In our study, we explored the feasibility of using HTS to perform metatranscriptomics and metagenomics on an experimental cheese microbial community throughout ripening. Our methodological strategy is based on the establishment of a reference database of all the genomes of the community, onto which RNAseq data can be mapped. The choice of the cheese community is guided by a low level of microbial complexity (≈10 species) which is sufficient to reproduce the complex metabolic pattern of cheese maturation (Bonaïti et al., 2005; Mounier et al., 2008). Furthermore, features (eg: aroma production, color, microbial growth, substrates uptake) generated by this microbial community are measurable in a reduced time scale (3 - 10 weeks). Finally, most of these microorganisms are cultivable with sequenced genomes available. A first objective of our investigation centers on developing and optimizing experimental, analytical and statistical pipelines designed for meta-transcriptomics, genomics and physiological data analyses. By applying a joint metagenomic, metatranscriptomic and physiological approach, a second objective aimed to give a global view of this very peculiar microbial community being firstly regarded as a whole (= meta-organism) and, secondly, down to the species and the gene level. To address these aims, we also considered the cheese microbial community in a dynamic manner, samplings being made at different times of the ripening process. Our data show that DNA datasets did not parallel RNA datasets. Furthermore, although transcriptional activity was essentially brought by yeasts – and overrepresented by Geotrichum candidum, the relative importance of surface-ripening bacteria - in terms of DNA, RNA and cells counts - tends to increase with time.
Less...