key: cord-0732019-6gzhpssv
authors: Yang, Yihan; Liu, Jinchuan; Zhou, Xiaohong
title: A CRISPR-based and post-amplification coupled SARS-CoV-2 detection with a portable evanescent wave biosensor
date: 2021-10-15
journal: Biosens Bioelectron
DOI: 10.1016/j.bios.2021.113418
sha: ae9f578791384015153cbaf591bd919d9c778306
doc_id: 732019
cord_uid: 6gzhpssv

The continuing pandemic of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, which causes coronavirus disease 2019 (COVID-19), has spread globally and its reliable diagnosis is one of the foremost priorities for protecting public health. Herein a rapid (<1 h), easy-to-implement, and accurate CRISPR-based evanescent wave fluorescence biosensing platform for detection of SARS-CoV-2 is reported. The collateral effect of Cas13a is combined with a universal autonomous enzyme-free hybridization chain reaction (HCR) by designing a cleavage hairpin reporter, which is cleaved upon target recognition, and hence releasing the initiator sequence to trigger the downstream HCR circuits. Detection of HCR assemblies is accomplished by first adsorbing to the desthiobiotin-modified optical fiber, followed by fluorescence emission induced by an evanescent field. Three Cas13a crRNAs targeting the genes of S, N and Orf1ab of SARS-CoV-2 are programmed to specifically target SARS-CoV-2 or broadly detect related coronavirus strains, such as MERS-CoV and SARS-CoV. The HCR amplification coupled Cas13a-based biosensing platform is capable of rapid detection of SARS-CoV-2 with attomolar sensitivity. This method is further validated by adding target RNA of SARS-CoV-2 in negative oropharyngeal swabs. The good discrimination capability of this technique demonstrates its promising potential for point-of-care diagnosis of COVID-19.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, which causes 29 coronavirus disease 2019 (COVID-19), has spread globally since its discovery in Wuhan city, 30

China in December 2019 ). Up to January 3, 2021, the disease has spread to 31 at least 219 countries/area, infected at least 83 million people, and has resulted in at least 1.8 32 million deaths globally. The apparent ease with it spreading from human to human presents 33 an imminent threat to the global public health ; Wrapp et al. 2020 ). Simple, low-34

cost and yet accurate, sensitive, and quantitative detection of this virus has been one of the 35 foremost priorities for facilitating public health interventions. 36

The SRAS-CoV-2 virus is enveloped and generally spherical with diameter ranging from 60 37 nm to 140 nm, and its solar corona-like morphology is consistent with the family 38

Coronaviridae (Zhu et al. 2020 ). SARS-CoV-2 has a single-stranded positive-sense RNA 39 genome that is ~30,000 nucleotides in length. Its first complete genome was discovered by 40 using a combination of Sanger, Illumina, and nanopore sequencing, providing important data 41 for researchers to design primers and to probe sequences for other nucleic acid tests ( The CRISPR-Cas13a was diluted by using 6× loading buffer with a volume ratio of 5:1 for 152 gel electrophoresis. A 15% urea-denaturing polyacrylamide gel was prepared with 10× TBE 153 buffer. After pre-running and warming the gel for 30 min at 50 V and then loading the 154 samples, electrophoresis was run at 80 V for 100 min in 1× TBE buffer. Gels were soaked in 155

1× TBE for about 10 min and then stained with SYBR Green I Gel Stain and imaged under 156 UV illumination using FLS-5100 film (Fuji Photo Film Co., Ltd., Tokyo, Japan). 157

To characterize the formation of HCR assemblies, a 15% fresh polyacrylamide gel was 158 prepared with 5× TBE buffer. Electrophoresis was run at 60 V for 100 min in 1× TBE buffer. 159

Gels were stained and imaged as the way in the denaturing polyacrylamide gel 160 electrophoresis. 161

Under the optimized conditions, 2 μL LwaCas13a in storage buffer (63.3 μg ml -1 stock 164 concentration), 2 μL crRNA (225 nM stock), 2 μL Reporter 2 (1.25 μM stock), 2 μL synthetic 165 SARS-CoV-2 genes at varying concentrations, 0.5 μL RNase inhibitor and 11.5 μL nuclease 166 assay buffer were mixed for Cas13a recognition. After incubation of the mixture for 20 min, 167 20 μL SPSC buffer, 5 μL annealed Bio-H1 (1 μM stock) and 5 μL annealed Cy5.5-H2 (1 μM 168 stock) were added and then the mixtures further incubated at 37 ℃ for 30 min. The annealing 169 process contained that 1 μM Bio-H1 and Cy5.5-H2 stock were denatured respectively at 95 ℃ 170 for 5 min, and then cooled at room temperature for over 1 h to form annealed hairpins. Lastly, 171 500 μL SPSC buffer was added into the reaction tube that contained amplified HCR products 172 to fit the volume requirement for evanescent wave fluorescent biosensing detection. The 173

optical-fiber based detection proceeded as follows: 1) 20 nM streptavidin was first delivered 174 into the flow cell to bind to DTB immobilized onto the sensing surface of fiber for 2 min; 2) 175 J o u r n a l P r e -p r o o f SPSC buffer was pumped to elute the unbounded SA; 3) amplified HCR products with biotin 176 labels were pumped to react with interface-bound streptavidin for 10 min; 4) washing buffer 177 was pumped to elute interface-bound streptavidin and HCR products for 10 min; 5) 178 equilibration buffer was pumped for 1 min to regenerate the sensing surface for next round of 179 reaction cycle. (Table S1 ). After annealing, the 226 formation of hairpin reporter placed fluorophore in close proximity to its quencher, resulting 227 in a low fluorescence intensity background; after the specific recognition, Reporter 1 beacon 228 was cleaved and separated FAM from BHQ1, leading to an obvious florescence recovery. 229

Cas13a crRNAs are programmed to specifically target SARS-CoV-2 or broadly detect related 231 coronavirus strains. As shown in Fig. S3 , the N gene and S gene of SARS-CoV and MERS-232

CoV were significantly different from that of SARS-CoV-2. While, there were only two-base 233 difference between Orf1ab gene of three viral sequences. Therefore, theoretically, the N gene 234 and S gene crRNAs used in the assay were specific for SARS-CoV-2, however failed to 235 detect SARS-CoV and MERS-CoV, whereas the Orf1ab gene crRNA was capable to detect 236 three highly pathogenic coronavirus strains. As revealed by the net fluorescence subtracting 237 that of the blank sample in Fig. 2a , we demonstrated that the CRISPR-Cas13a system was 238 able to distinguish SARS-CoV-2 with no cross-reactivity for other two highly pathogenic 239 coronaviruses using the crRNAs targeting N gene and S gene, however, with expected cross-240 reactivity for the related coronavirus strains using the crRNA targeting Orf1ab gene. The 241 CRISPR-Cas13a system showed higher sensitivity when targeting S gene compared with that 242 targeting N gene. 243

Denaturing polyacrylamide gel electrophoresis (PAGE) assay revealed that the targets and 244 nucleic acid reporters, Reporter 1 beacon, were cleaved into short RNA fragments due to the 245 cleavage activity of CRISPR-Cas13a (Fig. 2b) respectively, as shown in Table S1 , were examined by using the home-made evanescent wave 288 biosensor (Insert of Fig. 2d) . After that, we designed the I sequence-triggered TSD of hairpins 289 (Bio-H1 and H2-Cy5.5) to generate multiple linear duplex I·(H1·H2) n assemblies as signal 290 amplification products. By using a home-made optical-fiber evanescent wave fluorescence 291 biosensor as the signal collector, the spontaneous hybridization of H1 and H2 hairpins was reporter with a stem of 12 hybridized base pairs, i.e. Reporter 2, was used in the following 300 experiments. The formation of assembled products was also be demonstrated by native PAGE 301 (Fig. 2e) , where the formed HCR assemblies in lane 5 exhibited lower and variable migration 302 speeds than other short strands. Annealing of sequences showed a slight improvement on the 303 biosensor signal (Fig. S5a) . For the sake of time and cost saving, the concentrations of H1 304 and H2, temperature and optimal reaction time for the viral recognition triggered HCR 305 reactions were 5 µL, 37 ℃, 0.5 h, respectively (Fig. S5b to d) . We also compared the 306 performance of adopting two individual steps (i.e. CRISPR-Cas13a recognition and HCR 307 amplification) with that of merging them into one-pot. The biosensor signal exhibited only 308 half of the original value after merging two steps together (Fig. 2f) . Hence, one-pot reaction 309

was not recommended even though prolonging the reaction time helped to improve the signal 310 to some extent. More investigations to develop a common buffer that could accommodate 311 both the CRISPR-mediated recognition and the HCR amplification are needed(Ladha 2020). 312

To reduce the cost of single detection, we used DTB molecules with moderate binding affinity 314 toward streptavidin to prepare a reusable optical-fiber surface ). Biotinylated 315 HCR assemblies are designed to be anchored onto DTB-functionalized fiber surface via 316 streptavidin "glue", providing approximately unified physical distances for labeled 317 fluorophores. Signals emitted by these Cy5.5-labeled HCR assemblies under the evanescent 318 wave excitation can be measured by the biosensing platform. Moreover, streptavidin-HCR 319 complexes can be effectively washed off to regenerate the DTB modified fiber surface. The 320 performances of DTB-functionalized optical fiber in governing the balance of HCR assembly 321 capture and surface regeneration were evaluated. Specifically, BSA-DTB conjugates with 322 optimal molar ratios of 50 X were synthesized and covalently immobilized on the fiber 323 surface. The "glue" streptavidin layer was robust enough to resist buffer elution (Wang et al. 324 2019), which was applied before the HCR assembly injection to guarantee the minimal 325 interferences from unbound SA, and could be efficiently washed off using washing buffer 326 (0.5% SDS, pH 1.9) (Wang et al. 2019) . Cas-HCR biosensing technique using crRNA_N towards the N genes of SARS-CoV-2 and 350 crRNA_Orf1ab towards the Orf1ab of SARS-CoV-2 were investigated, respectively. Results 351 showed that this technique was able to detect down to 100 aM, i.e. 60 copies/µL, SARS-CoV-352 2 N gene and 10 aM, i.e. 6 copies/µL, SARS-CoV-2 Orf1ab gene, respectively (Fig. 3c, d) . 353

The R 2 of curve-fitting between the signal and the concentrations of N genes and Orf1ab 354 genes were 0.997 and 0.966, respectively, over the range of 0-1 nM. 355 fiber allowed to be reused over 100 successive cycles with less than 2.5% signal decrease 375 using washing buffer (Fig. S8a) . In general, we used the fiber for testing if we continued the 376 experiment within one month for no more than 150 successive cycles. For example, three 377 optical fibers were used to get all data in Fig. 3 . Moreover, we evaluated the carryover issue 378 to treat the fiber with 4 nM Cy5.5-labeled streptavidin protein. By observing the sensorgram 379 during the binding, washing buffer elution, and equilibration buffer regeneration as shown in 380 for no more than 10 min, and then equilibration buffer for 1 min, indicating the washing 382 buffer was effective to break the binding interaction between the DTB and streptavidin, hence 383 the potential carryover issue should be negligible to ensure the surface regeneration 384 efficiently. 385

Last, we investigated the application ability of this Cas-HCR biosensing technique by 386

CDC Diagnostic Tests for COVID-19

This work was supported by Tsinghua University Spring Breeze Fund and National Nature 431 Science Foundation of China (52091541). 432

detecting S genes of SARS-CoV-2 spiked in RNA extracts of oropharyngeal swabs from three 387 healthy donors. The SARS-CoV-2 RNA was confirmed negative by the RT-qPCR technique. 388Right inset of Fig. 3b shows the heat map of three concentration levels (10 aM, 100 aM and 389 1000 aM) of S genes of SARS-CoV-2 measure by using this technique. The color bar ratio 390 represents the relative signal intensity calculated by Eq. 1, which showed significant 391 difference from the control one and also good correlation between the target concentration 392and Cas-HCR signal. Notably, we observed the decreased signal against the same 393 concentration in the oropharyngeal swabs compared with those obtained in buffer, reflecting 394 (Table S2) . This attempt provides a valuable reference for the 408 subsequent development of more post amplified CRISPR technologies. 

The authors declare that they have no known competing financial interests or personal 428 relationships that could have appeared to influence the work reported in this paper. 429

Supplementary data to this article can be found online. 434References 435