Degenerate sequence-based CRISPR diagnostic for Crimean-Congo hemorrhagic fever virus

Abstract

CRISPR (clustered regularly interspaced short palindromic repeats), an ancient defense mechanism used by prokaryotes to cleave nucleic acids from invading viruses and plasmids, is currently being harnessed by researchers worldwide to develop new point-of-need diagnostics. In CRISPR diagnostics, a CRISPR RNA (crRNA) containing a "spacer" sequence that specifically complements with the target nucleic acid sequence guides the activation of a CRISPR effector protein (Cas13a, Cas12a or Cas12b), leading to collateral cleavage of RNA or DNA reporters and enormous signal amplification. CRISPR function can be disrupted by some types of sequence mismatches between the spacer and target, according to previous studies. This poses a potential challenge in the detection of variable targets such as RNA viruses with a high degree of sequence diversity, since mismatches can result from target variations. To cover viral diversity, we propose in this study that during crRNA synthesis mixed nucleotide types (degenerate sequences) can be introduced into the spacer sequence positions corresponding to viral sequence variations. We test this crRNA design strategy in the context of the Cas13a-based SHERLOCK (specific high-sensitivity enzymatic reporter unlocking) technology for detection of Crimean-Congo hemorrhagic fever virus (CCHFV), a biosafety level 4 pathogen with wide geographic distribution and broad sequence variability. The degenerate-sequence CRISPR diagnostic proves functional, sensitive, specific and rapid. It detects within 30-40 minutes 1 copy/μl of viral RNA from CCHFV strains representing all clades, and from more recently identified strains with new mutations in the CRISPR target region. Also importantly, it shows no cross-reactivity with a variety of CCHFV-related viruses. This proof-of-concept study demonstrates that the degenerate sequence-based CRISPR diagnostic is a promising tool of choice for effective detection of highly variable viral pathogens.

Fig 1. The CRISPR/Cas13a-based diagnostic assay (SHERLOCK) for CCHFV detection.

The polarity of an RNA strand is labelled for positive sense (+) or negative sense (-). In the RNA reporter, “F” and “Q” mean fluorophore and quencher, respectively.

Fig 2. Design and function test of CCHFV CRISPR sets.

A and B. Sequences and genomic locations targeted by CCHFV CRISPR sets. The S segment sequences of CCHFV strains, labelled with GenBank accession numbers, were aligned at the assay design regions for CRISPR set 1 (in Panel A) and CRISPR set 2 (in Panel B). Nucleotides differing from the majority are highlighted in dark shades. Sequences and locations targeted by RT-RPA primers and crRNAs are indicated, based on the positive-sense sequence of the S segment. The CCHFV strains, representing each of all the seven clades/sub-clades, are detailed in Table 1, except that a Clade V variant, DQ211643, with a two-nucleotide deletion at genomic positions 31 and 32 is included only in this figure. Note that degenerate nucleotides were introduced into the crRNA of CRISPR set 2, to cover variations at each position in the targets, and are displayed with the standard codes by International Union of Pure and Applied Chemistry (IUPAC). C and D. Specific signal amplification by CRISPR Set 2, but not CRISPR set 1. The CRISPR sets were tested in the CRISPR/Cas13a-based SHERLOCK assay, in the presence of chemically synthesized CCHFV RNA fragments and different concentrations of stock crRNAs. Sequences of RT-RPA primers, crRNAs and synthetic RNA templates are listed in Table 2. Amplification plots show relative fluorescent units (RFU) at indicated time points in the T7-Cas13a reaction. C. Test of CRISPR set 1 (with constant crRNA spacer). PC RNA (positive control RNA, specific target) = Target_RNA_1–120; NC RNA (negative control RNA, non-specific target) = Target RNA_641–723. NTC (no-template control) = H2O. Graph represents five independent experiments showing similar patterns. D. Test of CRISPR set 2 (with degenerate crRNA spacer). PC RNA = Target RNA_641–723; NC RNA = Target_RNA_1–120. NTC = H2O. Graph represents four independent experiments showing similar patterns.

Fig 3. The degenerate sequence-based CRISPR diagnostic detects CCHFV in a sensitive, specific and rapid manner.

A–G. Detection of all CCHFV clades/sub-clades. The S segment genomic RNAs representing different CCHFV clades/sub-clades were produced by in vitro transcription, serially diluted and tested for detection by the degenerate CRISPR set (CRISPR set 2). Amplification plots show relative fluorescent units (RFU) at indicated time points in the T7-Cas13a reaction. RNA template concentrations in the T7-Cas13a reaction are shown as copies/μl (cp/μl). Red curves: CCHFV RNA-containing samples; green solid curve (0 cp/μl): NTC (no-template control, H2O); and green dash curve (Neg): the negative threshold = Mean + 3X standard deviation of NTCs, calculated based on 12 NTC replicates. Any RFU value above the negative threshold is considered CCHFV positive. Graphs represent three independent experiments showing similar patterns. H. Limit of detection (LoD). The S segment genomic RNA from the Hoti strain was produced by in vitro transcription and tested at the indicated concentrations for detection by the degenerate CRISPR set. Positivity% values are shown based on 10 replicates per target concentration type. I. The degenerate sequence-based CRISPR diagnostic is specific for CCHFV, with no cross-reactivity against closely related viruses. RNAs from CCHFV and related viruses were tested for detection by the degenerate CRISPR set. NSD: Nairobi sheep disease virus; RVFV: Rift Valley fever virus; and NTC: no-template control (H2O). CCHFV and Rift Valley fever virus RNAs were extracted from cell culture-derived viruses. Other viral RNAs were the S segment genomic RNAs produced by in vitro transcription. RNA template concentrations in the T7-Cas13a reaction were all 40 pg/μl. Amplification plots show relative fluorescent units (RFU) at indicated time points in the T7-Cas13a reaction and represent three independent experiments showing similar patterns.

Fig 4. The degenerate sequence-based CRISPR diagnostic is capable of detecting emerging CCHFV variants with new and more complex mutations.

Chemically synthesized CCHFV RNA targets were tested for detection at 1 cp/μl by the degenerate sequence-based CRISPR set, using crRNA_669–696 (crRNA_Deg), as in Fig 3. A constant sequence-based crRNA, derived from the Hoti strain, crRNA_669–696_5.Hoti (crRNA_Con), was compared to the degenerate sequence-based crRNA in the same context. The targeted CCHFV variants were isolates from India (IND, 2018) and United Arab Emirates (UAE, 2017) in Panels A and B, respectively, as well as a recombinant (Rec) between Clades I and IV. Sequences of synthetic RNA templates, RT-RPA primers and crRNAs and GenBank accession numbers of CCHFV strains are provided in Table 2. Sequence of the IND, UAE or Rec variant targeted by CRISPR was each aligned with the degenerate and constant spacer sequences of the crRNAs (all as CCHFV sense sequences). Each nucleotide in the crRNA spacers that differs from that in the targeted CCHFV variant is shaded. In crRNA_Deg, the region is underlined where each degenerate nucleotide covers its counterpart in the target sequence, meaning that one of the mixed nucleotide types represented by the degenerate code matches the counterpart nucleotide in the target. Arrow indicates the nucleotide in crRNA_Deg that does not cover/match the counterpart nucleotide in the target. In Panel C, regions in the target sequence with mutations characteristic of Clade I and Clade II are demarcated, respectively. Each amplification plot shows relative fluorescent units (RFU) at indicated time points in the T7-Cas13a reaction, in the presence (+) or absence (-) of the target CCHFV RNA, using an indicated crRNA. Data are represented as mean of three technical replicates.