Validation of AIOD-CRISPR detection system
To evaluate our AIOD-CRISPR assay, we prepared and tested different reaction systems with various components (Fig. 1). A plasmid containing a 1,200 bp fragment of the jujube witches’ broom (JWB) phytoplasma 16SrRNA gene was used as the target sequence. The activity and visual signal of the ssDNA-FQ were detected. Results showed that only reactions containing all components (target sequence, Cas12a, crRNAs and RPA Mix) produced a visibly obvious, bright fluorescence signal after incubation at 37 ℃ within 5 min (Fig. 1). The fluorescence signal was almost invisible when the detection system lacked any one component, such as LbCas12a, crRNA or target DNA (Fig. 1a). Real-time fluorescence showed a significantly increased fluorescence signal within 5 min only when all components were added to the reaction system (Fig. 1b). Results suggested that this AIOD-CRISPR assay can successfully detect the phytoplasma target nucleic acid within 15 min.
Optimization of AIOD-CRISPR assay
We first evaluated the effect of different recognition sites on the phytoplasma-detecting efficiency. Four reactions (Test 1-4) containing various RPA primers and crRNAs were prepared (Fig. 2a). In test 1, the RPA assay was carried out with the RPA-F1/RPA-R1 primer pair and the recognition of crRNA1 was not adjacent to the RPA primers recognition sites in the target sequence, so the crRNA1 then requires a PAM motif in the target sequence to active the Cas12a protein. Fluorescence signal was detected in this test, but a crRNA primer not dependent on a PAM motif is preferred. If the crRNA is adjacent to the RPA primer, no PAM motif is needed, as shown design for test 2, 3, and 4 in Fig. 2a. Test 2 contained two crRNAs adjacent to the RPA primers, test 3 and 4 each contained the single crRNA2 or crRNA3, respectively. All the sequences of crRNAs and their targets were shown in Fig. S2.
After incubation at 37 °C for 15 min, test 2 produced a visible and bright fluorescence signal, directly observable under UV light. The fluorescence signals of test 3 and 4 was difficult to distinguish from reactions lacking the target. All assays showed a change from colorless to green that was visible by the naked eyes 25 min after reaction, but test 2 showed an elevated fluorescence signal compared to test 3 and 4 (Fig. 2b) which was consistent with the visual detection results (Fig. 2c). These results indicated that the AIOD-CRISPR assay system can detect the phytoplasma target sequence and that the use of two crRNAs can greatly improve the intensity of the reaction when compared with those containing a single crRNA. Interestingly, test 1 showed a florescence intensity compared to test 3 and 4 (Fig. 2b), reaching 1500 RFU within 15 min and exceeding 4000 RFU within 25 min (Fig. 2c). Although the AIOD-CRISPR system can detect the phytoplasma successfully without the PAM sequence (Test 2), the PAM sequence does improve the detection efficiency by enhancing the activity of Cas12a when only one crRNA is used. In sum, test 2 offered the best detection system.
Based on the detection system with two crRNAs, we further explored the optimal contents of ssDNA-FQ, LbCas12a and crRNAs (Fig. 3). We determined the dosage of ssDNA-FQ that would create a visual fluorescence intensity detectable by both LED and UV visual detection and real-time fluorescence detection (Fig. 3a, 3b). The higher the concentration of the ssDNA-FQ reporter, the shorter the threshold time in real-time fluorescence detection (Fig. 3b). Consistent with the changes in absorbance, the color transition was distinctly observed when the dosage of ssDNA-FQ was 4,000 nM, 2,000 nM and 1,000 nM (Fig. 3a). The fluorescence signal can still be easily distinguished with the naked eye when the concentration of ssDNA-FQ was 1,000 nM. For better visual colorimetric detection, 1,000 nM ssDNA-FQ was the best choice.
After optimizing the concentration of the ssDNA-FQ, the concentration of the LbCas12a/crRNA complex was optimized in assays using ssDNA-FQ at the optimized dose (1,000 nM). For endpoint imaging and UV visual detection, the fluorescence brightness was remarkably changed when the ratio of LbCas12a/crRNA2/crRNA3 complex was 1.6 μM: 0.8 μM: 0.8 μM (Fig. 3c). The result was further confirmed by real-time fluorescence detection (Fig. 3d). Therefore, an optimized system was established containing 1,000 nM ssDNA-FQ and a 2:1:1 ratio of LbCas12a/crRNA1/crRNA2 complex with a 0.8 μM concentration as 1.
Sensitivity of the AIOD-CRISPR system
The minimum detection limit of the LbCas12a/crRNA assay was initially determined using 10-fold gradient dilutions of pEASY-16SrRNA clones. At the optimum reaction conditions, visible fluorescence signals were observed with around 3.37E+2 copies per reaction (Fig. 4b), reach the same level as PCR detection (Fig. 4a). Consistent with the color transition in the endpoint imaging, the fluorescence intensity was significantly higher than control when the copy number of the pEASY-16SrRNA clone was above 3.37E+2 in real-time fluorescence detection (Fig. 4c). In sum, we confirmed that the LOD of the AIOD-CRISPR assay for phytoplasma detection was as low as 3.37E+2 copies per reaction when using the optimized detection system.
Specificity and accuracy analysis of AIOD-CRISPR system for phytoplasma detection
To confirm the specificity of the AIOD-CRISPR detection assay, DNA samples suspected to carry phytoplasma from sixteen species (Trema tomentosa, Crotalaria juncea,Crotalaria pallida,Cajanus scarabaeoides,Justicia gendarussa,Vigna unguiculata,Arachis hypogaea,Solanum lycopersicum,Raphanus sativus,Parthenium hysterophorus, Catharanthus roseus, Triticum aestivum, Ziziphus jujuba, Paulownia fortunei, Prunus persica and Bambusoideae) were prepared for AIOD-CRISPR detection using the optimized reaction conditions. Healthy plant samples of these sixteen species were also collected as negative controls for AIOD-CRISPR detection. First, the DNA of the plant samples were used as templates for PCR to confirm the presence of phytoplasma. Twelve of the samples amplified a target fragment (1,300 bp) (S1, S3, S4, S5, S6, S7, S8, S9, S10, S11, S12, S14 and S16), while no amplified products were observed in their negative controls (Fig. 5a). All the amplification products were recovered for sequencing, and all of fragments corresponded to the phytoplasma 16sRNA sequence, indicating that they were infected with phytoplasma (data not shown). Then each DNA sample was subjected to the AIOD-CRISPR assay. All samples except S2, S13 and S15 showed obvious fluorescence in the optimal reaction system, while the fluorescence signal was almost invisible for the corresponding negative controls (Fig. 5b). The results were consistent with those of PCR detection. Results showed that the specificity of AIOD-CRISPR system for phytoplasma detection was very high and can be used to specifically detect phytoplasma in plant DNA extracts.
HS represent healthy samples of the species.
To evaluate the accuracy of the AIOD-CRISPR phytoplasma detection in the field, forty jujube seedlings were randomly sampled. The total DNA extracts were subjected to the AIOD-CRISPR system and PCR assay. For PCR-gel electrophoresis assay, the phytoplasma 16SrRNA universal primers were used (Chen et al.2022). Thirty-one samples were confirmed that to be infected by JWB phytoplasma (Fig. 6a), while samples 2, 3, 16, 18, 21, 33, 36, 37 and 38 were negative for phytoplasma 16SrRNA gene fragment. These forty JWB phytoplasma-infection samples were also selected to further evaluate the accuracy of the AIOD-CRISPR system, all of the samples except samples 2, 3, 16, 18, 21, 33, 36, 37 and 38 showed powerful fluorescence in the optimal reaction system (Fig. 6b), which was perfect consistent with that of PCR detection. The results indicated that the accuracy of AIOD-CRISPR system reached 100% and can be used for the accurate detection of phytoplasma in the field.
Diagram of AIOD-CRISPR detection system of phytoplasma
A schematic of the visual detection system for phytoplasma in plant tissues and based on an All-in-one dual (AIOD)-CRISPR/LbCas12a platform is generated (Fig.7). Within the AIOD-CRISPR platform, the Cas12a-crRNA complexes are prepared prior to being adding into the reaction solution containing RPA primers, single-stranded DNA fluorophore-quencher (ssDNA-FQ) reporter, RPA Mix and sample DNA. All components were added together and incubated in one tube at 37 °C. On one hand, once the phytoplasma is existed in the reaction, the crRNAs immediately bind the target DNA, which activates the Cas12a endonuclease that cleaves the ssDNA-FQ reporters to produce fluorescence. RPA products also provide additional binding sites for the Cas12a-crRNA complexes, effectively amplifying the process and the fluorophore signal through continuous triggering of CRISPR-Cas12a-based collateral cleavage activity. On the other hand, if there is no phytoplasma in the reaction, crRNAs loss the target, which will not activates the Cas12a endonuclease and the ssDNA-FQ keeps stable. No fluorescence is produced in these reactions (Fig. 7).