Bovine- and avian-PPD were purchased from PRONABIVE (Mexico City, Mexico). Ficoll-hypaque was acquired from Amersheim Biosciences (Uppasala, Sweden). The kit for the detection of bovine gamma interferon was from BOVIGAM™ (Applied Biosystems™ Thermo Fisher Scientific). RPMI-1640 culture medium, peroxidase-labeled protein G, L-glutamine, ortho-phenylendiamine, t-octylphenoxy polyeth oxyethanol (Triton X-100), polyoxyethylenesorbitan monolaurate (Tween 20), Canavalia ensiformis (Concanavalin A, Con A), ethidium bromide, and 2-mercaptoethanol were from Sigma Aldrich (St. Louis, MO, USA). TRIzol reagent, RNase inhibitor, dithiothreitol (DTT), recombinant DNase I (RNase-free), ampliTaq DNA polymerase, superScript III reverse transcriptase, and deoxyribonucleotides were acquired from Invitrogen Life Technologies (Carlsbad, CA, USA). Diethyl-pyrocarbonate (DEPC) was from Amresco Inc. (Solon, OH, USA). AmpliTaq DNA polymerase and oligo-d(T)₁₆ were from Applied Biosystems Inc. (Hammonton NJ, USA). Chloroform and 2-propanol were acquired from Merck Chemicals Inc. (Gibbstown, NJ, USA). Sulfuric acid and salts were from JT. Baker Inc (Phillipsburg, NJ, USA).

The study design and sampling methodology were approved by Bioethics Committee for care and reasonable use of experimental animals in research projects of the Centro Nacional de Investigación Disciplinaria en Salud Animal e Inocuidad (CENID-SAI) belonging to the National Institute for Forestry, Agriculture and Livestock Research (INIFAP) in Mexico City, Mexico. Permission to perform the fieldwork was also obtained from CENID-SAI and INIFAP. Collection of blood samples and Single Intradermal Comparative Tuberculin (SICCT) test were performed by qualified veterinarians following official procedures from the Norma Oficial Mexicana (NOM-041-ZOO1995) of the National Campaign against tuberculosis in Animals 15.

The study was conducted in a Holstein-Friesian dairy herd located in a rural region of the State of Hidalgo, México, bTB is endemic in the region. In a first field work session, the SICCT test was applied to all the animals of the herd. The same day blood samples were taken from all animals to carry out Enzyme-Linked Immunosorbent Assay (ELISA) to evaluate immune humoral response against M. bovis-antigens. The following week, after SICCT and ELISA tests were carried out and results obtained, heparinized blood samples were taken from reactor cows and cows with high antibody titers, as well as from non-reactor cows, but which showed high titers of antibodies to mycobacterial antigens, to perform the Interferon-Gamma (IFN-γ) Release Assay (BOVIGAM). Based on the results obtained in the three diagnostic tests, twenty-three cows were selected and grouped, as: Thirteen, positive to all the mentioned tests, formed Group 1; the other ten cows, that were only positive to ELISA, formed Group 2. A third Group included ten healthy cows, and tested negative by all the three diagnostic tests, served as control. The latter came from a bTB-free accredited herd. For the purpose of analysis of mRNA relative expression of IFN-γ, IL-2, IL-4, and IL-10 in PBMC culture of the animal groups, blood samples were collected into tubes containing sodium heparin anticoagulant (BD Vacutainer, Franklin Lakes NJ, USA). PBMC were isolated by Ficoll-Hypaque (GE Healthcare Ficoll-Paque™ PLUS Media (density 1.077 g/mL)) density gradient centrifugation for its subsequent in vitro stimulation with bovine PPD (20 μg/ml) and with the mitogen Concanavalina A.

Intradermal inoculation of bovine PPD and avian PPD in two sites 12 cm apart on the cow mid-neck was performed. The first site was injected 100 µl containing 3,250 IU/ml bovine PPD and the second site was injected with 100 µl containing 5,000 IU/ml avian PPD. The skin thickness was measured by a digital Vernier caliper before the intradermal injection and at 72 h later. The results were recorded as an increase in skin thickness at 72 h compared to the thickness pre-injection. Values of both, avian and bovine PPD were plotted. According to the official graphic of the Mexican Official Norm, the result was the value at the intersection of avian and bovine PPD values 15.

Heparinized whole blood samples were individually processed within 2–3 h after collection. They were distributed in 1.5 ml aliquots into four individual wells per animal in sterile flat-bottomed 24-well cell culture plate (Nunclon, Roskilde, Denmark). The first well of the different cultures was stimulated with 20 mg/ml bovine PPD, the second with 20 mg/ml avian PPD, the third with 10 mg/ml polyclonal mitogen Con A, and the fourth well was used as control without stimulation. Blood cultures were incubated at 37°C in a 5% CO₂ humidified atmosphere for 24 h. After this time, the supernatant plasma was harvested and stored at -70°C until analyzed. The IFN-γ levels in the plasma supernatants were measured using a sandwich ELISA following the instructions from a commercial kit (Applied Biosystems™ Bovigam™ Tb Kit, Thermo Fisher Scientific) 16. Absorbance of standards and plasma samples were read at 450 nm using an ELISA plate reader (Benchmark-Plus Microplate Spectrophotometer Bio-Rad Laboratories, Hercules CA, USA). Results were evaluated considering the optical density at 450 nm (OD₄₅₀) mean value of stimulated plasma samples. An animal was positive to the test if value to bovine PPD OD₄₅₀ minus value to avian PPD OD₄₅₀ were >0.1 OD, and a value for the OD₄₅₀ with bovine PPD minus the OD₄₅₀ nil antigen culture of >0.1.

Culture-filtrate protein extracts (CFPE) was prepared from M. bovis strain AN5 and M. avium strain D4. The bacterial were cultured individually at 37°C for over six weeks in modified Dorset-Henley liquid synthetic medium 17. At the completion of the incubation period, proteins from the cultures filtrates were precipitated by ammonium sulfate [(NH₄)₂ SO₄] at a final saturation of 80%, at 4°C for 24h in constant agitation18. Then, the solution was centrifuged at 15,000 g for 60 min at 4°C, the pellet was suspended in 10 ml of phosphate buffered saline pH 7.2 (PBS) and dialyzed against PBS using dialysis bags with MW cutoff point from 10 kDa (Cellu.Sep H1, Membrane Filtration Products, Inc. Seguin TX, USA) at 4°C for 36h. Protein concentration in the CFPE was determined by the Bradford method 19, and it was adjusted to 3.5 mg/ml and divided into aliquots of 1 ml, which were stored at -70°C until used.

A comparative ELISA was carried out using CFPE from M. bovis and M. avium antigens separately to assess IgG-class antibodies to M. bovis in serum. 20. Ninety-six-well flat-bottom ELISA plates (Nunclon, Denmark) were incubated with 5 mg/well of either CFPE from M. bovis or M. avium diluted in 0.1 M carbonate/bicarbonate buffer of pH 9.6 at 4°C for 24h. Antigen-coated ELISA plates were washed 3 times by 0.3% Triton X-100 in PBS (PBS-Triton), 3 times by 0.1% Tween-20 in PBS (PBS-Tween) and blocked in 200ml of a blocking buffer (3 % skim milk in PBS-Tween) with incubation for 1h at 37°C. After 6 washes in PBS-Tween, 100 µl blood serum (diluted 1:100 in blocking buffer) was added in each well and incubated for 1h at 37°C. The ELISA plates were washed again and 100 ml peroxidase-labeled protein G diluted 1:10,000 in blocking buffer was added per well, with incubation for 1h at 37°C. Another stage of washes was performed and subsequently 100 µl chromogenic substrate solution containing 0.04% O-phenylenediamine (Sigma P-3804) and 0.04% of hydrogen peroxide in citrate buffer pH 4.5 was added for detection of enzyme reaction. Reaction was stopped with 50 μl/well of 2 M sulfuric acid. Optical densities were obtained at 492 nm (OD_492nm) using an automated ELISA plate reader (Benchmark-Plus Microplate Spectrophotometer). All assays were carried out in duplicate. Antibody response to M. bovis or M. avium CFPE was established previously by setting the cutoff value of the mean OD₄₉₂ plus two standard deviations of sera from healthy animals.

PBMC were isolated from bovine blood sample collected using heparin as anticoagulant. For this, 6 ml of blood was diluted 1:2 in PBS, carefully placed over 2.5 ml of Ficoll-Hypaque gradient in a 15 ml centrifuge tube, and centrifuged at 1,600 g for 40 min at room temperature 21. The mononuclear cell layer was isolated and washed twice in RPMI-1640 culture medium by centrifugation at 440 g for 10 minutes. PBMC (5 x 10⁶) were placed into each well of a 24-wells flat-bottom culture plate (Nunclon Denmark) and suspended in 1 ml RPMI-1640 culture medium supplemented with 10 mM HEPES buffer, 50 mM 2-mercaptoethanol, 2 mM L-glutamine, 100 U/ml penicillin, 0.1 mg/ml streptomycin, and 10% heat-inactivated fetal calf serum. Twenty mg/ml bovine PPD, 20 mg/ml avian PPD and 10 mg/ml Con A were added to the corresponding wells, and incubated for 72 h at 37°C in a 5% CO₂ humidified atmosphere. Cells without antigen stimulation were used as negative control. At the end of the incubation period, the cells, were harvested and washed in RPMI-1640 medium by centrifugation at 440 g for 3 min at 4°C discarding supernatant before RNA extraction.

Total RNA from bovine PPD- or avian PPD- or Con A-stimulated PBMC and control without stimulus were extracted as described 22. After RPMI-1640 medium washing, cells (5 x 106) were incubated in 1 ml of TRIzol for 5 minutes at room temperature. A second incubation was performed in 200 ml chloroform for 3 min at 4°C. After centrifugation at 12,000 g for 15 minutes at 4°C, the aqueous phase was recovered and RNA was precipitated by incubation 10 minutes in 500 ml 2-propanol. The pellet was centrifuged at 12,000 g for 10 min at 4° C, next suspended in 1 ml 75% ethanol and centrifuged at 7,500 g/5 min/4°C. Subsequently, the RNA was suspended in 50 ml of DEPC water and heated for 10 minutes at 55°C. The extracted RNA was measured in a fluorometer (VersaFluor^TMFluorometer System Bio-Rad, Richmond CA, USA). To eliminate any genomic DNA contamination, a recombinant DNase I, RNase-free was used to treat the extracted RNA.

The reverse transcriptase-polymerase chain reaction (RT-PCR) technique was used to evaluate cytokine transcript expression by specific primers for amplification of each one of them [23-27 (Table 1). Beta-actin (β-actin) mRNA gene was used as internal control 27,28. The cDNA was synthesized from total RNA obtained from stimulated and unstimulated PBMC using the enzyme superScript III reverse transcriptase. The reaction mixture was: 5 ml of Oligo(dT)16 (50 mM); 10 ml of 5X first-strand buffer (250 mM); 2.5 ml of dNTP Mix (10mM); 5 ml of 0.1M DTT; 2.5 ml 40U RNase inhibitor; 5 ml 250 U enzyme superScript III reverse transcriptase and 250 ng total RNA in a total volume per reaction  of 50 ml. The final reaction mixtures were incubated at room temperature for 10 min, followed by incubation at 42°C for 50 min and then by heat inactivation at 70°C for 15 min. Then, the cDNA was amplified by PCR, which was conducted in a final volume of 20 ml. The mixture contained 2 ml 10X buffer (200 mM Tris-HCl, pH 8.4, 500 mM KCl MgCl₂); 2 ml of dNTP Mix (10mM); 1 ml of each primer (20 pmol) for the cytokines under study, 0.15 ml ampliTaq DNA polymerase (0.75U), 4 ml of each cDNA and DEPC water (up until 20 ml) 29. Polymerase chain reaction was conducted in a thermocycler iCicler IQ^TM (Bio-Rad. Hercules CA, USA) according to previously established conditions (Table 1). PCR product of each cytokine RNA was analyzed in an agarose gel dyed in ethidium bromide. Subsequently, it was compared with the β-actin mRNA level and denoted as relative intensity (cytokine intensity/β-actin intensity) using the software LabWorks 4.0.

BTB-suspected animals were examined at post mortem in the abattoir for visible evidence of tuberculosis. The procedure involved visual inspection, palpation and incision of lungs, liver, kidneys and tracheobronchial, mediastinal and prescapular lymph nodes. Tissue samples were collected from cattle with suspected TB lesions during post mortem examination. The collected tissues were sliced at 0.5 to 1 cm intervals followed by evaluation to determining the severity of gross lesions. Pathology scoring method described by Vordermeier., et al. (2002) 30was used to evaluate the severity of the lesions. A score of 0, was assigned for absence of lesions; a score of 1, to a few mild lesions (1 to 2 mm in diameter); a score of 2, to a few medium lesions (1 to 5 mm in diameter) and necrotic tissue of a dimension of 5 by 5 mm; a score of 3 to multiple medium lesions presenting a size of greater than10 mm in diameter; and a score of 4 to severe multifocal lesions and large necrotic regions.

Thirteen from 23 tuberculosis-suspected cattle were extreme reactors to tuberculin test. These animals displayed a mean skin thickness value of >37 mm (± 9.7 mm) for bovine PPD, (Table 2). In contrast, the remaindaining animals in the test groups and the 10 clinically healthy animals in the control group did not show a significant reaction in skin after inoculation of M. bovis PPD or M. avium PPD. Group 1, was formed by 13 cows, which were positive to both Single Intradermal Comparative Cervical Tuberculin (SICCT) test and IFN-γ assay. Group 2, included 10 cows that were negative to tuberculin test and IFN-γ assay. Group 3, was constituted by 10 clinically healthy cows used as control group, which were negative to the different immunodiagnostic tests realized. Values denote mean ± standard deviation of optical density (OD492) obtained in serum samples from cattle.

Plasma samples from 13 tuberculin-reactor animals showed an IFN-γ OD₄₅₀ mean value upon bovine PPD-stimulation, which was 7.47-fold greater than non-stimulated samples (0.605 ± 0.313 vs 0.080 ± 0.033, respectively). Correspondingly, the IFN-γ level in response to stimulation by avian PPD was 0.393 ± 0.290, whereas that the upon Con A-stimulation the mean value was 1.026 ± 0.297. Thus, the group was considered positive (Table 2), according to the criteria of the BOVIGAM^TM test. On the contrary, 10 bTB-suspected cattle (group 2, anergic) but negative to tuberculin test exhibited low IFN-γ levels after stimulation with either bovine or avian PPD (0.100 ± 0.026 and 0.124± 0.060, respectively). As a result, they were considered negative to this test (Table 2). Likewise, a low mean level production was obtained when stimulated by Con A (0.260 ± 0.173) in whole blood cultures of anergic animals. Therefore, IFN-γ level was 3.94-fold lower than that produced by tuberculin-positive animals.

The IFN-γ production in plasma samples from 10 tuberculin-negative healthy control animals was as follows: IFN-γ level in non-stimulated whole blood showed OD₄₅₀ mean value of 0.069± 0.017, which did not increase after stimulation by either bovine or avian PPD (0.095 ± 0.015 and 0.117 ± 0.029, respectively). However, a high level of IFN-γ production was observed upon Con A stimulation, 2.923 ± 0.780; which was 2.85 fold higher than the reactor group, and 11.2 fold higher than the anergic group.

Serum antibodies in response to M. bovis antigen in tuberculin-positive cattle showed an OD₄₉₂ mean value of 0.489 ± 0.100, which was 2.5 fold higher than the control (0.197 ± 0.079). While, bTB-suspected tuberculin-negative cattle with low IFN-γ release in response to Con A stimulation, showedmean value of 0.771 ± 0.102 (Table 2). The positive ELISA reaction detected in tuberculin-negative cattle showed a mean value of 1.57 fold higher than tuberculin-positive cattle (0.771 ± 0.102 vs 0.489± 0.100). With regard to ELISA reactions using M. avium antigen, the mean values of sera from tuberculin-positive cattle and bTB-suspected tuberculin-negative cattle were similar (0.332± 0.130 and 0.355 ± 0.123, respectively). However, the above results were 2.0 to 2.13-fold higher than the OD₄₉₂ mean value(0.166 ± 0.048) when compared to the control group (Table 2).

PBMC from cattle of each group were individually stimulated with bovine-PPD, avian-PPD, and Con A mitogen for 72h. Cytokine transcript responses were analyzed by semi-quantitative RT-PCR. The mRNA levels were calculated as the ratio between the band intensity of each cytokine and the corresponding β-actin (Figure 1). In group 1, the production of IFN-γ and IL-2 (108.2 ± 19 and 91.8 ± 9.8, respectively) in PBMC stimulated by bovine-PPD, was similar to Con A-stimulated cells, with the exception of IL-4 (70.2 ± 9.8). The mRNA levels upon stimulation by Con A were IFN-γ (118 ± 13), IL-2 (78.7 ± 9.5), and IL-4 (74.8 ± 9.8). In contrast, low transcript levels to IFN-γ (43.9 ± 6.6), IL-2 (19.3 ± 4.6), and IL-4 (5.9 ± 1.3) were detected in response to avian PPD. The expression of IL-10 was not detected in PBMC stimulated by bovine PPD, avian PPD nor Con-A (Figure 2A). On the contrary, in group 2, IL-10 had a significantly high production (85.7 ± 7.1) only upon stimulation by bovine PPD (Figure 2B). IL-10 was not detected in groups 1 and 3. Though there were transcripts to IL-10 upon PPD avium- or Con A-stimulation (31.4 ± 5.7 and 42.8 ± 4.6, respectively), the levels were similar to samples from non-stimulated cells (23.3 ± 3.8) (Figure 2 B). Bovine PPD also stimulated production of IL-4 (54 ± 7.7), which was 1.3 fold lower than group 1. The level of IL-4 transcript upon stimulation by PPD avium exhibited 3.7 fold more relative intensity (21.7 ± 2.9) than the bovines from group 1 (5.9 ± 1.3) (Figure 2. A and B). Similar production of IL-4 was also observed upon Con A-stimulation (20 ± 3.9). However, although cells produced IFN-γ and IL-2 upon stimulation by bovine PPD, avian PPD, or Con A, results were below the background level of the control group (Figure 2B).

As expected, PBMC from healthy control group produced very low IFN-γ levels that were only detected after stimulation by bovine PPD (12 ± 3). In this group, IL-2, IL-4 and IL-10 transcripts were not observed upon stimulation, with neither bovine PPD nor avian PPD (Figure 2 C). Similarly, as expected, the control group produced IFN-γ (110 ± 13.6) and IL-2 (70.1 ± 10.7), but not IL-4 or IL-10 transcripts in response to stimulation by Con A.

Lymphatic tissue lesions were measured in parotid, submandibular, lateral and medial retropharyngeal, deep and superficial cervical, pulmonary, thoracic-tracheal, mediastinal, hepatic, inguinal, iliac, and mesenteric lymph nodes. In group 1, nine cows displayed mainly median lesions in deep cervical and lateral retropharyngeal lymph nodes. Twenty-three to 38% animals exhibited small lesions in examined lymph nodes, while 3 animals did not show any bTB-visible lesion in all the examined tissue. In group 2, 50-60% cows showed severe lesions in medial retropharyngeal, deep and superficial cervical lymph nodes, which were scored 3 and 4. Moreover, these animals showed severe pulmonary lesions that were considered as score 3. However, 2 cows did not show bTB-associated lesions in lymphatic tissues nor in lung lobes sliced into thin sections (Table 3).