Maths on microbes: adding microbial ecophysiology to metagenomics

Microbial ecology has shifted in the last 30 years from a field strongly depending on culturing and microscopy to a discipline dominated by cultivation-independent molecular analysis. Nowadays, metagenomics (for simplicity, I include in metagenomics also high-throughput sequencing of phylogenetic and functional marker genes) allows unprecedented insight into the composition and functional potential of microbial communities, while metatranscriptomics, metaproteomics and metabolomics inform on the expression of functional potential. Yet it is easy to lose sight of the forest for the trees: certainly for engineered systems, like waste water treatments and bioremediation, one is in the end more interested in what all species together do than what their individual identities and activities are. While enormous amounts of metagenomic data are generated, these data are currently heavily underexplored. First, for many genes we currently do not know what they encode for. More important, the data are generally explored by ‘simple’ multivariate analysis on gene lists. To illustrate this point, take oil-polluted marine sediments in which hydrocarbon oxidation occurs by sulfate reducers and aerobic microorganisms: standard data analysis will give reads of genes belonging to these two different functional groups equal weight even though the sulfate reducers had to degrade about 10 times more hydrocarbon to produce one new cell than aerobic microorganisms. I foresee that by incorporating microbial ecophysiology into metagenomic data analysis, we will gain much better insight into microbial communities and their functioning. Many of the components to achieve this goal are already available and will be synergetic when combined. A key component is the concept of ‘reverse ecology’: the evolution and ecology of a species are anchored in its genome, and by large-scale comparison of the ecophysiological properties of species to their genomes and genome-derived metabolic networks we can move forward to linking metagenomic data to community functioning (Röling and Van Bodegom, 2014). ‘Reverse ecology’ uses systems biology modelling of metabolic networks, generally of species that have been isolated and characterized in monoculture (Levy and Borenstein, 2012). Availability of monocultures is an important drawback still to microbial ecology; we only have access to a relatively small proportion of microbial species found in nature, heavily biased to a few phyla. Thus, more emphasis is needed on physiological characterization of notyet-cultured species. Cultivation-independent single-cell approaches are rapidly advancing, and single prokaryotic cell transcriptomics is coming into view (Kang et al., 2011) and will contribute to revealing in situ physiology. Yet, to obtain more extensive physiological information (e.g. growth rates, growth yields, substrate uptake rates and affinities) for use in ‘reverse ecology’, culturing still appears to be the most effective approach for the near future. In fact, cultivation-independent single-cell analysis may also assist here: metabolic network reconstruction on the basis of single-cell genomic data, overlaid with transcriptomic data, will inform how the microorganism of interest connects to its environment, e.g. which nutrients it may need to take up from its environment, and contribute to the design of isolation strategies. Thus, microbial ecologists need in part to return to the situation before the advent of molecular microbial ecology, and focus more on culturing. However, enrichment and isolation strategies were in the past, and still are, often ill-considered in light of the environmental context of the microorganism of interest. The favourite method was batch cultivation in liquid medium with unrealistic high concentrations of nutrients, which hardly mimics the environment most microorganisms live in: microbes tend to grow attached, at very low growth rates with low concentrations of nutrients. Nowadays, high-throughput microculturing under environmental relevant conditions is possible. An approach like the micro-Petri dish (Ingham et al., 2007) can be combined with gradients in nutrients and slow diffusion of substrates, enabling slow growth after dilution-to-extinction of inocula, and allowing for subsequent screening of micro-cultures by microscopy and (meta)genomics. Received 29 September, 2014; accepted 7 October, 2014. *For correspondence. E-mail [email protected]; Tel. +31 20 5987192; Fax +31 20 5987229. Microbial Biotechnology (2015) 8(1), 21–22 doi:10.1111/1751-7915.12233 Funding Information No funding information provided. bs_bs_banner