Closed‐reference metatranscriptomics enables <i>in planta</i> profiling of putative virulence activities in the grapevine trunk disease complex

Grapevines, like other perennial crops, are affected by so-called 'trunk diseases', which damage the trunk and other woody tissues. Mature grapevines typically contract more than one trunk disease and often multiple grapevine trunk pathogens (GTPs) are recovered from infected tissues. The co-existence of different GTP species in complex and dynamic microbial communities complicates the study of the molecular mechanisms underlying disease development, especially under vineyard conditions. The objective of this study was to develop and optimize a community-level transcriptomics (i.e. metatranscriptomics) approach that could monitor simultaneously the virulence activities of multiple GTPs in planta. The availability of annotated genomes for the most relevant co-infecting GTPs in diseased grapevine wood provided the unprecedented opportunity to generate a multi-species reference for the mapping and quantification of DNA and RNA sequencing reads. We first evaluated popular sequence read mappers using permutations of multiple simulated datasets. Alignment parameters of the selected mapper were optimized to increase the specificity and sensitivity for its application to metagenomics and metatranscriptomics analyses. Initial testing on grapevine wood experimentally inoculated with individual GTPs confirmed the validity of the method. Using naturally infected field samples expressing a variety of trunk disease symptoms, we show that our approach provides quantitative assessments of species composition, as well as genome-wide transcriptional profiling of potential virulence factors, namely cell wall degradation, secondary metabolism and nutrient uptake for all co-infecting GTPs.