Modular Nuclease-Responsive DNA Three-Way Junction-Based

Subscriber access provided by GAZI UNIV

Article

Modular nuclease-responsive DNA three-way junctions-based dynamic assembly of DNA device and its sensing application Jing Zhu, Lei Wang, Xiaowen Xu, Haiping Wei, and Wei Jiang Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b04889 • Publication Date (Web): 04 Mar 2016 Downloaded from http://pubs.acs.org on March 6, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Modular nuclease-responsive DNA three-way junctions-based dynamic assembly of DNA device and its sensing application Jing Zhu a, Lei Wang b, Xiaowen Xu a, Haiping Wei a, Wei Jiang *, a a

Key Laboratory for Colloid and Interface Chemistry of Education Ministry, School of Chemistry and Chemical Engineering, Shandong University, 250100 Jinan, P.R. China b School of Pharmaceutical Sciences, Shandong University, 250012 Jinan, P.R. China ABSTRACT: Here we explored a modular strategy for rational design of nuclease-responsive three-way junctions (TWJs) and fabricated a dynamic DNA device in a ‘plug-and-play’ fashion. Firstly, the inactivated TWJs were designed which contained three functional domains: the inaccessible the toehold and branch migration domains, the specific sites of nucleases and the auxiliary complementary sequence. The actions of different nucleases on their specific sites in TWJs caused the close proximity of the same toehold and branch-migration domains, resulting in the activation of the TWJs and the formation of a universal trigger for the subsequent dynamic assembly. Secondly, two hairpins H1 and H2 was introduced which could coexist in metastable state initially to act as the components for the dynamic assembly. Once the trigger initiated the opening of H1 via TWJs-driven strand displacement, the cascade hybridization of hairpins immediately switched on, resulting in the formation of the concatemers of H1/H2 complex appending numerous integrated G-quadruplexes which were used to obtain label-free signal readout. The inherent modularity of this design allowed us to fabricate a flexible DNA dynamic device and detect multiple nucleases through altering the recognition pattern slightly. Taking uracil-DNA glycosylase and CpG methyltransferase M.SssI as models, we successfully realized the butt joint between the uracil-DNA glycosylase and M.SssI recognition events and the dynamic assembly process. Furthermore, we achieved ultrasensitive assay of nuclease activity and the inhibitor screening. The DNA device proposed here will offer an adaptive and flexible tool for clinical diagnosis and anticancer drug discovery.

DNA three-way junction (TWJ), a typically Y-shaped DNA structure composed of three individual DNA strands hybridized with each other, is demonstrated to be a flexible building block in DNA dynamic assemblies.18-23 Branched structures offer a unique window into strand displacement reaction. Because the toehold and branch migration domains can be allocated on two of branches, the third branch can be flexibly adapted according to different targets to trigger the following strand displacement reaction.24 The toehold and branch migration domains are inaccessible and cannot form an 40 activated trigger until the specific recognition events have been executed.25, 26 Based on this, we proposed a modular TWJ-based toehold activation strategy. Taking different nucleases as specific targets, multiple nucleases activities could be translated into a readily available trigger sequence by different transducing mechanisms, and could further act as the trigger module to initiate a unique strand displacement reaction. Nucleases, such as DNA modified enzymes, are vital to the regulation of gene expression and the maintenance of ge27, 28 50 nomic integrity. Aberrant expression of DNA modified enzymes would interfere with the epigenetic reprogramming of gene patterns that alter chromatin structures and become the initiator of many diseases, including cancers, Huntington’s disease, and psychiatric disorders.29, 30 Therefore, the activities assay of DNA modified enzymes represent a valuable tool for clinical diagnostics and therapeutics. This strategy is demonstrated using two model enzymes, uracil-DNA glycosylase and CpG methyltransferase M.SssI. The action of these two enzymes on their respective DNA substrate can induce the prox60 imity of toehold and branch-migration domains in TWJs 30

INTRODUCTION DNA has been proven to be a powerful and intriguing material in the field of nanotechnology that enables programming of the kinetically controlled assembly of DNA nanostructures in a predictable manner.1-3 By exploitation of the toeholdmediated strand displacement reaction,4, 5 a series of dynamic DNA devices, including nanomachines,6, 7 logic circuits,8, 9 and catalytic amplifiers,10, 11 have been rationally constructed. Generally, the strand displacement reaction is initiated by hy10 bridization at the toehold domain and completed through a branch migration process. Thus engineering control of dynamic DNA devices can be realized by programming sequestration and activation of the toehold. The addition of specific molecular inputs (analytes or “fuels”) or the environmental changes such as pH could activate the strand displacement reaction.12-14 Recently, Liu et al. developed a “hidden toehold” activation strategy by caging a toehold domain in an arched structure and activating it by subsequent binding of ATP and the aptamer. 15 In addition, the “metallo-toehold” and the “DNA tetraplexes20 based toehold” activation strategies have been developed, in which the toeholds are T-rich DNA sequence or DNA tetraplex fragment and activated by binding with specific targets (such as Hg2+ or Sr2+). 16, 17 However, in these strategies the recognition sites have been confined in toehold domains and thus the strand displacements reactions need to be welldesigned for each specific target, limiting the flexibility of dynamic DNA devices. Therefore, it is desirable to develop a toehold activation strategy which can activate a unique strand displacement reaction for multiple rather than one target.

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

through DNA hairpin-reconfiguration or enzyme-blocked digestion mechanism, resulting in the formation of the TWJ module. Furthermore, this developed TWJ module was integrated with assembly module containing two partial complementary hairpins in a ‘plug-and-play’ fashion, which, in turn, significantly extends application of TWJ module in designing target-fuelled dynamic DNA devices.

EXPERIMENTAL SECTION Materials and Apparatus Oligonucleotides used in this work were synthesized and purified by Shanghai Sangon Biological Engineering Technology & Services Co., Ltd. (China), and their sequences and modifications were listed in Table S1. Thermodynamic parameters and secondary structures of all oligonucleotides were calculated using bioinformatics software (http://www.bioinfo.rpi.edu/applications/). Uracil-DNA glycosylase (UDG), uracil glycosylase inhibitor (UGI), human 8oxoG DNA glycosylase 1 (hOGG1), M.SssI CpG methyltransferase (M.SssI MTase) supplied with 10 × NEBuffer 2, re20 striction endonuclease HpaII supplied with 10 × CutSmart buffer, HaeIII methyltransferase (HaeIII MTase), AluI methyltransferase (AluI MTase), and S-adenosyl-L-methionine (SAM) were obtained from New England Biolabs Ltd. (Beverly, MA, USA). Exonuclease I (Exo I) and Exonuclease III (Exo III) were purchased from Thermo Fisher Scientific Ltd. (China). 5-fluorouracil (5-FU) was obtained from Med Chem Express (USA). 5-Azacytidine (5-Aza) and 5-aza-2’deoxycytidine (5-Aza-dC) were from Sigma-Aldrich (St. Louis, MO, USA). N-methyl mesoporphyrin IX (NMM) was pur30 chased from J&K Scientific Ltd. (Beijing, China). The BCA protein assay kit was purchased from Beyotime Institute of Biotechnology (Haimen, China). All other reagents were of analytical grade and used as received. The ultrapure water which was obtained from a Millipore Milli-Q water purification system (>18.25 MΩ) was used to prepare all of the solutions. 10

Modular TWJ-based dynamic DNA device for assaying UDG/ M.SssI MTase and inhibition For assaying UDG activity, 250 nM recognition probe of 40 UDG (RUP) and various amounts of UDG were added to the reaction buffer (20 mM Tris-HCl, 1 mM EDTA, 1 mM DTT, 10 mM NaCl, 2 mM MgCl2, pH 8.0) to give a total volume of 10 µL. The mixture was incubated at 37 °C for 45 min to allow the base excision reaction to take place. After the base excision reaction, H1, H2 and 5 × TNaK buffer (100 mM Tris, 700 mM NaCl, 25 mM KCl, pH 7.5) were added. Then the assembly of DNA device in a final volume of 40 µL was carried out at 37 °C for 2 h. At last, NMM was added to the resulting product and incubated at 37 °C for 1 h. The fluores50 cence intensity of the mixture solution was measured on a Hitachi F-7000 spectrofluorophotometer (Hitachi, Japan). The instrument was operated under the following parameters: λex = 399 nm (bandpass 10 nm), λem = 609 nm (bandpass 10 nm), PMT detector voltage = 700 V. The control experiment was carried out under the same condition without UDG. For assaying the inhibition of UDG, the equivalent UGI and UDG were firstly added to the enzyme reaction buffer containing 250 nM RUP, and the mixture solution was incubated at 37 °C for 45 min. Then the dynamic assembly process and the fluorescence 60 measurement were accordant with the operations in UDG assay. For the inhibitor 5-FU, different concentrations of 5-FU and UDG were firstly added in the enzyme reaction buffer

Page 2 of 10

containing 250 nM RUP, and the mixture solution was incubated at 37 °C for 45 min. Then the dynamic assembly process and the fluorescence measurement were accordant with the operations in UDG assay. For assaying M.SssI MTase activity, the dsDNA recognition probe (RMP) of M.SssI MTase was first incubated with 1 × NEBuffer 2 (10 mM Tris-HCl, 1 mM DTT, 50 mM NaCl, 70 10 mM MgCl2, pH 7.9) containing 160 µM S-adenosyl-Lmethionine (SAM) and various amounts of M.SssI MTase with a total volume of 10 µL at 37 °C for 2 h to achieve the methylation at the CpG dinucleotide sites of RMP. Then the resulting mixture was incubated with 1 × CutSmart buffer (50 mM CH3COOK, 20 mM Tris-CH3COOH, 10 mM Mg(CH3COO)2, 100 µg/mL BSA, pH 7.9) containing 25 U/mL of HpaII restriction endonuclease at 37 °C for 2 h to allow the cleave reaction to take place. At last, the dynamic assembly and the fluorescence measurement were performed 80 under the same experimental conditions as mentioned above in the UDG activity assaying. To further study the inhibition effect of anticancer drug, such as 5-Aza and 5-Aza-dC, on the M.SssI MTase activity, different concentrations of inhibitors and 8 U/mL M.SssI were added to the 1 × NEBuffer 2 containing 160 µM SAM and 250 nM RMP, and the mixture solution was incubated at 37 °C for 2 h. Then the subsequent HpaII cleavage reaction, the dynamic assembly and the fluorescence measurement were all accordant with the operations in M.SssI MTase activity assay. 90

Fluorescent Experiment for Demonstrating the Conformational Change of RUP To demonstrate the conformational change of the RUP probe, 250 nM RUP was mixed with 1 × SYBR Green I in the absence and presence of 5 U/mL UDG, respectively. Then the mixture was incubated at 37 °C. The fluorescence intensity of the mixture was measured on F-7000 spectrofluorophotometer with the excitation wavelength at 497 nm and the emission wavelength at 520 nm. To monitor the time tendency of the 100 conformational change process, we measured the fluorescent spectra of the mixture from 0 to 18 min. Furthermore, a duallabeled RUP was designed to verify the conformational change. 250 nM dual-labeled RUP was mixed with 5 U/mL UDG, and the mixture was incubated at 37 °C for 45 min. Afterward, the fluorescent spectra was measured with the excitation wavelength at 495 nm and the emission wavelength at 529 nm. The control experiment was carried out under the same condition without UDG. 110 Characterization

Experiments The gel electrophoresis was conducted using the DNA sample (8 µL per well) on a 15% polyacrylamide gels. The electrophoresis in 1 × TBE running buffer (89 mM Tris, 89 mM Boric Acid, 2.0 mM EDTA, pH 8.3) was run at a constant current of 30 mA for about 2 h. The gels were stained with ethidium bromide for 5 min, destained in distilled water for 5 min, and then visualized under UV imaging system (Bio-RAD Laboratories Inc. USA).

120 Preparation

of HeLa cells lysate The HeLa and MCF-7 cells samples were pelleted by centrifugation (5 min, 3000 rpm, 4 ºC) and resuspended in 20 mL of lysis buffer (20 mM Tris, 150 mM NaCl, 1% Triton X100, sodium pyrophosphate, β-glycerophosphate, EDTA, Na3VO4 and leupeptin, pH 7.5) on ice using a sonicator (four pulses at 200 W for 30 s with a tapered microtip). The mixture 2


Page 3 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


solution was then centrifuged at 12,000 rpm for 30 min at 4 ºC to remove insoluble material. The resulting supernatant was collected and filtered through a 0.45 µm filter membranes, yielding crude lysate.

Scheme. 1 Schematic illustration of the modular DNA three-way junction-based DNA device for UDG activity assay.a

RESULTS AND DISCUSSION Principle of the TWJ-based dynamic DNA device for UDG activity A TWJ trigger on the basis of hairpin-reconfiguration mechanism was firstly constructed using uracil-DNA glyco10 sylase (UDG) as model system. The assembly rationale of dynamic DNA device was displayed in Scheme 1. In this model, the UDG recognition probe (RUP) which is composed of a central, uracil-containing DNA sequence (in black), toehold domain (in blue), branch migration domain (in green), and two auxiliary complementary sequences (in red) was tailored. The uracil-containing DNA sequence was designed to embed a part of the branch migration domain in an intramolecular secondary structure, so that the toehold and the branch migration domains were sequestrated in the rigid-hairpin and 20 would not form the activated TWJ of dynamic DNA device in the absence of UDG. In UDG activity assay, the uracil bases in the stem of RUP could be specifically recognized and hydrolyzed by UDG, generating apyrimidinic (AP) sites and leading to a lower melting temperature of RUP. Scheme 1A depicted the recognition process of UDG and hairpinreconfiguration of RUP. In this process, RUP was enforced to reconfigure into an activated TWJ (denoted as RUP’) accompanying the association of the toehold and branch migration domains for initiating the dynamic assembly. Herein, the RUP 30 probe as the recognition element could convert the detection of UDG activity to the detection of a DNA TWJ. The Scheme 1B depicted the dynamic assembly process. Two partially complementary hairpins (H1 and H2) were designed. The fragment I of H1 (yellow line) was complementary to the fragment I’ of H2 (yellow line). The fragment II’ (black line) of H2 could hybridize with the fragment II (black line) of H1. G-quadruplex subunits (III and III’) were sequestrated in the stem and two terminals of H2 (in purple). When H2 was in its hairpin configuration, intact G-quadruplex could not 40 form due to the block of stem and steric hindrance from the two terminals. In a typical assembly procedure, H1 was unfolded by RUP’ and released the fragment II of H1 via a strand-displacement reaction. Then the newly exposed fragment II of H1 nucleated at the sticky end of H2 and opened the hairpin to expose the fragment I’ of H2. This fragment was identical to the toehold and branch migration domains in sequence, and could hybridize with H1. In this way, the alternant hybridization of H1 and H2 could form a nicked double-helix attaching to multiple integrated G-quadruplexes. Herein, the 50 dynamic DNA device was successfully assembled and could act as the signal amplifier for the sensitive detection of UDG activity. NMM is selected as the signal reporter which is a commercially available unsymmetrical anionic porphyrin characterized by a prominent structural selectivity for Gquadruplex DNA. It exhibits weak fluorescence in free form, but significantly enhanced fluorescence signal via binding to G-quadruplex DNA.31 Verification of the hairpin-reconfiguration mechanism 60 As the recognition element, the hairpin-reconfiguration of the RUP was the key factor for the success of our strategy. Hence, we confirmed firstly that hairpin-reconfiguration of the RUP could happen under the action of UDG by determining

a

RUP is the recognition element and the cascaded H1-H2 hydrization product attaching to multiple integrated G70 quadruplexes is the signal amplifier for the detection of the UDG activity. the fluorescent recovery of the RUP probe with fluorophore and quencher pair (Fig 1A). As the Fig 1B shown, the fluorescence of this probe was brightened (curve a) after the base excision reaction compared to its low fluorescence before the addition of UDG (curve b), convincing that uracil in the stemof RUP was successfully removed and the hairpin reconfigured into another conformation. In addition, the hairpin80 reconfiguration of the RUP was confirmed by determining the time-dependent fluorescent curve (Fig S1). In this experiment, SYBR Green I was used as the signal reporter, which showed an obvious fluorescence intensity when it was intercalated into the groove of dsDNA. Upon the addition of SYBR Green I, the fluorescence change was not obvious when the RUP itself was added into the system. However, in the presence of UDG, an obvious decrease was observed within 18 min, suggesting that uracil in the stem of RUP was successfully removed and the hairpin reconfigured into another conformation which con90 tained less basepairs in the stem compared to the RUP. The native polyacrylamide gel electrophoresis experiment was further used to verify the hairpin-reconfiguration of the RUP (Fig. 1C). After the UDG treatment, the mobility of the bright band in lane 2 was retarded to some extent compared with the RUP free of UDG in lane 3. Such a result evidenced the formation of the RUP’ in lane 2 because of the mobility difference of the same DNA sequence with different conformation.32 The band in lane 1 was a synthetic hairpin oligo which has the same sequence with RUP’. The parallel bright 100 bands in lane 1 and lane 2 further evidenced the formation of the RUP’ after the action of UDG. Investigation of the hairpin-reconfiguration capability of different probes Furthermore, we investigated the effect of the A-U basepairs in the RUPs on the hairpin-reconfiguration capability. In one design, a total of five thymidines in the stem of the RUP were completely displaced with the uracils, resulting in the multiple 3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig. 1 (A)Validation of the hairpin-reconfiguration mechanism. (B) The fluorescent experiment. (a) 250 nM RUP with fluorophore and quencher pair + 5 U/mL UDG; (b) 250 nM RUP with fluorophore and quencher pair. (C) The native polyacrylamide gel electrophoresis experiment. Lane 1: 250 nM synthetic hairpin oligo which has the same sequence with RUP’; lane 2: 250 nM RUP + 5 U/mL UDG; lane 3: 250 nM 10 RUP. recognition sites of UDG (denoted as RUP-5). In addition, four other designs contained four, three, two and one uracils in the stem of RUP which were denoted as RUP-4, RUP-3, RUP2 and RUP-1, respectively. As shown in Fig 2A, when the concentration of UDG was 5 U/mL, the fluorescence intensity increased gradually through altering the number of A-U basepairs from 1 to 5 in the RUP. These results signified that the capability of the hairpin-reconfiguration of the RUP was grad20 ually depressed following the decrease of the A-U basepairs in the RUP. However, the reverse trend was obtained when the detection was performed in the condition that the concentration of UDG was 0.02 U/mL (Fig 2B). The fluorescence intensity increased gradually through altering the number of A-U basepairs from 5 to 1 in the RUP. Such an unexpected result could be explained that the insufficient UDG worked on the excessive and the same concentrations of RUP which lead to the gradually increasing yield of RUP’ along with the decrease of the number of the A-U basepairs in RUP. These results 30 indicated that the recognition probes which contained multiple uracil bases were facile to induce the hairpin-reconfiguration in a high concentration of UDG such as 5 U/mL because the excision of multiple uracils led to the reduction of the melting temperature of RUP sufficiently. However, these probes might make the estimation of UDG activity relatively insensitive in a low concentration of UDG such as 0.02 U/mL since they could give little signal even if some (not all) of the uracil sites in each recognition probe were removed but insufficient to induce the hairpin-reconfiguration. 40

Fig. 2 The effect of different A-U basepairs number on the capability of hairpin-reconfiguration under the different concentrations of UDG. (A) UDG, 5 U/mL; (B) UDG, 0.02 U/mL.

Page 4 of 10

Feasibility of the TWJ-based dynamic DNA device for UDG activity In addition, to confirm that the dynamic assembly of the DNA device was initiated by the end-to-end joint of the toe50 hold and branch migration domains on DNA TWJ, we introduced two additional strands: one contained the same sequence as most sequence in RUP, but without toehold domain, and the other contained the same sequence as most sequence in RUP, with the exception of branch migration domain (denoted as RUP-NT and RUP-NB). As shown in Fig 3A (curve b and curve a), the negligible fluorescence intensity was obtained which evidenced the frustrated assembly of DNA device, suggesting that the cross-opening of H1 and H2 must be initiated by successful association of the toehold and branch 60 migration domains. In deed, the system exhibited significant fluorescence enhancement by exploiting the initial RUP as reactant which confirmed that the association of the toehold and the branch migration domains launched subsequent assembly of DNA device (curve d). In addition, the weak fluorescent intensity was observed (curve c) when only the RUP, H1 and H2 were added in the system. The reason might be that the unintended secondary structure of RUP induced the slight dynamic assembly of DNA device. However, there was a big distinction between the fluorescent values in the presence and 70 absence of UDG, evidencing that the dynamic DNA device was assembled effectively. Furthermore, Fig. 3B showed the characterization of the proposed strategy using the native polyacrylamide gel electrophoresis experiment. In the absence of UDG, the incubation of the RUP and the mixture of H1 and H2 lead to little dynamic assembly product band (lane 2), indicating a poor assembly between the RUP and the mixture of H1 and H2. This result was in accordance with the deduction from the fluorescent experiment. However, when the UDG and RUP were both added into the mixture of H1 and H2, the 80 observed strong band (lane 1) and the invisible band of RUP signified that the base excision reaction induced the hairpinreconfiguration of RUP and initiated the dynamic assembly effectively.

Fig. 3 Validation of the modular DNA TWJ-based DNA device for UDG activity assay. (A) Fluorescence experiment. (a) RUP-NB + UDG + H1 + H2; (b) RUP-NT + UDG + H1 + H2; (c) RUP + H1 + H2; (d) RUP + UDG + H1 + H2. (B) The native polyacrylamide gel electrophoresis experiment. Lane 1: 90 RUP + UDG + H1 + H2; lane 2: RUP + H1 + H2; lane 3: H1 + H2; lane 4: H2; lane 5: H1; lane 6: RUP + UDG; lane 7: RUP. The concentrations of the reagents used in this condition: RUP, 250 nM; H1, 500 nM; H2, 550 nM; UDG, 5 U/mL. Optimization of experimental conditions The signal amplification capability of the dynamic DNA device was affected by the probes concentration and the reaction time which were investigated using F - F0 as standard in order to obtain the optimal analytical parameters, where F and F0 were the enhanced fluorescence intensities of the G4


Page 5 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


quadruplex-NMM complex in the presence and absence of UDG, respectively. The concentrations of RUP, H1 and H2, the reaction time of dynamic assembly, the concentration of NMM were optimized. Fig S2-S6 depicted typical fluorescence response in correlation to these assay conditions. The concentration of RUP has a crucial effect on the execution of the dynamic assembly process. Though RUP in a low concentration could reconfigure into RUP’ mostly, the rate of the toehold-mediated strand displacement between the RUP’ and 10 H1 could be influenced and little product of dynamic assembly could be obtained in this condition. With the increase of the concentration of RUP, the dose of RUP’ produced by the conformation change increased gradually, resulting in the growing rate of the toehold-mediated strand displacement between the RUP’ and H1 which could accelerate the total progress of the dynamic assembly. Furthermore, the increased product of dynamic assembly could be obtained. Thus we varied the concentration of RUP from 100 nM to 350 nM. As shown in Fig S2, the fluorescence response gradually increased and became 20 almost leveled off at the concentration of 250 nM which was used throughout subsequent experiments. Furthermore, we optimized the concentrations of other probes. As shown in Fig S3-S4, the F - F0 reached the maximum when the concentrations H1 and H2 were 500 nM and 550 nM, respectively. The influence of NMM concentration on F - F0 was also studied. The result showed in Fig S5 indicated that the optimized concentration for the system was 400 nM. It was well known that the dynamic assembly performance which determined the amplification capability was also influenced by the reaction 30 time. Considering the different target concentrations might lead to different efficiency of dynamic assembly, thus we chose the low (0.01 U/mL), middle (0.025 U/mL) and high (5 U/mL) concentrations of UDG to optimize the assembly time. As shown in Fig S6, while the slopes under the 0.01 U/mL and 0.025 U/mL were gentle compared with the slope under the concentration of the 5 U/mL, all of the net fluorescent signals leveled off on the 120 min, indicating that the amplification capability of this DNA device achieved the maximum in this time. Herein, the dynamic assembly time 120 min was applied 40 in all the concentrations of the enzyme. Quantitative Activity Detection of UDG using TWJ-based dynamic DNA device Here, we demonstrated dynamic assembly of a DNA device on the basis of UDG-responsive TWJ toehold activation and applied it to determine the concentration of UDG in reaction buffer. As depicted in Fig 4A, the fluorescence intensity enhanced progressively as the concentration of UDG increased from 0 to 5 U/mL, suggesting that the formation of the fluo50 rescent G-quadruplex-NMM complex was highly dependent on the concentration of UDG. As shown in Fig 4B, a dynamic increase of the fluorescence intensity was observed along with the UDG concentrations in a range of 0.0005 U/mL - 0.01 U/mL (R = 0.9982) with a detection limit of 0.000043 U/mL calculated by the triple signal-to-noise method, which has improved more than one order of magnitude as compared with most reported strategies for UDG activity detection (Table S2). The excellent sensitivity of our method was most probably ascribed to the recognition element RUP-1 which was 60 characterized by one uracil removal in each recognition event. To further demonstrate the deduction, we investigated the fluorescent responses of the as-proposed strategy to different concentrations of UDG when the RUP-2, RUP-3, RUP-4 and RUP-5

70

80

90

Fig. 4 (A) Fluorescence responses to the different concentrations of UDG in reaction buffer. (B) Variance of the net fluorescent intensity (F-F0) with UDG concentrations. The inset shows the linear responses at low UDG concentrations. Error bars represent the standard deviations of the results from three 100 independent experiments. worked as the recognition probes (Fig S7-S10). The detection limits which were calculated by the triple signal-to-noise method were 0.000089 U/mL, 0.00023 U/mL, 0.00092 U/mL and 0.0014 U/mL, respectively. These results verified that the high sensitivity was almost based on the design of the recognition probe with one uracil. In addition, the repeatability of the UDG activity assay was investigated through 5 successive samples in the presence of 0.02 U/mL of UDG. The relative standard deviation (RSD) was determined to be 1.7%, indicat110 ing an acceptable repeatability of the as-proposed strategy. Specificity of the TWJ-based dynamic DNA device in blank solution and cell lytates The specificity of the developed strategy was further evaluated. We studied fluorescence responses towards UDG against other nucleases including another base excision enzyme hOGG1, nonspecific protein Exo I and Exo III. For hOGG1, 8-oxoG sites should be included in the DNA substrate, and thus it can not work on the RUP probe. The system 120 did not response to Exo I because that the single strand terminals of the RUP, H1, and H2 were digested by the Exo I. In addition, Exo III is characterized that it can catalyze the stepwise removal of mononucleotides from 3’-hydroxyl terminus 5



Page 6 of 10

as references. The results of 0.27 and 0.16 U mg−1 (U UDG per mg total protein) were obtained in our study, respectively, which were close to the doses in the previous report.33

50 serve

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Inhibition screening of UDG enzyme using the TWJ-based dynamic DNA device The utility of this DNA device for assaying the inhibition of UDG was also studied. Herein, we selected UGI and 5-FU which were useful biochemical tools and potential drugs as the models for this study. UGI can form a tight and physiological60 ly irreversible complex with UDG in 1:1 molar stoichiometry.34 As shown in Fig 6A, an obvious fluorescence decrease was observed when the concentrations of the UDG and UGI were both 5 U/mL. The IC50 value of 2.5 mM (the inhibitor concentration required to reduce enzyme activity by 50%) was obtained from the plot of relative activity versus the concentration of 5-FU (Fig 6B). The results demonstrated that the proposed strategy can be successfully applied in screening UDG inhibitor screening and is a potentially useful tool for anticancer drug discovery.

10

70

Fig. 5 Fluorescence responses towards UDG against other interfering protein in buffer and cell lysates. (A) The fluores20 cence response to UDG (5 U/mL) to Exo I (5 U/mL), Exo III (5 U/mL) and hOGG1 (5 U/mL) in reaction buffer. (B) The fluorescence response to UDG against the multiple components in MCF-7 and Hela cell lysates, and the inhibition effect of UGI to UDG. of duplex DNA and its preferred substrates are blunt or recessed 3’-terminus. In this strategy, all of the DNA probes had overhung 3’-terminus and thus the system can not response to Exo III. As shown in Fig 5A, only UDG caused significant 30 fluorescence intensity, whereas hOGG1, Exo I and Exo III gave low intensity which was comparable to that in the blank solution. In general, the strategy we proposed here exhibited a good performance for discriminating UDG against other interfering proteins. To demonstrate the applicability of the proposed strategy to detect UDG activity, it was also used for the analysis of UDG activity in HeLa and MCF-7 cell lysates. As shown in Fig 5B, both cell lysates induced fluorescence enhancement due to the presence of UDG activity. In contrast, lysis buffer 40 could only induce very low fluorescence signal. Furthermore, UGI was also added to the cell lysate in order to verify that the enhanced fluorescence was solely generated by UDG rather than any other component in the lysate. These results signified that the developed strategy could be tolerant toward the cellular components such as large amounts of other proteins. In addition, the UDG activities in lysed MCF-7 and HeLa cells were measured and calculated based on the standard addition method (Fig S11). The total amounts of protein in the cell lysates were firstly determined by a BCA protein assay kit to

Fig. 6 The inhibited effect of UGI and 5-FU to the UDG activity. (A) Fluorescence response induced by adding UGI; (B) Measurement of IC50 for 5-FU as a small molecular UDG inhibitor. Modular applicability of the TWJ-based dynamic DNA device 80 In order to investigate the modular applicability of our strategy, another DNA TWJ recognition element (denoted as RMP) for the CpG methyltransferase M.SssI (M.SssI MTase) activity assay and MTase inhibitor was designed, which was 6


Page 7 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


formed via the anneal process of toehold-carrying strand and branch migration domain-carrying strand. In the RMP, the proximal toehold and branch migration domains constituted the intact trigger of DNA device, and the specific M.SssI MTase and HpaII recognition sequence (5’-CCGG-3’) was contained in the stem. As shown in Fig 7A, after being digested by HpaII, the toehold and branch migration domains were subject to separate, resulting in the split of the intact trigger and the failure of the dynamic assembly, and thus a negligible 10 fluorescence intensity was observed (Fig 12S, curve a). However, the cleavage of HpaII endonuclease was blocked once the CpG dinucleotide site in the 5’-CCGG-3’ sequence is methylated, and the remaining RMP after the HpaII cleavage successfully induced the assembly of DNA device and generated the obviously enhanced fluorescence response (Fig 12S, curve b), i. e. the alternant hybridization of H1 and H2 formed a nicked double-helix attaching to multiple integrated Gquadruplexes. Furthermore, the native polyacrylamide gel electrophoresis experiment confirmed the rationality of the 20 proposed strategy. As shown in Fig 13S, after the methylation and digestion steps that were catalyzed by M.SssI MTase and HpaII, the bright band of RMP in lane 3 was not darkened compared to the untreated RMP in lane 2, signifying the methylated RMP could block the cleavage of the HpaII. In contrast, the RMP band almost disappeared after the cleavage of HpaII in lane 1. After the methylated and the untreated RMP incubating with the HpaII, respectively, the mixture of H1 and H2 were then added to confirm the occurrence of dynamic assembly. The observed strong band of as 30 sembly product (lane 8) and the invisible band of RMP signified that the methylated RMP initiated the dynamic assembly effectively. In lane 7, the bands of RMP and assembly product were invisible, indicating that the thorough digestion of HpaII to RMP separated the toehold and the branch migration domain, and resulted in ineffectual assembly of the DNA device. We also investigated the sensitivity and selectivity for the MTase activity assay and MTase inhibitor screening by determining fluorescence intensity. It was observed that the increasing concentrations of M.SssI resulted in the progressively 40 enhanced fluorescence intensity in Fig 7B, suggesting that the formation of the fluorescent G-quadruplex-NMM complex was highly dependent on the concentration of M.SssI. As shown in Fig 7C, a dynamic increase of the fluorescence intensity was observed along with the M.SssI concentrations in a range of 0.5 U/mL - 80 U/mL (R = 0.9979) on logarithmic scales with a detection limit of 0.019 U/mL calculated by the triple signal-to-noise method, which has improved one order of magnitude or comparable to the reported fluorescent amplification strategies for MTase activity detection (Table S3). In 50 selectivity experiment, two other cytosine MTases (HaeIII and AluI) were selected as the potential interfering targets.29 As shown in Fig 14S, the proposed assay displayed high specificity to M.SssI which was evidently due to that the CpG dinucleotide site in the HpaII recognition sequence cannot be methylated by HaeIII or AluI. To investigate the inhibitor screening ability of the proposed assay, 5-Aza (5-azacytidine) and 5Aza-dC (5-aza-2’-deoxycytidine) were selected as model inhibitors. These two compounds are nucleoside analogs and representative anticancer drugs used in the majority of methyl60 ation inhibition experiments and a large number of clinical trials, proven to be able to be incorporated into DNA to trap and inactivate MTase.35 As shown in Fig 15S, the relative activity of M.SssI MTase decreased gradually with the increasing doses of 5-Aza and 5-Aza-dC, exhibiting a dose-

Fig. 7 (A) Schematic illustration of the modular DNA TWJbased DNA device for M.SssI MTase assay; (B) Fluorescence 70 response of this method under the different concentrations of M.SssI; (C) Linear relationship between the fluorescence response and the concentration of M.SssI on logarithmic scales. dependent inhibition. The IC50 value for 5-Aza and 5-Aza-dC were acquired from the plots of the relative activity of M.SssI MTase versus the concentrations of inhibitors, and they were 0.4 µM and 2.9 µM, respectively. These results indicated that the proposed DNA device could be successfully applied in 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

MTase inhibitor screening and studying the inhibited effects of anticancer drugs on MTase. Importantly, this dynamic DNA device determined by the intrinsic modular nature should be fairly extended for the detection of other DNA modified enzymes such as thymine DNA glycosylase (TDG),36 human (Dnmt1) methyltransferase,37 through the expedient of changing of its recognition domains. For the measurement of TDG, the recognizable G/T mismatches can be installed on the stem of the hairpin recog10 nition probe which is designed to be similar to RUP, and the hairpin-reconfiguration mechanism can be introduced to obtain activated TWJ for dynamic assembly. Similarly, the unmethylated 5’-CG-3’ sequence can be substituted for the hemimethylated 5’-mCG-3’ sequence on the stem of the dsDNA recognition probe for Dnmt1 which is analogous to RMP, and the enzyme-blocked digestion mechanism can be employed to sustain the integrity of the TWJ for subsequent dynamic assembly. Herein, the capability of this modular DNA device to convert the inactive DNA TWJ recognition probes for differ20 ent nucleases to a unique associated trigger for dynamic assembly is the key feature of the DNA device. Namely, while the recognition probes for each enzyme were designed with a bit difference, the modular DNA device could be switched on by the respective recognition events. As compared to other dynamic assemblies in the literatures, 38-41 a notable advantage of this DNA device is that the associated trigger is formed by the butt joint of the same toehold and branch migration domains, and thus the connected regions between them could be designed to the diverse recognition sites for various enzymes, 30 facilitating the universal dynamic assemblies. Furthermore, the constant hairpins sequence for the assembly units was sufficient for all of the dynamic assemblies due to the resultant triggers of them, i.e. the association of the same toehold and branch migration domains. Compared with the diverse hairpins schemes for each target in traditional dynamic assembly, the universal hairpins enable the facile detection of multiple enzymes in a ‘plug-and-play’ mode.

Page 8 of 10

Supporting Information DNA sequences (Table S1), designs for sequence domains of the probes used in this method (Scheme S1), investigation of hairpin-reconfiguration mechanism (Fig S1), reaction conditions optimization (Fig S2-S6), linear range contrast under different conditions (Fig S7-S10), the standard addition method for UDG in cell lysis (Fig S11), the feasibility study of the 70 method for M.SssI (Fig S12-S13), the selectivity and inhibitor screening study for M.SssI (Fig S14 and Fig S15), the comparison of analytical performance for UDG activity detection by our strategy and those reported in literature (Table S2), the comparison of analytical performance for M.SssI activity detection by our strategy and those reported in literature (Table S3). This material is available free of charge via the Internet at http://pubs.acs.org

AUTHOR INFORMATION 80 Corresponding

author: Tel: +86 531 88363888; fax: +86 531 88564464. E-mail: [email protected]

ACKNOWLEDGMENT This project was supported by the National Natural Science Foundation of China (Grant Nos. 21175081, 21175082, 21375078 and 21475077).

REFERENCES 90

(1) Seeman, N. C. Angew. Chem. 1998, 110, 3408-3428. (2) Seeman, N. C. Nature. 2003, 421, 427-431. (3) Yin, P.; H. Choi, M.T.; Calvert, C. R.; Pierce, N.A. Nature. 2008, 451, 318-323.

CONCLUSION In conclusion, we have successfully developed a modular TWJ-based toehold activation strategy by introducing the mechanisms of DNA hairpin-reconfiguration and enzyme-blocked digestion, and then translated the action of two target nucleases into a universal trigger for a simple and facile assembly of DNA device in ‘plug-and-play’ fashion. This proposed strategy has several advantages: 1) The exquisite design of TWJs and the cascade hybridization-based signal amplification of DNA device make us achieve highly sensitive detection for UDG and CpG methyltransferase M.SssI with detection limits 0.000043 U/mL and 0.019 U/mL, 50 respectively. 2) The modular design and the flexible assembly facilitate the universal detection of this DNA device, and thus it can be applied to more enzymes that could interrupt the base pairs in DNAs. 3) This new strategy to initiate the dynamic assembly of hairpins through the action of DNA modified enzymes on DNA specific sites opens up opportunities to further expand the advanced target-responsive DNA devices for diverse applications. 40 nuclease-responsive

60

(4) Zhang, D. Y.; Seelig, G. Nat. Chem. 2011, 3, 103-113. 100 (5)

Zhu, J.; Zhang, L.; Li, T.; Dong, S.; Wang, E. Adv. Mater. 2013, 25, 2440-2444.

(6) Sherman, W. B.; Seeman, N. C. Nano Lett. 2004, 4, 12031207. (7) Omabegho, T.; Sha, R.; Seeman, N. C. Science. 2009, 324, 67-71. (8) Frezza, B. M.; Cockroft, S. L.; Ghadiri, M. R. J. Am. Soc. 2007, 129, 14875-14879.

110 Chem.

(9) Qian, L.; Winfree, E. Science. 2011, 332, 1196-1201. (10) Dirks, R. M.; Pierce, N. A. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 15275-15278. (11) Li, B.; Ellington, A. D.; Chen, X. Nucleic Acids Res. 2011, 39, e110. 120 (12)

ASSOCIATED CONTENT

Zheng, J.; Jiao, A.; Yang, R.; Li, H.; Li, J.; Shi, M.; Ma, C.; Jiang, Y.; Deng, L.; Tan, W. J. Am. Chem. Soc. 2012, 134, 19957-19960. 8


Page 9 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry (34) Kaushal, P. S.; Talawar, R. K.; Krishna, P. D.; Varshney, U.; Vijayan, M.; Acta Crystallogr. D Biol. Crystallogr. 2008, 64, 551-560.

(13) Qi, X. J.; Lu, C. H.; Liu, X. Q.; Shimron, S.; Yang, H. H.; Willner, I. Nano Lett. 2013, 13, 4920-4924.

(35) Christman, J. K. Oncogene. 2002, 21, 5483-5495.

(14) Bi, S.; Ji, B.; Zhang, Z. P.; Zhu, J. J. Chem. Sci. 2013, 4, 1858-1863. (15) Xing, Y. Z.; Yang, Z. Q.; Liu, D. S. Angew. Chem. Int. Ed. 2011, 50, 11934-11936.

(36) Chen, C.; Zhou, D.; Tang, H.; Liang, M.; Jiang, J. Chem. 2013, 49, 5874-5876.

70 Commun.

(37) Muren, N. B.; Barton, J. K. J. Am. Chem. Soc. 2013, 135, 16632-16640.

10

(16) Ding, W.; Deng, W.; Zhu, H.; Liang, H. J. Chem. Commun. 2013, 49, 9953-9955.

(38) Niu, S.; Jiang, Y.; Zhang, S. Chem. Commun. 2010, 46, 3089-3091.

(17) Tang, W.; Wang, H. M.; Wang, D. Z.; Zhao, Y.; Li, N.; Liu, F. J. Am. Chem. Soc. 2013, 135, 13628-13631. (18) Boer, D. R.; Kerckhoffs, J. M. C. A.; Parajo,Y. Angew. Chem. Int. Ed. 2010, 49, 2336-2339.

(39) Li, F.; Zhang, H.; Wang, Z.; Li, X.; Li, X. F.; Le, X. C. J. Am. Chem. Soc. 2013, 135, 2443-2446. 80

(40) Chen, J.; Qiu, H.; Zhang, M.; Gu, T.; Shao, S.; Huang, Y.; Zhao, S. Biosens. Bioelectron. 2015, 68, 550-555.

20 (19)

He, Y.; Chen, Y.; Liu, H.; Ribbe, A. E.; Mao, C. J. Am. Chem. Soc. 2005, 127, 12202-12203.

(41)Huang, J.; Gao, X.; Jia, J.; Kim, J. K.; Li, Z. Anal. Chem. 2014, 86, 3209-3215.

(20) Shu, D.; Shu, Y.; Haque, F.; Abdelmawla, S.; Guo, P. Nat. Nanotechnol. 2011, 6, 658-667. (21) Muller, J.; Lippert, B. Angew. Chem. Int. Ed. 2006, 45, 2503-2505. (22) Hamada, S.; Murata, S. Angew. Chem. Int. Ed. 2009, 48, 30 6820-6823.

90

(23) Chen, J.; Zhou, S.; Wen, J. Angew. Chem. 2015, 127, 456-460. (24) Li, F.; Lin, Y.; Le, X. C. Anal. Chem. 2013, 85, 1083510841. (25) Chen, X. J. Am. Chem. Soc. 2012, 134, 263-271. 40 (26)

Tang, W.; Hu, S.; Wang, H. Zhao, Y. Li, N.; Liu, F. Chem. Commun. 2014, 50, 14352-14355. (27) Jones, P. A.; Takai, D. Science. 2001, 293, 1068-1070.

(28) Khanna, K. K.; Jackson, S. P. Nat. Genet. 2001, 27, 247254.

100

(29) Esteller, M. Nat. Rev. Genet. 2007, 8, 286-298. 50 (30)

Coppedè , F.; Migheli, F.; Ceravolo, R.; Bregant, E.; Rocchi, A.; Petrozzi, L.; Unti, E.; Lonigro, R.; Siciliano, G.; Migliore, L. Toxicology. 2010, 278, 199-203. (31) Zhao, J. J.; Zhang, L. L.; Jiang, J. H.; Shen, G. L.; Yu, R.Q. Chem. Commun. 2012, 48, 4468-4470. (32) Orita, M.; Iwahana, H.; Kanazawa, H.; Hayashi, K.; Sekiya, T. Proc. Natl. Acad. Sci. USA. 1989, 86, 2766-2770.

60 (33)

Tao, J.; Song, P.; Sato, Y.; Nishizawa, S.; Teramae, N.; Tong, A.; Xiang, Y. Chem. Commun. 2015, 51, 929-932. 110

Graphical Abstract: For Table of Contents only. 9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 10

10

10


Modular Nuclease-Responsive DNA Three-Way Junction-Based

Recommend Documents