|
|
ChloroflexiThe phylum Chloroflexi is an early diverging clade of eubacteria whose first characterized member, Chloroflexus aurantiacus, was reported by Pierson and Castenholz in 1974 (1). This moderately thermophilic, filamentous anoxygenic phototroph, or FAP, is still the best-characterized member of the Chloroflexi (2). Organisms closely related to C. aurantiacus have been isolated from slightly alkaline hot springs around the world; two of the original isolates were recovered from Octopus Spring in the Lower Geyser Basin in Yellowstone National Park (1, 2). Although these organisms have at times been referred to as “green non-sulfur bacteria,” several members of the phylum can in fact oxidize sulfide as a source of electrons for photosynthesis (e.g., Candidatus Chlorothrix halophila and Oscillochloris trichoides) (2-5). Hanada and others have described “red Chloroflexus” strains, for example Roseiflexus castenholzii, which was originally isolated from the Nakabusa Hot Springs in Japan (6-8). Unlike C. aurantiacus, this organism lacks the ability to synthesize bacteriochlorophyll c and thus does not have chlorosomes as light-harvesting antennae. More recently, a second Roseiflexus sp. strain, RS-1, was isolated from Octopus Spring (9). This Roseiflexus sp. accounts for a substantial proportion of the biomass in the euphotic portion of the microbial mat community.
Until very recently, the only genomic sequence data for members of the Chloroflexi was an incomplete draft sequence (~5.2 Mb) for Chloroflexus aurantiacus strain J10-fl, an isolate from Japan (1, 2) and some complete sequence data for Dehalococcoides spp., which are only very distantly related to FAPs. In collaboration with JGI-DOE, we have now obtained/are now obtaining genomic sequence data for several additional Chloroflexi isolates. Roseiflexus castenholzii (5,723,298 bp) and Roseiflexus sp. strain RS-1 (5,801,598 bp) have been completely sequenced. No plasmids were found in either strain, and only a single circular chromosome was detected. Chloroflexus aggregans, another species of Chloroflexus from Japan (2, 10, 11), is nearly complete and has significantly smaller genome (~4.5 Mb) than other FAPs. The genome of the mesophilic, non-phototrophic Chloroflexi, Herpetosiphon aurantiacus (2, 12), has also been completed (6,346,587 bp; plasmids: 339639 bp and 99,204 bp). Although this organism synthesizes carotenoids, no remnant genes suggestive of chlorophyll biosynthesis or photosynthesis were detected in its genome. However, several genes putatively involved in secondary metabolite production have been found in this genome, and it is possible that some of these will confer the ability to make novel compounds. DNA from an enrichment culture for Candidatus Chlorothrix halophila (3) was also sequenced, but the complexity of this DNA makes the data similar to that obtained for a metagenome. Without substantially greater sequence coverage, it is not clear how much information can be extracted from the current data. Finally, two additional genomes are still in progress. Chloroflexus sp. Y-400-fl and Chloroflexus sp. Y-396-1 are axenic strains that were isolated from Octopus Spring in the Lower Geyser Basin of Yellowstone (1, 2). The Y-400-fl strain is predicted to be more similar to the J-10-fl strain because of its 16S rRNA sequence similarity to that strain, and but the similarities and differences among these strains remains to be determined. It is hoped that these data will provide much better anchor genomes for the analysis of Octopus and Mushroom Springs metagenomic sequences. With respect to carbon fixation, some of these data have already been been analyzed (13). Although C. aurantiacus J-10-fl can be grown photoautotrophically, no Roseiflexus sp. has yet been grown autotrophically under laboratory conditions. Interestingly, the genomes of all of both Roseiflexus sp., C. aggregans, and C. aurantiacus all contain the complete set of genes for the 3-hydroxypropionate pathway for carbon dioxide fixation. None of the genes unique to this pathway are present in the Herpetosiphon aurantiacus genome. Previous studies had shown that Oscillochloris trichoides has Rubisco and can fix carbon by the reductive pentose phosphate pathway (14). The data obtained for Candidatus Chlorothrix halophila strongly suggest that this organism also utilizes the reductive pentose phosphate pathway for carbon dioxide fixation. These observations are of interest, since no other phylum containing chlorophototrophs contains strains with different mechanisms of carbon dioxide fixation. Strikingly, the reaction centers in Candidatus Chlorothrix halophila are not of the type 1 class, but are type 2 reaction centers most similar to those found in Proteobacteria. However, the enzymes of bacteriochlorophyll biosynthesis seem to be more similar to those of Chlorobi and Cyanobacteria. Further characterization of these organisms is clearly warranted because of the mosaic nature of their photosynthetic traits, which imply that there could have been a lateral gene transfer that led to the replacement of ancestral type 1 reaction centers in the Chloroflexi by type 2 reaction centers derived from an ancestor of modern chlorophototrophic Proteobacteria. References:
Support from   
|
