Stevia rebaudiana (Bertoni) is a highly valuable crop for the steviol glycoside content in its leaves, which are no-calorie sweeteners hundreds of times more potent than sucrose. The presence of health-promoting phenolic compounds, particularly flavonoids, in the leaf of S. rebaudiana adds further nutritional value to this crop. Although all these secondary metabolites are highly desirable in S. rebaudiana leaves, the genes regulating the biosynthesis of phenolic compounds and the shared gene network between the regulation of biosynthesis of steviol glycosides and phenolic compounds still need to be investigated in this species. To identify putative candidate genes involved in the synergistic regulation of steviol glycosides and phenolic compounds, four genotypes with different contents of these compounds were selected for a pairwise comparison RNA-seq analysis, yielding 1136 differentially expressed genes. Genes that highly correlate with both steviol glycosides and phenolic compound accumulation in the four genotypes of S. rebaudiana were identified using the weighted gene co-expression network analysis. The presence of UDP-glycosyltransferases 76G1, 76H1, 85C1, and 91A1, and several genes associated with the phenylpropanoid pathway, including peroxidase, caffeoyl-CoA O-methyltransferase, and malonyl-coenzyme A:anthocyanin 3-O-glucoside-6″-O-malonyltransferase, along with 21 transcription factors like SCL3, WRK11, and MYB111, implied an extensive and synergistic regulatory network involved in enhancing the production of such compounds in S. rebaudiana leaves. In conclusion, this work identified a variety of putative candidate genes involved in the biosynthesis and regulation of particular steviol glycosides and phenolic compounds that will be useful in gene editing strategies for increasing and steering the production of such compounds in S. rebaudiana as well as in other species.

Much has been said about sunflower (Helianthus annuus L.) retrotransposons, representing the majority of the sunflower’s repetitive component. By contrast, class II transposons remained poorly described within this species, as they present low sequence conservation and are mostly lacking coding domains, making the identification and characterization of these transposable elements difficult. The transposable element Tetu1, is a non-autonomous CACTA-like element that has been detected in the coding region of a CYCLOIDEA (CYC) gene of a sunflower mutant, tubular ray flower (turf). Based on our knowledge of Tetu1, the publicly available genome of sunflower was fully scanned. A combination of bioinformatics analyses led to the discovery of 707 putative CACTA sequences: 84 elements with complete ends and 623 truncated elements. A detailed characterization of the identified elements allowed further classification into three subgroups of 347 elements on the base of their terminal repeat sequences. Only 39 encode a protein similar to known transposases (TPase), with 10 TPase sequences showing signals of activation. Finally, an analysis of the proximity of CACTA transposons to sunflower genes showed that the majority of CACTA elements are close to the nearest gene, whereas a relevant fraction resides within gene-encoding sequences, likely interfering with sunflower genome functionality and organization.

Genome divergence by repeat proliferation and/or loss is a process that plays a crucial role in species evolution. Nevertheless, knowledge of the variability related to repeat proliferation among species of the same family is still limited. Considering the importance of the Asteraceae family, here we present a first contribution towards the metarepeatome of five Asteraceae species. A comprehensive picture of the repetitive components of all genomes was obtained by genome skimming with Illumina sequence reads and by analyzing a pool of full-length long terminal repeat retrotransposons (LTR-REs). Genome skimming allowed us to estimate the abundance and variability of repetitive components. The structure of the metagenome of the selected species was composed of 67% repetitive sequences, of which LTR-REs represented the bulk of annotated clusters. The species essentially shared ribosomal DNA sequences, whereas the other classes of repetitive DNA were highly variable among species. The pool of full-length LTR-REs was retrieved from all the species and their age of insertion was established, showing several lineage-specific proliferation peaks over the last 15-million years. Overall, a large variability of repeat abundance at superfamily, lineage, and sublineage levels was observed, indicating that repeats within individual genomes followed different evolutionary and temporal dynamics, and that different events of amplification or loss of these sequences may have occurred after species differentiation.