At a time when metagenomic studies are capturing the limelight, why is culturing important? In short, cultures are important because they provide complete genomes and the means to test the hypotheses that emerge from genomic data. Cultured bacterial and archaeal isolates that have yielded complete genome sequences have also proved to be important for evaluating metagenomic data — a recent paper that presented environ- mental whole-genome shotgun-sequencing data from Craig Venter’s Global Ocean Survey relied heavily on complete genome sequences from bacterial plankton to evaluate metagenomic data, because procedures for reconstructing genomes from whole-genome shotgun sequences are not yet reliable 1,2 . Cultures, supported by predictions from genomes, together with the information about natural variation that is provided by metagenomics are a powerful combination. On the one hand, live cells allow the study of the whole organism, rather than trying to infer physiological properties from an incom- plete list of parts. On the other hand, genomic and metagenomic data can reveal hidden metabolic potential, metabolic pathways, regulatory circuits and conservation among cultured and uncultured bacteria (FIG. 1). The TIMELINE lists selected important events in marine microbial cultivation. Historically, there has been debate about whether heterotrophic microbial cells can grow at the extremely low nutrient concentra- tions that are found in natural ecosystems 3 . However, now there are many strains that have been cultured which replicate well in sterilized seawater, but do not grow in media that contain concentrated organic nutrients (BOX 1). The main challenge of designing artificial media to mimic natural seawater is the accurate reconstruction of the complex composition of dissolved organic carbon (DOC) and trace elements (TABLE 1). To easily obtain dense, turbid bacterial or archaeal cul- tures, most researchers have relied on media that contain high concentrations of organic carbon. ZoBell’s Marine Medium 4 , which is still widely used for culturing heterotrophic bacteria, has 170-fold more DOC than natu- ral seawater. To make matters worse, most of the DOC in seawater is derived from recalci- trant matter, so that the difference in useable organic carbon probably exceeds three orders of magnitude (TABLE 1). The first culture of an obligate oligotroph was obtained from a freshwater sample in 1981 (REF. 5). In the same year, an influential review of oligotrophy was published 6 . Subsequently, several research groups, most notably that of Don Button, developed methods for the successful cultivation of oligotrophs and began working with cultures of marine bacteria that could be grown in autoclaved seawater 7–9 . The laboratory of S.G. modified Button’s methods by redu- cing the size of cultures to microtitre dish volumes, implementing screening methods and adopting clean procedures that had been developed by oceanographers for han- dling seawater 10 . Additional modifications included segregating containers that were used for culturing cells, avoiding detergents and using plastics such as polycarbonate and Teflon that do not inhibit the growth of bacterial cultures. These methods, which have been referred to as high-throughput extinction-culturing methods, led to the isolation into pure culture of the most important new heterotrophic marine strains that have been cultured so far. Other approaches to cultivation have used microencapsulation techniques 11 , chambers that allow the exchange of small molecules with the environment 12 , floating filters 13,14 , micromanipulation 15 or, simply, low-nutrient agar plates 16 to cultivate novel microorgan- isms. Although the details of these approaches differ, they are all based on the same princi- ples of re-creating the low-nutrient conditions of natural environments and overcoming the tendency of rapidly growing cells to over- whelm species that divide less often. These approaches have promise, as do traditional enrichment strategies that target organisms with unusual metabolisms 17 . Although genomic information has not been a major factor in the isolation of new marine organ- isms, it has been exploited to design isolation strategies for bacteria that are present in diff- erent environments 18,19 , and it is only a matter of time before genomic or metagenomic information contributes to the isolation of a marine bacterial or archaeal species. Strains that represent many of the most abundant marine clades have been isolated into pure culture in recent years, but other important clades still lack any cultured representatives 20,21 (FIG. 2). Some of the uncul- tured clades originate from phyla that are associated with unusual metabolisms, which suggests that their cultivation could depend on successfully determining their ecological niche. For example, the recent cultivation of Nitrosopumilus maritimus SCM-1, which was the first cultured representative of the Group I Marine Crenarchaeota, succeeded because of a specific strategy that excluded bacteria with antibiotics and aimed to isolate archaeal species that can oxidize ammonium to yield energy 22 . The SAR202 clade, which lacks cultured representatives, belongs to the phylum Chloroflexi, which includes photo- heterotrophs and anaerobes that have the capacity to oxidize halogenated compounds. The SAR202 clade is one of several that have been relegated to regions that are just below the euphotic zone 23 . Another important group that lacks cultured representatives is the SAR86 group of gammaproteobacteria, which account for approximately 10% of the total prokaryotic microbial community in the ocean surface layer 24 . The discovery of the light-driven, proton pump proteo- rhodopsin in the fragmentary genome data OPINION The importance of culturing bacterioplankton in the ‘omics’ age Stephen Giovannoni and Ulrich Stingl Abstract | Progress in the culturing of microorganisms that are important to ocean ecology has recently accelerated, and technology has been a factor in these advances. However, rather than a single technological breakthrough, a combination of methods now enable microbiologists to screen large numbers of cultures and manipulate cells that are growing at the low biomass densities that are characteristic of those found in seawater. The value of ribosomal RNA databases has been reaffirmed, as they provide nucleic-acid probes for screening to identify important new species in culture. The new cultivation approaches have focused on specific targets that ecological studies suggest are significant for geochemical transformations, such as SAR11. Here, we review how to cultivate marine oligotrophs and why it is worth the effort. PERSPECTIVES 820 | OCTOBER 2007 | VOLUME 5 www.nature.com/reviews/micro PERSPECTIVES © 2007 Nature Publishing Group