CRISPR/Cas12a-based on-site diagnostics of Cryptosporidium parvum IId-subtype-family from human and cattle fecal samples

Cryptosporidium parvum is an enteric protozoan parasite with zoonotic importance and can cause cryptosporidiosis in humans as well as domestic and wild animals worldwide. The IId subtype family (SF) is one of the most prevalent subtypes of C. parvum. Some clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated (Cas) protein systems have been developed to detect nucleic acid with high flexibility, sensitivity and specificity. By integrating recombinase polymerase amplification and the Cas12a/crRNA trans-cleavage system (termed ReCTC), we established end-point diagnostics by observing fluorescence readouts with the naked eye under blue light and on-site diagnostics using a lateral flow strip (LFS) biosensor. Our ReCTC-based diagnoses can detect as little as a single copy of a cloned C. parvum 60-kDa glycoprotein (GP60) gene, 10 oocysts per gram (OPG), clinical fecal sample without tedious extraction of genomic DNA and have no cross-reactivity with other SFs of C. parvum or other common enteric parasitic protozoa. This study provided a new strategy for direct identification of the IId SF of C. parvum free of highly trained operators and expensive special equipment.


Background
Cryptosporidium is an enteric protozoan parasite that can infect various vertebrate species, including humans, nonhuman primates and domestic and wild animals worldwide [1]. It has been demonstrated that cryptosporidiosis is the second greatest cause of diarrhea and mortality in children after rotavirus [2][3][4]. Currently, 44 valid Cryptosporidium species have been identified [5,6], among which Cryptosporidium parvum and Cryptosporidium hominis are the predominant species that mainly infect humans and cause > 90% of the human infections [5,7]. C. parvum is also one of the most common species detected in ruminants, especially in youngaged animals [8].
Based on the analysis of the DNA sequence encoding the 60-kDa glycoprotein (GP60), a subtyping tool has been established and most commonly used in studies of the transmission of C. parvum in humans and ruminants [7]. A total of 19 subtype families (SFs) (IIa-IIi and IIk-IIt) have been identified for C. parvum [5], among which the IId SF is found in both humans and ruminants and

Open Access
Parasites & Vectors *Correspondence: zhanglx8999@henau.edu.cn 1 College of Animal Science and Veterinary Medicine, Longzihu Campus of Henan Agricultural University, No. 15 Longzihu University Area, Zhengzhou New District, Zhengzhou 450046, People's Republic of China Full list of author information is available at the end of the article is responsible for zoonotic cryptosporidiosis reported in Europe, Middle East, Asia and Australia [7,9]. Moreover, in China, all C. parvum isolates characterized thus far belong to IId subtypes [9].
Because of the common existence and high virulence of the C. parvum IId SF, a point-of-care diagnostic method, which can directly distinguish infections caused by the IId SF, is necessary. The diagnosis of Cryptosporidium spp. can be divided into the following categories: optical microscopy, antigens or genetic material detection, and the detection of antibodies against Cryptosporidium spp. [5]. Among these diagnoses, direct identification of IId SF can only be approached by DNA detection such as PCR amplification combined with nucleotide sequencing [10]. However, this method is time-consuming and complicated, requires highly trained operators and expensive special equipment and is not applicable for on-site clinical disease monitoring, especially in resource-limited areas.
The clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated (Cas) protein systems have been reported to recognize and cleave specific nucleic acid sequences (namely cis-cleavage) [11,12]. Some Cas proteins, including Cas12a, Cas12b, Cas13a and Cas14, have exhibited nonspecific trans-cleavage activity by cleaving non-target sequences when they are activated by recognizing a specific target sequence [13][14][15][16]. This collateral effect has been developed into Cas12a/Cas13a-based nucleic acid detection methods termed HOLMES (one-hour low-cost multipurpose highly efficient system), DETECTR (DNA endonucleasetargeted CRISPR trans reporter) and SHERLOCK (specific high-sensitivity enzymatic reporter unlocking) [14,17,18]. In DETECTR, CRISPR RNA (crRNA) is designed specifically to target double-strand DNA (dsDNA) located downstream of a short T-rich protospacer-adjacent motif (PAM); therefore, the target DNA works as an activator to trigger both cis-and trans-cleavage of Cas12a. The fluorophore quencher labeled reporter (FQ Reporter) in the system is then cut, and a fluorescence signal is released and measured [14].
This universal molecular diagnostic methodology has been applied in recent studies to detect African swine fever virus (ASFV) [19,20] and beta-coronavirus severe acute respiratory syndrome (SARS)-CoV-2 [21], mycoplasma [22] and to identify Mycobacterium tuberculosis infection and nontuberculous mycobacteria infection [23]. In the present study, by designing a SF-specific crRNA probe, integrating recombinase polymerase amplification and the Cas12a/crRNA trans-cleavage system (termed ReCTC), we established an end-point diagnostic method by observing fluorescence readouts with the naked eye under blue light and an on-site diagnostic method using a lateral flow strip (LFS) biosensor to detect C. parvum IId SF nucleic acid rapidly and accurately.

Reagents and oligonucleotides
All the primers, quenched fluorescent DNA reporter (FAM-TTATT-BHQ1) and lateral flow strip test reporter (FAM-TTATT-biotin) were ordered from Sangon Biotech (Shanghai, China). Partial sequences of C. parvum gp60 gene of six SFs (IIa-IIf ) were synthesized and cloned into pUC57 vectors and transformed into E. coli DH5α, respectively (also from Sangon Biotech). All the nucleotide sequences are listed in Additional file 1:

Cryptosporidium parvum oocyst counting and DNA extraction
Subtype IIdA19G1 oocysts of C. parvum were used in this study. Oocysts were purified and counted according to a previous study [24] and stored in phosphate buffer solution (PBS) at 4 °C. N-lauroylsarcosine sodium salt (LSS) was used in the crude DNA extraction, and a homemade DNA banding column was used to purify DNA from the crude extracts (Additional file 1: Figure  S1). One μicroliter 10% LSS solution was added to 100 μl PBS containing C. parvum oocysts. After intensive mixing, the oocyst suspension was incubated in boiling water for 5 min. Then, heating of the water was stopped and the incubation was retained with the afterheat of boiled water for another 10 min. The lysate was then added to the homemade DNA banding column using a 1-ml syringe and was pushed through the column under the syringe pressure; 100 μl 10 mM Tris solution was added to the column using a new syringe. After incubating for 5 min at room temperature, DNA was eluted in Tris solution under the syringe pressure.

Design and preparation of crRNA
The gp60 gene has been used in the identification of SFs of C. parvum; thus, we chose it as the target sequence for the design of crRNA. To determine the highly conserved regions with SF specificity, sequences of different SFs were downloaded from GenBank and aligned using ClustalX 2.1 (http:// www. clust al. org/ clust al2/), and a 24-nucleotide (nt) sequence closely following a T nucleotide-rich protospacer adjacent motif (PAM) was chosen as the target sequence and synthesized containing T7 promotor (Additional file 1: Table S1). The two synthesized reverse complementary crDNA oligonucleotides were annealed to form dsDNA templates, and the crRNA was transcribed using a HiScribe ™ T7 High Yield RNA Synthesis Kit followed by DNase I digestion and NucAway ™ Spin Column purification. The concentration of purified crRNA was measured using NanoDrop One (Thermo Fisher Scientific Inc., Waltham, MA, USA).

Recombinase polymerase amplification (RPA) assay
Isothermal recombinase polymerase amplification (RPA) was used to promote the concentration of target dsDNA. Ten pairs of RPA primers were designed (Additional file 1: Table S1), and TwistAmp ® Basic kit was used to perform the RPA assay according to the instruction manual. Briefly, the 50-μl total reaction volume was made up of 29.5 μl rehydration buffer, 500 nM of each primer, 5 μl extracted DNA, 2.5 μl 280 nM magnesium acetate (MgOAc) and sterile nuclease-free water up to 50 μl. The reaction was performed at 37 °C for 30 min. The MgOAc was added onto the inner surface of the tube lid while all the other reagents were added to the bottom of the tube that was then carefully capped. A short spin to mix the MgOAc into the solution was applied, which triggered the RPA reactions.

Construction of the LFS assay
To enable the on-site diagnosis of C. parvum IId SF, a lateral flow readout based on the cleavage of the designed FAM-TTATT-biotin ssDNA reporter was used, which allowed reading the results without sophisticated instruments and skilled technicians. The FAM-TTATT-biotin ssDNA reporter can specifically bind to anti-FITC antibody conjugated with Au nanoparticles and form a complex. When the ssDNA in the reporter is not degraded by FnCas12a, this complex can be captured by the biotin ligand fixed at the control line. On the contrary, when the ssDNA is degraded, biotin is released from the complex, and the complex without biotin cannot be captured by the biotin ligand fixed at the control line but by the IgG antibody at the test line.
In the LFS assay, 20 μl of the ReCTC reaction system was incubated at 37 °C for 1 h followed by mixing with 250 μl of ddH 2 O. Then, 80 μl of the mixture was loaded onto the LFS pad and incubated at room temperature for 3 min. The strips were then photographed using a smartphone camera.

PCR amplification of Cryptosporidium parvum gp60 gene
To assess our ReCTC-based detection, a nested PCR amplification based on the SSU rRNA loci of Cryptosporidium was used, mainly according to the previous study [26]. Primers used are listed in the Additional file 1: Table S1, and PCR was performed in 25-μl reaction mixtures consisting of 2 μl of DNA preparation (or first PCR product), 1 × KOD-Plus buffer, 200 μM dNTPs, 1 mM MgSO 4 , 300 nM of each primer and 1.5 units of KOD-Plus DNA polymerase (Toyobo, Osaka, Japan). The amplification program consisted of an initial denaturation step at 94 °C for 5 min, followed by 35 cycles of 95 °C for 45 s, 55 °C for 45 s and 72 °C for 1 min, with a final extension of 72 °C for 10 min. Then, the PCR products were analyzed electrophoretically on 1% agarose gel stained with DNAGREEN (Tiandz, Beijing, China) and visualized under UV light. To determine the species and subtype of the Cryptosporidium detected, positive secondary PCR products were sequenced bidirectionally by SinoGenoMax Biotechnology Co., Ltd. (Beijing, China), and the generated sequences were aligned in Clustal X software, version 2.1 (http:// www. clust al.org/), with reference sequences downloaded from GenBank.

Design and preparation of crRNA
By multiple alignment of the gp60 gene, we identified a fragment that was conserved in IId SF but not in other SFs. A 24-nt sequence of this fragment was chosen as the target sequence, and the crRNA sequence was determined (Additional file 1: Table S1; Figure S2). The synthesized single-strand crDNA oligonucleotides were annealed and transcribed to crRNA. Following DNase I digestion and NucAway ™ Spin Column purification, the concentration was measured using NanoDrop One. To ensure the stability of the in vitro transcription assay, three sets of replications were carried out, and concentrations of the final crRNA ranged from 577.1 to 608.3 ng/ μl with a mean concentration of 593.2 ng/μl ( Fig. 1 and Additional file 1: Figure S3).

Optimization of the RPA assay
A number of conserved RPA primers were designed based on the gp60 gene of C. parvum IId SF (GenBank accession no. FJ839877). A set of primers named F01333 and R32535 (Additional file 1: Table S1) showed the highest amplification efficiency by using the TwistAmp Basic Kit (TwistDx, UK) (Additional file 1: Figure S4) and was chosen as the best primer pair. According to the manufacturer's instructions, the optimal reaction temperature was 39 °C. To enable our whole ReCTC assay workflow worked well under 37 °C, we also compared the RPA efficiency at 37 and 39 °C, and the result showed that the primers F01333 and R32535 could work well at both temperatures (Additional file 1: Figure S5).

Feasibility verification of ReCTC-based detection
Based on the ReCTC system, we intended to establish the C. parvum IId SF detection methods using both the fluorescence and LFS as read-out reporters. Together with the FnCas12a, the designed crRNA and the target dsDNA sequence with a TTN PAM sequence form a triplex. The recognition of the target dsDNA by the designed crRNA can induce the activation of the FnCas12a, which subsequently cleaves the FAM-TTATT-BHQ1 to emit 520-nm fluorescence under 488-nm light or cleaves the FAM-TTATT-biotin reporter to show visible test lines on the LFS (Fig. 2).
The fluorescence signal was strong enough to be observed by the naked eye, and distinct differences between positive and negative samples were observed using the CRISPR/Cas12a-based fluorescence detection assay (Fig. 3a, b). To monitor the real-time progress of  (Fig. 4). At 15 nM or lower concentrations, there was not enough FAM-TTATT-biotin ssDNA reporter to combine all the Au nanoparticles labeled by the anti-FITC antibody, so a portion of the nanoparticles that did not bind to the FAM-TTATT-biotin ssDNA reporter kept moving forward until they reached the test band to cause the false-positive result. Therefore, 20 nM was chosen as the optimal concentration of reporter for the ReCTC-based LFS detection.

Sensitivity of the ReCTC-based detection
The limit of detection (LOD) of the ReCTC was evaluated using a cloned C. parvum IId family gp60 gene with a conserved PAM (TTC) region cloned in pUC57 plasmid (Additional file 1: Table S2) and verified with PCR amplification. The cloned plasmid DNA was serially diluted to different concentrations and subsequently applied to the ReCTC. Results of fluorescence detection showed that samples with concentrations of 1.0 × 10 -18 M or higher target plasmid DNA showed an obvious fluorescence signal compared to samples with lower concentrations and negative control (Fig. 5a). Similar results were also observed in the ReCTC-based LFS detection. The lowest concentration of samples that showed a test line was 1.0 × 10 -18 M, indicating the LOD of the ReCTC-based LFS detection was also 1.0 × 10 -18 M target plasmid DNA (Fig. 5b).
The sensitivity of the ReCTC-based fluorescence detection and LFS detection was also tested using crude DNA extracted from purified oocysts. The concentration of purified oocysts was adjusted to 1 × 10 6 oocysts per milliliter before the DNA extraction. The concentration of extracted DNA was 9.3 ng/μl and serially diluted to concentrations of 9.3 × 10 -7 to 9.3 × 10 -1 ng/μl, which were equivalent to 0.1 to 1 × 10 5 oocysts per milliliter, respectively. In the ReCTC-based fluorescence detection, samples with crude genomic DNA of 9.3 × 10 -6 or higher showed observable fluorescence signals, representing an LOD of one oocyst per milliliter ( Fig. 5c and Additional file 1: Figure S6). In the ReCTC-based LFS detection assay, the sample with 9.3 × 10 -5 ng/μl DNA showed a faint test line and samples with higher input target DNA showed strong test lines, indicating that ten oocysts per milliliter was detected with the ReCTC-based LFS detection (Fig. 5d).

Specificity of the ReCTC-based detection
The specificity of the ReCTC-based detection was verified using the recombinant pUC57 plasmid DNA containing the gp60 gene of six different C. parvum SFs (IIa, IIb, IIc, IId, IIe and IIf ) (Additional file 1: Table S2). Genomic DNA extracted from several prevalent Cryptosporidium species (C. andersoni, C. hominis, C. meleagridis, C. muris, C. bovis and C. ryanae) and other intestinal protozoa (Enterocytozoon bieneusi, Giardia duodenalis, Blastocystis hominis and Cyclospora cayetanensis) were also included in the specificity test. As shown in Fig. 6a, only samples of C. parvum IId SF exhibited strong fluorescence signal, which could be observed by the naked eye and verified by the qTOWER 3 G real-time PCR system (Additional file 1: Figure S7). The LFS detection further confirmed the specificity of the ReCTC, and a clear test line was observed only on the LFS where C. parvum IId SF recombinant pUC57 plasmids DNA was added (Fig. 6b).

Performance of the ReCTC-based detection of C. parvum IId SF on clinical samples
To assess the clinical utility of the ReCTC-based detection, fresh fecal specimens were collected from 30  , d) were indicated on the LFS pads preweaned calves. The specimens were filtered through a 7.62-cm-diameter sieve with a pore size of 45 μm and applied to the crude DNA extraction. The extracted crude DNA samples were initially screened using nested PCR based on the gp60 gene of C. parvum, and results of sequencing and alignment revealed that the positive rate of C. parvum IId SF was 23.3% (7/30) (Fig. 7a). All the crude DNA samples were then tested by ReCTC-based detection. Consistent with the PCR results, distinct fluorescence signals were observed in all the PCR-positive samples by the naked eye and no such signal in all the PCR-negative samples, showing an accordance rate of 100% with PCR-based nucleotide sequencing (Fig. 7b). The ReCTC-based LFS detection also was in 100% accordance with the conventional PCR sequencing method (Fig. 7c).
To validate the ReCTC diagnostic accuracy in human samples, eight clinical DNA samples from diarrhea patients that had been detected as positive for C. parvum IId SF were tested using ReCTC-based detection. Fluorescence signals were observed in all eight samples by the naked eye in the ReCTC-based fluorescence assay (Fig. 8a). The positive results were also confirmed by the fluorescence values recorded by the qTOWER 3 G qPCR system (Additional file 1: Figure S8). In the ReCTC-based LFS assay, all eight clinical DNA samples showed the test bands on the LFS pads (Fig. 8b).

Discussion
In this study, we combined isothermal recombinase polymerase amplification with the Cas12a/crRNA transcleavage system to establish the ReCTC method and used fluorescence or chromatography readout as signal reporting. We found that ReCTC was able to detect C. parvum IId SF with high sensitivity and specificity. The use of CRISPR/Cas12a to detect infectious antigens has been reported in many other studies on detecting ASFV [20], human papillomavirus (HPV) [14] and the recently emerged coronavirus, SARS-CoV-2 [21]. Here, we first applied this method in detection of the zoonotic parasitic protozoan C. parvum, and the ReCTC-based fluorescence or LFS assay exhibited advantages in point-of-care use without the need for technical expertise, ancillary equipment or power (LFS assay).
In previous Cas12a off-target studies, a mismatch tolerance pattern was observed [27,28], and PAM mutations or mismatches in the PAM-adjacent region inhibited trans-cleavage when the activator was a dsDNA [14]. This stringent principle gives the Cas12a trans cleavagebased diagnosis good specificity, which was confirmed in our specificity test with several other common intestinal pathogens and other subtypes of C. parvum.
The currently reported CRISPR/Cas12a biosensing systems generally consist of three key parts: signal amplification, signal transducing and signal reporting [29]. In our ReCTC-based diagnosis, the RPA was chosen as the signal amplification method for its mild temperature requirements. Our results indicated that the RPA assay worked at 37 °C as well as 39 °C, which made the ReCTC more convenient as the downstream Cas12a trans-cleavage system was also conducted at 37 °C. A water/metal bath kettle, constant temperature incubator or even body heating can carry out the ReCTC reaction.
Two elements were used in the signal reporting part in this study: a fluorescence readout and LFS biosensor, corresponding to different reporters, FAM-TTATT-BHQ1 and FAM-TTATT-biotin, respectively. When the FAM-TTATT-BHQ1 reporter was used, bright green fluorescence signals could be observed by the naked eye under blue light with a Tanon-5200 Multi Fluorescence Imager. In resource-limited areas, a much cheaper device, Blue Light Gel Imager (Sangon Biotech. Inc., Shanghai, China; 440-485 nm), is recommended [22]. When FAM-TTATT-biotin was used in the signal reporting part, the readout could be observed using a LFS biosensor, which did not rely on technical expertise, ancillary equipment or power. This trait makes ReCTC-based LFS detection a real on-site diagnostic tool for C. parvum IId SF in field conditions.
As a biosensor, the CRISPR/Cas12a/crRNA system acts as a signal converter, converting the existence of target dsDNA to the presence of fluorescence or colorimetric signals by cutting the collateral ssDNA reporter [29]. Once the high endonuclease activity of FnCas12a was activated by the amplified target dsDNA, it could cleave a large number of collateral ssDNA reporters [25]. Together with RPA, the signal amplification part, the efficient cleavage activity of FnCas12a makes ReCTC-based detection highly sensitive.
In conclusion, by integrating recombinase polymerase amplification and Cas12a/crRNA trans-cleavage (termed ReCTC), we established an end point diagnostic method by observing fluorescence readouts with the naked eye under blue light and an on-site diagnostic method using a lateral flow strip (LFS) biosensor. Our ReCTC-based diagnoses could detect the IId subtype family of C. parvum from clinical fecal samples independent of professional technicians, expensive instruments or cumbersome operations. The newly established ReCTC-based detection displayed one and ten copy sensitivity in pure and complex samples, respectively, and the specificity was also confirmed to be robust. The ReCTC-based detection also agreed 100% with the conventional PCR sequencing method, which is the current gold standard commonly used in the diagnosis of C. parvum IId SF. Further optimization of the ReCTC assay as a one-pot reaction should be done in our future research to detect C. parvum IId SF from the clinical samples in the field more rapidly and simply.
Additional file 1: Figure S1. Absorbance curves of purified crRNA. The crRNA was transcribed from crDNA annealed from two reverse complementary single-strand oligonucleotides. The transcribed crRNA was treated with DNase I and was purified using the NucAway ™ Spin Column. Figure S2.  Figure S6. Specificity of the ReCTC-based detection. Recombinant pUC57 plasmid DNA containing the gp60 gene of IIa, IIb, IIc, IId, IIe and IIf SFs of C. parvum (1)(2)(3)(4)(5)(6) and genomic DNA of C. andersoni, C. hominis, C. meleagridis, C. muris, C. bovis, C. ryanae, Enterocytozoon bieneusi, Giardia duodenalis, Blastocystis hominis and Cyclospora cayetan (7)(8)(9)(10)(11)(12)(13)(14)(15)(16) were included. A. Specificity test of the ReCTC-based fluorescence detection assay. Only the sample of C. parvum IId SF exhibited a strong fluorescence signal. B. Specificity test of the ReCTC-based LFS detection assay. A clear test line was observed only on the LFS where C. parvum IId SF recombinant pUC57 plasmid DNA was added. Figure S7. ReCTC-based detection of C. parvum IId SF on clinical cattle samples. Clinical fecal samples from 30 dairy cattle were tested by a conventional nested PCR sequencing method (A) and our ReCTC-based fluorescence (B) and LFS (C) detection. Both the ReCTC-based fluorescence and LFS detection agreed 100% with the conventional nested PCR sequencing method. Figure S8. ReCTC-based detection of C. parvum IId SF on positive clinical human samples. Clinical human fecal DNA samples collected from inpatients that had been identified as positive for C. parvum IIdA19G1 were subjected to ReCTC-based fluorescence (A) and LFS (B) detection. N: negative control. Table S1. Nucleotide sequences used in this study. Table S2. Partial sequences of C. parvum gp60 gene of six subtype families (IIa-IIf ) cloned into the pUC57 vectors.