The taxonomy and ecology of the genus Thermus

Abstract

The phenotype distribution of Thermus isolates was examined by numerical classification using isolates from New Zealand, Iceland, New Mexico and Yellowstone Park. Analysis of the New Zealand isolates showed that there was a relationship between the hot pool temperatures and pHs, and the phenotypes which could be isolated from them. Evidence also emerged which suggested that geographical factors, between distant thermal areas, may also influence the phenotypes isolated from those areas. Analysis of the Icelandic isolates supported these assertions by showing a lack of similarity to the New Zealand isolates. A classification was constructed using all of the available isolates and this showed the same patterns. All of the isolates could be divided into 8 major clusters and it is suggested that these could represent species. In this classification T. ruber and T. aquaticus would be type strains of two of the species as would the invalidly named “T. thermophilus”. Continuous culture experiments and analysis of isolates taken from hot pools over a period of time, did not support the proposal of genetic instability in Thermus. Varying the pH and temperature of continuous culture did not significantly change the phenotype. Analysis of isolates from the same pools over a time course showed there to be resident and transient strains, as well as showing a limitation of the phenotypes which could be recovered from any one pool. A new isolate is described, which formed trichomes that further coiled to form braiding. Trichomes also exhibited swellings and necridia. Thin-section electronmicroscopy showed a similar cell wall to Thermus with the addition of 2 extra layers which appeared to hold the cells together in the form of trichomes. The isolate showed many features in common with Thermus and the only significant difference lay in the morphology. It is suggested that this organism be adopted as a new species of Thermus and named Thermus filiformis, to reflect its filamentous nature. The aminopeptidase test was negative for T. ruber, T. aquaticus and “T. thermophilus” which is characteristic of Gram type positive organisms, while the KOH lysis test was positive, for these organisms, which is characteristic of Gram type negative organisms. This evidence adds to the contradictory literature derived data on the genus, which although staining Gram-negatively shows many Gram-positive characteristics. It is suggested that these anomalies may represent adaptations to life at high temperatures or may reflect the divergent evolutionary line of descent of the genus.