- Email: [email protected]
Target-protecting dumbbell molecular probe against exonucleases digestion for sensitive detection of ATP and streptavidin
Biosensors and Bioelectronics 83 (2016) 221–228
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Target-protecting dumbbell molecular probe against exonucleases digestion for sensitive detection of ATP and streptavidin Jinyang Chen, Yucheng Liu, Xinghu Ji, Zhike He n Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education), College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072, China
art ic l e i nf o
a b s t r a c t
Article history: Received 24 February 2016 Received in revised form 3 April 2016 Accepted 18 April 2016 Available online 22 April 2016
In this work, a versatile dumbbell molecular (DM) probe was designed and employed in the sensitively homogeneous bioassay. In the presence of target molecule, the DM probe was protected from the digestion of exonucleases. Subsequently, the protected DM probe specifically bound to the intercalation dye and resulted in obvious fluorescence signal which was used to determine the target molecule in return. This design allows specific and versatile detection of diverse targets with easy operation and no sophisticated fluorescence labeling. Integrating the idea of target-protecting DM probe with adenosine triphosphate (ATP) involved ligation reaction, the DM probe with 5′-end phosphorylation was successfully constructed for ATP detection, and the limitation of detection was found to be 4.8 pM. Thanks to its excellent selectivity and sensitivity, this sensing strategy was used to detect ATP spiked in human serum as well as cellular ATP. Moreover, the proposed strategy was also applied in the visual detection of ATP in droplet-based microfluidic platform with satisfactory results. Similarly, combining the principle of targetprotecting DM probe with streptavidin (SA)-biotin interaction, the DM probe with 3′-end biotinylation was developed for selective and sensitive SA determination, which demonstrated the robustness and versatility of this design. & 2016 Elsevier B.V. All rights reserved.
Keywords: Dumbbell molecular probe Fluorescence detection Droplet microfluidics DNA ATP Streptavidin
1. Introduction As an important small molecule in life activities, adenosine-5′triphosphate (ATP) is a multifunctional nucleotide, which is not only a universal energy source, but also an extracellular signaling mediator and involved in many biological processes including DNA replication, membrane ion-channel pump, biosynthesis, hormonal and neuronal activities (Abraham et al., 1997; Newman et al., 1997; Szewczyk et al., 1998). It has also been used as an indicator of living organisms for cell viability and cell injury (Eguchi et al., 1997). Therefore, the highly sensitive and selective detection of ATP is essential for biochemical research, food quality control, environmental analysis, as well as clinical diagnosis. Many methods based on host-guest receptors, peptides, conjugated polymers, DNA/RNA aptamers, and ATP-dependent ligation reactions have been developed for ATP detection (Li et al., 2005; Lin et al., 2014; Lu et al., 2010, 2011; McCleskey et al., 2003; Mizukami et al., 2002; Zhou et al., 2011). Among these reported strategies, some of them exhibit only moderate sensitivity with detection limits for ATP in the micromolar or nanomolar range. In addition, even though n
Corresponding author. E-mail address: [email protected] (Z. He).
http://dx.doi.org/10.1016/j.bios.2016.04.055 0956-5663/& 2016 Elsevier B.V. All rights reserved.
some methods have shown desirable analytical performance, they usually require pre-labeling of a signal source, which needs considerable time-consumption and may suffer from higher cost. The investigation of small molecule-protein interaction is of great significance in unveiling the mystery of cell development involving small molecules, as well as in the drug discovery and molecular diagnostics and therapeutics (Gao et al., 2004; Howitz et al., 2003; Overington et al., 2006; Stockwell, 2004). There are numerous analytical methods available for the assays of small molecules or their protein receptors, including capillary electrophoresis, affinity chromatography, surface plasmon resonance (SPR), fluorescence resonant energy transfer, and fluorescence anisotropy (Bachovchin et al., 2009; Drabovich et al., 2009; Goldman et al., 2005; Mano et al., 2006; Petrov et al., 2005; Wear et al., 2005). Typical of many molecule-protein interactions, the binding of biotin to streptavidin (SA) has been the subject of considerable fundamental and applied interest. To study the SA-biotin interaction, a novel method called terminal protection of small moleculelinked DNA was developed, which translated the small moleculeprotein interaction into the detection of DNA (Wu et al., 2009). In view of the unique characteristics of DNA, such as specificity, stability, sequence coding, and the assists of various enzymes as well as enormous achievements of DNA sensors (Li et al., 2010; Zhang et al., 2013; Zhao et al., 2015), the strategy of terminal
222
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
protection of small molecule-linked DNA opened a new window for studying the small molecule-protein interaction, and has been applied in numerous bioassays (Cao et al., 2012; He et al., 2013; Wang et al., 2013; Wu et al., 2011; Zhou et al., 2013). Inspired by the idea of building the bridge of DNA detection for other molecules sensing, and based on our previous efforts in small molecule and protein assays (Chen et al., 2014a, 2016), a versatile dumbbell molecular (DM) probe was designed for multiplex and sensitive bioassays. In the presence of target analyte, the DM probe was protected from the degradation of exonucleases, consequently exhibiting obvious fluorescence response resulted from the protected DM probe bound to intercalation dye. United with the ATP dependent ligation reaction, the DM probe with 5′-end phosphorylation was constructed for sensitive detection of ATP which was chosen as target model of small molecule. Changing the same DM probe to the one with 3′-end biotinylation, the strategy of target-protecting DM probe against exonucleases digestion was also employed for the sensing of protein SA based on the specific SA-biotin interaction. In addition to the versatility and compatibility of the proposed strategy, the results indicated that both of the detections of ATP and SA were of high sensitivity and selectivity, as well as excellent resistance to matrix interferences. What's more, benefiting from the user-friendly control, this sensing strategy was also applied in the visual colorimetric analysis in droplet-based microfluidic platform.
2. Experimental 2.1. Materials and reagents All oligonucleotide with different sequences were synthesized and HPLC purified by Sangon Biotechnology Co., Ltd (Shanghai, China). The sequences of the oligonucleotide used in this work are listed in Table S1 (Supplementary information). Exonuclease I (Exo I), Exonuclease III (Exo III), and Shrimp alkaline phosphatase (SAP) were purchased from the Takara Biotechnology Co., Ltd. (Dalian, China). T4 DNA ligase was purchased from Thermo Fisher Scientific. SYBR Green I (SG I) (10,000 ), Streptavidin (SA), adenosine triphosphate (ATP) and its analogues were obtained from SigmaAldrich. Human serum sample was supplied by the Zhongnan Hospital of Wuhan University (Wuhan, China). Cell culture dishes were was obtained from NEST Biotechnology (Beijing, China). Culture medium and fetal bovine serum (FBS) were purchased from Gibco BRL (Grand Island, NY, USA). The silicon wafer was purchased from Institute of Microelectronics of Chinese Academy of Sciences. The poly(dimethylsiloxane) (PDMS) and AZ 50XT photoresist were obtained from RTV615 GE Toshiba Silicones Co., Ltd. and AZ Electronic Materials USA Corp., respectively. All chemical reagents were of analytical grade and used without further purification. All solutions were prepared with ultrapure water (18.25 MΩ cm) from a Millipore system.
solution was diluted to 400 μL with Tris–HCl buffer and the fluorescence spectrum was obtained with a RF-5301PC spectrophotometer (Shimadzu, Japan) equipped with a 150 W xenon lamp (Ushio Inc, Japan). 2.3. Agarose gel electrophoresis After the ligation and digestion reactions mentioned above, the mixture was stained with 100 SG I (1 μL). Three percent of agarose gel was prepared using 1 TAE buffer (40 mM Tris-AcOH, 2 mM Na2EDTA, pH ¼8.5). The electrophoresis was carried out at 100 V for approximately 40 min in 1 TAE buffer. Then the gel was imaged by a ChemiDoc XRD system (Bio-Rad). 2.4. Cellular ATP assay HepG2 cells were maintained in Dulbecco's modified eagle's medium (DMEM) supplemented with 5% fetal bovine serum (FBS), 10 mM Hepes, antibiotics (50 U/mL pencillin and 50 μg/mL streptomycin), and 2 mM glutamine at 37 °C under 5% CO2 atmosphere. The collected cells were centrifuged at 3000 rpm for 3 min at 4 °C and washed three times by PBS buffer (20 mM phosphate, 100 mM NaCl, and 5 mM KCl, pH ¼7.4) and then suspended in 25 mM Tris– HCl buffer (100 mM NaNO3, 6 mM magnesium acetate, pH ¼7.6). Cell lysis was performed by repeated cycles of freezing and thawing, and then the lysed cells were ready for ATP assays. For control experiment, the above HepG2 cell lysate (approximately 20,000 cells/mL, 400 μL) was treated with 10 units SAP at 37 °C for 20 min to remove the ATP. 2.5. Visual detection of ATP in droplet platform The droplet microchip designed and fabricated according to the previous protocol (Chen et al., 2015b). For ATP assay, the DM probe 1 was sealed by adding 8 μL of T4 DNA ligase (5 U/μL), 5 μL different concentrations of ATP and 17 μL of Tris–HCl buffer into the 10 μL of 5 μM DM probe 1 solution and allowing the ATP-triggered ligation reaction at room temperature for 30 min. After that, 5 μL of Exo I (5 U/μL) and 5 μL of Exo III (200 U/μL) were added into the mixture solution to induce the digestion for 60 min. After that, the mixture solution was delivered from the sample container into the microchip for droplets generation by using the technique previously reported (Chen et al., 2015a). Finally, the droplets containing the protected DM probe 1 were mixed with the SG I reagent through the droplet dosing strategy. The fluorescence images were recorded by an inverted fluorescence microscope (Axio Observer.A1, Zeiss, Germany) in conjunction with a light source system (excitation filter was set as 490 nm) and a Spot RT3 chargecoupled device (CCD, Diagnostic Instruments, Inc., USA). The fluorescence intensities of the droplets were obtained by ImagePro Plus software.
2.2. Procedures for ATP assay 2.6. Procedures for SA detection In a typical procedure, first, the DM probe 1 was diluted with 25 mM Tris–HCl buffer (100 mM NaCl, 5 mM MgCl2, 1 mM DTT, pH ¼7.6) and denatured at 95 °C for 10 min followed with a slow annealing treatment for 1 h before use. Then the DM probe 1 was sealed by adding 4 μL of T4 DNA ligase (5 U/μL), 5 μL different concentrations of ATP and 100 μL of Tris–HCl buffer into the 10 μL of 0.1 μM DM probe 1 solution and allowing the ATP-triggered ligation reaction at room temperature for 30 min. After that, 4 μL of Exo I (5 U/μL) and 10 μL of Exo III (20 U/μL) were added into the mixture solution to induce the digestion for 60 min. After that, 2 μL of SG I (50 ) was added into the solution. Finally, the
First, the DM probe 2 was diluted with 25 mM Tris–HCl buffer (100 mM NaCl, 5 mM MgCl2, 1 mM DTT, pH ¼7.6). 10 μL of 0.2 μM DM probe 2 was mixed with 5 μL different concentrations of SA and followed by adding 100 μL Tris–HCl buffer. The solution was incubated in room temperature for 30 min Then, 4 μL of Exo I (5 U/ μL) and 10 μL of Exo III (20 U/μL) were added into the mixture solution to induce the digestion for 60 min. After the digestion reaction, 2 μL of SG I (50 ) was added into the solution. Finally, the solution was diluted to 400 μL with Tris–HCl buffer and the fluorescence intensity was measured.
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
223
3. Results and discussion 3.1. Probe design and detection principle In this work, a dumbbell-shaped DNA molecule was designed as the probe for the detections of ATP and SA. This DM probe consists of two cytosine hairpin loops and a double-helical stem which is closed by the loops. In subsequent sensing applications, the fluorescence signal mainly results from the stem of the DM probe after binding with an intercalation dye, SG I. So the sequence and length of the stem should be carefully evaluated. First, the dumbbell DNA molecules with different sequences of stems were stained with SG I to obtain the fluorescence signals. It is shown that the fluorescence intensity of the dumbbell DNA with AT-rich stem stained by SG I is stronger than those of dumbbell DNA with CG-rich stem and dumbbell DNA with random sequence stem (Fig. 1A), which is also consistent with previous report (Zipper et al., 2004). This result indicates that the DM (AT-rich) is more suitable employed as the probe in the subsequent assay compared with DM (CG-rich) and DM (random). Moreover, the length of AT pairs in the stem of dumbbell DNA molecule was evaluated. As shown in Fig. 1B, the fluorescence intensity of the dumbbell DNA stained by SG I dramatically increased with the increase of the length of AT pairs range from 16 to 32. It is noted that the increase rate of the fluorescence intensity became slow when the length of AT pairs reached to 24. What's more, the DNA cost also obviously increased with the increase of the stem length of the DM probe. In consideration of the enhancement of fluorescence intensity and the experiment cost, the dumbbell DNA containing 24 AT pairs in the stem was chosen as the most suitable DM probe in this work. After that the optimal design of DM probe was obtained, the versatile sensing strategy for sensitive detection of different biomolecules was presented. For ATP detection, the DM probe was phosphorylated in its 5′-end, which was called as DM probe 1. The principle of ATP detection is shown in Scheme 1A. In the absence of cofactor ATP, the ligation reaction could not be initiated by T4 DNA ligase. So the DM probe 1 was completely digested by Exo I and Exo III and no double-helical stems bound to SG I, which provided a low background for the sensing system. However, in the presence of ATP, the DM probe 1 was sealed as close structure that could resist the digestion by Exo I and Exo III due to the T4 ligase-catalyzed ligation reaction. As a result, the protected DM probe 1 was easily stained with SG I, and a strong fluorescence signal was obtained. For SA detection, the DM probe was biotinylated in its 3′-end, which was called as DM probe 2. The schematic demonstration of SA detection was shown in Scheme 1B. In
Scheme 1. Schematic illustration of the mechanism for the sensitive detection of ATP (A) and SA (B).
the absence of SA, the DM probe 2 without any protections was degraded by exonucleases, resulting in a low background signal. Whereas, when SA was added to the solution, it bound to the biotin modified in the 3′ end of DM probe 2 due to the specific SAbiotin interaction. The steric hindrance of protein SA prevents exonucleases from approaching and cleaving the phosphodiester bond adjacent to the 3′ terminus. So the DM probe 2 was protected from the degradation by exonucleases. Consequently, an obvious fluorescence response was detected when the protected DM probe 2 bound to SG I.
Fig. 1. (A) Fluorescence spectra of the DM (AT-rich), DM (CG-rich) and DM (random) stained by SG I, respectively. (B) Fluorescence spectra of DM (AT16), DM (AT20), DM (AT24), DM (AT28), DM (AT32) stained by SG I, respectively.
224
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
Fig. 2. (A) Fluorescence spectra of the system under different conditions: (a) DM probe 1 þExo I þ Exo III; (b) DM probe 1þ T4 ligase þ Exo Iþ Exo III; (c) DM probe 1 þT4 ligase þ 2 nM ATP þExo Iþ Exo III. (B) Agarose gel electrophoresis of the products of different experiments. Lane M: DNA ladder; lane 1: DM probe 1; lane 2: DM probe 1 þT4 ligase; lane 3: DM probe 1þExo I þ Exo III; lane 4: DM probe 1þ T4 ligase þ Exo I þ Exo III; lane 5: DM probe 1þ T4 ligaseþ ATP þ Exo Iþ Exo III.
3.2. Feasibility of ATP detection To verify the designed strategy, the feasible experiment was carried out and the result is illustrated in Fig. 2A. In the absence of ATP, no obvious fluorescence emission was detected, which was almost the same as the fluorescence signal in the condition that DM probe 1 was completely digested by Exo I and Exo III. It is revealed that this sensing strategy possesses a highly reduced background. In contrast, a significant fluorescence emission was observed in the presence of 2 nM ATP. These results suggest that the proposed method is feasible for ATP detection. In order to further confirm the feasibility of the proposed strategy, the experiment of agarose gel electrophoresis was carried out. As shown in Fig. 2B, in the absence of Exo I and Exo III, a strong fluorescence band was easily identified when the solution contained only DM probe 1 (lane 1) or DM probe 1 and T4 ligase (lane 2). Further, when Exo I and Exo III were added in the solutions mentioned above, no obvious bands were obtained (lane 3, 4). It is revealed that the unprotected DM probe 1 could be completely degraded by the exonucleases due to the failed ligation reaction. However, when the solution contained DM probe 1, ATP, T4 ligase, and exonucleases, the DM probe 1 could be sealed by T4 ligation reaction and resistant to the digestion of exonucleases. As expected, an evident band was observed in lane 5, and the band indicated the same mobility with that in the control experiments (lane 1, 2). The above results successfully verified that the proposed strategy was entirely feasible for the ATP detection. For ATP detection, the digestion procedures of exonucleases play vital roles in the performance of the sensing system, because they are in direct relation to the background signal. So, the effects of the exonucleases concentrations and digestion time on the background signal were evaluated first. The optimization experiments were carried out in the absence of ATP, and the exonucleases concentrations and digestion time were defined as optimum value when the background signal reached to the minimum. Through the successive investigations, the optimal concentrations of Exo I and Exo III were 0.05 and 0.5 U/μL, respectively (Fig. S1, 2). The digestion time was fixed at 60 min (Fig. S3). Besides, as key factors for fluorescence enhancement in the presence of ATP, the T4 ligase concentration and ligation time were optimized as well. The optimal concentration of T4 ligase was 0.05 U/μL (Fig. S4), and the optimal ligation time was 30 min (Fig. S5). Because both of the digestion reaction and ligation reaction were carried out at room temperature in purpose to simplify the experiments. So it is
reasonable infer that the digestion and ligation times were shorter when the experiments were performed at the optimum temperature of enzyme reactions, such as 37 °C. Moreover, the effect of SG I concentration on the fluorescence signal was also investigated, and the optimal concentration was fixed at 50 in subsequent experiments (Fig. S6). 3.3. Performance of ATP detection Under the optimized conditions discussed above, the emission titration experiment with increasing concentrations of ATP was carried out, and the linear response range of the proposed strategy for ATP detection was evaluated. As shown in Fig. 3A, the fluorescence intensity of the protected DM probe 1 stained with SG I dramatically increased with the increase of the ATP concentration. The fluorescence intensity versus ATP concentration is plotted in the range from 10 to 4000 pM (Fig. 3B). As shown in Fig. 3C, there is a good linear relationship (R2 ¼0.9926) between the fluorescence intensities and the concentrations of ATP with a detection limit of 4.8 pM (signal-to-noise ratio of 3). To the best of our knowledge, this detection limit is superior to those of previously reported fluorescence sensing strategies, and even can be comparable with or exceed those of some previously reported electrochemical methods (Table S2). Among the methods listed in Table S2, the detection limit of the electrochemical sensor based on multiple signal amplification strategy is 0.5 pM, which is lower than that of this work (Chen et al., 2014b). However, the construction of the electrochemical sensor involves tedious procedures of nanocomposite preparation, aptamer conjugation, and magnetic separation. By contrast, we put forward a simple, rapid and homogeneous method for ATP detection in this work. The high sensitivity of this proposed strategy is mainly attributed to the low background signal caused by exonucleases degradation as well as the strong fluorescence response of the protected DM probe 1 stained with SG I. Selectivity is a critical parameter to assess the performance of a sensing strategy. In this work, the selectivity of this strategy has been evaluated by comparing the fluorescence signal of samples containing ATP with those of its analogues, including cytidine triphosphate (CTP), guanosine triphosphate (GTP), uridine triphosphate (UTP), and adenosine diphosphate (ADP). As shown in Fig. 3D, when ATP, CTP, GTP, UTP, and ADP were added to the sensing system, respectively, only ATP caused a marked fluorescence enhancement. However, the others made no obvious
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
225
Fig. 3. (A) Fluorescence spectra of the system in response to different concentrations of ATP. (B) Relationship between the fluorescence intensity and ATP concentration. (C) Calibration curve for ATP detection. (D) Selectivity of the developed sensing strategy for ATP compared to ATP analogues and blank.
changes in fluorescence intensity, even though the concentration of the analogues was one hundred times as that of ATP. This desirable selectivity of the method mainly benefited from the high dependence of T4 DNA ligase for the ATP cofactor. 3.4. Real sample analysis To evaluate the detection toward the target in complex biological matrix, the proposed strategy was performed to detect ATP in human serum samples. A serious of ATP standard solutions at 0, 0.25, 1.0, 1.5, and 2.0 nM were separately spiked into the 10% diluted human serum and then the samples were tested. As shown
in Fig. 4A, the fluorescence signals detected in the diluted human serum had slightly differences with those obtained in buffer. Despite these differences, the presence of ATP in human serum still could be detected with this developed method. It is demonstrated that the ATP assay can be performed in complex biological environment with this sensing strategy. Therefore, the ATP assay in real sample was further carried out through the detection of ATP in cancer cells (HepG2 cells). As illustrated in Fig. 4B, in the absence of cell lysates, no obvious fluorescence emission was observed (curve a). However, in the presence of lysates from different numbers of HepG2 cells, all the obvious fluorescence signals were obtained (curve b, c, and d). Furthermore, when the cell lysates
Fig. 4. (A) Fluorescence responses of the sensing platform for ATP detection in buffer and 10% diluted human serum samples, respectively. (B) Fluorescence responses of the system in the absence of freshly lysed cells (a), and in the presence of freshly lysed cells with different concentrations: approximately 15,000 cells/mL (b), 30,000 cells/mL (c), 60,000 cells/mL (d), as well as in the presence of SAP-treated lysed cells (60,000 cells/mL) (e).
226
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
Fig. 5. (A) Schematic illustration of visual detection of ATP in droplet-based platform. Fluorescence responses of the droplets in the absence (B) and presence (C) of ATP. (D) Fluorescence responses of the droplets with the different concentrations of ATP. Scale bars indicate 100 mm.(For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
were treated with shrimp alkaline phosphatase (SAP) to remove the ATP, almost no significant fluorescence enhancement was observed as anticipated (curve e). The above results clearly showed the robustness of the proposed method in real sample analysis. 3.5. Visual detection of ATP in droplet-based platform In our previous work, to make the operation of droplet-based analysis more convenient and efficient, we presented a novel droplet dosing strategy and a flexible sample-introduction technique in microfluidic system (Chen et al., 2015a, 2015b). Integrating these previously reported techniques with the sensing strategy developed in this work, the visual detection of ATP was successfully realized in droplet-based microfluidic platform. The schematic illustration of visual detection of ATP based on droplet platform is illustrated in Fig. 5A. First, the ATP-dependent ligation reaction and exonucleases-involved degradation were successively carried out in the centrifuge tube. Then the solution of mixtures was directly delivered into the microchannel of chip from the centrifuge tube and generated as numerous droplets. Based on the dosing strategy previously reported, the intercalation dye (SG I) was added into the droplets (Chen et al., 2015b). So the protected DM probe 1 contained in the droplet was stained with SG I and resulted in an obvious fluorescence response. The qualitative analysis of ATP was achieved from the fluorescence image of droplet with the naked eye, and the quantitative analysis was performed by colorimetric method. As shown in Fig. 5B, when the ATP was absent, the droplets demonstrated as dark color in the fluorescence images, so no obvious fluorescence response was obtained. While in the presence of ATP, the droplets presented bright green color in the fluorescence images, which was easily discriminated from that color of blank droplets with the naked eye (Fig. 5C). To evaluate the analytical performance of visual detection of ATP, a serious of standard samples containing various
concentrations of ATP were detected in the droplet-based platform. As shown in Fig. 5D, the green color of the droplet in fluorescence image turned brighter and brighter as the increase of ATP concentration in the range from 25 to 1000 nM, and did not change obviously when the concentration increased to 1250 nM. What's more, as indicated in Fig. S7, the fluorescence intensity obtained from droplet image showed a clear linear relationship with the ATP concentration over the range from 25 to1000 nM (R2 ¼0.9932). The detection limit was calculated as 8.9 nM (signalto-noise ratio of 3), which is about 2 amol per droplet. To our knowledge, the sensitivity of this visual detection can be comparable with or exceeds those of previously reported strategies with fluorescent spectrometry (Table S2). 3.6. Sensitive and selective detection of SA The proposed strategy was also extended to the protein detection by integrating the idea of target-protecting DM probe against exonucleases degradation with SA-biotin interaction (Scheme 1B). The fluorescence responses of the sensor to target SA at different concentrations are illustrated in Fig. 6A, B, from which it can be seen that the fluorescence intensity increased significantly with the increase of SA concentration. There is a good linear relationship (R2 ¼ 0.9903) between the fluorescence intensities and the concentrations of SA in the range from 50 to 4000 pM (Fig. 6C). The detection limit was determined to be 12.7 pM (signal-to-noise ratio of 3), which is suitable to serve as a biosensor for detecting SA. To investigate the selectivity of the strategy, 2 nM of SA and 20 nM of other proteins were tested with this biosensor. As shown in Fig. 6D, the results showed that only SA could significantly enhance the fluorescence intensity of the protected DM probe 2 stained with SG I. Whereas no significant change in emission intensity was observed upon the addition of the other proteins, such as thrombin, trypsin, lysozyme, bovine serum albumin (BSA), and pepsin, even though the concentration
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
227
Fig. 6. (A) Fluorescence spectra of the system in response to different concentrations of SA. (B) Relationship between the fluorescence intensity and SA concentration. (C) Calibration curve for SA detection. (D) Selectivity of the developed sensing strategy for SA compared to other proteins and blank. The concentration of SA is 2 nM, while the concentrations of other proteins are 20 nM.
of the other proteins was ten times as that of SA. This high selectivity was mainly attributed to the specific interaction of SAbiotin. It is notable that the proposed strategy could be extended to detect different proteins just by modifying the DM probe with corresponding small molecule, which demonstrates the versatility and generality of this design.
4. Conclusion In summary, a dumbbell-shaped DNA probe was designed and successfully applied in the homogeneous bioassays based on the strategy of target-protecting DM probe against exonucleases degradation. This strategy is simple but versatile, which allows sensitive detection of diverse targets such as ATP and SA. The advantages of this strategy are as follows. Apart from the phosphorylation of DM probe for ATP assay or the biotinylation for SA detection, neither laborious and time-consuming fluorescence labeling, nor tedious separation steps were required in this method. In addition, the fluorescence signal readout was carried out by staining the protected DM probe with SG I, which is easy and fast to realize. The proposed strategy also possesses significantly high selectivity and satisfactory sensitivity which allows both of the detection limits for small molecule and protein in picomolar level. Benefiting from the excellent detection performance, the method was successfully employed in the detection in complex environment as well as real sample analysis. To broaden the application of the developed strategy, the visual detection of ATP was also accomplished by integrating the strategy with droplet-based microfluidics. And the detection limit still could reach to nanomolar
level just by the naked eye. In a word, all the results as mentioned above clearly indicate the robustness and versatility of the proposed strategy.
Acknowledgments This work was supported by the National Natural Science Foundation of China (21275109, 21205089, 21475101), the Suzhou Nanotechnology Special Project (ZXG2013028).
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2016.04.055.
References Abraham, E.H., Okunieff, P., Scala, S., Vos, P., Oosterveld, M.J., Chen, A.Y., Shrivastav, B., 1997. Science 275, 1324–1326. Bachovchin, D.A., Brown, S.J., Rosen, H., Cravatt, B.F., 2009. Nat. Biotechnol. 27, 387–394. Cao, Y., Zhu, S., Yu, J., Zhu, X., Yin, Y., Li, G., 2012. Anal. Chem. 84, 4314–4320. Chen, C., Xiang, X., Liu, Y., Zhou, G., Ji, X., He, Z., 2014a. Biosens. Bioelectron. 58, 205–208. Chen, J., Ji, X., He, Z., 2015a. Anal. Chim. Acta 888, 110–117. Chen, J., Liu, Y., Ye, T., Xiang, X., Ji, X., He, Z., 2015b. Sens. Actuator B-Chem. 215, 206–214. Chen, J., Ji, X., Tinnefeld, P., He, Z., 2016. ACS Appl. Mater. Interfaces 8, 1786–1794. Chen, X., Ge, L., Guo, B., Yan, M., Hao, N., Xu, L., 2014b. Biosens. Bioelectron. 58, 48–56.
228
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
Drabovich, A.P., Berezovski, M.V., Musheev, M.U., Krylov, S.N., 2009. Anal. Chem. 81, 490–494. Eguchi, Y., Shimizu, S., Tsujimoto, Y., 1997. Cancer Res. 57, 1835–1840. Gao, X., Cui, Y., Levenson, R.M., Chung, L.W., Nie, S., 2004. Nat. Biotechnol. 22, 969–976. Goldman, E.R., Medintz, I.L., Whitley, J.L., Hayhurst, A., Clapp, A.R., Uyeda, H.T., Deschamps, J.R., Lassman, M.E., Mattoussi, H., 2005. J. Am. Chem. Soc. 127, 6744–6751. He, Y., Xing, X., Tang, H., Pang, D., 2013. Small 9, 2097–2101. Howitz, K.T., Bitterman, K.J., Cohen, H.Y., Lamming, D.W., Lavu, S., Wood, J.G., Zipkin, R.E., Chung, P., Kisielewski, A., Zhang, L.L., Scherer, B., Sinclair, D.A., 2003. Nature 425, 191–196. Li, C., Numata, M., Takeuchi, M., Shinkai, S., 2005. Angew. Chem. Int. Ed. 44, 6371–6374. Li, D., Song, S., Fan, C., 2010. Acc. Chem. Res. 43, 631–641. Lin, C., Cai, Z., Wang, Y., Zhu, Z., Yang, C.J., Chen, X., 2014. Anal. Chem. 86, 6758–6762. Lu, C.H., Li, J., Lin, M.H., Wang, Y.W., Yang, H.H., Chen, X., Chen, G.N., 2010. Angew. Chem. Int. Ed. 49, 8454–8457. Lu, L.M., Zhang, X.B., Kong, R.M., Yang, B., Tan, W., 2011. J. Am. Chem. Soc. 133, 11686–11691. Mano, N., Sato, K., Goto, J., 2006. Anal. Chem. 78, 4668–4675. McCleskey, S.C., Griffin, M.J., Schneider, S.E., McDevitt, J.T., Anslyn, E.V., 2003. J. Am.
Chem. Soc. 125, 1114–1115. Mizukami, S., Nagano, T., Urano, Y., Odani, A., Kikuchi, K., 2002. J. Am. Chem. Soc. 124, 3920–3925. Newman, E.A., Zahs, K.R., 1997. Science 275, 844–847. Overington, J.P., Al-Lazikani, B., Hopkins, A.L., 2006. Nat. Rev. Drug. Discov. 5, 993–996. Petrov, A., Okhonin, V., Berezovski, M., Krylov, S.N., 2005. J. Am. Chem. Soc. 127, 17104–17110. Stockwell, B.R., 2004. Nature 432, 846–854. Szewczyk, A., Piku̵ a, S., 1998. Biochim. Biophys. Acta 1365, 333–353. Wang, G., He, X., Wang, L., Zhang, X., 2013. Biosens. Bioelectron. 42, 337–341. Wear, M.A., Patterson, A., Malone, K., Dunsmore, C., Turner, N.J., Walkinshaw, M.D., 2005. Anal. Biochem. 345, 214–226. Wu, Z., Zhen, Z., Jiang, J.H., Shen, G.L., Yu, R.Q., 2009. J. Am. Chem. Soc. 131, 12325–12332. Wu, Z., Wang, H., Guo, M., Tang, L.J., Yu, R.Q., Jiang, J.H., 2011. Anal. Chem. 83, 3104–3111. Zhang, H., Li, F., Dever, B., Li, X.F., Le, X.C., 2013. Chem. Rev. 113, 2812–2841. Zhao, Y., Chen, F., Li, Q., Wang, L., Fan, C., 2015. Chem. Rev. 115, 12491–12545. Zhou, G., Zhang, X., Ji, X., He, Z., 2013. Chem. Commun. 49, 8854–8856. Zhou, Z., Du, Y., Dong, S., 2011. Anal. Chem. 83, 5122–5127. Zipper, H., Brunner, H., Bernhagen, J., Vitzthum, F., 2004. Nucleic Acids Res. 32, e103.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Target-protecting dumbbell molecular probe against exonucleases digestion for sensitive detection of ATP and streptavidin Jinyang Chen, Yucheng Liu, Xinghu Ji, Zhike He n Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education), College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072, China
art ic l e i nf o
a b s t r a c t
Article history: Received 24 February 2016 Received in revised form 3 April 2016 Accepted 18 April 2016 Available online 22 April 2016
In this work, a versatile dumbbell molecular (DM) probe was designed and employed in the sensitively homogeneous bioassay. In the presence of target molecule, the DM probe was protected from the digestion of exonucleases. Subsequently, the protected DM probe specifically bound to the intercalation dye and resulted in obvious fluorescence signal which was used to determine the target molecule in return. This design allows specific and versatile detection of diverse targets with easy operation and no sophisticated fluorescence labeling. Integrating the idea of target-protecting DM probe with adenosine triphosphate (ATP) involved ligation reaction, the DM probe with 5′-end phosphorylation was successfully constructed for ATP detection, and the limitation of detection was found to be 4.8 pM. Thanks to its excellent selectivity and sensitivity, this sensing strategy was used to detect ATP spiked in human serum as well as cellular ATP. Moreover, the proposed strategy was also applied in the visual detection of ATP in droplet-based microfluidic platform with satisfactory results. Similarly, combining the principle of targetprotecting DM probe with streptavidin (SA)-biotin interaction, the DM probe with 3′-end biotinylation was developed for selective and sensitive SA determination, which demonstrated the robustness and versatility of this design. & 2016 Elsevier B.V. All rights reserved.
Keywords: Dumbbell molecular probe Fluorescence detection Droplet microfluidics DNA ATP Streptavidin
1. Introduction As an important small molecule in life activities, adenosine-5′triphosphate (ATP) is a multifunctional nucleotide, which is not only a universal energy source, but also an extracellular signaling mediator and involved in many biological processes including DNA replication, membrane ion-channel pump, biosynthesis, hormonal and neuronal activities (Abraham et al., 1997; Newman et al., 1997; Szewczyk et al., 1998). It has also been used as an indicator of living organisms for cell viability and cell injury (Eguchi et al., 1997). Therefore, the highly sensitive and selective detection of ATP is essential for biochemical research, food quality control, environmental analysis, as well as clinical diagnosis. Many methods based on host-guest receptors, peptides, conjugated polymers, DNA/RNA aptamers, and ATP-dependent ligation reactions have been developed for ATP detection (Li et al., 2005; Lin et al., 2014; Lu et al., 2010, 2011; McCleskey et al., 2003; Mizukami et al., 2002; Zhou et al., 2011). Among these reported strategies, some of them exhibit only moderate sensitivity with detection limits for ATP in the micromolar or nanomolar range. In addition, even though n
Corresponding author. E-mail address: [email protected] (Z. He).
http://dx.doi.org/10.1016/j.bios.2016.04.055 0956-5663/& 2016 Elsevier B.V. All rights reserved.
some methods have shown desirable analytical performance, they usually require pre-labeling of a signal source, which needs considerable time-consumption and may suffer from higher cost. The investigation of small molecule-protein interaction is of great significance in unveiling the mystery of cell development involving small molecules, as well as in the drug discovery and molecular diagnostics and therapeutics (Gao et al., 2004; Howitz et al., 2003; Overington et al., 2006; Stockwell, 2004). There are numerous analytical methods available for the assays of small molecules or their protein receptors, including capillary electrophoresis, affinity chromatography, surface plasmon resonance (SPR), fluorescence resonant energy transfer, and fluorescence anisotropy (Bachovchin et al., 2009; Drabovich et al., 2009; Goldman et al., 2005; Mano et al., 2006; Petrov et al., 2005; Wear et al., 2005). Typical of many molecule-protein interactions, the binding of biotin to streptavidin (SA) has been the subject of considerable fundamental and applied interest. To study the SA-biotin interaction, a novel method called terminal protection of small moleculelinked DNA was developed, which translated the small moleculeprotein interaction into the detection of DNA (Wu et al., 2009). In view of the unique characteristics of DNA, such as specificity, stability, sequence coding, and the assists of various enzymes as well as enormous achievements of DNA sensors (Li et al., 2010; Zhang et al., 2013; Zhao et al., 2015), the strategy of terminal
222
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
protection of small molecule-linked DNA opened a new window for studying the small molecule-protein interaction, and has been applied in numerous bioassays (Cao et al., 2012; He et al., 2013; Wang et al., 2013; Wu et al., 2011; Zhou et al., 2013). Inspired by the idea of building the bridge of DNA detection for other molecules sensing, and based on our previous efforts in small molecule and protein assays (Chen et al., 2014a, 2016), a versatile dumbbell molecular (DM) probe was designed for multiplex and sensitive bioassays. In the presence of target analyte, the DM probe was protected from the degradation of exonucleases, consequently exhibiting obvious fluorescence response resulted from the protected DM probe bound to intercalation dye. United with the ATP dependent ligation reaction, the DM probe with 5′-end phosphorylation was constructed for sensitive detection of ATP which was chosen as target model of small molecule. Changing the same DM probe to the one with 3′-end biotinylation, the strategy of target-protecting DM probe against exonucleases digestion was also employed for the sensing of protein SA based on the specific SA-biotin interaction. In addition to the versatility and compatibility of the proposed strategy, the results indicated that both of the detections of ATP and SA were of high sensitivity and selectivity, as well as excellent resistance to matrix interferences. What's more, benefiting from the user-friendly control, this sensing strategy was also applied in the visual colorimetric analysis in droplet-based microfluidic platform.
2. Experimental 2.1. Materials and reagents All oligonucleotide with different sequences were synthesized and HPLC purified by Sangon Biotechnology Co., Ltd (Shanghai, China). The sequences of the oligonucleotide used in this work are listed in Table S1 (Supplementary information). Exonuclease I (Exo I), Exonuclease III (Exo III), and Shrimp alkaline phosphatase (SAP) were purchased from the Takara Biotechnology Co., Ltd. (Dalian, China). T4 DNA ligase was purchased from Thermo Fisher Scientific. SYBR Green I (SG I) (10,000 ), Streptavidin (SA), adenosine triphosphate (ATP) and its analogues were obtained from SigmaAldrich. Human serum sample was supplied by the Zhongnan Hospital of Wuhan University (Wuhan, China). Cell culture dishes were was obtained from NEST Biotechnology (Beijing, China). Culture medium and fetal bovine serum (FBS) were purchased from Gibco BRL (Grand Island, NY, USA). The silicon wafer was purchased from Institute of Microelectronics of Chinese Academy of Sciences. The poly(dimethylsiloxane) (PDMS) and AZ 50XT photoresist were obtained from RTV615 GE Toshiba Silicones Co., Ltd. and AZ Electronic Materials USA Corp., respectively. All chemical reagents were of analytical grade and used without further purification. All solutions were prepared with ultrapure water (18.25 MΩ cm) from a Millipore system.
solution was diluted to 400 μL with Tris–HCl buffer and the fluorescence spectrum was obtained with a RF-5301PC spectrophotometer (Shimadzu, Japan) equipped with a 150 W xenon lamp (Ushio Inc, Japan). 2.3. Agarose gel electrophoresis After the ligation and digestion reactions mentioned above, the mixture was stained with 100 SG I (1 μL). Three percent of agarose gel was prepared using 1 TAE buffer (40 mM Tris-AcOH, 2 mM Na2EDTA, pH ¼8.5). The electrophoresis was carried out at 100 V for approximately 40 min in 1 TAE buffer. Then the gel was imaged by a ChemiDoc XRD system (Bio-Rad). 2.4. Cellular ATP assay HepG2 cells were maintained in Dulbecco's modified eagle's medium (DMEM) supplemented with 5% fetal bovine serum (FBS), 10 mM Hepes, antibiotics (50 U/mL pencillin and 50 μg/mL streptomycin), and 2 mM glutamine at 37 °C under 5% CO2 atmosphere. The collected cells were centrifuged at 3000 rpm for 3 min at 4 °C and washed three times by PBS buffer (20 mM phosphate, 100 mM NaCl, and 5 mM KCl, pH ¼7.4) and then suspended in 25 mM Tris– HCl buffer (100 mM NaNO3, 6 mM magnesium acetate, pH ¼7.6). Cell lysis was performed by repeated cycles of freezing and thawing, and then the lysed cells were ready for ATP assays. For control experiment, the above HepG2 cell lysate (approximately 20,000 cells/mL, 400 μL) was treated with 10 units SAP at 37 °C for 20 min to remove the ATP. 2.5. Visual detection of ATP in droplet platform The droplet microchip designed and fabricated according to the previous protocol (Chen et al., 2015b). For ATP assay, the DM probe 1 was sealed by adding 8 μL of T4 DNA ligase (5 U/μL), 5 μL different concentrations of ATP and 17 μL of Tris–HCl buffer into the 10 μL of 5 μM DM probe 1 solution and allowing the ATP-triggered ligation reaction at room temperature for 30 min. After that, 5 μL of Exo I (5 U/μL) and 5 μL of Exo III (200 U/μL) were added into the mixture solution to induce the digestion for 60 min. After that, the mixture solution was delivered from the sample container into the microchip for droplets generation by using the technique previously reported (Chen et al., 2015a). Finally, the droplets containing the protected DM probe 1 were mixed with the SG I reagent through the droplet dosing strategy. The fluorescence images were recorded by an inverted fluorescence microscope (Axio Observer.A1, Zeiss, Germany) in conjunction with a light source system (excitation filter was set as 490 nm) and a Spot RT3 chargecoupled device (CCD, Diagnostic Instruments, Inc., USA). The fluorescence intensities of the droplets were obtained by ImagePro Plus software.
2.2. Procedures for ATP assay 2.6. Procedures for SA detection In a typical procedure, first, the DM probe 1 was diluted with 25 mM Tris–HCl buffer (100 mM NaCl, 5 mM MgCl2, 1 mM DTT, pH ¼7.6) and denatured at 95 °C for 10 min followed with a slow annealing treatment for 1 h before use. Then the DM probe 1 was sealed by adding 4 μL of T4 DNA ligase (5 U/μL), 5 μL different concentrations of ATP and 100 μL of Tris–HCl buffer into the 10 μL of 0.1 μM DM probe 1 solution and allowing the ATP-triggered ligation reaction at room temperature for 30 min. After that, 4 μL of Exo I (5 U/μL) and 10 μL of Exo III (20 U/μL) were added into the mixture solution to induce the digestion for 60 min. After that, 2 μL of SG I (50 ) was added into the solution. Finally, the
First, the DM probe 2 was diluted with 25 mM Tris–HCl buffer (100 mM NaCl, 5 mM MgCl2, 1 mM DTT, pH ¼7.6). 10 μL of 0.2 μM DM probe 2 was mixed with 5 μL different concentrations of SA and followed by adding 100 μL Tris–HCl buffer. The solution was incubated in room temperature for 30 min Then, 4 μL of Exo I (5 U/ μL) and 10 μL of Exo III (20 U/μL) were added into the mixture solution to induce the digestion for 60 min. After the digestion reaction, 2 μL of SG I (50 ) was added into the solution. Finally, the solution was diluted to 400 μL with Tris–HCl buffer and the fluorescence intensity was measured.
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
223
3. Results and discussion 3.1. Probe design and detection principle In this work, a dumbbell-shaped DNA molecule was designed as the probe for the detections of ATP and SA. This DM probe consists of two cytosine hairpin loops and a double-helical stem which is closed by the loops. In subsequent sensing applications, the fluorescence signal mainly results from the stem of the DM probe after binding with an intercalation dye, SG I. So the sequence and length of the stem should be carefully evaluated. First, the dumbbell DNA molecules with different sequences of stems were stained with SG I to obtain the fluorescence signals. It is shown that the fluorescence intensity of the dumbbell DNA with AT-rich stem stained by SG I is stronger than those of dumbbell DNA with CG-rich stem and dumbbell DNA with random sequence stem (Fig. 1A), which is also consistent with previous report (Zipper et al., 2004). This result indicates that the DM (AT-rich) is more suitable employed as the probe in the subsequent assay compared with DM (CG-rich) and DM (random). Moreover, the length of AT pairs in the stem of dumbbell DNA molecule was evaluated. As shown in Fig. 1B, the fluorescence intensity of the dumbbell DNA stained by SG I dramatically increased with the increase of the length of AT pairs range from 16 to 32. It is noted that the increase rate of the fluorescence intensity became slow when the length of AT pairs reached to 24. What's more, the DNA cost also obviously increased with the increase of the stem length of the DM probe. In consideration of the enhancement of fluorescence intensity and the experiment cost, the dumbbell DNA containing 24 AT pairs in the stem was chosen as the most suitable DM probe in this work. After that the optimal design of DM probe was obtained, the versatile sensing strategy for sensitive detection of different biomolecules was presented. For ATP detection, the DM probe was phosphorylated in its 5′-end, which was called as DM probe 1. The principle of ATP detection is shown in Scheme 1A. In the absence of cofactor ATP, the ligation reaction could not be initiated by T4 DNA ligase. So the DM probe 1 was completely digested by Exo I and Exo III and no double-helical stems bound to SG I, which provided a low background for the sensing system. However, in the presence of ATP, the DM probe 1 was sealed as close structure that could resist the digestion by Exo I and Exo III due to the T4 ligase-catalyzed ligation reaction. As a result, the protected DM probe 1 was easily stained with SG I, and a strong fluorescence signal was obtained. For SA detection, the DM probe was biotinylated in its 3′-end, which was called as DM probe 2. The schematic demonstration of SA detection was shown in Scheme 1B. In
Scheme 1. Schematic illustration of the mechanism for the sensitive detection of ATP (A) and SA (B).
the absence of SA, the DM probe 2 without any protections was degraded by exonucleases, resulting in a low background signal. Whereas, when SA was added to the solution, it bound to the biotin modified in the 3′ end of DM probe 2 due to the specific SAbiotin interaction. The steric hindrance of protein SA prevents exonucleases from approaching and cleaving the phosphodiester bond adjacent to the 3′ terminus. So the DM probe 2 was protected from the degradation by exonucleases. Consequently, an obvious fluorescence response was detected when the protected DM probe 2 bound to SG I.
Fig. 1. (A) Fluorescence spectra of the DM (AT-rich), DM (CG-rich) and DM (random) stained by SG I, respectively. (B) Fluorescence spectra of DM (AT16), DM (AT20), DM (AT24), DM (AT28), DM (AT32) stained by SG I, respectively.
224
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
Fig. 2. (A) Fluorescence spectra of the system under different conditions: (a) DM probe 1 þExo I þ Exo III; (b) DM probe 1þ T4 ligase þ Exo Iþ Exo III; (c) DM probe 1 þT4 ligase þ 2 nM ATP þExo Iþ Exo III. (B) Agarose gel electrophoresis of the products of different experiments. Lane M: DNA ladder; lane 1: DM probe 1; lane 2: DM probe 1 þT4 ligase; lane 3: DM probe 1þExo I þ Exo III; lane 4: DM probe 1þ T4 ligase þ Exo I þ Exo III; lane 5: DM probe 1þ T4 ligaseþ ATP þ Exo Iþ Exo III.
3.2. Feasibility of ATP detection To verify the designed strategy, the feasible experiment was carried out and the result is illustrated in Fig. 2A. In the absence of ATP, no obvious fluorescence emission was detected, which was almost the same as the fluorescence signal in the condition that DM probe 1 was completely digested by Exo I and Exo III. It is revealed that this sensing strategy possesses a highly reduced background. In contrast, a significant fluorescence emission was observed in the presence of 2 nM ATP. These results suggest that the proposed method is feasible for ATP detection. In order to further confirm the feasibility of the proposed strategy, the experiment of agarose gel electrophoresis was carried out. As shown in Fig. 2B, in the absence of Exo I and Exo III, a strong fluorescence band was easily identified when the solution contained only DM probe 1 (lane 1) or DM probe 1 and T4 ligase (lane 2). Further, when Exo I and Exo III were added in the solutions mentioned above, no obvious bands were obtained (lane 3, 4). It is revealed that the unprotected DM probe 1 could be completely degraded by the exonucleases due to the failed ligation reaction. However, when the solution contained DM probe 1, ATP, T4 ligase, and exonucleases, the DM probe 1 could be sealed by T4 ligation reaction and resistant to the digestion of exonucleases. As expected, an evident band was observed in lane 5, and the band indicated the same mobility with that in the control experiments (lane 1, 2). The above results successfully verified that the proposed strategy was entirely feasible for the ATP detection. For ATP detection, the digestion procedures of exonucleases play vital roles in the performance of the sensing system, because they are in direct relation to the background signal. So, the effects of the exonucleases concentrations and digestion time on the background signal were evaluated first. The optimization experiments were carried out in the absence of ATP, and the exonucleases concentrations and digestion time were defined as optimum value when the background signal reached to the minimum. Through the successive investigations, the optimal concentrations of Exo I and Exo III were 0.05 and 0.5 U/μL, respectively (Fig. S1, 2). The digestion time was fixed at 60 min (Fig. S3). Besides, as key factors for fluorescence enhancement in the presence of ATP, the T4 ligase concentration and ligation time were optimized as well. The optimal concentration of T4 ligase was 0.05 U/μL (Fig. S4), and the optimal ligation time was 30 min (Fig. S5). Because both of the digestion reaction and ligation reaction were carried out at room temperature in purpose to simplify the experiments. So it is
reasonable infer that the digestion and ligation times were shorter when the experiments were performed at the optimum temperature of enzyme reactions, such as 37 °C. Moreover, the effect of SG I concentration on the fluorescence signal was also investigated, and the optimal concentration was fixed at 50 in subsequent experiments (Fig. S6). 3.3. Performance of ATP detection Under the optimized conditions discussed above, the emission titration experiment with increasing concentrations of ATP was carried out, and the linear response range of the proposed strategy for ATP detection was evaluated. As shown in Fig. 3A, the fluorescence intensity of the protected DM probe 1 stained with SG I dramatically increased with the increase of the ATP concentration. The fluorescence intensity versus ATP concentration is plotted in the range from 10 to 4000 pM (Fig. 3B). As shown in Fig. 3C, there is a good linear relationship (R2 ¼0.9926) between the fluorescence intensities and the concentrations of ATP with a detection limit of 4.8 pM (signal-to-noise ratio of 3). To the best of our knowledge, this detection limit is superior to those of previously reported fluorescence sensing strategies, and even can be comparable with or exceed those of some previously reported electrochemical methods (Table S2). Among the methods listed in Table S2, the detection limit of the electrochemical sensor based on multiple signal amplification strategy is 0.5 pM, which is lower than that of this work (Chen et al., 2014b). However, the construction of the electrochemical sensor involves tedious procedures of nanocomposite preparation, aptamer conjugation, and magnetic separation. By contrast, we put forward a simple, rapid and homogeneous method for ATP detection in this work. The high sensitivity of this proposed strategy is mainly attributed to the low background signal caused by exonucleases degradation as well as the strong fluorescence response of the protected DM probe 1 stained with SG I. Selectivity is a critical parameter to assess the performance of a sensing strategy. In this work, the selectivity of this strategy has been evaluated by comparing the fluorescence signal of samples containing ATP with those of its analogues, including cytidine triphosphate (CTP), guanosine triphosphate (GTP), uridine triphosphate (UTP), and adenosine diphosphate (ADP). As shown in Fig. 3D, when ATP, CTP, GTP, UTP, and ADP were added to the sensing system, respectively, only ATP caused a marked fluorescence enhancement. However, the others made no obvious
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
225
Fig. 3. (A) Fluorescence spectra of the system in response to different concentrations of ATP. (B) Relationship between the fluorescence intensity and ATP concentration. (C) Calibration curve for ATP detection. (D) Selectivity of the developed sensing strategy for ATP compared to ATP analogues and blank.
changes in fluorescence intensity, even though the concentration of the analogues was one hundred times as that of ATP. This desirable selectivity of the method mainly benefited from the high dependence of T4 DNA ligase for the ATP cofactor. 3.4. Real sample analysis To evaluate the detection toward the target in complex biological matrix, the proposed strategy was performed to detect ATP in human serum samples. A serious of ATP standard solutions at 0, 0.25, 1.0, 1.5, and 2.0 nM were separately spiked into the 10% diluted human serum and then the samples were tested. As shown
in Fig. 4A, the fluorescence signals detected in the diluted human serum had slightly differences with those obtained in buffer. Despite these differences, the presence of ATP in human serum still could be detected with this developed method. It is demonstrated that the ATP assay can be performed in complex biological environment with this sensing strategy. Therefore, the ATP assay in real sample was further carried out through the detection of ATP in cancer cells (HepG2 cells). As illustrated in Fig. 4B, in the absence of cell lysates, no obvious fluorescence emission was observed (curve a). However, in the presence of lysates from different numbers of HepG2 cells, all the obvious fluorescence signals were obtained (curve b, c, and d). Furthermore, when the cell lysates
Fig. 4. (A) Fluorescence responses of the sensing platform for ATP detection in buffer and 10% diluted human serum samples, respectively. (B) Fluorescence responses of the system in the absence of freshly lysed cells (a), and in the presence of freshly lysed cells with different concentrations: approximately 15,000 cells/mL (b), 30,000 cells/mL (c), 60,000 cells/mL (d), as well as in the presence of SAP-treated lysed cells (60,000 cells/mL) (e).
226
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
Fig. 5. (A) Schematic illustration of visual detection of ATP in droplet-based platform. Fluorescence responses of the droplets in the absence (B) and presence (C) of ATP. (D) Fluorescence responses of the droplets with the different concentrations of ATP. Scale bars indicate 100 mm.(For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
were treated with shrimp alkaline phosphatase (SAP) to remove the ATP, almost no significant fluorescence enhancement was observed as anticipated (curve e). The above results clearly showed the robustness of the proposed method in real sample analysis. 3.5. Visual detection of ATP in droplet-based platform In our previous work, to make the operation of droplet-based analysis more convenient and efficient, we presented a novel droplet dosing strategy and a flexible sample-introduction technique in microfluidic system (Chen et al., 2015a, 2015b). Integrating these previously reported techniques with the sensing strategy developed in this work, the visual detection of ATP was successfully realized in droplet-based microfluidic platform. The schematic illustration of visual detection of ATP based on droplet platform is illustrated in Fig. 5A. First, the ATP-dependent ligation reaction and exonucleases-involved degradation were successively carried out in the centrifuge tube. Then the solution of mixtures was directly delivered into the microchannel of chip from the centrifuge tube and generated as numerous droplets. Based on the dosing strategy previously reported, the intercalation dye (SG I) was added into the droplets (Chen et al., 2015b). So the protected DM probe 1 contained in the droplet was stained with SG I and resulted in an obvious fluorescence response. The qualitative analysis of ATP was achieved from the fluorescence image of droplet with the naked eye, and the quantitative analysis was performed by colorimetric method. As shown in Fig. 5B, when the ATP was absent, the droplets demonstrated as dark color in the fluorescence images, so no obvious fluorescence response was obtained. While in the presence of ATP, the droplets presented bright green color in the fluorescence images, which was easily discriminated from that color of blank droplets with the naked eye (Fig. 5C). To evaluate the analytical performance of visual detection of ATP, a serious of standard samples containing various
concentrations of ATP were detected in the droplet-based platform. As shown in Fig. 5D, the green color of the droplet in fluorescence image turned brighter and brighter as the increase of ATP concentration in the range from 25 to 1000 nM, and did not change obviously when the concentration increased to 1250 nM. What's more, as indicated in Fig. S7, the fluorescence intensity obtained from droplet image showed a clear linear relationship with the ATP concentration over the range from 25 to1000 nM (R2 ¼0.9932). The detection limit was calculated as 8.9 nM (signalto-noise ratio of 3), which is about 2 amol per droplet. To our knowledge, the sensitivity of this visual detection can be comparable with or exceeds those of previously reported strategies with fluorescent spectrometry (Table S2). 3.6. Sensitive and selective detection of SA The proposed strategy was also extended to the protein detection by integrating the idea of target-protecting DM probe against exonucleases degradation with SA-biotin interaction (Scheme 1B). The fluorescence responses of the sensor to target SA at different concentrations are illustrated in Fig. 6A, B, from which it can be seen that the fluorescence intensity increased significantly with the increase of SA concentration. There is a good linear relationship (R2 ¼ 0.9903) between the fluorescence intensities and the concentrations of SA in the range from 50 to 4000 pM (Fig. 6C). The detection limit was determined to be 12.7 pM (signal-to-noise ratio of 3), which is suitable to serve as a biosensor for detecting SA. To investigate the selectivity of the strategy, 2 nM of SA and 20 nM of other proteins were tested with this biosensor. As shown in Fig. 6D, the results showed that only SA could significantly enhance the fluorescence intensity of the protected DM probe 2 stained with SG I. Whereas no significant change in emission intensity was observed upon the addition of the other proteins, such as thrombin, trypsin, lysozyme, bovine serum albumin (BSA), and pepsin, even though the concentration
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
227
Fig. 6. (A) Fluorescence spectra of the system in response to different concentrations of SA. (B) Relationship between the fluorescence intensity and SA concentration. (C) Calibration curve for SA detection. (D) Selectivity of the developed sensing strategy for SA compared to other proteins and blank. The concentration of SA is 2 nM, while the concentrations of other proteins are 20 nM.
of the other proteins was ten times as that of SA. This high selectivity was mainly attributed to the specific interaction of SAbiotin. It is notable that the proposed strategy could be extended to detect different proteins just by modifying the DM probe with corresponding small molecule, which demonstrates the versatility and generality of this design.
4. Conclusion In summary, a dumbbell-shaped DNA probe was designed and successfully applied in the homogeneous bioassays based on the strategy of target-protecting DM probe against exonucleases degradation. This strategy is simple but versatile, which allows sensitive detection of diverse targets such as ATP and SA. The advantages of this strategy are as follows. Apart from the phosphorylation of DM probe for ATP assay or the biotinylation for SA detection, neither laborious and time-consuming fluorescence labeling, nor tedious separation steps were required in this method. In addition, the fluorescence signal readout was carried out by staining the protected DM probe with SG I, which is easy and fast to realize. The proposed strategy also possesses significantly high selectivity and satisfactory sensitivity which allows both of the detection limits for small molecule and protein in picomolar level. Benefiting from the excellent detection performance, the method was successfully employed in the detection in complex environment as well as real sample analysis. To broaden the application of the developed strategy, the visual detection of ATP was also accomplished by integrating the strategy with droplet-based microfluidics. And the detection limit still could reach to nanomolar
level just by the naked eye. In a word, all the results as mentioned above clearly indicate the robustness and versatility of the proposed strategy.
Acknowledgments This work was supported by the National Natural Science Foundation of China (21275109, 21205089, 21475101), the Suzhou Nanotechnology Special Project (ZXG2013028).
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2016.04.055.
References Abraham, E.H., Okunieff, P., Scala, S., Vos, P., Oosterveld, M.J., Chen, A.Y., Shrivastav, B., 1997. Science 275, 1324–1326. Bachovchin, D.A., Brown, S.J., Rosen, H., Cravatt, B.F., 2009. Nat. Biotechnol. 27, 387–394. Cao, Y., Zhu, S., Yu, J., Zhu, X., Yin, Y., Li, G., 2012. Anal. Chem. 84, 4314–4320. Chen, C., Xiang, X., Liu, Y., Zhou, G., Ji, X., He, Z., 2014a. Biosens. Bioelectron. 58, 205–208. Chen, J., Ji, X., He, Z., 2015a. Anal. Chim. Acta 888, 110–117. Chen, J., Liu, Y., Ye, T., Xiang, X., Ji, X., He, Z., 2015b. Sens. Actuator B-Chem. 215, 206–214. Chen, J., Ji, X., Tinnefeld, P., He, Z., 2016. ACS Appl. Mater. Interfaces 8, 1786–1794. Chen, X., Ge, L., Guo, B., Yan, M., Hao, N., Xu, L., 2014b. Biosens. Bioelectron. 58, 48–56.
228
J. Chen et al. / Biosensors and Bioelectronics 83 (2016) 221–228
Drabovich, A.P., Berezovski, M.V., Musheev, M.U., Krylov, S.N., 2009. Anal. Chem. 81, 490–494. Eguchi, Y., Shimizu, S., Tsujimoto, Y., 1997. Cancer Res. 57, 1835–1840. Gao, X., Cui, Y., Levenson, R.M., Chung, L.W., Nie, S., 2004. Nat. Biotechnol. 22, 969–976. Goldman, E.R., Medintz, I.L., Whitley, J.L., Hayhurst, A., Clapp, A.R., Uyeda, H.T., Deschamps, J.R., Lassman, M.E., Mattoussi, H., 2005. J. Am. Chem. Soc. 127, 6744–6751. He, Y., Xing, X., Tang, H., Pang, D., 2013. Small 9, 2097–2101. Howitz, K.T., Bitterman, K.J., Cohen, H.Y., Lamming, D.W., Lavu, S., Wood, J.G., Zipkin, R.E., Chung, P., Kisielewski, A., Zhang, L.L., Scherer, B., Sinclair, D.A., 2003. Nature 425, 191–196. Li, C., Numata, M., Takeuchi, M., Shinkai, S., 2005. Angew. Chem. Int. Ed. 44, 6371–6374. Li, D., Song, S., Fan, C., 2010. Acc. Chem. Res. 43, 631–641. Lin, C., Cai, Z., Wang, Y., Zhu, Z., Yang, C.J., Chen, X., 2014. Anal. Chem. 86, 6758–6762. Lu, C.H., Li, J., Lin, M.H., Wang, Y.W., Yang, H.H., Chen, X., Chen, G.N., 2010. Angew. Chem. Int. Ed. 49, 8454–8457. Lu, L.M., Zhang, X.B., Kong, R.M., Yang, B., Tan, W., 2011. J. Am. Chem. Soc. 133, 11686–11691. Mano, N., Sato, K., Goto, J., 2006. Anal. Chem. 78, 4668–4675. McCleskey, S.C., Griffin, M.J., Schneider, S.E., McDevitt, J.T., Anslyn, E.V., 2003. J. Am.
Chem. Soc. 125, 1114–1115. Mizukami, S., Nagano, T., Urano, Y., Odani, A., Kikuchi, K., 2002. J. Am. Chem. Soc. 124, 3920–3925. Newman, E.A., Zahs, K.R., 1997. Science 275, 844–847. Overington, J.P., Al-Lazikani, B., Hopkins, A.L., 2006. Nat. Rev. Drug. Discov. 5, 993–996. Petrov, A., Okhonin, V., Berezovski, M., Krylov, S.N., 2005. J. Am. Chem. Soc. 127, 17104–17110. Stockwell, B.R., 2004. Nature 432, 846–854. Szewczyk, A., Piku̵ a, S., 1998. Biochim. Biophys. Acta 1365, 333–353. Wang, G., He, X., Wang, L., Zhang, X., 2013. Biosens. Bioelectron. 42, 337–341. Wear, M.A., Patterson, A., Malone, K., Dunsmore, C., Turner, N.J., Walkinshaw, M.D., 2005. Anal. Biochem. 345, 214–226. Wu, Z., Zhen, Z., Jiang, J.H., Shen, G.L., Yu, R.Q., 2009. J. Am. Chem. Soc. 131, 12325–12332. Wu, Z., Wang, H., Guo, M., Tang, L.J., Yu, R.Q., Jiang, J.H., 2011. Anal. Chem. 83, 3104–3111. Zhang, H., Li, F., Dever, B., Li, X.F., Le, X.C., 2013. Chem. Rev. 113, 2812–2841. Zhao, Y., Chen, F., Li, Q., Wang, L., Fan, C., 2015. Chem. Rev. 115, 12491–12545. Zhou, G., Zhang, X., Ji, X., He, Z., 2013. Chem. Commun. 49, 8854–8856. Zhou, Z., Du, Y., Dong, S., 2011. Anal. Chem. 83, 5122–5127. Zipper, H., Brunner, H., Bernhagen, J., Vitzthum, F., 2004. Nucleic Acids Res. 32, e103.