Zhou et al., Sci. Adv. 2020; 6 : eabb0695 29 May 2020

S C I E N C E A D V A N C E S | R E S E A R C H A R T I C L E

1 of 8

M O L E C U L A R B I O L O G Y

Three-dimensional DNA tweezers serve as modular DNA intelligent machines for detection and regulation of intracellular microRNALin Zhou1, Mingxuan Gao1, Weiling Fu1, Yunxia Wang1, Dan Luo2, Kai Chang1*, Ming Chen1,3,4*

MicroRNAs (miRNAs) as novel biological targets are hardly applied in diagnostic and treatment of diseases, as they are difficult to be accurately detected and regulated. Here, we demonstrated a modular DNA intelligent machine named three-dimensional tweezers (TD-tweezers) to image and regulate miRNAs in living cells simultaneously. Fluorophore or miRNA inhibitors are introduced as detecting or regulating parts to construct different types of TD-tweezers, and the conformational state of TD-tweezers is controlled by the target miRNAs. The TD-tweezers exhibit excellent sensitivity, specificity, stability, and biocompatibility in vitro and in vivo, and their function of regulating miRNAs was confirmed by the up-regulated expression of downstream genes and proteins. Moreover, the TD-tweezers have been tested in whole blood, preliminarily verifying their clinical application potential. This design provides a multifunctional platform that can achieve efficient detection and regulation of targets within living cells and promote the development of DNA intelligent machines.

INTRODUCTIONMicroRNAs (miRNAs) are an endogenous, noncoding small single- stranded RNAs that have been confirmed to play important roles in the proliferation, development, and maturation of cells (1, 2). Al-though inhibiting their abnormal expression can slow diseases pro-gression, they do not perform as expected due to lack of methods to target and trigger drug release (3, 4). Conventional detection methods of miRNAs rely on Northern blot or reverse transcription polymerase chain reaction (RT-PCR) (5). For the regulation methods, liposomal, lentiviruses, or polymers are used as delivery vehicles to transfer drugs or miRNA regulators into cells (6, 7). These methods are time- consuming, expensive, and require technical laboratory training. Furthermore, the detection and regulation methods are completely separated and without any connection, making it still challenging to precisely target drugs to cells of interest. There are few methods that can integrate both detection and regulation into a whole to achieve comprehensive miRNA detection and regulation (8–10).

Synthetic macromolecules, such as dendrimers, polymers, and nanoparticles, have been shown to have potential for diagnostic and drug-loading applications (11–13). Their structures make it possible to integrate detecting and regulating to achieve targeted therapies not available in traditional methods. However, before using them for biological application, in addition to their own functions, it is important to consider their toxicity, stability, and biocompatibility as exogenous substances in cells (14, 15). DNA can also be used as a building block for structurally tunable nanomaterials, in addition to its principal role of being the carrier of genetic information (16, 17). The programmatic identification and assembly of DNA can achieve precise control of nanomaterial topologies and functions. Besides,

DNA-based strategies are more advantageous in in situ detection and regulation, eliminating the need to consider issues such as bio-compatibility and toxicity. Among DNA nanostructures, DNA ma-chines have been constructed in an attempt to mimic the working mechanisms of machines at the macro level (18, 19). These DNA machines, such as “walkers,” “spiders,” “gears,” and “tweezers,” are constructed via hybridization of nucleic acid strands and may have the structural flexibility required for intelligent biological applica-tions (20–22). Benefiting from their response to specific molecules, DNA machines have been applied in biological diagnostics. However, the lack of regulatory features limits current DNA machines to only the detection of the presence or abnormal expression of specific tar-gets, but cannot restore them to normal through further regulation, thereby limiting their development from simple machines to intel-ligent machines. Judging from the structures of DNA machines, many domains are wasted and can be designed as other functional parts (23, 24). Thus, it is possible to construct a novel DNA machine platform that integrates detection and regulation to innovate the working mode of DNA machines.

We demonstrate the strategy by designing a tetrahedron-shaped DNA intelligent machine to directly target and regulate miRNAs. DNA tetrahedron has been reported to be rapidly endocytosed into cells by the caveolin-mediated pathway and to be maintained sub-stantially in cells for more than 48 hours (25, 26). This machine is designed to detect overexpression of miRNAs and release miRNA inhibitors to regulate their concentrations. Because of the integrated design and calculation, this customizable intelligent DNA machine capable of incorporating sensing motifs with regulating motifs may effectively solve the problem of diagnostic and negative feedback regulation of miRNAs in living cells.

RESULTSPreparation and characterization of platform of three-dimensional tweezersAs shown in Fig. 1A, after slow cooling, the nanostructure of three- dimensional tweezers (TD-tweezers) was synthesized using six single- stranded oligonucleotides. P1 and P2 were partially hybridized to

1Department of Clinical Laboratory Medicine, Southwest Hospital, Army Medical University, 30 Gaotanyan, Shapingba District, Chongqing 400038, China. 2Depart-ment of Biological and Environmental Engineering, Cornell University, Ithaca, NY 14853-5701, USA. 3College of Pharmacy and Laboratory Medicine, Army Medical University, 30 Gaotanyan, Shapingba District, Chongqing 400038, China. 4State Key Laboratory of Trauma, Burn and Combined Injury, Army Medical University, 30 Gaotanyan, Shapingba District, Chongqing 400038, China.*Corresponding author. Email: chming1971@126.com (M.C.); changkai0203@163.com (K.C.)

Copyright © 2020 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution NonCommercial License 4.0 (CC BY-NC).

on October 17, 2020

http://advances.sciencemag.org/

Dow

nloaded from

Zhou et al., Sci. Adv. 2020; 6 : eabb0695 29 May 2020

S C I E N C E A D V A N C E S | R E S E A R C H A R T I C L E

2 of 8

construct the axis of the TD-tweezers. Then, P3, P4, P5 and P6 are partially hybridized with P1 and P2, respectively, to form the four sides of the TD-tweezers, and the middle parts of P5 and P6 were designed to hybridize with the target to control the “on” or “off” state of the TD-tweezers. To verify the efficacy of the TD-tweezers in this experiment, miRNA-34a was used as the target molecule (i.e., the molecule of interest). The process of TD-tweezer construction was displayed using 2% agarose gel electrophoresis (Fig. 1B). After a simple annealing process, the sequences were combined into TD- tweezers (lines 1 to 3) in the form of two chains (P1 + P2), four chains (P1 + P2 + P3 + P4), and complete TD-tweezers (P1 + P2 + P3 + P4 + P5 + P6). In addition, after our target miRNA was intro-duced, the reduction in electrophoretic mobility decreased due to increased molecular mass and more complex spatial structure (line 4). This result illustrated the successful construction of the TD-tweezers and demonstrated the feasibility of detecting and regulating miRNAs using this structure. The change in fluorescence intensity, caused by the fluorescence resonance energy transfer (FRET) of the fluorophore Cy3 (labeled on the 5′-end of P5 and 3′-end of P6) and the quencher BHQ-2 (labeled on the 3′-end of P5 and 5′-end of P6), was also used to confirm the feasibility of TD-tweezers (Fig. 1C). When the TD- tweezer hybridized with miRNA-34a, the distance between the fluorophore and the quencher increased, increasing the fluores-cence intensity of the solution. The fluorescence response to the addition of the miRNA target confirmed the feasibility of the de-signed nanostructure.

Thermodynamics calculations and kinetic analysisThermodynamics and kinetics are two important elements to eval-uating the feasibility of a chemical reaction, since they are used to determine whether a reaction will take place and the reaction’s duration. The whole process is shown in Fig. 1 and can be sim-plified to

A + 2B → C (1)

where A is the TD-tweezers, B is the target, and C is the product. The entire reaction was carried out in an isothermal and isobaric environment. Therefore, the van’t Hoff equation is applicable to this reaction process.

∆G = ∆ G C − ∆ G B − ∆ G A + RT lnQ (2)

where R is the gas constant (8.314 J/mol∙K), T is the temperature in kelvin unit, and Q is the reaction quotient. ∆GA, ∆GB,and ∆GC are the standard free energies under standard conditions, which can be calculated using NUPACK software (Caltech NUPACK team)

∆G − RT lnQ = ∆ G C − ∆ G A − ∆ G B = − 25.15 kcal / mol

When the reaction reaches equilibrium, ∆G = 0, ln Q = lnK, giving

RT lnQ = 25.15 kcal / mol

Q = 5.398 × 1017

Meanwhile

 K = [C] ─ [A ] [B] 2

(3)

assuming that initial concentration of target (B) is 100 nM and the initial concentration of A is x. Then, if 99% B can hybridize to A, then the final concentration of C would be 49.5 nM, at which point we can write the following equation

49.5 × 10 −9 ───────────────── [(x − 49.5 ) × 10 −9 ] (1 × 10 −9 )

2 = 5.398 × 10 17

Fig. 1. Design and characterization of three-dimensional tweezers. (A) Schematic illustration of TD-tweezers synthesis based on internal hybridization of six oligo-nucleotides. (B) A 2% agarose gel to analyze the assembly and conversion of the TD-tweezers. Lane M, DNA ladder; lane 1, strand P1 + strand P2; lane 2, strand P1 + strand P2 + strand P3 + strand P4; lane 3, TD-tweezers; and lane 4, TD-tweezers with target miRNA. (C) Fluorescence response under a fluorescence microscope in the presence or absence of target miRNA. (D) Time-dependent fluorescence spectra of TD-tweezers and ordinary DNA tweezers in 100 nM miRNA-34a. a.u., arbitrary units.

on October 17, 2020

http://advances.sciencemag.org/

Dow

nloaded from

Zhou et al., Sci. Adv. 2020; 6 : eabb0695 29 May 2020

S C I E N C E A D V A N C E S | R E S E A R C H A R T I C L E

3 of 8

Via this method, x can be calculated as 141 nM, meaning that 141 nM TD-tweezers is sufficient to achieve a reaction efficiency of 99% of 100 nM targets without considering the reaction time.

Time-dependent fluorescence intensity analysis was performed to study the reaction kinetics in a homogeneous solution. The fluo-rescence recovery of Cy3 to 100 nM target miRNA was measured at 530 nm for 80 min (Fig. 1D). Since the fluorescence intensity of TD-tweezers itself slowly decreased over time, the difference in value between blank TD-tweezers and miRNA-34a hybridization TD- tweezers was used to analyze the reaction kinetics. Upon addition of target miRNA-34a to the TD-tweezer solution, a time-dependent substantial increase in fluorescence was observed (black line). These results demonstrated that the hybridization of miRNA-34a to TD- tweezers completed in 15 min, indicating that TD-tweezers rapidly responded to the target molecule. On the other hand, because the design inspiration came from the traditional DNA machine (DNA tweezers), we also designed a solution of traditional DNA tweezers (27, 28) (fig. S1) and recorded their response kinetics for 80 min to 100 nM target miRNA. Compared with DNA tweezers in a ho-mogeneous solution (red line), TD-tweezers resulted in a higher fluorescence enhancement. The fluorescence enhancement of our TD-tweezers is 2.6 times higher than that of traditional DNA tweezers, which is consistent with the quantitative difference in fluorophores. Thus, the complex nanostructure of TD-tweezers did not influence the reaction efficiency, and the double FRET-based probe design produced a significant fluorescence enhancement.

In vitro studies of TD-tweezersAfter demonstrating the successful construction and hybridization reaction of TD-tweezers, detection performances of the structure were investigated in the following experiments since detection is the basis of regulation. First, we monitored the fluorescence intensity changes of the TD-tweezers as the concentration of miRNA-34a was increased. As shown in Fig. 2 (A and B), as the concentration of miRNA target increased, a gradual enhancement of fluorescence emission peak was observed. Upon the addition of miRNA target, the fluorescence emission intensity increased rapidly, while the TD- tweezers displayed only low fluorescence signals in the absence of the miRNA target, and the fluorescence intensity of TD-tweezers alone is significantly different from that of the TD-tweezers that hy-bridize to miRNA-34a. Figure 2C illustrates the positive linear rela-tionship observed between fluorescence intensities and miRNA- 34a concentrations. The limit of detection (LOD) was calculated to be 1.499 nM based on the 3/slope rule. This indicated that our DNA intelligent machines could effectively hybridize to the target miRNA and correspondingly changed the structure from “signal off” to “signal on,” resulting in high fluorescence signals. And, a linear relationship between fluorescence intensities and miRNA-34a con-centration was also detected when using traditional DNA tweezers (fig. S2), and in this case, the LOD was 5.4941 nM according to 3/slope rule. The slightly lower LOD of TD-tweezers indicated that changes in nanostructure do not affect the efficiency of DNA tweezers, and the design of double fluorophores actually helped to improve the sensitivity, which is consistent with the assumptions made during kinetic analysis.

In addition to sensitivity, specificity is another crucial parameter that needs to be evaluated before using TD-tweezers as an effective detection method. To test the specificity of TD-tweezers, single-base mismatched miRNA-34a, double-bases mismatched miRNA-34a,

three-bases mismatched miRNA-34a, and other three miRNAs (miRNA-21, miRNA-27, and miRNA-155) were designed as contras-tive substances (Fig. 2, D and E). In these tests of miRNA-34a–targeted TD-tweezers, only miRNA-34a triggered the reaction effectively and showed enhanced fluorescence, with all the other miRNAs showing negligible fluorescence responses. In addition, the normal DNA tweezers also showed the same ability to recognize target miRNA- 34a (fig. S3). The above results indicated that TD-tweezers have enough reliable specificity to distinguish target miRNAs from simi-lar molecules, effective enough to reduce false positives. In vitro experiments with TD-tweezers and traditional DNA tweezers show

Fig. 2. Detection performance of TD-tweezers in vitro and their feasibility in vivo. (A) Fluorescence responses in the presence of different concentrations of miRNA-34a. (B) The relationship between fluorescence intensities and target miRNA concentrations. (C) The linear relationship between signals and miRNA concentra-tions. (D and E) Selectivity of TD-tweezers toward single-base mismatched miRNA, double-bases mismatched miRNA, three-bases mismatched miRNA, and different types of miRNAs (miR-21, miR-27, and miR-155). (F and G) Fluorescence character-ization of TD-tweezers and DNA tweezers degradation by incubating in 10% fetal bovine serum for the stated times. (H) Cell viability assay (CCK-8): K562 cells treated with TD-tweezers (50 and 100 nM) for 2, 4, 6, 12, and 24 hours at 37°C.

on October 17, 2020

http://advances.sciencemag.org/

Dow

nloaded from

Zhou et al., Sci. Adv. 2020; 6 : eabb0695 29 May 2020

S C I E N C E A D V A N C E S | R E S E A R C H A R T I C L E

4 of 8

that the performance of the designed TD-tweezers is sufficient for use as DNA machines.

The stability and biocompatibility of TD-tweezersThe biggest challenges faced when applying many DNA intelligent machines to intracellular detection are instability and lack of bio-compatibility, because these properties determine whether the probes can be delivered into cells, work properly, and function without producing false-positive signals (29, 30). Before the application of TD-tweezers in the intracellular environment, the stability of TD- tweezers was first investigated by incubating them with 10% fetal bovine serum (FBS) to simulate physiological conditions. As shown in Fig. 2 (F and G), the fluorescence intensities of TD-tweezers hardly changed during the 0- to 8-hour incubation in 10% FBS, in-dicating that the TD-tweezers may effectively resist damages caused by the environment. However, when traditional DNA tweezers were incubated in the presence of 10% FBS, an obvious increase in fluo-rescence intensity was observed after 6 hours of incubation. It is be-lieved that the biological stability of TD-tweezers could be improved via the steric hindrance effect of the designed backbone structure. Nevertheless, the electrophoretic characterization of TD-tweezers also showed no degradation and that the tweezers could maintain their conformation at least 12 hours (fig. S4). These results indicated that TD-tweezers were more stable than the traditional DNA tweezers and could be used in complex biological samples.

Good biocompatibility is another crucial feature for the applica-tion of TD-tweezers to in vivo detection and regulation. Therefore, a standard colorimetric Cell Counting Kit-8 (CCK-8) assay was per-formed on K562 cells (Fig. 2H). Different concentrations of TD-tweezers (50 and 100 nM) were incubated with K562 cells for a period of time (2, 4, 6, 12, and 24 hours). It was observed that the viability of K562 cells was approximately 100%, and no obvious cytotoxicity or side effects in living cells occurred as the result of TD-tweezer incubation. These results confirmed that TD-tweezers exhibited excellent biocompatibility and could potentially provide reliable results during intracellular detection and regulation.

In vivo study of TD-tweezersEncouraged by their success in biological stability and biocompatibility, we further studied the ability of TD-tweezers to function in living cells. To confirm the stability of TD-tweezers and traditional DNA tweezers in vivo, K562 cells were incubated with TD-tweezers or traditional DNA tweezers for 2, 4, 6, 8, and 10 hours (fig. S5). TD- tweezers maintained their fluorescence intensity within the cells for 6 hours, and then this intensity increased rapidly, indicating that the nanostructure could resist the damages caused by enzymes or other molecules for 6 hours. This time was shorter than that ob-served for TD-tweezers in 10% FBS, since the environment of cells is more complex than that of FBS. In contrast, the fluorescence of traditional DNA tweezers initially increased and gradually stabilized after 8 hours. These results validated our previous assumption that normal DNA tweezers would be destroyed in cells and, therefore, produce fluorescence. These results, combined with the results of a previous in vitro stability experiment, confirmed the ability of TD- tweezers to maintain their structure and function in vitro and in vivo. Therefore, TD-tweezers are believed to be useful as probes in intra-cellular applications.

K562 cells (human myeloid leukemia cells) were used to verify the feasibility of TD-tweezers in an in situ visualization of intracellular

miRNA-34a (31, 32). Before the application of TD-tweezers in vivo, the appropriate incubation times and concentrations were studied to ensure optimal reaction conditions (fig. S6). The fluorescence in-tensity was monitored in the living cells, demonstrating that the fluorescence signals gradually increased with increasing incubation time and stabilized at 2 hours. These results indicated that this time point could be used as the optimal reaction time for the subsequent experiments. In addition, by incubating TD-tweezers with different concentrations of cells, the fluorescence signals increased with in-creasing cell mass, indicating that the optimal cell concentration for detecting reaction was higher than 1 × 105 cells/ml.

Live-cell imaging of TD-tweezersThe in vivo detection process is briefly illustrated in Fig. 3A, where TD-tweezers hybridized with target miRNAs to produce a stronger fluorescence. The potential of TD-tweezers for the quantitative eval-uation of miRNAs in living cells needed to be evaluated, as miRNA expression levels vary within same cancer cells during various stages of tumorigenesis. To test this in our study, K562 cells were divided into the following three groups: a group treated with miRNA-34a mimic (aimed at increasing miRNA-34a expression), a group treated

Fig. 3. Study of TD-tweezers for detecting miRNAs in vivo. (A) Schematic illus-tration of the mechanism of TD-tweezers for specific miRNA detection in living cells. (B) Confocal fluorescence imaging of TD-tweezers to detect miRNA-34a in K562 cells treated with miRNA-34a mimic, K562 cells treated with inhibitor, and untreated K562 cells. (C) Fluorescence intensity of the K562 cells treated with miRNA-34a mimic, K562 cells treated with inhibitor, and untreated K562 cells. (D) Confocal flu-orescence imaging of TD-tweezers for detection of miRNA-34a in K562 cells, Mino cells, and Jurkat cells. (E) Fluorescence intensity of the K562 cells, Mino cells, and Jurkat cells. Scale bars, 10 m.

on October 17, 2020

http://advances.sciencemag.org/

Dow

nloaded from

Zhou et al., Sci. Adv. 2020; 6 : eabb0695 29 May 2020

S C I E N C E A D V A N C E S | R E S E A R C H A R T I C L E

5 of 8

with miRNA-34a inhibitor (aimed at decreasing miRNA-34a expres-sion), and an untreated control group. The fluorescence was observed from cells incubated with 100 nM TD-tweezers for 2 hours. A strong red fluorescence signal of Cy3 was observed in the miRNA-34a–overexpressing group, while cells with low expression of miRNA- 34a exhibited faint fluorescence signals under the same conditions (Fig. 3B). Untreated K562 cells expressed fluorescence signals inter-mediate to the previous two. These results suggested that this strategy was appropriate for detecting miRNA expression levels in living cancer cells. To confirm the accuracy of these intracellular imaging results, the quantified fluorescence intensities of each cells were counted (Fig. 3C) and compared with the result of RT-PCR (fig. S7). The same trend indicated that TD-tweezers were able to detect miRNA expression semiquantitatively.

After verifying that TD-tweezers were able to recognize the aber-rant expression of miRNA-34a, it was necessary to investigate whether they could simultaneously be applied to detect the expres-sion of miRNA-34a and image it in cells. Since the expression level of miRNA is different in different cell lines, Jurkat and Mino cells were used to investigate whether TD-tweezers could be applied to different cell lines. As shown in Fig. 3D, an obvious miRNA-34a fluorescence signal was observed in K562 cells, but almost no fluo-rescence signal was observed in Jurkat cells and Mino cells. The flu-orescence intensity of K562 cells was 11.7 and 5.3 times that of the Jurkat and Mino cells, respectively (Fig. 3E). The standard RT-PCR further confirmed that the expression levels of miRNA-34a in these three groups of cells were consistent with the results of intracellular detection (fig. S7). Similarly, traditional DNA tweezers were also incubated in the above cells, as shown in figs. S8 and S9. Although red fluorescence signals were observed, there was no significant difference among the three groups of cells with different expres-sion levels of miRNA-34a. These consistent fluorescence signals may have been generated by the spontaneous disassembly of the tweez ers, as it is difficult for traditional DNA tweezers to maintain their structures in vivo. Overall, these findings indicated that TD-tweez ers were able to distinguish target miRNA expression levels in different cells.

Living cell regulation of TD-tweezersIn addition to TD-tweezers for miRNA imaging in living cells, regulatory sections can also be introduced into the structure of TD- tweezers since the advantage of the modular DNA intelligent ma-chine platform is that other modules can be added to its structure. As shown in Fig. 4A, miRNA regulatory moieties were added by replacing miRNA inhibitors with P3 and P4. Hybridization of the target miRNA and TD-tweezers allows the fixed area of the inhibitors to be replaced, thereby facilitating their release. The introduction of this part enables the TD-tweezers to release miRNA inhibitors while detecting miRNA and play a role in regulating miRNA expression. The electrophoresis result showed that TD-tweezers added with regulatory modules met our expectation (Fig. 4B). The addition of miRNA can open their structure and release their inhibitors, which are a special type of sequences that can hybridize with miRNAs. Since the sequences of miRNA inhibitors are longer than those of miRNAs, the electrophoretic mobility increases after the regulatory reaction, and the bands where the released inhibitors and the miRNAs hybridize are also displayed (line 7).

To verify whether the regulatory part played its intended role, the related biological reactions were detected. As we all know, miRNAs

could regulate gene expression at the posttranscriptional level by inhibiting mRNA transcription. Therefore, the expression of down-stream genes and proteins can be used to determine whether miRNAs were regulated by TD-tweezers. Therefore, we designed miRNA-34a–, miRNA-188–, miRNA-574–, and miRNA-152–related TD-tweezers and detected the expression of related mRNAs and proteins by PCR and Western blot (33–35). As can be seen in Fig. 4 (C and D), the mRNA expression levels were up-regulated after miRNAs are in-hibited. This is because miRNAs act as trans-acting factors in gene regulation, so miRNA inhibitors can reduce the effect of miRNAs on downstream gene expression. Similarly, the expression of corre-sponding proteins also increased with increasing mRNA expression. Since the expression levels of downstream genes are affected by many factors, the efficiency of the designed regulatory platform is limited, which indicated that the process of cell activity is much more complicated than that of machines. The multifarious potential of TD- tweezers stimulated us to develop them as a universal DNA intelligent machine for working autonomously in living cells, similar to macro-level intelligent machines.

Clinical application of TD-tweezersThe good performance of this DNA intelligent machine in living cells motivated us to further investigate its feasibility in blood. Leukocytes, which were extracted from blood after centrifugation, expressed dif-ferent fluorescence intensity between samples acquired from patients with myeloid leukemia and healthy individuals (Fig. 5) (36, 37). The leukocytes from patients with myeloid leukemia expressed high flu-orescence signals, while no apparent fluorescence signal was observed in that of healthy individuals. Quantitative analysis showed that the myeloid leukemia cells generated 4.3-fold higher fluorescence than that of the healthy individuals. The satisfying performance of TD- tweezers in blood further increases our confidence in applying these machines in intracellular molecule detection and regulation. There-fore, we believe that our design of TD-tweezers will contribute to the intracellular biological location of miRNAs or other specific mole-cules and promote the development of DNA intelligent machines for in vivo applications. Moreover, TD-tweezers may even be used as an miRNA regulation method due to their modular characteristics and ability to react to the environment.

DISCUSSIONIn summary, we have designed a modular DNA intelligent machine that can respond to specific miRNAs targets within living cells. With precise design and calculation, TD-tweezers were formed with high efficiency and could achieve complete hybridization with miRNA targets. Because of the unique DNA tetrahedral framework, TD- tweezers showed significantly enhanced biological stability and cel-lular delivery efficiency, which are prerequisites for DNA intelligent machine use in in vivo applications. In addition, the modular design allows us to add components such as regulatory components to their original structure. We have successfully demonstrated their detec-tion functions and miRNA regulatory function in vivo. In addition to miRNAs inhibitors, these TD-tweezers may also carry other reg-ulatory molecules and even drugs. The in vivo multifunction of this designed DNA intelligent machine can promote the development of DNA intelligent machines and further improve them to intelli-gent machines. Furthermore, with the assistance of the sustained- release or targeted-release structures, the TD-tweezers will hopefully

on October 17, 2020

http://advances.sciencemag.org/

Dow

nloaded from

Zhou et al., Sci. Adv. 2020; 6 : eabb0695 29 May 2020

S C I E N C E A D V A N C E S | R E S E A R C H A R T I C L E

6 of 8

work more accurately and be put to practical application in the fu-ture. In brief, this modular functional platform is a new attempt to realize comprehensive detection and regulation of biomolecules in the future.

MATERIALS AND METHODSMaterialsAll oligonucleotides were synthesized by Sangon Biotech Co., Let (China), and TaKaRa Biotechnology (China). miRNA-34a mimic and miRNA-34a inhibitor were purchased from Genechem (China). CCK-8 was purchased from DOJINDO (Japan). MiRNeasy Mini Kit, miScript II RT Kit, and miScript PCR Starter Kit were obtained from QIAGEN Bioinformatics (Germany). RPMI-1640 medium and FBS were pur-chased from Biological Industries (USA). Penicillin- streptomycin was obtained from Biosharp (China). The human myeloid leukemia cells K562, mantle cell lymphoma cells Mino, and acute T cell leukemia cells Jurkat were from our laboratory preservation. All chemical reagents used were of analytical grade.

InstrumentsGel electrophoresis analysis was carried out on a Bio-Rad Electro-phoresis System and imaged on Bio-Rad ChemDoc XRS (Bio-Rad, USA). The fluorescence spectra and CCK-8 assay were recorded

Fig. 4. Study of TD-tweezers for regulating miRNAs in vivo. (A) Schematic illustration of the mechanism of TD-tweezers for specific miRNA regulation in living cells. (B) Electrophoresis result to verify the construction of regulatory TD-tweezers. Lane M, DNA ladder; lane 1, target miRNA; lane 2, miRNA inhibitor; lane 3, the hybridization of miRNA and inhibitor; lanes 4 to 6, the process of the construction of regulatory TD-tweezers; and lane 7: the regulatory TD-tweezers reacted with target miRNA. (C) The changes of downstream mRNA expression after the function of regulatory TD-tweezers. (D) The changes of downstream protein expression after the function of regulatory TD-tweezers.

Fig. 5. Application of TD-tweezers in clinic. (A) Simple illustration of the detection process with whole blood. I, centrifugation; II, separation; and III, incubation with TD-tweezers. (B) Fluorescence intensity of myeloid leukemia sample and normal sample. (C) Confocal fluorescence imaging of myeloid leukemia sample and normal sample. Scale bar, 10 m.

on October 17, 2020

http://advances.sciencemag.org/

Dow

nloaded from

Zhou et al., Sci. Adv. 2020; 6 : eabb0695 29 May 2020

S C I E N C E A D V A N C E S | R E S E A R C H A R T I C L E

7 of 8

using a Thermo Scientific Variskan Flash (Thermo, USA). RT-PCR was performed on an ABI 7500 Real-Time PCR machine (ABI, Applied Biosystems, USA). The confocal fluorescence imaging was performed on a Zeiss LSM780 laser scanning confocal microscope (Zeiss, Germany).

Preparation of TD-tweezers and normal DNA tweezersTD-tweezers consist of six custom single-stranded oligonucleotide strands (P1, P2, P3, P4, P5, and P6; P3 and P4 are replaced by special miRNA-related inhibitors when we constructed regulatory TD- tweezers). DNA tweezers consist of four single-stranded oligo-nucleotide strands (D1, D2, P3, and P5). In brief, these oligonucleotides were mixed in an equal molar ratio in buffer (10 mM tris-HCl, 1 mM EDTA, and 20 mM MgCl2). The final concentration of 1 M solu-tion was heated to 95°C for 10 min, and then slowly cooled to room temperature and kept at 4°C for further use. Then, the successful formations of TD-tweezers and DNA tweezers were verified using electrophoresis.

Fluorescence measurementsFor the kinetic assays, TD-tweezers were incubated with miRNA-34a, and the fluorescence emission signal per minute was recorded. For the sensitivity experiment, different concentrations of miRNA-34a were incubated with synthetic TD-tweezers. After incubation for 1 hour at 37°C, the fluorescence spectra of 560 to 620 nm were mea-sured at an excitation wavelength of 530 nm. For the specificity experiments, single-base mismatched miRNA-34a, double-bases mismatched miRNA-34a, three-bases mismatched miRNA-34a, miRNA-21, miRNA-27, miRNA-155, and miRNA-34a were separately added to the solution containing TD-tweezers, incubated at 37°C for 1 hour, and the fluorescence emission spectra were then collected. For the stability experiment, we incubated TD-tweezers with 10% FBS continued recording the signal of Cy3 for 8 hours. For compar-ison with the traditional DNA tweezers, we also performed all of the above experiments using DNA tweezers.

Cell cultureK562 cells and Jurkat cells were cultured in RPMI-1640 medium supplemented with 10% FBS and 1% penicillin-streptomycin, and Mino cells were cultured in RPMI-1640 medium containing 15% FBS and 1% penicillin-streptomycin. All cells were incubated in 5% CO2, 37°C incubator.

CCK-8 assayK562 cells (100 l per well) were incubated within 96-well microtiter plates for 24 hours at 37°C in 5% CO2. Then, cells were treated with different concentrations of TD-tweezers (0, 50, and 100 nM) and incubated for 2, 4, 6, 12, and 24 hours. Subsequently, 10 l of CCK-8 solution was added into each well. After incubation at 37°C in 5% CO2 for 1 hour, the cytotoxic effects were determined by measuring the absorbance at optical density (OD) 450 nm.

Uptake of miRNA-34a mimic and inhibitorEighty-six microliters of K562 cells (105 cells/ml) was incubated with 10 l of miRNA-34a mimic [3 × 107 toxic unit (TU)/ml] or inhibitor. Besides, 4 l of HitransG A was added to promote the efficient lenti-virus infection of cells. These cells were incubated at 37°C in 5% CO2 for 72 hours, and 100 l of RPMI 1640 medium containing 10% FBS was added daily to ensure the normal cell growth. RT-PCR was used to verify the results of cells treated with mimic and inhibitor.

RT-PCR analysisTotal cellular miRNA was extracted using the miRNeasy Mini Kit according to the manufacturer’s instruction. For detecting expression miRNAs, the complementary DNA (cDNA) samples were prepared using the miScript II RT Kit. In addition, RT quantitative PCR (qPCR) analysis was performed with the miScript PCR Starter Kit. For de-tecting the expression of mRNAs, the cDNA samples were prepared using the QuantiNava Reverse Transcription Kit, and RT-qPCR anal-ysis was performed with QuantiNava SYBR Green PCR Kit. The primers used in these experiments are listed in table S1. All the results of miRNAs/mRNA expression were evaluated by normalizing the expression of U6/glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and using the 2−Ct method.

Western blotProteins were extracted from cells treated with miRNA-34a–, miRNA- 188–, miRNA-574–, miRNA-152–related TD-tweezers for 48 hours and quantified using the BCA (bicinchonininc acid) assay. Then, the proteins were separated by 12% SDS–polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes. After a period of incubation with antibodies for a short period of time, the immunoreactive bands were visualized and observed.

Confocal fluorescence imagingAll cells were seeded on confocal culture dishes for 24 hours. Then, TD-tweezers or DNA tweezers were added and incubated with cells for 2 hours at 37°C in 5% CO2. After washing twice with PBS, three volumes of 4′,6-diamidino-2-phenylindole (DAPI) (5 g/ml) were mixed with washed cells and incubated for 1 hour. Confocal fluo-rescence imaging was performed on a Zeiss LSM780 laser scanning confocal microscope. For the semiquantitative experiment, miRNA- 34a mimic– or inhibitor–treated cells and untreated K562 cells were used. For the specificity experiment, K562 cells, Mino cells, and Jurkat cells were used. For the stability experiment, TD-tweezers or DNA tweezers were incubated with K562 cells for 2, 4, 6, 8, and 10 hours at 37°C in 5% CO2.

Preparation and detection of blood samplesSerums were obtained from healthy volunteers and patients with myeloid leukemia at the Southwest Hospital of Army Medical Uni-versity with ethical approval. The whole samples were dealt with red blood cell lysis buffer according to the manufacturer’s instruction, and the obtained white cells were washed with PBS and resuspended by RPMI-1640 medium supplemented with 10% FBS. Ten microliters of cells (>105/ml) was incubated with 10 l of TD-tweezers (10 M) for 2 hours at 37°C in 5% CO2, and the results were imaged by a confocal microscope.

SUPPLEMENTARY MATERIALSSupplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/6/22/eabb0695/DC1

View/request a protocol for this paper from Bio-protocol.

REFERENCES AND NOTES 1. Y. Tutar, MiRNA and cancer; computational and experimental approaches. Curr. Pharm.

Biotechnol. 15, 429 (2014). 2. M. Westphal, K. Lamszus, Circulating biomarkers for gliomas. Nat. Rev. Neurol. 11,

556–566 (2015). 3. T. Tian, J. Wang, X. Zhou, A review: microRNA detection methods. Org. Biomol. Chem. 13,

2226–2238 (2015).

on October 17, 2020

http://advances.sciencemag.org/

Dow

nloaded from

Zhou et al., Sci. Adv. 2020; 6 : eabb0695 29 May 2020

S C I E N C E A D V A N C E S | R E S E A R C H A R T I C L E

8 of 8

4. R. Deng, K. Zhang, J. Li, Isothermal amplification for microRNA detection: From the test tube to the cell. Acc. Chem. Res. 50, 1059–1068 (2017).

5. H. Dong, J. Lei, L. Ding, Y. Wen, H. Ju, X. Zhang, MicroRNA: Function, detection, and bioanalysis. Chem. Rev. 113, 6207–6233 (2013).

6. R. Rupaimoole, F. J. Slack, MicroRNA therapeutics: Towards a new era for the management of cancer and other diseases. Nat. Rev. Drug Discov. 16, 203–222 (2017).

7. D. Mellis, A. Caporali, MicroRNA-based therapeutics in cardiovascular disease: screening and delivery to the target. Biochem. Soc. Trans. 46, 11–21 (2018).

8. K. Ren, Y. Xu, Y. Liu, M. Yang, H. Ju, A responsive “nano string light” for highly efficient mRNA imaging in living cells via accelerated DNA cascade reaction. ACS Nano 12, 263–271 (2018).

9. L. He, D. Lu, H. Liang, S. Xie, X. Zhang, Q. Liu, Q. Yuan, W. Tan, An mRNA-initiated, three-dimensional DNA amplifier able to function inside living cells. J. Am. Chem. Soc. 140, 258–263 (2018).

10. Y. Wu, X. Yao, Y. Chen, Y. Li, W. Tian, Advance of DNA and CCPs-based nanocarriers in drug delivery systems. Biomed. Mater. Eng. 28, S255–S261 (2017).

11. L. Scheller, T. Strittmatter, D. Fuchs, D. Bojar, M. Fussenegger, Generalized extracellular molecule sensor platform for programming cellular behavior. Nat. Chem. Biol. 14, 723–729 (2018).

12. K. Tian, K. Decker, A. Aksimentiev, L. Q. Gu, Interference-free detection of genetic biomarkers using synthetic dipole-facilitated nanopore dielectrophoresis. ACS Nano 28, 1204–1213 (2017).

13. L. Huang, J. Wan, X. Wei, Y. Liu, J. Huang, X. Sun, R. Zhang, D. D. Gurav, V. Vedarethinam, Y. Li, R. Chen, K. Qian, Plasmonic silver nanoshells for drug and metabolite detection. Nat. Commun. 8, 220 (2017).

14. S. Li, L. Xu, W. Ma, X. Wu, M. Sun, H. Kuang, L. Wang, N. A. Kotov, C. Xu, Dual-mode ultrasensitive quantification of MicroRNA in living cells by chiroplasmonic nanopyramids self-assembled from gold and upconversion nanoparticles. J. Am. Chem. Soc. 138, 306–312 (2016).

15. K. Zhang, S. Song, S. Huang, L. Yang, Q. Min, X. Wu, F. Lu, J. J. Zhu, Lighting up microRNA in living cells by the disassembly of lock-like DNA-programmed UCNPs-AuNPs through the target cycling amplification strategy. Small 14, e1802292 (2018).

16. P. Cramer, Organization and regulation of gene transcription. Nature 573, 45–54 (2019). 17. E. H. Finn, T. Misteli, Molecular basis and biological function of variability in spatial

genome organization. Science 365, eaaw9498 (2019). 18. M. Endo, H. Sugiyama, DNA origami nanomachines. Molecules 23, E1766 (2018). 19. L. Voskoboinik, M. Amiel, A. Reshef, R. Gafny, M. Barash, Laundry in a washing machine

as a mediator of secondary and tertiary DNA transfer. Int. J. Legal Med. 132, 373–378 (2018).

20. J. L. He, Y. Zhang, T. T. Mei, L. Tang, S. Y. Huang, Z. Cao, Telomerase-triggered DNAzyme spiders for exponential amplified assay of cancer cells. Biosens. Bioelectron. 144, 111692 (2019).

21. S. Lei, L. Xu, Z. Liu, L. Zou, G. Li, B. Ye, An enzyme-free and label-free signal-on aptasensor based on DNAzyme-driven DNA walker strategy. Anal. Chim. Acta 1081, 59–64 (2019).

22. X. Liu, J. Zhang, M. Fadeev, Z. Li, V. Wulf, H. Tian, I. Willner, Chemical and photochemical DNA "gears" reversibly control stiffness, shape-memory, self-healing and controlled release properties of polyacrylamide hydrogels. Chem. Sci. 10, 1008–1016 (2018).

23. C. Teller, I. Willner, Functional nucleic acid nanostructures and DNA machines. Curr. Opin. Biotechnol. 21, 376–391 (2010).

24. M. Liu, J. Fu, C. Hejesen, Y. Yang, N. W. Woodbury, K. Gothelf, Y. Liu, H. Yan, A DNA tweezer-actuated enzyme nanoreactor. Nat. Commun. 4, 2127 (2013).

25. K. E. Sapsford, L. Berti, I. L. Medintz, Materials for fluorescence resonance energy transfer analysis: Beyond traditional donor-acceptor combinations. Angew. Chem. Int. Ed. Eng. 45, 4562–4589 (2006).

26. M. Acunzo, C. M. Croce, MicroRNA in cancer and cachexia—A mini-review. J. Infect. Dis. 212, S74–S77 (2015).

27. B. Kou, Y. Chai, Y. Yuan, R. Yuan, Dynamical regulation of enzyme cascade amplification by a regenerated DNA nanotweezer for ultrasensitive electrochemical DNA detection. Anal. Chem. 90, 10701–10706 (2018).

28. J. Elbaz, M. Moshe, I. Willner, Coherent Activation of DNA Tweezers: A "SET-RESET" logic system. Angew. Chem. Int. Ed. Eng. 48, 3834–3837 (2009).

29. R. Gao, C. Hao, L. Xu, C. Xu, H. Kuang, Spiny nanorod and upconversion nanoparticle satellite assemblies for ultrasensitive detection of messenger RNA in living cells. Anal. Chem. 90, 5414–5421 (2018).

30. A. S. Walsh, H. Yin, C. M. Erben, M. J. Wood, A. J. Turberfield, DNA cage delivery to mammalian cells. ACS Nano 5, 5427–5432 (2011).

31. F. Navarro, D. Gutman, E. Meire, M. Cáceres, I. Rigoutsos, Z. Bentwich, J. Lieberman, MiR-34a contributes to megakaryocytic differentiation of K562 cells independently of p53. Blood 114, 2181–2192 (2009).

32. S. Y. Xie, Y. J. Li, P. Y. Wang, F. Jiao, S. Zhang, W. J. Zhang, MiRNA-regulated expression of oncogenes and tumor suppressor genes in the cisplatin-inhibited growth of K562 cells. Oncol. Rep. 23, 1693–1700 (2010).

33. Z. Y. Nie, L. Yang, X. J. Liu, Z. Yang, G. S. Yang, J. Zhou, Y. Qin, J. Yu, L. L. Jiang, J. K. Wen, J. M. Luo, Morin inhibits proliferation and induces apoptosis by modulating the miR-188-5p/PTEN/AKT regulatory pathway in CML cells. Mol. Cancer Ther. 18, 2296–2307 (2019).

34. H. Yang, J. Zhang, J. Li, F. Zhao, Y. Shen, X. Xing, Overexpression of miR-574-3p suppresses proliferation and induces apoptosis of chronic myeloid leukemia cells via targeting IL6/JAK/STAT3 pathway. Exp. Ther. Med. 16, 4296–4302 (2018).

35. L. Wang, Y. Wang, J. Lin, MiR-152-3p promotes the development of chronic myeloid leukemia by inhibiting p27. Eur. Rev. Med. Pharmacol. Sci. 22, 8789–8796 (2018).

36. Y. J. Tian, Y. H. Wang, A. J. Xiao, P. L. Li, J. Guo, T. J. Wang, D. J. Zhao, Long noncoding RNA SBF2-AS1 act as a ceRNA to modulate cell proliferation via binding with miR-188-5p in acute myeloid leukemia. Artif .Cells Nanomed. Biotechnol. 47, 1730–1737 (2019).

37. S. Jinlong, F. Lin, L. Yonghui, Y. Li, W. Weidong, Identification of let-7a-2-3p or/and miR-188-5p as prognostic biomarkers in cytogenetically normal acute myeloid leukemia. PLOS ONE 10, e0118099 (2015).

Acknowledgments: We sincerely thank Y. Wang for providing K562, Jurkat, and Mino cell lines. We also thank the staff of Southwest Hospital and all participants. Funding: We gratefully thank the financial support from projects 81430053, 81972027, and 81401751 by the National Natural Science Foundation of China (NSFC). Author contributions: M.C. designed the overall approach with K.C. and L.Z. L.Z. carried out experiments and wrote the manuscript with M.G. W.F. and Y.W. contributed to the analytical experiments. D.L. designed the regulating part of TD-tweezers. All the authors discussed the results and commented on the manuscript. Competing interests: The authors declare that they have no competing interests. The authors declare that a patent has been filed relating to this work. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors.

Submitted 28 January 2020Accepted 25 March 2020Published 29 May 202010.1126/sciadv.abb0695

Citation: L. Zhou, M. Gao, W. Fu, Y. Wang, D. Luo, K. Chang, M. Chen, Three-dimensional DNA tweezers serve as modular DNA intelligent machines for detection and regulation of intracellular microRNA. Sci. Adv. 6, eabb0695 (2020).

on October 17, 2020

http://advances.sciencemag.org/

Dow

nloaded from

and regulation of intracellular microRNAThree-dimensional DNA tweezers serve as modular DNA intelligent machines for detection

Lin Zhou, Mingxuan Gao, Weiling Fu, Yunxia Wang, Dan Luo, Kai Chang and Ming Chen

DOI: 10.1126/sciadv.abb0695 (22), eabb0695.6Sci Adv 

ARTICLE TOOLS http://advances.sciencemag.org/content/6/22/eabb0695

MATERIALSSUPPLEMENTARY http://advances.sciencemag.org/content/suppl/2020/05/21/6.22.eabb0695.DC1

REFERENCES

http://advances.sciencemag.org/content/6/22/eabb0695#BIBLThis article cites 37 articles, 4 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.Science AdvancesYork Avenue NW, Washington, DC 20005. The title (ISSN 2375-2548) is published by the American Association for the Advancement of Science, 1200 NewScience Advances

License 4.0 (CC BY-NC).Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution NonCommercial Copyright © 2020 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on October 17, 2020

http://advances.sciencemag.org/

Dow

nloaded from