The novel coronavirus (CoV) disease termed COVID-19 (coronavirus disease-19) caused by SARS-CoV-2 (severe acute respiratory syndrome coronavirus-2)1 is causing a massive pandemic worldwide, threatening public health systems across the globe. During this ongoing COVID-19 outbreak, nucleic-acid detection has played an important role in early diagnosis2. To date, four protocols based on CRISPR for detecting SARS-CoV-2 have been published3,4,5,6. Using lateral flow protocols, RNA samples harboring more than 1 × 104–1 × 105 copies/mL (SHERLOCK) or 1 × 104 copies/mL (DETECTR) can be detected within 1 hour. In addition to these reported efforts, we have also established a SARS-CoV-2 detection protocol based on our previously reported platform—CDetection (Cas12b-mediated DNA detection)7. By combining sample treatment protocols and nucleic-acid amplification methods with CDetection, we have established an integrated viral nucleic-acid detection platform—CASdetec (CRISPR-assisted detection). The detection limit of CASdetec for SARS-CoV-2 pseudovirus is 1 × 104 copies/mL, with no cross-reactivity observed. Here, we present our assay design and optimization process, which could provide guidance for future CRISPR-based nucleic-acid detection assay development and optimization.
To optimize the output of fluorescence signal, we designed and synthesized poly-T fluorescence-quenchers of varying nucleotide lengths, including 4 nt, 5 nt, 7 nt, 12 nt, 17 nt, 22 nt, and 27 nt. Of all the lengths tried, the 7 nt poly-T reporter provided the highest signals in the shortest time (Supplementary Fig. S1a, b). Based on this observation, we applied the 7 nt poly-T reporter in later experiments.
According to the published SARS-CoV-2 whole-genome sequence8, we designed seven sgRNAs around the RdRp locus, as recommended by World Health Organization (WHO)2 (Supplementary Fig. S2a and Table S1). Owing to its high similarity to SARS-CoV, we ran initial experiments on both SARS-CoV-2 and SARS-CoV plasmids. According to fluorescence kinetics studies, sgRNA-3 stood out in not only being able to distinguish between the two similar coronaviruses but also being able to produce the most distinct fluorescence signal (Fig. 1a and Supplementary Fig. S2b–g).
As previously demonstrated, CRISPR is unable to detect any target DNA when there is <1–10 nM of amplification product within the reaction mix7. Hence, increasing the molecular collisions between CRISPR and target would be essential to improve sensitivity. We found out that increasing the sgRNA concentration by threefolds not only enhances the fluorescence signal and the signal-to-background ratio, but also increases the rate of reaction (Fig. 1b and Supplementary Fig. S3a, b).
Given that the average viral load in the plasma of SARS patients ranged from <1 to ~1000 copies per microliter9, or 1 × 103–1 × 106 copies/mL, nucleic-acid amplification techniques are needed to produce sufficient DNA for CRISPR-based DNA detection methods. Recombinase-aided amplification (RAA) can amplify substrates 1010 times at most (from 1 am to 10 nM) within 10–30 minutes at constant temperature between 37 °C and 42 °C, complementing the needs of CRISPR-based detection. Thus, we designed and screened RAA primers that matched our previously optimized sgRNA-3 (Supplementary Fig. S4a and Table S1). Based on our screenings, we found that by using the best primer pairs together with sgRNA-3, we can detect SARS-CoV-2 RNA samples containing 5 × 103 copies/mL (Fig. 1c and Supplementary Fig. S4b, c).
In addition, we analyzed all SARS-CoV-2 sequences that have been uploaded to GISAID up till March 26th 2020. Out of 1792 sequences on GISAID, 1673 of them contained sequences matching our chosen primers and sgRNA. Only three of them have one mismatch to the forward primer and only two of them have one mismatch to the reverse primer (Supplementary Fig. S5), suggesting that our selected sgRNA and primers can be used for nearly all of the reported SARS-CoV-2 genomes. Meanwhile, we aligned the selected primers and sgRNAs to twelve typical human coronaviruses to evaluate their specificity, and found that none of the whole set of primers and sgRNA showed high similarity (Supplementary Fig. S6a–c).
The experiments above were conducted by executing RT-RAA nucleic-acid amplification and CDetection separately. However, it would be best to conduct both reactions within a single tube for convenience and, more importantly, to prevent aerosol contamination, which happens when the reaction mixture has to be exposed to the environment midway through the protocol. Hence, we tried to execute both the RT-RAA and CDetection concurrently within a single tube. However, the combination resulted in a drastic decrease in sensitivity (Fig. 1c and Supplementary Fig. S7). Therefore, in order to keep both the RT-RAA and CRISPR reactions within a single tube, we executed the RT-RAA reaction within the tube while keeping the CDetection reagents within the cap of the tube for 30 minutes, following which, the CDetection reagents were spun down into the tube for nucleic-acid detection, and the resultant reaction mixture was imaged for fluorescence.
To validate the specificity of our method for SARS-CoV-2 nucleic-acid detection, we tested our protocols against six coronaviruses known to cause respiratory diseases (SARS-CoV, MERS-CoV, CoV-HKU1, CoV-229E, CoV-OC43, and CoV-NL63). Consistent with alignment analysis (Supplementary Fig. S6a–c), no cross-reactivity with other endemic human coronavirus were detected (Fig. 1d). Our results suggested our set of sgRNA and primers showed high sensitivity and specificity.
However, viral genomes are packaged inside capsid protein and need to be released. Thus, to investigate the virus handling process, we produced pseudoviruses by packaging the target sequences of SARS-CoV-2, SARS-CoV, MERS-CoV into actual lentivirus particles (Supplementary Fig. S8). These pseudoviruses were diluted serially and treated with either virus genome extraction kits (spin column) or lysis buffer, respectively. Our results demonstrated that the spin column treatment gave a lower detection limit of 1 × 104 copies/mL (Fig. 1e). On the other hand, the lysis buffer offered higher usability, but raised the detection limit to 5 × 104 copies/mL (Fig. 1f). Owing to the difference in detection limit between spin column and lysis buffer, we suggest using the spin columns in hospitals, and using the lysis buffer for point-of-care testing (POCT).
To make CASdetec more amenable for POCT, we have also constructed a portable dark box containing a blue LED and demonstrated that the positive fluorescence signal generated from the protocol can be visualized upon illumination by a blue LED (Fig. 1g). The detection limit was deceased under blue LED compared with quantitative PCR instrument. Optimizing the extensity of excitation light and the sensitivity of optical detector may increase the detection limit of optical visualization. What’s more, a more impersonal readout method, such as taking photos followed by mobile software-mediated analyses, can be developed to analyze the optical results.
In conclusion, we have established a CASdetec platform, which consists of procedures including virus handling, nucleic-acid amplification, and CRISPR-based detection (Fig. 1h). CASdetec can detect pseudovirus samples with >1 × 104 copies/mL, with no cross-reactivity to other endemic human coronaviruses. In addition, we optimized the workflow to run both reactions within one single tube without lid opening. This will thus prevent aerosol contamination and reduce the false positive rate. To guarantee the accuracy of CASdetec, efforts should be taken to avoid nucleic-acid aerosol contamination, including (1) conducting reagent preparation, sample treatment, amplification, and product analyzation in separate rooms and (2) being careful on handling samples and reagents to avoid contamination from touch.
Coronaviridae Study Group of the International Committee on Taxonomy of Viruses. The species severe acute respiratory syndrome-related coronavirus: classifying 2019-nCoV and naming it SARS-CoV-2. Nat. Microbiol. 5, 536–544 (2020).
Corman, V. M. et al. Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR. Eur. Surveill. 25, 2000045 (2020).
Zhang, F., Abudayyeh, O. O. & Gootenberg, J. S. A protocol for detection of COVID-19 using CRISPR diagnostics. https://www.broadinstitute.org/files/publications/special/COVID-19%20detection%20(updated).pdf (2020).
Metsky, H. C. et al. CRISPR-based COVID-19 surveillance using a genomically-comprehensive machine learning approach. bioRxiv. https://doi.org/10.1101/2020.02.26.967026 (2020).
Lucia, C., Federico, P. B. & Alejandra, G. C. An ultrasensitive, rapid, and portable coronavirus SARS-CoV-2 sequence detection method based on CRISPR-Cas12. bioRxiv. https://doi.org/10.1101/2020.02.29.971127 (2020).
Broughton, J. P. et al. CRISPR–Cas12-based detection of SARS-CoV-2. Nat. Biotechnol. https://doi.org/10.1038/s41587-020-0513-4 (2020).
Teng, F. et al. CDetection: CRISPR-Cas12b-based DNA detection with sub-attomolar sensitivity and single-base specificity. Genome Biol. 20, 132 (2019).
Zhang, Y. Novel 2019 coronavirus genome. http://virological.org/t/novel-2019-coronavirus-genome/319 (2020).
Grant, P. R. et al. Detection of SARS coronavirus in plasma by real-time RT-PCR. N. Engl. J. Med. 349, 2468–2469 (2003).
We thank professor Ng Shyh-Chang from Institute of Zoology, CAS for his critical support with this study. We thank Hanxing Zhang from Institute of Microbiology, CAS for her kind help on equipment. This work was supported by the National Key Research and Development Program (2020YFA070009 to R.R.W), the Key Research Projects of the Frontier Science of the Chinese Academy of Sciences (QYZDY-SSW-SMC002 to Q.Z. and QYZDB-SSW-SMC022 to W.L.), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA16030400 to W.L.).
Conflict of interest
The authors declare that they have no conflict of interest.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Guo, L., Sun, X., Wang, X. et al. SARS-CoV-2 detection with CRISPR diagnostics. Cell Discov 6, 34 (2020). https://doi.org/10.1038/s41421-020-0174-y
Systematic analysis, identification, and use of CRISPR/Cas13a–associated crRNAs for sensitive and specific detection of the lcrV gene of Yersinia pestis
Diagnostic Microbiology and Infectious Disease (2021)
Expert Review of Molecular Diagnostics (2021)
Biosensors and Bioelectronics (2021)
Biosensors and Bioelectronics (2021)