Development and evaluation of a Mycobacterium bovis interferon-γ enzyme-linked immunospot (ELISpot) assay for detection of bovine tuberculosis

Bovine tuberculosis (bTB) caused by Mycobacterium bovis is an important zoonotic disease. This infection is difficult to control because of the limited ability of the tuberculin skin test (TST) and ancillary IFN-γ release assay to detect all infected animals. In this study, we aimed to develop an efficient assay based on the enzyme-linked immunospot (ELISpot) technique for the diagnosis of bTB, with IFN-γ monoclonal antibodies 3E9 and Bio-labeled 6F8 used as capture and detection antibodies, respectively. As expected, there were significantly more M. bovis -specific spot-forming units (SFU) in bTB-infected cattle than in healthy cattle when an M. bovis -specific antigen, CFP-10-ESAT-6 fusion protein (CE protein), was used. The M. bovis IFN-γ ELISpot assay demonstrated a high level of agreement (90.83%) with the BOVIGAM ELISA test (Thermo Fisher Scientific) for detecting bTB. Furthermore, 3 of 109 cattle tested negative by both the TST and the BOVIGAM ELISA tests, but positive by the ELISpot assay (TST − ELISA − ELISpot + ). During subsequent long-term monitoring, these 3 cattle became TST + ELISA + ELISpot + . These results suggest that the M. bovis IFN-γ ELISpot assay we established could detect infected cattle earlier than the BOVIGAM ELISA test.


INTRODUCTION
Bovine tuberculosis (bTB), caused by Mycobacterium tuberculosis complex (MTBC), is a zoonotic disease that primarily affects cattle, as well as other domestic and wild mammals (MacHugh et al., 2009;Pérez-Lago et al., 2014;Ramanujam et al., 2021).Mycobacterium bovis is the bacilli within the MTBC most often detected in tuberculous lesions in cattle (de la Rua-Domenech et al., 2006).The World Organization for Animal Health (OIE) has classified bTB as a notifiable animal disease (OIE, 2021a).As a chronic wasting disease, bTB is a huge hindrance to livestock-related products, impairing productivity and imposing large economic burdens worldwide (Tsairidou et al., 2016;El-Sayed et al., 2016).Zoonosis is also a major problem accounting for approximately 10% of all human tuberculosis cases caused by M. bovis, with most cases being recorded in developing countries (Aagaard et al., 2006;Muller et al., 2013).There are several methods for detecting tuberculosis, including bacteriological, molecular biological, and immunological methods (Schiller et al., 2010a;OIE, 2021b).Approved and implemented diagnostic tests for bTB are mainly based on specific cell-mediated immune responses, including the tuberculin skin test (TST) with purified protein derivative (PPD) antigens (Monaghan et al., 1994) or IFN-γ release assays (IGRA; Buddle et al., 2009;Schiller et al., 2010b) or both.The TST, the only OIE-stipulated tool, is generally the primary screening test used for monitoring bTB (Schiller et al., 2011;OIE, 2021a).The sensitivity of the TST has been estimated to be 68 to 95% in several overseas studies (Monaghan et al., 1994;Buddle et al., 2009).The IGRA is also considered as a variable tool of the OIE and a complementary instrument to assist the TST (Wood et al., 1990;Clegg et al., 2017) and has been integrated into the controllable planning of bTB in many countries.The BOVIGAM M. bovis Gamma Interferon Test Kit (BOVIGAM ELISA test) based on the IFN-γ in vitro release test, which is certified as an ancillary assay for the TST by the OIE, has become the most important technology in the current diagnosis of bTB (Wood and Jones, 2001;Ghielmetti et al., 2021).The BOVIGAM ELISA test can be used to maximize the detection of infected cattle

Development and evaluation of a Mycobacterium bovis interferon-γ enzymelinked immunospot (ELISpot) assay for detection of bovine tuberculosis
because of its high sensitivity and specificity (Gormley et al., 2006).However, even a combination of the TST and ELISA has not been able to achieve a 100% detection rate for positive animals.This could be explained by the presence of PPD, a crude water-soluble protein extracted from a heat-treated culture of M. bovis (Griffa, et al., 2020), which may impair the specificity of tests because of the cross-reaction of antigens shared by both pathogenic and nonpathogenic mycobacterial species (Aagaard et al., 2006;Allen et al., 2010;Mon et al., 2014;Klepp et al., 2019).
The enzyme-linked immunospot (ELISpot) assay has been used as a detection method in human tuberculosis because of its higher sensitivity and precision in assessing T-cell-specific responses (Sattah et al., 2012).In addition, the ELISpot assay has been used to explore the contribution of different phenotypes of T cells in the immune response to M. bovis (Blunt et al., 2015;Maggioli et al., 2015) and the potential to distinguish different stages of bTB (Veerasami et al., 2011;Parthasarathy et al., 2012;Steinbach et al., 2019).However, a detailed evaluation of the ELISpot assay versus official agency-approved methods for detecting bTB is lacking.Early secretory antigenic target 6 (ESAT-6) and culture filtrate protein 10 (CFP-10) are encoded by genes of the difference-1 region (RD1; Vordermeier et al., 2016) in M. bovis, a region present only in the MTBC but absent in Mycobacterium avium and bacillus Calmette-Guérin (Encinas et al., 2018).Both ESAT-6 and CFP-10 are the major target antigens recognized by host cells in the first phase of Mycobacterium tuberculosis infection (Rodríguez-Hernández et al., 2020).The ESAT-6-CFP-10 fusion protein used as TB-specific antigen can improve the accuracy of diagnosis compared with the traditional PPD in IFN-γ ELISA in humans (Hill et al., 2005).
In the present study, we aimed to screen mAb against bovine IFN-γ (BoIFN-γ) and develop a M. bovis IFN-γ ELISpot assay with CFP-10-ESAT-6 fusion protein (CE protein), which could be used as a promising method for bTB elimination combined with the TST and the BOVIGAM ELISA test.

Study Design and Sampling Strategy
The evaluation of the ELISpot assay was conducted in conjunction with the regular application of the bTB surveillance program in China between 2017 and 2019.Blood samples were collected aseptically in Vacutainer blood collection tubes containing sodium heparin (BD Biosciences) from dairy farms for ELISA and ELISpot assay, and transported to our laboratory at ambient temperature (20 ± 5°C) for processing.The blood samples collected during continuous monitoring were used to evaluate the ELISpot assay of cattle and define cut-offs for optimal sensitivity and specificity.

Selection of Farms
The farms were selected for detecting bTB by TST over the past 5 yr.Bovine tuberculin PPD (2,000 tuberculin forms per 0.1 mL) in the TST was obtained from the China Institute of Veterinary Drugs Control.The test was conducted as per the manufacturer's instructions: a 2.5-cm 2 area in the middle third of the neck of the animal was shaved and 0.1 mL of bovine tuberculin PPD was injected intradermally.A positive reaction to tuberculin was recorded 72 h post intradermal injection if there was an increase of 4 mm or more in skin thickness with or without clinical signs.Routine TST were performed once every 3 mo and farms without bTB were classified as bTB-free farms (farm 1, farm 2, and farm 3) and those with a high percentage of skin reactors were classified as bTB high-prevalence farms (positive incidence >5%; farm 4; Aagaard et al., 2010).

Animals and Ethics
Thirty-eight cattle were chosen to develop the M. bovis IFN-γ ELISpot assay.Healthy cattle (n = 19) from farm 1 tested negative by both the TST and the BOVIGAM ELISA test (Thermo Fisher Scientific) during continuous monitoring for the past 3 yr, whereas bTB-exposed cattle (n = 19) from farm 4 tested positive by both tests (Supplemental Table S1, https: / / doi .org/ 10 .17632/72vtdxxkkb .3;Li, 2021b).To assess the specificity of the method, 31 cattle from bTB-free farms 1, 2, and 3 that tested bTB negative were randomly selected.Another 109 cattle were selected from dairy farms to compare the results of the ELISpot assay with the BOVIGAM ELISA test (Supplemental Table S2; https: / / doi .org/ 10 .17632/72vtdxxkkb .3;Li, 2021b).All procedures were approved by the Institutional Animal Experimental Committee of Yangzhou University.All experiments were performed according to the national guidelines for animal welfare.

Evaluation of IGRA
Blood samples were collected from the caudal veins of cattle.Heparinized whole blood (1.5 mL) was incubated with 100 μL of Bovine Tuberculin PPD 3000 (PPDB; Thermo Fisher Scientific), Avian Tuberculin PPD 2500 (PPDA; Thermo Fisher Scientific), and PBS for 16 to 24 h in 24-well cell plates with 5% CO 2 at 37°C.Plasma was collected and IFN-γ was measured using the BO-

Establishment of M. bovis IFN-γ ELISpot Assay
Peripheral blood mononuclear cells (PBMC) were isolated using peripheral blood lymphocyte separation tubes (Dakewe) according to the manufacturer's instructions.
The ELISpot plates (Sigma-Aldrich) were coated with 100 μL/well mAb 3E9 (our laboratory) at concentrations of 1.25, 2.5, 5.0, and 10.0 μg/mL and incubated for 16 h at 4°C.The PBMC, at concentrations of 2.5 × 10 4 , 5.0 × 10 4 , 1.0 × 10 5 , and 2.0 × 10 5 cells/ well, were dispensed in ELISpot plates incubated at 37°C in a 5% CO 2 incubator for 36 h.Lectin from Phytolacca Americana (PWM; Sigma-Aldrich, 5 μg/mL)  and RPMI 1640 medium (Gibco) were used as positive and negative controls, respectively.The CE protein (1.25, 2.5, 5, and 10 μg/mL) was added as the specific stimulant to establish the M. bovis IFN-γ ELISpot assay for detecting bTB-infected cattle.The plates were washed 5 times with 250 μL of PBS and then incubated with 100 μL/well Bio-labeled mAb 6F8 (Bio-6F8; our laboratory) at concentrations of 0.5, 1.0, and 1.5 μg/ mL for 2 h at 37°C in a 5% CO 2 incubator.After washing 5 times with 250 μL of PBS, 100 μL/well working concentration of streptavidin-alkaline phosphate conjugate (BD Biosciences) was added and incubated at 37°C in a 5% CO 2 incubator for 2 h.The plates were washed 5 times with 250 μL of PBS, and 100 μL/ well BCIP/NBT (Sigma-Aldrich) was added.After the formation of visible spots, the reaction was terminated by gently rinsing the wells with distilled water.The ELISpot plate was then dried.Spots were inspected and counted by Immunospot Series 6 ENTRY Analyzer (Cellular Technology Ltd.).The ELISpot units were defined as spot-forming units (SFU) by default, and SFU was denoted as SFU per 1 × 10 5 PBMC.

Statistical Analysis
Prism V9.01 software (GraphPad; GraphPad Software Inc.) was used for statistical analyses.Data were shown as mean ± standard error of the mean, and a 2-tailed unpaired t-test was used to assess differences between bTB-infected and healthy cattle.Statistical significance was determined at P-values of <0.0001.Receiver operator characteristic analysis was performed to establish cut-offs.
induce optimal scattered spots (Figure 1A and B).Further, 1.0 × 10 5 PBMC per well were chosen to induce optimal spots with a clearer background (Figure 1C).
The CE protein (5 μg/mL) was used as the specific stimulation to establish the M. bovis IFN-γ ELISpot assay, which produces an appropriate number of spots in bTB-infected cattle, but none in healthy cattle (Figure 2).As expected, when SFU was used as the detection target, all infected animals (median 164 SFU, interquartile range: 72-187 SFU) responded to CE, whereas control animals (1, interquartile range: 0-1) did not (Figure 3).Based on the above results, receiver operating characteristic analysis was performed, and a cut-off value of 10 SFU was chosen for the assay with a sensitivity of 100% (95% CI: 79.08-100%) and a specificity of 94.7% (95% CI: 71.89-99.72%; Figure 4).
All 31 cattle randomly selected from bTB-free farms also tested negative by the M. bovis IFN-γ ELISpot assay, indicating that our ELISpot assay had good specificity in bTB detection.

Application of M. bovis IFN-γ ELISpot Assay for the Diagnosis of BTB
To evaluate whether the M. bovis IFN-γ ELISpot assay represents a potential approach in bTB detection, the level of agreement in detecting bTB in 109 cattle between our ELISpot assay and the BOVIGAM ELISA test was analyzed.Results showed that 26 cattle tested positive by both the BOVIGAM ELISA test and the M. bovis IFN-γ ELISpot assay (ELISA + ELISpot + ) and 73 tested negative (ELISA − ELISpot − ).Statistical analysis showed that the positive level of agreement of the M. bovis IFN-γ ELISpot assay and the BOVIGAM ELISA test was 83.87%; the negative level of agreement was 93.59%.The total compliance rate was 90.83% (Table 1).The M. bovis IFN-γ ELISpot assay had a high level of agreement with the BOVIGAM ELISA test.

Long-Term Monitoring on Cattle
In the above tests, 3 of 109 cattle tested as negative by the TST and the BOVIGAM ELISA tests but positive by the ELISpot assay (TST − ELISA − ELISpot + ).These 3 cattle were fed separately to avoid outside interference and monitored for a long period of time.After monitoring for 6 mo, 3 cattle tested positive (TST + ELISA + ELISpot + ; Figure 5).A month later, the TST and the BOVIGAM ELISA tests showed the same positive results again.

Sensitivity Determination of M. bovis IFN-γ ELISpot Assay and BOVIGAM ELISA Test
To test the hypothesis that differential detection levels of cytokine-secreting cells contributed to the distinct outcomes of diagnosis by ELISpot or ELISA, an experiment was designed to determine the limit of detection.Results demonstrated the strengths of the M. bovis IFN-γ ELISpot assay for detecting IFN-γ secretion.The PBMC stimulated with PWM were 2-fold serial dilutions at a starting concentration of 2.0 × 10 6 cells/mL, and IFN-γ was detected by the ELISpot assay and the BOVIGAM ELISA test.When the number of PBMC was diluted to 8.0 × 10 3 cells/mL, IFN-γ was not detected by the BOVIGAM ELISA test but 20 SFU per well could still be counted in the ELISpot assay (Table 2).

DISCUSSION
The ELISpot assay is based on the principle that cytokines secreted by inflammatory cells are trapped by a capture antibody on the membrane and revealed through the binding of a secondary detection antibody to the substrate, which precipitates to form visible spots at the captured cytokine sites (Sedegah, 2015;Ji and Forsthuber, 2016).To obtain more information about host immune responses, the FluoroSpot assay was developed, which is fluorescence-based ELISpot technology that is able to simultaneously detect multiple cytokines (Steinbach et al., 2019;Kim et al., 2020).In this study, a pair of antibodies was selected from 43 strains of BoIFN-γ mAb to establish the ELISpot assay.To obtain suitable specific spots, 5 μg/mL CE protein and 1.0 × 10 5 /well PBMC were used.In addition, the M. bovis IFN-γ ELISpot assay has good specificity in the detection of bTB.The BOVIGAM ELISA test has been approved by the OIE as an ancillary test to confirm or negate the results of an intradermal skin test.The M. bovis IFN-γ ELISpot assay was applied to clinical testing in dairy farms.Our ELISpot assay demonstrated a high level of agreement (89.19%) with the BOVIGAM ELISA test for detecting bTB.This result suggests that the M. bovis IFN-γ ELISpot assay could be used as an auxiliary tool to diagnose of bTB.
Compared with conventional ELISA assays, the ELISpot assay can both measure the cytokine response at the single-cell level and detect specific cytokine-  producing cells with higher sensitivity (Faresjö, 2014;Janetzki, 2015;Ji and Forsthuber, 2016).This may be why 3 TST − ELISA − ELISpot + cattle became TST + ELISA + ELISpot + after 6 mo.To test this hypothesis, we designed an experiment to identify the lowest detection level of cytokine-secreting cells by our ELISpot assay and the BOVIGAM ELISA test, and the results showed that the detection level of cytokine-secreting cells by our ELISpot assay was 50 to 100 times higher than that by the BOVIGAM ELISA test.Thus, our ELISpot assay can detect bTB at an earlier stage to avoid missed detections.These findings indicated that the M. bovis IFN-γ ELISpot assay has potential as an ancillary diagnostic method for the detection of bTB and can be used for the purification of dairy farms.In addition, the assay provides the possibility of mixed detection of multiple samples due to its high sensitivity, which can significantly reduce the cost of detection, especially with large sample sizes.
There are a few limitations to our M. bovis IFN-γ ELISpot assay, which lacks bTB lesions or postmortem test results as a gold standard.As reported, the BO- The number of cells is rounded up.OD 450 = optical density at 450 nm.SFU = spot-forming units per well.VIGAM ELISA kit has a sensitivity varying between 81.8% and 100% and a specificity between 94% and 100% for the diagnosis of bTB (Wood and Jones, 2001), which could be used as a gold standard due to the reliability and accuracy (Gan et al., 2013).Further, when the TST is combined with the IFN-γ assay, 95.2% of infected animals can be identified (Gormley et al., 2006).Therefore, the BOVIGAM ELISA test and the TST were used as the standard in this study.Meanwhile, compared with supernatant-based assays, the advantages of ELISpot are the ability to enumerate cells within a given population that respond to stimulation and the strength of this response (in spot size and intensity; Leehan and Koelsch, 2015).
In conclusion, the present study confirms the potential application of the M. bovis IFN-γ ELISpot assay for the diagnosis of bTB in dairy farms by using 2 strains of BoIFN-γ mAb and specific antigen CE protein.

Figure 4 .
Figure 4. Receiver operator characteristic (ROC) analysis of IFN-γ-producing cells to distinguish animals with bovine tuberculosis (bTB) from healthy animals.Nineteen bTB-infected cattle from farm 4 and 19 healthy cattle from farm 1 were subjected to the Mycobacterium bovis IFN-γ enzyme-linked immunospot (ELISpot) assay to determine the cut-off value.AUC = area under the curve.
Li et al.: INTERFERON-γ ELISPOT ASSAY FOR BOVINE TUBERCULOSIS VIGAM ELISA test according to the manufacturer's protocol.Optical density values were measured at 450 nm using a microplate reader (BioTek).