Programmed catalysis within stimuli-responsive mechanically unlocked nanocavities in DNA origami tiles

The assembly of reversible stimuli-responsive locked DNA origami tiles being unlocked, in the presence of appropriate triggers, to form nanocavities in the origami rafts, is introduced. In the presence of ATP, K+-ion-stabilized G-quadruplexes or pH-responsive T-A·T triggers and appropriately engineered “helper units”, the origami rafts are unlocked to form nanocavities. By the application of appropriate counter-triggers, the nanocavities are relocked, thus establishing the switchable and reversible “mechanical” opening and closure mechanism of the nanocavities. The interconnection of the stimuli-responsive origami tiles into dimer structures enables the programmed triggered unlocking of each of the origami tiles, or both of the origami tiles, to yield dictated nanocavity-containing tiles. In addition, the functionalization of the opposite faces of the origami tiles with Mg2+-ion-dependent DNAzyme subunits leads, upon the triggered unlocking of the nanocavities, to the self-assembly of the active DNAzymes in the confined cavities. By the cyclic opening and closure of the cavities the reversible “ON”/“OFF” activation of the Mg2+-ion-dependent DNAzyme is demonstrated. Furthermore, upon the tethering of different Mg2+-ion-dependent subunits to the opposite faces of stimuli-responsive dimer origami tiles, the triggered programmed catalytic operation of different Mg2+-ion-dependent DNAzymes in the confined nanocavities, associated with the origami tiles, is demonstrated.

For the FRET test, the strand Ha-F, internally modified with Cy3, and the anchor strand A1-F modified with Cy5 were used for preparation of the K + -ion-responsive origami tile. The fluorescence features of the locked configuration of the origami tiles, 24 nM, were measured at λex = 532 nm. Subsequently, the origami tiles were unlocked by K + ions and the added hairpins H1 and H2. The fluorescence spectrum of the origami tiles in the open cavity-containing configuration was measured (λex = 532 nm). For deriving the calibration curve, mixtures of the strands Ha-F (24 nM) and A1-F (24 nM) were subjected to variable concentrations of H1-F (0, 6, 12, 18, and 24 nM) and their fluorescence spectra of the resulting hybrids were measured (λex = 532 nm).
To unlock the ATP-responsive origami tiles (2 nM, 80 μL, in TAE buffer with 6 mM Mg 2+ and 5 mM Na + ), the sample was treated with ATP (50 mM, 10 μL) and the helper strands H1 and H2 (100 nM, 10 μL) allowed to react for 10 hours at 25 °C. For the reverse re-locking process, the anti-helper strands (H1a'/H1b'/H2a'/H2b') (500 nM, 10 μL) were added into the sample to remove the strands H1 and H2 (25 °C for 2 h) and the counter-ATP aptamer strand C-ATPa (200 nM, 10 μL) was added to remove ATP from the ATP-aptamer complex followed by centrifuging to remove the free ATP from the sample (100 k NMWL, 3000 × g, 10 min, four times). Then the anti-C-ATPa strand, C-ATPa' (1 μM, 10 μL) was applied to remove C-ATPa and to form the M/M' locking duplex units of the "window". For the control measurements, only the ATP or the helper Switchable catalysis in the nanocavities of the single DNA origami tiles. For switching the catalytic process in ATP-responsive tiles, the ATP-responsive origami tiles functionalized with the E1a and E1b were prepared (30 nM, 200 μL). To induce the opening of the nanocavities and to activate the Mg 2+ -ion-dependent DNAzyme in the nanocavity, the tiles were treated with the ATP (60 mM, 20 μL) and the helper hairpins H1/H2 (1.5 μM, 20 μL) for 10 hours at 25 °C. To a sample, 49 μL of 15 nM, of the opencavity origami tiles was added with the ROX/BHQ2-modified substrate S1 (100 μM, 1 μL) and the catalytic activity of the DNAzyme was probed by following the fluorescence of the ROX modified fragmented product (λex = 550 nm). To reclose the open cavity tiles, the system was treated with the blocker strands B1a/B1b (3 μM, 10 μL) and the system was allowed to anneal from 30 °C to 10 °C for 2 hours. Subsequently, the anti-helper strands, H1a'/H1b'/H2a'/H2b' (15 μM, 10 μL) and C-ATPa (6 μM, 10 μL) were added to the system that was allowed to react for 2 hours. The resulting tiles were S4 four times centrifuged to remove the free ATP from the sample (100 k NMWL, 3000 × g, 10 min), and the strand C-ATPa' (12 μM, 10 μL) and the anti-blocker strands B1a′/B1b′ (10 μL, 6 μM) were added to the system to relock the cavities to yield the initial state tiles. The catalytic functions of the system were checked as described above. The reopening and reclosure of the cavities were performed by repeating the cycle and probing the catalytic activities of the system in the open/closed nanocavity states.
For switching the catalytic process in K + -ion-responsive tiles, the origami tiles functionalized with the E2a and E2b were prepared (30 nM, 200 μL). The tiles were treated with the K + ions (1.2 M, 20 μL) and the helper hairpins H1/H2 (1.5 μM, 20 μL) for 10 hours at 25 °C . To a sample, 49 μL of 15 nM, of the open-cavity origami tiles was added with the Cy5/BHQ2-modified substrate S2 (100 μM, 1 μL) and the catalytic activity of the DNAzyme was probed by following the fluorescence of the Cy5 modified fragmented product (λex = 630 nm). To reclose the open cavity tiles, the system was treated with the blocker strands B2a/B2b (1 μM, 30 μL) and the system was allowed to anneal from 30 °C to 10 °C for 2 hours. Subsequently, the anti-helper strands, H1a'/H1b'/H2a'/H2b' (5 μM, 30 μL) and crown-ether (CE, 57 mM, 2100 μL) were added to the system that was allowed to react for 2 hours at 25 °C to relock the cavities. The resulting tiles were washed and, centrifuged four times, to remove the excess strands and CE from the sample (100 k NMWL, 3000 × g, 10 min). The sample was treated with the anti-blocker strands B2a′/B2b′ (30 μL, 2 μM) to regenerate the initial state. The catalytic functions of the system were examined as described above. The re-opening and reclosure of the cavities were performed by repeating the cycle and probing the catalytic activities of the system in the open/closed nanocavity states.
For switching the catalytic process by pH-responsive tiles, the origami tiles functionalized with the E3a and E3b were prepared (30 nM, 200 μL). To induce the opening of the nanocavities and to activate the Mg 2+ -ion-dependent DNAzyme in the nanocavity, the tiles were treated with the helper hairpins H1/H2 (3 μM, 10 μL) for 10 hours at pH = 9.5. To a sample, 49 μL of 15 nM, of the open-cavity origami tiles was added the FAM/BHQ1-modified substrate S3 (100 μM, 1 μL) and the catalytic activity of the DNAzyme was probed by following the fluorescence of the FAM modified fragmented product (λex = 495 nm). To reclose the open cavity tiles, the system was treated with the blocker strands B3a/B3b (3 μM, 10 μL) and the system was allowed to anneal from 30 °C to 10 °C for 2 hours. Subsequently, the anti-helper strands, H1a'/H1b'/H2a'/H2b' (15 μM, 10 μL) were added to the system that was allowed to react for 2 hours at 25 °C to relock the cavities at pH 6. The sample was subjected to the antiblocker strands B3a′/B3b′ (10 μL, 6 μM) to yield the initial state. The catalytic functions of the system were checked as described above. The re-opening and reclosure of the cavities were performed by repeating the cycle and probing the catalytic activities of the system in the open/closed nanocavity states.
Programmed catalytic activities in the nanocavities within the ATP-/K + -ionresponsive origami dimers D1. To induce the programmed transition of the E1a/E1b and E2a/E2b-functionalized ATP-/K + -ion-responsive dimer D1 into the catalytically active dimer in state II, the dimer in state I (100 μL, 30 nM in a TAE buffer that contained S5 Mg 2+ , 6 mM, and Na + , 5 mM) was subjected to ATP (10 μL, 60 mM) and the hairpin strands H1/H2 (10 μL, 1.5 μM), for a time interval of 10 hours, to unlock the nanocavities in the ATP-responsive tile, and assemble the Mg 2+ -ion-dependent DNAzyme in the cavities. For the programmed opening of the K + -ion-responsive tile (marked) in state I into state III, the dimer in state I (100 μL, 30 nM) was subjected to K + ions (10 μL, 1.2 M) and the hairpin strands H1/H2 (10 μL, 1.5 μM), for a time interval of 10 hours, to unlock the nanocavities in the K + -ion-responsive tile, and assemble the Mg 2+ -ion-dependent DNAzyme in the cavities. For the programmed unlocking of the nanocavities in the two tiles of state I (100 μL, 30 nM) into state IV, the respective steps detailed for unlocking the nanocavities in the ATP-responsive tile and the K + -ionresponsive tile were applied. For the activation of the catalytic transformations in the different dimer configurations the origami mixtures were subjected to a mixture of the ROX/BHQ2-modified substrate S1 (100 μM, 1 μL) and the Cy5/BHQ2-modified substrate S2 (100 μM, 1 μL). The system were allowed to react for a time-interval of 6 hours at 25 °C. The fluorescence spectra of the resulting fluorophore-labeled fragments generated by the respective DNAzymes, associated with the different dimers were recorded (ROX, λex = 550 nm; Cy5, λex = 630 nm).
Programmed catalytic activities in the nanocavities within the pH-/K + -ionresponsive origami dimers D2. To induce the programmed transition of the E3a/E3b and E2a/E2b-functionalized pH-/K + -ion-responsive dimer D2 into the catalytically active dimer in state II, the dimer in state I (100 μL, 30 nM in a TAE buffer that contained Mg 2+ , 6 Mm, and Na + , 5 mM) was subjected to the hairpin strands H1/H2 (10 μL, 1.5 μM) at pH = 9.5, for a time interval of 10 hours, to unlock the nanocavities in the pHresponsive tile, and assemble the Mg 2+ -ion-dependent DNAzyme in the cavities. For the programmed opening of the K + -ion-responsive tile (marked) in state I into state III, the dimer in state I (100 μL, 30 nM) was subjected to K + ions (10 μL, 1.2 M) and the hairpin strands H1/H2 (10 μL, 1.5 μM), for a time interval of 10 hours, to unlock the nanocavities in the K + -ion-responsive tile, and assemble the Mg 2+ -ion-dependent DNAzyme in the cavities. For the programmed unlocking of the nanocavities in the two tiles of state I into state IV, the respective steps detailed for unlocking the nanocavities in the pH-responsive tile and the K + -ion-responsive tile were applied. For the activation of the catalytic transformations in the different dimer configurations the origami mixtures were subjected to a mixture of the FAM/BHQ1-modified substrate S3 (100 μM, 1 μL) and the Cy5/BHQ2-modified substrate S2 (100 μM, 1 μL). The system were allowed to react for a time-interval of 6 hours at 25 °C. The fluorescence spectra of the resulting fluorophore-labeled fragments generated by the respective DNAzymes, associated with the different dimers were recorded (FAM, λex = 495 nm; Cy5, λex = 630 nm).
AFM imaging. For the AFM measurements, 2 μL of the respective origami samples were deposited on the freshly peeled mica. After adsorbing for 5 min, the samples were imaged in an aqueous buffer solution under tapping mode using SNL-10 probes (Bruker, Multimode Nanoscope VIII). Figure S1. Schematic of the designed origami tile responsible to K + -ion-stabilized Gquadruplexes and crown ether. The lock strands L1/L1' and L2/L2' shows in green, the handles (Ha, Hb) are purple loops and the anchoring tethers (A1, A2) are short purple lines. The red square shows the designed "nano-cavity" patch in the origami tile.
The "nano-cavity" is designed in the origami tile through opening an inner patch and binding to the nearby part of the origami tile. The patch is linked to the origami raft through the M13 strand (as hinges for its opening process) on its right side and the locks (green) on its left side. The lock strands L1 and L2 contain the G-rich sequences and form the G-quadruplexes in the presence of K + ions. The top and bottom sides of the designed "nano-cavity" have Ha and Hb handles (purple). One end of the handle is linked to the staple strand of the origami tile and the other end of the handle is linked to the staple of the "nano-cavity". A1 and A2 anchoring footholds are extending from the staple strands on the right side of the origami tile. The handles can hybridize with the helper strands to stabilize the path (in the form of Ha/H1/A1 and Hb/H2/A2). Figure S2. Four AFM images of the initial locked origami tiles before the treatment with K + ions and the helper hairpins H1/H2. Scale bar: 200 nm. Table S1. Statistical analysis of the yields of the unlocked and locked origami tiles before the treatment with K + ions and the helper hairpins H1/H2.  Figure S3. Four AFM images of the unlocked origami tiles upon the treatment of the tiles with K + ions and the helper hairpins H1/H2 (first cycle). Scale bar: 200 nm. S10 Table S2. Statistical analysis of the yields of the unlocked and locked origami tiles in the unlocked state after the treatment with K + ions and the helper hairpins H1/H2.  Figure S4. Four AFM images of the regenerated and locked origami tiles upon the treatment of the G-quadruplex unlocked tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b') and crown ether (CE) (first cycle). Scale bar: 200 nm. S12 Table S3. Statistical analysis of the yields of the unlocked and locked origami tiles upon the treatment of the G-quadruplex unlocked tiles with the anti-helper strands (H1a'/H1b' and H2a'/H2b') and crown ether (CE).  Figure S5. Four AFM images of the G-quadruplex unlocked origami tiles upon the treatment of the tiles with K + ions and the helper hairpins H1/H2 (second cycle). Scale bar: 200 nm. S14 Table S4. Statistical analysis of the yields of the unlocked and locked origami tiles upon the treatment of the locked tiles with K + ions and the helper hairpins H1/H2 (second cycle).  Figure S6. Four AFM images of the regenerated and locked origami tiles upon the treatment of the G-quadruplex unlocked tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b') and CE (second cycle). Scale bar: 200 nm. S16 Table S5. Statistical analysis of the yields of the unlocked and locked origami tiles upon the treatment of the G-quadruplex unlocked tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b') and CE (second cycle).  Figure S7. AFM images of the control experiment testing the unlocking of the Gquadruplex-responsive origami tiles in the presence of the helper hairpins H1/H2 and in the absence of K + ions. No tiles with nanoholes are observed, implying that the K + ions are essential to unlock the origami tiles to yield the nano-cavities. Scale bar: 200 nm.  Figure S8. AFM images of the control experiment testing the K + -ion-driven unlocking of the G-quadruplex-responsive origami tiles in the absence of the helper hairpins H1/H2. Only few origami tiles are with nanoholes. The results imply that the unlocked "window" exists in a flexible configuration that retains the nanohole closed without H1/H2 that stretch the "window" to a rigid open configuration by the hairpins/handles/anchor site units. Scale bar: 200 nm. Table S7. Statistical analysis of the yields of the unlocked and locked origami tiles in the presence of K + ions and in the absence of the helper hairpins H1/H2.  Figure S9. Electrophoretic Gel image of the locked and unlocked K + -ion stabilized Gquadruplexes/crown ether-responsive origami tiles (lane 2 and 3, respectively). Lane 1 is 1kb reference ladder. Figure S10. Schematic of the designed origami tile responsible to ATP molecules. The lock strands M1/M1' and M2/M2' shows in yellow, the handles (Ha, Hb) are purple loops and the anchoring tethers (A1, A2) are short purple lines. The red square shows the designed "nano-cavity" patch in the origami tile. The lock strands M1 and M2 contain the ATP aptamer sequences and can form the aptamer/ATP complex in the presence of ATP.

Statistical analysis
S23 Figure S11. Four AFM images of the initial locked ATP-responsive origami tiles before the treatment of the tiles with ATP and the helper hairpins H1/H2. Scale bar: 200 nm.
S24 Table S8. Statistical analysis of the yields of the unlocked and locked ATP-responsive origami tiles before the treatment of the tiles with ATP and the helper hairpins H1/H2.  Figure S12. Four AFM images of the unlocked ATP-responsive origami tiles upon the treatment of the tiles with ATP and the helper hairpins H1/H2 (first cycle). Scale bar: 200 nm.

Statistical analysis
S26 Table S9. Statistical analysis of the yields of the unlocked and locked ATP-responsive origami tiles in the unlocked state after the treatment with ATP and the helper hairpins H1/H2 (first cycle).  Figure S13. Four AFM images of the regenerated and locked ATP-responsive origami tiles upon the treatment of the unlocked cavity-containing tiles with the anti-helper strands (H1a'/H1b' and H2a'/H2b'), C-ATPa and C-ATPa' (first cycle). Scale bar: 200 nm.

Statistical analysis
S28 Table S10. Statistical analysis of the yields of the unlocked and locked ATP-responsive origami tiles upon the treatment of the unlocked tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b'), C-ATPa and C-ATPa' (first cycle).  Figure S14. Four AFM images of the unlocked ATP-responsive origami tiles upon the treatment of the locked ATP-responsive tiles with ATP and the helper hairpins H1/H2 (second cycle). Scale bar: 200 nm.

Statistical analysis
S30 Table S11. Statistical analysis of the yields of the unlocked and locked ATP-responsive origami tiles upon the treatment of the locked ATP-responsive tiles with ATP and the helper hairpins H1/H2 (second cycle).  Figure S15. Four AFM images of the regenerated and locked ATP-responsive origami tiles upon the treatment of the unlocked cavity-containing ATP-responsive tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b'), C-ATPa and C-ATPa' (second cycle). Scale bar: 200 nm.

Statistical analysis
S32 Table S12. Statistical analysis of the yields of the unlocked and locked origami tiles upon the treatment of the unlocked cavity-containing ATP-responsive origami tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b'), C-ATPa and C-ATPa' (second cycle).  Figure S16. AFM images of the control experiment testing the unlocking of the ATPresponsive origami tiles in the presence of the helper hairpins H1/H2 and in the absence of ATP. No tiles with nanoholes are observed, implying that the ATP is essential to unlock the origami tiles to yield the nanoholes. Scale bar: 200 nm.

Statistical analysis
S34 Table S13. Statistical analysis of the yields of the unlocked and locked ATP-responsive origami tiles in the presence of the helper hairpins H1/H2 and in the absence of ATP.  Figure S17. AFM images of the control experiment testing the ATP-driven unlocking of the origami tiles in the absence of the helper hairpins H1/H2. Only few origami tiles are with nanoholes. The results imply that the unlocked "window" exists in a flexible configuration that retains the nanohole closed without H1/H2 that stretch the "window" to a rigid configuration by the hairpins/handles/anchor site units. Scale bar: 200 nm. Yield (%) 7.6 84.8 Figure S18. Schematic of the designed origami tile responsible to pH. The lock strands N1/N1' and N2/N2' shows in blue, the handles (Ha, Hb) are purple loops and the anchoring tethers (A1, A2) are short purple lines. The red square shows the designed "nano-cavity" patch in the origami tile. The lock strands can form the triplex structures at pH 6 and separate at pH 9.5.  Figure S20. Four AFM images of the initial locked pH-responsive origami tiles before the treatment of the tiles at pH = 9.5 with the helper hairpins H1/H2. Scale bar: 200 nm.

S36
S40 Table S15. Statistical analysis of the yields of the unlocked and locked pH-responsive origami tiles before the treatment of the tiles at pH = 9.5 with the helper hairpins H1/H2.  Figure S21. Four AFM images of the unlocked pH-responsive origami tiles upon the treatment of the tiles at pH = 9.5 with the helper hairpins H1/H2 (first cycle). Scale bar: 200 nm. Table S16. Statistical analysis of the yields of the unlocked and locked pH-responsive origami tiles in the unlocked state after the treatment at pH = 9.5 with the helper hairpins H1/H2 (first cycle).  Figure S22. Four AFM images of the regenerated and locked origami tiles upon the treatment of the unlocked pH-responsive tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b') at pH = 6 (first cycle). Scale bar: 200 nm. Table S17. Statistical analysis of the yields of the unlocked and locked origami tiles upon the treatment of the unlocked pH-responsive tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b') at pH = 6 (first cycle).  Figure S23. Four AFM images of the unlocked origami tiles upon the treatment of the pH-responsive tiles at pH = 9.5 with the helper hairpins H1/H2 (second cycle). Scale bar: 200 nm. Table S18. Statistical analysis of the yields of the unlocked and locked origami tiles upon the treatment of the pH-responsive tiles at pH = 9.5 with the helper hairpins H1/H2 (second cycle).  Figure S24. Four AFM images of the regenerated and locked pH-responsive origami tiles, upon the treatment of the unlocked tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b') at pH = 6 (second cycle). Scale bar: 200 nm.

Statistical analysis
S48 Table S19. Statistical analysis of the yields of the unlocked and locked origami tiles upon the treatment of the unlocked pH-responsive tiles with anti-helper strands (H1a'/H1b' and H2a'/H2b') at pH = 6 (second cycle).  Figure S25. AFM images of the control experiment testing the unlocking of the pHresponsive origami tiles in the presence of the helper hairpins H1/H2 at pH = 6. No tiles with nanoholes are observed, implying that the pH = 9.5 is essential to unlock the origami tiles to yield the nanoholes. Scale bar: 200 nm. Table S20. Statistical analysis of the yields of the unlocked and locked pH-responsive origami tiles in the presence of the helper hairpins H1/H2 at pH = 6.  Figure S26. AFM images of the control experiment testing the pH-driven unlocking of the pH-responsive origami tiles in the absence of the helper hairpins H1/H2. Only few origami tiles are with nanoholes. The results imply that the unlocked "window" exists in a flexible configuration that retains the nanohole closed without H1/H2 that stretch the "window" to a rigid configuration by the hairpins/handles/anchor site units. Scale bar: 200 nm. Table S21. Statistical analysis of the yields of the unlocked and locked pH-responsive origami tiles at pH = 9.5 and in the absence of the helper hairpins H1/H2.  Figure S27. Four AFM images of the initial locked ATP-/K + -ion-responsive origami dimers (state I) before the treatment with K + ions or ATP and the helper hairpins H1/H2. Scale bar: 200 nm. Table S22. Statistical analysis of the yields of the unlocked and locked ATP-/K + -ionresponsive origami dimers (state I) before the treatment with K + ions or ATP and the helper hairpins H1/H2.  Table S23. Statistical analysis of the yields of the unlocked and locked ATP-/K + -ionresponsive origami dimers (state II) upon the treatment with ATP and the helper hairpins H1/H2.  Table S24. Statistical analysis of the yields of the unlocked and locked ATP-/K + -ionresponsive origami dimers (state III) upon the treatment with K + ions and the helper hairpins H1/H2.  Table S25. Statistical analysis of the yields of the unlocked and locked ATP-/K + -ionresponsive origami dimers (state IV) upon the treatment with K + ions/ATP and the helper hairpins H1/H2.

Dimer Monomer
Incomplete   Figure S32. Four AFM images of the pH-/K + -ion-responsive origami dimers (state I) before the treatment with K + ions or pH 9.5 buffer and the helper hairpins H1/H2. Scale bar: 200 nm. Table S26. Statistical analysis of the yields of the unlocked and locked pH-/K + -ionresponsive origami dimers (state I) before the treatment with K + ions or pH 9.5 buffer and the helper hairpins H1/H2.  Figure S33. Four AFM images of the pH-/K + -ion-responsive origami dimers with unlocked nanocavity on the unmarked tile (state II) upon the treatment of the dimers at pH 9.5 buffer and the helper hairpins H1/H2. Scale bar: 200 nm. Table S27. Statistical analysis of the yields of the unlocked and locked pH-/K + -ionresponsive origami dimers (state II) upon the treatment in pH 9.5 buffer and the helper hairpins H1/H2.  Table S28. Statistical analysis of the yields of the unlocked and locked pH-/K + -ionresponsive origami dimers (state III) upon the treatment with K + ions and the helper hairpins H1/H2.  Figure S35. Four AFM images of the pH-/K + -ion-responsive origami dimers with unlocked nanocavities on both tiles (state IV) upon the treatment with K + ions and the helper hairpins H1/H2 at pH 9.5. Scale bar: 200 nm. Table S29. Statistical analysis of the yields of the unlocked and locked pH-/K + -ionresponsive origami dimers (state IV) upon the treatment with K + ions and the helper hairpins H1/H2 at pH 9.5.

Dimer Monomer
Incomplete  Figure S36. Schematic of engineering of the origami tile (only showing the core part of the tile) and the switchable catalysis of the Mg 2+ -ion-dependent DNAzyme in the nanocavity. The tile includes the ATP-driven unlocking apparatus (discussed in Fig. 3 and text). Protruding tethers T1/T3 and T2/T4 are designed on the opposite sides of the origami tile. The duplex E1a/B1a/B1b and the strand E1b are hybridized with the tethers T1/T3 and T2/T4, respectively. Prior to the unlocking of the tile, E1a/B1a/B1b are unblocked by the strand displacement process, using appropriate anti-blockers (B1a'/B1b'). The deblocked strand E1a and strand E1b correspond to the Mg 2+ -iondependent DNAzyme subunits. The ATP induced unlocking of the origami tile, in the presence of the helper hairpins H1 and H2 leads to the formation of the nanocavity. The strands E1a and E1b bind together and form the active Mg 2+ -ion-dependent DNAzyme that cleave the ROX/BHQ2-modified substrate S1 to produce the ROX-modified fragment in the confined nanocavity. Treatment of the catalytic system with the blockers B1a and B1b separates the Mg 2+ -ion-dependent DNAzyme subunits, and the subsequent treatment of the tiles with C-ATPa and C-ATPa', in the presence of the respective counter-helper units (H1a'/H1b'/H2a'/H2b'), leads to the closure of the nanocavity. The scheme represents the mechanistic path for the cyclic switching of the activity of the Mg 2+ -ion-dependent DNAzyme in the confined nanocavity, associated with the origami tile. The reversible switching of the fluorescence of ROX provides the readout signal for the "ON"/"OFF" switching of the catalytic functions of the system. Figure S37. Fluorescence spectra corresponding to the cyclic activation and deactivation of the Mg 2+ -ion-dependent DNAzyme in the ATP-responsive origami tiles. Panel I corresponds to the fluorescence response of the locked ATP-responsive origami tile. Panel II corresponds to the activated DNAzyme in the nanocavities as a result of the ATP, H1/H2-stimulated unlocking of the cavities. Panel III to Panel V represent the fluorescence spectra of the system upon the cyclic closure-opening and reclosure of the nanocavities, The closed ATP-responsive origami tiles are generated by treatment of the open ATP-responsive origami tiles with the counter strands H1a' and H1b', H2a' and H2b', C-ATPa and C-ATPa'. The re-opening of the ATP-responsive tiles involved the treatment of the closed state with ATP and the hairpins H1/H2. Figure S38. Schematic of engineering of the G-quadruplex-responsive origami tile (only showing the core part of the tile) and the switchable catalysis of the Mg 2+ -iondependent DNAzyme in the nanocavity. The tile includes the K + -ion-driven unlocking system. Protruding tethers T5/T7 and T6/T8 are designed on the opposite sides of the origami tile. The duplex E2a/B2a/B2b and the strand E2b are hybridized with the tethers T5/T7 and T6/T8, respectively. Prior to the unlocking of the tile, E2a/B2a/B2b are unblocked by the strand displacement process, using appropriate anti-blockers (B2a'/B2b'). The deblocked strand E2a and strand E2b correspond to the Mg 2+ -iondependent DNAzyme subunits. The K + ions induced unlocking of the origami tile, in the presence of the helper hairpins H1 and H2 leads to the formation of the nanocavity. The strands E2a and E2b assemble into formation of the active Mg 2+ -ion-dependent DNAzyme that cleave the Cy5/BHQ2-modified substrate S2 to produce the Cy5modified fragment in the confined nanocavity. Treatment of the catalytic system with the blockers B2a and B2b separates the Mg 2+ -ion-dependent DNAzyme subunits and the subsequent treatment of the tiles with crown ether (CE), in the presence of the respective counter-helper units (H1a'/H1b'/H2a'/H2b'), leads to the closure of the nanocavity. The scheme represents the mechanistic path for the cyclic switching of the activity of the Mg 2+ -ion-dependent DNAzyme in the confined nanocavity associated with the origami tile. The reversible switching of the fluorescence of Cy5 provides the readout signal for the "ON"/"OFF" switching of the catalytic functions of the system. Figure S39. Fluorescence spectra corresponding to the cyclic activation and deactivation of the Mg 2+ -ion-dependent DNAzyme in the G-quadruplex-responsive origami tiles. Panel I corresponds to the fluorescence response of the locked Gquadruplex-responsive origami tile. Panel II corresponds to the activated DNAzyme in the nanocavities as a result of the K + ions, H1/H2-stimulated unlocking of the cavities.

S76
Panel III to Panel V represent the fluorescence spectra of the system upon the cyclic closure-opening and reclosure of the nanocavities, The closed G-quadruplexresponsive origami tiles are generated by treatment of the open origami tiles with the counter strands H1a ' and H1b',H2a' and H2b',and crown ether (CE). The re-opening of the G-quadruplex-responsive tiles involved the treatment of the closed state with K + ions and the hairpins H1/H2. Figure S40. Schematic of engineering of the pH-responsive origami tile (only showing the core part of the tile) and the switchable catalysis of the Mg 2+ -ion-dependent DNAzyme in the nanocavity. The tile includes the pH-driven unlocking system. Protruding tethers T9/T11 and T10/T12 are designed on the opposite sides of the origami tile. The duplex E3a/B3a/B3b and the strand E3b are hybridized with the tethers T9/T11 and T10/T12, respectively. Prior to the unlocking of the tile, E3a/B3a/B3b are unblocked by the strand displacement process, using appropriate anti-blockers (B3a'/B3b'). The deblocked strands E3a and E3b correspond to the Mg 2+ -ion-dependent DNAzyme subunits. The pHinduced unlocking of the origami tiles, in the presence of the helper hairpins H1 and H2, leads to the formation of the nanocavities. The strands E3a and E3b assemble and form the active Mg 2+ -ion-dependent DNAzyme that cleaves the FAM/BHQ1-modified substrate S3 to produce the FAM-modified fragment in the confined nanocavity. Treatment of the catalytic system with the blockers B3a and B3b separates the Mg 2+ -iondependent DNAzyme subunits, and the subsequent treatment of the tiles at pH = 6, and in the presence of the respective counter-helper units (H1a'/H1b'/H2a'/H2b'), results in the closure of the nanocavities. The scheme represents the mechanistic path for the cyclic switching of the activity of the Mg 2+ -ion-dependent DNAzyme in the confined nanocavity associated with the origami tile. The reversible switching of the fluorescence of FAM provides the readout signal for the "ON"/"OFF" switching of the catalytic functions of the system. S80 Figure S41. Fluorescence spectra corresponding to the cyclic activation and deactivation of the Mg 2+ -ion-dependent DNAzyme in the pH-responsive origami tiles. Panel I corresponds to the fluorescence response of the locked pH-responsive origami tile. Panel II corresponds to the activated DNAzyme in the nanocavities as a result of the pH = 9.5, H1/H2-stimulated unlocking of the cavities. Panel III to Panel V represent the fluorescence spectra of the system upon the cyclic closure-opening and reclosure of the nanocavities, The closed pH-responsive origami tiles are generated by treatment of the open origami tiles with the counter strands H1a' and H1b', H2a' and H2b', at pH = 6. The re-opening of the pH-responsive tiles involved the treatment of the closed state with K + ions and the hairpins H1/H2. Figure S42. Fluorescence spectra corresponding to the programmed activation of the two Mg 2+ -ion-dependent DNAzymes in the confined nanocavities in the ATP-/K + -ionresponsive origami dimers. The programmed activation of the catalytic functions of the two Mg 2+ -ion-dependent DNAzymes were performed by unlocking the nanocavities in the ATP-/K + -ion-responsive origami dimer D1 by treatment with the K + ions and/or ATP and the help hairpins H1/H2. Panel I -Fluorescence spectra of ROX (left) and Cy5 (right) generated by the locked origami dimers (shown in panel I of Fig. 8). Panel II -Fluorescence spectra of ROX (left) and Cy5 (right) generated upon selective unlocking of the dimer on the left with ATP (cf. panel II in Fig. 8). The fluorescence spectrum of ROX shows the enhanced intensity over the background signal generated in the confined cavity. Panel III -Fluorescence spectra of ROX (left) and Cy5 (right) upon subjecting the dimer to K + ions (cf. process shown in panel III in Fig. 8). No fluorescence change of ROX above the background signal is observed, while the fluorescence of Cy5 is enhanced as compared to the background fluorescence signal. Panel IV -Fluorescence spectra of ROX (left) and Cy5 (right) generated upon subjecting the dimer to the K + ions and ATP (cf. process shown in panel IV of Fig. 8).

S82
The fluorescence of ROX and the fluorescence of Cy5 are intensified as compared to the background signals, consistent with the activation of the two DNAzymes in the nanocavities. Figure S43. Fluorescence spectra corresponding to the programmed activation of the two Mg 2+ -ion-dependent DNAzymes in the confined nanocavities of the pH-/K + -ionresponsive origami dimers. The programmed activation of the catalytic functions of the two Mg 2+ -ion-dependent DNAzymes were performed by unlocking the nanocavities in the pH-/K + -ion-responsive origami dimer D2 by treatment with the K + ions and/or pH = 9.5 buffer and the help hairpins H1/H2. Panel I -Fluorescence spectra of FAM (left) and Cy5 (right) generated by the locked origami dimers D2 (shown in panel I of Fig. 9). Panel II -Fluorescence spectra of FAM (left) and Cy5 (right) generated upon selective unlocking of the dimer D2 on the left at pH = 9.5 (cf. panel II in Fig. 9). The fluorescence spectrum of FAM shows the enhanced intensity over the background signal generated in the confined cavity. Panel III -Fluorescence spectra of FAM (left) and Cy5 (right) upon subjecting the dimer D2 to K + ions (cf. process shown in panel III in Fig. 9). No fluorescence change of FAM above the background signal is observed, while the fluorescence of Cy5 is enhanced as compared to the background fluorescence signal. Panel IV -Fluorescence spectra of FAM (left) and Cy5 (right) generated upon subjecting the dimer D2 to the K + ions at pH = 9.5 (cf. process shown in panel IV of Fig. 9). The fluorescence of FAM and the fluorescence of Cy5 are intensified as compared to the background signals, consistent with the activation of the two DNAzymes in the nanocavities.