RESUMEN
BACKGROUND: Testing of antibiotic resistance of intact vaginal microbiota in pure culture is not feasible. METHODS: Metronidazole, antiseptic octenisept®, antimycotic ciclopirox, bacterial probiotic Lactobacillus crispatus, yeast probiotic Saccharomyces boulardii, Gardnerella-phage-endolysin named phagolysin and phagolysin in combination with probiotics were tested for bacteriolytic activity. Included were vaginal swabs from 38 random women with Amsel-confirmed bacterial vaginosis (BV). Test aliquots were incubated by 37° for 2 and 24 h. Gardnerella, low G+C, Atopobium, lactobacilli, Lactobacillus iners and crispatus, Prevotella-Bacteroides, and Gammaproteobacteria microbial groups were quantified using fluorescence in situ hybridization (FISH). RESULTS: The probiotic strain Lactobacillus crispatus demonstrated the weakest bacteriolytical effects, followed by metronidazole. Both had no impact on Gardnerella species, instead lysing Prevotella-Bacteroides, Enterobacteriaceae (by L.crispatus) or LGC, Atopobium and Prevotella-Bacteroides (by metronidazole) groups of the microbiota. Cytolytic activity on Gardnerella was highly pronounced and increased from octenisept to ciclopirox, phagolysin, phagolysin with L.crispatus, being best in the combination of phagolysin with S.boulardii. Universally active ciclopirox and octenisept® suppressed nearly all microbial groups including those which are regarded as beneficial. Phagolysin had no effect on naturally occurring Lactobacillus crispatus. Conclusions: FISH susceptibility testing allows unique efficacy evaluation of individually adjusted topical therapy without microbial isolation facilitating optimal therapy choice.
RESUMEN
Alternative treatments for Escherichia coli infections are urgently needed, and phage therapy is a promising option where antibiotics fail, especially for urinary tract infections (UTI). We used wastewater-isolated phages to test their lytic activity against a panel of 47 E. coli strains reflecting the diversity of strains found in UTI, including sequence type 131, 73 and 69. The plaquing host range (PHR) was between 13 and 63%. In contrast, the kinetic host range (KHR), describing the percentage of strains for which growth in suspension was suppressed for 24 h, was between 0% and 19%, substantially lower than the PHR. To improve the phage host range and their efficacy, we bred the phages by mixing and propagating cocktails on a subset of E. coli strains. The bred phages, which we termed evolution-squared ε2-phages, of a mixture of Myoviridae have KHRs up to 23% broader compared to their ancestors. Furthermore, using constant phage concentrations, Myoviridae ε2-phages suppressed the growth of higher bacterial inocula than their ancestors did. Thus, the ε2-phages were more virulent compared to their ancestors. Analysis of the genetic sequences of the ε2-phages with the broadest host range reveals that they are mosaic intercrossings of 2-3 ancestor phages. The recombination sites are distributed over the whole length of the genome. All ε2-phages are devoid of genes conferring lysogeny, antibiotic resistance, or virulence. Overall, this study shows that ε2-phages are remarkably more suitable than the wild-type phages for phage therapy.