The transcription and translation signals of the S-layer gene (slpA) from Thermus thermophilus HB8 have been used to express a thermostable kanamycin adenyl transferase gene in this organism. The chimaeric resistance gene was inserted in vitro into slpA to produce different inactive forms of the gene, which were used to transform T. thermophilus HB8. After 48 hours of incubation at 70 degrees C, only two constructions that contained the kat gene flanked by Thermus sequences from both sides of slpA were able to produce protein layer (P100)-defective mutants. The mutants obtained with both constructions showed identical protein patterns, in which a major 50 kDa protein and two other minor proteins were tentatively identified as P100 fragments, expressed from the extreme 5' end of slpA. They also exhibited important phenotypic defects, such as slow growth in liquid broth, a tendency to aggregate as 'rotund bodies', a twisted filamentous shape, and an extreme sensitivity to lysozyme, suggesting protective and shaping roles for the S-layer in T. thermophilus HB8. These results also demonstrate for the first time the feasibility of using selective antibiotic-resistance markers in extreme thermophiles.

No abstract

Despite the fact that the extreme thermophilic bacteria belonging to the genus Thermus are classified as strict aerobes, we have shown that Thermus thermophilus HB8 (ATCC 27634) can grow anaerobically when nitrate is present in the growth medium. This strain-specific property is encoded by a respiratory nitrate reductase gene cluster (nar) whose expression is induced by anoxia and nitrate (S. Ramı́rez-Arcos, L. A. Fernández-Herrero, and J. Berenguer, Biochim. Biophys. Acta, 1396:215–1997). We show here that this nar operon can be transferred by conjugation to an aerobic Thermus strain, enabling it to grow under anaerobic conditions. We show that this transfer takes place through a DNase-insensitive mechanism which, as for the Hfr (high frequency of recombination) derivatives ofEscherichia coli, can also mobilize other chromosomal markers in a time-dependent way. Three lines of evidence are presented to support a genetic linkage between nar and a conjugative plasmid integrated into the chromosome. First, the naroperon is absent from a plasmid-free derivative and from a closely related strain. Second, we have identified an origin for autonomous replication (oriV) overlapping the last gene of thenar cluster. Finally, the mating time required for the transfer of the nar operon is in good agreement with the time expected if the transfer origin (oriT) were located nearby and downstream of nar.