XNAzymes targeting the SARS-CoV-2 genome inhibit viral infection

The study addresses whether fully synthetic nucleic acid enzymes (XNAzymes) can be rapidly designed to cleave SARS-CoV-2 genomic RNA under physiological conditions and thereby inhibit viral infection. The context is the urgent need for adaptable antiviral platforms highlighted by COVID-19, and limitations of existing nucleic acid therapeutics (siRNA, ASO, CRISPR-based) including specificity, biostability, immunogenicity, delivery, and viral escape. Classic DNAzymes (e.g., 10–23) often require unphysiological Mg2+ concentrations, limiting intracellular activity. The purpose is to demonstrate rapid design and screening of FANA-based XNAzymes against SARS-CoV-2 and to enhance stability and safety by assembling them into a nanostructure, testing both biochemical cleavage of viral RNA and antiviral efficacy in cells.

Prior work established nucleic-acid-based antivirals (RNAi, ASOs, CRISPR) as rapid-response modalities, but with challenges in specificity, stability, and delivery. DNAzymes such as the 10–23 motif have been retargeted to viral RNAs (including SARS-CoV) but typically need >10 mM Mg2+ for optimal folding, reducing activity at physiological Mg2+ (0.5–1 mM). Observed in vivo knockdown with DNAzymes may reflect antisense/RNase H or cytotoxicity rather than true catalysis. XNA chemistries expand nucleic acid functionality: modified DNAzymes (e.g., X10-23) and fully modified XNAzymes have been developed, including the modular FANA XNAzyme FR6.1 (FR6-I) capable of cleaving long, structured RNAs under physiological conditions and mediating allele-specific silencing in cells. These works motivate using FANA XNAzymes as precise, programmable antivirals against SARS-CoV-2.

• Target selection and XNAzyme design: Starting from the modular FANA XNAzyme FR6.1, the SARS-CoV-2 reference genome (NC_045512.2) was scanned for sites compatible with FR6.1’s preferred cleavage motif and structural accessibility (assessed by RNAfold). Off-target similarity to the human transcriptome was avoided. Initial sites analogous to the original Ebola target were tested; core mutations previously found to aid retargeting were screened to enhance activity under quasi-physiological conditions (37 °C, 1 mM Mg2+, 150 mM KCl, pH ~7.4).
• In vitro cleavage assays on short substrates: FANA XNAzymes and fluorophore-labelled RNA substrates (typically 35 nt) were annealed and incubated under quasi-physiological and high Mg2+ buffers. Urea-PAGE quantified cleavage. Single-turnover kinetics (kobs) were derived by fitting time courses. Turnover (multiple turnover) was assessed over 48 h.
• Catalytic nanostructure assembly: To enhance nuclease resistance and biosafety, three SARS-CoV-2-targeting XNAzymes (targeting ORF7b, spike, and ORF1 regions) were extended with complementary FANA sequences to self-assemble into a three-component, fully-FANA nanostructure (TF2s/TFz; ~225 nt; ~74.3 kDa) presenting one XNAzyme per edge. Assembly was confirmed by native PAGE, with further purification by depleting monomer/dimer components (ultrafiltration). Retained catalytic activity was validated on short substrates.
• Serum stability: The nanostructure and a single-stranded component were incubated in human serum (37 °C), sampled over time, and analyzed by urea-PAGE with SYBR Gold staining to determine half-lives.
• Ex vivo genomic RNA cleavage: Purified XNAzymes or nanostructure were incubated with extracted SARS-CoV-2 genomic RNA (with added total cellular RNA and RNase inhibitor) under quasi-physiological conditions for 5 h. Post-reaction removal/inactivation of XNAzymes was rigorously performed to avoid artefactual RT inhibition. Droplet digital RT-qPCR quantified depletion of amplicons spanning the ORF7b cleavage site relative to a non-target reference (CDC N2 site), with appropriate controls (irrelevant XNAzyme, omission controls).
• RNase H assays: Recombinant human RNase H1 assays were used to distinguish RNase H-mediated cleavage (antisense effect) from XNAzyme catalysis by analyzing product end-chemistries and PAGE mobility, comparing single-stranded XNAzymes vs nanostructure-embedded XNAzymes.
• Cell-based infection assay: HEK293T reporter cells overexpressing ACE2 and harboring a SARS-CoV-2 papain-like protease-activatable firefly luciferase (normalized by Renilla luciferase) were transfected with XNAzymes or the nanostructure via electroporation or lipofection. Cells were infected with authentic SARS-CoV-2 (MOI ~0.01–0.1). After ~16–24 h, infection was quantified as FFluc/Rluc ratio. Dose–response established IC50. Specificity controls included an irrelevant XNAzyme and a catalytically inactive nanostructure variant (single core mutation per XNAzyme component, [C627A]). Statistical analyses used t-tests or ANOVA as appropriate.
• Oligonucleotide synthesis and purification: FANA oligonucleotides were enzymatically synthesized using FANA triphosphates and polymerase D4K, with biotinylated DNA templates and streptavidin bead workflows, followed by primer hydrolysis and PAGE purification. Detailed buffer compositions and cycling conditions for PCR/ddPCR are provided.

• Rapid design and screening identified five active FANA XNAzymes targeting sites across the SARS-CoV-2 genome (including ORF7b, spike, ORF1a/1b). The three most active (targeting ORF7b, spike, and ORF1 regions) were advanced.
• Biochemical activity: XNAzymes specifically cleaved target RNAs under quasi-physiological conditions, showing single-turnover kinetics comparable to prior FR6.1 activities on long RNAs. Some variants demonstrated modest rates (e.g., an early ORF1b-targeting design exhibited kobs ≈ 0.1 ± 0.01 h−1), yet all selected XNAzymes displayed measurable activity and achieved 2–8 catalytic turnovers over 48 h.
• Catalytic nanostructure: The three XNAzymes self-assembled into a fully FANA three-component nanostructure (TF2s), retained catalytic activity equivalent to component enzymes, and exhibited increased serum stability (approximately fourfold increase in half-life vs single-stranded FANA).
• Ex vivo genomic RNA cleavage: ddPCR quantification revealed site-specific depletion at the ORF7b cleavage site relative to the non-target N2 site after 5 h incubation: ~60% depletion with the single ORF7b XNAzyme and ~70% with the TF2s nanostructure, indicating cleavage of full-length genomic SARS-CoV-2 RNA.
• Cellular antiviral activity: In HEK293T reporter cells infected with authentic SARS-CoV-2, transfection with the active TF2s nanostructure reduced infection in a dose-dependent manner (IC50 ≈ 54.7 pmol per 10^6 cells). At the highest tested dose (100 pmol per 10^6 cells), infection was reduced by ~75%, with an estimated 60–70% of the effect attributable to specific XNAzyme catalysis rather than antisense mechanisms.
• Specificity controls: An irrelevant XNAzyme did not inhibit infection. A catalytically inactive TF2s variant produced only minor inhibition, supporting that antiviral activity primarily derives from catalytic cleavage. RNase H assays indicated that embedding XNAzymes within the nanostructure limits RNase H-mediated cleavage, favoring bona fide XNAzyme catalysis.

The work demonstrates that fully synthetic FANA XNAzymes can be rapidly retargeted to multiple SARS-CoV-2 genomic sites and function under physiological conditions to cleave full-length viral genomic RNA. Assembling three XNAzymes into a catalytic nanostructure improves biostability (serum half-life) and reduces antisense-type RNase H contributions, enhancing biosafety and interpretability of antiviral effects. In cell culture, the active nanostructure significantly inhibits authentic SARS-CoV-2 infection with low to mid-picomole per 10^6 cells potency (IC50 ~55 pmol), achieving up to ~75% inhibition at higher doses. These findings address the initial hypothesis that XNAzymes can serve as a rapidly adaptable antiviral platform, complementing existing RNA-targeted approaches. Compared with reported siRNAs, microRNAs, and CRISPR-Cas13 strategies where many guides yield <50–65% inhibition, the XNAzyme approach achieves competitive inhibition with a small number of rationally designed catalysts and benefits from catalytic turnover and modularity. The results suggest XNAzymes could be rapidly deployed against emerging variants and other RNA viruses, with potential improvements through expanded target site screening, structural optimization of catalytic cores, and alternative chemistries to further enhance stability and activity.

This study establishes a rapid pipeline to design, synthesize, and screen FANA XNAzymes targeting SARS-CoV-2, identifies multiple active catalysts, and demonstrates that a three-component catalytic XNA nanostructure retains activity, exhibits improved serum stability, cleaves full-length genomic RNA ex vivo, and inhibits authentic SARS-CoV-2 infection in cells. The platform highlights XNAzymes as adaptable antivirals with potential advantages in specificity and modularity. Future work should focus on broader screening for optimal cleavage sites on long viral RNAs, structural and mechanistic elucidation of the XNAzyme core to refine folding and catalysis, strategies to increase turnover (e.g., binding arm tuning), evaluation of additional chemistries (e.g., LNAs), and exploration of targeting negative-sense RNA. Advancing delivery and in vivo pharmacology will be essential for translation.

• Catalytic rates for several retargeted XNAzymes are modest with slow turnover (few turnovers over 48 h), which may limit efficacy without optimization.
• The extent to which nanostructures remain assembled inside cells is unclear; disassembly could impact biostability and activity.
• Some inhibition observed with inactive nanostructures suggests minor non-catalytic effects (e.g., protein interactions or low-level RNase H contributions), although controls and RNase H assays indicate the dominant role of catalysis.
• Target site selection was not exhaustive; accessibility and local RNA structure likely affect performance.
• Studies relied on transfection in cell culture; in vivo delivery, biodistribution, and immunogenicity were not addressed.
• Quantitative linkage between ex vivo cleavage fraction and in-cell antiviral effect is indirect; precise intracellular concentrations and assembly states are not measured.