A changing of the guard


*E-mail phugenholtz@lbl.gov; Tel. (+1) 925 296 5725; Fax (+1) 925 296 5720.

After 30 articles beginning in May 2004, Michael Galperin is handing over the reins on the genomics update editorials. I have enjoyed Michael's regular updates and commentaries on microbial genomics over the past four and a half years and hope to provide similarly interesting articles for the Environmental Microbiology readership. I have enlisted the help of Nikos Kyrpides who has his finger on the pulse of upcoming microbial genome projects through the Genomes OnLine Database (GOLD; Liolios et al., 2008). We will take a slightly different tack for this news feature by spotlighting individual bacterial or archaeal phyla (divisions) in the context of recently sequenced genomes (or at least that is the plan). We will also intersperse these articles with solicited commentaries from other researchers with something to say on microbial genomics or metagenomics.1 Finally, since Environmental Microbiology is becoming a rather thick journal, we will do our part by keeping our editorials short and not overly frequent.

There are now more than 1500 publicly available draft or complete microbial genomes (Table 1) and many more slated for sequencing. Some would argue that these are more than enough data to get a handle on the microbial world (Whitworth, 2008) but, in truth, we have barely made a dent on our genomic exploration of microbial diversity. The staggering number of microorganisms in our biosphere in and of itself should tell us this, but the point is really driven home from a phylogenetic perspective. Most genomes (80%) still come from only three bacterial phyla, the Proteobacteria, Firmicutes and Actinobacteria, because these are the most readily culturable groups (Table 1). However, we now have draft and/or complete genome sequences in hand for 24 bacterial phyla; 26 if you include the single unreleased genome representatives of the Fibrobacteres (Fibrobacter succinogenes, GOLD id Gi00251) and Gemmatimonadetes (see below). And many of these have a higher percentage of sequenced isolates than the top three bacterial phyla; for example, ∼16% of all chlamydial isolates have been sequenced (Table 1). Phylogenetically conspicuous lineages found repeatedly in environmental surveys but with only a few isolated representatives are particularly attractive targets for genome sequencing. A case in point is the Gemmatimonadetes that comprises over 500 environmental clone sequences (> 1200 nt; DeSantis et al., 2006), but only five isolated representatives: Gemmatimonas aurantiaca after which the phylum is named (Zhang et al., 2003) and four strains isolated from the same soil (Joseph et al., 2003; Davis et al., 2005). Gemmatimonas aurantiaca has been the subject of a sequencing project (Gi01301) by NITE in Japan and its 4.6 Mb genome is complete and submitted for publication (S. Hanada, pers. comm.). A sequencing project (Gi03757) for one of the soil strains, Ellin5290, has just been initiated at the JGI as part of a pilot project to target genomic gaps in the bacterial and archaeal domains. There is still a lot of ground to cover though; G. aurantiaca and Ellin5290 represent only one of at least four main lines of descent in the Gemmatimonadetes (Zhang et al., 2003). Other notable isolates are the first cultured representatives of candidate phyla OP10 (Stott et al., 2008) and OP5 (Mori et al., 2008), both from geothermal habitats. One of the two OP10 isolates (strain T49) is scheduled to be sequenced at the University of Hawaii beginning in February 2009 (M. Stott and P. Dunfield, pers. comm.) and sequencing has already begun on the OP5 isolate at NITE where it was isolated (K. Mori, pers. comm.).

Table 1.  Estimates of number of isolates and genome sequences for each phylum in the bacterial and archaeal domains as of December 2008.
Phylum-level lineageaNumber of isolates (%)bNumber of genomes (%)c% of isolates sequencedCandidatus genomesc
  • a. 

    Candidate phylum TM7 is not included in this table as published candidatus TM7 genomes (Marcy et al., 2007; Podar et al., 2007) are not draft-level. An additional 30 bacterial and 5 archaeal candidate phyla, each comprising at least 20 environmental clone sequences, also are not included in this list as they presently have neither isolates nor candidatus genome sequences.

  • b. 

    With 16S rRNA gene sequences > 1200 nt long; data derived from greengenes283274.arb (DeSantis et al., 2006) using a search limited to records with ‘isolate’ in the sequence_type field and ‘clone’ absent in the full_name field.

  • c. 

    Publicly available (with the exceptions marked1) draft plus complete genomes from cultured isolates or uncultured ‘Candidatus’ species; data source GOLD (Liolios et al., 2008).

 Proteobacteria31 597 (46.7)727 (47.6)2.310
 Firmicutes15 059 (22.2)364 (23.8)2.45
 Actinobacteria12 460 (18.4)130 (8.5)1.0 
 Bacteroidetes2 889 (4.3)53 (3.5)1.84
 Cyanobacteria2 274 (3.4)51 (3.3)2.2 
 Spirochaetes964 (1.4)36 (2.4)3.7 
 Chlamydiae122 (0.2)20 (1.3)16.41
 Chlorobi109 (0.2)12 (0.8)11.0 
 Chloroflexi82 (0.1)12 (0.8)14.6 
 Thermotogae97 (0.1)10 (0.7)10.3 
 Verrucomicrobia62 (0.1)8 (0.5)12.9 
 Thermi325 (0.5)6 (0.4)1.8 
 Fusobacteria118 (0.2)6 (0.4)5.1 
 Planctomycetes113 (0.2)6 (0.4)5.31
 Aquificae90 (0.1)6 (0.4)6.7 
 Synergistetes42 (0.1)3 (0.2)7.1 
 Acidobacteria25 (< 0.1)2 (0.1)8.01
 Deferribacteres23 (< 0.1)2 (0.1)8.7 
 Dictyoglomi4 (< 0.1)2 (0.1)50.0 
 Lentisphaerae2 (< 0.1)2 (0.1)100.0 
 Nitrospirae62 (0.1)1 (< 0.1)1.6 
 Thermodesulfobacteria7 (< 0.1)1 (< 0.1)14.3 
 Chrysiogenetes3 (< 0.1)1 (< 0.1)33.3 
 Elusimicrobia (TG1)1 (< 0.1)1 (< 0.1)100.01
 Fibrobacteres56 (0.1)1 (< 0.1)11.8 
 Gemmatimonadetes5 (< 0.1)1 (< 0.1)120.0 
 OP102 (< 0.1)   
 Caldithrix1 (< 0.1)   
 OP51 (< 0.1)   
 WWE1   1
 Euryarchaeota1 000 (1.5)47 (3.1)4.72
 Crenarchaeota136 (0.2)16 (1.0)11.8 
 Thaumarchaeota2 (< 0.1)1 (< 0.1)50.01
 Nanoarchaeota1 (< 0.1)1 (< 0.1)100.0 
 Korarchaeota   1
Total67 734 (100.0)1528 (100.0)2.328

Cultured microorganisms are great to have for many reasons, but to really accelerate the genomic mapping of the tree of life we cannot wait for fortuitous isolation of novel organisms – we need to target and sequence populations directly from the environment. This has already started, in a mostly untargeted fashion, and in fact there are almost 30 complete or near-complete genomes from uncultured ‘Candidatus’ species representing 11 phylum-level lineages (Table 1). Importantly, these include the first genome sequences of phyla that completely lack cultured representatives; Candidatus Cloacamonas acidaminovorans (Gc00715) from the bacterial candidate phylum WWE1 (Pelletier et al., 2008) and Candidatus Korarchaeum cryptofilum (Gc00747) from the archaeal candidate phylum Korarchaeota (Elkins et al., 2008). Mostly these candidatus genomes have been derived from dominant populations in simple communities or enrichments, but a promising step forward is the specific targeting of small numbers of cells that are then randomly amplified to provide enough DNA for genome sequencing. This has the benefit of limiting the genomic heterogeneity that may exist in the targeted population (by using fewer cells from a smaller area) thereby improving the likelihood of complete recovery of a candidatus genome, especially when coupled with next generation sequencing technolo gies. An elegant example of this approach is the recent determination of a complete genome sequence for Candidatus strain Rs-D17 (Gc00763) obtained by amplifying ∼1000 Rs-D17 cells from a single protist host in a termite hindgut (Hongoh et al., 2008). This is only the second genome representative of the bacterial phylum Elusimicrobia (formerly Termite Group I, A. Brune, pers. comm.; Table 1). With these technical developments and a greater appreciation of the need to systematically map the tree of life, it will be interesting to see when the number of (draft or complete) genomes from uncultured organisms exceeds those from isolates. It may be sooner than we think.


  • 1

    Please contact the corresponding author if you have a topic that you would like to bring up in this forum.