The rational use of IS6110 fingerprinting for studies of the molecular epidemiology and evolution of Mycobacterium tuberculosis requires understanding of the dynamics of transposition. In laboratory model systems, it has been shown that transposition is context-sensitive, i.e. it is influenced by the nature of the site in which the insertion sequence is presented. Stimulation of transposition by activation of an adjacent promoter supports the hypothesis that transposition occurs more readily from transcriptionally active locations. In addition, it has been shown that transposition can be enhanced by the expression of the transposase in trans. These findings imply that the frequency of transposition will vary substantially between different strains of M. tuberculosis, and furthermore that a hitherto stable strain may develop more rapid variation due to transposition into an active site. The use of IS6110 fingerprinting for the analysis of longer-range relationships between M. tuberculosis isolates therefore needs to be interpreted with care.

SummaryThe Mycobacterium tuberculosis-speci®c insertion sequence IS6110/986 has been widely used as a probe because of the multiple polymorphism observed among different strains. To investigate transposition of IS6110, a series of arti®cially constructed composite transposons containing IS6110 and a kanamycin resistance marker were constructed. The composite transposons were inserted into a conditionally replicating, thermosensitive, Escherichia coli ±mycobacterial shuttle vector and introduced into M. smegmatis mc 2 155. Lawns of transformants were grown at the permissive temperature on kanamycin-supplemented agar and subsequently prevented from further growth by shifting to the nonpermissive temperature. Under normal atmospheric conditions, kanamycin-resistant papillae appeared after only about 5±6 weeks of incubation. However, these events were not associated with transposon mobilization. In contrast, lawns that were exposed to a 48 h microaerobic shock generated kanamycinresistant papillae after only 6±14 days. These events were generated by conservative transposition of the IS6110 composite transposon into the M. smegmatis chromosome, with loss of the shuttle vector. In common with other IS3 family elements, transposition of IS6110 is thought to be controlled by translational frameshifting. However, we were unable to detect any signi®cant frameshifting within the putative frameshifting site of IS6110, and the level of frameshifting was not affected by microaerobic incubation. The ®nding that transposition of IS6110 is stimulated by incubation at reduced oxygen tensions may be relevant to transposition of IS6110 in M. tuberculosis harboured within TB lesions.

Mosaic penicillin-binding proteins (PBP) 1A, 2X and 2B genes were cloned from four clinical isolates of Streptococcus pneumoniae with levels of susceptibility to penicillin ranging from 1.5 to 16 micrograms benzylpenicillin ml-1. In each instance it was possible to transform either the penicillin-sensitive laboratory strain R6 or a sensitive clinical isolate 110K/70 to the full level of penicillin resistance with these three penicillin-binding proteins alone. Until now it has not been possible to clearly determine whether alterations to PBP1A, 2X and 2B alone were sufficient to attain high level penicillin resistance.

Infections caused by vancomycin-resistant enterococci (VRE) are becoming increasingly prevalent throughout the world. Control measures include detection and isolation of carriers of VRE. A selective medium to detect faecal carriage of VRE is described. The medium has a high productivity ratio (90.4%) for VRE with VanA resistance phenotype, a moderate productivity ratio (79.2%) for VRE with VanB resistance phenotype, and a relatively low productivity ratio (65.5%) for VRE with VanC resistance phenotype. There was no breakthrough of vancomycin-susceptible enterococci. The medium selected the growth of all three types of VRE, which were used to spike faecal specimens. In a limited clinical trial, six faecal specimens of carriers and contacts were screened using the selective medium. Vancomycin-resistant enterococci (Enterococcus faecalis, VanA phenotype) were detected in four of the specimens. In all four specimens the growth of VRE was nearly pure and easily identifiable.