Ali, Zahir; Aman, Rashid; Mahas, Ahmed; Rao, Gundra Sivakrishna; Tehseen, Muhammad; Marsic, Tin; Salunke, Rahul; Subudhi, Amit K; Hala, Sharif M; Hamdan, Samir; Pain, Arnab; Alofi, Fadwa S; Alsomali, Afrah; Hashem, Anwar M; Khogeer, Asim; Almontashiri, Naif A M; Abedalthagafi, Malak; Hassan, Norhan; Mahfouz, Magdy M

How to develop a rapid test with CRISPR cas protein for detection of COVID19 ?

• Choose an amplification system, such as RDA, RPA, LAMP isothermal amplification.

• Get the right CRISPR Cas protein.

• Get the target viral gene sequence.

• Engineering the cas protein with the sequence.

• Serial dilutions of samples for LoD determination.

• Optimize testing to get the best result.

Abstract

1. Introduction

The COVID-19 pandemic caused by SARS-CoV-2 affects all aspects of human life. Detection platforms that are efficient, rapid, accurate, specific, sensitive, and user friendly are urgently needed to manage and control the spread of SARS-CoV-2. RT-qPCR based methods are the gold standard for SARS-CoV-2 detection. However, these methods require trained personnel, sophisticated infrastructure, and a long turnaround time, thereby limiting their usefulness. Reverse transcription-loop-mediated isothermal amplification (RT-LAMP), a one-step nucleic acid amplification method conducted at a single temperature, has been used for colorimetric virus detection. CRISPR-Cas12 and CRISPR-Cas13 systems, which possess collateral activity against ssDNA and RNA, respectively, have also been harnessed for virus detection.

CRISPR-Cas systems, including CRISPR-Cas12 and CRISPR-Cas13, exhibit robust collateral activity against single-stranded DNA (ssDNA) and RNA targets, respectively. Such collateral activity provides the basis for highly specific, sensitive approaches for nucleic acid detection. For example, SHERLOCK (Specific High-sensitivity Enzymatic Reporter UnLOCKing) and DETECTR (DNA Endonuclease-Targeted CRISPR Trans Reporter) detection have attracted substantial attention. CRISPR-based diagnostic methods exploit the efficiency and simplicity of different isothermal amplification approaches, such as LAMP and recombinase polymerase amplification, which achieve highly specific, sensitive amplification of a few copies of the targeted nucleic acid at single temperature in a short period of time, eliminating the need for thermocycling steps, and are therefore favored for POC diagnostics where low-cost and ease of use are needed. Although these systems hold great promise for developing an effective diagnostic, employing these techniques as POC diagnostics amenable to massive-scale field deployment remains challenging.

In the current study, to address the challenge of SARS-CoV-2 detection, we established a system involving RT-LAMP coupled with CRISPR- Cas12 for the rapid, specific, accurate, sensitive detection of SARS-CoV- 2 and named this system iSCAN (in vitro Specific CRISPR-based Assay for Nucleic acids detection).

Our iSCAN platform is

1) rapid, as the RT- LAMP and CRISPR-Cas12 reaction takes less than 1 h;

2) specific, because detection depends on the identification and subsequent cleavage of SARS-CoV-2 genomic sequences by the Cas12 enzyme;

3) field- deployable, as only simple equipment is required; and

4) easy to use, as the colorimetric reaction coupled to lateral flow immunochromatography makes the assay results easy to assess. Our iSCAN approach is suitable for large-scale, in-field deployment for the early detection of SARS-CoV-2 carriers, allowing them to be effectively isolated and quarantined, thus limiting the spread of the virus. Our SARS-CoV-2 detection kit has been validated using extracted RNAs from clinical samples from COVID-19 positive patients, providing the possibility of widespread deployment for virus detection.

2. Results

2.1. Establishment of RT-LAMP for rapid detection of SARS-CoV-2 in a low-resource environment SARS-CoV-2 testing faces many hurdles, including the availability of critical reagents, highly-trained technical personnel, and sophisticated equipment. Hence, the ability to produce diagnostic reagents in-house is vital in epidemic situations. Therefore, we set out to build a SARS-CoV-2 end-to-end detection platform that is sensitive, specific, easy-to-use, and low cost to facilitate its field deployment on a massive scale. LAMP reactions employ DNA polymerases that possess strand-displacement activities, such as the Bst DNA polymerase large fragment, for efficient amplification of the target DNA. By supplementing the LAMP reaction with a suitable isothermal reverse transcriptase, reverse transcription LAMP (RT-LAMP) can be used to detect RNA by reverse transcribing the target RNA and subsequent DNA amplification in one step. Therefore, we expressed and purified recombinant Bst DNA polymerase large fragment (referred to as Bst LF) from Geobacillus stearothermophilus, and the synthetic, directed evolution-derived RT ‘xenopolymerase’ RTx enzymes (Ellefson et al., 2016). The catalytic activities and performance of the purified proteins for LAMP with the purified Bst LF, and RT-LAMP with RTx and Bst LF were assayed and compared with commercial Bst and RTx enzymes using published RT-LAMP primers designed against the SARS-CoV-2 genome (Zhang et al., 2020). Our in-house produced proteins showed efficient and consistent LAMP and RT-LAMP amplifications comparable to the commercial enzymes.

Next, we designed, built, and tested several sets of LAMP primers targeting the SARS-CoV-2 genome. The SARS-CoV-2 genome consists of ~30 kb positive single-stranded RNA with a 5′-cap structure and 3′ poly- A tail containing several genes characteristic of coronaviruses, such as S (spike), E (envelope), M (membrane), and N (nucleocapsid) genes (Fig. 1A).

Fig. 1

Other elements of the genome, such as ORF1a and ORF1b, encode non-structural proteins, including RNA-dependent RNA polymerase (RdRp) (Lu et al., 2020a; Kim et al., 2020). We targeted two regions in the N and E genes. The N gene at the 3′ end of the virus genome is highly conserved among coronaviruses. We designed LAMP primers to generate ~200 bp amplification products to ensure robust amplification sufficient for LAMP-based detection. We conducted RT-LAMP experiments on synthetic SARS-CoV-2 sequences. We identified primer sets, for N and E genes, which were able to specifically and efficiently amplify the synthetic virus fragments, but not the controls.

2.2. Efficient detection of SARS-CoV-2 via RT-LAMP coupled with CRISPR-Cas12a The sensitivity of RT-LAMP is comparable with the RT-PCR. However, the specificity of the reactions and the visualization of the results may be complicated due to primer-dimer formation, non-specific amplification, and cross-contamination when running a large number of samples simultaneously. Therefore, to develop a highly sensitive and specific detection system, we coupled CRISPR-Cas12 with RT-LAMP target amplification to establish a binary system for SARS-CoV-2 detection. To this end, we purified the class II, type V Lachnospiraceae bacterium ND2006 (LbCas12a) orthologue. We assessed its programmable (specific) cis-cleavage activity against dsDNA targets and (collateral) trans-cleavage activity on ssDNA non- targeted sequences. Purified LbCas12a assembled with crRNAs targeting dsDNA fragments of the SARS-CoV-2 N or E genes exhibited endonuclease activity on the targeted dsDNA, confirming the catalytic cis- cleavage activity of the purified Cas12a. To assess the collateral activity of the purified LbCas12a protein, the specific dsDNA targeting reaction was supplemented with non-targeted fluorophore quencher (FQ)-labeled ssDNA reporters. Cas12a catalyzed the efficient degradation of the ssDNA FQ reporters as measured by the fluorescent signal, which was generated only in the presence of the specific dsDNA targets and specific crRNA, indicating the active collateral cleavage activity of the purified enzyme. To establish our iSCAN system for the detection of SARS-CoV-2, we used RT-LAMP for virus detection and amplification coupled with CRISPR-Cas12 as a specificity factor. The positive signal from Cas12 provides a specific and accurate signal for virus detection. Collateral ssDNA-FQ reporter is cleaved upon Cas12a binding and cleavage of the target virus sequence in the RT-LAMP amplification products (Fig. 1B). We designed crRNAs specific to the SARS-CoV-2 N and E gene genomic regions amplified with the primer sets identified in the primer screening assays. We tested the activity and specificity of our two-pot iSCAN system using an FQ reporter assay. We performed RT-LAMP using the E gene- specific RT-LAMP primers and synthetic SARS-CoV-2 genome. Subsequently, we introduced the RT-LAMP product to Cas12a protein pre- assembled with crRNAs specific to the E gene amplicon, non-specific crRNA, or without crRNA in the presence of FQ-ssDNA reporter. Cas12a cleaved the ssDNA reporters only in the presence of specific crRNA and RT-LAMP amplicons, confirming the specificity and activity of our systems. Next, we quantitatively compared the sensitivity of our system to that of the approved SARS-CoV-2 CDC RT-qPCR assays to determine the limit of detection (LoD). We conducted our two-pot iSCAN assays with various dilutions of the synthetic viral RNA ranging from 0 to 20,000 copies per reaction. The LoD assays showed that the iSCAN reactions could detect down to 10 RNA copies per reaction, versus 5 copies per reaction for the CDC RT-qPCR assays (Lu et al., 2020b), indicating the high sensitivity of our assay (Fig. 1D). To further optimize the assay, we systemically evaluated different RT-LAMP and Cas12a detection timing to determine the time needed to detect the signal. Using the lowest concentration identified in the LoD assay, we added RT-LAMP products from different amplification time points, including 5, 10, 20, and 30 min to crRNA-Cas12a detection complex along with the fluorescence reporter. The samples were incubated at 37 ◦C for different times (5, 10, 20, and 30 min), and the fluorescent signal generated from the reporter by the Cas12a collateral activity upon target recognition was measured. Our assay reliably detected the positive signal with 20 min of RT-LAMP followed by 20 min of Cas12a detection comparable to the stronger fluorescent signal with prolonged RT-LAMP and Cas12a detection.

2.3. Validation of RT-LAMP coupled CRISPR-Cas12a for the detection of SARS-CoV-2 in clinical samples To validate our detection system for application with real samples from patients, we first assessed the capability of our iSCAN assay to detect SARS-CoV-2 nucleic acid extracted from nasopharyngeal swabs of five different patients who tested positive for SARS-CoV-2 with RT-qPCR assays and two patients who tested negative (Supplementary File 2). The positive samples had different Ct values ranging from 15 to 40 with the CDC qPCR N gene primer sets (IDT, Catalogue #10,006,606). Using fluorescence-based detection with at least three replicates for each sample, our iSCAN system targeting the SARS-CoV-2 E gene showed 100 % agreement with RT-qPCR results. To facilitate the effective detection of SARS-CoV-2 in a low-resource environment, we sought to simplify the use of iSCAN for the detection of SARS-CoV-2. Therefore, we coupled lateral flow immunochromatography with our iSCAN detection system. Lateral flow strips are user- friendly and straight-forward, and can facilitate the development of home testing for SARS-CoV-2. We evaluated our system for POC diagnostics with lateral flow readouts using commercial lateral flow strips. Lateral flow assays showed 100 % concordance with results obtained with fluorescence-based detection (Fig. 1F). Next, we evaluated our iSCAN assay by testing on total RNAs extracted from an additional 24 nasopharyngeal swab samples obtained from 21 SARS-CoV-2-positive and 3 negative patients. We assayed the performance of our system with E and N gene primers and crRNAs. We observed low sensitivity of our assay when using the E gene target, with 8 out of 21 positive samples testing positive, and 3 out of 3 negative samples testing negative with fluorescence- based readouts. However, when we used the N gene assay, we observed a significant increase in the sensitivity of the assay, where 18 samples tested positive out of the 21 positive samples with qPCR (~86 %), and the 3 negative samples were diagnosed correctly in agreement with the qPCR data (100 %). We next tested the performance our iSCAN N gene assay using the simple lateral flow readouts. We observed high agreement between the fluorescent- based detection and the lateral flow readout results.

2.4. SARS-CoV-2 detection via RT-LAMP coupled CRISPR-Cas12b Our two-pot iSCAN systems, employing RT-LAMP coupled with CRISPR-Cas12a, reliably and specifically detected SARS-CoV-2 from clinical patient samples. However, developing virus detection modalities suitable for POC testing might require minimal liquid handling and single-pot reactions to facilitate wide adoption and in-field deployment. Therefore, we attempted to further simplify our iSCAN system and develop a one-pot assay by employing the thermophilic variants of Cas12b that can catalytically function in the same temperature range as RT-LAMP. To this end, we expressed and purified AacCas12b (Alicyclobacillus acidoterrestris) and AapCas12b (Alicyclobacillus acidophilus) variants from E. coli (BL21) DE3 (Teng et al., 2018; Shmakov et al., 2015). We confirmed the cis activity of these proteins on PCR products of the target sequences using sgRNA sequence of the AacCas12b variant. Our data show that both Cas12b variants were capable of inducing double-strand breaks using the same sgRNA at 62 ◦C, and had robust collateral activities when incubated with ssDNA FQ reporter. However, because AapCas12b showed enhanced cis cleavage activity at 62 ◦C, we chose to proceed with this Cas12b variant. The goal of utilizing the thermophilic Cas12b variants was to develop a simple one-pot and single temperature detection modality that minimizes liquid handling. This can be achieved by mixing all amplification and CRISPR- based detection reagents simultaneously in one pot, which allows simultaneous target amplification and detection. Thus, using the N gene assay, we attempted to verify the feasibility of performing SARS- CoV-2 detection in a one-pot assay format using the synthetic virus RNA in one-pot detection reaction mixtures containing AapCas12b, sgRNA, ssDNA-reporters, and RT-LAMP amplification reagents, and incubated the reaction mixtures at 62 ◦C for 1 h. The one-pot detection system was capable of detecting the viral RNA, albeit at lower efficiency compared to the two-pot system. When we visualized the RT-LAMP product from one-pot assay on an agarose gel, we found weak amplification of the target virus RNA. Therefore, we hypothesized that the concurrent presence of the active Cas12b–sgRNA complex in the same pot leads to the digestion of the initial RT-LAMP product, which significantly affects the performance of the RT-LAMP amplification, and thus the robustness of the detection. To improve the performance of the one-pot detection, we separated the Cas12b enzyme from the rest of the detection components by adding the Cas12b protein in a droplet on the tube wall, then allowed the RT- LAMP reaction to proceed before mixing the Cas12b with the other reaction components. We observed an enhanced RT-LAMP amplification of the target and improved detection performance using this “spotted” one-pot reaction with synthetic viral RNA. In the spotted one-pot reaction, AapCas12b consistently achieved a LoD of 10 copies per reaction. Next, we tested the ability of the AapCas12b-based detection system to detect SARS-CoV-2 RNAs in clinical samples (3 positive and 1 negative) using fluorescence-based and lateral flow-based readouts. AapCas12b was capable of detecting the SARS-CoV-2 RNA using spotted one-pot and all-mixed one-pot reactions. Using the spotted one-pot AapCas12b detection system, we tested the 24 clinical samples used with the Cas12a based detection system. The performance of AapCas12b system was very similar to the performance of Cas12a-based system, where 18 out of the 21 qPCR- positive samples showed a positive signal, and the 3 qPCR negative samples showed negative results using fluorescence-based readouts. However, when we tested the system on 17 of the positive samples and 2 negative samples using lateral flow readouts, we consistently observed weak performance measured by the weak signal on lateral flow readouts in positive samples compared to negative ones, which compromised the interpretation of the results.