Figure 8

Taxonomic rearrangements undergone by the E. agglomerans/E. herbicola complex in the last decades and attempts to assign still unassigned biotypes to known species. Strains belonging to the E. agglomerans/E. herbicola complex were described as early as 1888 [59] and included organisms that were saprophytes or plant pathogens [60, 61] or (opportunistic) pathogens in humans [61]. The name E. agglomerans was proposed by Ewing and Fife [50] after comparing plant and animal isolates as a subjective synonym for all three Erwinia species in the Herbicola group which was created in the meantime, i.e., E. herbicola, E. stewartii (now P. stewartii) and E. uredvora [62]. In this process, other Enterobacter strains may have been included in the new species. Brenner et al. [41] attempted to classify E. agglomerans strains by DNA hybridization and phenotypic tests deciding upon 13 biotypes. Subsequent classification efforts assigned several of the Brenner biotypes to new species, including P. agglomerans, P. dispersa, P. ananatis or Leclercia adecarboxylata [1, 52, 54], but for most reclassification with definitive assignment remains open. For these still unnamed biotypes an approximate classification, based on strain phylogeny (Figure 1 & 2) or 16S rDNA and gyrB sequence similarity (see Additional file 2- Table S2) is projected above.