Methods and Applications: Development of A Novel EvaGreen-Dye Based Recombinase Aided Amplification Assay Using Self-Avoiding Molecular Recognition System Primers

Standard strains of Bacillus cereus (ATCC13525) and Listeria monocytogenes (ATCC11778) were purchased from the BeNa Culture Collection. Nucleic acid was extracted using the FastPure^® Microbiome DNA Isolation Kit (Vazyme Biotech, Nanjing, China).

All intermediates and final products were characterized by ¹H NMR (Nuclear Magnetic Resonance) or ³¹P NMR, with spectral data consistent with previously reported findings (9). All oligonucleotides containing modified nucleobases were synthesized by Sangon Biotech.

The B subunit of the DNA gyrase gene (gyrB) and alkaline metalloprotease gene (aprX) were selected as specific targets for the identification of BC and PF, respectively. Highly conserved regions of gyrB and aprX genes were downloaded from NCBI GenBank and subjected to sequence alignment using BioEdit software. The designed primers and probes were evaluated using Oligo7 and Amplifx software. The specific sequences of primers and probes are presented in Table 1.

The target fragments of gyrB and aprX genes were cloned into pUC57 vector for the construction of recombinant plasmids, respectively, which were synthesized by TsingKe Biotech Crop (Beijing, China). Plasmids were quantified using the Qubit® dsDNA HS Assay Kit (Thermo Fisher Scientific, Waltham, MA, United States), and the copy number per microliter was calculated according to the equation: copies/reaction=[(6.02 × 10²³) × concentration (ng/μL) × 10^-9] / [number of bases (bp) × 660]. The lyophilized plasmids were reconstituted by adding 40 μL of TE buffer, quantified, and stored at –80 ℃ until use.

The RAA reaction was performed in a total volume of 50 μL, containing a dry powder pellet, 2 μL of RAA standard nucleotide (STD) primer, 4 μL of betaine, 0.6 μL of probe, 5 μL of template, and 29.4 μL of reaction buffer (Amp-Future, Changzhou, China). Additionally, 2.5 μL of magnesium acetate (280 mmol/L) was added to the reaction lid as an initiator for the RAA reaction, followed by brief centrifugation to ensure rapid reaction initiation at 39 ℃ for 30 min.

The EvaGreen RAA reaction differed from the standard protocol by the addition of 2.5 μL of EvaGreen dye, 2 μL of betaine, and 1.2 μL each of SAMRS forward and reverse primers (10 μmol/L). All other reagents remained identical to those described above.

Serial dilutions of recombinant plasmids (10¹–10⁶ copies/μL) were used as templates for probe-free RAA reactions with both STD and SAMRS primers for BC and PF at 39 °C for 30 min. Following amplification, products underwent a melting curve analysis from 60 to 95 °C with a temperature gradient of 0.2 °C/s. Fluorescence signals were detected in the SYBR green channel, and high-resolution melting curves (HRM) were generated by plotting the negative inverse of fluorescence values against temperature. In addition to HRM analysis, primer dimer formation was evaluated using agarose gel electrophoresis (AGE).

To validate the sensitivity of both RAA and EvaGreen RAA assays, serial dilutions of BC or PF recombinant plasmids (10⁰–10⁵ copies/reaction) were used as templates, with DEPC water serving as a negative control. Each reaction was performed in eight replicates.

The specificity of both assays was evaluated using 18 common foodborne pathogens, including BC (ATCC11778), PF (ATCC13525), Staphylococcus aureus (ATCC29213), Listeria monocytogenes (ATCC19115), Pseudomonas aeruginosa (ATCC27853), Klebsiella pneumoniae (ATCC11296), Escherichia coli (isolated strains), Staphylococcus epidermidis (isolated strains), Enterococcus faecium (isolated strains), Enterococcus faecalis (isolated strains), Enterobacter cloacae (isolated strains), Pseudomonas maltophilia (isolated strains), Proteus mirabilis (isolated strains), Candida albicans (ATCC753), Candida tropicalis (ATCC750), Candida parapsilosis (ATCC22019), Candida glabrata (ATCC2001), and Candida krusei (ATCC6258). DEPC water was used as a negative control, with each reaction performed in eight replicates.

Standard BC and PF strains were inoculated into Nutrient Broth (NB) and Luria-Bertani (LB) liquid media, respectively, and cultured overnight on a shaker. Bacterial suspensions were quantified spectrophotometrically, with OD values between 0.45–0.5 corresponding to approximately 10⁸ CFU/mL. The accuracy of spectrophotometric quantification was verified by direct microscopic enumeration using a bovine abalone blood counting chamber. Following sequential ten-fold dilutions, appropriate bacterial concentrations were added to sterilized milk to prepare simulated specimens. Nucleic acids were extracted from these specimens, and both fluorescent RAA and EvaGreen RAA assays were performed in parallel to determine their respective detection sensitivities.

The dual EvaGreen-RAA reaction system comprised 29.4 μL of buffer, 1 μL each of SAMRS forward and reverse primers for PF and BC, 2.5 μL of EvaGreen dye, 2 μL of betaine (5 mol/L), 2.5 μL of magnesium acetate (280 μmol/L), and 2.5 μL of templates (corresponding to 200 CFU/mL) for each BC and PF. The RAA reaction was initially performed at 39 ℃ for 30 min, followed by high-resolution melting curve analysis on a PCR instrument.

When STD primers were used in the RAA system, both high product peaks and non-specific primer dimer peaks at approximately 30–35 °C were observed (Figure 1A and 1E). In contrast, SAMRS primers produced only distinct high product peaks at 84 °C and 91 °C, with complete elimination of non-specific product peaks (Figure 1B and 1F). To validate these findings, amplification products from both methods were analyzed by agarose gel electrophoresis (3%). STD primers generated prominent dimer bands (Figure 1C and 1G), whereas SAMRS primers yielded no detectable dimer formation (Figure 1D and 1H). These results conclusively demonstrate that incorporating SAMRS primers into the RAA reaction system effectively inhibited or completely eliminated primer dimer formation.

Figure 1. 

Effect of SAMRS primers on primer dimerization. (A) Merged HRM results of the STD primer for BC; (B) Merged HRM results of SAMRS primers for BC; (C) Merged HRM results of the STD primer for PF; (D) Merged HRM results of SAMRS primers for PF; (E) AGE results of STD primers for BC; (F) AGE results of SAMRS primers for BC; (G) AGE results of STD primers for PF; (H) AGE results of SAMRS primers for PF.

Note: Lane M: Marker; Lanes 1–6: 10¹, 10², 10³, 10⁴, 10⁵ and 10⁶ copies/μL BC or PF recombinant plasmids.

Abbreviation: SAMRS=self-avoiding molecular recognition system; AGE=agarose gel electrophoresis; STD=standard nucleotide; HRM=high-resolution melting curves; NC=negative control; BC=Bacillus cereus; PF=Pseudomonas fluorescens.

The sensitivity of the RAA reaction was 10 copies/μL for both BC and PF (Figure 2A and 2B), while the EvaGreen-RAA reaction demonstrated enhanced sensitivity at 1 copy/μL for both pathogens using recombinant plasmids (Figure 2C and 2D). In simulated milk specimens, the detection sensitivity of the RAA assay for BC and PF was 400 CFU/mL and 600 CFU/mL (Figure 2E and 2F), respectively, whereas the EvaGreen-RAA assay exhibited superior sensitivity at 100 CFU/mL and 200 CFU/mL (Figure 2G and 2H), respectively. These results demonstrate that the utilization of SAMRS primers consistently achieves enhanced detection sensitivity compared to STD primers in equivalent assay conditions.

Figure 2. 

Sensitivity assessment of RAA and EvaGreen-RAA on recombinant plasmids. (A) The sensitivity of RAA for BC was 10 copies/μL; (B) The detection sensitivities of RAA for PF was 10 copies/μL; (C) The sensitivity of EvaGreen-RAA for BC was 1 copy/μL; (D) The detection sensitivities of EvaGreen-RAA for PF was 1 copy/μL. Detection of RAA and EvaGreen-RAA on simulated milk specimens; (E) The detection sensitivity of RAA for BC in simulated specimens was 400 CFU/mL; (F) The detection sensitivity of RAA for PF in simulated specimens was 600 CFU/mL; (G) The detection sensitivity of EvaGreen-RAA for BC in simulated specimens was 100 CFU/mL; (H) The detection sensitivity of EvaGreen-RAA for PF in simulated specimens was 200 CFU/mL.

Abbreviation: RAA=recombinase aided amplification; BC=Bacillus cereus; PF=Pseudomonas fluorescens.

Both the RAA and EvaGreen-RAA assays demonstrated high specificity for BC and PF detection, with no cross-reactivity observed against 16 other common foodborne pathogens (Table 2).

Amplification products with distinct nucleotide compositions exhibit characteristic melting temperature (Tm) values. High-resolution melting curve analysis of the dual EvaGreen-RAA amplification products revealed two clearly distinguishable product peaks at 84 °C and 91 °C (Figure 3). These distinct peaks corresponded precisely to the individual melting curves of BC and PF, demonstrating that the EvaGreen-RAA assay with HRM analysis can effectively identify and differentiate mixed BC and PF contamination in simulated milk specimens at concentrations as low as 200 CFU/mL for each pathogen.

Figure 3. 

Dual EvaGreen-RAA assay using melting curves for mixed BC and PF in simulated milk samples with a concentration of 200 CFU/mL for each pathogen.

Abbreviation: RAA=recombinase aided amplification; NC=negative control; BC=Bacillus cereus; PF=Pseudomonas fluorescens.

While probe-based RAA fluorescence assays demonstrate sensitivity and specificity comparable to real-time PCR, they require probes of considerable length (45–52 bp) with a tetrahydrofuran site positioned centrally within the sequence. These requirements substantially increase both the complexity of probe design and associated costs, particularly when targeting highly variable pathogens. Furthermore, multiplexed RAA reactions present significant challenges due to interference between different probe and primer sets within a single reaction vessel.

Previous investigations have demonstrated the utility of SAMRS in various isothermal amplification techniques, including recombinant polymerase amplification (RPA) and deconjugate-dependent amplification (HDA), where strategic placement of SAMRS nucleotides at the 3’ end of primer-binding sites effectively prevents the formation of most primer dimers (10–12). In contrast to previously reported approaches, our study retained the T base at the 3’ terminus of primers while selectively replacing only A, G, or C bases, thereby enhancing the stability of the 3’-end of the SAMRS primer.

Eva-green dye exhibits preferential binding to the minor groove region of all dsDNA double helices, offering advantages of enhanced sensitivity and cost-effectiveness. However, its non-specific binding capacity to DNA double strands, including primer dimers, can potentially generate false positive results (13–14). The integration of SAMRS primers in our study effectively addressed this non-specificity issue. Both AGE and melting curve analyses confirmed that SAMRS primers successfully eliminated primer dimer formation.

The workflow of the EvaGreen-RAA assay, illustrated in Supplementary Figure S1, demonstrates enhanced sensitivity and quantitative capability, as the amplicons generated by the reaction consist exclusively of target products, corroborating our most recent findings (15). Furthermore, the EvaGreen-RAA assay enables dual-target detection through HRM curve analysis by exploiting the differential Tm values of distinct amplification products. However, this study did not optimize primer concentration or the number of non-natural bases incorporated. An additional limitation concerns the persistent challenges in synthesizing non-natural bases. Despite these constraints, the EvaGreen-RAA assay provides a valuable platform for simple, rapid, and sensitive detection of pathogens of public health concern in primary laboratory settings.

The EvaGreen-RAA assay utilizing SAMRS primers demonstrates superior sensitivity and specificity for detecting B. cereus and P. fluorescens in under 30 minutes, eliminating the need for complex RAA probe design. This approach provides a novel strategy for multiplexed pathogen detection in isothermal amplification systems.

All the participants in this study.