Open AccessCCS ChemistryRESEARCH ARTICLES6 Oct 2022Construction of a Stimuli-Responsive DNAzyme-Braked DNA Nanomachine for the Amplified Imaging of miRNAs in Living Cells and Mice Yingying Chen†, Mingzhu Yan†, Yushi Wang, Jinhua Shang, Shizhen He, Yuhui Gao, Chen Hong, Xiaoqing Liu and Fuan Wang Yingying Chen† College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072 †Y. Chen and M. Yan contributed equally to this work.Google Scholar More articles by this author , Mingzhu Yan† College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072 †Y. Chen and M. Yan contributed equally to this work.Google Scholar More articles by this author , Yushi Wang College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072 Google Scholar More articles by this author , Jinhua Shang College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072 Google Scholar More articles by this author , Shizhen He College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072 Google Scholar More articles by this author , Yuhui Gao College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072 Google Scholar More articles by this author , Chen Hong College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072 Google Scholar More articles by this author , Xiaoqing Liu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072 Google Scholar More articles by this author and Fuan Wang *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202202171 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Synthetic DNA motors have spurred considerable research interest as a way to interrogate biochemical processes that can further facilitate disease diagnosis. However, current direct sense-on-demand DNA motors are hampered by undesired signal leakage originating from their nonspecific stimulation prior to their arrival at target sites. Herein, we propose a DNAzyme-guided DNA amplification motor through the facile built-in of a stimuli-responsive DNAzyme brake to facilitate the in vivo sensitive imaging of biomarkers by minimizing the off-site signal leakage in living systems. To realize the on-site activation, the inactive DNAzyme DNA motor was encapsulated into the folate (FA)-modified zeolitic imidazolate framework-8 (ZIF-8)/poly(d,l-lactic-co-glycolic acid (PLGA) nanovesicles. The FA interaction-promoted tumor-cell-selective delivery and the pH-responsive disassembly of ZIF-8/PLGA permitted sufficient release of DNAzyme cofactors to achieve on-site stimulation of DNAzyme for restoring the temporally caged DNA amplification motor in tumor cells where the endogenous miRNA target was thus specifically and sensitively analyzed. By integration of the cell-selective delivery and the site-specific stimulation, our DNAzyme-guided DNA motor can realize the reliable imaging of tumor cells and provide a new toolbox for disease diagnosis. Download figure Download PowerPoint Introduction Natural biomolecular motors, like functional proteins, can execute stimulus-specific commands via conformation changes to retain the viability of live cells.1,2 Inspired by cellular motors, numerous artificial molecular motors have thus been developed to uncover underlying biomolecular mechanisms and functions and eventually to realize the diagnosis-guided treatment of various diseases.3–6 Due to their high biocompatibility/controllability and variable molecular recognition, synthetic DNA motors have supplemented an attractive toolbox for interrogating biochemical processes.7–13 Particularly, by integrating a built-in amplification function, extensive progress has been made with catalytic DNA motors for sensitive biosensing and bioimaging research.14–17 However, current direct sense-on-demand DNA motors have met with undesired signal leakage originating from their off-site stimulation in living systems. Therefore, the engineering of on-site stimulation of DNA amplification motors is highly appealing to realize the specific bioimaging applications by preventing the undesired leakage reaction. To prevent the off-site signal leakage, these DNA motors were further stabilized by introducing an additional functional DNA brake unit, which was then specifically liberated by using photoactivation,18–20 microRNAs (miRNAs),21,22 and protein enzymes23–26 to control the on-site stimulation of DNA motors. Despite satisfactory controllability, external photoactivation can be restricted by limiting the penetration depth and forward exposure area. Meanwhile, the trace expression and varied distributions of miRNAs and enzymes are unfavorable for achieving the stimuli-responsive activation of DNA motors in specific tumor cells. Worse still, the limited availability of exotic enzymes severely impedes their in-depth utility in live systems.27,28 This requires the development of a more appropriate toolbox for achieving the precise activation of DNA motors. As a self-sustainable built-in stimulus, the self-hydrolyzing catalytic nucleic acid (DNAzyme) represents an ideal candidate for programming the on-site regulation of DNA motors,29–31 yet it is still unexplored in living systems. To achieve DNAzyme regulation, the accurate DNAzyme delivery and sufficient intracellular cofactors supply need to be fulfilled because DNAzyme biocatalysis requires the intrinsic metal-ion cofactors.32,33 Thus, it is imperative to program a synthetic DNA nanomachine to achieve sufficient cofactor supply as well as the on-site stimulation of DNAzyme. Especially, the cell-selective delivery and programmable stimuli-responsive activation of the built-in DNAzyme is highly appealing to realize the accurate operation of DNA motors and to obtain the sensitive imaging of given tumor cells. Based on the above-mentioned considerations, herein we propose a successive and multiple cell-specific stimuli-guaranteed DNA amplification motor through the built-in stimuli-responsive DNAzyme brake to facilitate the spatially selective imaging of endogenous biomarkers by minimizing the nonspecific background signal. As shown in Scheme 1, the hybridization chain reaction (HCR)-based DNA motor consisted of two hairpin probes H1 and H2-X. The toehold of H2-X was caged tightly by an additional DNA-hydrolyzing DNAzyme I-R3 that acted as a self-supplied molecular switch for the stimuli-responsive motor activation. For sustaining the tumor-specific stimulation of DNAzyme, the DNAzyme-functional HCR motor was loaded into zeolitic imidazolate framework-8 (ZIF-8) nanoparticles that were further encapsulated by folate-modified poly(d,l-lactic-co-glycolic acid) (FA-PLGA) membranes. Under normal physiological conditions, the DNAzyme-functional HCR motor-laden FA-PLGA/ZIF-8 (F-ZD) system, which refers to nanostructures obtained by encapsulating HCR motor-laden ZIF-8 nanostructures into folate-decorated PLGA nanovesicles. Then these multivalent FA conjugates can specifically bind with the overexpressed FA receptors of the tumor cells, thus facilitating the selective delivery of the DNAzyme-functional HCR motor to tumor cells where the acidic endo/lysosomal microenvironment triggers the pH-responsive disassembly of ZIF-8 to release a sufficient amount of Zn2+ cofactors. These generated Zn2+ ions can then catalyze the cleavage of H2-X to produce the active H2 probe, thus restoring the miRNA-sensing HCR motor to achieve the more precise bioimaging in vivo. As robust guidance, the DNAzyme can realize the on-site activation of the DNA motor and benefit the tumor-specific miRNA imaging by minimizing the nonspecific background readout during the systemic delivery process. Scheme 1 | (A) Fabrication of the pH-responsive FA-functionalized DNA amplification motor-laden nanocapsule. (B) Scheme of the Zn2+-mediated self-cleaving DNAzyme-guided DNA amplification motor for tumor-cell-specific miRNA imaging. Download figure Download PowerPoint Experimental Methods Flow cytometry analysis The MCF-7 cells (human breast tumor cell line), HeLa cells (cervical tumor cell line), and MCF-10A cells (human breast normal cell line) were seeded at a density of 2 × 105 per well onto a 6-well plate and cultured. When the cells reached 90% confluency, the F-ZD nanoparticles dispersed in the Opti-minimum essential medium (MEM) were transfected into the cells. After incubation for 4 h, the cells were trypsinized, centrifuged, and washed with phosphate-buffered saline (PBS) for flow cytometry assay. MiR-21 imaging in live cells The MCF-7 cells, HeLa cells, and MCF-10A cells were seeded at a density of 2 × 105 per well onto a 6-well plate and cultured for transfection. The F-ZD nanoparticles were incubated with the plated cells (for 3 h at 37 °C), by replacing the Dulbecco’s modified Eagle’s medium with Opti-MEM medium containing F-ZD nanoparticles. All these treated cells were then washed with PBS to remove the noninternalized nanoparticles for the purpose of confocal microscopy imaging. The FA/FD (fluorescence emission ratio of acceptor to donor) was used as the output of fluorescence imaging to increase the anti-interference ability of the activatable DNA motor against complex intracellular environment. The antimiR-21 inhibitor assay was performed by transfecting the MCF-7 cell line with inhibitor antisense oligonucleotide (0.1 nmol) using commercial lipofectamine 3000 (300 μL) for 2 h at first, followed by the F-ZD nanoparticles transfection as above. Live animal imaging The female BABL/c nude mice (6–8-weeks old) were purchased from Vital River (Beijing, China) to build up the tumor model. All animal experiments were guided by the Regulations for the Administration of Affairs Concerning Experimental Animals approved by the State Council of the People’s Republic of China, which were approved by the Wuhan University Center for Animal Experiment/Animal Biosafety Level-III Laboratory (S07918050D). Tumor volume was calculated with the equation: V = (L × W2)π/6 (L and W are the longest and shortest diameters of the tumor, respectively). When the volume of the MCF-7 tumor reached ∼200 mm3, the nude mice were arbitrarily divided into five groups and intravenously injected with the c-ZD nanoparticles loaded with oligonucleotide ( H1* + H2-12*) or F-ZD nanoparticles loaded with oligonucleotide ( H1* + H2-12*, H1* + mH2-12*, miR-21 inhibitor + H1* + H2-12*, or miR-21 inhibitor + H1* + mH2-12*, 10 nmol/kg each). Then the in vivo imaging was collected by IVIS Spectrum (Ex: 640 nm and Em: 680 nm) at 1, 4, 8, and 12 h after intravenous injection. The ex vivo imaging of main organs (heart, liver, spleen, lung, and kidney) and tumors was carried out after the mice were euthanized. Results and Discussion Work mechanism of the DNAzyme-guided DNA motor for the amplified microRNA imaging To construct the DNAzyme-controlled molecular motor for tumor cell-specific miRNA imaging, we engineered a polymer nanocomposite by encapsulating the Zn2+-mediated self-cleaving DNAzyme-guided HCR system and the ZIF-8 adjuvant into the FA-modified PLGA nanovesicles (Scheme 1A). By conjugating FA on PLGA nanovesicles, the as-designed F-ZD nanocapsules could specifically bind with the FA receptor that were overexpressed in various tumor cells,34,35 endowing the DNA motor with tumor-targeting capacity as compared to normal cells.36 These multivalent FA conjugates trafficked quickly into lysosomes during FA receptor-mediated endocytosis.37–39 Thus, the acidic microenvironment (pH 5.0) induced the disassembly of the pH-responsive DNA motor-laden ZIF-8 nanostructures (ZD) nanoparticles to release the DNAzyme cofactors (Zn2+ ions) for mediating the DNAzyme-guided HCR system. As shown in Scheme 1B and Supporting Information Scheme S1, Tables S1 and S2, the released DNAzyme-activatable HCR system contains two metastable hairpin DNA substrates, a fluorophore acceptor Cy5-labeled H1 and a fluorophore donor Cy3-labeled H2-X. The hairpin H2-X contained an additional loop region ( e–c* and f) that corresponded to the Zn2+-dependent DNA-cleaving DNAzyme I-R3,40 which acted as a built-in stimuli-responsive molecular brake. Since the stem of hairpin H2-X was firmly caged by duplex region g–g*, the cross-reaction between H1 and H2-X was thus substantially retarded by the strong intramolecular hybridization of H2-X. With the addition of Zn2+ ions, the intact and “inactive” H2-X probe was decaged by I-R3 DNAzyme for generating the “active” H2 probe. Then the recognition between target miRNA and hairpin H1triggered the autonomous HCR process, involving the cross-opening of H1 and the newly exposed H2 to produce the dsDNA polymeric nanowires with an amplified Förster resonance energy transfer (FRET) transduction signal. This stimuli-responsive on-site activation strategy controlled the DNA motors in tumor cells and distinguished tumor cells from normal cells, thus realizing the more sensitive and tumor-specific miRNA imaging. Preparation and characterization of the F-ZD nanocapsules The stimulus-responsive F-ZD nanocapsules were synthesized by the following steps: (1) construction of ZD, (2) preparation of FA-conjugated diblock copolymer PLGA-PEG-FA, and (3) encapsulation of ZD nanoparticles into FA-decorated PLGA nanovesicles (FP). The nanosized ZD (∼80 nm in diameter, Supporting Information Figure S1) was synthesized by a facile one-step process to obtain the loading of the DNA motor.41,42 Next, to construct the FA-conjugated diblock copolymer PLGA-PEG-FA, the amine-terminated diblock copolymer PLGA-PEG was prepared in advance. The reaction of PLGA with an excess polyethylene glycol (PEG) bis-amine generated the copolymer PLGA-PEG ( Supporting Information Figure S2A), and the 1H NMR spectra illustrated the characteristic peaks of PLGA (1.5, 4.7, and 5.1 ppm, respectively) and PEG (3.5 ppm). The PLGA-PEG was further reacted with the activated FA to yield the conjugate PLGA-PEG-FA, and the characteristic peaks of FA (6.7, 7.7, and 8.7 ppm, respectively) demonstrated the successful preparation of PLGA-PEG-FA ( Supporting Information Figure S2B).43,44 UV–vis spectra exhibited the characteristic absorbance of FA at 357 nm for conjugate PLGA-PEG-FA ( Supporting Information Figure S3) and revealed the successful coupling of FA with PLGA-PEG precursor.44 Finally, the F-ZD nanocapsules were prepared by the double emulsion solvent evaporation method with a 71% encapsulation efficiency of ZD as determined by the inductively coupled plasma atomic emission spectroscopy (ICP-AES). The morphological features of F-ZD nanocapsules were confirmed by scanning electron microscopy (SEM) (Figure 1A) and transmission electron microscopy (TEM) (Figure 1A inset). Further element analysis by SEM energy dispersive X-ray spectroscopy (EDX) spectrum (Figure 1B) and dark-field scanning TEM and energy-dispersive system mapping ( Supporting Information Figure S4) verified the coexistence of carbon, nitrogen, oxygen, zinc, and phosphorus elements, suggesting the encapsulation of ZD into FP nanovesicles. As shown in Figure 1C, the F-ZD nanocomposites displayed an average diameter of 305 nm (polydispersity index, PDI, 0.13), as compared with FP nanocarriers (297 nm, PDI, 0.17), thus the encapsulation of ZD had no significant effect on the morphology of nanocapsules. The zeta potential of nanoparticles was also explored. Compared with the positively charged ZIF-8 nanocarriers (14.6 mV), the negatively charged ZD (−25.1 mV) implied the incorporation of the DNAzyme-guided DNA amplification motor (Figure 1D). As compared with FP nanovesicles with a negative charge (−19.1 mV), the ultimate F-ZD nanocapsules were encoded with more negative charges (−33.6 mV), indicating the successful ZD encapsulation. The pH-stimulated Zn2+ release from F-ZD nanoparticles wasalso investigated by ICP-AES analysis ( Supporting Information Figure S5). The results showed an almost complete release of Zn2+ cofactors in the lysosomal condition (pH 5.0), and nearly no dissolution of F-ZD in the physiological condition (pH 7.4), thus facilitating the on-site activation of the DNA motor for intracellular imaging application. Figure 1 | (A) SEM and TEM analyses of F-ZD nanocapsules. (B) SEM EDX spectrum of F-ZD nanocapsules. (C) Dynamic light scattering analysis of the FP and F-ZD nanoparticles. (D) Zeta potential measurements of ZIF-8, ZD, FP, and F-ZD nanoparticles. Download figure Download PowerPoint Optimization of Zn2+-mediated DNAzyme cleavage reaction and DNAzyme-functional DNA probe For DNAzyme-guided H2-X, where X refers to the sequence length of duplex g–g* segment ( Supporting Information Scheme S1), hairpin H2-12 was selected as the optimal candidate, and over 80% of H2-12 was cleaved within 1 h of incubation ( Supporting Information Figure S6a). And the more favorable DNAzyme cleavage was conducted in the physiological condition (pH 7.4, Figure 2A). The concentration of Zn2+ ions was optimized as 0.5 mM, under which more than 80% of substrate was cleaved by the built-in DNAzyme ( Supporting Information Figure S6B). Then the concentration of intracellular Zn2+ was estimated by ICP-AES for the intact and F-ZD-treated HeLa cells. After the transfection of F-ZD (240 μg/mL) into Hela cells, the intracellular concentration of Zn2+ was acquired to be 538 μM (Figure 2B), which was sufficient to activate the DNAzyme-guided HCR system in live cells. Interestingly negligible toxicity was observed for both normal cells and cancer cells that were treated with F-ZD under the same dosage ( Supporting Information Figure S7), thus revealing the high biocompatibility of the F-ZD nanocomposite. To realize the on-site activation of the DNA motor in target tumor cells, its miRNA-sensing function should be temporally caged without DNAzyme biocatalysis and then can be restored after the DNAzyme cleavage reaction. Accordingly, the length of caging duplex g–g* was optimized for the H2-X probe ( H2-8, H2-10, H2-12, and H2-16, Supporting Information Scheme S1). As shown in Supporting Information Figure S8A, the length variation of the caging sequence had scarcely no influence on the DNAzyme cleavage efficiency. With increasing length of the caging sequence, the background signal decreased, and H2-12 with a 12-base-long caging sequence was sufficient to retard the miRNA-initiated HCR assembly prior to DNAzyme cleavage ( Supporting Information Figure S8B). Thus, H2-12 was selected for the following assays to maximally lower the possible off-site activation and nonspecific signal leakage. Figure 2 | (A) Denatured PAGE analysis of DNAzyme-guided H2-12 activation at different pH values. (B) ICP-AES quantification of Zn2+ for the intact and F-ZD-treated HeLa cells. (C) Time-dependent fluorescence responses to 10 nM miR-21 (FA/FD, the fluorescence emission ratio of Cy5 to Cy3). The intact HCR system: curve a, H1+H2-12, curve b, miR-21 +H1+H2-12. And the Zn2+-mediated DNAzyme-guided HCR system: curve c, Zn2++H1+H2-12, curve d, Zn2++miR-21+H1+H2-12. (D) Illustration of the respective reaction response of the intact and DNAzyme-guided HCR system to miRNA analyte. (E) Fluorescence responses of the intact HCR system (curve a, H1+H2-12) and the DNAzyme-guided HCR system (curve b, Zn2++H1+H2-12) to increased concentrations of miR-21. (F) The linear relationship (curve b, panel E) between FRET signals and miR-21 (0.05–10 nM). (G) Evaluation of the Zn2+-specific DNAzyme cleavage. (H) Selectivity toward different types of miRNAs. Data were means ± SD (n = 3). Download figure Download PowerPoint Feasibility of the DNAzyme-powered DNA motor The feasibility of our DNAzyme-powered DNA motor was investigated by fluorescence analysis. The inactive probe H2-X was expected to be incapable of executing the HCR process since the stem region was tightly locked by the long duplex segment. As depicted in Figure 2C, in the presence or absence of miR-21, a negligible FRET signal was observed for the intact HCR system ( H1+ H2-12) (curves a and b, Figure 2C), which minimized the undesired off-site activation during cell delivery. Upon the addition of Zn2+ cofactors, the I-R3 DNAzyme transformed the HCR-inactive H2-12 into the HCR-active H2 via the self-cleavage reaction. Yet a much more slowly increased FRET readout was recorded without miR-21 (curve c, Figure 2C). Only in the presence of miR-21 was a substantially higher FRET signal produced (curve d, Figure 2C), indicating the crucial role of self-cleaving DNAzyme in executing the miR-21-triggered HCR motor. These fluorescence results were in good agreement with the corresponding native polyacrylamide gel electrophoresis (PAGE, Supporting Information Figure S9). Without miR-21, no background signal was observed in the mixture of H1+ H2-12 (lane 7) and Zn2++ H1+ H2-12 (lane 8). With miR-21, the DNAzyme-powered DNA motor (Zn2++ H1+ H2-12) produced lots of high-molecular-weight HCR products with low electrophoretic mobility (lane 10), which was in sharp contrast with the caged intact HCR system ( H1+ H2-12) without the generation of any product (lane 9). The self-cleaving DNAzyme was indispensable for recovering the miR-21-sensing capability of the HCR motor. Thus, these in vitro results validate the successful construction of our DNAzyme-powered DNA motor. The construction of a DNAzyme-manipulated molecular motor encouraged us to further explore its live cells analysis (Figure 2D). Initially, without Zn2+-mediated DNAzyme digestion, the intact HCR system ( H1+ H2-12) responded much more slowly to the increased concentrations of miR-21 (see curve a in Figure 2E and Supporting Information Figure S10A), and the limit of detection was calculated to be 4.4 nM from the 3σ calculation method ( Supporting Information Figure S10B). In stark contrast, the DNAzyme-guided HCR system (Zn2++ H1+ H2-12) revealed a notably enhanced sensitivity to miR-21 (see curve b in Figure 2E and Supporting Information Figure S10C). And the FRET signal showed a linear correlation with miR-21 concentration ranging from 0.05 to 10 nM (Figure 2F), from which a limit of detection of 30.5 pM was achieved. Hence, the built-in DNAzyme brake temporally caged the DNA amplification motor, and the DNAzyme cleavage reaction restored the miR-21-sensing function of DNA motor, thus paving the way for on-site imaging of miRNA. In addition to sensitivity, selectivity and specificity were also assessed. Considering the existence of trace metal ions in live cells, the DNAzyme biocatalysis was first evaluated by comparing the catalytic efficiency of Zn2+ ions with many other interfering metal ions (see Figure 2G and Supporting Information Figure S11A). Obviously, Zn2+ ions caused a distinctly higher DNAzyme biocatalysis than other interfering metal ions. Afterward, the miR-21-sensing capability of the DNAzyme-motivated DNA motor was explored to discriminate miR-21 from these mismatched miR-21 and interfering miRNAs. Target miR-21 displayed a significantly enhanced FRET signal, as compared to the mismatched ones ( Supporting Information Figure S11B and S11C), including single-base mismatched (1-Mut), double-base mismatched (2-Mut), or three-base mismatched (3-Mut) miR-21. While these interfering miRNAs (miR-155, miR-199a, let-7a, miR-429, and miR-141) produced negligible fluorescence change (Figure 2H and Supporting Information Figure S11D), thus indicating the specific miR-21-sensing performance of the activatable DNA motor. Thus, this Zn2+-DNAzyme-guided DNA motor can attain sensitive and selective miR-21 assay in live cells. Intracellular miR-21 imaging via the DNAzyme-guided DNA motor Prior to the intracellular assay, the robustness of our DNAzyme-propelled DNA motor was tested by treating it with 5% and 10% fetal bovine serum.45 In the simulated physiological condition, the FRET signal was hardly changed as compared with that in 4-(2-hydroxyethyl)piperazine-1 ethanesulfonic acid sodium salt (HEPES) buffer ( Supporting Information Figure S12). Thus, the robust DNA motor fulfilled the requirement of intracellular imaging. As illustrated in Figure 3A, the FA-coupled PLGA nanovesicles facilitated the delivery of F-ZD into FA receptor-overexpressed tumor cells with improved tumor-specific delivery performance. Then the tumor-targeting effect of FA modification was explored by flow cytometry. As a control, the control c-ZD nanoparticles were also prepared without tumor-targeting property (Figure 3B,C). The F-ZD nanocapsules achieved a 2.9-fold higher delivery efficacy than the c-ZD, thus confirming the FA-promoted delivery to tumor cells. In addition, the multivalent FA-conjugated nanocomposites are prone to internalize into lysosomes through the abundant FA receptor-mediated endocytosis in tumor cells.37–39 As shown in Supporting Information Figure S13, the F-ZD nanocapsules colocalized well with lysosome at 1 h, as revealed by the high overlapping of nanocapsules (Cy5-labeled H1) and LysoTracker Green (staining lysosome). However, after 3 h of transfection, the red fluorescence puncta diffused into the cytoplasm, thus manifesting the pH-responsive disassembly of F-ZD nanocapsules in the acidic lysosome. These released Zn2+ cofactors could activate the DNAzyme for recovering the HCR-amplified miRNA imaging. Figure 3 | (A) Schematic illustration of the FA-assisted DNAzyme-powered DNA motor to imaging MCF-7 cells with different miR-21 levels. (B) Flow cytometry analysis of the FA-assisted cellular uptake, .three cells were chosen, including the intact MCF cells as control (Ctrl), cells incubated with c-ZD nanocapsules (loaded with Cy5-labeled probe H1), and cells incubated with F-ZD nanocapsules (loaded with Cy5-labeled probe H1). (C) MFI of panel B. (D) Confocal fluorescence image of MCF-7 cells; the histogram analysis of FRET readout and (E) the corresponding image. Scale bar refers to 20 μm. ****p < 0.0001 (one-way analysis of variance [ANOVA] followed by Tukey’s multiple comparisons test). Data were means ± SD (n = 3). Download figure Download PowerPoint Visualization of tumor-related miRNAs and distinguishing their relative expressions supplies crucial information for tumor diagnosis. Inspired by the high sensitivity and robustness of the DNAzyme-guided HCR motor, we further studied its applicability in live cell imaging by confocal laser scanning microscopy (CLSM), where the fluorescence emission ratio of acceptor to donor (FA/FD) was adapted as the output for high anti-interference concern. The incubation duration of F-ZD was optimized at 3 h ( Supporting Information Figure S14), which was thus selected in the following transfection assays. To further clarify that the DNAzyme-guided DNA motor can evaluate the varied miRNA expression of tumor cells, MCF-7 cells were pretreated with miR-21 inhibitor (to downregulate miR-21 expression) and miR-21 mimic (to upregulate miRNA-21 expression) before the transfection of F-ZD. Compared with the unpretreated cells (group b, Figure 3D,E, and Supporting Information Figure S15a), the attenuated or intensified fluorescence signals were respectively collected for the miR-21 inhibitor (see group a, Figure 3D,E, and Supporting Information Figure S15A) or miR-21 mimic (see group c, Figure 3D,E, and Supporting Information Figure S15A) pretreated cells. Additionally, the quantitative reverse transcription-polymerase chain reaction (qRT-PCR) analysis was conducted to obtain the relative expression levels of miR-21 in MCF-7 cells. As shown in Supporting Information Figure S15B, the qRT-PCR analysis was in line with that of the quantified fluorescence readout. Therefore, these findings verified that the on-site DNAzyme-stimulation DNA amplification motor appropriately differentiated the varied miR-21 expressions in live cells. In addition, z-section images of the F-ZD nanocapsules-incubated MCF-7 cells revealed that the miR-21 molecules were distributed in the cytoplasm ( Supporting Information Video S1 and Figure S16), based on the spatial distr