SARS-CoV-2 pandemics: An update of CRISPR in diagnosis and host–virus interaction studies

Abstract

Since December 2019, the Coronavirus disease 2019 (COVID-19) outbreak caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) has spread rapidly around the world, overburdening healthcare systems and creating significant global health concerns. Rapid detection of infected individuals via early diagnostic tests and administration of effective therapy remains vital in pandemic control, and recent advances in the clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR-associated proteins (Cas) system may support the development of novel diagnostic and therapeutic approaches. Cas-based SARS-CoV-2 detection methods (FnCAS9 Editor Linked Uniform Detection Assay (FELUDA), DNA endonuclease-targeted CRISPR trans reporter (DETECTR), and Specific High-sensitivity Enzymatic Reporter Unlocking (SHERLOCK)) have been developed for easier handling compared to quantitative polymerase chain reaction (qPCR) assays, with good rapidity, high specificity, and reduced need for complex instrumentation. Cas-CRISPR-derived RNA (Cas-crRNA) complexes have been shown to reduce viral loads in the lungs of infected hamsters, by degrading virus genomes and limiting viral replication in host cells. Viral-host interaction screening platforms have been developed using the CRISPR-based system to identify essential cellular factors involved in pathogenesis, and CRISPR knockout (CRISPRKO) and activation screening results have revealed vital pathways in the life cycle of coronaviruses, including host cell entry receptors (ACE2, DPP4, and ANPEP), proteases involved in spike activation and membrane fusion (cathepsin L (CTSL) and transmembrane protease serine 2 (TMPRSS2)), intracellular traffic control routes for virus uncoating and budding, and membrane recruitment for viral replication. Several novel genes (SWI/SNF Related, Matrix Associated, Actin Dependent Regulator of Chromatin, subfamily A, member 4 (SMARCA4), ARIDIA, and KDM6A) have also been identified via systematic data mining analysis as pathogenic factors for severe CoV infection. This review highlights how CRISPR-based systems can be applied to investigate the viral life cycle, detect viral genomes, and develop therapies against SARS-CoV-2 infection.

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) and coronavirus disease 2019 (COVID-19)

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is an enveloped positive-sense (+) single-stranded RNA (ssRNA) virus [1]. The virus belongs to the β-Coronavirus subgenus of the Coronaviridae family, which is part of the order, Nidovirales [2]. SARS-CoV-2 shares 79% similarity with SARS-CoV and 50% similarity with Middle East Respiratory Syndrome Coronavirus (MERS-CoV) [3]. It is one of seven types of CoV that are known to infect humans, including CoVs that cause the common cold (NL63, OC43, HKU1, 229E), and the severe CoVs (SARS-CoV, MERS-CoV, SARS-CoV-2) that cause mild symptoms to severe pneumonia, dyspnea, and even death [4]. The genome sequence of SARS-CoV-2 consists of a single RNA sequence of about 30 kb in length that encodes sixteen non-structural proteins (nsp), four structural proteins, and nine putative accessory factors [5] [Fig. 1]. For virion replication to begin, the spike protein first specifically binds to the host cell surface receptor, angiotensin-converting enzyme 2 (ACE2), together with other host factors. This binding functions as a protein primer to allow the precursor form of the spike protein to be cleaved to S1 and S2 subunits by the cellular protease transmembrane protease serine 2 (TMPRSS2) [6]. The conformation change of the spike protein is critical for the virion to penetrate into human cells via fusion with cellular or endosome membranes [6]. Once inside the cytoplasm, the virus releases and uncoats the genomic RNA, which is subjected to immediate translation of two large open reading frames (ORF), ORF1a and ORF1b [7,8]. The resulting translation products, polyproteins pp1a and pp1ab, are co-translationally and post-translationally processed into the 16 individual nsp that form the viral replication and transcription complex (RTC) [9,10]. The genomic and sub-genomic RNAs are replicated or transcribed by RTCs in double-membrane vesicles (DMVs) that originate from the endoplasmic reticulum (ER) and provide protection from cellular exonucleases and the innate immune system [11,12]. The structural spike (S), envelope (E), membrane (M), and nucleocapsid (N) proteins are translated by ribosomes and bound to the ER [13], then transported to the ER-to-Golgi intermediate compartment (ERGIC), where the structural proteins are assembled with genomic RNA to form virions. The virions subsequently pass through the secretory route to the cell surface, where they are released via exocytosis to find another host cell for infection [14,15]. Each step of this SARS-CoV-2 life cycle involves intercommunications between viral and host proteins, and researchers have sought to screen for host factors that are essential for the replication of multiple CoV, in order to develop antiviral drugs with broad-spectrum activities.

Fig. 1.

Schematic of the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) virion. (A) The morphology, structural protein composition and topology, and RNA genome of SARS-CoV-2. (B) Genome organization of SARS-CoV-2. The non-structural polyprotein has two primary fragments, pp1a and pp1ab, that are co-translationally and post-translationally processed into individual functional proteins associated with viral replication and transcription. The downstream portion consists of the structural and accessory proteins.

Abbreviations: nsp: nonstructural proteins.

SARS-CoV-2 can be transmitted rapidly through direct contact with droplets or tiny airborne particles generated by coughing and sneezing from infected carriers, or indirect contact with contaminated surfaces [16]. Reports have shown that the infectivity period of SARS-CoV-2 ranges from 2 to 14 days, and transmission is possible prior to the emergence of symptoms [1,17]. Effective diagnostics and treatments are needed to combat the virus, and since the outbreak of the current COVID-19 pandemic caused by SARS-CoV-2, numerous diagnostics with high specificity have been developed. These can be roughly divided into two main groups: immunological assays and molecular assays. Immunological assays such as the enzyme-linked immunosorbent assay (ELISA) can be used to identify viral antigens in the respiratory secretions of infected individuals, or antiviral antibodies in their blood. Currently, molecular assays are based on the identification of SARS-CoV-2 RNA in nasopharyngeal samples. The standard COVID-19 diagnostic method as recommended by the World Health Organization (WHO) is reverse transcription-quantitative polymerase chain reaction (RT-qPCR) based on TaqMan primers and probes to detect SARS-CoV-2 genes for the E, N, and RNA-dependent RNA polymerase (RdRp) proteins [18,19]. However, being an RNA virus, the viral genome of SARS-CoV-2 frequently mutates, and since the beginning of the pandemic, the WHO has already recognized five SARS-CoV-2 mutants as variations of concern: alpha, beta, gamma, delta, and omicron [20,21]. SARS-CoV-2 viral mutations may affect the results of immunological and molecular assays. In response to the urgent nature of the pandemic and the shifting landscape of viral variants, research and development of accurate diagnostic tools and appropriate therapeutics are currently in fast-track mode.

The CRISPR-Cas system

The clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR-associated proteins (Cas) system was initially identified in prokaryotic organisms, such as bacteria and archaea, as an adaptive immune system that acts against bacteriophage infections and invading plasmids [22]. A typical CRISPR-Cas locus in a bacterial genome consists of a CRISPR array, comprising a series of short palindromic repeats (25–35 base pairs (bp) each) separated by several DNA fragments known as spacer sequences that are 30–40 bp in length [23] [Fig. 2]. Adaptation, Cas-CRISPR-derived RNA (crRNA) biogenesis, and target interference represent three distinct stages of the CRISPR–Cas defense mechanism [24]. The CRISPR array is formed during the adaptation stage, when foreign DNA fragments from invading viruses and plasmids, known as protospacers, are identified and integrated between two adjacent repeats in the CRISPR locus region. The protospacer adjacent motif (PAM), a short length of conserved nucleotides located proximal to the protospacer, serves as a non-self recognition motif for cleavage of the invading DNA by bacterial Cas nuclease. In the crRNA biogenesis stage, CRISPR arrays are transcribed by RNA polymerase into precursor CRISPR RNA (pre-crRNA), which further forms an RNA duplex with trans-activating CRISPR RNA (tracrRNA) and is processed by RNase III. In the interference stage, the crRNA/tracrRNA hybrid binds with and guides Cas endonuclease to form a complex with the foreign DNA. This enables cleavage of the DNA sequence when bacteria encounter a secondary infection of the target pathogen or virus, thereby providing immunity to the host cell [[24], [25], [26]].

Fig. 2.

Mechanistic overview of CRISPR-Cas-mediated adaptive immunity. The CRISPR-Cas system was initially identified in bacteria, and incorporates genome fragments from past viral invaders to create cellular memory and provide adaptive immune protection, with three progressive stages: I. adaptation or spacer acquisition, in which viral DNA sequences are incorporated into bacterial DNA; II. crRNA biogenesis, in which Cas9 complexes with tracrRNA and crRNA, followed by cleavage with RNase III; and III. target interference, in which the cleaved crRNA/tracrRNA hybrid binds with invading viral DNA and guides the Cas9 endonuclease to create genotoxic double-stranded breaks.

Abbreviations: crRNA: CRISPR-derived RNA; pre-crRNA: precursor crRNA; tracrRNA: trans-activating crRNA.

CRISPR-Cas systems can be divided into two distinct classes according to their effector modules, and each class contains three different types and several subtypes [27]. Class 1 CRISPR-Cas systems include type I, type III, and type IV systems, which contain multiple Cas proteins and crRNA for nucleic acid degradation, while Class 2 systems comprise type II, type V, and type VI, which use only one Cas protein and crRNA [27] [Fig. 3]. Class 2 systems are the most commonly used, with type II systems utilizing Cas9 endonuclease, and type V and type VI systems respectively employing Cas12 and Cas13 endonucleases [28] [Fig. 3]. Cas9 is guided by mature crRNA-foreign DNA complexes to the PAM sequence and makes a blunt double-stranded DNA (dsDNA) break. Cas12 is a compact and efficient enzyme that similarly requires a PAM sequence in the target and cis-cleaves specific target dsDNA to generate a 5′ overhang staggered cut while showing trans-cleavage activity towards nearby non-specific single-stranded DNA (ssDNA). Cas13 contains two enzymatically active higher eukaryotes and prokaryotes nucleotide-binding (HEPN) RNase domains that recognize protospacer flanking sequence (PFS) instead of PAM sequence to define and specifically degrade the target ssRNA [28,29] [Fig. 3]. The intrinsic trans-cleavage activity of Cas12 and Cas13 towards neighboring non-specific ssDNA and ssRNA, respectively, was applied to SARS-CoV-2 diagnosis (see the section of CRISPR-based Diagnostics).

Fig. 3.

Schematic presenting the characteristics of Class II CRISPR-Cas effector complexes. The target motifs, substrates, cleavage patterns, nuclease domains, and trans-cleavage activity of Cas9, Cas12, and Cas13 are presented.

Abbreviations: dsDNA: double-stranded DNA; HEPN: higher eukaryotes and prokaryotes nucleotide-binding; N: no; PFS: protospacer flanking site; sgRNA: single-guide RNA; ssDNA, single-stranded DNA; Y: yes.

To date, CRISPR systems have been utilized in various investigative applications, playing essential roles in efficient, multiplexable, and specific DNA/RNA knockdowns to examine genes of interest. CRISPR has also been extensively employed against different infectious diseases to assess viral–host interactions and screen potential therapeutic options [[30], [31], [32]]. With the urgent need for efficient diagnostic testing and the lack of efficient treatment in the current COVID-19 pandemic crisis, innovative approaches utilizing CRISPR technology may offer a promising alternative to existing approaches, and potential applications of CRISPR-Cas in COVID-19 diagnosis, large-scale screening to identify key host factors, and treatment will be reviewed in the following sections.

CRISPR-based diagnostics

Although the RT-qPCR and metagenomic next-generation sequencing (mNGS) assays now widely used in COVID-19 screening are highly effective at detecting SARS-CoV-2 in patient samples [35], these approaches have limited cost-effectiveness, and require expensive equipment, sophisticated processing, and lengthy detection times. CRISPR-Cas systems have therefore been proposed to speed up the diagnostic process, with potential advantages including convenient operations, minimal reagent and instrument requirements, and low cost. Several CRISPR-Cas systems are under development for use in SARS-CoV-2 detection, including Cas9, Cas12, and Cas13 systems. Currently, there are approximately 20 CRISPR-based platforms with high specificity that can be used to distinguish single nucleotide differences and improve diagnostic sensitivity for viral levels as low as 0.4 copies per μL, with minimal off-target activity both in vitro and in vivo. These platforms also have the potential to reduce diagnostic turnaround times to approximately 40 min, a vast improvement on conventional methods [Table 1]. Additionally, CRISPR-based platforms display specificity comparable to that of RT-qPCR (CRISPR: 80%–97.1%; RT-qPCR: 95%–100%), with some platforms showing 100% specificity in the negative ratio of the assay when the virus is absent in samples (similar to RT-qPCR).

Table 1.

Summary of CRISPR-Cas-based platforms for COVID-19 diagnosis.

Platform Name	Effector	Nucleic Acid Extraction	Sample Amplificationa, b, c, d	Single-Base Resolution	Limit of Detection (copies/μL)	Number of Clinical Samples	Sensitivity/Specificitya, b, c, d	Turnaround Timec	Target Gene	One Pot vs Two Pots	Readoutsd	Ref
FELUDA	Cas9	Y	RT-PCR/RT-RPA	N	∼400	473	100%/97%	∼2h	N	Both	L	[33]
CASLFA	Cas9	Y	RT-RPA	N	4	64	100%/97%	∼1h	ORF1ab, E	Two pots	L	[34]
RAY	Cas9	Y	RT-PCR	Y	NA	NA	NA	∼1.5h	S	Two pots	L, A	[35]
DETECTR	Cas12	Y	RT-LAMP	N	10	11	95%/100%	∼1 h 40 min	E, N	Two pots	F, L	[36]
OR-DETECTOR	Cas12	Y	RT-RPA	N	2.5	NA	NA	∼1 h 50 min	RdRp, N	One pot	L	[37]
MeCas12a	Cas12	Y	RT-RAA	Y	5	24	100%/100%	∼1 h 45 min	E	Two pots	N	[38]
Contamination-free visual Cas12a	Cas12	Y	RT-LAMP	N	20	10	100%/100%	∼1 h 45 min	ORF	One pot	N	[39]
CRISPR-Cas12a-NER	Cas12	Y	RT-RAA	N	10	31	100%	∼1 h 45 min	E	Two pots	N	[40]
CASdetec	Cas12	Y	RT-RAA	N	10	NA	NA	∼2 h	RdRp	One pot	N	[41]
AIOD-CRISPR	Cas12	Y	N	N	5	28	NA	∼1 h 40 min	N	One pot	N	[42]
opvCRISPR	Cas12	Y	RT-LAMP	N	5	26	100%/100%	∼1 h 45 min	S	One pot	N	[43]
CRISPR-FDS	Cas12	Y	RT-RPA	N	2	29	100%	∼1 h 40 min	ORF1ab, N	Two pots	F	[44]
ENHANCE	Cas12	Y	RT-LAMP	N	3–300	NA	NA	∼1 h 50 min	N	Two pots	L	[45]
iSCAN	Cas12	Y	RT-LAMP	N	0.4	24	86%/100%	∼1 h 40 min	E, N	Two pots	F, L	[46]
Microfluidic ITP-CRISPR-based assay	Cas12	N	RT-LAMP	N	10	64	93.8%/100%	∼40 min	E, N	–	E	[47]
SHERLOCK	Cas13	Y	RT-RPA	N	NA	380	100%/100%	∼1 h 40 min	ORF1ab, S and N	Both	F	[48,49]
SHINE	Cas13	N	N	N	>1000	50	90%/100%	∼55 min	ORF1a	One pot	N, L	[50]
DisCoVER	Cas13	N	N	N	200–400	63	93.9%/100%	40 min	N	Two pots	F	[51]
CARMEN	Cas13	Y	RT-RPA	N	NA	NA	NA	NA	NA	Two pots	Chip	[52]
Ultralocalized Cas13 assay	Cas13	Y	LAMP	Y	1	NA	Single-molecule/single nucleotide	NA	ORF1a, N	NA	F	[53]

^aAbbreviations: LAMP, loop-mediated isothermal amplification; N, not required; RT-LAMP, reverse transcription-LAMP; RT-PCR, reverse transcription-polymerase chain reaction; RT-RAA, reverse transcription-recombinase-aided amplification; RT-RPA, reverse transcription-recombinase polymerase amplification.

^b: sensitivity: the positive percent of the assay which compares with the positive ratio of RT-qPCR; specificity: the negative ratio of the assay when the virus is absent in samples (performance relative to RT-qPCR).

^c: time required for nucleic acid extraction, amplification, and detection.

^dAbbreviations: A: agarose electrophoresis; E: epifluorescence microscopy; F: fluorometry, L: lateral flow device; N: naked eyes; CRISPR/Cas12a-NER: CRISPR/Cas12a-based detection with naked eye readout; CRISPR-FDS: CRISPR-based fluorescent detection system; ITP: isotachophoresis; NA, not available; N, no; opvCRISPR: one-pot visual reverse transcription-LAMP-CRISPR; Y, yes.

CRISPR-Cas9-based diagnostics for SARS-CoV-2

FnCAS9 Editor Linked Uniform Detection Assay (FELUDA) was developed using a Cas9 ortholog from Francisella novicida, and has demonstrated high specificity, sensitivity, and minimal off-target activity both in vitro and in vivo for SARS-CoV-2 virus detection [54]. Similar to the current gold-standard of RT-qPCR detection, the FELUDA assay collects a nasopharyngeal swab sample for RNA extraction, followed by reverse transcription of viral RNA to complementary DNA (cDNA) and subsequent amplification in a single step using biotinylated primers [Fig. 4]. The amplification step is critical to enhancing detection probability. The FELUDA assay uses 6-carboxyfluorescein (FAM)-labeled single-guide RNA (sgRNA) designed with PAM to bind with the biotinylated reverse transcription-polymerase chain reaction (RT-PCR) amplicons (nsp8 and N genes) of SARS-CoV-2, following which a combination of amplified viral DNA, guide RNA (gRNA), and Cas9 protein is prepared and immersed with a paper strip. The mixture moves due to lateral flow. Strikingly, a FELUDA-based lateral flow assay was able to recognize SARS-CoV-2 synthetic DNA on a paper strip when SARS-CoV-2-specific ribonucleocapsid (RNP) was used to interrogate the substrate, with a single line indicating a negative result and a double line signifying a positive outcome, akin to a home pregnancy test. Results are available within 2 h [54]. Furthermore, FELUDA is capable of detecting SARS-CoV-2 variants [54,55].

Fig. 4.

Schematic of experimental workflows for COVID-19 diagnostic tests based on the CRISPR-CAS system. Nucleic acids from different sources of clinical samples (e.g. nasopharyngeal swabs, blood samples, or saliva samples), depending on the CRISPR-Cas system used, are isolated and amplified. In Cas9-based diagnostic tests, isolated RNA is converted to complementary DNA (cDNA) and amplified to multiple copies via biotinylated primers in a one-step reaction. In Cas12-based diagnostic tests, viral RNA signals are amplified via reverse transcription-loop-mediated isothermal amplification (RT-LAMP) under isothermal conditions. However, in Cas13-based diagnostic tests, viral RNA is amplified using reverse transcription-recombinase polymerase amplification (RT-RPA) and cDNA is converted to RNA via in vitro transcription. The Cas-crRNA (CRISPR-derived RNA) complex is added to the amplified nucleic acid mixture, thereby enabling recognition of the SARS-CoV-2 specific sequence and triggering Cas enzyme cleavage activity against substrate or probes. In the readout step, many methods have been developed, including fluorescent probes, colorimetric dyes, lateral flow immunoassay (LFIA) cartridges, or detection by electrokinetic or microfluidic devices.

Abbreviations: RT-LAMP: reverse transcription loop-mediated isothermal amplification; RT-RPA: reverse transcription-recombinase polymerase amplification.

Introductions of point mutations (D10A and H840A) into each domain of the Cas9 endonuclease generates deactivated Cas9, also known as dead Cas9 (dCas9). Such dCas9 can be repurposed to deliver functional cargo to programmed sites by binding to other proteins for gene activation, repression, and base editing in the genome [56]. A colorimetric viral detection approach has been developed based on the CRISPR/dCas9 system, with dCas9-guide RNA (gRNA) complexes fixed in wells, to which viral lysate and biotin-PAMmer (biotin-protospacer adjacent motif (PAM)-presenting oligonucleotide) are then added. Following incubation and washing, wells were treated with streptavidin horseradish peroxidase (HRP) and the colorimetric substrate. The presence of viral RNA is manifested as a yellow color generated by substrate oxidation [57]. These CRISPR-Cas9 detection platforms are summarized in Table 1.

CRISPR-Cas12-based diganostics for SARS-CoV-2 infection

DNA endonuclease-targeted CRISPR trans reporter ( DETECTR) is a diagnostic technique that utilizes the trans-cleavage activity of Cas12a to collaterally cleave a ssDNA reporter (probe) containing a fluorophore and a quencher at each end, which subsequently releases quenchers that induce a fluorescent signal, thereby allowing fast and effective detection even at low viral concentrations [58] [Fig. 4]. DETECTR uses reverse transcription-loop-mediated isothermal amplification (RT-LAMP) to amplify the E and N genes of SARS-CoV-2, after which cleavage of ssDNA confirms virus detection. DETECTR has also been optimized for processing at 62 °C to enable faster detection (30–40 min) than current RT-qPCR methods, and can provide visualized results on a lateral flow strip [45].

RT-LAMP enables multiplexing with various methods such as magnetic precipitation, colorimetric substrates, agarose gel electrophoresis, and fluorescence, and this offers distinct advantages in terms of rapidity, sensitivity, and specificity, with the potential to detect SARS-CoV-2 or any infectious pathogen in less than 1 h, even at low concentrations [59,60]. Efforts to optimize processing procedures and simplify reagent and instrument requirements are ongoing, and include the use of manganese-enhanced Cas12a (MeCas12a) or Cas12b, engineered crRNAs and target sequences, and all-in-one tube (one-pot) reactions to improve sensitivity, specificity, and rapidity, while also reducing the possibility of carryover contamination [44,46,47]. Other optimization mechanisms such as microfluidics, electrochemical and electronic methods, and fluorescence are also being developed [Table 1].

CRISPR-Cas13-based diganostics for SARS-CoV-2 infection

Specific High-sensitivity Enzymatic Reporter Unlocking (SHERLOCK) utilizes isothermal recombinase polymerase amplification (RPA) or reverse transcription-RPA (RT-RPA), and makes use of the Cas13a-crRNA complex to precisely recognize target RNA [61,62] [Fig. 4]. Similar to Cas12, Cas13a can cleave collateral ssRNA alongside target RNAs, and this results in the cleavage of non-target RNA coupled to a fluorescent reporter, similar to DETECTR, to induce a fluorescent signal for fast and effective detection. Since the outbreak of COVID-19, the Feng Zhang lab has redesigned primers and guided RNA to target the SARS-CoV-2 ORF1ab and S genes [63]. Within an hour, SHERLOCK can identify virus-specific nucleic acid sequences and provide a visual readout via a lateral flow device [61,62]. Several Cas13-based systems have demonstrated strong diagnostic performance against SARS-CoV-2 [Table 1], and researchers are continuing to optimize these methods by simplifying RNA extraction procedures, facilitating direct detection of viral RNA to skip the amplification and/or in vitro transcription process, or integrating microfluidic technology to enable multiplexed analysis of multiple pathogens [[50], [51], [52],[63], [64], [65]].

CRISPR-Cas-based diagnostic systems for COVID-19 have shown tremendous promise in speed, precision, sensitivity, and cost-effectiveness, and can be rapidly modified to target new infections caused by emerging viruses, but their clinical efficacy and false positive rate remain unclear. Cas12-based DETECTR and Cas13-based SHERLOCK have been shown to be as accurate as RT-qPCR, but are much faster and avoid key RT-qPCR limitations such as the need for expert personnel and a sophisticated thermocycler, as well as requirements for lab consumables, extraction kits, and personal protective equipment. However, one of the main drawbacks of CRISPR-Cas-based diagnostics is the off-target effect, which can result in misdiagnosis. The use of sophisticated bioinformatics algorithms to select the most relevant gRNAs may limit such unwanted consequences, but a precise understanding of the target virus genome is necessary to inform these algorithms, and it can be time-consuming to collect and validate this information [66,67]. The Cas9-based FELUDA approach has demonstrated high sensitivity and specificity, with minimal technical expertise and equipment required, and thus the assay is easier and quicker than RT-qPCR [33]. Furthermore, FELUDA combined with rapid variant assay (RAY) has been developed to distinguish SARS-CoV-2 variants and other coronavirus strains using dFnCas9 [33,35]. For DETECTR and other fluorescence readout-based approaches, another important disadvantage is the inability to distinguish between positive and negative tests. To address this issue, fluorescence values from fluorometry or ELISA reader were normalized to the highest value (within the N gene, E gene, or RNase P set), with a positive threshold of five standard deviations above the background [36]. In SHERLOCK, a threshold has also been set to eliminate the negative effect of fluorescence background on detection accuracy [68]. In addition, the loop-mediated isothermal amplification (LAMP) and RPA isothermal amplification techniques widely used in CRISPR-Cas-based diagnostics face a lack of commercial resources, particularly for RPA (primarily supported by the TwistDx company), and purified recombinant Cas13 and Cas12 are not broadly available commercially, thus presenting barriers to the widespread utilization of these diagnostics [66].

Identification of host factors in severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) replication by CRISPR-based screening

A better understanding of host factors involved in the life cycle of SARS-CoV-2 and other human coronaviruses can facilitate the development of therapies against COVID-19 and other CoV-induced diseases. CRISPR knockout (CRISPRKO) and CRISPR activation (CRISPRa) are now the preferred methods to induce complete and sustained loss or gain of function in target gene expression, and CRISPRKO (loss of function) screening can identify pro-viral genes by removing host dependency factors that drive viral infection and cytopathic impact in virus-susceptible cells, while CRISPRa (gain of function) screening can uncover antiviral factors that improve viral restriction when upregulated. These methods have expanded current capabilities in identifying novel host factors that are critical for host–pathogen interaction, and which may be used to develop broad-spectrum antiviral therapeutics [Fig. 5A].

Fig. 5.

Overview of CRISPR-Cas-based screening to identify critical host factors for human coronaviruses. (A) Experimental workflow of CRISPR screening. Coronavirus-permissive cell lines, including Huh 7, Huh 7.5, Vero, and Vero E6 cell lines (non-respiratory related cell lines), as well as A549 and Calu-3 cell lines (lung cancer cell lines), are transduced with either a genome-wide Cas9 knockout library (loss of function) or a deactivated Cas9 (dCas9) activator library (gain of function). In addition, single-guide RNA (sgRNA) library cells were infected with coronaviruses, and the surviving cells were harvested and compared with mock control cells in terms of sgRNA abundance using sequence analysis. (B) Venn diagram depicting the overlap of host factors interacting with coronaviruses according to CRISPR screening results. The top rank of host factors for common cold CoVs (HCoV-229E, NL63, and OC43) and severe CoV (SARS-CoV, MERS-CoV, and SARS-CoV-2) from CRISPR screens are grouped and shown as pan-CoV factors and severe CoV factors. ∗Indicates genes also detected in SARS-CoV-2 protein interactome studies. Red: Pan-CoV factors; Green: Severe-CoV factors.

The performance of the CRISPR screens was robust, as indicated by the top resistance hits of viral entry receptors identified, such as dipeptidyl peptidase-4 (DPP4) for MERS-CoV, human aminopeptidase N (ANPEP) for HCoV-229E, and ACE2 for SARS-CoV, SARS-CoV-2, and HCoV-NL63 [[69], [70], [71], [72], [73], [74]]. Unlike other coronaviruses, HCoV-OC43 lacks a recognized proteinaceous receptor, but alternatively relies on sialic acid [75] or glycosaminoglycans (Neu5, 9Ac2) for cell entry [76]. Comparison of CRISPR screens for SARS-CoV-2 and HCoV-OC43 indicate that they share several overlapping sets of host factors to complete their life cycle [76]. Several genes involved in pathways related to heparan sulfate biosynthetic genes (EXT1, EXT2, EXTL3, B3GALT6, B3GAT3, B4GALT7, SLC35B2, XYLT2, and NDST1) were enriched in the HCoV-OC43 and SARS-CoV-2 screens conducted in Huh-7.5 hepatoma cells [69,76].

Exostosin 1 and 2 (EXT1, EXT2) are endoplasmic reticulum-resident type II transmembrane glycosyltransferase proteins, and have also been identified as pan-CoV cellular proteins associated with virus replication [summarized in Fig. 5B] [69,76,77]. Heparan sulfate chains, which are proteoglycans expressed on the cell surface as part of the extracellular matrix, are extended by EXT1 and EXT2 [78]. The heparan sulfate biosynthesis pathway has also been demonstrated to serve as a common cellular factor for multiple viruses, including the six human coronaviruses in Fig. 5 [79,80]. Consistent with the essential role of heparan sulfate as an attachment factor for mediating interactions with the positively-charged S1 domain of the SARS-CoV-2 spike protein when viruses attach to the ACE2 receptor, several host proteases involved in viral cell entry have also been identified, including TMPRSS2, furin, and cathepsin B or L1, depending on the target cell type [73,[81], [82], [83]]. This is highlighted in screens performed using lung epithelial origin Calu-3 and A549 cell lines, as well as the hepatocellular carcinoma cell line Huh-7, which show ACE2, TMPRSS2, and cathepsin L (CTSL) as top hits [69,76,77,[84], [85], [86], [87]].

Two high-throughput bidirectional CRISPR screens demonstrate that the adaptin AP-1 complex subunit gamma-1 (AP1G1) and the membrane-tethered mucins serve as SARS-CoV-2 essential factors in viral replication [85,87]. AP1G1 is a clathrin-adaptor protein that is expressed in the trans-Golgi network and the recycling endosomes, and is responsible for the proper localization of other plasma membrane components. AP1G1 adaptins also play essential roles in SARS-CoV-2 attachment, endocytosis, or egress in the host cell. In addition, the ubiquitously present membrane-associated mucins (MUC1, MUC4, MUC13, and MUC21) were shown in lung epithelial Calu-3 cells to have antiviral properties, acting to hide surface receptors and suppress viral penetration of the host cell.

Comparison of the results from CRISPRKO and CRISPRa screens conducted in different cell types (Huh 7, Calu3, A549, Vero, and Vero 6) for SARS-CoV, SARS-CoV-2, and other coronaviruses identified 8 pan-CoV genes (transmembrane protein 41B (TMEM41B), EXT1, EXT2, PIK3C3, KDM6A, RAB7A, SCAP, and SLC35B2) and 3 severe-CoV genes (CTSL, SWI/SNF Related, Matrix Associated, Actin Dependent Regulator of Chromatin, subfamily A, member 4 (SMARCA4), and ARIDIA) of interest, and protein–protein interaction (PPI) studies subsequently revealed expression changes in 9 of these genes (TMEM41B, EXT1, EXT2, RAB7A, SCAP, SLC35B2, CTSL, SMARCA4, and ARIDIA) following SARS-CoV-2 infection [[88], [89], [90]] [Fig. 5B]. A summary of the CRISPRKO and CRISPRa screens reveals that the dominant factors overlap pro-viral/antiviral host factors in identified virus strains and different cell lines, and the proposed roles of host factors identified for human coronaviruses to date are presented in Fig. 6.

Fig. 6.

Overview of putative host factors involved in various stages of the SARS-CoV-2 life cycle as derived from CRISPR screening. The indicated cellular localization of viral proteins and putative host factors (Red: Pan-CoV factors; Green: Severe-CoV factors) is based on published references. Further elucidation and verification of the interaction network between viral proteins and putative host factors is needed.

Abbreviations: DMV: double-membrane vesicle; ER: endoplasmic reticulum; ERGIC: ER-to-Golgi intermediate compartment; nsps: non-structural proteins.

Transmembrane protein 41B (TMEM41B), a protein involved in autophagosome formation, was identified as a common host factor in human coronavirus replication [69,73,76,77,91,92]. TMEM41B knockout has been shown to trigger the abnormal accumulation of intracellular lipid droplets in the early stage of autophagosome formation [93]. During viral replication, coronaviruses modify ER membranes to form membrane-protected viral RNA replication-associated organelles, and TMEM41B is known to be involved in ER membrane modification and rearrangement [92]. Another key pan-viral host factor, PIK3C3, appeared in several CRISPR screens of Huh 7 cell lines [81,84,85] [Fig. 5B]. PIK3C3 encodes the type 3 catalytic subunit of the phosphatidylinositol 3-kinase (PI3K) complex, which mediates the formation of phosphatidylinositol 3-phosphate and is known to be involved in multiple cellular processes, including vesicular trafficking and autophagy [94]. This suggests that CoVs may hijack the host PI3K pathway during the early stages of the viral life cycle, including membrane fusion, endocytosis, translation, and replication initiation [69,76,77].

Besides TMEM41B, the transmembrane protein TMEM106B has also been shown in genome-wide CRISPRKO screens to be a SARS-CoV-2 essential gene [92,93]. TMEM106B is involved in lysosome trafficking and activity in dendritic cells and neurodegenerative disorders [92,93], and given the importance of lysosomes in SARS-CoV-2 infection, TMEM106B may act as a pro-viral host factor in the entry stage of the viral replication cycle [77]. Other key pro-viral factors include SWI/SNF Related, Matrix Associated, Actin Dependent Regulator of Chromatin, subfamily A, member 4 (SMARCA4) and AT-Rich Interaction Domain 1A (ARID1A), both members of the SWItch/Sucrose Non-Fermentable (SWI/SNF) family [87]. SMARCA4 demonstrates helicase and ATPase activities that can modulate gene expression via interaction with ARID1A, which serves as a bridge for the formation of the complex required to remodel chromatin structure [87]. Interestingly, PFI-3, an inhibitor of SMARCA4, shows anti-SARS-CoV-2 activity, implying that chromatin accessibility is essential for severe CoV replication [87]. Further research to identify yet to be discovered host factors in glycosaminoglycan biosynthesis, lipid droplet formation, and vesicle-Golgi membrane fusion is warranted [Fig. 6]. CRISPRKO screening has also been applied to identify host factors involved in the replication of SARS-CoV-2 variants, and results show that GATA6, a zinc finger transcription factor, is an essential factor for regulating ACE2 expression, with corresponding effects on SARS-CoV-2 entry; while the two host factors, KREMEN2 and SETDB1, are Alpha variant-specific candidates involved in viral replication [72]. These results highlight the powerful role of CRISPRKO and CRISPRa screens in revealing host factors that may serve as putative therapeutic targets for current and future coronavirus pandemics.

CRISPR as a treatment platform for COVID-19

Unlike the Cas9 and Cas12 systems, the CRISPR-Cas13d system does not require any PAM or PFS motif to cleave the target sequence, and the prophylactic antiviral CRISPR in human cells (PAC-MAN) system, comprising Cas13d and gRNAs with spacer sequences that precisely complement target viral RNA, was recently developed for the treatment of influenza A virus and SARS-CoV-2 [95] [Fig. 7A]. As bioinformatics analyses of the conserved regions of SARS-CoV-2 suggest that disruption of the RdRp and N sequences may considerably impact virus function and virion formation, infection reporters consisting of the green fluorescent protein (GFP) gene fused with the SARS-CoV-2 RdRp or N sequences were generated and transfected to human lung epithelial A549 cells expressing Cas13d, followed by transfection with different pools of crRNAs to activate the PAC-MAN system [95]. Compared to a pool of non-targeting control crRNAs, the RdRp and N targeting crRNA pools were shown to inhibit GFP reporter expression by 81 and 90%, respectively [95]. Another recent bioinformatics study identified three groups of 2, 6, and 22 crRNAs that could respectively target 50%, 91%, and 100% of all sequenced coronaviruses [95], and the PAC-MAN system can employ such multiple crRNA pools to target different regions of SARS-CoV-2, different SARS-CoV-2 variants, or different coronaviruses to provide comprehensive treatment coverage. This represents a potentially powerful approach for suppressing viral protein function and virus reproduction.

Fig. 7.

Schematic of potential CRISPR-Cas-based therapeutic strategies for COVID-19. (A) Mechanistic overview of the CRISPR-Cas13-based prophylactic antiviral CRISPR in human cells (PAC-MAN ) strategy. (B) Schematic of the proposed AntiBody And CAS fusion systems (ABACAS) strategy and mechanism of action.

Blanchard et al. have recently reported a messenger RNA (mRNA)-encoded Cas13a system that can protect hamsters from SARS-CoV-2 [96], in which the most potent crRNAs targeting SARS-CoV-2 RdRp and N genes, as determined by cell-based assays [95], were delivered along with polymer-formulated Cas13a mRNA via a nebulizer to the respiratory tract of study hamsters [96]. Treatment decreased SARS-CoV-2 replication and related symptoms, and this strategy demonstrates progress in the in vivo use of Cas13a against SARS-CoV-2 [96].

Antibody–drug conjugates were initially developed as anti-cancer drugs, and consist of an antibody covalently linked to a small molecule drug. The AntiBody And CAS fusion systems (ABACAS), which utilize such antibody–drug conjugates, were recently proposed as a potential therapy against COVID-19 [Fig. 7B] [97]. ABACAS comprises Cas13 and an antibody fragment specific to the S protein of SARS-CoV-2, which can recognize and bind to SARS-CoV-2 to promote the selective delivery of Cas13 into infected cells alongside the virus [Fig. 7B]. Upon entering infected cells, ABACAS can identify and cleave viral RNA; moreover, the antibody fragment neutralization activity of ABACAS may interfere with the binding of the viral S protein to ACE2 receptors, thereby reducing endocytosis-mediated viral entry into host cells. In this way, ABACAS can provide a dual strategy to suppress both viral replication and viral infectivity. REGN-COV2 and LY-COV555 represent key SARS-CoV-2 S antibody candidates for the ABACAS system, and both these candidates are currently under investigation [95,[97], [98], [99]]. Alternatively, the antibody moiety can be replaced with a small peptide of the ACE binding site with the S protein of SARS-CoV-2, such as the peptidase domain, to ensure selective delivery of the CRISPR part into the infected cells to neutralize viral particles [97].

Conclusion and future perspectives

Although the CRISPR-Cas system has significant therapeutic potential against COVID-19, off-target effects remain a major issue. Such effects may be reduced by optimizing the design of sgRNA and improving the specificity of the Cas protein. Another significant hurdle to clinical deployment involves the safe and effective delivery of these therapies in vivo [96]. In addition, most research to date has been conducted on animal models, and further studies in humans are needed.

The development of vaccines and drugs is quite time-intensive, and the rapid mutation rate of SARS-CoV-2 variants may render existing vaccines and drug treatments ineffective [100, 101]. Although the development of a new generation of vaccines has accelerated the time from development to injection, it still cannot match the pace of virus mutation. In this sense, the CRISPR-Cas system has an inherent advantage against SARS-CoV-2 in that the sequenced genomes of new variants can be rapidly assessed using bioinformatics methods, which can quickly identify potent crRNAs that can then be applied to develop or update CRISPR-Cas-based diagnostics and treatments. CRISPR-based screens combined with proteomics studies can also help to efficiently build viral–host interaction networks that can be employed to understand the mechanics of viral replication and pathogenesis, and assist with drug target/candidate selection.

In conclusion, CRISPR-Cas technology has transformed gene editing, and enabled advancements in screening, diagnostics, and treatment against infectious agents, including the ongoing SARS-CoV-2 pandemic and future emerging pathogens.

Funding

This study was financially supported by the Chang Gung Memorial Hospital, Taoyuan, Taiwan (grants BMRP416, CMRPD1K0241-2 and 1M0881-3) and the Ministry of Science and Technology (grants 109-2320-B-182-026-MY3, 109-2327-B-182-003, and 111-2321-B-182-001). This research also received funding from the Research Center for Emerging Viral Infections (Featured Areas Research Center Program) within the framework of the Higher Education Sprout Project conducted under the auspices of the Ministry of Education and the Ministry of Science and Technology (grants 110-2634-F-182-001 and 111-2634-F182-001). The funders had no role in the design of the study, the collection, analysis, and interpretation of data, the preparation of the manuscript, and the decision to submit the article for publication.

Availability of data and materials

Not applicable.

Ethics approval and consent to participate

Not applicable.

Conflicts of interest

None.