Possible Misidentification of Mycobacterium yongonense

To the Editor: Tortoli et al. (1) reported pulmonary disease caused by M. yongonense strains isolated from patients in Italy; these strains were identified by sequencing the 16S rRNA, hsp65, rpoB, and sodA genes and the internal transcribed spacer 1 (ITS1) region. The 16S rRNA gene sequence of these isolates showed 100% similarity with those of M. yongonense and M. marseillense. The isolates were more closely related to M. yongonense than to M. marseillense in terms of the hsp65 gene and ITS1 region; however, the rpoB gene sequence showed a higher degree of similarity to that of M. intracellulare (99.4%) than to that of M. marseillense (97.4%). The authors did not mention the similarity of the isolates with M. intracellulare in these sequences except for the rpoB gene. However, because these sequences showed high similarity to M. yongonense, a high degree of similarity to M. intracellulare could be inferred.

The initial description of M. yongonense highlighted its unique molecular character (2). The 16S rRNA and hsp65 genes and ITS1 region are closely related to those of M. intracellulare ATCC 13950^T; however, the rpoB gene is closely related to that of M. parascrofulaceum ATCC BAA-614^T (99.4%). No consensus guidelines are available for mycobacterial identification, but the rpoB gene has been used widely as a target gene; multilocus sequence analysis also has been used recently (3,4). Although the authors suggest that a variant of M. yongonense preceded the acquisition of the rpoB gene from M. parascrofulaceum by a lateral gene transfer event (3), the isolates described are more similar to M. intracellulare than to M. yongonense on the basis of the rpoB gene sequence and multilocus sequence analysis. It is also possible that the isolates are a M. yongonense strain that preceded the acquisition of the rpoB gene but that are not the same as the initially described M. yongonense.