Microalgae are an outstanding source of high value products with applications in food, feed or biofuel production. Among these promising organisms, the haptophyte Tisochrysis lutea is commonly used as a feed for shellfish and shrimps in aquaculture. Its capacity to produce high amounts of docosahexanoic acid and storage lipids is also of economic interest for nutrition and energy production. Consequently, understanding its lipid accumulation under nitrogen deprivation is of major interest.

Here, we aimed to identify Transcription Factors (TFs) involved in the establishment of the mutant phenotype of the 2Xc1 strain of T. lutea, which accumulates double the quantity of storage lipids under nitrogen deprivation than the wild type strain (WTc1). Strains were grown in chemostats and subjected to different nitrogen availability (limitation, repletion and depletion). Using RNA-seq data, the differentially expressed genes (DEGs) between strains were identified and summarized as a co-expression network to pinpoint putative major TFs in mutant phenotypes. This analysis was followed by a complementary Weighted Gene Correlation Network Analysis in order to classify genes based on their relative importance to mutant phenotype features, regardless of annotation biases due to the lack of functional annotation of the Tisochrysis lutea draft genome. This network-like strategy allowed the identification of seven TF candidates related to key functions in the mutant strain compared with WTc1. In particular, MYB-2R_14 and NF-YB_2 TFs are related to photosynthesis, oxidative stress response and triacylglycerol synthesis. GATA_2, MYB-rel_11 and MYB-2R_20 TFs are likely to be related to nitrogen uptake or carbon and nitrogen recycling, feeding carbohydrate synthesis in the form of chrysolaminarin. Finally, a q-RT-PCR approach further characterized the role of MYB-rel_11 and MYB-2R_20, revealing an expression pattern dependent on nitrogen availability.