Differentiation of phylogenetically related slowly growing mycobacteria based on 16S-23S rRNA gene internal transcribed spacer sequences

Abstract

Interspecific polymorphisms of the 16S rRNA gene (rDNA) are widely used for species identification of mycobacteria. 16S rDNA sequences, however, do not vary greatly within a species, and they are either indistinguishable in some species, for example, in Mycobacterium kansasii and M. gastri, or highly similar, for example, in M. malmoense and M. szulgai. We determined 16S-23S rDNA internal transcribed spacer (ITS) sequences of 60 strains in the genus Mycobacterium representing 13 species (M. avium, M. conspicuum, M. gastri, M. genavense, M. kansasii, M. malmoense, M. marinum, M. shimoidei, M. simiae, M. szulgai, M. triplex, M. ulcerans, and M. xenopi). An alignment of these sequences together with additional sequences available in the EMBL database (for M. intracellulare, M. phlei, M. smegmatis, and M. tuberculosis) was established according to primary- and secondary-structure similarities. Comparative sequence analysis applying different treeing methods grouped the strains into species-specific clusters with low sequence divergence between strains belonging to the same species (0 to 2%). The ITS-based tree topology only partially correlated to that based on 16S rDNA, but the main branching orders were preserved, notably, the division of fast-growing from slowly growing mycobacteria, separate branching for M. simiae, M. genavense, and M. triplex, and distinct branches for M. xenopi and M. shimoidei. Comparisons of M. gastri with M. kansasii and M. malmoense with M. szulgai revealed ITS sequence similarities of 93 and 88%, respectively. M. marinum and M. ulcerans possessed identical ITS sequences. Our results show that ITS sequencing represents a supplement to 16S rRNA gene sequences for the differentiation of closely related species. Slowly growing mycobacteria show a high sequence variation in the ITS; this variation has the potential to be used for the development of probes as a rapid approach to mycobacterial identification.

FIG. 1

Alignment of 16S-23S rDNA ITS sequences including those of 13 mycobacterial species investigated in this study together with sequences of 4 other species published elsewhere (9, 10, 14, 34) (Table 1). Sequevar designations are shown in parentheses. The sequence of M. simiae (Msi-B) was published by De Smet et al. (4). The length of the ITS is indicated at the end of the sequences in nucleotides. The complete ITS sequence between the end of the 16S rRNA gene and the beginning of the 23S rRNA gene is shown. With the exception of M. conspicuum, whose ITS ends with GG, the 3′ end of the ITS was inferred to be GTGT. Dots indicate identity, and hyphens represent alignment gaps. Possible features of secondary structures predicted in analogy to those previously proposed for pre-rRNA transcripts of mycobacteria are indicated by labelled arrows, whereas the stem-loop structures designated leader and spacer 2 interact with leader and 23S-5S rRNA spacer regions, respectively (14, 15).

FIG. 1

Alignment of 16S-23S rDNA ITS sequences including those of 13 mycobacterial species investigated in this study together with sequences of 4 other species published elsewhere (9, 10, 14, 34) (Table 1). Sequevar designations are shown in parentheses. The sequence of M. simiae (Msi-B) was published by De Smet et al. (4). The length of the ITS is indicated at the end of the sequences in nucleotides. The complete ITS sequence between the end of the 16S rRNA gene and the beginning of the 23S rRNA gene is shown. With the exception of M. conspicuum, whose ITS ends with GG, the 3′ end of the ITS was inferred to be GTGT. Dots indicate identity, and hyphens represent alignment gaps. Possible features of secondary structures predicted in analogy to those previously proposed for pre-rRNA transcripts of mycobacteria are indicated by labelled arrows, whereas the stem-loop structures designated leader and spacer 2 interact with leader and 23S-5S rRNA spacer regions, respectively (14, 15).

FIG. 2

Distance matrix tree showing the divergence of ITS sequences of the investigated mycobacteria. All alignment positions which are occupied by residues were used for the calculation of binary distance values. The topology of the tree was evaluated and corrected according to the results of maximum-parsimony and maximum-likelihood analyses. Multifurcations indicate that a relative branching order could not be unambiguously determined or that a common branching order was not supported by the different treeing methods. The corresponding sequences of the fast-growing mycobacteria M. phlei and M. smegmatis were used as outgroup references. Numbers in brackets indicate the numbers of strains sequenced. The bar represents 10% estimated sequence divergence.

FIG. 3

16S rRNA-based distance matrix tree for a selection of mycobacterial species. The tree was reconstructed with all available, at-least-90%-complete (with respect to the homologous E. coli molecule), 16S rRNA primary structures of gram-positive bacteria with a high DNA G+C content as well as from a selection of reference organisms of the other major bacterial phyla. The topology of the tree was evaluated and corrected by the methods applied to ITS sequences (Fig. 2). The bar indicates 5% estimated sequence divergence.