Evaluation of MIRU-VNTR for typing of Mycobacterium bovis isolated from Sika deer in Northeast China

Background Bovine tuberculosis has led to serious economic losses for Sika Deer producers in China. Strategies for controlling the spread of Mycobacterium bovis are often hampered by a lack of epidemiological data. Specifically, tracing infections requires the ability to trace back infections, which, in turn, requires the ability to determine isolates with a common source. This study was planned to assess the discriminatory power of each mycobacterial interspersed repetitive unit (MIRU)-variable number tandem repeats (VNTR) locus and evaluate the most appropriate combination of MIRU-VNTR loci for molecular epidemiological studies on Sika Deer in China. Results The discriminatory power of MIRU-VNTR typing based on 22 known loci (12 MIRUs, 2 ETRs, 4 QUBs, and 4 Mtubs) were assessed in 96 Mycobacterium bovis strains collected sequentially from Sika Deer at a slaughterhouse in northeastern China. We defined four loci (MIRU4, ETRA, QUB11b, and Mtub4) as highly discriminative, eight loci (MIRU2, MIRU23, MIRU27, MIRU31, MIRU39, MIRU40, QUB26, and Mtub21) as moderately discriminative, and three loci (MIRU16, Mtub30, and Mtub34) as poorly discriminative. The final locus showed no polymorphism between strains. MIRU-VNTR typing as a whole was highly discriminative, with an overall allelic diversity of 0.897. Of the loci tested, the four highly discriminative loci and eight moderately discriminative loci proved to be most appropriate for first line typing of M. bovis from Sika Deer, with the same resolving ability as all 22 loci (H = 0.897). Conclusions MIRU-VNTR typing is quick and effective for typing bovine tuberculosis isolates from Sika Deer in China. Electronic supplementary material The online version of this article (doi:10.1186/s12917-015-0402-0) contains supplementary material, which is available to authorized users.


Background
Sika Deer tuberculosis is caused by the Mycobacterium tuberculosis complex (MTBC); most commonly, M. bovis. This disease has led to significant economic losses for deer producers in China. However, previous efforts in controlling MTBC have been hampered by a lack of data, as genotyping MTBC members is an important tool in epidemiological analysis for studying the spread of MTBC [1]. In 2001, VNTR genotyping method using MIRU units was developed [2] and used to study the molecular epidemiology of Mycobacterium tuberculosis [3]. MIRU-VNTR has high efficiency and reproducibility, because MIRU-VNTR typing results are easier to compare between different laboratories than restriction fragment length polymorphisms (RFLP), which were the previous "gold standard" for MTBC typing [4]. Another advantage of MIRU-VNTR is that it does not require large quantities of template DNA for typing [5].
In addition to M. tuberculosis, this technique has also been proven in M. bovis in different countries and epidemiologic scenarios [1,[6][7][8][9][10][11]. Earlier studies frequently used the loci designated as mycobacterial interspersed repeat units (MIRU) [2,12] and exact tandem repeats (ETR) [13]. To obtain better resolution, other VTNR loci were investigated [14]. Recently, novel M. tuberculosis MIRU-VNTR loci groups, such as the Queen's University Belfast (QUB) and Mtub groups [15], have been identified. Various combinations of MIRU-VNTR loci have been reported to give better differentiation of M. tuberculosis strains [16][17][18][19]. However, no specific set of loci have been agreed upon as a standard, and the allelic diversity of loci can vary from country to country and between M. tuberculosis complex species, requiring localized selection of suitable loci.
This study was therefore developed to determine the MIRU-VNTR loci most appropriate to use for typing MTBC organisms isolated from Sika deer in China.
The discrimination ability was compared with different combinations of loci ( Table 3). The four highly discriminatory loci and the eight moderately discriminatory loci, also identified 12 genotypes (H = 0.897). The MIRU loci with moderate discriminatory power (MIRU2, MIRU23, MIRU27, MIRU31, MIRU39, and MIRU40) resolved the 96 isolates into 10 types (H = 0.877). If the MIRU4 and MIRU16 loci were also used, 11 types (H = 0.893) were resolved. With the Mtub4, Mtub21, Mtub30, and Mtub34 loci, 8 types (H = 0.836) were resolved. Based on these results, we consider ETRA, MIRU4, QUB11b, and Mtub4 to be robust enough for initial molecular epidemiological studies when rapid results are of paramount importance.

Discussion
M. bovis is the primary pathogen for tuberculosis in Sika Deer. Because it also infects humans, molecular epidemiologic studies on M. bovis are of major importance in tuberculosis prevention. Genotyping of MTBC isolates is a useful tool, not only for regional investigations, such as tracebacks, but to allow comparisons of tuberculosis isolates worldwide [20]. Different typing methods published for M. bovis include IS6110, RFLP, VNTR, and spoligotyping [21]. While IS6110 has been the most common, the discriminatory power of IS6110 is generally low for M. bovis typing [22,23]. Because the resolution provided by MIRU-VNTR is adequate for most situations and this method is relatively cheap and rapid, it is quickly becoming the standard as the initial step in M. bovis epidemiology [14].
The individual discriminatory power of 22 MIRU-VNTR loci were analyzed. The QUB11b locus produced the highest individual discriminatory index (h = 0.739), followed by Mtub4, ETRA, and MIRU4. This correlates well with previous studies [14,15].
Mtub loci exhibit good application potential; however, the discriminatory power of Mtub4, Mtub21, Mtub30, and Mtub34 differed greatly, with h ranging from 0.108 to 0.613. These loci exhibited a high enough discrimination ability to be useful in early research [24,25], and may prove useful in future M. bovis typing studies. The ETRC, MIRU10, MIRU20, MIRU24, MIRU26, QUB1895, and QUB3336 loci showed no allelic diversity in this test.  The allelic diversity of MIRU-VNTR loci are higher in our study than previous studies in China [26,27], Published data on M. tuberculosis from France [28], South Africa [29], and the United States [30] found that the discriminatory power of MIRU-VNTR loci in Sika Deer is higher than in humans. Although these MIRU-VNTR loci showed enough discrimination ability for genotyping current M. bovis isolates, additional studies with new MIRU-VNTRs are needed to ensure accurate genotyping and epidemiological resolution for field studies.
The discriminatory power of MIRU-VNTR loci may vary between countries, geographical regions, and epidemiological scenarios. Analysis of 15 VNTR loci showed the expected results: when compared with previous studies the test had a relatively high Hunter-Gaston discriminatory index (HGDI) and low clustering rates [31][32][33]. Therefore, the discriminatory power of VNTR loci appears greatly improved by applying novel combinations of VNTR loci.
The 15 VNTR loci methodology is a highly discriminatory method for first line typing of M. bovis isolated from Sika deer and should be considered as a replacement for the original 12 MIRU loci method. However, for in-depth studies, we recommend using ETRA, MIRU4, QUB11b, and Mtub4 loci as a first set to obtain initial molecular epidemiology data quickly before proceeding on to the full set of loci.

Conclusions
The results of this study will be used in a molecular epidemiology study to identify risk factors for recent

Bacterial strains
A total of 96 M. bovis strains were used in this study. These strains were isolated from samples collected at the slaughterhouse for Sika Deer in China covering five different regions: Shuang Yang, Liao Yuan, Tong Hua, Si Ping, and Song Yuan between 2008 and 2012. All 96 strains were characterized by real-time PCR [34].

Ethical considerations
This research was performed according to the international, national and institutional rules regarding animal experimentation. The research was approved by the Institutional Animal Care and Use Committee of Jilin Agricultural University (reference number:2008627).

Preparetion of chromosomal DNA
DNA was extracted with the method followed. Cells were suspended in 400 μL TE buffer (10 mM Tris-CI and 1 mM EDTA at pH 8.0) and inactivated at 80°C for 30 min followed by centrifugation at 5000 × g. The supernatant was discarded and the pellet resuspended in 400 μL TE buffer and 45 μL 10% SDS. This was heated in a 37°C water bath for 2 h after incubation with 3 μL of 20 μg/mL proteinase K. The suspension was recentrifuged and the supernatant extracted using an equal volume of phenol/chloroform/isoamyl alcohol (25:24:1 [v/v]). The resulting supernatant was diluted with a double volume of anhydrous ethanol for 10 min and precipitated at −20°C after centrifugation at 12000 × g for 20 min. The precipitate was washed with 70% ethanol twice, and the ethanol evaporated at room temperature for 20 min. The final pellet was dissolved in 30 μL sterile deionized water and stored at −20°C until use.

PCR amplification
The primers and designations used in this study for each MIRU-VNTR locus are described by Supply et al. [14]. The primer sequences and PCR product sequencing in this study was described in Additional file 2. PCR reactions were performed using the method described in

Allelic diversity and VNTR typing
The individual allelic diversity (h) of each VNTR locus was calculated using the following equation: h = 1-∑xi 2 [n/(n − 1)], where n is the number of isolates and xi the frequency of the ith allele at the locus [35]. We considered h > 0.6 as highly discriminative, 0.3 ≤ h ≤ 0.6 as moderately discriminative, and h < 0.3 as poorly discriminative [36]. The HGDI was calculated as previously described [37]. The HGDI varies between 0.00 and 1.00 and represents the discriminatory power for VNTR loci in combination. Clustering analyses were performed using the online tool at http://www.MIRU-VNTRplus.org.

ALL MIRUs
Start at a denaturing step of 15 min at 95°C, after denaturation, the PCR was performed for 40 cycles of 1 min at 94°C, 1 min at 59°C and 1.5 min at 72°C. Terminated by an incubation of 10 min at 72°C.
ETRA, C and QUB26 Start at a denaturing step of 12 min at 95°C, after denaturation, the PCR was performed for 40 cycles of 0.5 min at 94°C, 1 min at 60°C and 2 min at 72°C. Terminated by an incubation of 7 min at 72°C.
QUB3336, 1895, 2163 Start at a denaturing step of 15 min at 95°C, after denaturation, the PCR was performed for 40 cycles of 0.5 min at 94°C, 1 min at 55°C and 2 min at 72°C. Terminated by an incubation of 7 min at 72°C.
Mtub4, 21,30,34 Start at a denaturing step of 15 min at 95°C, after denaturation, the PCR was performed for 40 cycles of 1 min at 94°C, 1 min at 50°C and 0.5 min at 72°C. Terminated by an incubation of 7 min at 72°C.