Size-selective molecular recognition based on a confined DNA molecular sieve using cavity-tunable framework nucleic acids

Abstract

Size selectivity is an important mechanism for molecular recognition based on the size difference between targets and non-targets. However, rational design of an artificial size-selective molecular recognition system for biological targets in living cells remains challenging. Herein, we construct a DNA molecular sieve for size-selective molecular recognition to improve the biosensing selectivity in living cells. The system consists of functional nucleic acid probes (e.g., DNAzymes, aptamers and molecular beacons) encapsulated into the inner cavity of framework nucleic acid. Thus, small target molecules are able to enter the cavity for efficient molecular recognition, while large molecules are prohibited. The system not only effectively protect probes from nuclease degradation and nonspecific proteins binding, but also successfully realize size-selective discrimination between mature microRNA and precursor microRNA in living cells. Therefore, the DNA molecular sieve provides a simple, general, efficient and controllable approach for size-selective molecular recognition in biomedical studies and clinical diagnoses.

Fig. 1. Scheme illustration of DNA molecular sieve.

a Size-selective catalysis based on a traditional molecular sieve. b Size-selective molecular recognition in the confined space of DNA molecular sieve using cavity-tunable DNA nanocage framework nucleic acids. Large molecules, such as nucleases, proteins and larger analogues, are prohibited from contacting with inner recognition sites, while small target molecules are able to enter the cavity for efficient molecular recognition.

Fig. 2. Scheme illustration of CONFINED-DNAzyme.

The design of DNAzyme encapsulated in a cavity-tunable framework nucleic acid.

Fig. 3. Design and characterization of DNA nanocages with different cavity sizes.

a Schematic illustration of DNA nanocages construction. The cavity of DNA nanocages gradually increases from cage 1 to cage 5. b N-PAGE characterization of DNA nanocages from cage 1 (lane 1) to cage 5 (lane 5). c Atomic force microscopy characterization of DNA nanocages (Cage 5). The scale bars are 100 and 30 nm in large and small imaging, respectively. Source data are provided as a Source Data file.

Fig. 4. The size-dependence protective performance of DNAzyme encapsulated in the DNA nanocages.

a Schematic illustration of three kinds of DNAzyme probes: free DNAzyme, DNAzyme connected to the outside frame of Cage 2 (Czyme-out-2), and DNAzyme encapsulated inside Cage 2 (Czyme-in-2). b Time-dependent fluorescence changes of different probes after the addition of nuclease (F₀ fluorescence intensity without DNase I, F fluorescence intensity with DNase I). c Fluorescence response of free DNAzyme, Czyme-out-2 and Czyme-in-2 to l-histidine treated with and without SSB (F₀ fluorescence intensity without l-histidine, F fluorescence intensity with l-histidine); Data are presented as mean values ± s.d. (n = 3); ***p = 0.00070 < 0.001, **p = 0.0046 < 0.01, ns = 0.6800 > 0.05 (not significant), by two-tailed unpaired Student’s t-test. d Schematic illustration of the cavity effect; e Time-dependent fluorescence changes of different cage sizes after the addition of nuclease (F₀ fluorescence intensity without DNase I, F fluorescence intensity with DNase I). f Relative response intensity of different cage sizes to l-histidine after treatment with single-strand binding protein (SSB) (“Relative response intensity” refer to the ratio of (F/F₀) with and without SSB; F: fluorescence intensity with l-histidine, F₀ fluorescence intensity without l-histidine); Data are presented as mean values ± s.d. (n = 3); **p = 0.0014 (0.0040) for Czyme-in-4 (Czyme-in-5) < 0.01, ns = 0.43 (0.10) for Czyme-in-1 (Czyme-in-3) > 0.05 (not significant), by two-tailed unpaired Student’s t-test. Source data are provided as a Source Data file.

Fig. 5. The biosensing application of CONFINED-based biosensor.

a Fluorescence confocal imaging of HeLa cells incubated with Czyme-in-2 without and with l-histidine, respectively. The scale bars are 20 μm. b The relative fluorescence intensity of Czyme-in-2 confocal imaging that quantified by ImageJ; Data are presented as mean values ± s.d. (n = 3), Student’s t test, ***p = 0.00043 < 0.001, by two-tailed unpaired Student’s t-test. c Schematic illustration of ATP aptamer encapsulated in the DNA cages (Cage-apt-in-2). d Fluorescence response in the presence of different concentrations of ATP, ranging from 0 to 10 mM. Inset: relationship between fluorescence enhancement and concentrations, data are presented as mean values ± s.d. (n = 3). Source data are provided as a Source Data file.

Fig. 6. Size-selective sensing of mature miRNAs.

a Schematic illustration of size-selective molecular recognition to distinguish mature microRNA from precursor microRNA. b Schematic illustration of different lengths of nucleic acid targets containing the same recognition sequence of miRNA-21. c, d Fluorescence response of free MB (c) and Cage-MB-in (d) to the nucleic acid targets with different lengths (F: fluorescence intensity in the presence of target, F₀: fluorescence intensity in absence of target) ; Data are presented as mean values ± s.d. (n = 4), ns = 0.17 (0.30, 0.37) for T10-miRNA 21- T10 (T20-miRNA 21- T20, T30-miRNA 21- T30) to Free MB > 0.05 (not significant), ***p = 0.00089 (0.00021, 0.000068) for T10-miRNA 21-T10 (T20-miRNA 21-T10, T30-miRNA 21-T30) to Cage-MB-in <0.001, by two-tailed unpaired Student’s t-test. e–f Fluorescence study of the fluorescence response of Cage-MB-out (e) and Cage-MB-in (f) to pre-miRNA-21 and mature miRNA-21. Source data are provided as a Source Data file.

Fig. 7. Size-selective discrimination between mature microRNA and precursor microRNA in living cells.

a, b Fluorescence confocal imaging of MCF-7, HeLa and HEK293 cells incubated with 100 nM Cage-MB-in and Cage-MB-out (a) and the normalized mean fluorescence by ImageJ (b); Data are presented as mean values ± s.d. (n = 5), ***p = 0.00000068 (0.000000036), 0.0000014 for HeLa (HEK293) to Cage-MB-in, HEK293 to Cage-MB-out < 0.001, ns = 0.93 > 0.05 (not significant), by two-tailed unpaired Student’s t-test. c Fluorescence confocal imaging of HeLa cells treated with and without 5 μM miRNA-21 inhibitor PLL before incubating with 100 nM Cage-in. The scale bars are 20 μm. Source data are provided as Source Data file.