A systems biology approach for identifying candidate genes involved in the natural variability of biomass yield and chemical properties in black poplar

Lignocellulosic biomass is a renewable resource of interest for biorefinery. However, current poplarvarieties have not been selected for this specific purpose. The factors affecting biomass yield andchemical properties need thus to be studied. With this objective, we have initiated a systems biologyapproach, integrating genomic, transcriptomic and phenotypic data in natural populations of blackpoplar (Populus nigra). Up to now, we have focused on a subset of 12 genotypes from 6 populationsand trialled in a randomized complete block design located at INRA Orléans, France. Thetranscriptome of 2 biological replicates of each genotype has been explored through RNA sequencing(RNAseq) of pools of young differentiating xylem and cambium. Additionally, biomass yield wasevaluated through measurements of height and diameter on 6 replicates of each genotype acrossseveral years and rotations, while biomass propertieswere assessed through chemical analyses oflignin, cellulose and hemicellulose concentrations as well as saccharification potential on 3 replicatesof each genotype. The resulting data were used to build a weighted gene co-expression network andidentify gene modules whose expression was correlated with biomass yield and/or quality at thegenotypic level. Remarkably, the largest module (1,460 transcripts) was significantly associated withklason lignin content and displayed an enrichment in genes involved in secondary cell wall formation.Four candidate genes from this module were further selected to validate the detected quantitative traittranscripts (QTTs) on 2 new replicates of the 12 genotypes using RT-qPCR. The resulting expressionlevels were significantly correlated to those previously quantified by RNAseq and to the klason lignincontent in the wood samples. These results demonstrate the interest of our approach, and thus opensome prospects towards the identification of new candidate genes whose functions remain to beelucidated