Enhancing Clinical Detection Accuracy of Large Structured Viral RNA via DNAzyme Cleavage and Antisense-Assisted Rolling Circle Amplification

Abstract

Sensitive detection of viral RNA is critical for accurate diagnostic testing, particularly during outbreaks of emerging infectious diseases. Rolling circle amplification (RCA) is a powerful isothermal amplification strategy that can be directly primed by RNA, eliminating the need for reverse transcription. Previous approaches have used 10-23 DNAzymes to cleave viral RNA, generating 3'-ends for hybridization to circular DNA templates (CDTs). However, the resulting RNA fragments often retained secondary or tertiary structures that hindered CDT binding and limited RCA efficiency. To address this challenge, we developed antisense oligonucleotide-assisted RCA (ASO-RCA), a general strategy that uses short upstream antisense oligonucleotides (ASOs) to remodel RNA structure and expose the CDT-binding site. Using five DNAzyme-CDT systems targeting distinct regions of the SARS-CoV-2 genome, we show that ASO inclusion improves CDT hybridization and enhances RCA output-by up to 70-fold. This enhancement was observed using both linear and quasi-exponential RCA formats and remained effective in 50% pooled saliva. When applied to clinical saliva samples, ASO-assisted RCA markedly improved diagnostic performance, achieving 100% sensitivity and up to 97.5%-100% accuracy across multiple systems. These findings establish ASO-DNAzyme-RCA as a simple, robust, and clinically relevant platform for improving nucleic acid detection in structured RNA targets.

Keywords: Antisense oligonucleotides; Biosensors; DNAzymes; Rolling circle amplification; Structural elements.

Figure 1

Conceptual schematic of antisense oligonucleotide‐assisted rolling circle amplification (ASO‐RCA) for detecting structured genomic RNA. a) A 10–23 DNAzyme cleaves structured viral genomic RNA at a defined site, generating 5′ and 3′ cleavage products. b) In the absence of an antisense oligonucleotide (ASO), the 5′ cleavage product retains secondary structures that occlude the CDT‐binding region (shown in red), preventing effective hybridization with the circular DNA template (CDT) and thereby inhibiting RCA initiation. c) When a rationally designed ASO (magenta) is added, it binds upstream of the CDT‐binding site, disrupting inhibitory RNA structures and exposing the 3′ end for CDT hybridization. This structural remodeling enables efficient primer–template pairing and successful RCA initiation by phi29 DNA polymerase.

Figure 2

Secondary structure analysis of the ls584 RNA transcript (System 1) and the mechanism by which ASO1 enhances accessibility to the CDT1‐binding region. The full secondary structure of ls584, predicted using the Mathews Lab RNAstructure software, is shown at the top, with the CDT1‐binding region (Anti‐CDT1) in red and the ASO1‐binding region (Anti‐ASO1) in blue. The bottom‐left structure shows the DNAzyme dZ_12618a (shown in cyan) bound to ls584, cleaving the RNA between A523 and U524. The resulting 5′‐cleavage product (bottom center) retains significant secondary structure, with the Anti‐CDT1 region (red) still partially sequestered. Upon addition of ASO1 (magenta, bottom‐right), the upstream RNA structure is reorganized, freeing the Anti‐CDT1 region and enabling effective hybridization with the circular DNA template (CDT) for RCA initiation.

Figure 3

Native gel shift assay demonstrating that ASO1 enhanced hybridization between CDT1 and structured RNA1. In lane 6, CDT1 was incubated with 63 nt RNA1 alone, resulting in minimal complex formation, as indicated by the predominance of unshifted CDT1. In contrast, lane 7 showed that the addition of ASO1 promoted structural reorganization of RNA1, enabling efficient binding to CDT1 and resulting in a prominent shifted band. Corresponding secondary structure models (right) illustrate how ASO1 binding (magenta) exposed the Anti‐CDT1 region (red) by disrupting competing intramolecular structures formed by the Anti‐ASO1 region (blue), thus facilitating CDT hybridization.

Figure 4

ASO1 enhances the RCA signal following DNAzyme‐mediated cleavage of structured RNA (System 1) in both buffer and 50% pooled negative saliva. a) Real‐time RCA fluorescence curves show that ASO1 substantially increases signal output compared to reactions without ASO1. b) Relative fluorescence units (RFU) at 30 min (F ₃₀) normalized to the positive control (PC) in buffer. c) Fold enhancement in RFU by ASO1, calculated by dividing F ₃₀ with ASO1 by F ₃₀ without ASO1. Equivalent analyses in 50% negative pooled saliva. d) RCA fluorescence curves, e) relative F ₃₀ to PC, and f) fold enhancement with ASO1, showing signal improvements even in complex clinical matrices. g) Summary bar plots showing ASO‐mediated signal enhancement (F ₃₀‐fold change) across five different DNAzyme‐RCA systems in 50% pooled saliva, highlighting the general applicability of ASO‐assisted RCA for structured RNA detection.

Figure 5

ASOs improve sensitivity, specificity, and overall diagnostic accuracy in clinical SARS‐CoV‐2 detection using five DNAzyme‐RCA systems. a)–e) Bar plots showing F ₆₀ values (fluorescence at 60 min) for 20 positive samples (PS) and 20 negative samples (NS) tested with and without their respective ASOs for Systems 1–5. Each dotted line indicates the assay‐specific fluorescence cut‐off value used to distinguish positive and negative results. Inclusion of ASOs consistently elevated signal in PS and reduced background in NS, leading to improved assay performance. Error bars represent one standard deviation from triplicate measurements. NC, negative control (CDT only, no target RNA); PC, positive control (maximum RCA signal in buffer with ASO). f) Bar graph summarizing diagnostic accuracy (%) for each system with and without ASO assistance, highlighting notable improvements across all five systems.