The role of phase variation of lic1A, lic2A and lic3A in the ability of Haemophilus influenzae type b to colonize the nasopharynx, bloodstream and cerebrospinal fluid (CSF) of infants was investigated. This was achieved by using PCR to determine the number of 5′-CAAT-3′ repeats present in each gene, which is indicative of whether each ORF can be expressed. Multiple PCR products of different intensities were amplified from all three genes at each site sampled. This indicated that the nasopharynx, bloodstream and CSF were colonized by a heterogeneous population of organisms, expressing different combinations of lic genes. At each site however, a predominant PCR product was amplified from each gene, indicating that organisms with this genotype were the most abundant. The number of 5′-CAAT-3′ repeats in this predominant product varied depending upon whether organisms were isolated from the nasopharynx, bloodstream or CSF. These observations suggest that the expression of different combinations of lic genes may influence the efficiency with which H. influenzae colonizes the nasopharynx, bloodstream and CSF of infant rats.
Mycobacterium tuberculosis has innate resistance to a range of broad-spectrum antimicrobial agents. This may in part reflect the relative impermeability of the mycobacterial cell wall, but additional specific mechanisms may also be important. In the case of fosfomycin, it has been suggested that a key difference in the active site of the M. tuberculosis MurA enzyme might confer resistance. In Escherichia coli, fosfomycin covalently binds to a cysteine normally involved in the enzymic activity, while protein alignments predict an aspartate at this position in the M. tuberculosis MurA. In the present study, it is demonstrated that the wild-type M. tuberculosis MurA is indeed resistant to fosfomycin, and that it becomes sensitive following replacement of the aspartate residue in position 117 by a cysteine. In addition, the study illustrates the use of an inducible expression system in mycobacteria to allow functional characterization of an M. tuberculosis enzyme that is unstable during constitutive expression.
The population structure of Streptococcus pneumoniae in a sample of 134 carried antibiotic-susceptible isolates, and 53 resistant and susceptible invasive isolates, was examined using a DNA-based version of multilocus enzyme electrophoresis: multilocus restriction typing (MLRT). This involved RFLP analysis of PCR products generated from nine loci of housekeeping genes located around the pneumococcal chromosome. The combination of alleles at each of the nine loci gave an allelic profile or restriction type (RT). All carried (throat or nasopharyngeal) isolates from children or adults in Oxford and Manchester, UK, and from an HIV-seropositive cohort in Nairobi, Kenya, showed an epidemic population structure. Twelve carried clonal groups, each with different serotypes, were identified at both locations within the UK. Almost all of the carried clones examined (16/17) were found to possess identical RTs or sequence types (STs) to invasive isolates, indicating that frequently carried clones are also associated with cases of invasive disease. As expected from previous studies, the population of 53 invasive, mainly penicillin-resistant, isolates was also found to be at linkage equilibrium. Serotype switching was identified among 14% of RTs that possessed two or more members, or 5·7% of individual isolates within these RTs. In support of a population structure in which there is frequent recombination, there is also clear evidence that the trpA/B locus within pneumococci has evolved by horizontal gene transfer. A non-serotypable isolate from an HIV-seropositive patient in Kenya was clearly genetically distinct from other strains studied, with unique alleles at eight out of nine loci examined. However, it was initially identified as a pneumococcus by a 16S RNA gene probe (Gen-Probe), optochin susceptibility and the presence of pneumolysin and autolysin.
The standard Escherichia coli collection of reference (ECOR) strains was examined for ability to exert cytotoxicity towards mammalian cells. A group of strains with functional haemolysin expression caused strong cytotoxicity and detachment in J774 macrophage cells as measured by lactate dehydrogenase release and as observed under a microscope. The expression of haemolysin was monitored by using antisera recognizing the E. coli α-haemolysin, the HlyA protein, and by quantitative haemolysis assays. The presence of the hlyA gene, which may be part of a pathogenicity island, was also confirmed. These analyses revealed that different ECOR strains express quantitatively different levels of haemolysin. One putative enteroaggregative E. coli (EAEC) strain was also found in the ECOR collection. The EAEC strain was characterized by the clump formation assay, PCR amplification of the EAEC DNA probe sequence and confirmative sequence analysis of the amplified fragment. The EAEC heat-stable enterotoxin 1 gene, astA, was found in 14% (10/72) of the ECOR strains and a consensus sequence for astA was proposed by comparing these sequences with those from pathogens. The astA gene appeared to be plasmid-located. Based on evidence from the work of other laboratories and from the present findings, it is concluded that the ECOR collection contains strains that may represent pathogenic E. coli. It is noted that caution is necessary when handling or disposing of those potentially pathogenic ECOR strains.