We have developed a new way to isolate methanogens, by using syntrophs that can supply H2
at very low concentrations (coculture method). As expected, the RC-I methanogen could be cultivated by the coculture method whereas conventional enrichments from the same environmental samples with high levels of H2
yielded fast-growing methanogens such as Methanobacterium
spp. Assuming that strain SANAE is representative of the members of the RC-I group, our results suggest that the RC-I group has a higher affinity for H2
than other, fast-growing, members and thus may be well adapted to the low-H2
habitat of RPFs. This speculation was also suggested by two reports on stable-isotope-probing analysis (23
). Lu et al. reported that Methanobacteriales
C when rice root was incubated in a high-H2
atmosphere in the presence of 13
, while 13
C was preferentially incorporated into RC-I under an N2
atmosphere in which a low concentration of H2
was produced by fermentative bacteria from rice root materials (23
). According to the stable-isotope-probing analysis with 13
C-labeled propionate for RPF soil reported by Lueders et al., 13
C was incorporated not only into rRNA of syntrophic propionate-oxidizing bacteria but also into certain methanogenic archaeal groups, including RC-I (26
). These results imply that RC-I methanogens probably live syntrophically with propionate-oxidizing bacteria in RPF environments. Consequently, given that the members of RC-I may have a high affinity for H2
and live in syntrophy with fermentative bacteria such as syntrophic propionate oxidizers, it can be said that our coculture method is a suitable way to cultivate RC-I methanogens.
Very recently, a complete metagenomic sequence of an RC-I member was reported (9
). The sequence was retrieved from a paddy sample incubated at 50°C, indicating that the organism is most likely a thermophilic methanogen. Since the temperature of natural RPFs is normally not stable, it is reasonable that a variety of sequences within the RC-I cluster, which may represent either mesophilic or thermophilic members, have been detected from RPFs. That metagenomic study revealed that the thermophile has a unique set of genes encoding antioxidant enzymes such as catalase and superoxide dismutase. Considering that RPFs receive a seasonal change in redox potential (oxic to anoxic), these unique genes may be indispensable for survival in such an environment.
The 16S rRNA gene sequence similarity between the metagenome and strain SANAE is, however, only 92.0% (Fig. , RC-I fosmid 1B6mer2), indicating that the two organisms are phylogenetically distinct at the genus level. Therefore, further studies cross-linking both the metagenomic information of thermophilic RC-I and our mesophilic isolate SANAE are needed to better understand how RC-I methanogens contribute to global methane emission from RPF environments. There are several important questions that have to be addressed to fully understand the role of RC-I methanogens. (i) Can thermophilic RC-I methanogens also grow slowly with a low concentration of H2? (ii) What is the biochemical or physiological background of the low maximum specific growth rate and the assumed high affinity for hydrogen of RC-I and similar organisms? (iii) What is the difference in enzymatic characteristics, in particular, antioxidant enzymes, between the two different RC-I methanogens? To address these questions, we are now performing further biochemical (including examination of affinity for H2 in terms of Km) and genomic analyses of our isolate. We are also trying to isolate the thermophilic RC-I methanogen whose genomic sequence is complete.
Since the coculture method can mimic the natural ecosystem, where H2 is provided at low concentrations, the method was supposed to have potential for the cultivation of uncharacterized methanogens residing in other ecosystems. To prove this, we also applied the cultivation technique to a methanogenic digester that contains a relatively large amount of uncultured archaeal members within the order Methanomicrobiales. By strategy described above, we successfully isolated a new methanogen, designated strain NOBI-1 (16S rRNA gene sequence accession no. AB162774), which was one of the predominant methanogens in the original ecosystem but was not cultivated so far at the genus or family level. In this case, all of the primary enrichment cultures using a high partial pressure of H2 with the same inoculum resulted in the cultivation of well-known methanogens like those of the genus Methanobacterium. Detailed information about the enrichment, isolation, and physiological properties of the isolate will be reported in the near future. These findings strongly indicate that the coculture method should work for isolation of fastidious but ecologically important methanogens that would otherwise escape conventional isolation strategies.