Biosensors and Bioelectronics 73 (2015) 138–145
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Label-free and sensitive detection of T4 polynucleotide kinase activity via coupling DNA strand displacement reaction with enzymatic-aided amplification Rui Cheng, Mangjuan Tao, Zhilu Shi, Xiafei Zhang, Yan Jin n, Baoxin Li Key Laboratory of Applied Surface and Colloid Chemistry, Ministry of Education, Key Laboratory of Analytical Chemistry for Life Science of Shaanxi Province, School of Chemistry and Chemical Engineering, Shaanxi Normal University, Xi'an 710062, China
art ic l e i nf o
a b s t r a c t
Article history: Received 13 February 2015 Received in revised form 5 May 2015 Accepted 26 May 2015 Available online 28 May 2015
Several fluorescence signal amplification strategies have been developed for sensitive detection of T4 polynucleotide kinase (T4 PNK) activity, but they need fluorescence dye labeled DNA probe. We have addressed the limitation and report here a label-free strategy for sensitive detection of PNK activity by coupling DNA strand displacement reaction with enzymatic-aided amplification. A hairpin oligonucleotide (hpDNA) with blunt ends was used as the substrate for T4 PNK phosphorylation. In the presence of T4 PNK, the stem of hpDNA was phosphorylated and further degraded by lambda exonuclease (λ exo) from 5′ to 3′ direction to release a single-stranded DNA as a trigger of DNA strand displacement reaction (SDR). The trigger DNA can continuously displace DNA P2 from P1/P2 hybrid with the help of specific cleavage of nicking endonuclease (Nt.BbvCI). Then, DNA P2 can form G-quadruplex in the presence of potassium ions and quadruplex-selective fluorphore, N-methyl mesoporphyrin IX (NMM), resulting in a significant increase in fluorescence intensity of NMM. Thus, the accumulative release of DNA P2 led to fluorescence signal amplification for determining T4 PNK activity with a detection limit of 6.6 10 4 U/mL, which is superior or comparative with established approaches. By ingeniously utilizing T4 PNK-triggered DNA SDR, T4 PNK activity can be specifically and facilely studied in homogeneous solution containing complex matrix without any external fluorescence labeling. Moreover, the influence of different inhibitors on the T4 PNK activity revealed that it also can be explored to screen T4 PNK inhibitors. Therefore, this label-free amplification strategy presents a facile and cost-effective approach for nucleic acid phosphorylation related research. & 2015 Elsevier B.V. All rights reserved.
Keywords: Label-free Fluorescence Signal amplification DNA phosphorylation PNK activity Strand displacement reaction
1. Introduction The phosphorylation of nucleic acids plays a critical basis for quickly and effectively repairing DNA damage, such as singlestrand break (Whitehouse et al., 2001) base excision, (Wiederhold et al., 2004) and oxidative DNA damage (Breslin and Caldecott, 2009). Polynucleotide kinase (PNK) can catalyze the transfer of the γ-phosphate group from a nucleoside triphosphate (the most used one is ATP ) to the 5′-hydroxyl terminus of DNA or RNA (Richards, 1965; Novogrod et al., 1966), which plays a vital role in the phosphorylation of nucleic acids. Besides, the human PNK participates in maintaining the genetic integrity (Rasouli-Nia et al., 2004) and is closely associated with some vital human diseases, such as Werner syndrome, Rothmund Thomson syndrome and Bloom's syndrome (Shen et al., 2010; Tahbaz et al., 2012; Little, n
Corresponding author. Fax: þ86 29 81530727. E-mail address: [email protected] (Y. Jin).
http://dx.doi.org/10.1016/j.bios.2015.05.059 0956-5663/& 2015 Elsevier B.V. All rights reserved.
1967). Recently Freschauf's group reported the PNK inhibition should potentially increase the sensitivity of human tumors to γradiation (Freschauf et al., 2010, 2009). That means the inhibition of PNK activity might be a promising direction against cancers (Allinson, 2010). Therefore, the research on the PNK activity is of great importance to the biomedical research (Weinfeld et al., 2011). Recently, researchers employ T4 polynucleotide kinase (T4 PNK) which belongs to one kind of polynucleotide kinase family as the model target to explore the assay of PNK activity. They have made great progress. Conventionally, the activity of T4 PNK was detected via these sophisticated protocols, such as radioisotope 32 P-labeling, autoradiography, and polyacrylamide gel electrophoresis (Whitehouse et al., 2001; Wang et al., 2002). However, they were rather complex, time consuming, laborious, expensive and may necessitate stringent safety measures to control radiographic exposure. Up to now, some novel strategies including fluorescence detection (Song and Zhao, 2009; Jiao et al., 2012; Chen et al., 2013; Wu et al., 2011; Lin et al., 2011; Huang et al.,
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
2013; Liu et al., 2014a, 2014b; Tao et al., 2014; Guo et al., 2015; Zhu et al., 2014; Jiang et al., 2014), colorimetric assay (Jiang et al., 2013), nanochannel biosensor (Lin et al., 2013), and electrochemical methods (Wang et al., 2012, 2013) were developed for convenient determination of T4 PNK activity. Among them, fluorescence methods are particularly attractive due to their high sensitivity, easy readout, small sample volume, simple operation and feasibility of quantification. Song and Zhao (2009) proposed a simple fluorescence approach for real-time monitoring of the activity and kinetics of T4 PNK by employing a singly fluorophore-labeled DNA-hairpin probe. Wu et al. (2011) and Lin et al. (2011), respectively, realized the detection of T4 PNK activity by using graphene oxide as a super quencher. With the development of signal amplification technique, Chen et al. (2013) developed a novel amplified fluorescence approach for T4 PNK assay based on nicking endonuclease-mediated signal amplification. Based on double-labeled molecular beacon, Liu et al. (2014a, 2014b) proposed a dual amplification strategy for sensitive detection of T4 polynucleotide kinase activity by coupling split DNAzyme and ligation-triggered DNAzyme cascade amplification. Although these fluorescence strategies have made great advances toward the DNA phosphorylation assay, they always utilize fluorescence dye double-labeled hairpins (Chen et al., 2013; Liu et al., 2014a, 2014b; Tao et al., 2014) or single-labeled probes combining with nanomaterials as quenchers (Wu et al., 2011; Lin et al., 2011). Label-free strategy is a good choice for realizing facile, reliable and cost-effective assay of T4 PNK activity. However, label-free method is rare (He et al., 2014; Guo et al., 2015; Jiang et al., 2014). Therefore, the development of facile, label-free and sensitive fluorescence strategy is necessary and meaningful for studying T4 PNK activity and inhibition. We have addressed the limitation and report here a label-free strategy for sensitive detection of T4 PNK activity by coupling DNA strand displacement reaction with enzymatic-aided amplification. Yurke and co-workers (Yurke et al., 2000) pioneered the concept of toehold-mediated DNA SDR which can realize signal amplification without the help of functional enzymes. A hairpin oligonucleotide (hpDNA) with blunt ends was used as the substrate for T4 PNK phosphorylation. In the presence of T4 PNK, the stem of hpDNA was phosphorylated and further degraded by lambda exonuclease (λ exo) from 5′ to 3′ direction to release a singlestranded DNA as a trigger of DNA strand displacement reaction. The trigger drives SDR and subsequent restriction enzymatic-assisted cyclic amplification to continuously release signal DNA. It folds into a quadruplex structure in the presence of potassium ions and quadruplex-selective dye, leading to obvious fluorescence readout. Thus, label-free detection of T4 PNK activity can be facilely achieved by ingeniously utilizing T4 PNK-induced strand displacement reaction and cyclic amplification, which ensures high specificity as well as high sensitivity. More importantly, it also offers a potential platform for screening T4 PNK inhibitors.
2. Materials and methods 2.1. Chemicals and materials The DNA sequences (listed in Table S1 in supporting information) were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). The T4 polynucleotide kinase (10 unit/μl), EcoRI endonuclease, adenosine triphosphate (ATP), adenosine diphosphate (ADP) and dithiothreitol (DTT) were purchased from Sangon Biotech Co., Ltd. (Shanghai, China). The Lambda exonuclease (λ exo, 5000 unit/mL), Nicking enzyme (Nt.BbvCI), 10 NEB buffer 4 and Protein kinase A (PKA) were purchased from New England Biolabs (NEB, U.K.). Lysozyme and immunoglobin G (IgG) were purchased
139
from Beijing Dingguo Biotechnology Co., Ltd (Beijing, China). Bovine serum albumin (BSA) was purchased from Sigma-Aldrich Inc. N-Methyl mesoporphyrin IX (NMM) (analytical reagent grade) was obtained from Frontier Scientific, Inc (Logan, Utah, USA). All DNA samples were prepared in 70 mM Tris–HCl buffer (pH 8.0, 10 mM MgCl2) prior to use. All other chemicals were of analytical grade and were used without further purification. Ultrapure water obtained from a Millipore filtration system was used throughout all experiments. 2.2. Detection and inhibition evaluation of T4 PNK activity All fluorescence measurements were executed on a fluorometer F-7000 (Hitachi, Japan). The excitation and emission wavelengths of NMM were set at 399 nm and 612 nm. Before the toehold-mediated strand displacement reaction, the DNA P1 and DNA P2, hpDNA were heated to 95 °C for 5 min in 70 mM Tris–HCl buffer (pH ¼8.0, 10 mM MgCl2) followed by slowly cooling to room temperature for 1 h before use. Firstly the phosphorylation and λ exo cleavage reaction were simultaneously performed in 100 μL enzyme reaction buffer (70 mM Tris–HCl buffer, pH 8.0, 10 mM MgCl2, 1 mM ATP and 5 mM DTT). 0.5 μM hpDNA was incubated with certain amount of T4 PNK and 50 U/mL λ exo at 37 °C for 30 min and then the reaction solution was heated to 75 °C for 10 min to deactivate the enzymes and gradually cooled to 25 °C. Then 1 μM P1/P2 hybrid, 10 mM KCl, 1 μM NMM, and 100 μL 1 NEB buffer 4 were added into the above reaction solution and the fluorescence measurement was carried out to obtain the signal before amplification. Subsequently 40 U/mL Nt.BbvCI was added into the solution to activate the enzyme-aided circle amplification and kept at 37 °C for 1 h prior to fluorescence measurement. For the T4 PNK inhibition assay, both ADP and (NH4)2SO4 were used as the model inhibitors. Various concentrations of inhibitors were separately mixed with 0.5 μM hpDNA, 1 U/mL T4 PNK and 50 U/mL λ exo in 100 μL enzyme reaction buffer. The phosphorylation and enzymatic amplification reaction were carried out via similar procedures as those for T4 PNK activity assay stated above. 2.3. T4 PNK selectivity studies In order to investigate the influence of enzymes and proteins, 1 U/mL of EcoRI/PKA and 1.0 mM of IgG/Lysozyme/BSA were added into the buffer respectively. The phosphorylation and enzymatic amplification reaction were the same as those described in the aforementioned experiment for T4 PNK activity detection in reaction buffer. 2.4. Gel electrophoresis The label-free strategy for sensitive T4 PNK assay was validated by polyacrylamide gel electrophoresis (PAGE). The phosphorylation and enzymatic amplification reaction were carried out via similar procedures as those for T4 PNK assay stated above. 10 μL samples were added 1.0 μL 10 loading buffer and then loaded on a 12.5% native polyacrylamide gel. Electrophoresis was performed at a constant voltage of 200 V for 30 min in 1 Tris-borate-EDTA (TBE) buffer. After silver staining, the gel electrophoresis was visualized on the molecular imager system (JS 180, Shanghai Peiqing science & Technology. Co., Ltd). 2.5. Circular dichroism spectroscopy CD spectroscopy is measured by using a Chirascan Circular Dichroism Spectrometer (Applied Photophysics Ltd, England, UK). CD spectra was performed using an optical chamber (1 mm optical path length) with an instrument scanning speed of 100 nm/min, a
140
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
response time of 2 s at room temperature and was accumulated by taking the average of three scans made from 220 to 300 nm. 2.6. Assay of T4 PNK activity in diluted cell extracts HeLa cells were cultured in Dulbecco's modified Eagle's medium (DMEM, GIBCO) supplemented with 10% fetal calf serum (FCS, Sigma), penicillin (100 μg/mL), and streptomycin (100 μg/mL) at 37 °C in a humidified chamber, containing 5% CO2. Cell extracts were prepared according to the previous reports (Wu et al., 2014). The harvested cells were resuspended in 20 μL 10 mM Tris–HCl (pH 7.8) containing 150 mM NaCl. Equal volume of lysis buffer was added into the suspension containing 10 mM Tris–HCl (pH 7.8), 150 mM NaCl, 2 mM EDTA, 2 mM DTT, 40% glycerol, 0.2% NP-40, and 0.4 mM phenylmethylsulfonyl fluoride. The mixture was incubated for 1.5 h at 4 °C with occasional shake. Cell debris was removed by centrifugation at 16,000 rpm for 10 min, and the supernatant was recovered. Diluted cell extracts were added to the assay solution (50%). The detection procedure was the same as those described in the aforementioned experiment for T4 PNK detection in pure reaction buffer.
3. Results and discussion 3.1. Proof of principle The label-free fluorescent approach for T4 PNK assay is based on the PNK-triggered strand displacement reaction (SDR) and enzyme-aided cyclic amplification (Scheme 1). In this work, N-Methyl mesoporphyrin IX (NMM) was chosen as the signal reporter because NMM is an anionic porphyrin that has a high structural selectivity for quadruplexes but not for duplexes, triplexes, or single-stranded forms (Oh et al., 2010). NMM itself shows tiny fluorescence. However, a significant increase in its fluorescence can be achieved upon binding to G-quadruplex. A hairpin oligonucleotide (hpDNA) was designed as a T4 PNK substrate whose 5′-hydroxyl terminal could be phosphorylated by T4 PNK and subsequently digested by λ exonuclease (λ exo) to generate a trigger DNA. Nt.BbvCI is a nicking endonuclease that cleaves only one strand of DNA of a double-stranded DNA substrate at the specific recognition site. DNA P1 contains a recognition site of Nt.BbvCI. DNA P1/P2 hybrid keeps stable in the presence of Nt.BbvCI because the recognition site of Nt.BbvCI located
Fig. 1. Fluorescence spectra of NMM under different conditions. The concentrations of P1/P2, hpDNA, T4 PNK, λ exo and Nt.BbvCI were 1 μM, 0.5 μM, 1 U/mL, 50 U/mL, 40 U/mL, respectively.
Scheme 1. Schematic illustration of label-free DNA strand displacement reaction with enzymatic-aided amplification for monitoring T4 PNK activity.
outside the double-stranded domain of P1/P2 hybrid. Only when trigger DNA displaced P2 from P1/P2 hybrid to form longer and more stable dsDNA can DNA P1 was specifically cleaved by Nt. BbvCI to release trigger DNA again for the next cycle of SDR. The continuously released P2 led to fluorescence amplification because
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
the G-rich oligomer P2 can fold into a G-quadruplex structure in the presence of potassium ions and NMM. By the elaborate design, a label-free detection of T4 PNK activity could be performed with high sensitivity and specificity. To estimate the feasibility of the amplification strategy, the fluorescence spectra of NMM under different conditions was recorded in Fig. 1. The fluorescence intensity of NMM alone is very weak (Fig. 1A, curve a). The introduction of both P1/P2 hybrid and hpDNA led negligible fluorescence change (Fig. 1A, curve b and c). Hairpin DNA highly ensured the specificity of SDR. P1/P2 hybrid cannot open the hairpin structure to trigger SDR for releasing DNA P2. Therefore, the fluorescence intensity of NMM remained unchanged in the presence of hpDNA and P1/P2 hybrid. However, the fluorescence of NMM had a remarkably enhancement when hpDNA was incubated with T4 PNK and λ exo (Fig. 1A, curve e). That is, the trigger of SDR had been successfully released because λ exo is a highly processive 5′–3′ exonuclease which prefers a phosphate moiety at 5′ end of dsDNA (Higuchi and Ochman, 1989; Little, 1967) and immediately SDR occurred. Moreover, a further increase in fluorescence intensity of NMM was obtained when Nt. BbvCI was added to the above mixture. It is clear that a 286.727 0.64% increase in fluorescence signal was achieved (Fig. 1A, curve f). However in the absence of Nt.BbvCI, the corresponding fluorescence enhancement was only 66.64 70.52% with the same concentration of T4 PNK (Fig. 1A, curve e). It is mainly attributed to the circle amplification stimulated by nicking endonucleases which can cut one strand of a double-stranded DNA at a specific recognition nucleotide sequence known as recognition site of endonuclease (Morgan et al., 2000; Higgins et al., 2001; Wang and Hays, 2000). Therefore, these preliminary results demonstrated the feasibility of this signal amplification strategy. To further ensure reliability, more control experiments were conducted. Firstly, the influence of nonspecific cleavage was investigated. When the T4 PNK was inactive, the fluorescence intensity hardly changed (Fig. 1B, curve d) because the 5′ terminal of hpDNA still retained hydroxylation form which prohibited the digestion by λ exo. When Nt.BbvCI was introduced (Fig. 1B, curve e), a slight increase ( 3.4%) in fluorescence intensity was observed. It is mainly attributed to the cleavage of Nt.BbvCI due to the little and partial hybridization of hpDNA with P1. However, the above cleavage hardly affects the analysis of T4 PNK activity because it can be eliminated as background. Then, the effect of λ exo on the fluorescence was examined. As shown in Fig. 1C, the fluorescence intensity of NMM did not change when hpDNA has been successfully phosphorylated by T4 PNK without λ exo (Fig. 1C, curve b). Meanwhile, it is obvious that the fluorescence intensity was unchanged when hpDNA was co-incubated with inactive T4 PNK and λ exo (Fig. 1C, curve c). That means the releasing of trigger was originated from the T4 PNK-actuated λ exo digestion of the stem region of hpDNA (Fig. 1C, curve d). These results clearly suggested that only the combination of T4 PNK and λ exo can result in a remarkable fluoresence increase. Therefore, this label-free amplification strategy can reliably detect T4 PNK activity. 3.2. Validation by gel electrophoresis and circular dichroism spectroscopy To further verify the sensing mechanism, the strand displacement reaction and enzymatic amplification results have been proved using a 12.5% native polyacrylamide gel electrophoresis (PAGE). As shown in Fig. 2, the bands in lane 2, 3, 4 correspond to hpDNA, P1, P2, respectively. A distinct band of hybrid P1/P2 (the molecular weight is the sum of two single probes) is observed in lane 5. In the presence of T4 PNK and λ exo, the color of bands corresponding to both hpDNA and P1/P2 was made lighter, and a
141
Fig. 2. Native 12.5% polyacrylamide gel analysis for proof of principle. The concentrations of P1/P2, P1, P2 and hpDNA were 1.0 mM, 1.0 mM, 1.0 mM and 0.5 mM. The concentrations of T4 PNK, T4 PNK (inactive), λ exo and Nt.BbvCI were 1 U/mL, 1 U/mL, 50 U/mL and 40 U/mL, respectively.
new band corresponding to DNA P2 appeared, demonstrating that the phosphorylated hpDNA can be degraded by λ exo to release the trigger. The trigger DNA hybridized with P1 to form more stable double-stranded DNA as a result of replacing P2 (Fig. 2, lane 6, featured with red dotted line). It is accorded with the essence of strand displacement reaction that the molecular weight of new dsDNA was larger than that of P1/P2. Upon the addition of Nt. BbvCI, both that new band and P2 band deepened. Meanwhile, hpDNA and P1/P2 further diminished, and new bands of short DNA fragments appeared (Fig. 2, lane 7, featured with red solid line). The result indicated P1 was cleaved by Nt.BbvCI and the trigger can be released to replace P2 circularly. However, the P2 band and new band were not observed in the presence of inactive T4 PNK (Fig. 2, lane 8, 9). Both fluorescence and PAGE results illustrated that the amplification strategy really worked on the basis of the T4 PNK-triggered strand displacement reaction and Nt. BbvCI-aided circle amplification. To verify the formation of G-quadruplex, circular dichroism (CD) measurement was utilized to monitor the conformation changes of P2 under various conditions. It is known that the “parallel” quadruplexes have a CD spectrum characterized by a positive ellipticity maximum at 264 nm and a negative minimum at 240 nm (Dapic et al., 2003). In the presence of K þ and NMM, P2 can form a “parallel” quadruplex (Figure S1A). It is clear in Figure S1 A and B that K þ is the key element to induce P2 to form G-quadruplex which will facilitate the binding of NMM. 3.3. Optimization of the related factors NMM was used as the signal molecule for fluorescence detection of T4 PNK activity. Meanwhile, potassium ion plays an
142
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
important role in stabilizing quadruplex. So, firstly, the concentrations of NMM and K þ were optimized. The data in Figure S2 revealed that the highest fluorescence intensity was obtained when the final concentrations of NMM and K þ were 1 mM and 10 mM, respectively. Secondly, interaction time was a crucial parameter for the signal amplification. As shown in Figure S3, the strand displacement reaction and enzymatic amplification were finished in 20 and 30 min, respectively. It is well-known ATP is used as the donor for phosphorylation (Liu et al., 2013). Furthermore, the effect of ATP concentration on the T4 PNK activity was investigated. As shown in Figure S4A, the fluorescence intensity increased with the increase of ATP concentration and tended to be constant at 1 mM. Thus ATP concentration of 1 mM was used in the further experiments. As a key element for triggering signal transmission, the amount of λ exo was also tested. It can be seen from Figure S4B, the fluorescence enhancement efficiency increased gradually with the increasing of λ exo concentrations and reached a plateau beyond 50 U/mL. In order to obtain cost-effective digestion by λ exo, the concentration of 50 U/mL was adopted in this method. Because Nt.BbvCI was served as the tool for signal amplification, the concentration of Nt.BbvCI should be carefully controlled. The fluorescence enhancement efficiency greatly enhanced with the increasing concentrations of Nt.BbvCI and the optimal concentration was selected at 40 U/mL (Figure S4C). Due to the competion hybridization between P2 and trigger, the molar ratios of P1/P2 to hpDNA also significantly affect the signal amplification efficiency. Herein, the effects of molar ratio was further investigated with a mixed concentration of P1/P2 (1 mM) and various hpDNA concentrations from 0.1 μM to 1 μM (Figure S4D). It was observed that the fluorescence response was quickly enhanced as the hpDNA concentration increased. However, the increase of hpDNA concentrations also accompanied with the increasing background signal. Finally, the molar ratio of 2:1 (P1/P2: hpDNA) was used throughout the experiments. 3.4. Specificity and sensitivity of T4 PNK activity assay To verify the specificity of T4 PNK detection, the influences of five common enzymes and proteins, including EcoRI, PKA, Lysozyme, IgG and BSA were tested under the identical concentration (1 U/mL or 1 mM). As shown in Fig. 3A, there was no remarkable fluorescence increase upon the addition of above enzymes and proteins except T4 PNK. It again demonstrated that only phosphorylated hpDNA can be specifically digested by λ exo to release the trigger for cyclic amplification. So, it is a highly specific method for detection of T4 PNK activity. In addition, the specificity was further verified by PAGE. As shown in Fig. 3B, only T4 PNK produced two new and remarkable bands which stand for trigger/P1 hybrid and P2, respectively. It is consistent with fluorescent result, indicating that an excellent specificity has been ensured by smart design of DNA probe. Under the optimal conditions, the analytical performance of the proposed assay for the detection of T4 PNK activity is investigated. A continuous increase in fluorescence intensity has been found in Fig. 4A when T4 PNK concentration gradually increased in the absence of nicking endonuclease. Fig. 4A (inset) illustrates the relationship between T4 PNK concentrations and fluorescence intensity. The fluorescence intensity depends linearly on the concentrations of T4 PNK in the range from 0.05 to 10 U/mL with a correlation coefficient of 0.9911. According to the rule of three times standard deviation over the blank response, a detection limit of 1.9 10 2 U/mL is obtained. However, under the Nt.BbvCI-aided signal amplification (Fig. 4B), the fluorescent signal varied more obviously. A wider linear range was obtained from 0.005 to 5 U/mL with a correlation coefficient of 0.9955. The detection limit is calculated to be 6.6 10 4 U/mL based on 3 times standard
Fig. 3. Influence of common enzymes and proteins on the specificity of T4 PNK activity assay. (A) Bar chart of increase in fluorescence intensity in the presence of different enzymes and proteins utilizing strand displacement reaction coupled with enzymatic-aided amplification. Error bars were calculated from three replicate measurements. (B) Native 12.5% polyacrylamide gel analysis of specificity utilizing strand displacement reaction. In both (A) and (B), the concentrations of T4 PNK, T4 PNK (inactive), EcoRI, PKA were 1 U/mL and the concentrations of IgG, Lysozyme, BSA were 1.0 mM.
deviation over the blank response, which was superior or comparative with established approaches (Table S2). Based on the above performance, a solid conclusion is drawn that it offers a highly sensitive and specific method for detecting T4 PNK activity. In order to evaluate the reproducibility and stability of this system as a whole, 10 repetitive experiments for 0.1 U/mL T4 PNK were performed. The RSD of intra- (1.9%) and intergroup (4.4%) were all o5%. This indicates that our protocol has good reproducibility and stability. To further verify practicability, the capability to estimate T4 PNK activity in complex matrix was studied. 50% HeLa cell extracts were added into the reaction buffer to simulate the intracellular environment during the test procedure. As shown in Fig. 4C, a gradual increase in fluorescence intensity was observed as the concentration of T4 PNK increased in the range from 0.05 to
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
Fig. 4. (A) Fluorescence spectra of NMM with the increasing of T4 PNK concentrations in the absence of nicking endonuclease (Nt.BbvCI). (Inset) Dependence of fluorescence intensity on the logarithm of T4 PNK concentration from 0.05 U/mL to 10.0 U/mL. (B) Fluorescence spectra of NMM with the increasing of T4 PNK concentrations in the presence of nicking endonuclease (Nt.BbvCI). (Inset) Dependence of fluorescence intensity on the logarithm of T4 PNK concentration from 0.005 U/mL to 5.0 U/mL. Error bars were calculated from three replicate measurements. (C) Fluorescence response for detecting T4 PNK activity in pure buffer and diluted cell extracts (50%) respectively. Error bars were calculated from three replicate measurements.
143
Fig. 5. Inhibition effects of different concentrations of (A) ADP and (B) (NH4)2SO4 on phosphorylation utilizing strand displacement reaction coupled with enzymatic-aided amplification. In (A) and (B), the concentration of T4 PNK were 1 U/mL. The relative activity of T4 PNK was calculated via the ratio of (F(inhibit) F0) to (F F0), where F(inhibit) and F were the fluorescence intensity of NMM corresponding to 1 U/mL T4 PNK in the presence and absence of inhibitor, respectively, and the F0 was the fluorescence intensity of NMM/P1P2/hpDNA/PNK mixture. Error bars were calculated from three replicate measurements. (C) Native 12.5% polyacrylamide gel analysis of the inhibition effects utilizing strand displacement reaction. In (C), the concentrations of T4 PNK, ADP and (NH4)2SO4 were 1 U/mL, 4.0 mM and 4.0 mM, respectively.
144
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
5.0 U/mL. This result is in good agreement with that in pure reaction buffer without cell lysates, suggesting that this method could be applicable for T4 PNK detection in complex biological samples. 3.5. Inhibition evaluation of T4 PNK activity Recently Freschauf's group reported the PNK inhibition should potentially increase the sensitivity of human tumors to γ-radiation (Freschauf et al., 2010, 2009). That means the inhibition of PNK activity might be a promising direction against cancers (Allinson, 2010). Therefore the research on inhibition of PNK activity is worth conducting. We further investigated the effect of two model inhibitors (ADP and (NH4)2SO4) on the phosphorylation reaction of T4 PNK. With the fixed T4 PNK activity (1 U/mL), a gradual decrease in relative activity of T4 PNK was observed with the increase of ADP and (NH4)2SO4 concentrations (Figs. 5A and B), suggesting that T4 PNK activity has been inhibited. The relative activity of T4 PNK was calculated via the ratio of (F(inhibit) F0) to (F F0), where F(inhibit) and F were the fluorescence intensity of NMM corresponding to 1 U/mL T4 PNK in the presence and absence of inhibitor, respectively, and F0 was the fluorescence intensity of NMM/P1P2/hpDNA/PNK mixture. The IC50 value of ADP and (NH4)2SO4 were evaluated to be 1.45 mM and 11.0 mM, respectively, which are consistent with the reported IC50 values of 1 mM for ADP and 15 mM for (NH4)2SO4 (Liu et al., 2014b, 2014a; Du et al., 2014). The inhibition effect of ADP possibly resulted from the reversible phosphorylation reaction when ADP and 5′-phosphoryl nucleic acids coexisted in the reaction buffer (Hou et al., 2014). The salt effect on T4 PNK activity could be tentatively explained based on the following reasons. At high salt concentrations, the structure of hairpin DNA and dsDNA were more stable, which probably inhibited the activity of the 5′-hydroxyl group (Lillehaug et al., 1976). Also, the enzyme conformation might be affected by a high salts concentration which may weaken T4 PNK activity or the affinity between PNK and its substrates. To eliminate the possible influence of inhibitors (ADP and (NH4)2SO4) on the cyclic amplification and λ exo activity, 5′-phosphorylated double stranded P-DNA/C-DNA (sequences listed in Table S1) was designed as the substrate of λ exo. It is clear from Figure S5 that ADP and (NH4)2SO4 did not affect the activity of λ exo. Furthermore, the influence of the two inhibitors on the cyclic amplification and Nt.BbvCI activity was studied. As shown in Figure S6, ADP and (NH4)2SO4 did not affect the activity of Nt.BbvCI. These results clearly suggested that the proposed strategy may be extended to the study of inhibitors of T4 PNK. Furthermore, the gel electrophoresis result (Fig. 5C) reflected that ADP (lane 7) could inhibit T4 PNK activity more effectively than (NH4)2SO4 (lane 8) at same concentration and lead little strand displacement reaction compared with the control (lane 6). Thus ADP is a more effective T4 PNK inhibitor.
4. Conclusion In summary, a label-free fluorescent strategy has been developed for studying T4 PNK activity and inhibition. This approach integrated T4 PNK stimulated strand displacement reaction with enzyme-assisted amplification strategy to realize T4 PNK activity assay with good performance. The assay shares several distinct advantages. Firstly, this design allowed a homogeneous assay of T4 PNK activity without any additional modifications of DNA substrate, making the proposed strategy much facile, cost-effective. Meanwhile, excellent specificity is ensured by coupling high specificity of hairpin DNA with high accuracy of toehold-mediated strand displacement reaction. More importantly, the introduction
of nicking endonuclease further facilitates signal amplification, so as to achieve the aim of sensitive detection of T4 PNK activity with a detection limit of 6.6 10 4 U/mL. Additionally, the practicality of the proposed method was demonstrated by accurate detection of T4 PNK activity in complex matrix. Lastly, inhibition of phosphorylation can be reliably studied. Therefore, it provides a facile, easy readout and cost-effective manner for T4 PNK analysis and holds a great potential to be applied in the researches of DNA phosphorylation involved biological processes, drug development, and clinical diagnostics.
Acknowledgment This work was financially supported by the National Natural Science Foundation of China (No. 21075079, 21375086), the Fundamental Research Funds for the Central Universities (GK261001097), the Program for Changjiang Scholars and Innovative Research Team in University (No. 250123), the Program for Innovative Research Team in Shaanxi Province (No. 2014KCT28).
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.2015.05.059
References Allinson, S.L., 2010. Future Oncol. 6, 1031–1042. Breslin, C., Caldecott, K.W., 2009. Mol. Cell. Biol. 29, 4653–4662. Chen, F., Zhao, Y., Qi, L., Fan, C., 2013. Biosens. Bioelectron. 47, 218–224. Dapic, V., Abdomerovic, V., Marrington, R., Peberdy, J., Rodger, A., Trent, J.O., Bates, P.J., 2003. Nucleic Acids Res. 31, 2097–2107. Du, J., Xu, Q.F., Lu, X.Q., Zhang, C.Y., 2014. Anal. Chem. 86, 8481–8488. Freschauf, G.K., Mani, R.S., Mereniuk, T.R., Fanta, M., Virgen, C.A., Dianov, G.L., Grassot, J.M., Hall, D.G., Weinfeld, M., 2010. J. Biol. Chem. 285, 2351–2360. Freschauf, G.K., Karimi-Busheri, F., Ulaczyk-Lesanko, A., Mereniuk, T.R., Ahrens, A., Koshy, J.M., Rasouli-Nia, A., Pasarj, P., Holmes, C.F.B., Rininsland, F., Hall, D.G., Weinfeld, M., 2009. Cancer Res. 69, 7739–7746. Guo, Y.H., Wang, Q.L., Wang, Z.L., Chen, X., Xu, L.J., Hu, J.M., Pei, R.J., 2015. Sens. Actuators B 214, 50–55. He, H.Z., Leung, K.H., Wang, W., Chan, D.S., Leung, C.H., Ma, D.L., 2014. Chem. Commun. 50, 5313–5315. Higuchi, R.G., Ochman, H., 1989. Nucleic Acid Res. 17, 5865. Higgins, L.S., Besnier, C., Kong, H., 2001. Nucleic Acids Res. 29, 2492–2501. Hou, T., Wang, X.Z., Liu, X.J., Lu, T.T., Liu, S.F., Li, F., 2014. Anal. Chem. 86, 884–890. Huang, Y., Chen, J., Shi, M., Zhao, S., Chen, Z., Liang, H., 2013. J. Mater. Chem. B 1, 2018–2021. Jiang, C., Yan, C., Jiang, J., Yu, R., 2013. Anal. Chim. Acta 766, 88–93. Jiang, H.-X., Kong, D.-M., Shen, H.-X., 2014. Biosens. Bioelectron. 55, 133–138. Jiao, H., Wang, B., Chen, J., Liao, D., Li, W., Yu, C., 2012. Chem. Commun. 48, 7862–7864. Lin, L., Liu, Y., Zhao, X., Li, J.H., 2011. Anal. Chem. 83, 8396–8402. Lillehaug, J.R., Kleppe, R.K., Kleppe, K., 1976. Biochemistry 15, 1858–1865. Lin, L., Liu, Y., Yan, J., Wang, X.S., Li, J.H., 2013. Anal. Chem. 85, 334–340. Little, J.W., 1967. J. Biol. Chem. 242, 676–686. Liu, S.F., Ming, J.J., Lin, Y., Wang, C.F., Cheng, C.B., Liu, T., Wang, L., 2014a. Biosens. Bioelectron. 55, 225–230. Liu, S.F., Ming, J.J., Lin, Y., Wang, C.F., Liu, T., Cheng, C.B., Li, F., 2014b. Sens. Actuators, B 192, 157–163. Liu, S.F., Wang, Y., Zhang, C.X., Lin, Y., Li, F., 2013. Chem. Commun. 49, 2335–2337. Tao, M.J., Zhang, J., Jin, Y., Li, B.X., 2014. Anal. Biochem. 464, 63–69. Morgan, R.D., Calvet, C., Demeter, M., Agra, R., Kong, H., 2000. Biol. Chem. 381, 1123–1125. Novogrod, A., Tal, M., Traub, A., Hurwitz, J., 1966. J. Biol. Chem. 241, 2933–2943. Oh, S.S., Plakos, K., Lou, X., Xiao, Y., Soh, H.T., 2010. Proc. Natl. Acad. Sci. U. S. A 107, 14053–14058. Richards, C., 1965. Proc. Natl. Acad. Sci. U.S.A 54, 158–165. Rasouli-Nia, A., Karimi-Busheri, F., Weinfeld, M., 2004. Proc. Natl. Acad. Sci. U.S.A 101, 6905–6910. Shen, J., Gilmore, E.C., Marshall, C.A., Haddadin, M., Reynolds, J.J., Eyaid, W., Bodell, A., Barry, B., Gleason, D., Allen, K., Ganesh, V.S., Chang, B.S., Grix, A., Hill, R.S.,
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
Topcu, X., Caldecott, K.W., Barkovich, A.J., Walsh, C.A., 2010. Nat. Genet. 42, 245–249. Song, C., Zhao, M.P., 2009. Anal. Chem. 81, 1383–1388. Tahbaz, N., Subedi, S., Weinfeld, M., 2012. Nucleic Acids Res. 40, 3484–3495. Whitehouse, C.J., Taylor, R.M., Thistlethwaite, A., Zhang, H., Karimi-Busheri, F., Lasko, D.D., Weinfeld, M.K., Caldecott, W., 2001. Cell 104, 107–117. Wiederhold, L., Leppard, J.B., Kedar, P., Karimi-Busheri, F., Rasouli-Nia, A., Weinfeld, M., Tomkinson, A.E., Izumi, T., Prasad, R., Wilson, S.H., Mitra, S., Hazra, T.K., 2004. Mol. Cell 15, 209–220. Weinfeld, M., Mani, R.S., Abdou, I., Aceytuno, R.D., Glover, J.N.M., 2011. Trends Biochem. Sci. 36, 262–271. Wang, L.K., Lima, C.D., Shuman, S., 2002. EMBO J. 21, 3873–3880. Wu, W.H., Hu, H.Y., Li, F., Wang, L.H., Gao, J.M., Lu, J.X., Fan, C.H., 2011. Chem.
145
Commun. 47, 1201–1203. Wang, Y., He, X., Wang, K., Ni, X., Su, J., Chen, Z., 2012. Biosens. Bioelectron. 32, 213–218. Wang, G., He, X., Xu, G., Chen, L., Zhu, Y., Zhang, X., Wang, L., 2013. Biosens. Bioelectron. 43, 125–130. Wu, P., Tu, Y.Q., Qian, Y.D., Zhang, H., Cai, Ch.X., 2014. Chem. Commun. 50, 1012–1014. Wang, H., Hays, J.B., 2000. Mol. Biotechnol. 15, 97–104. Yurke, B., Turberfield, A.J., Mills, A.P., Simmel, F.C., Neumann, J.L., 2000. Nature 406, 605–608. Zhu, Z.-M., Yu, R.-Q., Chu, X., 2014. Anal. Methods 6, 6009–6014.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Label-free and sensitive detection of T4 polynucleotide kinase activity via coupling DNA strand displacement reaction with enzymatic-aided amplification Rui Cheng, Mangjuan Tao, Zhilu Shi, Xiafei Zhang, Yan Jin n, Baoxin Li Key Laboratory of Applied Surface and Colloid Chemistry, Ministry of Education, Key Laboratory of Analytical Chemistry for Life Science of Shaanxi Province, School of Chemistry and Chemical Engineering, Shaanxi Normal University, Xi'an 710062, China
art ic l e i nf o
a b s t r a c t
Article history: Received 13 February 2015 Received in revised form 5 May 2015 Accepted 26 May 2015 Available online 28 May 2015
Several fluorescence signal amplification strategies have been developed for sensitive detection of T4 polynucleotide kinase (T4 PNK) activity, but they need fluorescence dye labeled DNA probe. We have addressed the limitation and report here a label-free strategy for sensitive detection of PNK activity by coupling DNA strand displacement reaction with enzymatic-aided amplification. A hairpin oligonucleotide (hpDNA) with blunt ends was used as the substrate for T4 PNK phosphorylation. In the presence of T4 PNK, the stem of hpDNA was phosphorylated and further degraded by lambda exonuclease (λ exo) from 5′ to 3′ direction to release a single-stranded DNA as a trigger of DNA strand displacement reaction (SDR). The trigger DNA can continuously displace DNA P2 from P1/P2 hybrid with the help of specific cleavage of nicking endonuclease (Nt.BbvCI). Then, DNA P2 can form G-quadruplex in the presence of potassium ions and quadruplex-selective fluorphore, N-methyl mesoporphyrin IX (NMM), resulting in a significant increase in fluorescence intensity of NMM. Thus, the accumulative release of DNA P2 led to fluorescence signal amplification for determining T4 PNK activity with a detection limit of 6.6 10 4 U/mL, which is superior or comparative with established approaches. By ingeniously utilizing T4 PNK-triggered DNA SDR, T4 PNK activity can be specifically and facilely studied in homogeneous solution containing complex matrix without any external fluorescence labeling. Moreover, the influence of different inhibitors on the T4 PNK activity revealed that it also can be explored to screen T4 PNK inhibitors. Therefore, this label-free amplification strategy presents a facile and cost-effective approach for nucleic acid phosphorylation related research. & 2015 Elsevier B.V. All rights reserved.
Keywords: Label-free Fluorescence Signal amplification DNA phosphorylation PNK activity Strand displacement reaction
1. Introduction The phosphorylation of nucleic acids plays a critical basis for quickly and effectively repairing DNA damage, such as singlestrand break (Whitehouse et al., 2001) base excision, (Wiederhold et al., 2004) and oxidative DNA damage (Breslin and Caldecott, 2009). Polynucleotide kinase (PNK) can catalyze the transfer of the γ-phosphate group from a nucleoside triphosphate (the most used one is ATP ) to the 5′-hydroxyl terminus of DNA or RNA (Richards, 1965; Novogrod et al., 1966), which plays a vital role in the phosphorylation of nucleic acids. Besides, the human PNK participates in maintaining the genetic integrity (Rasouli-Nia et al., 2004) and is closely associated with some vital human diseases, such as Werner syndrome, Rothmund Thomson syndrome and Bloom's syndrome (Shen et al., 2010; Tahbaz et al., 2012; Little, n
Corresponding author. Fax: þ86 29 81530727. E-mail address: [email protected] (Y. Jin).
http://dx.doi.org/10.1016/j.bios.2015.05.059 0956-5663/& 2015 Elsevier B.V. All rights reserved.
1967). Recently Freschauf's group reported the PNK inhibition should potentially increase the sensitivity of human tumors to γradiation (Freschauf et al., 2010, 2009). That means the inhibition of PNK activity might be a promising direction against cancers (Allinson, 2010). Therefore, the research on the PNK activity is of great importance to the biomedical research (Weinfeld et al., 2011). Recently, researchers employ T4 polynucleotide kinase (T4 PNK) which belongs to one kind of polynucleotide kinase family as the model target to explore the assay of PNK activity. They have made great progress. Conventionally, the activity of T4 PNK was detected via these sophisticated protocols, such as radioisotope 32 P-labeling, autoradiography, and polyacrylamide gel electrophoresis (Whitehouse et al., 2001; Wang et al., 2002). However, they were rather complex, time consuming, laborious, expensive and may necessitate stringent safety measures to control radiographic exposure. Up to now, some novel strategies including fluorescence detection (Song and Zhao, 2009; Jiao et al., 2012; Chen et al., 2013; Wu et al., 2011; Lin et al., 2011; Huang et al.,
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
2013; Liu et al., 2014a, 2014b; Tao et al., 2014; Guo et al., 2015; Zhu et al., 2014; Jiang et al., 2014), colorimetric assay (Jiang et al., 2013), nanochannel biosensor (Lin et al., 2013), and electrochemical methods (Wang et al., 2012, 2013) were developed for convenient determination of T4 PNK activity. Among them, fluorescence methods are particularly attractive due to their high sensitivity, easy readout, small sample volume, simple operation and feasibility of quantification. Song and Zhao (2009) proposed a simple fluorescence approach for real-time monitoring of the activity and kinetics of T4 PNK by employing a singly fluorophore-labeled DNA-hairpin probe. Wu et al. (2011) and Lin et al. (2011), respectively, realized the detection of T4 PNK activity by using graphene oxide as a super quencher. With the development of signal amplification technique, Chen et al. (2013) developed a novel amplified fluorescence approach for T4 PNK assay based on nicking endonuclease-mediated signal amplification. Based on double-labeled molecular beacon, Liu et al. (2014a, 2014b) proposed a dual amplification strategy for sensitive detection of T4 polynucleotide kinase activity by coupling split DNAzyme and ligation-triggered DNAzyme cascade amplification. Although these fluorescence strategies have made great advances toward the DNA phosphorylation assay, they always utilize fluorescence dye double-labeled hairpins (Chen et al., 2013; Liu et al., 2014a, 2014b; Tao et al., 2014) or single-labeled probes combining with nanomaterials as quenchers (Wu et al., 2011; Lin et al., 2011). Label-free strategy is a good choice for realizing facile, reliable and cost-effective assay of T4 PNK activity. However, label-free method is rare (He et al., 2014; Guo et al., 2015; Jiang et al., 2014). Therefore, the development of facile, label-free and sensitive fluorescence strategy is necessary and meaningful for studying T4 PNK activity and inhibition. We have addressed the limitation and report here a label-free strategy for sensitive detection of T4 PNK activity by coupling DNA strand displacement reaction with enzymatic-aided amplification. Yurke and co-workers (Yurke et al., 2000) pioneered the concept of toehold-mediated DNA SDR which can realize signal amplification without the help of functional enzymes. A hairpin oligonucleotide (hpDNA) with blunt ends was used as the substrate for T4 PNK phosphorylation. In the presence of T4 PNK, the stem of hpDNA was phosphorylated and further degraded by lambda exonuclease (λ exo) from 5′ to 3′ direction to release a singlestranded DNA as a trigger of DNA strand displacement reaction. The trigger drives SDR and subsequent restriction enzymatic-assisted cyclic amplification to continuously release signal DNA. It folds into a quadruplex structure in the presence of potassium ions and quadruplex-selective dye, leading to obvious fluorescence readout. Thus, label-free detection of T4 PNK activity can be facilely achieved by ingeniously utilizing T4 PNK-induced strand displacement reaction and cyclic amplification, which ensures high specificity as well as high sensitivity. More importantly, it also offers a potential platform for screening T4 PNK inhibitors.
2. Materials and methods 2.1. Chemicals and materials The DNA sequences (listed in Table S1 in supporting information) were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). The T4 polynucleotide kinase (10 unit/μl), EcoRI endonuclease, adenosine triphosphate (ATP), adenosine diphosphate (ADP) and dithiothreitol (DTT) were purchased from Sangon Biotech Co., Ltd. (Shanghai, China). The Lambda exonuclease (λ exo, 5000 unit/mL), Nicking enzyme (Nt.BbvCI), 10 NEB buffer 4 and Protein kinase A (PKA) were purchased from New England Biolabs (NEB, U.K.). Lysozyme and immunoglobin G (IgG) were purchased
139
from Beijing Dingguo Biotechnology Co., Ltd (Beijing, China). Bovine serum albumin (BSA) was purchased from Sigma-Aldrich Inc. N-Methyl mesoporphyrin IX (NMM) (analytical reagent grade) was obtained from Frontier Scientific, Inc (Logan, Utah, USA). All DNA samples were prepared in 70 mM Tris–HCl buffer (pH 8.0, 10 mM MgCl2) prior to use. All other chemicals were of analytical grade and were used without further purification. Ultrapure water obtained from a Millipore filtration system was used throughout all experiments. 2.2. Detection and inhibition evaluation of T4 PNK activity All fluorescence measurements were executed on a fluorometer F-7000 (Hitachi, Japan). The excitation and emission wavelengths of NMM were set at 399 nm and 612 nm. Before the toehold-mediated strand displacement reaction, the DNA P1 and DNA P2, hpDNA were heated to 95 °C for 5 min in 70 mM Tris–HCl buffer (pH ¼8.0, 10 mM MgCl2) followed by slowly cooling to room temperature for 1 h before use. Firstly the phosphorylation and λ exo cleavage reaction were simultaneously performed in 100 μL enzyme reaction buffer (70 mM Tris–HCl buffer, pH 8.0, 10 mM MgCl2, 1 mM ATP and 5 mM DTT). 0.5 μM hpDNA was incubated with certain amount of T4 PNK and 50 U/mL λ exo at 37 °C for 30 min and then the reaction solution was heated to 75 °C for 10 min to deactivate the enzymes and gradually cooled to 25 °C. Then 1 μM P1/P2 hybrid, 10 mM KCl, 1 μM NMM, and 100 μL 1 NEB buffer 4 were added into the above reaction solution and the fluorescence measurement was carried out to obtain the signal before amplification. Subsequently 40 U/mL Nt.BbvCI was added into the solution to activate the enzyme-aided circle amplification and kept at 37 °C for 1 h prior to fluorescence measurement. For the T4 PNK inhibition assay, both ADP and (NH4)2SO4 were used as the model inhibitors. Various concentrations of inhibitors were separately mixed with 0.5 μM hpDNA, 1 U/mL T4 PNK and 50 U/mL λ exo in 100 μL enzyme reaction buffer. The phosphorylation and enzymatic amplification reaction were carried out via similar procedures as those for T4 PNK activity assay stated above. 2.3. T4 PNK selectivity studies In order to investigate the influence of enzymes and proteins, 1 U/mL of EcoRI/PKA and 1.0 mM of IgG/Lysozyme/BSA were added into the buffer respectively. The phosphorylation and enzymatic amplification reaction were the same as those described in the aforementioned experiment for T4 PNK activity detection in reaction buffer. 2.4. Gel electrophoresis The label-free strategy for sensitive T4 PNK assay was validated by polyacrylamide gel electrophoresis (PAGE). The phosphorylation and enzymatic amplification reaction were carried out via similar procedures as those for T4 PNK assay stated above. 10 μL samples were added 1.0 μL 10 loading buffer and then loaded on a 12.5% native polyacrylamide gel. Electrophoresis was performed at a constant voltage of 200 V for 30 min in 1 Tris-borate-EDTA (TBE) buffer. After silver staining, the gel electrophoresis was visualized on the molecular imager system (JS 180, Shanghai Peiqing science & Technology. Co., Ltd). 2.5. Circular dichroism spectroscopy CD spectroscopy is measured by using a Chirascan Circular Dichroism Spectrometer (Applied Photophysics Ltd, England, UK). CD spectra was performed using an optical chamber (1 mm optical path length) with an instrument scanning speed of 100 nm/min, a
140
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
response time of 2 s at room temperature and was accumulated by taking the average of three scans made from 220 to 300 nm. 2.6. Assay of T4 PNK activity in diluted cell extracts HeLa cells were cultured in Dulbecco's modified Eagle's medium (DMEM, GIBCO) supplemented with 10% fetal calf serum (FCS, Sigma), penicillin (100 μg/mL), and streptomycin (100 μg/mL) at 37 °C in a humidified chamber, containing 5% CO2. Cell extracts were prepared according to the previous reports (Wu et al., 2014). The harvested cells were resuspended in 20 μL 10 mM Tris–HCl (pH 7.8) containing 150 mM NaCl. Equal volume of lysis buffer was added into the suspension containing 10 mM Tris–HCl (pH 7.8), 150 mM NaCl, 2 mM EDTA, 2 mM DTT, 40% glycerol, 0.2% NP-40, and 0.4 mM phenylmethylsulfonyl fluoride. The mixture was incubated for 1.5 h at 4 °C with occasional shake. Cell debris was removed by centrifugation at 16,000 rpm for 10 min, and the supernatant was recovered. Diluted cell extracts were added to the assay solution (50%). The detection procedure was the same as those described in the aforementioned experiment for T4 PNK detection in pure reaction buffer.
3. Results and discussion 3.1. Proof of principle The label-free fluorescent approach for T4 PNK assay is based on the PNK-triggered strand displacement reaction (SDR) and enzyme-aided cyclic amplification (Scheme 1). In this work, N-Methyl mesoporphyrin IX (NMM) was chosen as the signal reporter because NMM is an anionic porphyrin that has a high structural selectivity for quadruplexes but not for duplexes, triplexes, or single-stranded forms (Oh et al., 2010). NMM itself shows tiny fluorescence. However, a significant increase in its fluorescence can be achieved upon binding to G-quadruplex. A hairpin oligonucleotide (hpDNA) was designed as a T4 PNK substrate whose 5′-hydroxyl terminal could be phosphorylated by T4 PNK and subsequently digested by λ exonuclease (λ exo) to generate a trigger DNA. Nt.BbvCI is a nicking endonuclease that cleaves only one strand of DNA of a double-stranded DNA substrate at the specific recognition site. DNA P1 contains a recognition site of Nt.BbvCI. DNA P1/P2 hybrid keeps stable in the presence of Nt.BbvCI because the recognition site of Nt.BbvCI located
Fig. 1. Fluorescence spectra of NMM under different conditions. The concentrations of P1/P2, hpDNA, T4 PNK, λ exo and Nt.BbvCI were 1 μM, 0.5 μM, 1 U/mL, 50 U/mL, 40 U/mL, respectively.
Scheme 1. Schematic illustration of label-free DNA strand displacement reaction with enzymatic-aided amplification for monitoring T4 PNK activity.
outside the double-stranded domain of P1/P2 hybrid. Only when trigger DNA displaced P2 from P1/P2 hybrid to form longer and more stable dsDNA can DNA P1 was specifically cleaved by Nt. BbvCI to release trigger DNA again for the next cycle of SDR. The continuously released P2 led to fluorescence amplification because
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
the G-rich oligomer P2 can fold into a G-quadruplex structure in the presence of potassium ions and NMM. By the elaborate design, a label-free detection of T4 PNK activity could be performed with high sensitivity and specificity. To estimate the feasibility of the amplification strategy, the fluorescence spectra of NMM under different conditions was recorded in Fig. 1. The fluorescence intensity of NMM alone is very weak (Fig. 1A, curve a). The introduction of both P1/P2 hybrid and hpDNA led negligible fluorescence change (Fig. 1A, curve b and c). Hairpin DNA highly ensured the specificity of SDR. P1/P2 hybrid cannot open the hairpin structure to trigger SDR for releasing DNA P2. Therefore, the fluorescence intensity of NMM remained unchanged in the presence of hpDNA and P1/P2 hybrid. However, the fluorescence of NMM had a remarkably enhancement when hpDNA was incubated with T4 PNK and λ exo (Fig. 1A, curve e). That is, the trigger of SDR had been successfully released because λ exo is a highly processive 5′–3′ exonuclease which prefers a phosphate moiety at 5′ end of dsDNA (Higuchi and Ochman, 1989; Little, 1967) and immediately SDR occurred. Moreover, a further increase in fluorescence intensity of NMM was obtained when Nt. BbvCI was added to the above mixture. It is clear that a 286.727 0.64% increase in fluorescence signal was achieved (Fig. 1A, curve f). However in the absence of Nt.BbvCI, the corresponding fluorescence enhancement was only 66.64 70.52% with the same concentration of T4 PNK (Fig. 1A, curve e). It is mainly attributed to the circle amplification stimulated by nicking endonucleases which can cut one strand of a double-stranded DNA at a specific recognition nucleotide sequence known as recognition site of endonuclease (Morgan et al., 2000; Higgins et al., 2001; Wang and Hays, 2000). Therefore, these preliminary results demonstrated the feasibility of this signal amplification strategy. To further ensure reliability, more control experiments were conducted. Firstly, the influence of nonspecific cleavage was investigated. When the T4 PNK was inactive, the fluorescence intensity hardly changed (Fig. 1B, curve d) because the 5′ terminal of hpDNA still retained hydroxylation form which prohibited the digestion by λ exo. When Nt.BbvCI was introduced (Fig. 1B, curve e), a slight increase ( 3.4%) in fluorescence intensity was observed. It is mainly attributed to the cleavage of Nt.BbvCI due to the little and partial hybridization of hpDNA with P1. However, the above cleavage hardly affects the analysis of T4 PNK activity because it can be eliminated as background. Then, the effect of λ exo on the fluorescence was examined. As shown in Fig. 1C, the fluorescence intensity of NMM did not change when hpDNA has been successfully phosphorylated by T4 PNK without λ exo (Fig. 1C, curve b). Meanwhile, it is obvious that the fluorescence intensity was unchanged when hpDNA was co-incubated with inactive T4 PNK and λ exo (Fig. 1C, curve c). That means the releasing of trigger was originated from the T4 PNK-actuated λ exo digestion of the stem region of hpDNA (Fig. 1C, curve d). These results clearly suggested that only the combination of T4 PNK and λ exo can result in a remarkable fluoresence increase. Therefore, this label-free amplification strategy can reliably detect T4 PNK activity. 3.2. Validation by gel electrophoresis and circular dichroism spectroscopy To further verify the sensing mechanism, the strand displacement reaction and enzymatic amplification results have been proved using a 12.5% native polyacrylamide gel electrophoresis (PAGE). As shown in Fig. 2, the bands in lane 2, 3, 4 correspond to hpDNA, P1, P2, respectively. A distinct band of hybrid P1/P2 (the molecular weight is the sum of two single probes) is observed in lane 5. In the presence of T4 PNK and λ exo, the color of bands corresponding to both hpDNA and P1/P2 was made lighter, and a
141
Fig. 2. Native 12.5% polyacrylamide gel analysis for proof of principle. The concentrations of P1/P2, P1, P2 and hpDNA were 1.0 mM, 1.0 mM, 1.0 mM and 0.5 mM. The concentrations of T4 PNK, T4 PNK (inactive), λ exo and Nt.BbvCI were 1 U/mL, 1 U/mL, 50 U/mL and 40 U/mL, respectively.
new band corresponding to DNA P2 appeared, demonstrating that the phosphorylated hpDNA can be degraded by λ exo to release the trigger. The trigger DNA hybridized with P1 to form more stable double-stranded DNA as a result of replacing P2 (Fig. 2, lane 6, featured with red dotted line). It is accorded with the essence of strand displacement reaction that the molecular weight of new dsDNA was larger than that of P1/P2. Upon the addition of Nt. BbvCI, both that new band and P2 band deepened. Meanwhile, hpDNA and P1/P2 further diminished, and new bands of short DNA fragments appeared (Fig. 2, lane 7, featured with red solid line). The result indicated P1 was cleaved by Nt.BbvCI and the trigger can be released to replace P2 circularly. However, the P2 band and new band were not observed in the presence of inactive T4 PNK (Fig. 2, lane 8, 9). Both fluorescence and PAGE results illustrated that the amplification strategy really worked on the basis of the T4 PNK-triggered strand displacement reaction and Nt. BbvCI-aided circle amplification. To verify the formation of G-quadruplex, circular dichroism (CD) measurement was utilized to monitor the conformation changes of P2 under various conditions. It is known that the “parallel” quadruplexes have a CD spectrum characterized by a positive ellipticity maximum at 264 nm and a negative minimum at 240 nm (Dapic et al., 2003). In the presence of K þ and NMM, P2 can form a “parallel” quadruplex (Figure S1A). It is clear in Figure S1 A and B that K þ is the key element to induce P2 to form G-quadruplex which will facilitate the binding of NMM. 3.3. Optimization of the related factors NMM was used as the signal molecule for fluorescence detection of T4 PNK activity. Meanwhile, potassium ion plays an
142
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
important role in stabilizing quadruplex. So, firstly, the concentrations of NMM and K þ were optimized. The data in Figure S2 revealed that the highest fluorescence intensity was obtained when the final concentrations of NMM and K þ were 1 mM and 10 mM, respectively. Secondly, interaction time was a crucial parameter for the signal amplification. As shown in Figure S3, the strand displacement reaction and enzymatic amplification were finished in 20 and 30 min, respectively. It is well-known ATP is used as the donor for phosphorylation (Liu et al., 2013). Furthermore, the effect of ATP concentration on the T4 PNK activity was investigated. As shown in Figure S4A, the fluorescence intensity increased with the increase of ATP concentration and tended to be constant at 1 mM. Thus ATP concentration of 1 mM was used in the further experiments. As a key element for triggering signal transmission, the amount of λ exo was also tested. It can be seen from Figure S4B, the fluorescence enhancement efficiency increased gradually with the increasing of λ exo concentrations and reached a plateau beyond 50 U/mL. In order to obtain cost-effective digestion by λ exo, the concentration of 50 U/mL was adopted in this method. Because Nt.BbvCI was served as the tool for signal amplification, the concentration of Nt.BbvCI should be carefully controlled. The fluorescence enhancement efficiency greatly enhanced with the increasing concentrations of Nt.BbvCI and the optimal concentration was selected at 40 U/mL (Figure S4C). Due to the competion hybridization between P2 and trigger, the molar ratios of P1/P2 to hpDNA also significantly affect the signal amplification efficiency. Herein, the effects of molar ratio was further investigated with a mixed concentration of P1/P2 (1 mM) and various hpDNA concentrations from 0.1 μM to 1 μM (Figure S4D). It was observed that the fluorescence response was quickly enhanced as the hpDNA concentration increased. However, the increase of hpDNA concentrations also accompanied with the increasing background signal. Finally, the molar ratio of 2:1 (P1/P2: hpDNA) was used throughout the experiments. 3.4. Specificity and sensitivity of T4 PNK activity assay To verify the specificity of T4 PNK detection, the influences of five common enzymes and proteins, including EcoRI, PKA, Lysozyme, IgG and BSA were tested under the identical concentration (1 U/mL or 1 mM). As shown in Fig. 3A, there was no remarkable fluorescence increase upon the addition of above enzymes and proteins except T4 PNK. It again demonstrated that only phosphorylated hpDNA can be specifically digested by λ exo to release the trigger for cyclic amplification. So, it is a highly specific method for detection of T4 PNK activity. In addition, the specificity was further verified by PAGE. As shown in Fig. 3B, only T4 PNK produced two new and remarkable bands which stand for trigger/P1 hybrid and P2, respectively. It is consistent with fluorescent result, indicating that an excellent specificity has been ensured by smart design of DNA probe. Under the optimal conditions, the analytical performance of the proposed assay for the detection of T4 PNK activity is investigated. A continuous increase in fluorescence intensity has been found in Fig. 4A when T4 PNK concentration gradually increased in the absence of nicking endonuclease. Fig. 4A (inset) illustrates the relationship between T4 PNK concentrations and fluorescence intensity. The fluorescence intensity depends linearly on the concentrations of T4 PNK in the range from 0.05 to 10 U/mL with a correlation coefficient of 0.9911. According to the rule of three times standard deviation over the blank response, a detection limit of 1.9 10 2 U/mL is obtained. However, under the Nt.BbvCI-aided signal amplification (Fig. 4B), the fluorescent signal varied more obviously. A wider linear range was obtained from 0.005 to 5 U/mL with a correlation coefficient of 0.9955. The detection limit is calculated to be 6.6 10 4 U/mL based on 3 times standard
Fig. 3. Influence of common enzymes and proteins on the specificity of T4 PNK activity assay. (A) Bar chart of increase in fluorescence intensity in the presence of different enzymes and proteins utilizing strand displacement reaction coupled with enzymatic-aided amplification. Error bars were calculated from three replicate measurements. (B) Native 12.5% polyacrylamide gel analysis of specificity utilizing strand displacement reaction. In both (A) and (B), the concentrations of T4 PNK, T4 PNK (inactive), EcoRI, PKA were 1 U/mL and the concentrations of IgG, Lysozyme, BSA were 1.0 mM.
deviation over the blank response, which was superior or comparative with established approaches (Table S2). Based on the above performance, a solid conclusion is drawn that it offers a highly sensitive and specific method for detecting T4 PNK activity. In order to evaluate the reproducibility and stability of this system as a whole, 10 repetitive experiments for 0.1 U/mL T4 PNK were performed. The RSD of intra- (1.9%) and intergroup (4.4%) were all o5%. This indicates that our protocol has good reproducibility and stability. To further verify practicability, the capability to estimate T4 PNK activity in complex matrix was studied. 50% HeLa cell extracts were added into the reaction buffer to simulate the intracellular environment during the test procedure. As shown in Fig. 4C, a gradual increase in fluorescence intensity was observed as the concentration of T4 PNK increased in the range from 0.05 to
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
Fig. 4. (A) Fluorescence spectra of NMM with the increasing of T4 PNK concentrations in the absence of nicking endonuclease (Nt.BbvCI). (Inset) Dependence of fluorescence intensity on the logarithm of T4 PNK concentration from 0.05 U/mL to 10.0 U/mL. (B) Fluorescence spectra of NMM with the increasing of T4 PNK concentrations in the presence of nicking endonuclease (Nt.BbvCI). (Inset) Dependence of fluorescence intensity on the logarithm of T4 PNK concentration from 0.005 U/mL to 5.0 U/mL. Error bars were calculated from three replicate measurements. (C) Fluorescence response for detecting T4 PNK activity in pure buffer and diluted cell extracts (50%) respectively. Error bars were calculated from three replicate measurements.
143
Fig. 5. Inhibition effects of different concentrations of (A) ADP and (B) (NH4)2SO4 on phosphorylation utilizing strand displacement reaction coupled with enzymatic-aided amplification. In (A) and (B), the concentration of T4 PNK were 1 U/mL. The relative activity of T4 PNK was calculated via the ratio of (F(inhibit) F0) to (F F0), where F(inhibit) and F were the fluorescence intensity of NMM corresponding to 1 U/mL T4 PNK in the presence and absence of inhibitor, respectively, and the F0 was the fluorescence intensity of NMM/P1P2/hpDNA/PNK mixture. Error bars were calculated from three replicate measurements. (C) Native 12.5% polyacrylamide gel analysis of the inhibition effects utilizing strand displacement reaction. In (C), the concentrations of T4 PNK, ADP and (NH4)2SO4 were 1 U/mL, 4.0 mM and 4.0 mM, respectively.
144
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
5.0 U/mL. This result is in good agreement with that in pure reaction buffer without cell lysates, suggesting that this method could be applicable for T4 PNK detection in complex biological samples. 3.5. Inhibition evaluation of T4 PNK activity Recently Freschauf's group reported the PNK inhibition should potentially increase the sensitivity of human tumors to γ-radiation (Freschauf et al., 2010, 2009). That means the inhibition of PNK activity might be a promising direction against cancers (Allinson, 2010). Therefore the research on inhibition of PNK activity is worth conducting. We further investigated the effect of two model inhibitors (ADP and (NH4)2SO4) on the phosphorylation reaction of T4 PNK. With the fixed T4 PNK activity (1 U/mL), a gradual decrease in relative activity of T4 PNK was observed with the increase of ADP and (NH4)2SO4 concentrations (Figs. 5A and B), suggesting that T4 PNK activity has been inhibited. The relative activity of T4 PNK was calculated via the ratio of (F(inhibit) F0) to (F F0), where F(inhibit) and F were the fluorescence intensity of NMM corresponding to 1 U/mL T4 PNK in the presence and absence of inhibitor, respectively, and F0 was the fluorescence intensity of NMM/P1P2/hpDNA/PNK mixture. The IC50 value of ADP and (NH4)2SO4 were evaluated to be 1.45 mM and 11.0 mM, respectively, which are consistent with the reported IC50 values of 1 mM for ADP and 15 mM for (NH4)2SO4 (Liu et al., 2014b, 2014a; Du et al., 2014). The inhibition effect of ADP possibly resulted from the reversible phosphorylation reaction when ADP and 5′-phosphoryl nucleic acids coexisted in the reaction buffer (Hou et al., 2014). The salt effect on T4 PNK activity could be tentatively explained based on the following reasons. At high salt concentrations, the structure of hairpin DNA and dsDNA were more stable, which probably inhibited the activity of the 5′-hydroxyl group (Lillehaug et al., 1976). Also, the enzyme conformation might be affected by a high salts concentration which may weaken T4 PNK activity or the affinity between PNK and its substrates. To eliminate the possible influence of inhibitors (ADP and (NH4)2SO4) on the cyclic amplification and λ exo activity, 5′-phosphorylated double stranded P-DNA/C-DNA (sequences listed in Table S1) was designed as the substrate of λ exo. It is clear from Figure S5 that ADP and (NH4)2SO4 did not affect the activity of λ exo. Furthermore, the influence of the two inhibitors on the cyclic amplification and Nt.BbvCI activity was studied. As shown in Figure S6, ADP and (NH4)2SO4 did not affect the activity of Nt.BbvCI. These results clearly suggested that the proposed strategy may be extended to the study of inhibitors of T4 PNK. Furthermore, the gel electrophoresis result (Fig. 5C) reflected that ADP (lane 7) could inhibit T4 PNK activity more effectively than (NH4)2SO4 (lane 8) at same concentration and lead little strand displacement reaction compared with the control (lane 6). Thus ADP is a more effective T4 PNK inhibitor.
4. Conclusion In summary, a label-free fluorescent strategy has been developed for studying T4 PNK activity and inhibition. This approach integrated T4 PNK stimulated strand displacement reaction with enzyme-assisted amplification strategy to realize T4 PNK activity assay with good performance. The assay shares several distinct advantages. Firstly, this design allowed a homogeneous assay of T4 PNK activity without any additional modifications of DNA substrate, making the proposed strategy much facile, cost-effective. Meanwhile, excellent specificity is ensured by coupling high specificity of hairpin DNA with high accuracy of toehold-mediated strand displacement reaction. More importantly, the introduction
of nicking endonuclease further facilitates signal amplification, so as to achieve the aim of sensitive detection of T4 PNK activity with a detection limit of 6.6 10 4 U/mL. Additionally, the practicality of the proposed method was demonstrated by accurate detection of T4 PNK activity in complex matrix. Lastly, inhibition of phosphorylation can be reliably studied. Therefore, it provides a facile, easy readout and cost-effective manner for T4 PNK analysis and holds a great potential to be applied in the researches of DNA phosphorylation involved biological processes, drug development, and clinical diagnostics.
Acknowledgment This work was financially supported by the National Natural Science Foundation of China (No. 21075079, 21375086), the Fundamental Research Funds for the Central Universities (GK261001097), the Program for Changjiang Scholars and Innovative Research Team in University (No. 250123), the Program for Innovative Research Team in Shaanxi Province (No. 2014KCT28).
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.2015.05.059
References Allinson, S.L., 2010. Future Oncol. 6, 1031–1042. Breslin, C., Caldecott, K.W., 2009. Mol. Cell. Biol. 29, 4653–4662. Chen, F., Zhao, Y., Qi, L., Fan, C., 2013. Biosens. Bioelectron. 47, 218–224. Dapic, V., Abdomerovic, V., Marrington, R., Peberdy, J., Rodger, A., Trent, J.O., Bates, P.J., 2003. Nucleic Acids Res. 31, 2097–2107. Du, J., Xu, Q.F., Lu, X.Q., Zhang, C.Y., 2014. Anal. Chem. 86, 8481–8488. Freschauf, G.K., Mani, R.S., Mereniuk, T.R., Fanta, M., Virgen, C.A., Dianov, G.L., Grassot, J.M., Hall, D.G., Weinfeld, M., 2010. J. Biol. Chem. 285, 2351–2360. Freschauf, G.K., Karimi-Busheri, F., Ulaczyk-Lesanko, A., Mereniuk, T.R., Ahrens, A., Koshy, J.M., Rasouli-Nia, A., Pasarj, P., Holmes, C.F.B., Rininsland, F., Hall, D.G., Weinfeld, M., 2009. Cancer Res. 69, 7739–7746. Guo, Y.H., Wang, Q.L., Wang, Z.L., Chen, X., Xu, L.J., Hu, J.M., Pei, R.J., 2015. Sens. Actuators B 214, 50–55. He, H.Z., Leung, K.H., Wang, W., Chan, D.S., Leung, C.H., Ma, D.L., 2014. Chem. Commun. 50, 5313–5315. Higuchi, R.G., Ochman, H., 1989. Nucleic Acid Res. 17, 5865. Higgins, L.S., Besnier, C., Kong, H., 2001. Nucleic Acids Res. 29, 2492–2501. Hou, T., Wang, X.Z., Liu, X.J., Lu, T.T., Liu, S.F., Li, F., 2014. Anal. Chem. 86, 884–890. Huang, Y., Chen, J., Shi, M., Zhao, S., Chen, Z., Liang, H., 2013. J. Mater. Chem. B 1, 2018–2021. Jiang, C., Yan, C., Jiang, J., Yu, R., 2013. Anal. Chim. Acta 766, 88–93. Jiang, H.-X., Kong, D.-M., Shen, H.-X., 2014. Biosens. Bioelectron. 55, 133–138. Jiao, H., Wang, B., Chen, J., Liao, D., Li, W., Yu, C., 2012. Chem. Commun. 48, 7862–7864. Lin, L., Liu, Y., Zhao, X., Li, J.H., 2011. Anal. Chem. 83, 8396–8402. Lillehaug, J.R., Kleppe, R.K., Kleppe, K., 1976. Biochemistry 15, 1858–1865. Lin, L., Liu, Y., Yan, J., Wang, X.S., Li, J.H., 2013. Anal. Chem. 85, 334–340. Little, J.W., 1967. J. Biol. Chem. 242, 676–686. Liu, S.F., Ming, J.J., Lin, Y., Wang, C.F., Cheng, C.B., Liu, T., Wang, L., 2014a. Biosens. Bioelectron. 55, 225–230. Liu, S.F., Ming, J.J., Lin, Y., Wang, C.F., Liu, T., Cheng, C.B., Li, F., 2014b. Sens. Actuators, B 192, 157–163. Liu, S.F., Wang, Y., Zhang, C.X., Lin, Y., Li, F., 2013. Chem. Commun. 49, 2335–2337. Tao, M.J., Zhang, J., Jin, Y., Li, B.X., 2014. Anal. Biochem. 464, 63–69. Morgan, R.D., Calvet, C., Demeter, M., Agra, R., Kong, H., 2000. Biol. Chem. 381, 1123–1125. Novogrod, A., Tal, M., Traub, A., Hurwitz, J., 1966. J. Biol. Chem. 241, 2933–2943. Oh, S.S., Plakos, K., Lou, X., Xiao, Y., Soh, H.T., 2010. Proc. Natl. Acad. Sci. U. S. A 107, 14053–14058. Richards, C., 1965. Proc. Natl. Acad. Sci. U.S.A 54, 158–165. Rasouli-Nia, A., Karimi-Busheri, F., Weinfeld, M., 2004. Proc. Natl. Acad. Sci. U.S.A 101, 6905–6910. Shen, J., Gilmore, E.C., Marshall, C.A., Haddadin, M., Reynolds, J.J., Eyaid, W., Bodell, A., Barry, B., Gleason, D., Allen, K., Ganesh, V.S., Chang, B.S., Grix, A., Hill, R.S.,
R. Cheng et al. / Biosensors and Bioelectronics 73 (2015) 138–145
Topcu, X., Caldecott, K.W., Barkovich, A.J., Walsh, C.A., 2010. Nat. Genet. 42, 245–249. Song, C., Zhao, M.P., 2009. Anal. Chem. 81, 1383–1388. Tahbaz, N., Subedi, S., Weinfeld, M., 2012. Nucleic Acids Res. 40, 3484–3495. Whitehouse, C.J., Taylor, R.M., Thistlethwaite, A., Zhang, H., Karimi-Busheri, F., Lasko, D.D., Weinfeld, M.K., Caldecott, W., 2001. Cell 104, 107–117. Wiederhold, L., Leppard, J.B., Kedar, P., Karimi-Busheri, F., Rasouli-Nia, A., Weinfeld, M., Tomkinson, A.E., Izumi, T., Prasad, R., Wilson, S.H., Mitra, S., Hazra, T.K., 2004. Mol. Cell 15, 209–220. Weinfeld, M., Mani, R.S., Abdou, I., Aceytuno, R.D., Glover, J.N.M., 2011. Trends Biochem. Sci. 36, 262–271. Wang, L.K., Lima, C.D., Shuman, S., 2002. EMBO J. 21, 3873–3880. Wu, W.H., Hu, H.Y., Li, F., Wang, L.H., Gao, J.M., Lu, J.X., Fan, C.H., 2011. Chem.
145
Commun. 47, 1201–1203. Wang, Y., He, X., Wang, K., Ni, X., Su, J., Chen, Z., 2012. Biosens. Bioelectron. 32, 213–218. Wang, G., He, X., Xu, G., Chen, L., Zhu, Y., Zhang, X., Wang, L., 2013. Biosens. Bioelectron. 43, 125–130. Wu, P., Tu, Y.Q., Qian, Y.D., Zhang, H., Cai, Ch.X., 2014. Chem. Commun. 50, 1012–1014. Wang, H., Hays, J.B., 2000. Mol. Biotechnol. 15, 97–104. Yurke, B., Turberfield, A.J., Mills, A.P., Simmel, F.C., Neumann, J.L., 2000. Nature 406, 605–608. Zhu, Z.-M., Yu, R.-Q., Chu, X., 2014. Anal. Methods 6, 6009–6014.
