Thrombin Ultrasensitive Detection Based on Chiral Supramolecular

Letter pubs.acs.org/ac

Thrombin Ultrasensitive Detection Based on Chiral Supramolecular Assembly Signal-Amplified Strategy Induced by Thrombin-Binding Aptamer Gang Shen,† Hong Zhang,*,† Chunrong Yang,‡ Qianfan Yang,*,‡ and Yalin Tang*,† †

National Laboratory for Molecular Sciences, Centre for Molecular Sciences, State Key Laboratory for Structural Chemistry of Unstable and Stable Species, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China ‡ College of Chemistry, Sichuan University, Chengdu, 610065, China S Supporting Information *

ABSTRACT: Thrombin plays a critical role in hemostasis and hemolysis. It is of high importance to develop a system toward thrombin detection with high sensitivity and high selectivity for both research and clinical diagnosis applications. In this paper, we developed a thrombin detection assay by taking advantage of the novel signal amplified strategy based on the chiral supramolecular assembly in physiological K+ background. This assay could detect thrombin as low concentration as about 2 pM and provided a highly specific selectivity among several common interferences. Furthermore, the assay can discriminate thrombin from other nonspecific analogous proteins with high selectivity and can be used to detect thrombin in diluted real human serum samples, which suggested its great potential for rapid detection of thrombin in the clinic.

T

complicated sample preparation, and low analysis speed. For the other, it is more critical that potassium ion (K+), besides thrombin, could also induce TBA folding into quite similar Gquadruplex structure. As a common ion in physiological condition, K+ would strongly disturb the detection of thrombin. And the attempts to keep TBA unfolding in the presence of K+ (such as using complementary oligonucleotides) would decrease the sensitivity of thrombin detection. In this work, we tried to solve the two problems by using a novel signal-amplified strategy based on chiral supramolecular assembly induced by TBA. We know that the cyanine dyes have a large conjugated π-system promoting self-assembly into supramolecular structures, called J-aggregates, with a narrow intense absorption band shifted toward longer wavelengths compared with the spectrum of monomers.17,18 In our previous work, we designed a cyanine dyes probe DMSB (3-ethyl-2-[3(3-ethyl-3H-benzoselenazol-2-ylidene)-2-methylprop-1-enyl] benzoselenazolium bromide) (shown in Figure 1A) which could specifically bind on TBA G-quadruplex and assemble to chiral J-aggregates, which greatly amplified the detecting signals.19 Also in this strategy, the cyanine dye (DMSB) loses its ability of self-assembly in solution. When the TBA Gquadruplex is present, the assembly potential of DMSB is

hrombin is a specific serine endoprotease that plays a critical role in hemostasis. It naturally serves as a blood clotting factor that solidifies the fibrin clot in terms of strength and elasticity.1−6 The concentration of thrombin in blood can change from the picomolar to micromolar level corresponding to the normal conditions and the coagulation process.7 As a key enzyme in pathologic processes such as leukemia, arterial thrombosis, and liver disease, the detection and quantification of thrombin in biological serum or other complex samples is clinically relevant.8−11 Detecting thrombin, especially in the picomolar range in blood is of importance to clinical diagnosis of the related disease. Thrombin-binding aptamer (TBA)12 is a 15-mer singlestranded DNA with the sequence 5′GGTTGGTGTGGTTGG-3′. When binding to thrombin, TBA could change its secondary structure from random coil to chairlike antiparallel G-quadruplex. Because of its high affinity and unique structure transformation, a variety of sensors was coupled with TBA and developed for thrombosis diagnostics in the past 2 decades.13−16 However, there are two problems that limit the application of these attempts. For one thing, the dissociation constant of TBA-thrombin has been reported in the nanomolar range, which does not fulfill the clinical detecting demand, around the picomolar range. It has been reported several ways to amplify the detecting signal and achieve to decrease the limit of detection to the pM level, but few of these methods were applied in clinical practice, due to their limitations of sophisticated and high cost instruments, © 2016 American Chemical Society

Received: October 31, 2016 Accepted: December 13, 2016 Published: December 13, 2016 548

DOI: 10.1021/acs.analchem.6b04247 Anal. Chem. 2017, 89, 548−551

Letter

Analytical Chemistry

room temperature at a rate of 1 °C min−1. The concentration of TBA stock solution was determined by measuring its absorbance at 260 nm. All oligonucleotide samples were stored for more than 24 h at 4 °C and then structurally identified by circular dichroism (CD).23 TBA (6.24 μL, 1.28 × 10−4 M) and DMSB (72 μL, 1.66 × 10−4 M) were successively added into 1.0 mL calibrated test tubes, then diluted to the mark with PBS and mixed thoroughly for 10 min at 4 °C. All the samples were kept in darkness overnight at 4 °C before measurement in order to realize the full complexation. Absorption and CD Spectra Measurements. Ultraviolet−visible (UV−vis) absorption spectra were measured by a JASCO J-815 spectrophotometer in a 10 mm quartz cell at room temperature. All the CD spectra were recorded on a JASCO J-815 spectrophotometer in a 10 mm quartz cell at room temperature. And all spectra were collected with a scan speed of 500 nm min−1 and a response time of 0.5 s between 200 and 700 nm with three scans averaged. Determination of Thrombin in Human Serum. The fresh human blood samples were supplied by the Peking University People’s Hospital. The human serum samples were obtained by centrifuging (1000 rpm, 5 min) the fresh blood samples. The serum samples were diluted 1000 times with PBS before detection. Dilution is a commonly adopted pretreatment procedure for protease detection in samples of high complexity such as human serum.24,25 The thrombin content in serum was derived from the standard curve and the regression equation. The test was performed by using the standard addition method.

Figure 1. (A) Molecular formula of DMSB and the sequence of TBA. (B) Strategy of thrombin ultrasensitive detection. (C) CD spectra of 12 μM DMSB with 1 μM TBA: (a) in tris-HCl buffer, (b) in PBS, and (c) in PBS with 5 pM thrombin.

reactivated so that it can assemble to J-aggregates and realizing the signal-ampilified strategy for TBA G-quadruple recognition. Applying DMSB into thrombin detection could achieve some unique properties that common TBA G-quadruplex probes do not have. First, owing to the novel assembly ability, DMSB could amplify the detection signal of TBA G-quadruplex and consequently strongly increase the sensitivity of thrombin detection. Meanwhile, the chiral assembly of DMSB is very sensitive to specific G-quadruplex structure. The slightly change in G-quadruplex structure would dramatically influence the assembly behavior of DMSB. It is reported that the TBA Gquadruplexes induced by thrombin and K+ have slight difference, which is hard for the unimolecular probe to identify but could be reflexed by the DMSB assembly state.20 In this way, DMSB has the potential to exclude the influence from physiological K+ background. To the best of our knowledge, the basis of supramolecular assembly for thrombin detection has not reported before. What is more, we also found that this novel thrombin detection assay can discriminate thrombin from other nonspecific analogous proteins with high selectivity and can be used to detect thrombin in diluted real human serum samples, which suggested its great potential for rapid detection of thrombin in clinic.

■

RESULTS AND DISCUSSION In order to develop a thrombin detection method based on the structural change of TBA, first of all, we need to confirm the conformation of TBA in different conditions. As shown in Figure S1, TBA exhibited very weak CD signal in 10 mM trisHCl buffer solution (without K+), indicating relative random coil conformation, while a typical antiparallel G-quadruplex CD signal (a positive signal at 293.5 nm and a negative one at 272.5 nm) in phosphate buffer (5 mM KH2PO4/K2HPO4, pH = 7.0, PBS). After adding thrombin into the PBS system, the positive CD signal of TBA slightly red-shifted to 297 nm, inferring slight structural change of TBA G-quadruplex.26 It is proved that antiparallel G-quadruplex structure can activate the assembly ability of DMSB and induce DMSB forming chiral J-aggregates while single-stranded DNA cannot. Furthermore, the slight difference in G-quadruplexes could strongly influence the formation of DMSB J-aggregates.27 In this work, we developed an ultrasensitive thrombin detecting method based on the unique assembly property of DMSB. As shown in Figure 1B, in the lack of K+ or thrombin, the assembly ability of DMSB was inactivated. DMSB stays in the form of monomer with single-stranded TBA and presents no induced CD signal (Figure 1C,a). While in the presence of K+, TBA would be folded into an antiparallel G-quadruplex which can activate the assembly ability of DMSB and facilitate the forming of its chiral J-aggregates. As shown in Figure 1C,b, the induced CD signal intensity of DMSB J-aggregates is around 33 mdeg. After adding thrombin into the PBS system, the G-quadruplex conformation of TBA would be slightly changed and the ability to facilitate DMSB assembly was dramatically enhanced. Only 5 pM thrombin increased the induced CD signal intensity of DMSB J-aggregates to around 116 mdeg (Figure 1C,c). The similar phenomenon is also observed in UV−vis absorption

■

EXPERIMENTAL SECTION Sample Preparation. The cyanine dye DMSB was synthesized according to Hamer’s and Brooker’s methods,21,22 and the purity was evaluated by mass spectrometry and nuclear magnetic resonance. The TBA oligonucleotide was purchased from Sangon Biotech Co., Ltd. (Shanghai, China) and purified by HPLC (purity 98%). The proteins (thrombin, human serum albumin, bovine serum albumin, immunoglobulin G), and lactic acid were purchased from Sigma-Aldrich (Beijing, China). Analytical grade methanol, KH2PO4, and K2HPO4 were purchased from Beijing Chem. Co. (China). Ultrapure water was prepared by the Milli-Q Gradient ultrapure water system (Millipore). The stock solution of DMSB was prepared by dissolving it in methanol to 166 μM and stored in the dark at 4 °C. The stock solution of TBA was prepared by dissolving it to 5 mM phosphate buffer (PBS, KH2PO4/K2HPO4, pH = 7.0) followed by filtering through a microfiltration membrane (Φ = 0.22 μm). Then it was heated to 90 °C for 5 min and gradually cooled to 549

DOI: 10.1021/acs.analchem.6b04247 Anal. Chem. 2017, 89, 548−551

Letter

Analytical Chemistry

ologies, which are popular tools used to develop thrombin detection methods. The signal-amplified function of DMSB can dramatically enhance the detection limit of CD and UV−vis spectroscopies and makes them comparable to fluorescence electrochemical methodology.13−15 Also, a comparison between this proposed method and other reported methods for thrombin determination with regards to detection limit is summed up in Table S1. It can be seen from the table that this proposed method has a comparable LOD to the electrochemical and fluorescent methods. And what is more, the sensitivity of this method was much better than that of the reported thrombin detection by using absorption spectroscopy. In order to evaluate the specificity of DMSB, the interactions between DMSB and 6 interferences are studied, including common proteins (human serum albumin, HSA, and immunoglobulin G, IgG), amino acids (cysteine, Cys, and glutamic acid, Glu), and organic acid (lactic acid, LA) in human organism and bull serum albumin (BSA), which is commonly used in many biomolecular treatments. As shown in Figure 2D, 2 pM thrombin can induce about 0.08 absorbance change at 641 nm (assigned to DMSB J-aggregates), while all interferences in the concentrations of 1 nM, 500-fold excessive than that of thrombin, can cause less than 0.02 absorbance change, which can hardly influence the detection of thrombin. In CD measurement, 2 pM thrombin can also enhance the detection signal by about 12 mdeg while the interferences even induced a decrease of the signal (Figure 2C). These above data revealed that our detection system is endowed with high thrombin selectivity against other common proteins and amino acids. Also clearly, thrombin can be easily differentiated when all the other interference components are present with a 500fold higher concentration. To further evaluate the detecting ability of DMSB in complex biological matrixes, we also applied the assay into human serum samples. Figure 3 is the variation of the CD

spectra (Figure S3). On the basis of the above results, the CD (positive signal at 644 nm) or the absorbance (peak around 641 nm) of the induced DMSB J-aggregates could be considered as unique signatures for thrombin detection. To achieve optimal conditions for thrombin detection, a key variable (the concentration ratio of [TBA]/[DMSB]) was optimized to obtain better analysis performance (Figures S4 and S5). These experimental data showed that 1:12 of the concentration ratio of [TBA]/[DMSB] was chosen as an optimum ratio to detect thrombin. Moreover, in order to simulate the physiological human serum system, phosphate buffer with 5 mM K+ (termed as PBS) was used in the whole experiment. Figure 2A,B is the variation of the detection

Figure 2. Variation of the CD signals intensity at 644 nm (A) and the absorbance at 641 nm (B) of 12 μM DMSB (with 1 μM TBA) in the presence of different concentrations of thrombin in PBS. The changes of CD signal intensities at 644 nm (C) and the absorbances at 641 nm (D) of 12 μM DMSB (with 1 μM TBA) in and the presence of 2 pM thrombin (red) and 1 nM of the interferences (pattern), including HSA, BSA, IgG, LA, Cys, and Glu in PBS.

signatures of DMSB J-aggregates in the presence of different concentrations of thrombin in PBS. It is indicated that the responses of CD signal intensity to thrombin concentration exhibits a good linear coefficient (R2) up to 0.989, suggesting excellent quantification ability. In experimental conditions but not necessarily a guarantee of the best conditions, as low as 2 pM thrombin is sufficient to induce observable and distinguishable detecting signals (S/N ≥ 3) (Figure 2A). For instrumental analysis, we could reveal a tinier amount of 2 pM (corresponding to about 75 pg mL−1) thrombin by using this signal-amplified detecting strategy. Besides, the absorption experiments showed the similar results. The limit of detection (LOD) by absorption spectroscopy is around 2 pM, and the linear coefficient R2 = 0.984 (Figure 2B). It is reported that the binding rate of TBA onto thrombin is 1:1 or 1:2.28,29 In the case of the unimolecular probe, one probe molecule can bind on only one TBA molecule. Since one TBA molecule can bind 1 or 2 thrombin, one probe can consequently detect 1 or 2 thrombin. The detection signal is quite limited. However, for the supramolecular probe (DMSB in this work), by using the signal-amplified strategy based on chiral assembly, one TBA molecule could induce the assembly of a large amount of DMSB molecule, which means consequently 1 or 2 thrombin could induce a much larger detection signal in the DMSB system. The LOD of DMSB against thrombin is 5000-fold lower than that against TBA G-quadruplex (10 nM). Furthermore, it is well-known that the sensitivities of CD and absorption spectroscopies are several orders of magnitude lower than those of fluorescent and electrochemical method-

Figure 3. Variation of the CD signals intensity at 644 nm of 12 μM DMSB (with 1 μM TBA) after adding different concentrations of thrombin in diluted human serum.

signal intensity at 644 nm of 12 μM DMSB (with 1 μM TBA) after adding different concentrations of thrombin in diluted human serum. It is indicated that DMSB presents as excellent determination ability in the human serum system as in PBS. The LOD is also around 2 pM and the linear coefficient R2 = 0.998. The recoveries of the samples (in the range 97.41− 101.43%) and the RSD values (less than 1.3%) summarized in Table 1 indicated that the accuracy and precision of the proposed method were satisfactory in diluted human serum and has the potential to be applied in clinical thrombin determination. 550

DOI: 10.1021/acs.analchem.6b04247 Anal. Chem. 2017, 89, 548−551

Letter

Analytical Chemistry

(2) Jain, A. S.; Grand, C. L.; Bearss, D. J.; Hurley, L. H. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 11593−11598. (3) Rankin, S.; Reszka, A. P.; Huppert, J.; Zloh, M.; Parkinson, A.; Todd, K.; Ladame, S.; Balasubramanian, S.; Neidle, S. J. Am. Chem. Soc. 2005, 127, 10584−10589. (4) Dexheimer, T. S.; Sun, D.; Hurley, L. H. J. Am. Chem. Soc. 2006, 128, 5404−5415. (5) Tan, J. H.; Gu, L. Q.; Wu, J. Y. Mini-Rev. Med. Chem. 2008, 8, 1163−1178. (6) Kumari, S.; Bugaut, A.; Huppert, J. L.; Balasubramanian, S. Nat. Chem. Biol. 2007, 3, 218−221. (7) Shuman, M. A.; Majerus, P. W. J. J. Clin. Invest. 1976, 58, 1249− 1258. (8) Alzeer, J.; Vummidi, B. R.; Roth, P. J. C.; Luedtke, N. W. Angew. Chem., Int. Ed. 2009, 48, 9362−9365. (9) Tera, M.; Iida, K.; Ikebukuro, K.; Seimiya, H.; Shin-ya, K.; Nagasawa, K. Org. Biomol. Chem. 2010, 8, 2749−2755. (10) Membrino, A.; Paramasivam, M.; Cogoi, S.; Alzeer, J.; Luedtke, N. W.; Xodo, L. E. Chem. Commun. 2010, 46, 625−627. (11) Drygin, D.; Jain, A. S.; O’Brien, S.; Schwaebe, M.; Lin, A.; Bliesath, J.; Ho, C. B.; Proffitt, C.; Trent, K.; Whitten, J. P.; Lim, J. K. C.; Hoff, D. V.; Anderes, K.; Rice, W. G. Cancer Res. 2009, 69, 7653− 7661. (12) Griffin, L. C.; Tidmarsh, G. F.; Bock, L. C.; Toole, J. J.; Leung, L. L. Blood 1993, 81, 3271−3276. (13) Cunningham, J. C.; Brenes, N. J.; Crooks, R. M. Anal. Chem. 2014, 86, 6166−6170. (14) Heydari-Bafrooei, E.; Amini, M.; Ardakani, M. H. Biosens. Bioelectron. 2016, 85, 828−836. (15) Wang, G. L.; Hu, X. L.; Wu, X. M.; Dong, Y. M.; Li, Z. J. Microchim. Acta 2016, 183, 765−771. (16) Hu, P.; Han, L.; Zhu, C. Z.; Dong, S. J. Chem. Commun. 2013, 49, 1705−1707. (17) Tian, Y.; Stepanenko, V.; Kaiser, T. E.; Würthner, F.; Scheblykin, I. G. Nanoscale 2012, 4, 218−223. (18) Würthner, F.; Kaiser, T. E.; Saha-MÖ ller, C. R. Angew. Chem., Int. Ed. 2011, 50, 3376−3410. (19) Gai, W.; Yang, Q. F.; Xiang, J. F.; Jiang, W.; Li, Q.; Sun, H. X.; Yu, L. J.; Shang, Q.; Guan, A. J.; Zhang, H.; Tang, Y. L. Analyst 2013, 138, 798−804. (20) Shim, J. W.; Tan, Q. L.; Gu, L. Q. Nucleic Acids Res. 2009, 37, 972−982. (21) Hamer, F. M. The Cyanine Dyes and Related Compounds; Interscience Publishers: New York, 1964. (22) Brooker, L. G. S.; White, F. L. J. Am. Chem. Soc. 1935, 57, 547− 551. (23) Shi, Y. H.; Sun, H. X.; Xiang, J. F.; Yu, L. J.; Yang, Q. F.; Li, Q.; Guan, A. J.; Tang, Y. L. Anal. Chim. Acta 2015, 857, 79−84. (24) Jie, G. F.; Yuan, J. X. Anal. Chem. 2012, 84, 2811−2817. (25) Gao, X.; Liu, X. C.; Lin, Z. H.; Liu, S. Y.; Su, X. G. Analyst 2012, 137, 5620−5624. (26) Nagatoishi, S.; Tanaka, Y.; Tsumoto, K. Biochem. Biophys. Res. Commun. 2007, 352, 812−817. (27) Gai, W.; Yang, Q. F.; Xiang, J. F.; Yu, L. J.; Guan, A. J.; Li, Q.; Sun, H. X.; Shang, Q.; Jiang, W.; Zhang, H.; Liu, Y.; Wang, L. X.; Tang, Y. L. Phys. Chem. Chem. Phys. 2013, 15, 5758−5761. (28) Russo Krauss, I.; Merlino, A.; Giancola, C.; Randazzo, A.; Mazzarella, L.; Sica, F. Nucleic Acids Res. 2011, 39, 7858−7867. (29) De Tito, S.; Morvan, F.; Meyer, A.; Vasseur, J.-J.; Cummaro, A.; Petraccone, L.; Pagano, B.; Novellino, E.; Randazzo, A.; Giancola, C.; Montesarchio, D. Bioconjugate Chem. 2013, 24, 1917−1927.

Table 1. Results of Thrombin Determination sample

added (pM)

founda (pM)

recovery (%)

RSD (%, n = 3)

1 2 3 4

2.0 3.0 4.0 5.0

2.00 3.04 3.89 5.06

100.35 101.43 97.41 101.13

0.80 0.62 0.86 1.26

a

Mean value of 3 replicates.

■

CONCLUSION In summary, taking advantage of the unique properties of the TBA with high specificity and affinity toward thrombin and the chiral assembly of DMSB is very sensitive to the slight difference of TBA G-quadruplexes structure induced by thrombin and K+, we reported, for the first time, a simple and ultrasensitive approach for thrombin detection. This method exhibited excellent sensitivity and ultrahigh selectivity and thus also showed promising practical applications. Moreover, in comparison, the present method is much simpler, easier to use, and less expensive as it only requires the cyanine dye DMSB as a signal reporter. This research represents a new application of supramolecular assembly and also provides a new insight toward the development of thrombin detection.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.6b04247. CD and UV−vis spectra of TBA in different buffer solutions; CD titration experiments; summarized different methods for determination of thrombin (Figures S1− S5 and Table S1) (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*Phone: 0086-10-6256-3101. Fax: 0086-10-8261-7302. E-mail: [email protected]. *Phone: 0086-10-6252-2090. Fax: 0086-10-8261-7302. Email:[email protected]. *Phone: 0086-10-6252-2090. Fax: 0086-10-8261-7302. E-mail: [email protected]. ORCID

Yalin Tang: 0000-0002-5325-2602 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Major National Basic Research Projects (973) (Grant Number 2013CB733701), the National Natural Science Foundation of China (Grant Numbers 21205121, 21472197, 21675162, and 21602226), and Strategic Priority Research Program of the Chinese Academy of Sciences (Grant Number XDA09030307). The measurements of CD and UV spectra were performed at the Center for Physicochemical Analysis and Measurements in ICCAS. The help from Liping Ding and Yuan Xu were acknowledged.

■

REFERENCES

(1) Parkinson, G. N.; Lee, M. P.; Neidle, S. Nature 2002, 417, 876− 880. 551

DOI: 10.1021/acs.analchem.6b04247 Anal. Chem. 2017, 89, 548−551