SEA detection kits were obtained from Qingdao Navid Biotechnology Co. Ltd. (China). Twenty bp DNA marker, 6 × DNA loading buffer and sodium dodecyl sulfate (SDS) were purchased from Sangon Biotech (Shanghai, China). Acrylamide and methylene diacrylamide were purchased from Sigma-Aldrich (St Louis, MO, USA). Recombinant DNase I was purchased from Takara Bio (Beijing, China). RNA-Be-Gone was purchased from Sangon Biotech (Shanghai, China). The bacterial strains including S. aureus, Listeria monocytogenes (L. monocytogenes), Salmonella typhimurium (S. typhimurium), Shigella castellani (S. castellani), Vibrio parahemolyticus (V. parahemolyticus), and Escherichia coli (E. coli) were stored in our laboratory.

A pair of specific primers (Table 1) were designed in the hypervariable region of S. aureus 16S rDNA with the NUPACK software (http://www.nupack.org/) and DINAMelt Web Server (http://unafold.rna.albany.edu/?q=DINAMelt). The primers were synthesized and purified by high performance liquid chromatography (HPLC) in Sangon Biotech (Shanghai, China).

The genomic DNA of S. aureus was extracted with TIANamp bacteria DNA kits (Tiangen Biotech, Beijing, China) according to the literature (Ulrich and Hughes, 2010). The target for colorimetric assay, specificity and anti-interference experiments. After enrichment culture for 24 h at 37°C, 1 mL bacterial liquid was heated at 95°C for 5 min, then centrifugated. One micro liter supernatant was used as target for colorimetric assay, specificity and anti-interference experiments.

The SEA reaction was performed according to the manufacturers' instructions under optimal conditions. The fluorescence signal of the SEA reaction was detected by the Gentier 48S Isothermal amplification fluorescence detection system at 1 min intervals. Gel images were recorded by the ChampGel 5,000 system (Saizhi Innovation Technology Co. Ltd, Beijing, China). The reaction temperature was optimized at 57°C, 58.2°C, 60.5°C, 62°C, 63.5°C, 65.3°C, 66.6°C, and 67°C, respectively.

The genomic DNA of S. aureus was used to demonstrate the feasibility of the SEA method. The sensitivity of the SEA detection method for S. aureus was evaluated with different dilutions of genomic DNA and DNA fragments. The culture fluids of S. aureus with L. monocytogenes, S. typhimurium, V. parahemolyticus, S. castellani, and E. coli were used to demonstrate the specificity and anti-jamming capacity of the SEA method.

Colorimetric kits were obtained from Qingdao Navid Biotechnology Co. Ltd. (China). Rapid nucleic acid extraction from bacterial fluids was used to demonstrate the feasibility of SEA colorimetric method.

The beef, pork, chicken, dried fish, and ham sausage real samples were collected from the small farmers markets (Qingdao, China). Samples were divided into 25 g each with a sterile knife. Then each sample was transferred to 225 mL 7.5% NaCl and then treated according to the literature with slight modifications (Abdalhai et al., 2014). After enrichment at 37°C for 24 h, 1 mL enrichment solution was taken from each sample for SEA and traditional plating detection, respectively.

SEA was based on the single-stranded denaturation bubbles of dsDNA at the reaction temperature, including a pair of primers, Bst DNA polymerase, and a constant temperature. The specific primers (P1 and P2) were designed with 50 bp amplification fragment in the hypervariable V2 region of 16S rDNA (Table 1), which has a high copy number in bacteria. Briefly, one pair of specific primers bound with the targeted DNA fragments by invading the denaturing bubbles and induced DNA polymerase to extend the chain. In addition, according to the Tm values of primers, the reaction temperature was optimized using the SEA kit with S. aureus genomic DNA as the template. Finally, 62°C was chosen as the optimal reaction temperature (Figure S1) in this work.

Genomic DNA of S. aureus was used as the template to demonstrate the feasibility of SEA method to detect S. aureus. As shown in Figure 1, compared with the no target control (NTC) group, the fluorescence signal with the addition of genomic DNA significantly increased, which indicated that the SEA method could effectively detect genomic DNA of S. aureus. This result was consistent with the targeted 50 bp amplification products in the gel electrophoresis (Figure 1 inserted), which further confirmed the feasibility of SEA method to detect S. aureus.

To evaluate the sensitivity of SEA detection method for S. aureus (Zhang et al., 2018), different concentrations of DNA fragments and genomic DNA of S. aureus were detected. As shown in Figure 2A, the fluorescence signals gradually increased with the increasing concentrations of targeted DNA fragments ranging from 1.0 × 10⁻¹¹ M to 1.0 × 10⁻¹⁴ M. In addition, as shown in Figure 2B, the threshold time (Tt) value increased linearly with the increasing negative logarithm (lg) value of concentrations of S. aureus DNA fragments, ranging from 1.0 × 10⁻¹¹ M to 1.0 × 10⁻¹⁴ M. The regression equation was Tt = 13.299 (-lgC_DNA)−134.35 (C_DNA was the concentration of S. aureus DNA fragments, R² = 0.9796). Moreover, SEA was also carried out using 10-fold serial dilutions of genomic DNA extracted from S. aureus. The results showed that SEA could be used to detect the concentration of S. aureus genomic DNA as low as 400 pg/μL (Figure 2C). As shown in Figure 2D, the Tt value increased linearly with the increasing lg of S. aureus genomic DNA concentrations in the range from 40 ng/μL to 400 pg/μL, which yielded a correlation equation of Tt = 13.32 (-lgC_DNA)+49.68 (C_DNA was the concentration of S. aureus genomic DNA, R² = 0.9993). In conclusion, the SEA method showed good linearity and sensitivity both in detecting genome DNA and DNA fragments.

RNA is readily degradable in the environment of pathogens in vitro, a biomarker of live bacteria, has been considered as a more suitable target for live bacteria detection. Therefore, using the SEA method to detect RNA and DNA in one step can not only take advantage of the abundance of RNA in live bacteria, but also make up for the low sensitivity of DNA detection. In this assay, total S. aureus nucleic acid (DNA and RNA), DNA and RNA were used as targets, respectively, to elucidate the nucleic acid detection efficiency of the SEA method. As shown in Figure 3, the SEA method could realize the detection of viable S. aureus by one-step detection of RNA by the fluorescence signal. We also found that the occurrence of the fluorescence signal was significantly delayed with DNA or RNA elimination, which indicated that detecting RNA and DNA simultaneously with the SEA method showed higher sensitivity than that of DNA or RNA independently. This result further verified that SEA method can not only detect DNA, but also RNA by one-step.

In addition, the targeted RNA fragments are relatively short in the SEA assay, so that both RNA and most of its incompletely-degraded RNA fragments could be used as amplification templates. This would undoubtedly improve the stability and sensitivity of SEA method for RNA detection.

Considering the wide existence in nature and severe effects of S. aureus, we have to make surveillance on it in various foods to ensure our health. At present, most rapid detection methods of S. aureus depend on complex and large-scale precision instruments, so that they are not suitable for developing regions. Therefore, it is necessary to develop a simple and rapid detection method for S. aureus. We developed a colorimetric method combined with SEA to detect S. aureus. As shown in Figure 4, the reaction mixture color could change from light yellow to red for positive samples and stayed light yellow for negative samples. Most importantly, the whole colorimetric detection procedure took no more than 50 min and was read out by the naked eyes, which was extremely simple and convenient. As shown in Figure 4 inset, the positive group changed its color from light yellow to red, while the NTC group remained the original yellow, which was consistent with the fluorescence results (Figure 4). Consequently, colorimetric detection results of the SEA method could be readily observed by the naked eyes instead of using the large and costly fluorescence detection equipment (Wang et al., 2014). In a word, the colorimetric SEA results could be performed without any complicated detection instruments or well-trained staff, which was especially applicable for the field detection of S. aureus in resource-limited environments.

Five other bacterial fluids, including L. monocytogenes, S. typhimurium, S. castellani, V. parahemolyticus, and E. coli were used to verify the specificity and anti-interference of the SEA method (Yang et al., 2016). As shown in Figure S2, the fluorescence signal only occurred in detecting S. aureus and no obvious effects were observed with the existence of other bacterial mixture, showing the good specificity and anti-interference of the SEA method. Thereafter, we carried out further artificially S. aureus-contaminated pork tests. The results showed that the detection limit of the SEA method on S. aureus was 10⁰ cfu/g (Figure S3), which met the requirements that S. aureus must not be detected out in foods. Furthermore, different meat real samples collected from the small farmers market were detected by the traditional plating method and SEA method for S. aureus simultaneously. As shown in Figure 5, the SEA method has a great advantage over the traditional plating method in detection time. The SEA method took only 1–2 h to detect real samples simply and rapidly, however, the traditional plating method need to spend as long as 72 h. In addition, positive rates of the SEA detection (Table 2) for S. aureus among those samples were consistent with that of the traditional plating method.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.