LwCas13a protein and its corresponding buffer were purchased from Magigen Biotech Co. Ltd. (Guangzhou, China). DNase I, T7 RNA polymerase, and recombinant RNase inhibitor were obtained from TaKaRa (Dalian, China). ChamQ Universal SYBR qPCR Master Mix was purchased from Vazyme (Guangzhou, China). Sangon Biotech (Shanghai, China) provided the NTP mixture (10 mM each), dNTP mixture (25 mM each), and Spin Column RNA Cleanup & Concentration Kit. DNA Constant Temperature Rapid Amplification Kit (Liquid Basic Type) and CRISPR Cas12/13 Hybridetect Test Note were purchased from Warbio Biotech (Nanjing, China).

In our laboratory, P. aeruginosa was isolated from spoiled wood paint lotion. Other pathogenic bacteria obtained from American Type Culture Collection (ATCC) and German Collection of Microorganisms and Cell Cultures included Staphylococcus aureus ATCC 6538P, Escherichia coli ATCC 8739, Vibrio parahaemolyticus ATCC 17802, Salmonella enterica ATCC 9115, Klebsiella pneumoniae ATCC 4352, Lactobacillus paracasei ATCC 334, Bacillus paralicheniformis ATCC 9945, and Bacillus subtilis DSM 23778. The nucleic acids of the above strains were extracted using the HiPure Bacterial DNA Kit (Magen Biotech Co., Ltd.; Guangzhou, China) according to the manufacturer’s instructions. The concentration of DNA was quantified using the Nanodrop One instrument (Thermo Fisher Scientific, Shanghai, China). Until they were used, all DNA samples were stored at-20°C.

The mexX gene sequences of P. aeruginosa were downloaded from the GenBank database of the National Center for Biotechnology Information (NCBI). ¹ After aligning gene sequences to identify conserved regions, 10 crRNAs were designed using the online Cas13 design tool. ² According to the selected target sequence design and relevant works of literatures (Liu et al., 2017; Gootenberg et al., 2018; Yang et al., 2023), a T7 promoter binding sequence was attached to the 5′-end of the designed crRNAs. Simultaneously, the PRIMER 5 program was used to design equivalent RPA primers between the upstream and downstream clips of the candidate target DNA. The RNA probe 1 is tagged with 6-FAM at the 5′ -end and uses BHQ1 at the 3′ -end (Zhou et al., 2020), whereas RNA probe 2 is designated with 6-FAM at the 5′ end and uses BIO at the 3′ end (Wang et al., 2023). The designed primers, probes, and oligonucleotides were synthesized by TsingkeBiotech (Beijing, China). All of the nucleic acid sequences used in this study are listed in Table 1.

The templates for an oligonucleotide incorporating T7 promoters, repetitive sequences, and interval sequences have been developed to create the double strand DNA (dsDNA). Forward and reverse oligonucleotide DNA (5 μmol/L) were added to produce dsDNA templates for transcription. And each single-chain template was then mixed, annealed for 5 min at 95°C and naturally cooled to room temperature naturally. The crRNA was synthesized by incubating the above mixtures for 2 h at 42°C with T7 RNA polymerase (TaKaRa). To carry out the process, the synthesized crRNA was digested with DNase I (TaKaRa) at 37°C for 1 h to remove DNA templates, and then purified using the RNA Rapid Concentration and Purification kit (Sangon Biotech, Shanghai, China) in accordance with the manufacturer’s instructions. The crRNA concentration was then determined using the NanoDrop spectrophotometer (Thermo Fisher Scientific, Shanghai, China) and was subsequently kept at-80°C until use.

For the one-tube RPA-Cas13a assay, RPA amplification and CRISPR-Cas13a were coupled in a one-tube reaction system (Figure 1). Briefly, the 50 μL one-tube reaction system consisted of 18 μL C buffer (Warbio Biotech Co. Ltd.; Guangzhou, China), 5 μL L buffer (Warbio), 12 μL P-core (Warbio), 1 μL each of 10 μM forward and reverse primers, 1.5 μL 10 mM dNTPs, 1 μL target DNA template, 2 μL 25 mM NTP mix, 1 μL recombinant RNase inhibitor, 1 μL 10 μM RNA-probe, 1 μL T7 RNA polymerase, 0.4 μL 2 μM Cas13a, 1 μL 12 nM crRNA, and added RNase-free ddH₂O to 50 μL. After gently spinning the reaction mixtures, 2.5 μL of 280 mM magnesium acetate (MgOAc) was added. Finally, the reactions were then run at 39°C for 20 min, and the fluorescence signal was collected every 1 min on a QuantStudio™ Real-Time PCR Software (Applied Biosystems, Waltham, MA, United States) with three replicates set for each sample.

In the two-step RPA-Cas13a assay, the crucial RPA reactions were first conducted according to the instructions of the DNA constant temperature rapid amplification kit (Warbio, China; Figure 1). Each 50 μL RPA reaction volume contained the following elements: 20 μL C buffer, 5 μL L buffer, 12 μL P-core, 2 μL each of 10 μM forward and reversed primers, 1.5 μL 10 mM dNTPs, 1 μL target DNA template, and 4 μL RNase-free ddH₂O. The reaction mixtures were gently vortexed and spun. Then, 2.5 μL of 280 mM MgOAc was added and thoroughly mixed to start the reaction, which would last for 30 min at 39°C. Subsequently, a 50 μL CRISPR-Cas13a reaction system was performed with 5 μL 10 × reaction buffer, 2 μL 25 mM NTP mix, 0.8 μL recombinant RNase inhibitor, 1 μL PA-crRNA, 1 μL 10 μM RNA-probe, 0.6 μL T7 RNA polymerase, 0.4 μL 2 μM Cas13a, 4 μL RPA products and added RNase-free ddH₂O to 50 μL. Three replicates were set up for each sample, and the reactions were carried out at 39°C for 20 min after mixing. The fluorescence signal was collected every minute using the QuantStudio^™ Real-Time PCR program.

The crRNA 1 with the highest predicted score in online website design was selected for the detection system to confirm the activity of the LwCas13a protein and examine if a detection system could be successfully established. In the first, second, and third groups, equivalent amounts of RNase-free H₂O were introduced to the reaction system in place of crRNA 1, DNA, and LwCas13a, which served as negative controls. Instead of DNA, crRNA 1, and LwCas13a protein, the blank control group received an equal amount of RNase-free H₂O. After mixing, the reactions were run for 20 min at 39°C, and the fluorescence signal was collected every 1 min on a QuantStudio^™ Real-Time PCR Software with three replicates.

To ensure the excellent activity of the Cas13a protein, 10 crRNAs were obtained via the online Cas13 design tool. And three pairs of RPA isothermal amplification primers were designed based on the conserved sequence to match the crRNA. Using the same batch of extracted DNA products, the CRISPR-Cas13a system, which contains 10 distinct crRNAs, was next tested for detection efficiency. The endpoint fluorescence value was recorded for comparison.

The Cas13a-crRNA-fluorescent measurement signal was used to optimize the detection system, and the fluorescence value was simultaneously selected for comparison. The samples for the optimization procedure are P. aeruginosa DNA samples with the same initial concentration. A single variable should be controlled to optimize the following parameters: reaction temperatures (37, 38, 39, 40, and 41°C), primer concentrations (0.6–1 μM), LwCas13a concentrations (0–40 nM), and crRNA concentrations (0–48 nM).

To compare the minimal LoD of these two approaches, the extracted P. aeruginosa nucleic acids are employed as a template to dilute a series of gradients 10 times, resulting in distinct concentration standards. The gradient diluted dsDNA standards were detected using the one-tube and two-step RPA-Cas13a methods. We used qRT-PCR as the reference test, which was conducted according to the guidelines provided in the following section. The specificity of the one-tube and two-step RPA-Cas13a methods was tested by using the genomes extracted from P. aeruginosa, S. aureus ATCC 6538P, E. coli ATCC 8739, V. parahaemolyticus ATCC 17802, S. enterica ATCC 9115, K. pneumoniae ATCC 4352, L. paracasei ATCC 334, B. paracaeniformis ATCC 9945, and B. subtilis DSM 23778.

In our laboratory, 38 P. aeruginosa were previously isolated from industrial products (spoiled daily chemical products and coatings; Supplementary Table 1). To validate the usefulness and practicality of the RPA-Cas13a assay, nucleic acids from these real specimens were extracted and detected using the one-tube and two-step RPA-Cas13a assay methods mentioned above. Meanwhile, qRT-PCR and PCR were also employed to detect these samples, and the results were finally compared with those obtained using RPA-Cas13a.

After a 10-fold gradient dilution, the DNA template was detected using a fluorescence quantitative PCR kit (TaKaRa). The reaction mixture contains 10 μL 2 × ChamQ Universal SYBR qPCR Master Mix, 0.4 μL of each primer pair at 10 μM, 1 μL template DNA, and ddH₂O to 20.0 μL. qRT-PCR was performed on the QuantStudio^™ Real-Time PCR software instrument, and the reaction was run as follows: 95°C for 30 s, followed by 40 cycles of 95°C for 10 s, and then 60°C for 30s.

For comparison with RPA-Cas13a, the traditional PCR was also used to amplify the targeted mexX gene from the strains used in this study. The following steps were taken to prepare the PCR reaction system: 5 μL 2 × Rapid Taq Master Mix (Vazyme, Nanjing, China), 0.2 μL of each primer (10 μmol/L), 4.4 μL ddH₂O, and 0.2 μL template DNA. The mixtures were subsequently subjected to 30 cycles of PCR using a PCR instrument (T100 Thermal Cycler; Bio-Rad Laboratories) with the following reaction parameters: 95°C for 5 min, 95°C for 15 s, annealing at 57°C for 15 s, and 30 s extending at 72°C. One more extension cycle at 72°C for 5 min was needed to complete the reaction.

The practical workflow for this test consists of three steps. First, the gene was amplified using RPA and translated to ssRNA. Then, this nuclease recognized crRNA and cut off the reporter molecule using its collateral cutting capability. Lastly, lateral flow dipstick assays were performed using the identical reaction components as the fluorescent detection assays, but using 400 nM FAM-20 U-Biotin instead of the fluorescent reporter. And the reaction was incubated for 30 min at 39°C. A lateral flow dipstick (Warbio, Nanjing, China) was then inserted into the tube and incubated at room temperature for 5 min, during which the result was recorded. It was noteworthy that, whereas negative samples should only have a C line, positive samples should display two lines (T and C lines).

The Cas13a-crRNA-fluorescence test signals were recorded and expressed as the mean of at least three independent reactions plus standard deviation (SD). GraphPad Prism 9.0 was used to do the analysis of variance and create the figures.

To verify whether the LwCas13a protein exhibited the expected ribonuclease activity, a comprehensive Cas13a reaction system is developed using positive genes as the detection template. Additionally, in order to quantify fluorescence kinetics, negative and blank control groups were set up. The results demonstrated that when any of the following elements were not present in the detection system: LwCas13a, crRNA 1, or DNA/RPA product, the fluorescence signal remained essentially unchanged (Figure 2). When LwCas13a, crRNA 1, and DNA/RPA products were all present at the same time, the fluorescence signal increased noticeably, indicating that the Cas13a protein was performing its collateral cleavage activity in this circumstance (Figure 2). The fluorescence signal developed quickly over time and it only took 5 min for a single tube to react (Figure 2A), whereas the two-step method took roughly 10 min (Figure 2B). Following that, whether utilizing a single tube or a two-step procedure, the fluorescence kinetic curves approached a plateaued stage (Figure 2). These results demonstrated that the LwCas13a protein can precisely recognize target RNA and exhibits the anticipated auxiliary cleavage activity to cleave the RNA fluorescence probe in the system, resulting in a signal with high fluorescence intensity. The one-tube and two-step Cas13a fluorescence detection systems based on RPA and Cas13a have been effectively constructed.

According to the earlier researches, crRNA is the primary factor affecting Cas13a cleavage activity (Liu et al., 2017; Yang et al., 2023). To improve the cleavage activity of LwCas13a, 10 crRNAs targeting the mexX gene were designed based on the Cas13a protospacer flanking site (PFS) motif. To identify the optimal crRNA for further research, the one-tube and two-step RPA-Cas13a fluorescence assays were performed. The fluorescence signals were recorded at 1 min intervals using a QuantStudio™ Real-Time PCR. The results demonstrated that the activity of the 10 different crRNA candidates is much higher than the no-target control (Figure 3). Within 10 min, all crRNAs produced fluorescence signals. However, crRNA 7 generated the maximum fluorescence intensity in both one-tube (Figure 3A) and two-step (Figure 3B) RPA-Cas13a detection systems, implying that crRNA 7 had enhanced activity for target DNA detection. Therefore, crRNA 7 was chosen as the optimal crRNA for the subsequent one-tube and two-step RPA-Cas13a detection optimizations.

In the one-tube RPA-Cas13a detection system, we first evaluated the effect of temperatures (37, 38, 39, 40, and 41°C) on fluorescence excitation of probes and discovered that 39°C exhibited the greatest fluorescence signals (Figure 4A). Then, the best fluorescence signal was obtained at an RPA primer concentration of 0.4 μM at 39°C conditions (Figure 4B). At 39°C and 0.4 μM primers, the optimum LwCas13a concentration was then determined. The results demonstrated that raising the LwCas13a concentration increased the fluorescence signal from 0 to 16 nM. There was only a minor fluorescence enhancement at LwCas13a concentrations greater than 16 nM (Figure 4C). Finally, we investigated the ideal crRNA concentration at 39°C, 0.4 μM primer, and 16 nM LwCas13a. Despite using less crRNA, the fluorescence generated is equivalent to 48 nM. Thus, the optimal crRNA concentration was determined to be 36 nM (Figure 4D). As a result, for one-tube, the ideal RPA-Cas13a conditions were 39°C, 0.40 μM primer, 16 nM LwCas13, and 36 nM crRNA concentration.

We first investigated the ideal reaction temperature for the two-step RPA-Cas13a reaction using a 0.40 μM primer concentration. The fluorescent signal reaches its maximum at 39°C, which was considered as the optimal temperature (Figure 5A). At 39°C, the ideal primer concentration was 0.2 μM (Figure 5B). Meanwhile, the fluorescence signal was enhanced at LwCas13a concentration between 0 and 16 nM, and this value hardly changed when the Cas13a concentration was over 16 nM (Figure 5C). Finally, at 39°C, 0.2 μM primer concentration, and 16 nM LwCas13a concentration, we explored the optimal crRNA concentration. It generated fluorescence comparable to the concentration range of 24–48 nM (Figure 5D). Thus, we determined that 24 nM was the optimal crRNA concentration (Figure 5D). Consequently, the optimal parameters for the two-step RPA-Cas13a were 39°C, 0.2 μM primer concentration, 16 nM LwCas13a, and 24 nM crRNA.

The DNA at a starting concentration of 100 pM was diluted 10 times with RNase-free H₂O to generate the detection concentration gradient. The sensitivity of one-tube and two-step RPA-Cas13a for P. aeruginosa was assessed using this gradient. The results demonstrated that the LoD for one-tube RPA-Cas13a was 10 aM (Figure 6A), slightly less than the LoD for two-step RPA-Cas13a (1 aM; Figure 6B). However, qRT-PCR exhibited a higher LoD of 100 aM (Figure 6C).

In nucleic acid testing, specificity is a fundamental issue: a suitable level of specificity improves detection accuracy by lowering false positive results and inaccurate conclusions (Antropov and Stepanov, 2023). Under optimal conditions, a particular experimental investigation was carried out to assay the degree of specificity of the RPA-Cas13a nucleic acid detection method. Only P. aeruginosa displayed detection signals among the nine common bacteria for specificity assessment without cross-reactions whether using one-tube (Figure 7A) or two-step RPA-Cas13a (Figure 7B). A comparable degree of specificity was also shown by qRT-PCR (Figure 7C). Thus, the target gene of mexX can be precisely identified using one-tube or two-step CRISPR/Cas13a-based detection method, which exhibits the same specificity as qRT-PCR.

The one-tube and two-step RPA-Cas13a methods were used to select and detect a total of 38 actual samples from various industrial products in order to assess the applicability of these two detection methods for P. aeruginosa in real samples. The results demonstrated that 38 samples, using both one-tube (Figure 8A) and two-step RPA-Cas13a (Figure 8B), tested positive for the target gene of mexX. Additionally, qRT-PCR (Figure 8C) and traditional PCR (Figure 8D) also demonstrated the same efficiency. We conclude that the one-tube and two-step procedures exhibit high agreement with qRT-PCR and traditional PCR, suggesting that these two RPA-Cas13a detection methods could be used to real samples.

To assess if the RPA-Cas13a detection method established in this study can directly apply the commercialized LFD to detect its results, a more practical and visually appealing detection strip method called RPA-Cas13a-LFD was developed. The results indicated that the initial template concentration affected the appearance of two lines and the LoD of the two-step RPA-Cas13a was 10 fM (Figure 9). The investigation demonstrated that commercial flow strips can be used to identify diagnostic results.