Within the skin: Grape berries during the mature stages of ripening
AuthorGhan, Ryan Marcus
StatisticsView Usage Statistics
A systems biology approach was used to investigate berry skins of three red- (Cabernet Sauvignon, Merlot, Pinot Noir) and two white-skinned (Chardonnay, Semillon) wine grape cultivars. Identical sample aliquots were analyzed for transcripts by a grapevine whole genome oligonucleotide microarray and RNAseq technologies, proteins by nano-liquid chromatography-mass spectroscopy, and metabolites by gas chromatography-mass spectroscopy and liquid chromatography-mass spectroscopy. Principal components analysis of each of five Omic technologies predicted similar variance between cultivars. Comparison of RNAseq and microarray data revealed a strong Pearson’s correlation (0.80), but concordance of protein with transcript data was low with a Pearson’s correlation of 0.27 and 0.24 for the RNAseq and microarray data, respectively. Integration of metabolite with protein and transcript data produced an expected model of phenylpropanoid biosynthesis, distinguishing red from white grapes, yet, provided detail of individual cultivar differences. The integration of multiple high-throughput Omic datasets revealed complex biochemical variation amongst five cultivars of an ancient and economically important crop species. Grape berry ripening occurs in the late stages of development with increases in sugar, changes in color, and decreases in malate concentration. In the final stages of ripening, fruit flavors and volatile aromas increase to signal readiness for seed dispersal. To identify the common transcriptional changes in the late stages of berry development in multiple grape cultivars, the transcriptomic responses of the berry skins of 7 cultivars of grapes that were grown in the same vineyard were determined using RNAseq at four different °Brix levels (20 to 26 °Brix). The abundance of thousands of transcripts changed significantly in the late stages of berry development. Gene set enrichment analysis of functional Gene Ontology terms provided evidence for a complex interplay of many gene ontology categories including those involved in the circadian clock, postembryonic development, photosynthesis, hormone signaling, reactive oxygen species (ROS), DNA methylation and transcriptional regulation. There were 809 transcription factors (TF) differentially expressed with increasing ˚Brix (~4% of all transcripts and ~32% of all TF), belonging to 81 families, including the C3H, MYB, AP2/ERF and bHLH families. Our analyses indicate that the circadian clock and epigenetic modification are major factors regulating transcription in mature berries.Finally, pathogenesis-related proteins that accumulated in skins of three red-skinned and two white-skinned cultivars: Cabernet Sauvignon, Merlot, Pinot Noir, Chardonnay and Semillon, were characterized in silico, using protein and transcript data. Large amounts of identified proteins were classified as pathogenesis-related in berry skins, more so than what was previously observed in shoot tips. Several PR-families had numerous protein members in skins, which maybe a tissue specific occurrence. The transcript abundance was well correlated to the protein abundance in thaumatins of PR-05, but not so in the L-ascorbate peroxidases of PR-09. Haze-forming proteins, while well represented, did not accumulate with more specificity in the white cultivars and were mostly higher in the red cultivar, Pinot Noir. Large accumulations of PR-proteins in skins at harvest provide support for a prolonged and possibly a constitutive defense mechanism that protects a maturing seed within the berry.