CRISPR-Based Detection of Viral Hemorrhagic Fevers at the Point of Care

Abstract

Viral hemorrhagic fevers (VHFs) are highly lethal diseases that often present non-specific, influenza-like symptoms in their early stages, making clinical recognition and differentiation from other febrile illnesses difficult. This overlap underscores the critical need for diagnostic tests that are both sensitive and specific. Point-of-care (POC) diagnostic tests are an invaluable tool for detecting and controlling the spread of pathogens that threaten public health, such as VHFs, as these require fast, accurate diagnostics to ensure biosafety and appropriate mobilization of resources during outbreaks. Current molecular and serological diagnostic tests, while efficient and effective, lack the characteristics required of a POC test (POCT) to quickly and easily respond to a VHF outbreak while maintaining a low cost. Clustered regularly interspaced short palindromic repeats (CRISPR)-based diagnostic tests have gained popularity as POCTs due to their inherent attractive qualities, including high sensitivity and specificity, adaptability, low cost, quick turnaround time, and ease of use. However, studies on the development of CRISPR-based POC diagnostic tests for VHFs are limited. This review summarizes the current CRISPR-based POCTs for VHFs, including Ebola virus (EBOV), Lassa virus (LASV), Dengue virus (DENV), and Crimean–Congo hemorrhagic fever virus (CCHF). The isothermal pre-amplification methods commonly paired with CRISPR-based tests, such as loop-mediated isothermal amplification (LAMP) and recombinase polymerase amplification (RPA), are also discussed.

1. Introduction

The viral hemorrhagic fever (VHF) disease group consists of many viruses from various families, including Filoviridae, Arenaviridae, Bunyaviridae, and Flaviviridae, which can cause severe diseases with high fatality rates in human populations [1,2]. The term ‘viral hemorrhagic fever’ is often used to describe these viruses due to similarities in disease progression and characteristics, including the onset of non-specific “flu-like” symptoms, followed by severe disease progression, increased transmissibility, high fatality rates, occasional propensity for hemorrhaging, and lack of treatment options [1,2,3]. The high transmissibility of VHFs was exemplified during the 2014–2016 West African Ebola virus (EBOV) outbreak, affecting mainly Guinea, Liberia, and Sierra Leone, resulting in over 28,000 cases and 11,000 deaths, with a fatality rate of 39% [4]. The magnitude of this outbreak also led to imported travel-associated cases to other countries, including Europe and North America, reinforcing the broader public health threat these viruses pose [4]. Similarly, Dengue virus (DENV), which has the potential to develop into severe dengue and hemorrhaging, is an endemic disease in over 100 countries [5,6]. Crimean–Congo hemorrhagic fever (CCHF) is another important VHF spreading outside its ecological niche, with contributing factors such as climate change and host migration increasing the risk of cases [7]. As well, Lassa Fever (LASV) is endemic in West Africa, with 100,000 to 300,000 cases reported annually, although the true number of cases is likely underreported [8,9]. In addition, high fatality rates are also of concern for CCHF and EBOV, which show fatality rates of up to 40% and 90%, respectively [10,11].

The presentation of non-specific symptoms and limited treatment and vaccination options available for these viruses add to the public health threat that these viruses pose. During early infection, symptoms often present as general and non-specific, including fever, chills, diarrhea, myalgia, and malaise, making it difficult to clinically differentiate between VHFs and other circulating diseases, such as malaria [1,2,3]. This difficulty can potentially lead to misdiagnosis, further spread, delayed patient care, and symptomatic treatments [1,3,6,12,13]. In addition to diagnostic uncertainty, lack of treatment options contributes to the difficulty in developing treatment plans and therefore often relies on supportive care and variably effective antiviral medications to address symptoms [11,14,15]. As well, there are no available vaccines for CCHF and LASV [11,14]. Although vaccines have been developed against DENV and EBOV, they come with specific recommendations and limitations [5,10]. With the increasing frequency of outbreaks, severity of infections, high fatality rates, and the presentation of non-specific symptoms, national and international efforts to control infections and prevent deaths through improved POC diagnostic capabilities are of paramount concern.

Early detection and differentiation of circulating viruses support public health by enabling timely testing, accurate diagnosis, appropriate treatment, and rapid isolation of VHF cases to prevent further transmission [1,16]. The World Health Organization (WHO) originally provided requirements for an ideal POCT termed ASSURED (Affordable, Sensitive, Specific, User-friendly, Rapid and Robust, Equipment-free and Deliverable to end users), which has since been updated by researchers to REASSURED, adding Real-time connectivity, Ease of specimen collection, and Environmental friendliness to reflect modern technology and current challenges [17]. In essence, the ideal POCT would be connected in real-time using limited equipment or mobile phones for streamlined data collection and consistent results, have simplified collection methodology, be manufactured using recyclable materials, have a low cost per assay, limit false positives and negatives, be easily used by the layperson, have a quick turnaround time, withstand logistics associated with travel, and be accessible in low resource settings [17,18].

Serological POCTs have been developed for VHFs such as DENV and EBOV (Sudan virus, Bundibugyo virus, and EBOV); however, there are several disadvantages to using serological-based tests to diagnose acute infections [19,20]. For example, seroconversion can take days to weeks, with some patients succumbing to VHF infection before producing detectable antibodies, thus severely limiting the use of serological-based testing at the POC [16,21,22]. Other disadvantages of antibody-based testing in the POC setting include difficulty differentiating current or past infection status and the effects of cross-reactivity [1,23,24].

Recently, the use of clustered regularly interspaced short palindromic repeats (CRISPR) in POC diagnostics has gained traction because of its ability to meet the REASSURED requirements, such as low cost per assay, adaptability, high sensitivity and specificity, and short turnaround time, in addition to deliverability to isolated communities and minimal equipment usage [25,26]. The goal of CRISPR-based diagnostics is to meet the REASSURED criteria in developing a rapid, field-deployable test for persons with limited laboratory skills to perform the test at the POC.

This review will discuss the current molecular-based diagnostic tests for VHFs, and the CRISPR technology currently being investigated for use in diagnostic testing will be outlined. In addition, isothermal pre-amplification methodologies, specifically those used in VHF detection, such as recombinase polymerase amplification (RPA), closely related recombinase aided amplification (RAA), and loop-mediated isothermal amplification (LAMP), will be discussed. Finally, the current development of CRISPR-based diagnostic tests for VHFs that have demonstrated progress toward meeting the POC criteria will be summarized.

2. Current Diagnostics

2.1. Polymerase Chain Reaction

Current diagnostic tests for VHFs include the gold standard for nucleic acid detection of pathogens, polymerase chain reaction (PCR) [1,2]. In the context of VHFs, which are all ribonucleic acid (RNA)-based, Quantitative Reverse-Transcription PCR (RT-qPCR) uses a reverse transcriptase enzyme, followed by temperature cycling for probe-based detection that tracks the fluorescence produced in real-time, to quantify samples as the complementary deoxyribonucleic acid (cDNA) template is amplified [2,27]. The use of probes allows for multiplexing, which can detect multiple targets at once [27]. While PCR is highly sensitive and specific, it typically lacks feasibility as a POC option due to the complexity of reactions and machinery [26,28]. PCR tests offer higher sensitivity than antigen-based rapid tests but are still prone to false negatives in the early stages of infection [3,29]. For example, it can take 3–10 days post-symptom onset for an RT-PCR test to display a positive EBOV result [3]. A false negative may be encountered if the test is administered too early during an infection, as there may not be enough RNA present to elicit a positive result, necessitating repeat testing at a later time point [3].

Other disadvantages of using PCR during an outbreak include its lengthy cycling times, reliance on expensive laboratory-based equipment and trained technicians, and the logistical challenges of transporting samples to a laboratory capable of performing diagnostics on risk group 4 pathogens [3,26,28,30,31]. In resource-limited regions, there may not be adequate or stable power sources to run specialized equipment such as a thermocycler or appropriate cold storage for samples and reagents [26,32]. As well, PCR can be costly, with a single reaction costing up to USD 125 [33]. These factors highlight the need for improved diagnostic tests and reinforce the importance of developing accurate, reliable, and robust POCTs that can be deployed in the field to provide results in a timely fashion without these special considerations.

2.2. Antigen-Based Tests

Antigen-based tests, including lateral flow-based tests or enzyme-linked immunosorbent assays (ELISAs), can detect circulating viral proteins in the body to indicate acute infections, a factor which is especially beneficial during outbreaks [31,34]. The benefits of antigen-based diagnostic tests include their ease of use outside the laboratory with minimal resources and training, affordability, rapid detection capabilities at the POC compared to PCR, and ability to be combined with antibody testing for further sensitivity and specificity [16,19,31,34]. In the case of EBOV, antigen-based tests have been described as preferable to antibody-based tests due to a detectable level of antigen that occurs prior to antibody development [22]. The ReEBOV Antigen Rapid Test detects viral protein 40 (VP40) in whole blood and plasma with a sensitivity and specificity of 100% and 92.2% when performed at the POC, respectively [31]. The ReLASV Pan-Lassa Antigen Rapid Test detects nucleoprotein in whole blood, plasma, and serum with a sensitivity and specificity over 80% and 90%, respectively [34]. Similarly, the Bio-Rad NS1 Ag Strip was shown to detect DENV with 61.6% sensitivity and 100% specificity [19].

However, antigen-based rapid tests do not meet all the qualities desired for a diagnostic tool in the POC setting. Limitations of antigen-based tests include cold storage, long production times, and variable sensitivity and specificity compared to PCR [2,31,34,35,36,37]. While potentially useful for triage, antigen-based tests may produce false negative results during early infection if the virus is not circulating in levels above the detection threshold, which may necessitate a follow-up test with PCR [31,36]. In other cases, infections may be cleared too rapidly, rendering antigen-based tests unsuitable, such as during DENV infection [2]. These limitations could result in repetitive testing, potentially leading to further spread of the pathogen and delayed patient care, both of which are of great concern in the case of a VHF infection [21,37].

Although these tests are unsuitable for POC diagnostics, they remain invaluable when used in specific contexts and in conjunction with other tests and clinical information. An easy-to-use, fast, and reliable CRISPR-based POCT that can address the limitations of current diagnostic tests could contribute to limiting transmission and increasing effective early patient management for future outbreaks [7,37].

3. CRISPR

3.1. CRISPR Background

CRISPR was first identified in Escherichia coli in 1987 as a distinctive genetic feature, but its role as a prokaryotic adaptive immune system was not recognized until years later [38]. Since its discovery, researchers have found that the characteristic short repetitive DNA sequences of the CRISPR array and its CRISPR-associated (Cas) enzymes protect prokaryotes against foreign genetic material that phages and plasmids can introduce [39,40]. It was soon realized that the unique ability of the CRISPR system to create genetic modifications could be exploited in genetic engineering and therapeutics. The applications of CRISPR range from preventing the transfer of gain-of-function elements, such as antibiotic resistance, as a therapy for sickle cell disease, to agricultural benefits [39,41,42]. CRISPR-based technology has expanded into molecular-based detection of infectious diseases and has since gained traction in POC diagnostics for its potential to meet the requirements of an ideal POC assay [25,26,43].

Briefly, the CRISPR genome is composed of the CRISPR array, which contains alternating spacer and repeat sequences that transcribe into CRISPR RNA (crRNA) and are adjacent to the trans-activating crRNA (tracrRNA) and Cas enzyme sequences [39,40,44,45,46]. The CRISPR response is commonly divided into three stages: adaptation, expression, and interference [39,44,45,46,47]. In the adaptation stage, the CRISPR system identifies and stores unique segments of foreign genetic material (spacers) between a series of short repetitive host DNA sequences (repeats) in the CRISPR array “library”, where it is stored for the next time the cell encounters the same pathogen or gene (Figure 1) [45,46]. To differentiate invading foreign nucleic acid, a short nucleotide sequence on the end of the spacer sequence, termed protospacer adjacent motif (PAM), is often identified to recognize the foreign pathogen and aid in binding [40,47,48]. If the same genetic material is reencountered, the expression stage begins, and the unique spacer sequences are then transcribed into crRNA, which pairs with a tracrRNA to form a guide RNA (gRNA) [45,49]. The gRNA then pairs with the Cas enzyme to form a ribonucleoprotein, which guides the Cas enzyme to the targeted, complementary sequence from the invading pathogen, termed the protospacer [45,49]. Once the target sequence and PAM site are detected, the Cas nuclease cleaves the target to destroy the foreign material and prevent infection of the cell, called the interference stage [45]. This property has been applied to POC diagnostic tests where certain Cas enzymes can alert the user of the results of the test using their trans-cleavage properties on fluorescent reporters, which will be discussed further below.

Figure 1.

Three stages of the CRISPR adaptive response. Adaptation: foreign genetic material is incorporated into the CRISPR array. Expression: ribonucleoprotein is formed following tracrRNA, crRNA, and Cas expression. Interference: foreign genetic material is cleaved. Adapted from [39,45]. Created in BioRender.com.

3.2. Cas Enzymes

CRISPR–Cas enzymes display endonuclease activity responsible for cleaving foreign genetic material, of which several types and subtypes have been identified and formally organized into two classes: class I and II [40,45,46,50]. Class I enzymes function as a multi-unit complex and characterize the majority of CRISPR loci, and are further divided into three subtypes: I, III, and IV [40,45,46]. Class II enzymes, which are most relevant to this review, function as a single enzyme and are also divided into three subtypes: II, V, and VI [40,45,46]. Class II type V CRISPR–Cas enzymes include DNA-cleaving Cas12 and Cas14, while type VI includes the RNA-cleaving Cas13 enzyme [40,45,46,51]. Certain enzymes within class II, such as Cas12, Cas13, and Cas14, have demonstrated trans-effects, also known as off-target, collateral, or indiscriminate cleavage [40,45]. Once the Cas enzyme becomes activated by target recognition, the enzyme demonstrates trans-cleavage activity of surrounding nucleic acids [40,46]. This property has been exploited by including nucleic acid reporters coupled with a fluorophore and quencher into the diagnostic test, which produces fluorescence once cleaved [40,46]. Activation of the Cas enzyme by target acquisition and the resulting off-target activity in a CRISPR-based POC diagnostic test can be measured by the increase in fluorescence due to the cleavage of a nearby probe, indicating a positive pathogen detection in the sample [40,46].

3.3. Pre-Amplification Methods

Amplification is a crucial step in nucleic acid-based detection to reach a suitable level of detection, such as during CRISPR-based diagnostics [26,52]. Several pre-amplification methods can be used to increase the genetic material present for detection. Specifically, isothermal pre-amplification methods are attractive for use in CRISPR-based POC diagnostic tests because of their minimal equipment requirements [26]. Many isothermal amplification methods have been paired with the CRISPR–Cas system for sensitive and specific pathogen detection, such as LAMP, RPA, the closely related RAA, nucleic acid sequence-based amplification (NASBA), strand displacement amplification (SDA), helicase-dependent amplification (HDA), and exponential amplification reaction (EXPAR) [7,35,37,48,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73]. While PCR is considered the gold standard for molecular-based diagnostics, PCR is not a viable option as a pre-amplification method prior to CRISPR-based detection at the POC due to the limitations previously discussed. This review will focus on the pre-amplification methods commonly used in CRISPR-based POC diagnostics, such as RPA/RAA and LAMP.

3.3.1. Recombinase Polymerase Amplification

RPA is one method of isothermal pre-amplification that can exponentially amplify targeted DNA in under 30 min [52]. This amplification method uses phage T4 uvsX recombinase, strand-displacing Bacillus subtilis (Bsu) polymerase I, and gp32 proteins for stabilization of single-stranded DNA (ssDNA) (Figure 2) [52]. The isothermal capabilities of RPA allow the reaction to run at 37 °C using only minimal equipment such as a fluorometer or lateral flow strips [52]. Due to the relatively low operating temperature, RPA could potentially run at ambient temperature or utilize body heat alone in locations where a heat block may not be feasible, creating a useful POC option [7]. RPA is highly sensitive, detecting as little as 10 copies of methicillin-resistant Staphylococcus aureus (MRSA) using lateral flow strips [52]. RPA also demonstrates multiplexing capabilities, allowing for the amplification and detection of multiple targets at the same time, as shown with MRSA, Zika virus (ZIKV), and DENV [52,62]. As well, one study found that most inhibitors of PCR do not affect RPA, even at high concentrations [74]. However, out of the inhibitors tested, whole blood and sodium dodecyl sulfate impeded the reaction [74]. Other limitations include RPA reagent sensitivity diminishing if not stored properly, which can be problematic in resource-limited areas; however, lyophilization is a possibility [26,69,75]. In terms of VHFs, RPA pre-amplification has been paired with CRISPR–Cas12 and Cas13 for the detection of DENV, EBOV, LASV, and CCHF, which are discussed in further detail below [7,35,37,73,76].

Figure 2.

RPA-based isothermal amplification. RPA creates one new copy of the template using strand-displacing Bsu polymerase, UvsX recombinase, and stabilizing gp32 proteins. With the addition of adenosine triphosphate (ATP), the recombinase displays cooperative binding of the forward and reverse primers and forms a nucleoprotein complex that scans the template for the target site. Once identified, strand invasion occurs, and the polymerase extends the DNA from both primer sites, synthesizing one new copy of the template. Adapted from [52]. Created in BioRender.com.

3.3.2. Recombinase Aided Amplification

Recombinase aided amplification (RAA) is similar to RPA and has been comprehensively reviewed in ref. [77]. In summary, RPA and RAA share similarities in the reaction temperature, sensitivity, specificity, and rapid diagnostic capabilities at the POC [77]. RAA employs a highly similar mechanism to RPA, involving recombinase binding of the primers and strand invasion of the target site aided with single-stranded DNA binding proteins (SSB), followed by a strand-displacing polymerase for isothermal amplification [70,77]. However, RAA has been more widely developed and differs in the origin of the enzymes used, such as bacterial- or fungal-derived recombinase, as opposed to the phage T4-derived proteins found in RPA [70,72,77]. RAA and reverse-transcription-coupled RAA (RT-RAA) have been used alone to detect EBOV and paired with CRISPR–Cas12 and Cas13 to detect LASV, DENV, and EBOV [70,71,72,78].

3.3.3. Loop-Mediated Isothermal Amplification

Similarly to RPA, LAMP or reverse-transcription-coupled LAMP (RT-LAMP) is another isothermal method of nucleic acid amplification that can produce billions of copies of DNA in under an hour [79]. LAMP shows high specificity by using four to six primers that target six to eight regions of the target strand [79]. The primers are termed forward inner primer (FIP), backward inner primer (BIP), forward outer primer (F3 or FOP), and backward outer primer (B3 or BOP) [79]. It was later found that by adding two additional loop primers, loop forward (Loop F) and loop backward (Loop B), amplification can occur even more rapidly [80]. Using these primers and a strand-displacing polymerase, LAMP amplifies and elongates the target sequence, creating many stem-loop and cauliflower-like structures of various sizes (Figure 3) [79].

Figure 3.

LAMP with additional loop primers. The binding of FIP synthesizes a complement to the target strand. F3 performs strand displacement synthesis, releasing a new strand containing the complement to the F1 region. The process is repeated with BIP and B3, forming a strand with complement regions on both ends that fold and form a dumbbell structure. Self-priming initiates the formation of stem-loop structures, followed by the amplification and elongation cycling of LAMP structures [79,80]. Created in BioRender.com.

The benefits of LAMP include its fast reaction time and minimal equipment requirements, such as a water bath or heating block, to maintain the reaction at approximately 60–65 °C, unlike PCR, which requires the use of a laboratory-based thermocycler [79]. While LAMP runs at a higher temperature than RPA, it can be paired with the thermostable Cas12b for CRISPR-based diagnostics [57,60]. LAMP constituents can be lyophilized, and multiple real-time and end-point readout types have been reviewed, such as turbidity, intercalating dyes, and gel electrophoresis [81]. As well, multiple techniques for multiplexing (mLAMP) have also been developed and thoroughly reviewed [81,82]. In terms of sensitivity, several studies have shown LAMP to outperform conventional PCR or real-time PCR in some contexts, while in other cases, LAMP is comparable to or less sensitive than PCR [9,81,83,84,85,86,87,88,89,90,91,92,93,94]. Similarly to RPA, LAMP is inhibited by the same inhibitors as PCR, but is more robust and can function at higher concentrations than PCR [95]. For example, urea is a well-known inhibitor of PCR, but LAMP can function at higher concentrations than what would normally be found in a clinical sample, surpassing the capability of PCR [95].

Limitations of LAMP include a lack of cloning ability and complicated sequencing due to the stem-loop and cauliflower-like structures it produces, and the risk of false positives [81,84]. Due to its high sensitivity, LAMP is prone to carryover contamination and is susceptible to amplification in non-template controls due to primer hybridizations, potentially contributing to false positives [81]. Care must be taken to prevent contamination by using dedicated spaces with ventilation, sterile techniques, and filtered pipette tips [81,84]. As well, easily confirming the amplified target by band size through gel electrophoresis is prevented by the “ladder” effect of LAMP, which can be mediated by running a restriction enzyme digest instead [79,81]. In addition, primer design can be difficult, so primer design tools are available, such as New England Biolabs LAMP Primer Design Tool (NEB LAMP) and PrimerExplorer V5 (PrimerExplorer V5), but the primer design may still need to be performed by hand if targeting a very specific area, or redesigned in the case of non-template amplification [81,96,97]. In terms of VHFs, and to the best of our knowledge, RT-LAMP has only been used alone and not paired with CRISPR to detect LASV, Rift Valley Fever virus (RVFV), Marburg virus (MARV), and EBOV [9,83,86,87,88,89,90,91,92,93,94,98,99,100,101].

3.3.4. Amplification-Free

Amplification-free CRISPR-based tests have been considered in order to streamline the process by removing the pre-amplification step for targets such as severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), human papillomavirus (HPV), and Monkey Pox viral plasmids [102,103,104]. The use of a pre-amplification method may introduce errors such as non-specific amplification, and extra steps in the workflow may introduce contamination, leading to false results [104,105]. Logistics such as running temperature of the assay and appropriate cold-storage of reagents must also be considered for ease of use at the POC [75,106]. Additionally, a pre-amplification step may add extra time to the assay when rapid results are desired [104].

To circumvent these challenges, amplification-free tests have been explored, and high sensitivity and quick detection times can still be achieved at the POC. One method couples Cas13 with the Csm6 enzyme for increased signal amplification [62,106]. In this reaction, Cas13 is first activated by target recognition, which then cleaves the Csm6 oligoadenylate activator, enabling the activation of Csm6 and subsequent reporter cleavage and fluorescence [106]. When Cas13 and Csm6 worked in tandem, 31 copies/uL of RNA could be detected in under one hour, highlighting its potential usefulness during an outbreak [106]. Removing the pre-amplification reagents could potentially lower the cost of the test as well [106].

In some contexts, removing the pre-amplification step may not be beneficial and may require more research into the feasibility of amplification-free methods on testing clinical samples [69,75]. For example, one study showed that the combination of RPA and Cas13a could detect a concentration of 2 attomolar (aM) of viral RNA, while running the assay without the pre-amplification step showed a sensitivity in the 50 femtomolar (fM) range, demonstrating that combining the reaction with RPA resulted in a significant increase in sensitivity [69]. Similar results were found with Cas12a, where the combination with RPA provided aM sensitivity, but not with Cas12a alone [61]. The ability to detect a lower concentration of viral RNA when combined with RPA shows that a pre-amplification step may continue to be a valuable addition to POC testing, especially during infections when the viral titer may be in the low aM range [69].

Amplification-free methods for VHFs such as EBOV and DENV are discussed in more detail in their respective sections [75,107].

3.4. CRISPR-Based Detection Methods

As discussed, CRISPR-based diagnostic tests endeavor to meet the ASSURED/REASSURED criteria, as they are easily adaptable to circulating strains or pathogens, provide faster readout than PCR at the POC, and demonstrate similar sensitivity and specificity as PCR in some contexts [26,35,37,61]. Furthermore, CRISPR-based tests demonstrate low cost, with two studies quoting less than USD 1 per test, and another quoting USD 6 per test [37,69,75]. CRISPR-based POCTs may also use minimal equipment and non-invasive sampling methods, such as saliva [37]. Even though PCR testing is at a disadvantage at the POC compared to other methodologies such as CRISPR or antigen-based assays, it remains the gold standard for diagnostic sensitivity and specificity to which new diagnostic assays are compared.

Several CRISPR-based tests that use different variations of pre-amplification methods and Cas enzymes have been developed, most notably DETECTR, SHERLOCK, HOLMES, and STOP [60,61,62,69,108,109]. The DETECTR (DNA Endonuclease-targeted CRISPR Trans Reporter) method combines Cas12a and RPA in a one-pot format to detect DNA targets within one hour [61]. It was discovered that Cas12a targets double-stranded DNA (dsDNA) and trans-cleaves ssDNA, and this method was applied to rapidly detect two types of cancer-associated HPV, HPV16 and HPV18, from patient anal swabs with high concordance with PCR [61].

Similarly to DETECTR, the SHERLOCK (Specific High-Sensitivity Enzymatic Reporter UnLOCKing) method combines RPA or reverse-transcription RPA (RT-RPA) and Cas13a coupled with T7 transcription to cleave RNA [69]. Using this method, ZIKV was detected in clinical serum and urine samples at aM concentrations [69]. As well, two different strains with single-nucleotide polymorphisms (SNPs) could be differentiated for both ZIKV and DENV [69]. The SHERLOCK method is inexpensive, sensitive, and maintains performance when lyophilized and reconstituted, showing promise for POC diagnostics [69].

In SHERLOCKv2, the multiplexing and signal amplification capabilities of SHERLOCK were further developed [62]. Multiplexing was achieved by combining four different Cas enzymes with their respective reporters containing their preferred motifs in the same reaction [62]. By using multiple fluorescent channels, different targets were detected, which can be applied to identify co-circulating viruses such as influenza or SARS-CoV-2 at the POC [62]. As well, quantitation and lateral flow readout were explored [62]. The combination of RPA, Cas13, and Csm6 was applied to lateral flow strips, which allowed for rapid, sensitive, and mobile detection [62].

Another method that was developed using the Cas12a enzyme is known as HOLMES (One Hour Low-Cost Multipurpose Highly Efficient System) [108]. Similarly to SHERLOCK, HOLMES could detect SNPs to differentiate between virus and vaccine strains and could also detect DNA and RNA viruses with aM sensitivity [108]. However, PCR was used as the pre-amplification method, which is unsatisfactory for POC diagnostics as previously discussed [108].

HOLMESv2 overcame the limitation of HOLMES by combining the thermophilic Cas12b enzyme with LAMP/RT-LAMP and further explored a one-pot system and quantitation [60]. Cas12b is thermostable and demonstrates nuclease activity at higher temperatures, with the optimal temperature determined to be 55 °C for Alicyclobacillus acidoterrestris (AacCas12b) in a one-pot reaction with LAMP, and up to 64 °C for Brevibacillus sp. (BrCas12b) [57,60]. Therefore, Cas12b pairs well with LAMP, which functions optimally at 60–65 °C [57,60,79]. HOLMESv2 demonstrates the practicality of LAMP in a fast, one-step CRISPR-based format for ease of use at the POC [60].

Lastly, the STOPCovid.v2 (SHERLOCK Testing in One-Pot, version 2) method also combines LAMP with Cas12b for rapid, sensitive, and specific detection [109]. The one-pot test uses magnetic bead-based extraction, which increased the sample input and resulted in the detection of concentrations as low as 33 copies/mL, much lower than the RT-qPCR test the assay was compared to [109]. The STOPCovid.v2 assay detected SARS-CoV-2 in less than 45 min with 93.1% sensitivity and 98.5% specificity [109].

4. CRISPR-Based Diagnostic Tests for VHFs at the POC

To the best of our knowledge, CRISPR-based diagnostic tests for VHFs that address use at the POC setting have been developed for DENV, EBOV, LASV, and CCHF. This review will focus on those that employ the isothermal amplification methods, RPA or RAA and LAMP.

4.1. Dengue Virus

A field-deployable diagnostic test for ZIKV and DENV was developed using the SHERLOCK method combining RPA and Cas13 [35]. In this study, a new technique was developed to bring viral nucleic acid extraction outside the laboratory environment, termed HUDSON (Heating Unextracted Diagnostic Samples to Obliterate Nucleases) [35]. HUDSON uses a mix of tris(2-carboxyethyl)phosphine hydrochloride–ethylenediaminetetraacetic acid (TCEP-EDTA) followed by a two-part heating procedure to extract viral nucleic acid and inactivate ribonucleases straight from a clinical sample [35]. Using SHERLOCK, DENV was detected in 24 RT-PCR-positive samples of extracted RNA, while the combination of SHERLOCK and HUDSON detected DENV in 8 clinical serum samples and 3 clinical saliva samples in under one hour [35]. Out of the different samples, saliva demonstrated a lower viral load but also potential for a rapid and non-invasive sample type, which meets one of the REASSURED POCT characteristics discussed above [35]. The assay was further developed to detect a positive sample out of a panel of flaviviruses (ZIKV, DENV, West Nile virus, and Yellow Fever), as well as any combination to detect a possible co-infection [35]. Similarly, the assay can detect which serotype (or multiple) of DENV (serotypes 1–4) is present [35]. The ability to differentiate between multiple viruses or serotypes that may be co-circulating with indistinguishable symptoms would be greatly beneficial in a readily deployable POCT. Lastly, to highlight the adaptability of SHERLOCK and its ability to detect SNPs, a successful new test was created within one week of identifying a new ZIKV mutation, demonstrating the usefulness of CRISPR-based POC diagnostic tests in future outbreaks [35].

In another study, Cas13a and Cas12a were combined to develop a field-deployable lateral flow test to detect DENV [107]. The assay did not use pre-amplification of the DENV RNA, rather it was found that using both Cas13a and Cas12a increased the sensitivity to 190 fM versus 741 fM when Cas13a was used alone, not quite reaching aM concentrations that other assays described here have demonstrated [69,107,108]. The assay was also highly specific to DENV-1, detecting serotype 1 and not serotypes 2, 3, or 4 [107]. While the trans-cleavage activity of the Cas enzymes is still taken advantage of, this test works differently in that fluorescence is observed when the pathogen is not present [107]. When DENV RNA is present, Cas13a is activated and the Cas12a gRNA is consequently degraded, leaving the DNA-based reporter intact to display the test line of the flow strip [107]. When DENV RNA is not present, Cas12a degrades the reporter, illuminating only the control line [107].

Recently, a unique one-pot system was developed to detect all four serotypes of DENV using RT-RPA and Cas12a [76]. In this reaction, RT-RPA was run at its optimal temperature (38 °C) in the reaction tube, while the Cas12a components were stored in the lid of the closed tube [76]. The tube was then centrifuged and the temperature increased (48 °C) for the Cas12a detection of the amplified targets [76]. The test detected plasmids containing DENV 1–4 in under one hour in both tube-based and lateral-flow formats [76]. In this study, no cross-reactivity was observed, and the detection limit was assessed to be lower for the tube-based format at approximately 91.7 copies/test (95% probability), while the lateral-flow test was found to be reliable above 250 copies/test [76].

Another one-pot test was recently developed, combining RAA with Cas13a, termed CRISPR-based Rapid and Efficient Test (CRISPRET) [72]. The one-pot test achieved high sensitivity and specificity when tested on plasma samples of all four serotypes, with an average of 95.8% and 96.6% when compared to qPCR, respectively [72]. As well, each assay was specific to only the serotype tested for [72]. It was noted that while the one-pot system is rapid and reduces the possibility of contamination, combining amplification and detection may result in lower sensitivity and specificity than if these reactions were performed separately [72].

4.2. Ebola and Lassa Virus

The SHERLOCK method was again employed to develop an inexpensive and easy-to-use POCT to successfully detect the Zaire strain of EBOV and clades II and IV of LASV from clinical samples in West Africa (Figure 4) [37]. The assay used both fluorescence and lateral flow readouts and showed that fluorescence-based tests were 10 times more sensitive than lateral flow strips for the detection of LASV, similar to the DENV test described above [37,76]. Fluorescent readout detected LASV clades II and IV to 10 copies and 100 copies/uL, respectively, and EBOV to 10 copies/uL [37]. The assays were also tested on 16 EBOV and 10 LASV clinical samples and found to agree 100% with RT-qPCR, with the LASV clade IV assay outperforming RT-qPCR [37]. Aside from sensitivity, both LASV and EBOV assays were highly specific down to the clade and showed no cross-reactivity with other VHFs [37]. As well, the safety and practicality of HUDSON as a POC extraction/inactivation technique were further assessed [37]. It was found that using either 70 °C for 30 min or 95 °C for 10 min worked to inactivate viruses in whole blood, urine, and saliva [37]. Among these sample types, saliva again showed the highest sensitivity [37]. Lastly, a smartphone application (app) was developed to ensure bias-free, consistent readings of the lateral test strips [37]. The HandLens app maintains high accuracy and demonstrates progress towards real-time connectivity, one of the REASSURED characteristics [37].

Figure 4.

SHERLOCK and HUDSON workflow. Patient sample (such as saliva) is collected, and the HUDSON technique is employed to inactivate and extract viral RNA. Reverse transcription and isothermal amplification of DNA occur, followed by transcription back into RNA. Recognition and cleavage by RNA-based enzyme Cas13 can be read through lateral flow strips or fluorescent real-time readout. Adapted from [37]. Created in BioRender.com.

Due to drawbacks of RPA, such as the complexity and stability of reagents, an amplification-free method was chosen in another study, where a small microfluidic device was developed for Cas13a POC detection of EBOV [75]. Compared to PCR, the assay is quick and inexpensive, with the ability to run 24 samples in 30 min and costing approximately USD 6/assay [75]. Furthermore, the assay does not require complicated sampling procedures, as it is compatible with finger prick tests, making it attractive for use in the POC setting [75]. The assay works by first extracting viral RNA from blood prior to running the sample on the microfluidic chip system, which is also capable of multiplexing and quantitation [75]. The Cas13a gRNA is first added to the detection reservoir, followed by extracted EBOV RNA, and the resulting fluorescence is measured using a small benchtop fluorometer aligned with the microfluidic chip [75]. Without target amplification, the assay has a limit-of-detection of 5.45 × 10⁷ copies/mL [75]. This can be compared to the mean viral load found in initial blood draw samples from patients who presented to the Liberian Ebola Treatment Unit and later succumbed to infection of 1.55 × 10⁷ copies/mL [110]. Due to the amplification-free nature, it is possible that not all infections will be detected, and the application of the device on clinical samples during the early stages of infection, as well as the incorporation of RNA extraction into the system, should be further studied [75].

Recently, a so-called “tube-sleeve-tube” one-pot assay was developed to detect a conserved region of the EBOV VP40 gene using RT-RAA paired with Cas12a [71]. In this assay, the RT-RAA reaction is run in the bottom of the tube, while a smaller, uncapped tube containing the CRISPR reaction is placed into the larger tube upside down [71]. Centrifugation allows the reactions to mix, and the resulting fluorescence can be measured [71]. The test is rapid, detecting 3.6 copies/uL of linearized plasmid in under one hour [71]. As well, the assay did not detect other Ebola virus species or viruses of concern, and when tested with extracted EBOV RNA, showed high concordance with RT-qPCR [71].

RT-RAA has also been paired with Cas13a to detect LASV [70]. In this study, lateral flow strips termed Easy-Readout and Sensitive-Enhanced (ERASE) strips were developed and used to detect plasmids containing a targeted NP region [70]. The detection limit was determined to be 100 copies/uL, which was less sensitive than the 10 copies/uL that fluorescence-based readout and qPCR achieved, a trend seen in other assays described here [37,70,76]. The test is also rapid, uses portable equipment for the POC, and does not detect other viruses such as DENV or ZIKV [70].

4.3. Crimean-Congo Hemorrhagic Fever Virus

In this proof-of-concept study, RPA and Cas13a were combined to target all clades of CCHF [7]. The high sequence diversity between the multiple CCHF clades creates difficulty in targeting a conserved region [7]. However, it was found that using degenerate nucleotides in the gRNA allowed the test to overcome the diversity between clades while maintaining high specificity, as the test did not detect other related viruses, including RVFV [7]. The two-step assay showed high sensitivity with a detection limit of 1 copy/µL and a fast turnaround time of under 40 min, making the assay a strong candidate for further development as a one-pot POCT [7]. It was also noted that the degenerate gRNA demonstrated similar efficacy as other non-degenerate-based gRNAs, while maintaining a greater ability to detect CCHF despite mutations [7]. Although the study did not include testing clinical samples, it successfully detected samples contaminated with DNA, a known contributor to false positives that could be found in clinical samples [7].

Due to the competitive nature and temperature differences of the components involved in designing a one-pot test, a microfluidic device with separate chambers for isothermal amplification and CRISPR–Cas detection was developed, termed the lift-heater centrifugal microfluidic platform (Lift-CM) [73]. The device is potentially suited for the POC, as it is quick, portable, compatible with lyophilization, and low-cost [73]. As well, the device is linked with a smartphone app to set up and monitor the activity of the assay such as the temperature and fluorescence produced in real-time [73]. The Lift-CM platform uses a spatially encoded centrifugal disc (termed SEC-disc) containing eight channels where the extracted samples are added, and dual-heating and centrifugation elements allow for pre-amplification and Cas detection to be carried out separately [73]. Plasmids containing the target such as CCHF and EBOV (10⁴ copies/uL) were detected using RPA combined with Cas12a [73]. The microfluidic device was also compatible with LAMP/Cas12a, highlighting its adaptability for the POC [73].

5. Discussion

The diagnostic landscape for VHFs has historically relied on established modalities such as PCR, antigen detection assays, and serology-based antibody tests [1,2]. While these platforms have proven indispensable in outbreak settings, each carries inherent limitations, including requirements for specialized infrastructure, time-intensive workflows, or reduced sensitivity during early infection [1]. In recent years, CRISPR–Cas systems have emerged as a promising alternative, offering rapid, highly specific, and potentially field-deployable diagnostics [26,43]. By leveraging isothermal amplification strategies and Cas-mediated nucleic acid recognition, CRISPR-based assays present a paradigm shift in point-of-care testing [26]. A critical comparison of these approaches highlights not only the strengths and shortcomings of conventional diagnostics for VHFs but also the potential of CRISPR-based platforms to expand capacity for timely and decentralized detection in future outbreak responses.

Strengths and Limitations

CRISPR-based diagnostics show promise in meeting the REASSURED criteria, as they boast low-cost and minimal equipment requirements, as well as demonstrate rapid detection times, which are characteristics that are of great importance when developing a POCT for areas with limited infrastructure [17,26,37]. As well, the CRISPR-based tests described here have demonstrated high sensitivity and specificity. The SHERLOCK test developed to detect DENV demonstrated 100% agreement with RT-PCR when tested on 24 positive DENV samples [35]. When applied to 10 clinical samples of LASV clades II and IV, SHERLOCK again achieved 100% agreement with both sequencing and RT-qPCR [37]. Specifically for clade IV, this level of sensitivity outperformed two different PCR assays, a Nikisin RT-qPCR assay (40% sensitivity) and an in-house RT-qPCR assay (50% sensitivity), which was attributed to differences in primer design [37]. The same results were found when tested on 16 clinical samples of EBOV, with SHERLOCK in 100% agreement with sequencing [37]. Furthermore, detection of concentrations as low as 1 copy/uL was achieved [7,35]. Each test was also shown to be highly specific to its respective target of DENV, LASV, EBOV, or CCHF after testing for cross-reactivity with other viruses [7,35,37,70,71,72,76,107]. Lastly, most of the assays showed rapid detection times under one hour [7,35,70,71,72,73,75,76,107]. While these tests show high sensitivity and specificity, the number of clinical samples tested is low and could benefit from greater sampling to better showcase the accuracy and reliability of CRISPR-based diagnostic tests.

There are other VHF’s that were not addressed in this review, including those that the WHO previously described as high-priority pathogens such as MARV or RVFV, for which no CRISPR-based POC diagnostic assays have been developed [111]. Furthermore, the assays described here for DENV, EBOV, LASV, and CCHF commonly use RPA/RAA pre-amplification paired with the Cas13a enzyme, with some variations including Cas12a [7,35,37,70,71,72,73,76]. Evidently, in addition to the limited number of CRISPR-based POCTs for VHFs, there is also an absence of variety in these assays. To circumvent this, other combinations of pre-amplification methods and Cas enzymes could be further explored as a POCT. For example, LAMP and Cas12 enzymes have unique features that may advance CRISPR-based POCTs for VHFs by adding more options to the CRISPR-based toolbox. The successful combination of LAMP and Cas12 has been previously demonstrated with several pathogens, which can be leveraged to contribute to a wide variety of CRISPR-based POCTs for future outbreaks of VHFs [57,59,60,109,112].

6. Conclusions

CRISPR-based assays are a valuable tool for the development of POC diagnostic tests due to their ability to be easily adapted to the current circulating strain of the pathogen and their feasibility for use in the field due to minimal, low-demand equipment, low cost per assay, and high sensitivity and specificity. CRISPR-based assays provide faster results than traditional nucleic acid tests like RT-qPCR, enabling more rapid responses, which is especially critical in resource-limited settings where timely diagnosis can significantly improve outbreak control and patient management. VHFs such as DENV, EBOV, LASV, and CCHF have successfully been detected using CRISPR-based diagnostics [7,35,37,70,71,72,73,75,76,107]. However, more development of CRISPR-based POCTs is needed to explore different combinations of pre-amplification methods and Cas enzymes to ensure public health safety in the event of future VHF outbreaks.

Abbreviations

The following abbreviations are used in this manuscript:

AacCas12b	Alicyclobacillus acidoterrestris Cas12b
aM	Attomolar
App	Application
ATP	Adenosine Triphosphate
BIP	Backward Inner Primer
BOP	Backward Outer Primer
BrCas12b	Brevibacillus sp. Cas12b
Bsu	Bacillus subtilis
Cas	CRISPR-Associated
CCHF	Crimean-Congo Hemorrhagic Fever Virus
cDNA	Complementary DNA
CRISPR	Clustered Regularly Interspaced Short Palindromic Repeats
crRNA	CRISPR RNA
CRISPRET	CRISPR-Based Rapid and Efficient Test
DENV	Dengue Virus
DETECTR	DNA Endonuclease-targeted CRISPR Trans Reporter
DNA	Deoxyribonucleic Acid
dsDNA	Double-Stranded DNA
EBOV	Ebola Virus
ELISA	Enzyme-Linked Immunosorbent Assays
ERASE	Easy-Readout and Sensitive-Enhanced
EXPAR	Exponential Amplification Reaction
FIP	Forward Inner Primer
fM	Femtomolar
FOP	Forward Outer Primer
gRNA	Guide RNA
HDA	Helicase Dependent Amplification
HOLMES	One Hour Low-Cost Multipurpose Highly Efficient System
HPV	Human Papillomavirus
HUDSON	Heating Unextracted Diagnostic Samples to Obliterate Nucleases
LAMP	Loop-Mediated Isothermal Amplification
LASV	Lassa Virus
Lift-CM	Lift-Heater Centrifugal Microfluidic Platform
Loop B	Loop Backward
Loop F	Loop Forward
MARV	Marburg virus
mLAMP	Multiplexing LAMP
MRSA	Methicillin-Resistant Staphylococcus aureus
NASBA	Nucleic Acid Sequence-Based Amplification
PAM	Protospacer Adjacent Motif
PCR	Polymerase Chain Reaction
POC	Point of Care
POCT	POC Test
RAA	Recombinase Aided Amplification
REASSURED	Real-time Connectivity, Ease of Specimen Collection, Affordable, Sensitive, Specific, User-friendly, Rapid and Robust, Equipment-Free and Environmentally Friendly, and Deliverable to End-Users
RNA	Ribonucleic Acid
RPA	Recombinase Polymerase Amplification
RT-LAMP	Reverse-Transcription LAMP
RT-qPCR	Quantitative Reverse-Transcription PCR
RT-RAA	Reverse-Transcription RAA
RT-RPA	Reverse-Transcription RPA
RVFV	Rift Valley Fever Virus
SARS-CoV-2	Severe Acute Respiratory Syndrome Coronavirus-2
SDA	Strand Displacement Amplification
SEC-disc	Spatially Encoded Centrifugal Disc
SHERLOCK	Specific High-Sensitivity Enzymatic Reporter UnLOCKing
SNP	Single Nucleotide Polymorphisms
ssDNA	Single-Stranded DNA
STOP	SHERLOCK Testing in One-Pot
TCEP-EDTA	Tris(2-carboxyethyl)phosphine hydrochloride–Ethylenediaminetetraacetic Acid
tracrRNA	trans-Activating crRNA
VHF	Viral Hemorrhagic Fever
WHO	World Health Organization
ZIKV	Zika Virus

Author Contributions

Conceptualization, K.W., A.B. and J.E.S.; writing—original draft preparation, K.W.; figures, K.W. created the figures using BioRender; writing—review and editing, L.G., A.B. and J.E.S. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

This work was supported by the Genomics Research and Development Initiative-8 and the Canadian Safety and Security Program (Grant number: CSSP-2022-CP-2546).