Diversification of bacterial community composition along a temperature gradient at a thermal spring

Abstract

To better understand the biogeography and relationship between temperature and community structure within microbial mats, the bacterial diversity of mats at a slightly alkaline, sulfide-containing hot spring was explored. Microbial mats that developed at temperatures between 75-52°C were collected from an area of approximately 1 m(2) in Nakabusa, Nagano, Japan. Bacterial 16S rRNA genes from these samples were examined by terminal restriction fragment length polymorphism (T-RFLP) and clone library analysis. T-RFLP profiles revealed 66 unique fragments (T-RFs). Based on total T-RFs observed in environmental profiles and clone libraries, a temperature effect on diversity was determined, with complexity in the community increasing as temperature decreased. The T-RF pattern indicated four distinct community assemblages related to temperature. Members of the Aquificales and particularly the sulfur-oxidizing bacterium Sulfurihydrogenibium were present at all temperatures and were the dominant component of mats taken at 75-67°C. Sulfide oxidation, which persisted throughout the temperature gradient, was the presumed dominant pathway of primary production above 67°C. As temperature decreased, successive additions of anoxygenic and oxygenic phototrophs increased primary productivity, allowing for diversification of the community.

Fig. 1

Regression between temperature and total number of T-RFs for each sample collected. T-RF signals were normalized by sample to the lowest sample value to correct for variations in loading between samples and for an unbiased estimate of richness. Temperature has a significant effect on the number of T-RFs observed (R²=0.6276, t=6.97 (17); P<0.0001). Raw values also showed a similar relationship, but with a lower R² value (0.39).

Fig. 2

UPGMA cluster analysis of T-RF patterns for all samples analyzed. Sample number indicates temperature (°C).

Fig. 3

Maximum likelihood phylogenetic tree of Sulfurihydrogenibium spp. based on partial 16S rRNA gene sequences. Clones from this study are in bold. Genbank accession numbers follow each sequence name. Bootstrap support >50 for ML and Bayesian posterior probabilities >0.7 are given at the nodes supported by both measures. Hydrogenothermus marinus and Persephonella sp. EPR486 were used as outgroup taxa.