Mycobacterial Interspersed Repetitive Unit-variable Number Tandem Repeat (MIRU-VNTR) Typing Lacks Discriminatory Power in the Genetic Analysis of Bovine Tuberculosis in Egypt

Tuberculosis (TB) is one of the most important infectious zoonotic diseases of vertebrates worldwide. TB in animals is primarily known from cases in cattle and other bovids for which the disease is generally referred to as bovine TB. The major causative agent of bovine TB is Mycobacterium bovis ( M. bovis ), a member of the Mycobacterium tuberculosis complex. Animal TB is a disease of high economic relevance within the context of livestock farming as it directly affects animal productivity and also influences international trade of animal products. In this study, we aimed at mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR) analysis of cases of TB infection in local cattle in Egypt. Therefore, various samples (milk and blood samples) were collected from cattle farms (in Damietta Province) that were positive for tuberculin test. Mycobacterial isolation was tested on milk samples, but it showed negative result. DNA was extracted from blood samples. Five mycobacterial genes ( IS6110 , katG , gyrA , oxyR , pncA ) were used for further confirmation of field TB infection. All blood samples were positive for Mycobacterial-specific genes. Twelve MIRU-VNTR loci were used to test their discriminatory power in the genetic analysis of TB, but such MIRU-VNTR loci typing failed to show any discriminatory power for the genetic analysis of bovine TB cases.


Introduction
Tuberculosis (TB) in animals is primarily known from cases in cattle and other bovids for which the disease is generally referred to as bovine TB [1,2]. The major causative agent of bovine TB is Mycobacterium bovis (M. bovis), a member of the Mycobacterium tuberculosis complex. Animal TB is a disease of high economic relevance within the context of livestock farming as it directly affects animal productivity and also influences international trade of animal products [3]. M. bovis infections have also been detected in wildlife and can have severe consequences for the ecosystem. Moreover, animal TB bears a zoonotic potential and is therefore of public health concern. Therefore, diagnosis of bovine TB is crucial for a proper control of disease transmission to animals and humans. To achieve that, a thorough knowledge of the available strains of M. bovis is required.
The identification of M. bovis by conventional bacteriological techniques is time consuming. Therefore, molecular identification techniques (spoligotyping and mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR) analysis) have been developed and were proved to be quicker and reliable in discriminating M. bovis from the other Mycobacterium tuberculosis complex [4][5][6][7][8][9][10][11]. Repeated sequences were identified in the genome of M. bovis and M. tuberculosis and were used to type M. bovis by MIRU-VNTR analysis [4][5][6][7][8][9][10][11]. Therefore, we aimed at MIRU-VNTR analysis of cases of TB infection in local cattle in Egypt. Here, we applied MIRU-VNTR typing on samples collected from tuberculin-positive cattle in local farms in Damietta Province.

Sample Collection
Samples were collected from different cattle farms, in Damietta Province. Samples were obtained from cattle that showed positive results for tuberculin test. Sample collection was performed under the recommendations and approval of Mansoura University Animal Care and Use Committee.

Milk Samples
Fifty milk samples were collected from animals (in two cattle farms in Damietta Province) that showed positive results for tuberculin test and had mastitis. Five ml of milk were collected aseptically in sterile tubes.

Blood Samples
One hundred blood samples were collected from cattle (in three different farms in Damietta Province) that showed positive results for tuberculin test.

Preparation of Culture Media Used for Isolation (Middlebrook 7H10 Agar (BBL, Difco))
Nineteen grams of the agar powder were suspended in 900 ml of purified water containing 5 ml of glycerol and mixed thoroughly. The mixture was heated with frequent agitation and boiled for one minute to completely dissolve the powder. The mixture was autoclaved and then 100 ml of Middlebrook OADC Enrichment were added to the medium when cooled to 50-55°C.

Preparation of Milk Samples
Sample preparation was performed according to a previously published procedure [12]. Two ml of 4% H 2 SO 4 were added to the sample and was incubated for 30 min at room temperature. Later, 16 ml of sterile distilled water were added to the previous mixture, and the whole mixture was centrifuged at 3000 rpm for 20 min. The supernatant was decanted into 5% phenol, and the sediment was used to make a direct smear, which was inoculated into two L-J medium slant (one contains 4% sodium pyruvate, and the other contains 5% glycerol) and was then incubated at 37°C for 3 weeks. Cultures were examined daily for one week and then once weekly for 6-8 weeks.

DNA Extraction from Blood Samples
DNA preparation from blood was performed by using DNA preparation Kit following the protocol of the manufacturer (Jenabioscience, Germany).

Confirmation of Mycobacterial Infection by Specific Gene Analysis
DNA, from blood samples collected from tuberculin-positive cattle farms, was tested for IS6110, katG, gyrA, oxyR, pncA genes (Table 1). Primer sequences for target genes were selected from previous literature [13] and were obtained from Metabion (Germany).

MIRU-VNTR Typing
MIRU-VNTR typing of DNA samples was performed using the primers shown in Table 1. Primer sequences for target loci were selected from previous reports [4,5,6]. Basically, the 20 μl PCR reaction consisted of 10 μl pf PCR mastermix (Intron Biotechnology, Korea), 1 μl (10 pmol) of each primer (forward or reverse), 2 μl of sample DNA and up to 20 μl nuclease-free water. The thermal cycling included 1 cycle of 12 min at 94°C, 40 cycles of 30 sec at 94°C, 1 min at the chosen annealing temperature according to the melting temperature of each primer sets (Table 1), 90 sec at 72°C, followed by 1 cycle of 7 min at 72°C. PCR products were resolved by electrophoresis using 1.5% agarose gels (in TAE buffer, Bioshop). The gel was placed in an electrophoresis tank containing 0.5% TBE and run for 1 h at 80 V. Markers (50 and 100 bp ladder, Qiagen and iNtTRON, respectively) were electrophoresed beside the DNA samples to measure the size of the DNA fragments. The DNA fragments were then visualised on a UV transilluminator using long wavelength ultraviolet light.

Data Analysis
Allele frequency for the single MIRU-VNTR locus was calculated as previously stated [14], where the genetic profile or the frequency of each assigned copy number (number of certain copy number appearance in the selected population) was divided by the number of samples/isolates in the selected population. Furthermore, genetic diversity for the locus or allele diversity (h) was calculated as h = 1 -Σxi 2 [n/(n -1)], where xi is the frequency of the ith allele at the locus, Σxi 2 is the summation of all recorded (allele frequency) 2 for a single locus, n is the number of isolates in the sample, and n/(n -1) is a correction for bias in small samples [14].

Mycobacterial Isolation
Milk samples (collected from farms that showed positive results for tuberculin test) did not show any positive result for mycobacterial isolation.

Confirmation of Mycobacterial Infection by Specific Gene Analysis
DNA, extracted from blood samples (collected from tuberculin-positive cattle farms), was tested for IS6110, katG, gyrA, oxyR, pncA genes. All DNA samples were positive for all tested genes, and gave the expected PCR product sizes (Table 2).

MIRU-VNTR Typing
Although the selected MIRU-VNTR loci gave positive PCR result for their relative copy number (Table 1), none of the selected MIRU-VNTR loci showed polymorphism (in the copy number) across all studied samples.

Allelic Diversity of Target MIRU-VNTR Loci
The allelic diversity was calculated for selected MIRU-VNTR loci (Table 3). Due to the lack of polymorphism (in the MIRU-VNTR copy number), none of the selected MIRU-VNTR loci showed any discriminatory power in the genetic analysis of field samples.
Mycobacterial isolation failed from collected milk samples from positive cases of tuberculin test. Milk was previously recorded as a poor carrier for bovine TB infection, where researchers obtained poor percentage (10%) for bovine TB isolation from positive or suspected cases of TB infection [15].
All tested DNA samples, extracted from blood samples (collected from tuberculin-positive cattle farms), were positive for the selected mycobacterial genes, IS6110, katG, gyrA, oxyR, pncA, at the expected product size. A previous study performed on M. bovis in Germany [13], revealed the positivity of some M. bovis-positive, IS6110-positive samples to katG, gyrA, oxyR, and pncA. However, in the same study, other M. bovis-positive, IS6110-positive samples were only positive for gyrA. In the same study, M. tuberculosis samples lacked one, two, or three of the four investigated targets (katG, gyrA, oxyR, and pncA), and only one sample contained the four targets.
The MIRU-VNTR loci analysis did not show any discriminatory power for bovine TB genetic analysis. Our findings were close to researchers in New Zealand who worked on M. bovis MIRU-VNTR typing [11], where MIRU-10, MIRU-16, MIRU-40, QUB-3232, QUB-3336, and QUB-1895 did not have any discriminatory power for Mycobacteria typing, and ETR-C and ETR-D had medium and high discriminatory power for Mycobacteria typing, respectively. However, in their study, ETR-B and ETR-A had medium and high discriminatory powers for Mycobacteria typing, respectively, and ETR-E did not have any discriminatory power for Mycobacteria typing. Our study findings were also close to the data revealed by Korean researchers who worked on M. bovis MIRU-VNTR typing, where MIRU-10, MIRU-16, MIRU-40, and ETR-C had no discriminatory power for bovine TB typing in Korea [6]. In their study, MIRU-26, QUB-1895, and ETR-B had minimal discriminatory power for bovine TB typing. On the other hand, QUB-3232 and ETR-A had medium discriminatory power, and QUB-3336 had the highest discriminatory power across all tested loci. Furthermore, MIRU-VNTR typing of bovine TB in Brazil, caused by M. bovis, revealed the high discriminatory power of ETR-A, ETR-B, ETR-C, and MIRU-16, when compared to the minimal to nil discriminatory power of MIRU-26, MIRU-40, and MIRU-10 [7]. The lack of discriminatory power in our result could be attributed to the small geographical area our investigation included.
In comparison with molecular typing of Mycobacterium tuberculosis, some ETRs were found polymorphic in Mycobacterium tuberculosis strains by other researchers in Thailand [9], where high discriminatory power for human TB typing was recorded by using ETR-B, ETR-D, and ETR-E. In contrast to our findings, where MIRUs did not have any discriminatory power, their work showed another high discriminatory power for TB typing by using MIRU-10, MIRU-16, and MIRU-26.

Conclusion
In the current study, we were able to perform MIRU-VNTR loci analysis for TB-infected cattle in Egypt. However, the MIRU-VNTR loci analysis did not any discriminatory power for bovine TB genetic analysis. Therefore, further analysis is required for clear distinction between field isolates/samples, either through using other MIRU-VNTR loci or through spligotyping.