A precise and portable CRISPR-based test to detect Brucella genetic material

Hundred fresh milk samples were collected from a dairy farm located in Chaoyang District, Beijing, Gansu Province, Lanzhou City. Additionally, 100 vaginal swab samples were collected from a sheep farm located in Bayannaoer City, Inner Mongolia Autonomous Region, and 123 whole blood samples were collected from patients with suspected brucellosis at the Mongolian Medicine Research Institute of Tongliao City, Inner Mongolia Autonomous Region. We collected 4 mL of fasting venous blood from the 123 patients with suspected brucellosis. These samples were used for Brucella serological testing in accordance with the “Diagnosis for brucellosis” (WS269-2019). Suspected cases are defined as patients with an epidemiological history and corresponding clinical manifestations (such as persistent fever, sweating, fatigue, and muscle and joint pain). Confirmed cases are defined as suspected cases with SAT antibody titers of 1:100 (++) or isolation of Brucella. All collected samples were stored in a -80°C freezer for less than one month. In order to protect the patients’ privacy, patient-specific information was removed by assigning numbers to the patients and referring to the samples by the same numbers. This study has been approved by the Medical Ethics Committee of the Mongolian Medical Research Institute in Tongliao City, Inner Mongolia (Approval No.: TLSMYYJS-2023-1-004), and informed consent was obtained from patients and their relatives.

The BCSP31 gene (GenBank: MH045846.1↗) of Brucella was inserted into the pUC57 vector for the synthesis of the recombinant positive reference plasmid, which was subsequently constructed and purified by Biomed. The droplet digital PCR (ddPCR) instrument was employed to quantify the copy number of plasmid concentration. The primers and probe utilized were identical to those used in the quantitative PCR (qPCR) method. In this study, reference strains of Brucella, namely, 544A, 16M, and 1330S, were cultured and inactivated in a BSL-3 laboratory by trained laboratory personnel. The bacterial cultures were heat treated at 100°C for 60 min to ensure complete inactivation, and no viable bacteria were detected on culture. Nucleic acids from the bacterial suspension, blood samples, milk samples, and vaginal swab samples were extracted using the DNeasy Blood & Tissue Kit (Qiagen), and the extracted nucleic acids were stored at -20°C for future use.

The RAA primers and crRNA designed in this study were developed entirely independently, without the aid of any software. The crRNA consists of two components. One is a 28 - base - pair spacer region, which was designed based on the Brucella Bcsp31 gene and avoiding the RAA primer region. The other is a repeat sequence that binds to the Cas13 protein. (Figure 1C). Based on the designed crRNA sequence, single-stranded DNA was synthesized and subjected to PCR amplification using primers containing the T7 promoter sequence to generate the complete crRNA template. The crRNA template was mixed with T7 RNA polymerase and transcribed in vitro using the NEB HiScribe T7 Quick High-Yield RNA Synthesis Kit (New England Biolabs Inc.). The 20 μL reaction system consisted of 10 μL NTP mix, 2 μL T7 RNA polymerase, 3 μL crRNA-1 to 5, and 5 μL RNase-free water. Following overnight incubation at 37°C, the synthesized crRNA was purified using Sangon Biotech’s RNA Rapid Concentration and Purification Kit. (Supplementary Tables 1 and 2).

The primers and probe for qPCR were sourced from the literature (Probert et al., 2004). The forward and reverse primers for the qPCR reactions were 5′-GCTCGGTTGCCAATATCAATGC-3′ and 5′-GGGTAAAGCGTCGCCAGAAG-3′, respectively. A FAM-BHQ1-labelled probe was used for fluorescence detection, with the following sequence: 5′-AAATCTTCCACCTTGCCCTTGCCATCA-3′. The primers and probe were provided by Sangon Biological Engineering (Shanghai) Co. Ltd. The qPCR reaction system consisted of 10 μL of the PCR mix, 1.5 μL each of the forward and reverse primers and probe, 2.5 μL of sterile enzyme-free water, and 3 μL of the template DNA. The reaction conditions included a pre-denaturation step at 95°C for 3 min, followed by denaturation at 95°C for 15 s, annealing at 61°C for 30 s, and a total of 45 cycles, and extension at 40°C for an additional duration of 30 s.

The amplification was carried out according to the instructions of the Hangzhou Public Test RAA amplification kit (basic type). According to the recommended reaction system, 25 μL A buffer, 13.5 μL water, and 2 μL upper and downstream primers (10 nM) were evenly mixed and added into an eight-strip tube containing lyophilized powder, and 5 μL sample DNA was added to the mixture. Next, 2.5 μL of Buffer B was added to the eight-strip tube cap, and the tube cap was closed. The octane discharge was centrifuged at the medium and low speed settings in a centrifuge for 15 s and then placed in a constant temperature metal bath at 39°C for 30 min.

The Brucella abortus, Brucella melitensis, and Brucella suis biotype I reference strains, that is, 544A, 16M, and 1330S, respectively, were cultured and inactivated by professional laboratory technicians in a biosafety level 3 laboratory to obtain bacterial broth. The bacteria were inactivated by heating at 100°C for 60 min. After culture, 500 µL of the inactivated bacterial solution was transferred to a 1.5-mL EP tube after ensuring that there was no viable bacterial growth. The blood samples used in this study were whole blood samples, and the milk samples and vaginal swab samples were centrifuged at high speed and then the precipitate was added with normal saline for use. The blood, milk, and vaginal swab samples were collected in accordance with the instructions of the DNeasy Blood & Tissue Kit for DNA extraction. During the nucleic acid extraction process of samples, the nucleic acids of all samples were extracted as soon as possible after collection. Before being tested, all the extracted nucleic acids were stored in a -20°C refrigerator to better ensure the accuracy of experimental data.

The extracted DNA was used as a template, and after RAA amplification, the amplified product was used as a CRISPR reaction template. The target single-stranded DNA obtained after the RAA amplification reaction was transcribed into ssRNA via T7 transcriptase. Next, the crRNA binds to ssRNA, and cas13a trans-cleavage activity is activated, which leads to the cleavage of RNA reporters. The cleaved RNA reporter can be detected by a fluorescence signal collection instrument or interpreted by a test strip (Figure 1A).

The CRISPR detection system consists of an RNase inhibitor (New England Biolabs Inc.), 1 μL LwCas13a, 1.5 μL crRNA, 2 μL NTP Mix (New England Biolabs Inc.), 0.5 μL of T7 RNA Polymerase (New England Biolabs Inc.), 0.5 μL of HEPES buffer solution (ThermoFisher Scientific Inc., USA), 0.25 μL MgCl₂ solution (1 M), 2.5 μL quenched fluorescence RNA reporter (RNAse Alert; Thermo Scientific, Waltham, MA, USA), 5 μL of target nucleic acid, and 10.75 μL of RNase-free water. The reaction system was continuously collected at 37°C in a fluorescence signal collector (Roche LC480 real-time fluorescence quantitative PCR instrument) for 60 min, and the fluorescence signal was collected every 2 min for 30 cycles.

The test strip was constructed with a sample pad consisting of a combination of colloidal gold-labeled FAM antibody pads, avidin lines, FAM antibody lines, and an absorption mat. The sample liquid was dripped onto the test paper using the siphon effect and passed through the gasket line (T) and control line (C). Upon detection of the target nucleotide by the Cas13a-crRNA complex, Cas13a initiates trans-cleavage activity to cleave the reporter RNA (FAM-reporter RNA-biotin). This results in cracking of the reporter RNA and prevents colloidal gold molecules from remaining at line (T), which indicates a positive result. Conversely, formation of the “colloidal gold-labeled FAM antibody-RNA FAM report-avidin biotin” complex allows colloidal gold molecules to reside at test line (T), which indicates a negative result. The CRISPR detection system consists of 2 μL RNase inhibitor (New England Biolabs Inc.), 2 μL LwCas13a, 3 μL crRNA, 4 μL NTP Mix (New England Biolabs Inc.), 1 μL of T7 RNA Polymerase (New England Biolabs Inc.), 1 μL of HEPES buffer solution (ThermoFisher Scientific Inc., USA), 0.5 μL MgCl₂ solution (1 M), 5 μL quenched fluorescence RNA reporter (5′-6-FAM/UUUUUUUUUUUUUUUUUUUU-Bio-3′), 5 μL of target nucleic acid, and 26.5 μL of RNase-free water. The reaction system, prepared with a volume of 50 μL, was incubated in a metal bath at 37°C for 30 min. Following the reaction, a volume of 45 μL of the product was transferred using a pipette onto a strip. After an additional 30-min incubation period, the results were read and recorded by capturing images.

The fluorescence signal data were analyzed by setting different reaction conditions with unique changes to screen for the best reaction conditions. Using Brucella recombinant positive plasmid at a concentration of 10⁵ copies/μL as the template, RAA was set at different reaction temperatures of 30°C, 33°C, 36°C, 39°C, 42°C, and 45°C, and different reaction times of 5 min, 15 min, 30 min, 45 min, and 60 min. The template volumes were also set to varying levels such as 2 µL, 4 µL, 6 µL, and 10 µL. For the RAA amplification reactions, different CRISPR reaction temperatures of 25°C, 28°C, 31°C, 37°C, 40°C, and 43°C were set to optimize the reaction conditions.

In order to verify the sensitivity of the established method, the reference strains of the recombinant positive plasmids of B. abortus, B. melitensis, and B. suis biotype I, namely, 544A, 16M, and 1330S, respectively, were diluted in gradient to obtain concentrations of 10⁵ copies/μL, 10⁴ copies/μL, 10³ copies/μL, 10² copies/μL, 10¹ copies/μL, 10⁰ copies/μL, and 10^-1 copies/μL. The above concentration gradient was used as the template for RAA amplification. After the completion of the amplification reaction, 5 μL of the mixture was used as the template for subsequent CRISPR/Cas13a detection reaction, and CRISPR fluorescence detection and dip-strip detection were performed to determine the detection limits of the two detection methods. Vibrio cholerae, Vibrio parahaemolyticus, Yersinia enterocolitica, Salmonella typhimurium, and Ochrobactrum anthropi were selected to determine the specificity of the detection methods.

GraphPad Prism software version 8.0 (GraphPad Inc., La Jolla, CA, USA) was used to determine the means and SDs for analysis of variance (ANOVA). A paired t-test was used to detect the mean differences for each quantitative index. All statistical tests were two-sided, and statistical significance was set at P < 0.05.

In order to enhance the detection of Brucella nucleic acid in actual test samples, we established methods to optimize the CRISPR/Cas13a system. We determined the optimal primer RAA reaction temperature, time, and template size, as well as the optimum CRISPR reaction temperature based the fluorescence signal intensity. To determine whether the filter condition was optimal, the strip method was used and the stripe color shades were assessed. The results of agarose gel electrophoresis showed that the fourth pair of primers (F4 and R4) had the highest amplification efficiency, and this was identified as the best primer pair for RAA (Figure 1B). crRNA-3 was selected as the best crRNA for the CRISPR reaction because it had high fluorescence intensity and the best amplification effect (Figure 1D). The results indicated that the optimal reaction temperature for RAA was 39°C; the optimal reaction time, 30 min; and the optimal template volume, 5 μL. With the fluorescence detection method, it was observed that the highest fluorescence intensity was recorded at 37°C (Figure 1E).

The RAA CRISPR/Cas13a Brucella fluorescence detection method and the strip detection capability evaluation method were examined for their sensitivity and specificity. The results indicated that the fluorescence signal values for the recombinant positive plasmids of Brucella and the RAA amplification products of B. abortus, B. melitensis, and B. suis biotype I reference strains 544A, 16M, and 1330S, respectively, at a concentration of 10⁰ copies/μL, were significantly higher than those of the negative control (Figure 2A). The fluorescence method had the lowest detection limit of 10⁰ copies/μL, while the strip detection method had a detection limit of 10¹ copies/μL (Figure 3A). Validation assays of the fluorescence detection method and the strip test demonstrated that a specific fluorescence curve and positive strip were only obtained for Brucella recombinant plasmids and reference strains 544A, 16M, and 1330S of Brucella species I in cattle, sheep, and pigs, respectively. In contrast, V. cholerae, V. parahaemolyticus, Y. enterocolitica, S. typhimurium, and B. palustris did not produce specific fluorescence curves or positive strip results (Figure 2B). Further, the established CRISPR fluorescence detection method for Brucella exhibited excellent specificity compared to the test strip (Figure 3B).

The results indicate that out of 100 milk samples, 80 positive cases were detected with qPCR. This corresponds to a positivity rate of 80%. Further, the positivity rate of ddPCR was 89%; CRISPR fluorescence method, 93%; and CRISPR strip method, 87%. With regard to the human whole blood samples (n = 48), qPCR detected 15 positive cases, which corresponds to a positivity rate of 31.25%. With ddPCR, the positivity rate was 45.83% (22 positive cases). With the CRISPR fluorescence method, the positivity rate was 62.5% (30 positive cases). The CRISPR strip method had a positivity rate of 39.58% (19 positive cases). With regard to the testing of the sheep vaginal swab samples (n = 100), qPCR detected 68 positive cases, corresponding to a positivity rate of 68%. The positivity rate for ddPCR was 84%; the CRISPR fluorescence method, 91%; and the CRISPR strip method, 76% (Table 1).