A fluorescence biosensor based on a cell-free transcription system for species-specific DNA sequence detection and seahorse product identification

Abstract

Seahorse products face rampant counterfeiting due to their high value and low production. The determination of species-specific DNA sequences is a key factor in identifying species. A conserved genetic fragment in the seahorse genome has been identified as a specific marker. In this study, a novel cell-free transcription system-based fluorescence biosensor has been designed and applied for seahorse product authenticity identification. A toehold-mediated triplex DNA complex has been designed. Within this complex, a single-stranded DNA (ssDNA) component is designed to hybridize with two other incomplete template strands, forming a complete aptamer transcription template. The target seahorse DNA, which exhibits greater complementarity with the other two strands in the triplex complex, can displace the ssDNA upon its presence and trigger a strand displacement reaction. The released ssDNA can be removed using magnetic beads easily since it is modified with biotin. The absence of ssDNA in the system prevented the formation of the complete transcription template, which inhibited the recognition by T7 RNA polymerase and reduced subsequent RNA transcription through the cell-free transtcription system. The resultant reduction in RNA production diminishes its binding to fluorescent aptamers, leading to a decrease in the detected fluorescence signal. Under optimized assay conditions, the fluorescence signal difference exhibited a linear correlation with the logarithm of seahorse DNA concentration over the range of 150 pM to 35 nM, achieving a limit of detection (LOD) of 65 pM. The proposed system has good reproducibility and selectivity, and has been successfully applied to seahorse product identification.

---

Introduction

Seahorses hold a significant position in traditional Chinese medicine due to their proven therapeutic effects. Due to their slow growth, low reproductive rate, and strict habitat requirements,^1–3 combined with the immaturity of artificial breeding techniques, the production of seahorses is relatively low. However, since market demand is strong, many counterfeit seahorse products have appeared in the market.⁴ Dried seahorse specimens (e.g., medicinal materials or crafts) are difficult to identify using morphological analysis alone. The determination of species-specific DNA sequences is a key factor in identifying species, as demonstrated by the use of internal transcribed spacer (ITS) sequences to authenticate valuable medicinal plants like the orchid Anoectochilus setaceus to ensure product authenticity,⁵ the application of entire chloroplast genome sequences to resolve phylogenetic relationships and accurately classify plant species,⁶ and the use of DNA barcoding to identify illegally traded bushmeat for combating wildlife crime, which involves endangered species such as pangolins and primates.⁷

Traditional detection techniques, including DNA barcoding,⁸ quantitative PCR (qPCR),⁹ and DNA sequencing,¹⁰ can be applied for species-specific DNA sequence detection. But most of these methods are time-consuming, rely on expensive instrumentation, and involve complex operational procedures, which limit their applicability in rapid on-site detection. In recent years, the emergence of cell-free expression systems has provided new insights for the development of biological detection technologies.^11–14 By integrating cell lysates with target gene circuits,^15,16 this system effectively bypasses the hindrance of cell membrane barriers to molecular detection, offering unique advantages such as modular design, rapid response, and in vitro operation.^17–19 Additionally, they can be coupled with different detection techniques such as fluorescence, colorimetric and electrochemical analysis to construct simple and efficient systems for rapid target analysis.^20–22 In particular, the fluorescence-based strategy, which leverages the high-affinity binding between in vitro transcribed RNA and small-molecule fluorogenic dyes, enables rapid and highly sensitive detection of a wide range of targets. For example, the Broccoli RNA aptamer binds to DFHBI-1T, emitting a strong fluorescence signal under UV light for the visual detection of Hg²⁺ and Pb²⁺ in water samples.²³ Similarly, transcription of the Mango RNA aptamer and its subsequent complex formation with thiazole orange (TO-1) provides a sensitive fluorescence readout in antibody-sensing gene circuits.²⁰ But to the best of our knowledge, this technique has not been applied in species identification.

In this study, a novel cell-free transcription-based fluorescence biosensor has been developed for seahorse product identification. To ensure species specificity, a conserved sequence region, serving as the target gene, was identified by aligning sequences from 18 known seahorse species obtained from the NCBI database. To achieve specific recognition of this target DNA, the biosensor employs a toehold-mediated strand displacement reaction (TMSDR), a strategy renowned for its high design flexibility, short recognition regions, and fast reaction kinetics.^24–26 The specific recognition of target analytes represents a crucial step in such cell-free biosensing systems.²⁷ However, a major challenge in applying the TMSDR is signal leakage caused by spontaneous nucleotide dissociation.²⁸ To mitigate leakage, DNA strands are typically designed with strategies such as terminating helices with C–G bonds,²⁹ introducing mismatches at the fraying regions,³⁰ or adding “clamps” at the end of the helices,^31,32 but these strategies complicate the design process. To address this issue, we innovatively combined magnetic separation technology with strand displacement reactions. By leveraging the high-efficiency separation capability of magnetic beads, this approach resolves the signal leakage problem while eliminating the need for complex DNA strand designs. The core detection mechanism is based on a cell-free transcription system, where the transcription template, once assembled, can be transcribed into RNA Mango aptamers that bind to their corresponding fluorogenic dyes, producing a measurable fluorescence signal. This strategy combines the flexible design capability of the TMSDR for target recognition, the powerful signal amplification of in vitro transcription, and the efficient separation and enrichment function of magnetic beads, overcoming the reliance of traditional methods on intact samples and high-precision equipment, providing a practical and efficient solution for the rapid identification of seahorse DNA.

Experimental section

Materials and reagents

The sequences of the oligonucleotides implemented in this research are enumerated below:

Blocker: 5′-CGCCA TTGTC CAATG AAT-3′

Scaffold: 5′-ATTCA TTGGA CAATG GCGAA AAAAC CAGCT TTGGG AGCCG GCGGT GGGAG TTC-3′

Target seahorse DNA: 5′-GAACT CCCAC CGCCG GCTCC CAAAG CTGGT-3′

Single-base mismatch DNA:

5′-GAACTCTCACCGCCGGCTCCCAAAGCTGGT-3′

Double-base mismatch DNA:

5′-GAACTCTCACCGCCGGCGCCCAAAGCTGGT-3′

Incomplete T7 promoter: 5′-TATTA GTATA TTTGT CTGCT TTTTT TTTGC AGACT TGTAT TCCTC ATTTT TGCTC CCAAA GCTGG TTTTT TTTTTT-biotin-3′

Incomplete template (1): 5′-TGAGG AATAC ATATA CTAAT ACGAC TCA CT ATAGG GCCGC CGGTA CCTCC GAAGG GACGG TGCGG AGAGG A GAGG GGGCA CTGGGC-3′

Incomplete template (2): 5′-GCCCA GTGCC CCCTC TCCTC TCCGC ACCGT CCCTT CGGAG GTACC GGCGG CCCTA TAGTG AGTCG-3′

HPLC purified oligonucleotides were purchased from Sangon Inc. (Shanghai, China). T7 RNA polymerase (50 U μL⁻¹), RNase inhibitor (40 U μL⁻¹), NTP mixture (25 mM each, nuclease-free), 1 M Tris-HCl solution (pH 7.5), and DEPC-treated water were obtained from the same supplier. The TO1-3PEG-Biotin fluorophore was purchased from abmgood (Canada). Streptavidin magnetic beads were purchased from Promega (USA). MgCl₂ and KCl were purchased from Sinopharm Chemical Reagent Co. Ltd. The tissue/cell DNA extraction kit was purchased from Hlingene Biotechnology Co., Ltd (Shanghai, China). RNase-free water was employed as the solvent throughout the experiment. All chemical reagents were of analytical grade.

Cell-free transcription procedure

First, equimolar ratios (2 μM) of the blocker, scaffold, and 3′-biotin-modified single-stranded DNA with a partial T7 promoter domain structure were incubated at 37 °C for 1 hour to allow complete reaction, forming a blocker/partial T7 promoter/scaffold ternary complex. Subsequently, varying concentrations of target seahorse DNA were then introduced to initiate toehold-mediated strand displacement. Following complete reaction, magnetic beads were added to specifically capture and remove the free single-stranded DNA (ssDNA) containing partial T7 promoter domains from the solution through streptavidin–biotin interaction.

Next, the magnetic beads, which had captured the partial T7 promoter strands, were separated and removed from the solution using a magnetic stand, and were thus excluded from subsequent reactions. Following magnetic separation, approximately 100 µL of supernatant was obtained, from which 2.5 µL was transferred to a 30 µL amplification reaction system. This amplification system contained a fluorescent aptamer template (130 nM), NTPs (10 mM), T7 RNA polymerase (10 U μL⁻¹), RNase inhibitor (10 U μL⁻¹), and 4 μL of 10× T7 RNA polymerase buffer. The reaction mixture was then incubated at 37 °C for 3 hours to facilitate transcriptional amplification, generating a large quantity of fluorescent RNA Mango aptamers.

Fluorescence detection procedure

Following the transcription step, 105 μL of buffer solution (10 mM Tris-HCl, 75 mM KCl, and 1 mM MgCl₂) and 7 μL of fluorescent dye (200 nM TO-1 for the Mango aptamer) were added to the reaction mixture. After gentle mixing, the solution was incubated under appropriate conditions for 15 min. Then the reaction products were directly subjected to fluorescence analysis.

Spectral experiments were carried out on a Hitachi F-4500 fluorescence spectrophotometer (Hitachi, Japan). The excitation wavelength for the Mango aptamer was set at 485 nm with a slit width of 10 nm, and emission signals were collected within the wavelength range of 510–610 nm.

Source of marine species samples and DNA extraction steps

The seahorse samples were donated by the Zhao'an Workstation of Zhangzhou Maritime Police Bureau and used as scientific research materials. The samples of farmed seahorses were sourced from the seahorse farm of Zhangzhou Lanying Technology Co., Ltd. Homogenates of different seahorse samples were prepared with a low-temperature tissue homogenizer, and subsequently, the DNA of seahorse tissues was rapidly extracted with a DNA extraction kit.

The remaining marine species samples (including sea bass, grouper, sea eel and grey bream) were purchased from the local market. The DNA extraction methods for different types of marine species samples are similar to those for seahorses.

Results and discussion

Principle of the cell-free system based fluorescence biosensor for seahorse product identification

Scheme 1 illustrates the principle of the fluorescent biosensor based on the cell-free RNA transcription system for seahorse product identification. A 3′-biotin-modified single-stranded DNA (ssDNA) containing a partial T7 promoter domain is synthesized, along with two partially complementary ssDNA strands (referred to as blocker and scaffold strands), forming a toehold-mediated three-strand complex. The partial T7 promoter ssDNA from this complex serves as a key component that, together with two additional incomplete DNA templates, assembles into a complete transcription template for the Mango aptamer. Under the action of T7 RNA polymerase, this template is transcribed, producing a large amount of the fluorescent RNA Mango aptamer. The aptamers specifically bind to the RNA dye TO1-Biotin, generating a strong fluorescence signal. When the target DNA is introduced, its binding affinity to the scaffold strand is stronger than that of the partial T7 promoter strand, causing the displacement and release of the partial T7 promoter strand. The displaced partial T7 promoter strand, modified with biotin, is subsequently removed from the system using streptavidin-coated magnetic beads. The remaining complex (a three-stranded structure of the blocker, scaffold, and target) cannot form a complete T7 promoter domain structure with the other two incomplete DNA templates, thereby inhibiting the transcription of the Mango aptamer. Consequently, a reduction in the fluorescence signal is observed. The difference in fluorescence signals is then used for subsequent quantitative analysis.

Scheme 1 Schematic of the cell-free system based fluorescence biosensor for seahorse DNA detection.

Feasibility of the proposed biosensor

The feasibility of the sensor pathway was verified via PAGE gel electrophoresis first. As shown in Fig. 1(A), the bands in lanes 1, 2, and 3 correspond to the scaffold strand, blocker strand, and incomplete T7 promoter strand, respectively. Lane 4 shows complex 1 formed by the hybridization of three DNA strands. Lane 6 displays complex 2, which consists of the scaffold strand, blocker strand, and target strand. Lane 5 displays the reaction product of complex 1 with target DNA, exhibiting two distinct bands that align with the positions of bands in lanes 3 and 6. This indicates that the product consists of complex 2 and the free T7 promoter strand, confirming that the target displaces the incomplete T7 promoter strand when present in the system.

Fig. 1 Feasibility of the cell-free RNA transcription system for seahorse DNA detection. (A and B) Polyacrylamide gel electrophoresis for the characterization of the ssDNA isolation process. (C) Fluorescence spectra in the presence and absence of seahorse DNA.

In Fig. 1(B), the band in lane 4 corresponds to complex 1, while the band in lane 5 shows the product of complex 1 after magnetic bead treatment. The band in lane 6 represents complex 2, and the band in lane 7 displays the product after the reaction between the target and complex 1, followed by magnetic bead treatment. The results in Fig. 1(B) demonstrate that lanes 4 and 5 exhibit bands at the same position, indicating that streptavidin-coated magnetic beads do not interfere with the structure of complex 1. Similarly, lanes 6 and 7 show bands at identical positions, confirming that the displaced T7 promoter strand is removed by the magnetic beads, leaving only complex 2 in the system.

Fluorescence analysis further validated the feasibility of the cell-free system based biosensor. The RNA dye used in the fluorescence experiments was thiazole orange-3PEG-biotin (TO-1), which produces strong fluorescence only upon binding to a specific RNA aptamer (i.e., Mango). In Fig. 1(C), the black curve represents the fluorescence signal in the presence of the target (experimental group), while the red curve corresponds to the signal in the absence of the target (control group). Compared to the red curve, the black curve exhibits significantly lower fluorescence intensity, as the presence of the target suppresses RNA transcription in the cell-free system, reducing the amount of RNA available for TO-1 binding. This result further confirms that the target regulates the transcriptional output of the sensing system, verifying the sensor's response mechanism.

Optimization of the experimental conditions

Several key parameters which affected the performance of the system were optimized to maximize the detection sensitivity. The target concentration was set at 35 nM for the optimization experiments, as this level falls within the dynamic range and is suitable for evaluating system performance under near-saturation yet analytically relevant conditions. First, the influence of the probe ratio on biosensor performance was investigated. As shown in Fig. 2(A), as the ratio of the incomplete T7 promoter strand to the scaffold strand and blocker strand increased, ΔI (ΔI is defined as the difference between the fluorescence signal measured in the blank control group and that measured in the experimental group.) gradually rose until stabilizing at a ratio of 1, indicating that the three-strand ratio had reached its optimum. Therefore, in subsequent experiments, the ratio of the incomplete T7 promoter strand to the scaffold and blocker strands was set to 1.

Fig. 2 Influence of the (A) ratio of the T7 promoter to the scaffold, (B) reaction time of the target, (C) concentration of streptavidin magnetic beads, (D) volume of the DNA template, (E) reaction time of cell-free transcription.

Next, the effect of reaction time between the target and complex 1 was studied. Fig. 2(B) shows that ΔI increased progressively with extension of the reaction time before eventually plateauing after 75 min. Thus, a reaction time of 75 min was selected for subsequent experiments. Fig. 2(C) illustrates the impact of streptavidin-coated magnetic bead concentration. As the bead concentration increased, ΔI gradually rose and then stabilized, leading to an optimal bead concentration of 2 mg mL⁻¹. Additionally, the amount of DNA template and the duration of the enzymatic reaction were critical factors influencing sensor performance. As shown in Fig. 2(D), ΔI increased significantly with higher DNA template amounts before stabilizing at 0.4 μL. Therefore, the optimal DNA template volume was determined to be 0.4 μL. Fig. 2(E) demonstrates the dynamic changes in ΔI over time after the introduction of T7 RNA polymerase. ΔI increased sharply in the early stages of the reaction before stabilizing after 3.5 h, indicating no further substantial enhancement. Thus, the optimal enzymatic reaction time was determined to be 3.5 h.

Performance of the proposed biosensor

Under optimized experimental conditions, the constructed biosensor was used to detect target DNA at varying concentrations. As shown in Fig. 3(A), the fluorescence signal exhibited a positive correlation with target concentrations ranging from 175 pM to 35 nM. The calibration curve for analyte quantification is illustrated in Fig. 3(B). The linear regression equation was determined as:

ΔI = −169.34 + 103.00 lg C_target (pM), R² = 0.999

Fig. 3 (A) ΔI responses for seahorse DNA at varying concentrations: (a) 175 pM, (b) 350 pM, (c) 787 pM, (d) 1.75 nM, (e) 3.50 nM, (f) 7.87 nM, (g) 17.5 nM, and (h) 35.0 nM. (B) Linear correlation between the fluorescence signal and seahorse DNA concentration, with error bars indicating the standard deviation of three replicate measurements. (C) Reproducibility of the proposed method. Target concentration: 1.75 nM and 35 nM. (D) Fluorescence signals detected in the presence of seahorse DNA (35 nM) and various control oligonucleotides/species (350 nM).

Here, C_target represents the analyte concentration and ΔI denotes the fluorescence signal difference, which is calculated as the fluorescence intensity in the absence of the target minus that in the presence of the target. And R² is the coefficient of determination reflecting the fit of the linear model. The calculated limit of detection (LOD) was 65 pM (S/N = 3).

To evaluate the sensor's reproducibility, five replicate tests were conducted for two different concentrations under identical conditions. As illustrated in Fig. 3(C), the relative standard deviations (RSD) were 2.25% and 1.81% (n = 5), demonstrating excellent repeatability.

Subsequently, the selectivity of the biosensor was further evaluated. We tested the sensor system with non-target sequences containing single-base mismatch DNA and double-base mismatch DNA. The target seahorse DNA (target), single-base mismatch sequence DNA, and double-base mismatch DNA sequence were, respectively, incubated with the sensor system, and the response signals were compared by measuring the resulting fluorescence intensity. As shown in Fig. 3(D), only the fully complementary target DNA sequence triggered a significant change in fluorescence signal. In contrast, both the single-base and double-base mismatch DNA sequences produced negligible fluorescence changes. This clear differentiation underscores the sensor's high selectivity and its ability to discriminate even single-base variations, highlighting its robustness against non-specific activation.

Seahorses, seamoths (Pegasus laternarius), and pipefishes (Solenognathus spp.) share high morphological similarity within the suborder Syngnathoidei, particularly as dried specimens. Given the scarcity and high market value of seahorses, the more abundant but less expensive seamoths and pipefishes are intentionally mislabeled as seahorses in trade. To evaluate the specificity of the biosensor, we tested it against DNA from potential adulterants: Syngnathus acus, Solenognathus hardwickii, Syngnathoides biaculeatus, and Pegasus laternarius. As shown in Fig. 3(D), the fluorescence signal increased significantly only in the presence of seahorse DNA, whereas negligible signals were observed with interfering species. This confirms the high specificity for seahorse DNA of the proposed system, enabling it to accurately distinguish seahorses from other counterfeiting species.

The practical application of the proposed biosensor

To further evaluate its analytical capability for species identification, short DNA fragments extracted from different parts of seahorse tissue (targets 1 and 2) and four other species (sea base, grouper, sea eel and grey bream) were tested. As shown in Fig. 4, the concentration of the seahorse (target) specific DNA sequence was 20 nM, while those of other species were approximately 200 nM. A significant increase in fluorescence signal was observed in the presence of seahorse samples, whereas only minimal signals were detected for the other species. The relative standard deviation (RSD) of real sample measurements ranged from 2.9% to 5.7%, which is considered acceptable according to the “Guideline for Validation of Chemical Analysis Methods” (GB/T 27417-2017). In conclusion, these results not only confirm the accuracy of our method but also demonstrate its broad application potential in the field of species identification.

Fig. 4 Fluorescence response from different marine species.

Conclusion

This research developed a novel fluorescent biosensor based on a cell-free expression system. The sensor incorporates conserved gene fragments derived from seahorse genomes and employs the TMSDR for highly specific target DNA recognition. By integrating magnetic beads for efficient separation, the system significantly reduces signal leakage, thereby improving detection accuracy and sensitivity. This biosensor enables rapid target DNA response and fluorescence signal generation without requiring sophisticated instrumentation or intact samples, allowing for precise identification of seahorse-derived products.

Most importantly, the core of this sensing strategy is its programmable and modular design. The specificity is determined by the sequence of the scaffold strand within the ternary complex, which can be rationally redesigned to recognize DNA targets from virtually any species. This inherent versatility suggests that the presented biosensor platform holds significant potential far beyond the authentication of seahorse products. Future work will focus on adapting this system for the detection of other endangered and trafficked species, enabling its direct application in biodiversity conservation and wildlife forensic investigations. This provides a simple and powerful tool for market regulation, which in turn positions it as a versatile platform with significant potential for species identification and conservation biology.

Author contributions

Linyue Tang: conceptualization, methodology, validation, formal analysis, writing – original draft. Qian Xie: validation, data curation. Ming Chen: supervision, visualization. Cuiying Lin: formal analysis, writing – review & editing. Fang Luo: supervision. Zhongqin Li: resources, visualization. Zhenyu Lin: methodology, writing – review & editing, supervision, funding acquisition.

Conflicts of interest

The authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.

Data availability

The data are available from the corresponding author on reasonable request.

Acknowledgements

This project was financially supported by the National Natural Science Foundation of China (22574024) and the Fujian Province Science and Technology Project (2021N0014).

References

1. L. Claassens, Estuar. Coast. Shelf Sci., 2016, 180, 1–10.
2. M. Correia, H. J. Koldewey, J. P. Andrade, E. Esteves and J. Palma, Environ. Biol. Fishes, 2018, 101, 1357–1367.
3. N. C. Perante, M. G. Pajaro, J. J. Meeuwig and A. C. J. Vincent, J. Fish Biol., 2002, 60, 821–837.
4. R. M. Y. Ng, A. C. O. Lim, C. N. B. Lee, N. B. A. Majib, M. A. S. Hussein and A. Y.-H. Then, Fish. Res., 2025, 281, 107188.
5. L. D. Chaca and B. B. Thinhb, Biol. Bull., 2023, 50, 1143–1156.
6. J. Dobrogojski, M. Adamiec and R. Luciński, Acta Physiol. Plant., 2020, 42, 98.
7. A. D. Dipita, A. D. Missoup, M. Tindo and P. Gaubert, Biol. Conserv., 2022, 269, 109552.
8. J. T. Boehm, E. Bovee, S. E. Harris, K. Eddins, I. Akahoho, M. Foster, S. K. Pell, M. J. Hickerson, G. Amato, R. DeSalle and J. Waldman, PLoS One, 2023, 18, e0291874.
9. A. Marín, R. Alfaro, C. Villegas-Llerena, L. E. Reyes-Flores, I. B. Alvarez-Jaque, C. Roble, C. Ingar, C. G. Yzasiga-Barrera, L. L. Calado and E. Zelada-Mazmela, J. Nat. Conserv., 2021, 62, 126030.
10. L. Zeng, A. Armani, J. Wen, H. Lin, Y. Xu, S. Fan, Y. Sun, C. Yang, Z. Chen, D. Chen, J. Zhao and X. Li, Food Control, 2019, 103, 175–181.
11. D. Kim, J. Kim, J. Han, J. Shin and K. S. Park, Biosens. Bioelectron., 2024, 261, 116517.
12. F. Miceli, S. Bracaglia, D. Sorrentino, A. Porchetta, S. Ranallo and F. Ricci, Nucleic Acids Res., 2025, 53, gkaf238.
13. M. W. Nirenberg and J. H. Matthaei, Proc. Natl. Acad. Sci. U. S. A., 1961, 47, 1588–1602.
14. M. Nirenberg and P. Leder, Science, 1964, 145, 1399–1407.
15. Y. Cui, X. Chen, Z. Wang and Y. Lu, BioDesign Res., 2022, 2022, 9847014.
16. A. Tinafar, K. Jaenes and K. Pardee, BMC Biol., 2019, 17, 64.
17. E. Amalfitano and K. Pardee, Nat. Chem. Biol., 2022, 18, 356–358.
18. P. Zhang, H. Feng, J. Yang, H. Jiang, H. Zhou and Y. Lu, J. Biotechnol., 2019, 300, 78–86.
19. C. A. Miller, J. M. L. Ho and M. R. Bennett, Biosensors, 2022, 12, 64.
20. S. Bracaglia, S. Ranallo and F. Ricci, Angew. Chem., Int. Ed., 2023, 62, e202216512.
21. G. E. Vezeau, L. R. Gadila and H. M. Salis, Nat. Commun., 2023, 14, 2416.
22. J. K. Bains, N. S. Qureshi, B. Ceylan, A. Wacker and H. Schwalbe, Nucleic Acids Res., 2023, 51, e82–e82.
23. Y. Zhang, C. Zhao, H. Bi, X. Zhang, B. Xue, Cu Li, S. Wang, X. Yang, Z. Qiu, J. Wang and Z. Shen, J. Hazard. Mater., 2022, 438, 129499.
24. D. Y. Zhang and E. Winfree, J. Am. Chem. Soc., 2009, 131, 17303–17314.
25. D. Y. Zhang, A. J. Turberfield, B. Yurke and E. Winfree, Science, 2007, 318, 1121–1125.
26. J. B. Kapadia, J. Daoud and J. Perreault, Analyst, 2025, 150, 2019–2028.
27. K.-H. Lee and D.-M. Kim, Front. Pharmacol., 2019, 10, 01166.
28. L. Tang, T. Luo, S. Fan, Y. Liu and J. Song, Biomater. Sci., 2023, 11, 5060–5077.
29. N. Srinivas, J. Parkin, G. Seelig, E. Winfree and D. Soloveichik, Science, 2017, 358, eaal2052.
30. Y. S. Jiang, S. Bhadra, B. Li and A. D. Ellington, Angew. Chem., 2014, 126, 1876–1879.
31. L. Qian and E. Winfree, Science, 2011, 332, 1196–1201.
32. F. Wang, H. Lv, Q. Li, J. Li, X. Zhang, J. Shi, L. Wang and C. Fan, Nat. Commun., 2020, 11, 121.

---