Cascaded, Feedback-Driven, and Spatially Localized Emergence of Constitutional Dynamic Networks Driven by Enzyme-Free Catalytic DNA Circuits

The enzyme-free catalytic hairpin assembly (CHA) process is introduced as a functional reaction module for guided, high-throughput, emergence, and evolution of constitutional dynamic networks, CDNs, from a set of nucleic acids. The process is applied to assemble networks of variable complexities, functionalities, and spatial confinement, and the systems provide possible mechanistic pathways for the evolution of dynamic networks under prebiotic conditions. Subjecting a set of four or six structurally engineered hairpins to a promoter P1 leads to the CHA-guided emergence of a [2 × 2] CDN or the evolution of a [3 × 3] CDN, respectively. Reacting of a set of branched three-arm DNA-hairpin-functionalized junctions to the promoter strand activates the CHA-induced emergence of a three-dimensional (3D) CDN framework emulating native gene regulatory networks. In addition, activation of a two-layer CHA cascade circuit or a cross-catalytic CHA circuit and cascaded driving feedback-driven evolution of CDNs are demonstrated. Also, subjecting a four-hairpin-modified DNA tetrahedron nanostructure to an auxiliary promoter strand simulates the evolution of a dynamically equilibrated DNA tetrahedron-based CDN that undergoes secondary fueled dynamic reconfiguration. Finally, the effective permeation of DNA tetrahedron structures into cells is utilized to integrate the four-hairpin-functionalized tetrahedron reaction module into cells. The spatially localized miRNA-triggered CHA evolution and reconfiguration of CDNs allowed the logic-gated imaging of intracellular RNAs. Beyond the bioanalytical applications of the systems, the study introduces possible mechanistic pathways for the evolution of functional networks under prebiotic conditions.

For emergence of a [2×2] CDN "O" composed of AA1, AB1, BA1 and BB1, four hairpins, HA, HA1, HB, and HB1, 1 μM each, were mixed, and then the primer P1, 1 μM or 10 nM, was subjected to the mixture, followed by incubating at 25 o C for 1 hour. The control experiment was performed under the same condition yet in the absence of P1.
For cascaded emergence of CDN "R" and CDN "S", eight hairpins, HG, HG1, HH, HH1, HI, HI1, HJ, and HJ1, 1 μM each, were mixed, and then the primer P3, 1 μM, was subjected to the mixture followed by incubating at 25 o C for 3 hours. The control experiment was performed under the same condition yet in the absence of P3.
For feedback-driven emergence of CDN "O", four hairpins, HA, HA1, HB, and HB1, and caged substrates S1 and S2 were mixed, and the primer P1, 10 nM, was subjected to the mixture followed by incubating at 30 o C. The non-cross-catalytic feedback circuit using non-ribonucleobase-containing substrates was performed under the same condition.
For emergence of CDN "T" attached on tetrahedra, the mixtures of SK, SK1, SL, SL1, HK, and HL, 1 μM each, were heated at 95 °C for 5 min and cooled to 4 o C within 1 min. S7 Subsequently, 1.5 μM HK1, and HJ1, were incubated with the as-prepared HK-/HLfunctionalized tetrahedron at 37 o C for 1 hour followed by purifying by ultrafiltration (30 kDa molecular weight cutoff) to remove the excessive HK1, and HL1 remaining in solution. The primer P4, 1 μM, was subjected to the mixture followed by incubating at 25 o C for 3 hours. The control experiment was performed under the same condition yet in the absence of P4.

Methods
Evaluation of the concentrations of the constituents in the CDNs. Taking CDN "O" as an example, 100 µL of as-prepared equilibrated CDN mixture was treated with the corresponding fluorophore/quencher substrate (sub). As an example, to probe constituent AB1 in CDN "O", 100 µL of the equilibrated CDN mixture was treated with the sub 12 (5 µL of 100 µM). Subsequently, the time-dependent fluorescence changes generated from the cleavage of sub 12 by the Mg 2+ -dependent DNAzyme associated with the AB1 were followed at 25 o C. Using the appropriate calibration curve corresponding to the rates of cleavage of the different substrates by different concentrations of the intact constituent (see detailed description in Figures S1 and S2), the contents of the constituent AB1 in the different CDN "O" were evaluated. It should be noted that the concentrations of the constituents were not evaluated in a single step, but followed separately, according to this procedure for each of the constituents using the appropriate Fi/Qi-substrate and the relevant calibration curve. The changes in the fluorescence intensities, as a result of the DNAzyme catalyzed cleavage of the respective substrates, ∆F, correspond to the measured fluorescence values, Fi, from which the background fluorescence of the respective fluorophore associated with the substrate, F0, was subtracted.
Fluorescence Assay. For detection of miRNAs using localized CHA system, different concentrations of miRNA-21 and miRNA-155 were added to 100 μL of 1 × PB buffer solution containing 90 nM tetrahedron sensing module, followed by incubation for 5 hours at room temperature. The FRET fluorescence spectra between Cy3 and Cy5 were S8 collected from 540 to 800 nm with excitation wavelength at 520 nm.
Human liver cancer cells (HepG2) were grown in 5% CO2 DMEM medium supplemented with 10% FCS and antibiotics (KeyGEN BioTECH, China). Cells were planted one day prior to the experiment on µ-slide 4 well glass bottom for confocal microscopy.
Confocal microscopy measurements. For cell imaging experiments, one day prior to the experiment, cells were planted in µ-slide 4 well glass bottom. Cells were incubated with the tetrahedral DNA nanostructures after washing with PBS. The tetrahedra (100 nM) were incubated with cells for 7 hours and then washed with DMEM-Hepes twice and replenished with the fresh medium for the measurement. An external 561 nm excitation with an accompanying emission ranging from 570 to 620 nm was selected for the green channel of fluorophore (Cy3) donor. The external 561 nm FRET stimulation with an accompanying emission signal collection ranging from 650 to 700 nm was selected for the yellow channel of fluorophore (Cy5) acceptor. To achieve a reliable FRET readout, the background FRET signal, originating from solely Cy3/Cy5 fluorophore, was subtracted from each of the samples. Figure S1. Time-dependent fluorescence changes generated upon the cleavage of the fluorophore/quencher-modified substrates by the respective Mg 2+ -ion-DNAzyme reporter units associated with the individual intact constituents at variable concentrations: (i) 1 μM, (ii) 0.8 μM, (iii) 0.6 μM, (iv) 0.4 μM, and (v) 0.2 μM. S10 Figure S2. Corresponding calibration curves of the catalytic rates of the different constituents as a function of their concentrations, derived from the data shown in Figure  S1. and BB1 at known concentrations (1 μM), we evaluated, using Image J software, the contents of the constituents in CDN "O", Table S1. As expected, we find that the contents of the constituents in the electrophoretically separated mixture of CDN "O" is similar to those evaluated by the DNAzyme reporter units, cf. Table 1 in the text. From the intensities of the stained separated bands for evolved CDN "P" shown in lane 10, Figure S4B, and using the stained bands of the individual intact constituents AA1, AB1, AC1, BA1, BB1, BC1, CA1, CB1, and CC1 at known concentrations (1 μM), we evaluated, using Image J software, the contents of the constituents in CDN "P", Table S2. As expected, we find that the contents of the constituents in the electrophoretically separated mixture of CDN "P" is similar to those evaluated by the DNAzyme reporter units.  Using ImageJ software and comparing the intensities of the separated bands to those of the individual constituents at known concentrations (1 μM), the contents of the constituents in CDN "Q" generated upon the P2-induced branched CHA reaction were evaluated, and the results were summarized in Table S3.  Using ImageJ software and comparing the intensities of the separated bands to those of the individual constituents at known concentrations (1 μM), the contents of the constituents in CDNs "R" and "S" generated upon the P3-induced two-layer CHA cascade were evaluated, and the results were summarized in Table S4. The concept of nonenzymatic two-layer CHA cascade amplification circuit and four fluorescence readout signals generated by DNAzyme reporter units associated with constituents in evolved CDN "R" were used to develop a universal sensing platform, as displayed in Figure S11. The miRNA-21 used in the present study is a unique oncogenic biomarker that is overexpressed in most types of cancer. The auxiliary hairpin "HP1" was introduced to recognize and hybridize with miRNA-21 analytes to release the trigger sequence P3, and upon opening, it activates the well-established twolayer CHA cascade amplification circuit. Figure S11 schematically depicts two-layer CHA cascade circuit for amplified analysis of miRNA-21. The analyte miRNA-21 hybridizes with and opens the auxiliary hairpin "HP1" to release the trigger sequence P3.

S9
The miRNA-21-triggered release of P3 stimulates the first-layer CHA to form four duplex constituents GG1, GH1, HG1, and HH1 comprising CDN "R", which in turn catalyzes the second-layer CHA reaction for the formation of CDN "S" composed of II1, IJ1, JI1, and JJ1. Each of the constituents in evolved CDN "S" includes a different Mg 2+ -ion-dependent DNAzyme that catalyzes the cleavage of the respective fluorophore/quencher-modified substrate, thus leading to a substantial fluorescence increase in the DNAzyme amplification stage. That is, the four intercommunicating constituents in the evolved CDN "S" are anticipated to yield four different fluorescence signals, providing four synergistic readout signals for reliable quantitative analysis of miRNA-21. Figure S11B, panel I shows fluorescence intensities generated by DNAzymes associated with four constituents, II1, IJ1, JI1, and JJ1 in CDN "S" before and after addition of miRNA-21. Upon the addition of miRNA-21, the fluorescence intensities associated with four constituents II1, IJ1, JI1, and JJ1 are intensified, demonstrating the multiple readout fluorescence signals for sensing of miRNA-21 is, indeed feasible. The sensing performance of the proposed two-layer CHA circuit was investigated using fluorescence intensity associated with constituent JI1 as the readout signal. Figure S11B, panel II and III, depicts the fluorescence spectra and derived calibration curve in the presence of variable concentrations of miRNA-21, respectively.
The detection limit for sensing miRNA-21 corresponds to 10 pM. Figure S11B, panel III, shows the selectivity features corresponding to the analysis of miRNA-21 by twolayer CHA cascade circuit. At a concentration of 50 nM of miRNAs, the signal transduced by two-layer CHA cascade circuit is approximately threefold enhanced in the presence of the target miRNA-21, as compared to the set of foreign miRNAs. To demonstrate the advantage of two-layer cascade circuit, the performance of single-layer CHA was also characterized. Under the same concentration of miRNA-21, the two-S21 layer cascade system shows a much higher fluorescence intensity than that of singlelayer CHA, Figure S11B, panel V, indicating a dramatic signal amplification efficiency of the two-layer CHA cascade. Two-layer CHA cascade amplification circuit reveals sensitivity improvements of two orders of magnitude over single-layer CHA, Figure   S11B, panel VI.  will be addressed in a comprehensive future report, we selected the feedback circuit introduced in Figure 5 as an example to assemble the kinetic model and to simulate computationally the experimental results.
The set of reactions summarized in eq. 1 -eq. 11 account for the temporal feedback driven emergence of the CDN "O" constituents from the set of hairpins HA, HA1, HB, and HB1, triggered upon the P1-triggered activation of the hairpins. Figure 5C, dotted curves represent the experimentally-evaluated dynamic evolution of constituent AA1 and AB1. These experimental results were fitted to the kinetic model outlined in eq. 1eq. 11. The best fitted computational temporal emergence of AA1 and of AB1 are overlayed as solid curves over the experimental data Figure S13. The computationallysimulated rate constants corresponding to the stepwise reaction framework comprising the kinetic model are summarized in Table S5. This framework of rate-constants, may, then, be applied to predict the non-linear temporal emergence of the constituents AA1 and AB1 at variable auxiliary conditions of the hairpins/P1 as inputs.      Similar to the two-layer cascade circuit, the feedback amplification circuit was also used to develop a universal sensing platform for analysis of miRNA-21, as displayed in Figure S14. The auxiliary hairpin "HP2" was introduced to recognize and hybridize with miRNA-21 analyte to release the trigger sequence P1, and upon opening, it activates the well-established feedback amplification circuit. Figure S14 schematically S26 depicts the feedback amplification circuit for analysis of miRNA-21. The analyte miRNA-21 hybridizes with and opens the auxiliary hairpin "HP2" to release the trigger sequence P1. The miRNA-21-triggered release of P1 stimulates the CHA to form four duplex constituents AA1, AB1, BA1, and BB1 comprising CDN "O". Subsequently, the DNAzymes associated with constituents BA1 and BB1 cleave the caged substrates S1

Kinetic equations of the feedback-driven CHA circuit shown in
and S2, respectively, to continuously generate protected P1 sequences that could, in turn, Under the same concentration of miRNA-21, the feedback system shows a much higher S27 fluorescence intensity than that of non-feedback circuit, Figure S14B, panel V, indicating a dramatic signal amplification efficiency of the feedback circuit. The feedback amplification circuit reveals sensitivity improvements of three orders of magnitude over non-feedback circuit, Figure S14B, panel V.
S28 Figure S15. Agarose gel (4%) electrophoretic image demonstrating the formation of hairpins-functionalized tetrahedra. Using ImageJ software and comparing the intensities of the separated bands to those of the individual constituents at known concentrations (1 μM), the contents of the constituents in CDNs "T", "U" and "V" were evaluated, and the results were summarized in Table S6.
S32 Figure S19. Selectivity of localized CHA sensor in the form of a bar presentation. Error bars derived from N = 3 experiments. Table S1. The contents of the constituents in CDN "O" evolved by the primer P1triggered activation of CHA reaction, shown in Figure 1. CDN AA1 AB1 BA1 BB1 O a 0.51 0.51 0.61 0.48 a The contents of the constituents (μM) were evaluated from the quantitative electrophoretic experiment in Figure S3. Table S2. The contents of the constituents in CDN "P" evolved by the primer P1triggered activation of CHA reaction shown in Figure Figures S1 and S2. b The contents of the constituents (μM) were evaluated from the quantitative electrophoretic experiment in Figure S4. Table S3. The composition of the constituents in CDN "Q" evolved by the primer P2triggered activation of the branched CHA, shown in Figure 3. CDN D1E1F1 D1E1F2 D1E2F1 D1E2F2 D2E1F1 D2E1F2 D2E2F1 D2E2F2 Q a 0.26 0.31 0.27 0.19 0.25 0.25 0.28 0.21 a The contents of the constituents (μM) were evaluated from the quantitative electrophoretic experiment in Figure S6.
It should be noted that the concentrations of the constituents are not equal. This is consistent with the fact that the concentrations of the constituents are dictated by the relative stabilities of the constituents that might include secondary intra-constituent structure, and eventually further interactions with auxiliary strands. Table S4. The composition of the constituents in CDN "R" and CDN "S" evolved by the primer P3-triggered activation of the two-layer CHA cascade circuit, shown in  Figures S8 and S9. b The contents of the constituent (μM) were evaluated from the quantitative electrophoretic experiments in Figure S10. Table S5. Rate constants derived from the feedback-driven CHA circuit shown in Figure 5. 0.25 --0.20 a The contents of the constituents (μM) were evaluated from the time-dependent fluorescence changes generated by the reporter units and using appropriate calibration curve in Figures S16 and S17. b The contents of the constituents (μM) were evaluated from the quantitative electrophoretic experiment in Figure S18.