key: cord-0980827-kliaxmb8
authors: Zhao, X.; Wang, Z.; Yang, B.; Li, Z.; Tong, Y.; Bi, Y.; Xia, X.; Chen, X.; Wang, W.; Tan, G.-Y.; Zhang, L.
title: Integrating PCR-free amplification and synergistic sensing for ultrasensitive and rapid CRISPR/Cas12a-based SARS-CoV-2 antigen detection
date: 2021-06-25
journal: nan
DOI: 10.1101/2021.06.17.21258275
sha: ef2a6d13131a349f885b445c013345e68238488d
doc_id: 980827
cord_uid: kliaxmb8

Antigen detection provides particularly valuable information for medical diagnoses; however, the current detection methods are less sensitive and accurate than nucleic acid analysis. The combination of CRISPR/Cas12a and aptamers provides a new detection paradigm, but sensitive sensing and stable amplification in antigen detection remain challenging. Here, we present a PCR-free multiple trigger dsDNA tandem-based signal amplification strategy and a de novo designed dual aptamer synergistic sensing strategy. Integration of these two strategies endowed the CRISPR/Cas12a and aptamer-based method with ultra-sensitive, fast, and stable antigen detection. In a demonstration of this method, the limit of detection was at the single virus level (0.17 fM, approximately two copies/L) in SARS-CoV-2 antigen nucleocapsid protein analysis of saliva or serum samples. The entire procedure required only 20 minutes. Given our system's simplicity and modular setup, we believe that it could be adapted reasonably easily for general applications in CRISPR/Cas12a-aptamer-based detection.

Antigens have been used as diagnostic hallmarks for many diseases, particularly infection with viruses, for example, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the cause of the current global coronavirus disease 2019 (COVID-19) pandemic [1] . Many methods have been developed for SARS-CoV-2 detection through nucleic acids, antigens, and antibodies. Although PCR-based nucleic acid detection is the most accepted and reliable approach, it is relatively slow and device-dependent, owing to the requirements of RNA extraction and reverse transcription. In addition, RNA is less stable than protein (antigen) in laboratory conditions [2, 3] . Moreover, antibodybased tests work well only in late stages of infection, after patients have mounted immune responses, and therefore are not suitable for early diagnosis [4] . In comparison, antigen-based SARS-CoV-2 detection has several unique merits in COVID-19 diagnosis. Given its robustness and simplicity, it provides an important addition to the toolbox for SARS-CoV-2 detection [3, 5, 6] . However, compared with mature nucleic acid detection approaches, recent state-of-the-art methods for SARS-CoV-2 antigen detection have some deficiencies in low abundance viral sample detection (e.g., early diagnosis) [3] . The underlying reason is that signal amplification of low amounts of antigen is not possible. In contrast, the amount of antibodies are already amplified in vivo after antigen stimulation, and nucleic acids can be amplified in vitro by PCR or LAMP (loop-mediated isothermal amplification) [7] , thus boosting both signals. To address the challenge of signal amplification for antigen detection, we endeavor to develop a novel antigen detection system with high sensitivity and excellent performance by amplifying the sensing signals.

CRISPR/Cas-based nucleic acid detection systems, such as SARS-CoV-2 DETECTR [8] , SHEROLOCK [9] , STOPCovid [10] , and AIOD-CRISPR [11] , have been developed and applied in rapid diagnostic tests for SARS-CoV-2. The advantages of CRISPR/Cas-based detection systems have been well demonstrated, and these methods have shown great potential during the COVID-19

pandemic, owing to their superior speed, portability, low cost, and comparable sensitivity to that of the traditional gold standard of RT-qPCR assays (one copy/μL of viral RNA) [12, 13, 14] . However, the CRISPR/Cas-based detection paradigm has not been expanded to SARS-CoV-2 antigen detection [15] .

To fill this gap, we proposed that the abovementioned antigen detection challenges could be overcome by coupling CRISPR/Cas and a de novo designed highly efficient and stable signal amplification system.

Aptamers as a general biosensing platform can be applied to biosensing or detection of a broad range of analytes [16] . Previously, CRISPR/Cas12a and aptamer-mediated systems for detection of diverse analytes (e.g., proteins or small molecules) have been developed [12, 17] . These platforms directly translate the signal of aptamer recognition target molecules into a CRISPR-mediated nucleic acid detection-based output signal. However, owing to the lack of stable amplification of the input signal, the detection sensitivity this method has been unable to meet the requirements for low abundance sample detection or early diagnosis of COVID-19, in which the viral titer is usually as low as several copies per microliter (e.g., in saliva) [18] .

Here, to improve the sensitivity or sensing efficiency of the CRISPR/Cas12a and aptamer-mediated method and achieve ultrasensitive detection, we innovated two aspects for the configuration of this new antigen detection platform. First, we developed a PCR-free amplification strategy to convert the aptamer-antigen recognition signal to more trigger dsDNA signal. Second, we undertook de novo design of a dual aptamer synergetic module to more sensitively sense the antigen independently of the affinity between the aptamer and antigen. In a demonstration of this platform, we successfully identified two copies/μL SARS-CoV-2 in samples by using our newly established system of antigen detection; the LOD was comparable to that of the gold standard of PCR-based nucleic acid detection, but the detection was more rapid (~ 20 min) and the costs were lower.

The SARS-CoV-2 viral surface contains four types of proteins, i.e., spike (S), nucleocapsid (N), membrane (M), and envelope (E), which have been used in antigen detection. In this study, to couple antigen recognition with a CRISPR-based detection system, we used a specific aptamer (A48, with an equilibrium dissociation constant (K D ) of 0.49 nM) [19] of N protein as the recognition element to develop a SARS-CoV-2 antigen biosensor with our recently established CRISPR/Cas12a-based biosensing platform [12] . Our prototype was denoted the antigen biosensing platform CaT-Smelor-Covid.v1 (CRISPR/Cas12a and aptamer-mediated detector of diverse analytes for COVID-19, version 1). The working principle of CaT-Smelor-Covid.v1 is illustrated in Fig. 1a . In brief, a hybrid DNA (HyDNA) containing the Cas12a triggering dsDNA and a single-stranded DNA (ssDNA) complementary to parts of the aptamer A48 sequence was anchored to A48-coated magnetic beads (MB) (Supplementary Fig. 1 ). SARS-CoV-2 nucleocapsid protein (Np), if present, interacts with A48 and releases the HyDNA. The released HyDNA then triggered the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 25, 2021. collateral ssDNA cleavage activity of Cas12a and then cut the fluorophore quencher (FQ)-labeled ssDNA probe, thus producing a readable fluorescence signal output. After optimizing the reaction system and signal-to-noise (S/N) ratio ( Supplementary Fig. 2 , 3), we found that the linear detection range of Np was 0.19-781 pM (R 2 > 0.99) (Fig. 1b, c, Supplementary Fig. 4 ) with a LOD of 32 fM (approximately 193 copies/μL). For nucleic acid detection, the LOD of the gold-standard RT-qPCR and CRISPR-based methods reached up to one copy/μL of viral RNA [6] . Thus, this prototype did not meet the requirements for detection of SARS-CoV-2 in saliva samples or early diagnosis.

Developing novel strategies for sensitive sensing and stable amplification is key for this CRISPRbased antigen detection paradigm.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 25, 2021.

In the CRSIR/Cas12a and FQ-based signal output and display modules, the slopes of the fluorescence signal are proportional to the concentration of the released triggering dsDNA [12] . To increase or amplify the signal generated by the Cas12a triggering dsDNA, we designed a cluster of multiple triggering dsDNAs in tandem. Considering the trade-off between the S/N ratio and amplified signal, we kept the length of dsDNA below 600 bp, a length accommodating as many as 10 copies of the Cas12a triggering dsDNA with a 30 bp interval sequence ( Supplementary Fig. 5 ).

Then the HyDNA containing different copy numbers (one to ten) of Cas12a triggering dsDNAs was anchored to A48-coated MB ( Supplementary Fig. 6 ). The slopes of the fluorescence signal were proportional to one to ten copies of Cas12a triggering dsDNA (HyDNA10; Fig. 2c , d, Supplementary 

To further improve the detection sensitivity of CaT-Smelor-Covid.v2, we pursued a de novo synergistic strategy in CaT-Smelor-Covid.v3 (version 3) by applying two different aptamers recognizing different epitopes of Np to release two cognate tandem HyDNA10 molecules ( Fig. 3a, b) .

When one aptamer binds the antigen and releases HyDNA, it promotes the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 25, 2021. ; . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 25, 2021. ;  response of the other aptamer to the same antigen and the release HyDNA due to proximity, thus producing a synergistic effect. Here, we introduced a biotin labeled Y-shaped DNA to simultaneously anchor the aptamer (A48) and another previously characterized aptamer (A61, K D = 2.74 nM) [19] on the MB. Then the two optimized HyDNA10 molecules ( Supplementary Fig. 9 ) were anchored on MB via interaction with aptamers A48 and A61. With CaT-Smelor-Covid.v3, we achieved excellent Np detection. Trace amounts of Np at concentrations of 0.19-2.98 fM were unambiguously and readily detected, with an LOD of 0.17 fM (Fig. 3c, d, Supplementary Fig. 10 ).

The LOD of SARS-CoV-2 was approximately two copies/μL, the lowest value reported to date for SARS-CoV-2 antigen detection through a CRISPR/Cas-based method [3, 15] . As expected, the response signal of CaT-Smelor-Covid.v3 allowed us to calculate a Hill coefficient of approximately 1.5, thus indicating positive cooperativity. Δ FI: fluorescence intensity increased in 1 min. The test results showed that the correct detection rate was 100%.

The LOD of CaT-Smelor-Covid.v3 for antigen detection was very close to that of the gold standard of RT-qPCR and CRISPR/Cas-based nucleic acid detection [8, 10] , thus indicating that this platform had great potential for real-life applications. We then tested the performance of our CaT-Smelor-Covid.v3 on authentic sample detection by using inactivated SARS-CoV-2 [20] . As shown in Fig. 4a , as few as two copies/μL SARS-CoV-2 could be efficiently detected (P < 0.05). Moreover, the entire . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) Fig. 4b, Supplementary Fig. 11-13 ). In another single-blind test, our method correctly distinguished the 30 positive and 30 negative samples in spiked human serum (P < 0.001; Fig. 4b ). These results indicated that CaT-Smelor-COVID.v3 was ultrasensitive and stable for SARS-CoV-2 detection. Table 1 . Comparison of reliable SARS-CoV-2 detecting methods.

Target

RT-LAMP RNA 50 copies/µL 2 µL 30 Y Ganguli A et al. [21] RT-qPCR RNA 100 copies/µL 5 µL 120 Y Vogels CBF et al. [22] Cas13a based detection RNA ~100 copies/µL ~0.3 µL 30 N Fozouni P et al. [14] DETECTR RNA 10 copies/µL 2 µL 30-40 N Broughton JP et al. [8] AIOD-CRISPR RNA ~5 copies/µL 1 µL >40 N Ding X et al. [11] STOPCovid RNA 33 copies/mL 50 µL >125 N Joung J et al. [10] Electrochemical biosensor RNA 200 copies/mL 10 µL >180 N Zhao H et al. [23] Microfluidic immunoassays Antibody N. D 10 µL 15 Y Lin Q et al. [24] MNPs biosensor Antigen 5.06×10 7 copies/μL 70 µL ~10 Y Zhong J et al. [25] LFIA Antigen 186 copies/μL <5 ng protein >20 N Lee JH et al. [3] CaT-Smelor-Covid.v3 Antigen 2 copies/μL 1 µL ~20 N This study Note: LOD, limit of detection. V, the volume of the sample required for per essay. TC, time cost. BIR: bulky instrumentation required. RT-LAMP, reverse transcription loop mediated isothermal amplification. DETECTR, detection ofSARS-CoV-2 DNA Endonuclease-Targeted CRISPR Trans Reporter. AIOD-CRISPR, All-In-One Dual CRISPR-Cas12a. STOPCovid, SHERLOCK One-Pot Testing. MNPs biosensor, homogeneous biosensing based on magnetic nanoparticles. LFIA, ACE2-based lateral flow immunoassay. N.D, not determined.

Comparison our new Cas12a-aptamer-based protein detection platform with other previously reported SARS-CoV-2 virus detection methods indicated that CaT-Smelor-Covid.v3 was more sensitive and accurate in rapid antigen detection. The results of our method are comparable to those . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 25, 2021. ;  of the gold-standard RT-qPCR assays and other CRISPR-based nucleic acid detection approaches [8, 10] (Table 1 ). In addition, our rational optimization procedures applied to the CaT-Smelor platform successfully boosted the signal about 200-fold. Given its rational and modular configuration, our system after optimization should be widely applicable in many protein detection-based applications. Figure 5 . Workflow chart of sample detection. Swab samples and serum samples were collected and then treated to release the Np. Next, lysed solution was incubated with the biosensor system. Finally, the fluorescence intensity of released HyDNA was detected with a portable fluorometer or microplate reader.

In parallel to the nucleic acid detection systems DETECTR, SHEROLOCK, and STOPCovid, the combination of CRISPR/Cas12a and aptamers represents a new detection paradigm for antigen detection [26] . Coupling with aptamers dramatically broadens the detection scope of CRISPR/Cas12a [12, 17] . However, how to stably amplify the signals sensed by aptamers and then translate the amplified signal into Cas12a cleavage activity has been an unresolved problem. Because of the lack of stable amplification of the signal sensed by the aptamer, this new detection paradigm has been unable to match the sensitivity of DETECTR and SHEROLOCK. Similarly, the sensitivity of Np antigen detection through the combination of CRISPR/Cas12a and aptamer in this study (LOD of 32 fM; ca. 193 copies/μL; Fig. 1, Supplementary Fig. 4 ) was much lower than the nucleic acid detection systems DETECTR, SHEROLOCK, and STOPCovid, although the performance of CaT-Smelor-. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 25, 2021. ; https://doi.org/10.1101/2021.06.17.21258275 doi: medRxiv preprint Covid.v1 was comparable to those of the previously reported immunology-based methods (186 viral copies/μL) [24] , Therefore, novel strategies were needed to dramatically enhance the sensitivity of this new detection paradigm. Furthermore, our results represent the general challenges of low sensing efficiency and detection sensitivity in the detection of target analytes through a combination of CRISPR/Cas12a and aptamers. Therefore, the main goal of this study was to address these problems and to provide a general solution for this detection paradigm.

In this study, multiple CRISPR/Cas12a trigger dsDNAs (up to 10 copies) were designed and constructed in tandem in a single DNA fragment (Fig. 2) . The use of these trigger dsDNAs significantly amplified the output fluorophore signal. Of note, when the trigger-dsDNA copy number was increased from one (HyDNA) to 10 (HyDNA10), the detection sensitivity also linearly increased approximately 10-fold (from 32 fM to 3.48 fM). Warranting the stabilized output signal and maximization of the S/N, we designed both the crRNA and trigger dsDNA sequences on the basis of the phage genome-derived prediction site [27] , which ensure optimal and robust performance of our CRISPR/Cas12a-based signal output system. In addition, the intersequence region between each trigger dsDNA was optimized. On the basis of our results, we believe that this well-characterized multiple trigger dsDNA tandem element could be used as a general tool in other CRISPR/Cas12a and aptamer-based detection systems, such as CaT-Smelor (16), in applications beyond Np detection.

For most aptamers, the affinity toward the corresponding target is usually at the nanomolar level (i.e., dissociation constant, K D ), and further increasing the binding affinity by systematic evolution of ligands by exponential enrichment (SELEX) may be difficult [28] . However, the K D of the aptamer directly determines the sensing efficiency and consequently affects the detection sensitivity.

Therefore, innovating the aptamer-based sensing module for a more sensitive response to antigen, independently of the affinity between aptamer and antigen, should be of general interest for all aptamer-based detection systems. In this study, we used a Y-shaped DNA and two aptamers, A48 and A61, recognizing different epitopes of the antigen, which were designed to sense the Np (Fig. 3) .

A significant synergistic sensing effect was observed, in which the binding of A48 to Np facilitated the binding of A61 and Np, and vice versa. According the response curve, the Hill coefficient reached 1.52, thus indicating a positive synergistic effect between A48 and A61 [29] . By combining the multiple trigger dsDNA and synergistic sensing strategies, our system achieved an LOD for Np detection of 0.17 fM (Supplementary Fig. 10) . Therefore, our dual aptamer synergistic sensing strategy, independently of binding affinity, not only increased the affinity limit of aptamers but also further improved the detection sensitivity (from 3.48 fM to 0.17 fM). Moreover, the synergistic . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 25, 2021. ; https://doi.org/10.1101/2021.06.17.21258275 doi: medRxiv preprint effects could be applied to the detection of any other antigen with two aptamers that recognize different epitopes of the antigen.

The output of the ssDNA cleavage activity of Cas12a can detected through many convenient methods, e.g., lateral-flow [30] . Here, we used a commercially available portable fluorescence detector to demonstrate the merits and potential of our method in clinical diagnosis. Saliva and serum spiked with different concentrations of inactivated SARS-CoV-2 were used to mimic clinical samples, and the positive detection rate reached 100% (Fig. 4b) . Despite the limitations of the experimental conditions, and the absence of clinical trials, we believe that our results confirmed the practicability of our detection approach. In addition, compared with the existing antigen detection methods, our method had significant advantages in that the sensitivity was greatly improved, and the testing is fast in time-spent and low in cost ( Table 1 ). The abovementioned advantages of CRISPR/Cas-based detection systems also extended to any antigen detection with our strategy.

Consequently, by focusing on the issues of signal amplification and sensing efficiency in the CRISPR/Cas12a and aptamer-based detection paradigm, we not only developed a PCR-free multiple trigger dsDNA tandem-based signal amplification strategy for high performance CRISPR/Cas-based detection but also designed a dual aptamer synergistic sensing strategy for highly efficient sensing.

Integrating these two aspects, we achieved the first reported ultrasensitive, fast, stable, highperformance antigen detection using a CRISPR/Cas12a and aptamer-based detection paradigm.

Given that aptamers recognizing different fragments of variable proteins (for example, multiple epitopes of antigens) can be generated by SELEX, we believe that the innovative strategy used in this study may mark the beginning of a wide range of applications in the development of detection methods for antigens and other target molecules.

Materials and reagents: Aptamers and primers were chemically synthesized by Beijing Tsingke Biotech Co. Ltd (Beijing, China) (Table S1) Table   S1 . To prepare the template for HyDNA amplification, a DNA sequence with ten copies of Cas12a trigger dsDNA was synthesized. Then, C3 modified forward primers (com1F, com2F, com3F, com4F, com5F, com6F, com7F, com8F, com9F, com10F, com11F, com12F, com13F, com14F, com15F, com16F, com17F, com18F, com19F, and com20F) combined with reverse primers (dsDNA-R1, dsDNA-R2, dsDNA-R3, dsDNA-R4, dsDNA-R5, dsDNA-R6, dsDNA-R7, dsDNA-R8, dsDNA-R9, and dsDNA-R10) were amplified by using the synthesized template DNA to produce the HyDNAs containing Cas12a trigger dsDNA and different com-ssDNA sequences (ssDNA complementary to part of the sequence of the aptamer) [31] . The PCR products were generated with Taq 2× Master Mix. The PCR conditions were 1 min at 95°C for activation followed by 30 cycles at 95°C for 30 s for denaturation, 55°C for 30 s for annealing, 68°C for 30 s for elongation, and a final . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. To prepare the template for crRNA (crRNA1, crRNA2, crRNA3, crRNA4, crRNA5, crRNA6, crRNA7, crRNA8, crRNA9, and crRNA10) synthesis, paired oligonucleotides containing a T7 priming site (Table S1) To construct the dual aptamer-based biosensor, MB, biotin labeled Y-DNA, adapter1-A48, adapter2-A61 [19] , com10-HyDNA and com11-HyDNA were used. represented the ssDNA trans-cleavage rate of CRISPR-Cas12a. The linear relationship between the trans-cleavage rate (or slope) and the SARS-CoV-2 standard concentration was obtained.

To demonstrate that SARS-CoV-2 screening with our platform would be possible outside of laboratory settings, we used a portable fluorometer (Qubit™ 4 Fluorometer) to quantify the fluorescence signal generated by the detection assay. The unactivated SARS-CoV-2 sample was diluted to three to five copies/μL, and Np was released with lysis buffer. The extracted Np sample (2 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

Supporting Information is available from the Wiley Online Library or from the author. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 25, 2021. ; https://doi.org/10.1101/2021.06.17.21258275 doi: medRxiv preprint

The Lancet. 2020

Genome Stability

129, 104529; b)

360, 439; b)

The authors declare no conflict of interest. 

Integrating PCR-free amplification and synergistic sensing for ultrasensitive and rapid Supplementary Tables   Supplementary Table 1 . Oligonucleotides used in this study. Oligo