In this study we report on the development of a multilocus variable-number tandem-repeat (VNTR) analysis (MLVA) method for the molecular typing of Mycoplasma pneumoniae. The genomic content of M. pneumoniae M129 was analyzed for VNTRs, and 5 of the 17 VNTRs identified were selected for use in an MLVA assay. The method was based on a GeneScan analysis of VNTR loci labeled with fluorescent dyes by multiplex PCR and capillary electrophoresis. This approach was applied to a collection of 265 isolates from various European countries, Japan, and Tunisia; and 26 distinct VNTR types were found. The VNTR assay was compared to the P1 adhesin PCRrestriction fragment length polymorphism (RFLP) typing method and showed a far better resolution than the P1 PCR-RFLP method. The discriminatory power of MLVA (Hunter-Gaston diversity index [HGDI], 0.915) for the 265 isolates was significantly higher than that of the P1 PCR-RFLP method (HGDI, 0.511). However, there was a correlation between the typing results obtained by MLVA and the P1 gene PCR-RFLP method. The potential value of MLVA of M. pneumoniae as an epidemiological tool is discussed, and the use of the VNTR markers in further investigations of the potential use of MLVA in outbreaks of M. pneumoniae infections is proposed.Mycoplasma pneumoniae represents one of the most common etiological agents of community-acquired respiratory tract infections, with the clinical courses ranging from mild forms of pharyngitis and tracheobronchitis to severe cases of interstitial pneumonia. M. pneumoniae infections are responsible for 20% or more of all cases of community-acquired pneumonia (22) and occur endemically, with epidemic peaks occurring every 4 to 7 years. School-age children and young adults are the populations that are the most affected.Strain subtyping by molecular methods is a powerful tool for surveillance and outbreak investigation. However, molecular typing is hampered by the fact that M. pneumoniae is a genetically homogeneous species and isolates are poorly differentiated by PCR-restriction fragment length polymorphism (RFLP) analysis of the P1 gene, the most common genotyping method. PCR-RFLP analysis of the gene encoding the P1 protein, a major adhesin that induces a strong humoral immune response during infection, enables the separation of isolates into two types, types 1 and 2 (1, 2, 17). More recent studies used repetitive regions, RepMp2/3 and RepMp4, which can be found in the P1 gene, for molecular typing. Besides the subtype classification, those methods allow the identification of variants of subtypes 1 and 2 (3,5,8).In addition to these molecular typing methods that target only one gene, other methods based on the study of the whole genome have been adapted to M. pneumoniae, such as pulsedfield electrophoresis (1) and multilocus sequence typing (MLST) (4). Pulsed-field electrophoresis is able to differentiate strains into two groups (groups 1 and 2), like P1 gene PCR-RFLP analysis can, and can subdivide group 2 into two subgroups, subgroups 2a and 2b (1). Due t...

Resistance of Helicobacter pylori to macrolides is a major cause of failure of eradication therapies. Single base substitutions in the H. pylori 23S rRNA genes have been associated with macrolide resistance in the United States. Our goal was to extend this work to European strains, to determine the consequence of this mutation on erythromycin binding to H. pylori ribosomes, and to find a quick method to detect the mutation. Seven pairs of H. pylori strains were used, the parent strain being naturally susceptible to macrolides and the second strain having acquired an in vivo resistance during a treatment regimen that included clarithromycin. The identity of the strains was confirmed by random amplified polymorphic DNA testing with two different primers, indicating that resistance was the result of the selection of variants of the infecting strain. All resistant strains were found to have point mutations at position 2143 (three cases) or 2144 (four cases) but never on the opposite DNA fragment of domain V of the 23S rRNA gene. The mutation was A-->G in all cases except one (A-->C) at position 2143. Using BsaI and BbsI restriction enzymes on the amplified products, we confirmed the mutations of A-->G at positions 2144 and 2143, respectively. Macrolide binding was tested on purified ribosomes isolated from four pairs of strains with [14C]erythromycin. Erythromycin binding increased in a dose-dependent manner for the susceptible strain but not for the resistant one. In conclusion we suggest that the limited disruption of the peptidyltransferase loop conformation, caused by a point mutation, reduces drug binding and consequently confers resistance to macrolides. Finally, the macrolide resistance could be detected without sequencing by performing restriction fragment length polymorphism with appropriate restriction enzymes.

No abstract