used methods for protein biomarker detection. Although ELISAs can realize sensitive detection of different types of protein biomarkers, these methods are inherently complicated and time-consuming with the requirements of multiple washing, probe immobilization and highly trained personnel, which limit their wide application in routine clinical diagnosis [6,7]. The development of protein assays without involving complex assay protocols, to achieve homogeneous assay of
n
Corresponding author. Tel.: þ 86 23 68252277; fax: þ 86 23 68254000. E-mail address: [email protected] (Y. Xiang).
http://dx.doi.org/10.1016/j.talanta.2015.04.005 0039-9140/& 2015 Elsevier B.V. All rights reserved.
proteins for example, will potentially facilitate the detection of protein biomarkers. The proximity ligation protein assays first developed by the Landegren group have significantly simplified the monitoring of proteins [8]. In this type of assays, oligonucleotides were respectively conjugated to a pair of antibody recognition probes. When the antibodies associated with the target proteins, the conjugated oligonucleotides were brought into close proximity and hybridized with a connector oligonucleotide, which served as the ligation template to join the two termini of the conjugated oligonucleotides to form a new DNA strand with the assistance of the DNA ligase. The DNA was then subjected to real-time PCR amplification to achieve indirect and sensitive detection of the protein targets. Following this mechanism, proximity binding-induced DNA annealing [9–11] and assembly [12,13] have also been suggested as alternatives for protein detection. Indeed, the proximity bindingbased assays have enabled the detection of proteins in homogenous solutions. Despite this advantage, the proximity binding assays using antibodies as the recognition probes encountered the stability and cost issues. This has intrigued the employment of aptamers as recognition probes in proximity binding assays. Aptamers with high binding specificity and selectivity to the corresponding targets [14,15] are synthetic single stranded oligonucleotides (DNA or RNA) selected from random nucleic acid libraries by the SELEX approach. Aptamers, unlike antibodies, can be
Y. Wei et al. / Talanta 141 (2015) 230–234
generated in vitro, which is independent on the animal immune system. Besides, aptamers can be easily manipulated and have longer shelf-life. Due to these obvious advantages, aptamers have been increasingly used as recognition probes for the detection of proteins [16], cells [17], and ions [18,19]. Considering the nucleic acid nature, aptamers are particularly suitable to be used as recognition probes in proximity binding assays [20,21]. Recently, several aptamer proximity binding assays have been reported for the detection of thrombin [22–25]. These methods show some improvements, yet, they still require the conjugation of the probes with fluorescent tags, and label-free proximity binding assays have been rarely reported. Herein, we report on the development of an aptamer proximity binding assay for label-free and homogenous fluorescent detection of human platelet-derived growth factor BB (PDGF-BB), a critical growth factor protein dimer found in human platelets that plays important roles in regulating cell growth and division [26,27]. It can stimulate autocrine growth of different types of tumor cells and is often overexpressed in human malignant tumors [28], making PDGF-BB a potential protein marker for cancer diagnosis [29]. In our sensing strategy, the association of PDGF-BB with two binding aptamers facilitates the formation of G-quadruplex structures, which bind to the organic dye, N-methylmesoporphyrin IX (NMM) [30,31], and results in significantly enhanced fluorescent emission for sensitive detection of PDGF-BB. This approach integrates aptamer probes and G-quadruplex/NMM complexes into the proximity binding assays, thus leading to simple, convenient and sensitive detection of protein biomarkers.
2. Experimental 2.1. Materials and reagents PDGF-BB was purchased from ExCell Biology Inc. (Shanghai, China). Tris–HCl, thrombin, mouse immunoglobin G (IgG) and lysozyme were purchased from Sigma (St. Louis, MO). The extended PDGF-BB binding aptamers (PBA1 and PBA2) and the blocking ssDNA (B-DNA) with the sequences listed in Table 1 were ordered from Shanghai Sangon Biological Engineering Technology and Services Co., Ltd. (Shanghai, China). NMM was obtained from J&K Scientific Ltd. (Beijing, China). Other reagents were obtained from Kelong Chemical Company (Chengdu, China). All reagents were of analytical grade and ultrapure water (specific resistance of 18.25 MΩ cm) was used to prepare all solutions during the experimental process.
231
The mixture of B-DNA/PBA1 (50 nM) and PBA2 (50 nM) was then incubated with various concentrations of PDGF-BB in Tris–HCl buffer for 45 min at 25 °C. This was followed by the addition of NMM (2 mM) and further incubation of the mixture for 30 min. Finally, the fluorescence intensities of the mixtures were measured on a RF-5301-PC spectrophotometer (Shimadzu, Tokyo, Japan) with the excitation at 399 nm. The fluorescence emission spectra were collected from 580 nm to 650 nm with the slit widths for excitation and emission both at 10 nm. 2.3. Native polyacrylamide gel electrophoresis (PAGE) The sample solutions were subjected to electrophoresis analysis on the 16% native polyacrylamide gel. Electrophoresis was carried out in 1 TBE (pH 8.3) at a 100 V constant voltage for 90 min. The gels were then stained with ethidium bromide for 10 min, followed by photographing with a digital camera under UV irradiation.
3. Results and discussion 3.1. Principle for fluorescent detection of PDGF-BB Our aptamer proximity binding assay approach for fluorescent detection of PDGF-BB involves three oligonucleotide sequences: two extended PDGF-BB binding aptamers (denoted as PBA1 and PBA2) and one blocking ssDNA (B-DNA). Extra sequences are linked to the PDGF-BB binding aptamers to validate the assay protocol while not affecting the binding affinity of the aptamers to PDGF-BB. The G-quadruplex forming sequence employed to generate the signal output is split into two segments (the red regions in Scheme 1), which are separately integrated into PBA1 and PBA2. The probes for PDGF-BB detection are thus composed of four regions (Scheme 1): the G-quadruplex forming sequences (the red regions), the complementary sequences (the green regions), the T25 spacer sequences (the black regions) that eliminate the steric hindrance to the hybridization between the complementary sequences when binding to PDGF-BB, and the original PDGF-BB binding sequences (the blue regions). The B-DNA is hybridized to the green region and part of the G-quadruplex forming sequences in PBA1 to prevent hybridization between the complementary
2.2. Proximity binding assay of PDGF-BB The B-DNA/PBA1 was first prepared by mixing B-DNA (1.2 mM) with PBA1 (1.0 mM) in Tris–HCl buffer (10 mM Tris–HCl, 100 mM NaCl, 10 mM KCl, 1 mM MgCl2, pH 7.4), followed by heating to 90 °C for 5 min and cooling down to 25 °C slowly in a period of 3 h. Table 1 The sequences of the oligonucleotides used in this work. Name
Sequence
PBA1
5′-TAC TCA GGG CAC TGC AAG CAA TTG TGG TCC CAA TGG GCT GAG TAT TTT TTT TTT TTT TTT TTT TTT TTTCTG CTA GGT CTG GGT AGG G-3′ PBA2 5′-TTG GGC GGG TGA CCT AGC AAT TTT TTT TTT TTT TTT TTT TTT TTT ACT CAG GGC ACT GCA AGC AAT TGT GGT CCC AAT GGG CTG AGT A-3′ B-DNA 5′-CAT AGC GAG ATC CAG ACC TAG CAG-3′
*The italic, bold and underlined sequences, respectively, indicated the PDGF-BB binding aptamers, G-quadruplex forming sequence and the complementary sequences.
Scheme 1. Illustration of the label-free and homogeneous aptamer proximity binding assay for fluorescent detection of PDGF-BB. (For interpretation of the references to color in this scheme, the reader is referred to the web version of this article.)
232
Y. Wei et al. / Talanta 141 (2015) 230–234
sequences of PBA1 and PBA2 and to inhibit the formation of the G-quadruplex structures in the absence of the target PDGF-BB, leading to low fluorescent emission upon the addition of the fluorescent dye NMM. On the contrary, when the target PDGF-BB is added to the probe solution, PBA1 and PBA2 simultaneously bind to PDGF-BB, which results in the increase in the local concentration of PBA1 and PBA2 and facilitates the hybridizations between the complementary sequences of PBA1 and PBA2 through strand exchange of the B-DNA. As a result, the B-DNA is released from PBA1 and the two segments of the G-quadruplex forming sequences are brought into close proximity, which promotes the formation of the G-quadruplex structures. Subsequently, NMM binds to the G-quadruplex structures and generates amplified fluorescent emission for sensitive detection of PDGF-BB. 3.2. Assay validation The binding of the probes to the PDGF-BB target proteins was first examined by using native PAGE. As shown in Fig. 1A, the two probes PBA1/B-DNA and PBA2 exhibit two clear bands (Lane a and b), respectively. The mixture of PBA1/B-DNA and PBA2 shows no obvious mobility shift (Lane c) compared to that of PBA1/B-DNA or PBA2, indicating insignificant binding between the complementary regions of PBA1/B-DNA and PBA2 through strand exchange of the B-DNA. However, when PDGF-BB is added to the mixture of PBA1/B-DNA and PBA2, a clear band with significantly low mobility (top band of Lane d) is observed with the reduction
in band intensity in Lane c, suggesting the successful binding of both PBA1 and PBA2 to PDGF-BB. In addition, a band with high mobility (the bottom band), is also observed in Lane d. This band corresponds to the B-DNA released through strand exchange between the complementary regions of PBA1/B-DNA and PBA2 with increased local concentrations upon binding to PDGF-BB. The proof-of-demonstration of the assay protocol for fluorescent detection of PDGF-BB was also verified by fluorescence microscopic spectra. Considering the fact that high concentration of probes could result in a high background noise [15], the concentration of PBA1/B-DNA and PBA2 was fixed at 50 nM. Fluorescence measurements were performed after incubation of the solutions with NMM (2 mM) for 20 min. As displayed in Fig. 1B, the solution of PBA1/B-DNA or PBA2 alone shows minimal fluorescent emission (curves a and b) upon incubation with the NMM dye. Despite that the mixture of PBA1/B-DNA and PBA2 exhibits slight increase in fluorescence emission (curve c), the addition of PDGFBB (100 nM) results in substantial increase in fluorescence intensity (curve d). The comparison between curve c and d clearly indicates that the B-DNA effectively inhibits the hybridization between the complementary regions of PBA1/B-DNA and PBA2 and the formation of the G-quadruplex structures with the absence of PDGF-BB (curve c) while the presence of PDGF-BB facilitates the strand exchange and the formation of the G-quadruplex structures (curve d). These results thus suggest great potential of the method for fluorescent detection of PDGF-BB. 3.3. Optimization of the assay conditions
Fig. 1. (A) Native PAGE analysis of different reaction mixtures: (a) PBA1/B-DNA, (b) PBA2, (c) PBA1/B-DNA and PBA2, (d) PBA1/B-DNA, PBA2 and PDGF-BB. The concentrations of PBA1/B-DNA, PBA2 and PDGF-BB were 1 mM, 1 mM and 500 nM, respectively. (B) Typical fluorescence emission spectra of different solutions: (a) PBA1/B-DNA, (b) PBA2, (c) PBA1/B-DNA and PBA2, (d) PBA1/B-DNA, PBA2 and PDGF-BB. The concentrations of PBA1/B-DNA, PBA2 and PDGF-BB were 50 nM, 50 nM and 100 nM, respectively. Fluorescence spectra were recorded by incubating the mixtures at 25 °C for 45 min, followed by further incubation with NMM (2 mM) for 20 min.
To achieve optimal conditions for PDGF-BB detection, experimental parameters including the concentration and incubation time of NMM that affect the assay performance were optimized. According to Fig. 2A, although elevated concentration of NMM from 1 mM to 10 mM leads to the increase in fluorescence intensity (F) with the presence of 100 nM PDGF-BB (with incubation time of 20 min), the background noise (F0, with the absence of PDGF-BB) is also increased due to the increase in the concentration of the fluorescent dye, NMM. The best value of signal-to-noise ratio in the investigated concentration of NMM is determined to be 2 mM, and this concentration of NMM is fixed for subsequent experiments. The effect of the incubation time of NMM was also investigated from 5 to 45 min with a time interval of 5 min. As shown in Fig. 2B, the fluorescence intensity increases with prolonged incubation of NMM (2 mM) with PBA1/B-DNA and PBA1 in the presence of PDGF-BB (100 nM) from 5 to 30 min and levels off thereafter, indicating 30 min of NMM incubation time is suitable for the assay protocol.
Fig. 2. Effects of (A) the concentration (with incubation time of 20 min) and (B) incubation time (with concentration of NMM at 2 mM) of NMM for fluorescent detection of PDGF-BB. F and F0 corresponded to the incubation of the mixture of PBA1/B-DNA (50 nM) and PBA2 (50 nM) with and without, respectively, the presence of PDGF-BB (100 nM). Error bars, SD, n¼ 3.
Y. Wei et al. / Talanta 141 (2015) 230–234
233
Fig. 3. (A) Typical fluorescence emission spectra of the developed method for the detection of PDGF-BB at various concentrations. From a to g: 0 nM, 10 nM, 20 nM, 50 nM, 100 nM, 150 nM, and 200 nM. (B) The corresponding calibration plot of the concentration of PDGF-BB (from 10 nM to 200 nM) vs. the fluorescence intensity. Error bars, SD, n¼3.
Table 2 Recovery tests for PDGF-BB spiked into 10% diluted human serum samples (n¼ 6). Sample
Added (nM)
Found (nM)
Rate of recovery (%)
RSD (%)
1 2 3
10.0 15.0 20.0
9.9 15.1 20.3
96.2–102.5 98.3–103.4 97.9–102.9
3.3 1.9 2.0
into 10% diluted human serum samples, and recovery tests for the spiked PDGF-BB were performed by using the developed method. As listed in Table 2, the recoveries for the spiked PDGF-BB are in the range of 96.2–103.4% with relative standard deviations from 1.9% to 3.3%, indicating the applicability of the sensing strategy for PDGF-BB in serum samples.
Fig. 4. Selectivity investigation of the method for the detection of the target PDGFBB (100 nM) against other control proteins of lysozyme (1 mM), mouse IgG (1 mM), thrombin (1 mM) and the mixture of PDGF-BB with the control proteins.
3.4. Sensitivity and selectivity of the proposed method for the detection of PDGF-BB The quantitative behavior of the developed method for the detection of PDGF-BB was evaluated with different concentrations of PDGF-BB under the optimized conditions. From Fig. 3A, it can be seen that increasing concentration of PDGF-BB from 10 nM to 200 nM (curves b–g) leads to gradual increase in fluorescence intensity. By plotting fluorescence intensity vs. the concentration of PDGF-BB, a linear correlation (R2 ¼0.9980) was obtained over the range from 10 nM to 200 nM (Fig. 3B). Based on the 3s rule, the detection limit for PDGF-BB is estimated to be 3.2 nM. To evaluate the selectivity of the proposed method for PDGF-BB detection, other control proteins, including lysozyme, mouse IgG and thrombin were tested. As shown in Fig. 4, despite the presence of the control proteins (1 mM for the corresponding protein) at 10-fold excess over the target PDGF-BB, the fluorescence intensities are similar to that of the blank test (background noise, F0). However, the presence of even lower concentration of PDGF-BB (100 nM) causes significant increase in fluorescence intensity, and the mixture of PDGF-BB and the control proteins exhibits only slight change on the fluorescence intensity, indicating high selectivity of the method that is basically associated with the high binding specificity of the aptamers to the PDGF-BB target proteins. 3.5. Detection of PDGF-BB in serum samples To evaluate the potential application of the proposed method for real samples, various concentrations of PDGF-BB were spiked
4. Conclusion In summary, we have demonstrated a label-free and homogenous approach for fluorescent detection of PDGF-BB protein biomarkers by using the aptamer-based proximity binding strategy. The developed method here takes the advantages of the aptamer recognition probes, avoids the conjugation of the fluorescent tags to the probes and achieves single-step, homogeneous detection of protein biomarkers. This protein detection method is also sensitive and selective and can be applied for the monitoring of PDGF-BB in diluted human serum samples. With the selection of appropriate dual aptamer binding targets, the developed method can be a convenient sensing platform for the detection of other small molecules such as adenosine triphosphate and cocaine with split binding aptamers [32,33].
Acknowledgments This work was supported by the National Natural Science Foundation of China (Nos. 21275004 and 21275119), the New Century Excellent Talent Program of MOE (NCET-12-0932) and the Fundamental Research Funds for the Central Universities (XDJK2014A012).
References [1] M.S. Pepe, R. Etzioni, Z.D. Feng, J.D. Potter, M.L. Thompson, M. Thornquist, M. Winget, Y. Yasui, J. Natl. Cancer Inst. 93 (2001) 1054–1061. [2] S. Hanash, Nature 422 (2003) 226–232. [3] V. Kulasingam, E.P. Diamandis, Nat. Clin. Pract. Oncol. 5 (2008) 588–599. [4] H. Lilja, D. Ulmert, A.J. Vickers, Nat. Rev. Cancer 8 (2008) 268–278. [5] F. Safi, I. Kohler, H.G. Beger, E. Röttinger, Cancer 68 (1991) 574–582.
234
Y. Wei et al. / Talanta 141 (2015) 230–234
[6] S.A. Bhakta, E. Evans, T.E. Benavidez, C.D. Garcia, Anal. Chim. Acta 836 (2014) 1–17. [7] B. Deng, Y.W. Lin, C. Wang, F. Li, Z.X. Wang, H.Q. Zhang, X.F. Li, X.C. Le, Anal. Chim. Acta 837 (2014) 1–15. [8] S. Fredriksson, M. Gullberg, J. Jarvius, C. Olsson, K. Pietras, S.M. Gustafsdottir, A. Ostman, U. Landegren, Nat. Biotechnol. 20 (2002) 473–477. [9] T. Heyduk, Biophys. Chem. 151 (2010) 91–95. [10] E. Heyduk, T. Heyduk, Anal. Biochem. 396 (2010) 298–303. [11] L.M. McGregor, D.J. Gorin, C.E. Dumelin, D.R. Liu, J. Am. Chem. Soc. 132 (2010) 15522–15524. [12] H. Zhang, X.F. Li, X.C. Le, Anal. Chem. 84 (2012) 877–884. [13] F. Li, H. Zhang, C. Lai, X.F. Li, X.C. Le, Angew. Chem. Int. Ed. 51 (2012) 9317–9320. [14] H. Li, M. Wang, C.Z. Wang, W. Li, W.B. Qiang, D.K. Xu, Anal. Chem. 85 (2013) 4492–4499. [15] Y.Y. Xu, W.J. Zhou, M. Zhou, Y. Xiang, R. Yuan, Y.Q. Chai, Biosens. Bioelectron. 64 (2015) 306–310. [16] Y.Y. Xu, J. Xu, Y. Xiang, R. Yuan, Y.Q. Chai, Biosens. Bioelectron. 51 (2014) 293–296. [17] X.H. Fang, W.H. Tan, Acc. Chem. Res. 43 (2010) 48–57. [18] Y.L. Wei, W.J. Zhou, Y.Y. Xu, Y. Xiang, R. Yuan, Y.Q. Chai, RSC Adv. 4 (2014) 39082–39086. [19] L.H. Wang, X.F. Liu, X.F. Hu, S.P. Song, C.H. Fan, Chem. Commun. (2006) 3780–3782. [20] Y.L. Zhang, Y. Huang, J.H. Jiang, G.L. Shen, R.Q. Yu, J. Am. Chem. Soc. 129 (2007) 15448–15449.
[21] B.Q. Liu, B. Zhang, G.N. Chen, H.H. Yang, D.P. Tang, Anal. Chem. 86 (2014) 7773–7781. [22] Y.S. Ang, L.Y. Lanry Yung, Nucleic Acids Res. 42 (2014) 9523–9530. [23] H. Wu, K. Zhang, Y.L. Liu, H.Y. Wang, J. Wu, F.F. Zhu, P. Zou, Biosens. Bioelectron. 64 (2015) 572–578. [24] J.Y. Zhuang, Y. He, G.N. Chen, D.P. Tang, Electrochem. Commun. 47 (2014) 25–28. [25] X.Y. Wang, J.M. Zhou, W. Yun, S.S. Xiao, Z. Chang, P.G. He, Y.Z. Fang, Anal. Chim. Acta 598 (2007) 242–248. [26] R.F. Doolittle, M.W. Hunkapiller, L.E. Hood, S.G. Devare, K.C. Robbins, S.A. Aaronson, H.N. Antoniades, Science 221 (1983) 275. [27] M.D. Waterfield, G.T. Scrace, N. Whittle, P. Stroobant, A. Johnsson, A. Wasteson, B. Westermark, C.H. Heldin, J.S. Huang, T.F. Deuel, Nature 304 (1983) 7. [28] Y. Nakamura, F. Tanaka, Y. Yoshikawa, K. Mimori, H. Inoue, K. Yanaga, M. Mori, Ann. Surg. Oncol. 15 (2008) 2129. [29] A.H. Shih, C. Dai, X. Hu, M.K. Rosenblum, J.A. Koutcher, E.C. Holland, Cancer Res. 64 (2004) 4783. [30] C. Lin, J.K. Nangreave, Z. Li, Y. Liu, H. Yan, Nanomedicine 3 (2008) 521–528. [31] T. Li, E. Wang, S. Dong, Anal. Chem. 82 (2010) 1515–1520. [32] X.L. Zuo, Y. Xiao, Kevin W. Plaxco, J. Am. Chem. Soc. 131 (2009) 6944–6945. [33] B.Y. Jiang, M. Wang, Y. Chen, J.Q. Xie, Y. Xiang, Biosens. Bioelectron. 32 (2012) 305–308.