In Vivo IS6110 Profile Changes in a Mycobacterium tuberculosis Strain as Determined by Tracking over 14 Years

Abstract

Transposition and homologous recombination of IS6110 appear in Mycobacterium tuberculosis along in vivo sequential infections. These events were checked in different clones of a successful strain, M. tuberculosis Zaragoza, with the focus on a variant in which integration of a copy of IS6110 in the origin of replication (oriC) region occurred.

TEXT

Clonal Mycobacterium tuberculosis variants appear along the sequential infections of hosts involved in the same transmission cluster (1). IS6110 in M. tuberculosis has been one of the most widely used elements in molecular epidemiology and has a transposition rate estimated at ∼18% over a period of 5 to 6 years (2). The IS6110 element itself modulates expression of neighboring genes and, depending on its location, could confer both transmission ability and virulence (3–5). Microevolution events due to the transposition of IS6110 are usually reflected in variations of restriction fragment length polymorphism (RFLP) patterns (6, 7). Minor variations in IS6110 fingerprints, spoligotypes, or mycobacterial interspersed repetitive-unit (MIRU)-variable-number tandem-repeat (VNTR) loci reveal microevolution in clonal infections (6, 8). These changes could have effects on the molecular structure of the mycobacterial genome and the biology of the bacillus (7).

Molecular epidemiological studies carried out for more than a decade have provided accuracy in the study of the transmission (9). The usefulness of the approach has been proven, even in countries with low incidences of tuberculosis (10). The M. tuberculosis Zaragoza strain spread in Zaragoza, Spain, and reached 18.7% of all isolates of M. tuberculosis in 2001 to 2004 (11). This strain was classified in principal genetic group 3 and in single-nucleotide polymorphism (SNP) cluster group 6a (12), it demonstrated the rare spoligotype SIT773, and it carried 12 localized copies of IS6110 (13). In addition, in the context of a high-throughput survey of in vivo IS6110 transposition in multiple M. tuberculosis genomes, an additional IS6110 copy was detected in the oriC region in one isolate (14). The purpose of this study was to review the in vivo genomic changes of the M. tuberculosis Zaragoza strain, focusing on this variant.

Among the 2,348 isolates collected between 2000 and 2013, 246 were identified as M. tuberculosis Zaragoza by IS6110 RFLP typing (15) and spoligotyping (16) in the context of tuberculosis surveillance in our region. By tracking the strain changes throughout time, we found 7 different variants, i.e., variants 1 to 7 (Fig. 1). An additional copy of IS6110 was detected in six evolved clones, namely, variants 1 to 6. Variant 7 lost one copy in the RFLP pattern and was the only variant that revealed a change in its spoligotyping pattern. Clones 1, 2, 4, 5, 6, and 7 were isolated from only one patient, whereas variant 3 was isolated from 12 patients, as described below.

FIG 1.

Dendrogram showing the IS6110 RFLP patterns of one M. tuberculosis Zaragoza (MTZ) strain and its seven variants, including the years of isolation. The molecular sizes are indicated at the top (in kilobases).

In variant 7, 10 of the 12 IS6110 insertion sites known for M. tuberculosis Zaragoza were confirmed by PCR (13). However, the two nonamplified copies were located a relatively small distance from each other (9,860 bp for the H37Rv reference strain). The suspicion that homologous recombination between the two copies had occurred was corroborated by amplification of the expected sequence using the following external primers: DR43-F, 5′-ACCCGGTGCGATTCTGCG-3′; 2823-R, 5′-AAGGTGATCGAGGAGAAGTACCGGC-3′. The amplicon obtained in variant 7 (1,660 bp) was sequenced, and BLAST searching was performed (http://genolist.pasteur.fr/TubercuList). The sequence confirmed the deletion of ∼10 kb, including the genes from Rv2816c to the point of insertion of IS6110 in Rv2823c. The loss of this region is reflected in the lack of spacers 1 to 24 of the spoligotyping pattern (SIT585), classified as a Beijing-like pattern in the SITVITWEB database, in contrast to the pattern of the M. tuberculosis Zaragoza strain (SIT773). This process of homologous recombination has been described to explain the deletion of other extensive regions (17). Our finding agrees with other studies indicating that strains with large numbers of IS6110 copies have lost genomic regions due to recombination between copies (18). This mechanism was previously illustrated for other isolates of M. tuberculosis, demonstrating evolution of the spoligotyping pattern (19). Curiously, this microevolved clone did not spread to other patients.

In variant 3, a copy of IS6110 was detected, in the context of a high-throughput survey of in vivo IS6110 transposition, in the oriC region of the clone, which was not reflected in the RFLP pattern. Accordingly, it has been described that RFLP analysis can underestimate the real copy number for the IS6110 element (14, 20). The oriC region in all available M. tuberculosis Zaragoza isolates was analyzed by PCR, which showed a fragment of 2,056 bp, reflecting a copy of IS6110 in this region, for 12 isolates (5.06%) (21). The exact point of IS6110 insertion (bp 1625 to 1627, with reference to H37Rv) and the direct orientation in the genome were confirmed by sequencing and gapped BLAST analysis. As a result of the IS6110 transposition, 3-bp (CAC) direct repeats were generated.

The oriC region has been considered a hot spot in the genome of M. tuberculosis, as at least 10 different insertion sites for IS6110 have been described (22). This location has been considered a specific marker to characterize the Beijing family (23, 24). The IS6110 insertion point in variant 3 of M. tuberculosis Zaragoza is distinct from the points described, although it is relatively close (33 bp) to the Beijing point of insertion.

Some studies have suggested that the presence of one IS6110 copy in the oriC region increases the duplication time for M. tuberculosis and this effect is independent of the orientation of the insertion element (25). It has also been reported that deletions in the oriC region can abolish the activity at the replication origin and that point mutations in the DnaA boxes severely reduce the activity of oriC. Concerning this, most IS6110 copies have been located outside the DnaA boxes, and the oriC activity has not been affected (26). In this sense, the insertion of IS6110 in the oriC region could be an evolutionary advantage (7). In our study, we observed that the insertion of IS6110 is located outside the described DnaA boxes; therefore, we can assume that IS6110 does not decrease the activity of the oriC region. Accordingly, and in contrast to the other variants, this variant was isolated from 12 patients.

A 24-locus MIRU-VNTR analysis (27) was performed for five M. tuberculosis Zaragoza isolates, and one difference in locus 2163b was observed in variant 3, with respect to the reference M. tuberculosis Zaragoza strain (Table 1). In order to study when this variation in the number of copies appeared, the 2163b VNTR was analyzed by simplex PCR amplification in all of the M. tuberculosis Zaragoza isolates. Nineteen of the isolates, including the 12 isolates of variant 3, showed 3 tandem repeats, in contrast to the rest of the isolates, which carried 4 copies. The fact that other isolates presented 3 copies of the repeated sequence in the 2163b locus in the absence of the IS6110 insertion in oriC suggests that the event in 2163b occurred before the transposition of IS6110 in the oriC region. Curiously, among the 7 isolates sharing this feature, two of the variants (variants 5 and 6) exhibited this variation, although no clinical significance could be attributed.

TABLE 1.

Twenty-four-locus MIRU-VNTR patterns for five M. tuberculosis Zaragoza isolates, including three M. tuberculosis Zaragoza pattern isolates and two isolates of variant 3, showing the signature (2163b) detected from 2006

Isolatea	No. of repeats at positionb:
	154	424	577	580	802	960	1644	1955	2059	2165	2347	2401	2461	2531	2687	2996	3007	3171	3192	3690	4052	4156	4348	2163b
MTZ-1	2	2	4	2	2	3	4	2	2	2	4	2	2	5	1	5	3	3	3	3	5	2	2	4
MTZ-2	2	2	4	2	2	3	4	2	2	2	4	2	2	5	1	5	3	3	3	3	5	2	2	4
MTZ-3	2	2	4	2	2	3	4	2	2	2	4	2	2	5	1	5	3	3	3	3	5	2	2	4
MTZ-v3-1	2	2	4	2	2	3	4	2	2	2	4	2	2	5	1	5	3	3	3	3	5	2	2	3
MTZ-v3-2	2	2	4	2	2	3	4	2	2	2	4	2	2	5	1	5	3	3	3	3	5	2	2	3

^aMTZ, M. tuberculosis Zaragoza; MTZ-v3, M. tuberculosis Zaragoza variant 3.

^bLoci are indicated by their positions (in kilobases) on the M. tuberculosis H37Rv chromosome.

In conclusion, in monitoring of the M. tuberculosis Zaragoza strain for 14 years, different clones have been detected. Although the turnover of the markers used is not in the range of the pace of transmission, the insertion of an IS6110 copy in oriC allowed us to distinguish one large subgroup. Variant 3 was able to be transmitted in a circumscribed area of the city. Interestingly, this well-studied outbreak will allow us to perform further research comparing variants belonging to the same isogenic group and therefore possessing the same genetic background. In addition, whole-genome sequencing would be useful for determining the microevolution of the M. tuberculosis Zaragoza strain during the well-documented outbreak and would be useful to validate the directionality and sequence of transmission (28).

The study and the protocols for collecting the bacterial strains were approved by the ethics committee of the Aragon government.