Skip to main content
NCBI home page
JACS Au logoLink to JACS Au
. 2026 May 13;6(6):3506–3514. doi: 10.1021/jacsau.6c00484

Functional DNAzyme-Guided Target Identification and Portable Detection of Colorectal Cancer

Yu Yan †, Hao Zheng †, Shenyang Zhao ‡, Lishuang Lin ‡, Qinqin Hu †,*, Jie Liu ‡,*, Hongzhou Gu †,§,∥,⊥,*
PMCID: PMC13291997  PMID: 42358703

Abstract

Rapid discrimination of colorectal cancer (CRC) at the time of tissue sampling remains clinically challenging. Although DNAzyme-based assays enable selective analysis of complex biological samples, their diagnostic translation is constrained by undefined endogenous activators. Here, using an iterative DNAzyme activity-guided isolation strategy that integrates chromatographic fractionation with affinity enrichment, we identified malic enzyme 1 (ME1) as the molecular activator of a DNAzyme newly selected against colorectal cancer tissue lysates. The ME1-triggered catalytic reaction was then translated into a lateral flow format, enabling instrument-free visual detection within 35 min. In clinical specimens, the platform discriminated CRC from low- and high-grade intraepithelial neoplasia with an AUC of 0.98. By defining the molecular basis of DNAzyme activation and coupling it to a portable readout, this work provides a direct route from evolving DNAzyme probes for sensing complex tissues to rapid, near-point-of-care detection of CRC.

Keywords: DNAzyme, colorectal cancer, target identification, malic enzyme 1, point-of-care diagnostics

graphic file with name au6c00484_0009.jpg

graphic file with name au6c00484_0007.jpg

Introduction

Colorectal cancer (CRC) is among the most prevalent and lethal malignancies worldwide, with both incidence and mortality projected to increase substantially in the coming decades. , Early and accurate diagnosis is critical for effective treatment and improved prognosis. In current clinical practice, the definitive diagnosis of CRC depends on colonoscopic biopsy followed by histopathological evaluation, including hematoxylin and eosin (H&E) staining and, when indicated, immunohistochemical or molecular analyses. − Although highly reliable, this tissue-based assessment is labor-intensive, requires specialized personnel, and typically takes several days to deliver results. The time and laboratory infrastructure required for pathological evaluation underscore the need for rapid and accessible approaches that can support specimen triage and clinical decision-making at or near the point of care.

Catalytic DNA molecules, known as deoxyribozymes or DNAzymes, are generated through in vitro selection from random DNA libraries. − As single-stranded catalysts, DNAzymes couple specific target recognition with a catalytic reaction that can be harnessed for signal amplification. For instance, RNA-cleaving DNAzymes (RCDs), − one of the most studied classes, can be evolved through iterative selection to respond selectively to complex biological mixtures, such as crude extracellular or intracellular mixtures derived from bacteria or human cell lines. − When paired with optical readouts such as fluorimetric − or colorimetric − methods, these functional probes enable rapid “mix-and-read” analysis directly in unprocessed samples. In many cases, the endogenous activators of DNAzymes selected against complex biological samples are believed to be proteinaceous in nature, yet their exact identities often remain unknown. As a result, assay outputs enable discrimination between samples but cannot be attributed to defined molecular targets. Without identifying the target molecule, mechanistic investigation is hindered, and systematic evaluation of assay performance, including specificity, sensitivity, and cross-reactivity becomes challenging, thereby limiting downstream diagnostic development and clinical translation.

To link functional sensing with molecular identification, we drew inspiration from bioassay-guided fractionation (BGF), a well-established strategy in natural product chemistry and drug discovery. , In BGF, iterative chromatographic fractionation is coupled with a bioassay to track biological activity and isolate active compounds. Adapting this concept to DNAzyme sensing, we replaced the traditional bioassay with DNAzyme-mediated cleavage as the trackable activity, thereby establishing a DNAzyme activity-guided strategy for identifying endogenous activators from complex biological samples. The catalytic amplification intrinsic to DNAzymes is particularly advantageous, as it enhances the sensitivity of activity tracking during fractionation. Establishing the molecular basis of DNAzyme activation opens the way for investigating disease-associated changes and supports the rational improvement of DNAzyme-based diagnostic approaches.

Building on this framework, we first performed in vitro selection against clinical CRC tissue lysates to obtain DNAzymes capable of selectively sensing tumor-derived complex mixtures. Using the DNAzyme activity-guided isolation strategy, we subsequently identified the cognate molecular activator of the selected DNAzyme, enabling mechanistic characterization of the sensing event. Finally, we translated the catalytic reaction into a portable lateral flow device (LFD) − for rapid, instrument-free detection of CRC tissue lysates. Collectively, this study integrates functional DNAzyme selection with molecular activator identification and portable detection, establishing a workflow from complex-sample screening to point-of-care-compatible analysis.

Results and Discussion

In Vitro Selection of CRC-Responsive DNAzymes

Clinical tissue samples were collected from 11 colorectal cancer patients (5 females, 6 males, ages 31–87) by endoscopic mucosal resection (Figure a). To mitigate inter-individual variability and enrich for shared CRC-associated molecular features, cancer tissues (CT) from all donors were pooled to generate a mixed target lysate, whereas paired adjacent normal tissues (NT) were combined to form the control lysate. DNAzyme selection was performed using a subtractive/positive systematic evolution of ligands by exponential enrichment (SELEX) strategy, , in which each round alternated between counter-selection against NT lysate and positive selection with CT lysate (Figure a).

1.

1

Open in a new tab

Selection of DNAzyme probes that specifically sense colorectal cancer tissue. (a) In vitro selection scheme. R12 represents the final enriched pool selected under stringent conditions (5 min incubation with CT lysate). Filled and hollow arrowheads indicate full-length (99 nt) and 3′ cleavage products (85 nt), respectively. (b) Sequences of the candidate DNAzyme probes. Gray shading represents the nucleotides conserved among these candidates, with the exception of those marked in green. (c) Selective sensing of CRC tissue lysate (CT) by Dz3. (d) Kinetic profiles of Dz3 in NT and CT lysates. Data are extracted from panel (c) and presented as mean ± s.d.; n = 3 independent experiments.

The initial library consisted of 99-mer oligonucleotides bearing a 40-nt randomized region and a single ribonucleoside adenosine (rA) cleavage site. Cleavage activity of the evolving pool was monitored in CT and NT lysates for each round (R). Sequence analysis was conducted on the twelfth-round (R12) pool (Figure a), as the signal difference between positive and subtractive selection had plateaued (Figure S1). Colony sequencing revealed 28 candidate DNAzyme sequences (Dz1–Dz28; Figure S2a; Figure b), all of which were confirmed to be capable of sensing the CT lysate (Figure S2b). Based on sequence abundance and catalytic performance, Dz3 and Dz5 were selected for further characterization. Both DNAzymes selectively recognized CT lysate and cleaved the RNA linkage in a time-dependent manner (Figure c; Figure S3a), with observed rate constant (k obs) values of 0.13 min–1 for Dz3 (Figure d) and 0.06 min–1 for Dz5 (Figure S3b), while showing negligible activity in NT lysate. To assess whether sequence truncation could further improve activity, we used Dz3 as a representative candidate and introduced a series of programmed 10-nt deletions at different positions (Figure S4). All variants remained active (46–67% cleavage), but none outperformed the full-length Dz3 (74%). Therefore, the full-length Dz3 was used in subsequent experiments.

Identification of the Target Molecule

Heat denaturation and proteinase K digestion prevented the CT lysate from triggering Dz3 (Figure S5a), indicating that the Dz3 activator is a protein; molecular-weight-cutoff (MWCO) filtration further estimated its apparent molecular weight to be below 100 kDa (Figure S5b). Because the amount of clinical tissue available per patient (<0.1 g) was limited, the starting material was insufficient for the multistep purification necessary to isolate the target protein. We therefore screened several colorectal cancer cell lines using Dz3 and identified HCT116 as expressing the highest level of the Dz3-responsive protein (Figure S5c). HCT116 cells were subsequently expanded to obtain ∼5 mL of cell pellet, yielding ∼10 mL of lysate at ∼10 mg mL–1 total protein, which was sufficient to support the subsequent purification workflow.

We implemented an iterative, Dz3-guided chromatographic fractionation strategy, in which Dz3 activity was used to track the target protein at each step. The workflow employed three orthogonal chromatographic steps in sequence: anion-exchange (AEX), cation-exchange (CEX), and size-exclusion chromatography (SEC). At pH 7.4, Dz3 probing confirmed that the target protein was retained on a 10 mL Source 15Q column; the strongest signals were observed in fractions F8–F10 (Figure a). These were pooled and subjected to CEX at pH 4.5, where the target protein bound efficiently to a Source 15S column (Figure S6). Dz3 analysis identified fraction F23 as the most enriched, which was further purified by SEC on a Superdex 75 Increase 10/300 GL column. Fraction F33 displayed the relatively highest Dz3 activity (Figure a), indicating efficient enrichment of the target protein. However, after this three-step purification process, only a few micrograms of total protein were recovered due to unavoidable losses during fractionation, which was insufficient for further chromatographic purification. To overcome this limitation, we implemented an affinity-based enrichment strategy.

2.

2

Open in a new tab

Pre-purification and pulldown for target identification. (a) Biochemical pre-purification of HCT116 cell lysate monitored by Dz3 cleavage. (b) Pulldown of the target molecule using deactivated Dz3 (dDz3). (c) Monitoring of the pulldown workflow by Dz5, another DNAzyme selective for CRC tissue lysate. (d) Identification of the target (malic enzyme 1, ME1) enriched by dDz3 from pre-purified lysate. The SDS-PAGE gel was silver-stained. (e) Sensitivity comparison of Dz3 and anti-ME1 antibody against the recombinant ME1 (rME1). (f) Induced Dz3 cleavage signal versus logarithm of the effector (rME1) concentration. Data were extracted from panel e. Apparent EC50: 3.8 ± 0.5 nM (30 min) and 1.2 ± 0.2 nM (2 h). (g) Flow cytometry analysis of dDz3 binding to rME1. The estimated K D is 3.0 ± 0.5 μM. RFU, relative fluorescence unit. Data are presented as mean ± s.d.; n = 3 independent experiments.

To isolate the protein, biotinylated RNA-to-DNA variants of Dz3 (dDz3) were immobilized on streptavidin magnetic beads and used as affinity tools for target pulldown (Figure b; Figure S7), building on established methods for converting catalytic DNAzymes into affinity reagents via RNA-to-DNA substitution. Because Dz3 and Dz5 differ by only a single nucleotide within the conserved region and were enriched during selection (Figure b; Figure S2), we hypothesized that they recognize the same molecular target, which was subsequently confirmed experimentally (Figure S8). Dz5 was therefore employed as an orthogonal reporter to monitor catalytic activity during the pulldown workflow (Figure c). Silver-stained SDS-PAGE showed that the pre-purified fraction (F33) contained fewer than ten visible protein bands, whereas the pulldown eluate was dominated by a single ∼60 kDa species (Figure d). This strong enrichment is consistent with the <100 kDa estimate obtained from MWCO filtration (Figure S5b) and demonstrates the effectiveness of the DNAzyme activity-guided isolation strategy.

Mass spectrometry identified malic enzyme 1 (ME1; calculated mass: 64.1 kDa) as the protein target (Figure S9a), which was independently validated by western blotting using an anti-ME1 antibody in both HCT116 lysate and the pulldown eluate (Figure S9b). To further validate ME1 as the functional target of Dz3, we generated ME1-knockout (KO) HCT116 cells using CRISPR/Cas9 (Figure S10a). Successful ME1 knockout was verified by PCR genotyping and western blotting (Figure S10b,c). As expected, Dz3 cleavage was completely abolished in KO lysates (Figure S10d), firmly establishing ME1 as the direct functional activator of Dz3.

ME1 is a cytosolic NADP+ -dependent enzyme that produces NADPH to support lipid biosynthesis and redox homeostasisprocesses commonly dysregulated in cancer metabolism. − Notably, ME1 is present at low abundance in colon tissue (∼50 ppm, http://pax-db.org/), posing the analytical challenge of its detection in clinical lysates.

To further characterize the Dz3–ME1 interaction, recombinant ME1 (rME1) was expressed and purified. Dz3 displayed EC50 values of 3.8 ± 0.5 nM (30 min) and 1.2 ± 0.2 nM (2 h), achieving a detection limit of ∼48 pMapproximately 81-fold more sensitive than anti-ME1 western blotting (∼3.9 nM) (Figure e). This superior sensitivity arises from Dz3’s catalytic mechanism: a single ME1 molecule triggers multiple DNAzyme turnovers, enabling signal amplification. At the EC50 concentration of rME1 (3.8 nM), half of the Dz3 (200 nM) underwent cleavage within 30 min (Figure e,f), equating to an average of ∼27 Dz3 molecules activated per rME1.

Furthermore, binding studies of the deactivated dDz3 with rME1 showed a dissociation constant (K D) of 3.0 ± 0.5 μM (Figure g; Figure S11a). Although the measured K D falls within the micromolar range, , such moderate affinity is compatible with rapid target dissociation, thus supporting catalytic amplification and enhanced sensitivity at low ME1 concentrations. dDz3 also exhibited high binding specificity for ME1 (Figure S11b), reflecting the sequence’s origin from a highly selective DNAzyme.

To evaluate the clinical relevance of ME1 as an analytical target in CRC, we analyzed an independent patient cohort from the GEO dataset GSE44076. ME1 transcripts were significantly elevated in CRC tissues compared with matched normal controls (98 pairs; Figure a; P = 4.0 × 10–7), demonstrating consistent upregulation in clinical specimens. Stratification of tumors into ME1-high and ME1-low groups revealed distinct transcriptomic patterns by gene set enrichment analysis (GSEA), with ME1-high tumors showing enrichment of proliferative and metabolic programs, whereas ME1-low tumors were associated with tumor microenvironment-related signatures (Figure b–c). ESTIMATE analysis further indicated higher stromal and immune scores in ME1-low tumors (Figure d–g). Collectively, these independent transcriptomic analyses corroborate the upregulation of ME1 in CRC and substantiate its analytical relevance as a molecular discriminator in tissue lysates.

3.

3

Open in a new tab

ME1 upregulation in CRC and associated transcriptomic and microenvironmental features. (a) ME1 mRNA expression levels in paired normal and CRC tissues from the GSE44076 cohort (n = 98 pairs). Gene expression data were processed using robust multiarray average (RMA) normalization. Boxes represent the median and interquartile range (IQR); whiskers indicate 1.5 × IQR; dots denote individual samples. P value was determined using a paired two-sided t-test. (b,c) Gene set enrichment analysis (GSEA) of representative Gene Ontology (GO) terms in ME1-high CRC (NES > 0, b) and ME1-low CRC (NES < 0, c). Adjusted P values are reported as false discovery rate (FDR) using the Benjamini–Hochberg correction. (d–g) ESTIMATE-derived stromal score (d), immune score (e), ESTIMATE score (f), and tumor purity score (g) compared between ME1-high and ME1-low CRC. Violin plots show distributions with individual samples overlaid. P values were determined using a two-sided Wilcoxon rank-sum test. P values are shown in the plots; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Lateral Flow Readout of DNAzyme Cleavage

Having defined ME1 as the molecular activator of Dz3, we next integrated the catalytic reaction with a lateral flow device (LFD) for visual detection (Figure ). − A dual-labeled Dz3 probe (5′-FAM, 3′-biotin) was designed to anchor onto streptavidin (SA)-agarose beads, forming an SA-bead-Dz3 complex that serves as the sensing module for ME1.

4.

4

Open in a new tab

Translation of DNAzyme cleavage into a lateral flow device (LFD) readout. (a) Schematic illustration of DNAzyme cleavage-induced signal generation for LFD readout. ME1-activated, bead-immobilized 5′-FAM–Dz3 cleaves the substrate, releasing a 5′ fragment that hybridizes with a biotinylated complementary strand to generate a dual-labeled FAM/biotin product. (b) Schematic of the Dz3-enabled positive LFD readout (TL, test line; CL, control line). (c) Selectivity of the Dz3-based LFD toward ME1 over a panel of control proteins, analyzed by LFD (top panel) and denaturing PAGE (dPAGE, bottom panel). (d) Sequence-scrambled Dz3 control (sDz3) analyzed by LFD and dPAGE. (e) Sensitivity of the Dz3-based LFD toward rME1. (f) Dependence of TL intensity on log­[rME1], yielding an EC50 of 3.8 ± 1.7 nM. Data are presented as mean ± s.d.; n = 3 independent experiments.

Upon ME1-dependent cleavage, the released 5′-FAM-tagged fragment hybridizes with a 5′-biotin-modified complementary strand, generating a duplex suitable for LFD analysis (Figure a). As the mixture migrates along the strip, the duplex associates with SA-functionalized gold nanoparticles (AuNPs-SA) and is captured intact at the test line by anti-FAM antibodies, producing a clear visible signal. Excess AuNPs-SA generates the control line by binding to immobilized BSA-biotin, where BSA stabilizes biotin presentation on the nitrocellulose membrane (Figure b).

To ensure robust strip performance, we optimized the AuNPs-SA conjugation conditions. Because passive adsorption of SA onto AuNPs is favored under mildly alkaline conditions, buffers of varying pH were evaluated, and pH 8.5 yielded the strongest control-line intensity (Figure S12). SA loading was further optimized on ∼25 nm AuNPs (characterized by TEM and DLS; Figure S13), with 4 μg mL–1 providing the maximal test-line signal (Figure S14), likely due to improved conversion efficiency of cleaved fragments into discrete AuNP-based readouts.

Specificity was evaluated using rME1 together with four selected nontarget proteins (SARS1, HSP70, BSA, and IgG1), including biologically relevant intracellular proteins and common assay-background proteins. Only rME1 generated a visible test line, confirming high selectivity. The LFD achieved a detection limit of ∼192 pM (Figure e,f), comparable to that of gel-based analysis (Figure e,f). This detection sensitivity is sufficient to resolve ME1 at concentrations relevant to CRC tissue lysates. Moreover, the LFD showed stable analytical performance over 180 days, with no significant reduction in test line intensity (Figure S15). Taken together, the LFD assay offers a user-friendly, instrument-free visual readout suitable for on-site CRC diagnostics.

Clinical Applicability of the Lateral Flow Device for CRC Detection

To assess the clinical applicability of the LFD, we analyzed an independent cohort of 14 paired CRC tumor (CT) and matched normal tissue (NT) lysates (Figure a). The sample-to-result workflow comprised ∼10 min tissue homogenization, 10 min DNAzyme cleavage, 5 min hybridization, and ∼10 min LFD readout, for a total assay time of ∼35 min. Across all patients, Dz3 produced stronger test-line (TL) signals in CT lysates than in matched NT controls (Figure b, top), with a mean TL intensity of 6146 ± 2114 for CT lysates versus 2316 ± 1286 for NT lysates (P < 0.0001). The observed inter-patient variability likely reflects heterogeneity in CRC metabolic states. Parallel denaturing PAGE analysis of the same lysates showed similar trends (Figure b, bottom panel), with average cleavage of 44.2 ± 17.6% in CT lysates versus 8.4 ± 7.9% in NT lysates (P < 0.0001).

5.

5

Open in a new tab

Evaluation of the LFD–Dz3 platform in CRC and premalignant colorectal tissues. (a) Representative hematoxylin and eosin (H&E)-stained images of CRC tissue (CT), high-grade intraepithelial neoplasia (HGIN) tissue (HT), and low-grade intraepithelial neoplasia (LGIN) tissue (LT). Scale bars, 100 μm. (b) Dz3-based LFD readouts (top panel) and dPAGE analysis (bottom panel) for paired CRC tissue (CT) and adjacent normal tissue (NT) lysates from 14 CRC patients. CL, control line; TL, test line. (c) Test-line intensity from LFD measurements and (d) induced Dz3 cleavage (%) from dPAGE analysis for paired CT–NT (n = 14), HT–NT (n = 10), and LT–NT (n = 10) lysates (biologically independent pairs). Data extracted from b and Figure S16a–d are presented as mean ± s.d.; paired samples are indicated by identical colors and marker shapes. P values were calculated using paired two-sided t-tests and are shown in the plots; NS, not significant; ****P < 0.0001.

Given the robust discrimination between CRC and matched normal tissues, we next evaluated whether the LFD-Dz3 platform responds to earlier-stage neoplastic precursor lesions. According to the World Health Organization’s two-tier histological classification of digestive neoplastic precursor lesions (low-grade vs high-grade intraepithelial neoplasia, LGIN and HGIN, respectively), we analyzed paired tissue lysates from 10 HGIN and 10 LGIN patients (with each lesion paired with matched normal tissue from the same patient) using both LFD and dPAGE readouts to assess Dz3 reactivity. In contrast to CRC, neither HGIN nor LGIN tissues exhibited significant differences relative to paired normal tissues in TL intensity or DNAzyme cleavage yield (all P > 0.05; Figure c,d and Figure S16).

Based on CRC-specific signal elevation, a binary classification model was established to distinguish CRC from premalignant lesions (LGIN/HGIN) (Figure ). Receiver operating characteristic (ROC) analysis showed excellent diagnostic performance for both LFD and dPAGE readouts, with identical areas under the curve (AUC = 0.98) (Figure a,b). Classification was based on the log2 fold change, defined as the lesion-to-matched-normal signal ratio. This patient-matched ratiometric readout enables built-in normalization and thus enhances robustness against signal fluctuation and sample matrix effects. , CRC samples showed significantly higher responses than premalignant lesions (P < 10–4; Figure c,d). Applying optimized cutoffs (0.54 for LFD and 0.61 for dPAGE), the assay achieved 100% sensitivity (14/14 CRC cases) in this cohort, with specificities of 90% (18/20) for LFD and 95% (19/20) for dPAGE, and overall accuracies of 94% and 97%, respectively (Figure e,f). These results highlight the assay’s ability to reliably discriminate between malignant and premalignant colorectal lesions.

6.

6

Open in a new tab

Dz3-based discrimination of colorectal cancer (CRC) from premalignant lesions (HGIN/LGIN). (a,b) ROC curves for the Dz3-based LFD (a) and dPAGE (b) assays distinguishing CRC from HGIN/LGIN (AUC = 0.98 for both). (c,d) Log2 fold change of test-line (TL) intensity (c) and induced Dz3 cleavage (d) in CRC versus HGIN/LGIN. Dashed lines indicate optimal cutoffs determined by Youden index. P values from unpaired two-sided t-tests: 3.3 × 10–5 (c) and 7.6 × 10–5 (d) (****P < 0.0001). (e,f) Confusion matrices for the LFD (e) and dPAGE (f) assays. Numbers represent sample counts; darker shading denotes correct classifications. Sensitivity, specificity, and accuracy are shown for each assay.

Conclusions

In summary, by leveraging the DNAzyme activity-guided isolation strategy, we identified ME1 as the molecular activator of the CRC-responsive DNAzyme evolved from complex tissue lysates. Translation of ME1-triggered catalysis into a lateral flow readout enabled rapid, instrument-free evaluation of excised colorectal tissues within 35 min, with robust discrimination of CRC from premalignant lesions (AUC = 0.98). Compared with conventional pathology-based workflows, which typically require tissue processing and expert histopathological interpretation, this platform offers a substantially shorter time-to-result, operational simplicity, and an instrument-free visual output, supporting its potential use as a complementary tool for rapid tissue assessment. In addition, the current DNAzyme preferentially responds to malignant CRC tissues rather than premalignant lesions and is therefore not suitable as a standalone probe for early CRC screening. Early detection would more likely require a panel of DNAzyme probes targeting different stages of colorectal neoplastic progression. More broadly, by resolving the molecular basis of functional DNAzyme responses and coupling it to portable readout, this work establishes a coherent pathway from complex-sample selection to mechanism-informed assay translation.

Supplementary Material

au6c00484_si_001.pdf (6.5MB, pdf)

Acknowledgments

We acknowledge funding by the National Natural Science Foundation of China (82121002 to J.L. and H.G.; 22525404, 82120108010, U24A20377 to H.G.), the Autonomous Project of the State Key Laboratory of Synergistic Chem-Bio Synthesis (sklscbs202570 to H.G.), and Shanghai Pilot Program for Basic Research. This research was also supported by the National Research Foundation, Prime Minister’s Office, Singapore, under its Campus for Research Excellence and Technological Enterprise (CREATE) program (CNSB). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.6c00484.

  • Detailed experimental procedures, DNA sequences, selection process, DNAzyme activity tests, LC–MS/MS analysis, immunoblot validation, CRISPR/Cas9-mediated knockout, flow cytometry analysis, LFD assays, and additional figures and tables (PDF)

H.G. applied for the funding. H.G. conceived the idea and supervised the project. Y.Y., H.Z., S.Z., and L.L. performed the experiments and data collection. Y.Y. analyzed the data. Y.Y., Q.H., and H.G. wrote the manuscript. Y.Y., H.Z., and S.Z. contributed equally.

The authors declare no competing financial interest.

References

  1. Bray F., Laversanne M., Sung H., Ferlay J., Siegel R. L., Soerjomataram I., Jemal A.. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. Ca-Cancer J. Clin. 2024;74(3):229–263. doi: 10.3322/caac.21834. [DOI] [PubMed] [Google Scholar]
  2. Morgan E., Arnold M., Gini A., Lorenzoni V., Cabasag C. J., Laversanne M., Vignat J., Ferlay J., Murphy N., Bray F.. Global burden of colorectal cancer in 2020 and 2040: incidence and mortality estimates from GLOBOCAN. Gut. 2023;72(2):338–344. doi: 10.1136/gutjnl-2022-327736. [DOI] [PubMed] [Google Scholar]
  3. Gurcan M. N., Boucheron L. E., Can A., Madabhushi A., Rajpoot N. M., Yener B.. Histopathological image analysis: a review. IEEE Rev. Biomed. Eng. 2009;2:147–171. doi: 10.1109/RBME.2009.2034865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chen W., Frankel W. L.. A practical guide to biomarkers for the evaluation of colorectal cancer. Mod. Pathol. 2019;32(Suppl 1):1–15. doi: 10.1038/s41379-018-0136-1. [DOI] [PubMed] [Google Scholar]
  5. Pollett A.. Colorectal cancer: microsatellite instability/mismatch repair testing in the era of digital pathology. Gastroenterology. 2020;159(4):1235–1237. doi: 10.1053/j.gastro.2020.08.008. [DOI] [PubMed] [Google Scholar]
  6. Breaker R. R., Joyce G. F.. A DNA enzyme that cleaves RNA. Chem. Biol. 1994;1(4):223–229. doi: 10.1016/1074-5521(94)90014-0. [DOI] [PubMed] [Google Scholar]
  7. Dokukina V., Silverman S. K.. Lanthanide ions as required cofactors for DNA catalysts. Chem. Sci. 2012;3(5):1707–1714. doi: 10.1039/c2sc01067d. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Sednev M. V., Mykhailiuk V., Choudhury P., Halang J., Sloan K. E., Bohnsack M. T., Höbartner C.. N6-methyladenosine-sensitive RNA-cleaving deoxyribozymes. Angew. Chem., Int. Ed. 2018;57(46):15117–15121. doi: 10.1002/anie.201808745. [DOI] [PubMed] [Google Scholar]
  9. Pandey R., Chang D., Smieja M., Hoare T., Li Y., Soleymani L.. Integrating programmable DNAzymes with electrical readout for rapid and culture-free bacterial detection using a handheld platform. Nat. Chem. 2021;13(9):895–901. doi: 10.1038/s41557-021-00718-x. [DOI] [PubMed] [Google Scholar]
  10. Wang Y., Wang Y., Song D., Sun X., Li Z., Chen J.-Y., Yu H.. An RNA-cleaving threose nucleic acid enzyme capable of single point mutation discrimination. Nat. Chem. 2022;14(3):350–359. doi: 10.1038/s41557-021-00847-3. [DOI] [PubMed] [Google Scholar]
  11. Gao L., Yi K., Tan Y., Guo C., Zheng D., Shen C., Li F.. Engineering gene-specific DNAzymes for accessible and multiplexed nucleic acid testing. JACS Au. 2024;4(4):1664–1672. doi: 10.1021/jacsau.4c00232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hu Q., Tong Z., Yalikong A., Ge L.-P., Shi Q., Du X., Wang P., Liu X.-Y., Zhan W., Gao X., Sun D., Fu T., Ye D., Fan C., Liu J., Zhong Y.-S., Jiang Y.-Z., Gu H.. DNAzyme-based faithful probing and pulldown to identify candidate biomarkers of low abundance. Nat. Chem. 2024;16(1):122–131. doi: 10.1038/s41557-023-01328-5. [DOI] [PubMed] [Google Scholar]
  13. Silverman S. K.. Deoxyribozymes: selection design and serendipity in the development of DNA catalysts. Acc. Chem. Res. 2009;42(10):1521–1531. doi: 10.1021/ar900052y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. McConnell E. M., Cozma I., Mou Q., Brennan J. D., Lu Y., Li Y.. Biosensing with DNAzymes. Chem. Soc. Rev. 2021;50(16):8954–8994. doi: 10.1039/D1CS00240F. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fan H., McGhee C. E., Lake R. J., Yang Z., Guo Z., Zhang X.-B., Lu Y.. A highly selective Mn­(II)-specific DNAzyme and its application in intracellular sensing. JACS Au. 2023;3(6):1615–1622. doi: 10.1021/jacsau.3c00062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jiang Y., Chen Y., Gong X., Weng B., He Y., Wang F.. An intelligent allosteric DNAzyme platform for on-site uracil-DNA glycosylase imaging and on-demand gene regulation. JACS Au. 2025;5(9):4288–4298. doi: 10.1021/jacsau.5c00672. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ali M., Almeida A. F., White D., Capretta A., Brennan J. D.. A combined aptamer pulldown-DNAzyme cleavage assay for intact methicillin resistant Staphylococcus aureus cells. Angew. Chem., Int. Ed. 2025;64(33):e202509598. doi: 10.1002/anie.202509598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Ali M. M., Mukherjee M., Radford K., Patel Z., Capretta A., Nair P., Brennan J. D.. A rapid sputum-based lateral flow assay for airway eosinophilia using an RNA-cleaving DNAzyme selected for eosinophil peroxidase. Angew. Chem., Int. Ed. 2023;62(38):e202307451. doi: 10.1002/anie.202307451. [DOI] [PubMed] [Google Scholar]
  19. Ali M. M., Aguirre S. D., Lazim H., Li Y.. Fluorogenic DNAzyme probes as bacterial indicators. Angew. Chem., Int. Ed. 2011;50(16):3751–3754. doi: 10.1002/anie.201100477. [DOI] [PubMed] [Google Scholar]
  20. Shen Z., Wu Z., Chang D., Zhang W., Tram K., Lee C., Kim P., Salena B. J., Li Y.. A catalytic DNA activated by a specific strain of bacterial pathogen. Angew. Chem., Int. Ed. 2016;55(7):2431–2434. doi: 10.1002/anie.201510125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Zhou Q., Zhang G., Wu Y., Zhang Q., Liu Y., Chang Y., Liu M.. In vitro selection of M2+-independent, fast-responding acidic deoxyribozymes for bacterial detection. J. Am. Chem. Soc. 2023;145(39):21370–21377. doi: 10.1021/jacs.3c06155. [DOI] [PubMed] [Google Scholar]
  22. Feng Q., Zakaria S., Morrison D., Tram K., Gu J., Salena B. J., Li Y.. A fluorogenic DNAzyme for a thermally stable protein biomarker from Fusobacterium nucleatum, a human bacterial pathogen. Angew. Chem., Int. Ed. 2023;62(35):e202306272. doi: 10.1002/anie.202306272. [DOI] [PubMed] [Google Scholar]
  23. Geng X., Zhang M., Wang X., Sun J., Zhao X., Zhang L., Wang X., Shen Z.. Selective and sensitive detection of chronic myeloid leukemia using fluorogenic DNAzyme probes. Anal. Chim. Acta. 2020;1123:28–35. doi: 10.1016/j.aca.2020.04.069. [DOI] [PubMed] [Google Scholar]
  24. He S., Qu L., Shen Z., Tan Y., Zeng M., Liu F., Jiang Y., Li Y.. Highly specific recognition of breast tumors by an RNA-cleaving fluorogenic DNAzyme probe. Anal. Chem. 2015;87(1):569–577. doi: 10.1021/ac5031557. [DOI] [PubMed] [Google Scholar]
  25. Rothenbroker M., McConnell E. M., Gu J., Urbanus M. L., Esmaeili Samani S., Ensminger A. W., Filipe C. D. M., Li Y.. Selection and characterization of an RNA-Cleaving DNAzyme activated by Legionella pneumophila . Angew. Chem., Int. Ed. 2021;60(9):4782–4788. doi: 10.1002/anie.202012444. [DOI] [PubMed] [Google Scholar]
  26. Ali M. M., Wolfe M., Tram K., Gu J., Filipe C. D. M., Li Y., Brennan J. D.. A DNAzyme-based colorimetric paper sensor for Helicobacter pylori . Angew. Chem., Int. Ed. 2019;58(29):9907–9911. doi: 10.1002/anie.201901873. [DOI] [PubMed] [Google Scholar]
  27. Li J., Khan S., Gu J., Filipe C. D. M., Didar T. F., Li Y.. A simple colorimetric Au-on-Au tip sensor with a new functional nucleic acid probe for food-borne pathogen Salmonellatyphimurium . Angew. Chem., Int. Ed. 2023;62(20):e202300828. doi: 10.1002/anie.202300828. [DOI] [PubMed] [Google Scholar]
  28. Ali M. M., Silva R., White D., Mohammadi S., Li Y., Capretta A., Brennan J. D.. A lateral flow test for Staphylococcus aureus in nasal mucus using a new DNAzyme as the recognition element. Angew. Chem., Int. Ed. 2022;61(3):e202112346. doi: 10.1002/anie.202112346. [DOI] [PubMed] [Google Scholar]
  29. Ali M., Nair P., Capretta A., Brennan J. D.. In-vitro clinical diagnostics using RNA-Cleaving DNAzymes. ChemBiochem. 2024;25(11):e202400085. doi: 10.1002/cbic.202400085. [DOI] [PubMed] [Google Scholar]
  30. Harvey A. L., Edrada-Ebel R., Quinn R. J.. The re-emergence of natural products for drug discovery in the genomics era. Nat. Rev. Drug Discovery. 2015;14(2):111–129. doi: 10.1038/nrd4510. [DOI] [PubMed] [Google Scholar]
  31. Meunier M., Schinkovitz A., Derbré S.. Current and emerging tools and strategies for the identification of bioactive natural products in complex mixtures. Nat. Prod. Rep. 2024;41(11):1766–1786. doi: 10.1039/D4NP00006D. [DOI] [PubMed] [Google Scholar]
  32. Zhang H., Tang Y., Zhou Y., Wang Y., Si H., Li L., Tang B.. DNAzyme-RCA-based colorimetric and lateral flow dipstick assays for the point-of-care testing of exosomal m5C-miRNA-21. Chem. Sci. 2024;15(24):9345–9352. doi: 10.1039/D4SC02648A. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mazumdar D., Liu J., Lu G., Zhou J., Lu Y.. Easy-to-use dipstick tests for detection of lead in paints using non-cross-linked gold nanoparticle-DNAzyme conjugates. Chem. Commun. 2010;46(9):1416–1418. doi: 10.1039/b917772h. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Fang Z., Huang J., Lie P., Xiao Z., Ouyang C., Wu Q., Wu Y., Liu G., Zeng L.. Lateral flow nucleic acid biosensor for Cu2+ detection in aqueous solution with high sensitivity and selectivity. Chem. Commun. 2010;46(47):9043–9045. doi: 10.1039/c0cc02782k. [DOI] [PubMed] [Google Scholar]
  35. Chen J., Zhou X., Zeng L.. Enzyme-free strip biosensor for amplified detection of Pb2+ based on a catalytic DNA circuit. Chem. Commun. 2013;49(10):984–986. doi: 10.1039/C2CC37598B. [DOI] [PubMed] [Google Scholar]
  36. Zhang C., Zheng T., Wang H., Chen W., Huang X., Liang J., Qiu L., Han D., Tan W.. Rapid one-pot detection of SARS-CoV-2 based on a lateral flow assay in clinical samples. Anal. Chem. 2021;93(7):3325–3330. doi: 10.1021/acs.analchem.0c05059. [DOI] [PubMed] [Google Scholar]
  37. Ran F. A., Hsu P. D., Wright J., Agarwala V., Scott D. A., Zhang F.. Genome engineering using the CRISPR-Cas9 system. Nat. Protoc. 2013;8(11):2281–2308. doi: 10.1038/nprot.2013.143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Jiang P., Du W., Mancuso A., Wellen K. E., Yang X.. Reciprocal regulation of p53 and malic enzymes modulates metabolism and senescence. Nature. 2013;493(7434):689–693. doi: 10.1038/nature11776. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Zhu Y., Gu L., Lin X., Liu C., Lu B., Cui K., Zhou F., Zhao Q., Prochownik E. V., Fan C., Li Y.. Dynamic regulation of ME1 phosphorylation and acetylation affects lipid metabolism and colorectal tumorigenesis. Mol. Cell. 2020;77(1):138–149. doi: 10.1016/j.molcel.2019.10.015. [DOI] [PubMed] [Google Scholar]
  40. Ju H. Q., Lin J. F., Tian T., Xie D., Xu R. H.. NADPH homeostasis in cancer: functions, mechanisms and therapeutic implications. Signal Transduction Targeted Ther. 2020;5(1):231. doi: 10.1038/s41392-020-00326-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wang M., Weiss M., Simonovic M., Haertinger G., Schrimpf S. P., Hengartner M. O., von Mering C.. PaxDb, a database of protein abundance averages across all three domains of life. Mol. Cell. Proteomics. 2012;11(8):492–500. doi: 10.1074/mcp.O111.014704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Green L. S., Jellinek D., Jenison R., Östman A., Heldin C. H., Janjic N.. Inhibitory DNA ligands to platelet-derived growth factor B-chain. Biochemistry. 1996;35(45):14413–14424. doi: 10.1021/bi961544+. [DOI] [PubMed] [Google Scholar]
  43. Jellinek D., Green L. S., Bell C., Janjić N.. Inhibition of receptor binding by high-affinity RNA ligands to vascular endothelial growth factor. Biochemistry. 1994;33(34):10450–10456. doi: 10.1021/bi00200a028. [DOI] [PubMed] [Google Scholar]
  44. Sanz-Pamplona R., Berenguer A., Cordero D., Molleví D. G., Crous-Bou M., Sole X., Paré-Brunet L., Guino E., Salazar R., Santos C.. et al. Aberrant gene expression in mucosa adjacent to tumor reveals a molecular crosstalk in colon cancer. Mol. Cancer. 2014;13:46. doi: 10.1186/1476-4598-13-46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Lew T. T. S., Aung K. M. M., Ow S. Y., Amrun S. N., Sutarlie L., Ng L. F. P., Su X.. Epitope-functionalized gold nanoparticles for rapid and selective detection of SARS-CoV-2 IgG antibodies. ACS Nano. 2021;15(7):12286–12297. doi: 10.1021/acsnano.1c04091. [DOI] [PubMed] [Google Scholar]
  46. Nagtegaal I. D., Odze R. D., Klimstra D., Paradis V., Rugge M., Schirmacher P., Washington K. M., Carneiro F., Cree I. A.. WHO Classification of Tumours Editorial Board. The 2019 WHO classification of tumours of the digestive system. Histopathology. 2020;76(2):182–188. doi: 10.1111/his.13975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Lee M. H., Kim J. S., Sessler J. L.. Small molecule-based ratiometric fluorescence probes for cations, anions, and biomolecules. Chem. Soc. Rev. 2015;44(13):4185–4191. doi: 10.1039/C4CS00280F. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Huang X., Song J., Yung B. C., Huang X., Xiong Y., Chen X.. Ratiometric optical nanoprobes enable accurate molecular detection and imaging. Chem. Soc. Rev. 2018;47(8):2873–2920. doi: 10.1039/C7CS00612H. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

au6c00484_si_001.pdf (6.5MB, pdf)

Articles from JACS Au are provided here courtesy of American Chemical Society

RESOURCES