First approach to molecular epidemiology of bovine tuberculosis in Colombia

Primera aproximación a la epidemiología molecular de la tuberculosis bovina en Colombia

This work is licensed under Creative Commons Attribution-ShareAlike 4.0 International.

Mycobacterium bovis, the causative agent of tuberculosis in cattle, is also a pathogen for large numbers of animals as wild cats, domestic animals and human or nonhuman primates among others; therefore is considered an important zoonotic agent (1). Although M. tuberculosis is the most common etiologic agent of human tuberculosis, this can also be caused by Mycobacterium bovis, the main etiological agent of bovine tuberculosis; hence the importance in knowledge the distribution of this pathogen in the cattle and the identification of other potential reservoirs in Colombia in order to contribute to public health policies in the country, considering that described multiple cases of human tuberculosis caused by M. bovis and multiresistant M. bovis in HIV infected patients in the world (2).

In the industrialized countries, the incident of bovine tuberculosis in animals and human beings has diminished with the pasteurization processes of milk and the judicious managing of the programs of elimination. Meanwhile, into developing countries the bovine tuberculosis has a significant impact owed probably to the generalized lack of pasteurization process for the country and absence of documentation on the sanitary condition of the production lines bovine (3).

In Colombia, the Colombian Agricultural Institute ICA, within the program of eradication of tuberculosis and brucellosis, has established the use of tuberculin (PPD) or derivative protean purified, for the diagnosis of the bovine tuberculosis in these animals. This program has normalized that the animals with positive PPD must be sacrificed, whereas the cattle estates with two negative results in the consecutive years must be declared like free of tuberculosis by ICA (4).

Although the PPD is a good tool to identify positives cattles for M. bovis and initiate the sanitation of farms for the control and eradication of bovine tuberculosis, this test has limitations in sensitivity and specificity and therefore may show false positive, generating economic losses (5, 6). This joined the national epidemiological and economic context, is essential improve, simplify and extend the diagnosis using more sensitive and specific tests, indicate that with greater certainty about the behaviors continue to animals, to contribute to the eradication of the disease, as well and the saving of resources within the National Program of Bovine Tuberculosis Eradication (7).

In Colombia, the prevalence of bovine tuberculosis detected postmortem is inaccurate and the number of farms officially certified since free of tuberculosis, has increased dramatically in the last years (4). In 2014, Colombia produced 6.717 million liters of milk and more than 837 million kilograms of beef (8), which meant a high proportion of revenue of currencies for the nation, therefore, a better knowledge and documentation of the situation of bovine tuberculosis it becomes necessary in order to adapt and to improve the epidemiological vigilance of this pathology guaranteeing the free revenue of the colombian products to the world markets (7).

In the field of the microbiological diagnosis of the bovine tuberculosis, this is delayed and costly, and is known that the excretion of the Mycobacterium in the milk is intermittent and less than 30% of the infected animals eliminate the Mycobacterium for this way (9).

Previous studies have been successful documenting the applicability of the PCR based in IS6110 as molecular method of detection of the members of the M. tuberculosis complex causative of tuberculosis in animals (10,11,12), especially for direct detection of M. bovis in tissues of bovine and milk where they have brought that provides rapid diagnosis of the disease.

On the other hand, in recent years, the develop of an alternative technology of molecular classification of the members of the complex M. tuberculosis called spoligotyping has been successfully applied in the investigations of molecular epidemiology of the tuberculosis, with special usefulness in bovine tuberculosis (13,14,15,16).

Considering the absence of true information and in order to document the current situation of the bovine tuberculosis in Colombia, that enables the adjustment of control measures in the frame of National Program of control and eradication of the bovine tuberculosis, we developed our first national investigation using the conventional and molecular tools.

Study type. An observational analytical study was developed in cattle estates from eight Colombians departments, selected based in the results of tuberculin tests applied by the Colombian Agricultural Institute (ICA), within national epidemiological vigilance program for trace the M. bovis causing of bovine tuberculosis (17).

Samples included in the study. Samples of lymph node of head, thorax and abdomen, and woven from tubercular injuries of bovines and buffaloes sacrificed according to resolution of the ICA were taken. The animals not slaughtered from positives farms were undergoing to collection of nasal swab, blood or milk according to the criterion of ICA veterinarians. Also, were included in this study raw milk and fresh cheeses samples from municipalities where there were estates included in the study.

Specimen collection and transport. The samples were gathered during the years 2008 to 2010, in cattle estates identified by ICA as positives to the tuberculin and in the plants of benefit, which were transported preserving the cold chain, to the Mycobacteria’s Laboratory at the National Institute of Health (INS) in Bogota Colombia.

Processing of samples, lymph node and tissues with macroscopic lesions. 2 to 5 g of samples previously washed, were individually macerated into sterile mortars. 1g of this macerated was placed into falcon plastic vial of 50 ml with 9ml of acid oxalic (4%) at 37°C during 20 minutes, after which we neutralized by NaOH (4 %) in the presence of red of phenol (1%). It was centrifuged to 4000xg for 10 minutes at 4°C. The supernatant was discarded and the sediment was washed by centrifugation twice and was re-suspended with distilled sterile water (18-19).

Processing of milk samples. Aseptically were collected 50ml of milk, which were centrifuged for 30 minutes at 4.000xg and 4°C. 21ml of the sediment was decontaminated with 3 ml of lauryl sulfate of sodium, incubating at ambient temperature during 30 minutes. Later, this was centrifuged during 30 minutes at 4.000xg and 4°C. The supernatant was discarded and the sediment was re-suspended with distilled sterile water (18-19).

Processing of nasal swabs samples. The samples of nasal swabs were dissolved by washing of hyssop, in 10ml of PBS. This product was centrifuged at 4.000xg during 30 minutes at 4°C. The supernatant was discarded, preserving the sediment, which was decontaminated by Kudoh method (18-19).

Processing of blood samples. Samples of blood were subjected to centrifugation at 2.000 xg during10 minutes at 4°C, to recover the white blood cells, which were lysates by adding 1 volume of sodium deoxycholate (0.4%) incubating at ambient temperature 30 minutes. Later, were centrifuged at 4.000xg during 30 minutes and 4°C. The supernatant was discarded and the sediment was re-suspended with 1 ml distilled sterile water (18-19).

Bacilloscopy. The sediment obtained by processing of each one of the samples of lymph nodes, tissues with gross lesions, blood, swabs and milk, were extended on slide to realize bacilloscopy, staining with the Ziehl-Neelsen coloration to looking the presence of bacilli acid alcohol resistant (BAAR), by observation of 100 fields with a white light microscope(18-19).

Culture. The sediment obtained by processing of each one of the samples of lymph nodes and tissues with gross lesions, blood, swab or milk, was cultured by duplicate in culture middles: Lowenstein-Jensen and Stonebrink-Giraldo’s modified (STG). The cultures were incubated at 37°C for growth readings, weekly up to completing 12 weeks when the negative cultures were discarded. The cultures with growth were submitted to Ziehl Neelsen coloration and those who presented BAAR were investigated by phenotypic tests of identification (18-19).

Phenotypic identification of species. The tests of species identification were carried out bearing in mind the recommendations from CDC of Atlanta. The cultures that presented growth were colored by Ziehl Neelsen to check the purity. So, those were processed by biochemical and enzymatic tests: niacin detection, nitrates reduction, catalase activity and semi quantitative catalase, tween hydrolysis, enzymatic activity of pyrazinamide and the acid phosphatase. In addition, we determined the inhibition of growth in LJ containing NaCl (5%); 2-tiofencarboxilic acid Hydrazide (TCH) and Hydroxylamine (HAS); in MacConkey agar, in Sauton and Sauton Picric (2%) (18-19).

DNA Extraction. An aliquot of 1 g macerated or sediment obtained by processing of each one of samples of lymph nodes and tissues with gross lesions, blood, swab or milk, were submitted to extraction of DNA according to the recommendations of van Soolingen et al (20), method based in CTAB, chloroform-isoamyl alcohol, and precipitating with isopropanol. Every DNA sample obtained was re-suspended in a buffer TE 0.1X. The quantification of the DNA was realized in the gel of agarose to 0.8% (w/v) by bromide of ethidium 0.5 ug/mL using DNA Ladder High MW Invitrogen (Cat N° 10496-016) as reference pattern (20).

Genotypic Identification. An aliquot of macerated or sediment obtained by processing of each one of samples of lymph nodes and tissues with gross lesions, blood, swab or milk, were submitted to identification of Mycobacterium belonging to the M. tuberculosis complex by amplification PCR-IS6110. For the discrimination of the members of the complex, specifically the M. bovis, we used the molecular methodology of spoligotyping.

PCR-IS6110. The detection of the sequence of insertion IS6110, typical of M. tuberculosis complex, previous standardization, were developed through a PCR-IS6110 described by Kremer et al (21) in a mixture of reaction of the final volume of 50µl that it contained 1.5mM of MgCl2, 0.125mM of dNTP’s; 0.125 µM of each primers INS1 (5’-CGTGAGGGCATCGAGGTGGC-3’) and INS2 (5’-GCGTAGGCGTCGGTGACAAA-3’) to amplified a 245 bp fragment of IS6110 insertion sequence. We used the C1000 Thermal Cycler Bio-Rad® to program of one initial cycle of denaturalization at 96°C, 5min and 30 cycles of 96°C x1min, 65°Cx1min and 72°Cx1min with a final extension of 72°C x 7min. The amplification product was visualized in agarose gel (1.5%) (w/v) with bromide of ethidium using molecular weight marker 50pb (Sigma®).

Spoligotyping. The discrimination of M bovis from other members of M. tuberculosis complex, was realized using the molecular methodology of spoligotyping before described by Kamerbeer et al (22). From macerated or sediment obtained by processing of each one of samples of lymph nodes and tissues with gross lesions, blood, swab or milk and positive cultures a region of the locus DR amplifies with primers specifics DRa and DRb and the products of amplification hybridize in a membrane (Immunetics) in which oligonucleotides synthetic they have been covalently close. The detection is realized using streptavidin peroxidase brought together (Roche Diagnostics Ltd, UK) and the casing ECL (GE, Biotech). The results were translated from binary code to octal code for comparison.

Analysis of results. Several databases were created in spreadsheets Excel 2007 for every analysis types, one for each characterization of animals and microbiological results of studied animals by culture, ZN and molecular markers PCR-IS6110 and spoligotyping. The results of spoligotyping were comparison with the molecular patterns found in the International base SpolDB4 of the Institute Pasteur of the Guadeloupe (23), equally, comparison was made with the database of M bovis spoligotypes (24) to determine the SIT (spoligo international type), family and international location.

Descriptive analysis. Descriptive analysis was realized of the animals characteristics studied according to the department and municipality of origin.

Independence analysis. To identify the relation between the animals’ variables as species, sex and race with the Mycobacterium infection was used independence tests at 0.05 significance level, If the awaited counts are superior to 5, the probability corresponds to a distribution of chi- square. If the counts are low, the probability corresponds to Fisher’s exact test (25).

Ethics statement. The Ethic Committee from National Institute of Health endorsed the study, considering that was an investigation without risk. The blood, tissues and nasal swabs samples from the cattle used for culture and/or DNA extraction, were collected under the directive of the Colombian Agricultural Institute, (ICA) within national epidemiological vigilance program (17) and their use in the present study for other purposes as to trace the M. bovis causing of bovine tuberculosis were not routine.

Origin of samples. 492 samples were studied in total. 86.6% was from bovine and 13.4% from buffaloes. The bovine ones of those who took the samples were of race Holstein (39.2%) and Zebu (22.1%), whereas the buffaloes were principally from race Murra (26%). To both species, the samples they belonged principally from females: 81.5% in bovines and 56.1% in buffaloes.

The samples were from Cundinamarca’s department n=197 (40%), followed by Boyacá n=107 (21.74%), Magdalena n=94 (19.10%) and Antioquia n=84 (17.07%); other departments contributed with minor proportion of samples. (Figure1).

Figure 1
Departments of origin of the samples included in the study and the M bovis isolated.

Type of samples studied. In the table 1 we discriminate the type of sample studied according to class of cattle and department of origin.

Table 1
Samples enrolled in this study, discriminated by type and from origin department

Microbiological Investigation.

Zielh Neelsen coloration. Were obtained 7.11% (n=35) positive samples that showed a concordance of 85.71% with regard to the culture results. The totality of positive samples by this procedure, were samples of ganglions, 54.28% from bovine and 45.71% from buffaloes.

Culture Investigation. In 492 studied samples we found 23.98% (n=118) positive samples to culture, of which 18.64% (n=22) were of nasal swabs, 1.69% (n=2) of milk and 79.66% (n=94) of lymphatic ganglions. In turn, 73.73% of these samples were from bovines and 26.27% from buffaloes.

The phenotypic identification of the positive cultures showed M. bovis in 71.19% (n=86) of them and 14.4% (n=17) M. tuberculosis complex, whereas 7.62% (n=9) corresponded to M. tuberculosis and 5.08% (n=6) to Mycobacterium spp.

it’s interesting to highlight that all isolations of M. bovis were achieved from lymphatic ganglions samples, whereas the M. tuberculosis and the Mycobacterium sp were obtained principally of nasal swabs (n=22) and in minor proportion from ganglions (n=8) and from milk (n=2).

The isolations achieved by this culture methodology coming only from Antioquia, Cundinamarca, Boyacá and Magdalena departments, also specifically from seven municipalities: Chibolo, Chiquinquirá, Puerto Nare, Santa Rosa de Osos, Simijaca, Tarazá y Yondó. In total 10 estates had positive cultures to Mycobacterium (Table 2).

Table 2
Microbiological and molecular results from samples studied by year and department.

Table 2 (Cont.)
Microbiological and molecular results from samples studied by year and department.

Molecular investigation

Results of PCR-IS6110. The molecular methodology of amplification by polymerase chain reaction using as molecular target the insertion sequence IS6110 that is specific of the Mycobacterium tuberculosis complex, It present a positivity of 51.02% (n=251) for which it was two times more sensitively that the culture and seven times more sensitive that ZN’s coloration.

Results of spoligotyping. 47.56% (n=234) of samples showed positivity for this methodology with a pattern belonging to M bovis in the 34.13% (n=86) and 65.87% (n=148) with a pattern from M tuberculosis. We identified 36 different genotypes and there were 22 samples that did not present pattern since did not achieve hybridization, although they amplified (Figure 2).

Figure 2
Dendogram of relation from spoligotypes of M. tuberculosis complex identified in the study.

Among the isolations with pattern from M bovis family, 7 were spoligotypes before reported: SB1075 (56 isolations), SB0822 (12 isolations), SB0961 (2 isolations), SB0121 (12 isolations), SB0295 (1 isolations) SB1190 (1 isolation) and 1 spoligotype SB0120 (2 isolations). Therefore, the great majority of samples that presented pattern from the M. bovis family already had reported in the database of M bovis Spoligotype (24).

Among the samples with spoligotypes not belonging to M. bovis family, we found six families where the majority was constituted by MANU 2 SIT 54 (n=64), followed by the genotype of M. tuberculosis NR (n=31) not reported, and for the genotype T1 SIT 53 (n=27). Also, there were present the genotypes U SIT 523; U SIT 1146 and MANU 2 SIT 1190 with 19, 5 and 2 isolates respectively (Figure 3).

Figure 3
Distribution of M. tuberculosis genotypes not belonging to M bovis family identified in the study

Like the results show, from the culture methodology we can observed that the nasal swabs (n=115) were the main contributors of genotypes, different from M. bovis family, though also some ones came from ganglions (n=16) milk (n=3), blood (n=10) and cheeses (n=4).

Relation between variables of the herd and the presence of Mycobacterium tuberculosis complex in the animals. In the evaluation of possible factors related to the presence of Mycobacterium belonging to the Mycobacterium tuberculosis complex in animals (according to the test of PCR-IS6110), met significant association for 79.8% of positivity versus 20.21% healthy animals from Magdalena’s department (p<0.0001). In contrast, Cundinamarca show minor risk of having Mycobacterium (p<0.0001, Table 3).

To other risk factors such as: species, sex within species, or race for cattle (buffaloes only report to one species), were found in Holstein, was found that have a negative percentage above than expected exhibiting a lesser presence of mycobacteria which should randomly. Contrary, in Zebu having values above those expected for animal reports with mycobacterial infection (Table 3).

Tabla 3
Resultados de las pruebas de Chi-cuadrado entre la presencia de Mycobacterium, las características del animal y procedencia

By the present research, it was possible to demonstrate for the first time that in Colombia exist at least 7 genotypes of M. bovis, which are infecting and sicken to bovines and buffaloes from four different Colombian’s departments. This is constituted in a serious threat of zoonosis not only for the manipulators of the estates infected, but also for the manipulators of the productive chain and the final consumers. This finding highlights the recommendation to strengthen surveillance for compliance with Decree 1880 of 2011 on guidelines for the marketing of raw milk for human consumption (26).

The genotype SB1075 that was the majoritary found in our study has been reported infecting cattle from United Kingdom and has been confirmed that is slightly frequent in the world’s statistics. Similar situation happens with the other 2 frequent genotypes in this study: the SB0822 and the SB0121 previously reported in Holland, Spain and Latin America (Brazil and Mexico)(27).

Therefore, it is extremely beneficial correlate the findings of present study with the very low infections incidence, historically observed and demonstrated by PPD and by microbiological investigation into Latin american human population (28), with the reported in colombian population where a single case caused by M. bovis, was confirmed in the INS in 2012 suggesting low infectivity/virulence of M. bovis circulating in Colombia, specifically from the mentioned departments.

Since by means of the statistical analysis it was possible to demonstrate that the high presence of M. bovis and not bovis was statistically significant in Magdalena’s department, it is there where principally there must be intensified the measures of vigilance and control in farms and products by the ICA.

Another powerful finding, it constitutes to have demonstrated that exists a considerable proportion of animals and samples carrying M. tuberculosis of different genotypes between which there stands out the MANU 2 and the T1, that are consider ancestors of the M. tuberculosis complex, which have a minor virulence for the bovine/buffalo, as for human being. Though possibly they are responsible for the positive tuberculin results in the herds, neither produce disease in the cattle and they do not even affect them in the magnitude that it does the M. bovis that is really pathogenic for the cattle and they could be competing for the host, displacing the M. bovis in ranching (29).

In this regard, there exist studies that demonstrate that the M. tuberculosis find in the ranching it’s a reflection of the situation of tuberculosis in human beings in such regions (30,31,32). This finding puts of present the significant efforts that are needed for the control of tuberculosis in human beings in Colombia.

Nevertheless, the true role of this M. tuberculosis, found in the studied animals, must be investigated at depth since this one can be the source responsible for the maintenance of the human tuberculosis in endemic form in our country. It is necessary to highlight the fact that it becomes necessary to complement this investigation with the study of other animals that they find in the estates so positives as negative, to demonstrate the presence of animals carrying Mycobacterium in subclinical state that they can be the reservoirs of this Mycobacterium.

The results of this investigation constitute topics of interest to Public Health & Social Protection’s responsible from Colombia, since it is their competition and responsibility developed the control of tuberculosis human as well as the prevention of the zoonosis.

In short and according to our findings, it is advisable to restore the program of epidemiological vigilance of the bovine tuberculosis using the methodologies proved in the present study, initially in the departments where we found positivity to M. bovis and/or M. tuberculosis.

Secondly one recommends realizing studies more deep with relation to the presence of Mycobacterium tuberculosis complex, different from the M. bovis as they are the MANU 2 and the T1 genotypes both in the wild environment as in the manipulators of the bovine production chain.

This study confirmed the circulation in Antioquia, Boyacá, Cundinamarca and Magdalena departments of at least 7 genotypes of Mycobacterium bovis, causing bovine tuberculosis in cattle from ranches. This information is useful for programs surveillance in public health and animal health addressed to articulation of joint actions to build regulatory frameworks for the elimination and control of zoonoses, all of which contribute to improve competitiveness in the chain cattle production, and will impact the fulfillment of international animal health standards to increase access to world markets, improving the economy and directly benefiting to public health.

All personnel working in the National Program for Prevention and Control of Bovine Tuberculosis and Brucellosis ICA in Colombia, for its accompaniment to the sampling areas.