Early and late responses of grapevine (Vitis vinifera L.) to water deficit: a proteomics perspective

Two different osmotic stress experiments using different proteomic methods were compared. In Experiment 1, two osmotic stress treatments of growing shoot tips of 'Cabernet Sauvignon' grapevines were compared over 16 days; one was a gradually-applied, long-term water deficit and the second was an equivalent salinity stress. Proteins were analyzed by two-dimensional polyacrylamide gel electrophoresis (2-D PAGE). In Experiment 2, a water deficit treatment was applied; protein extracts were analyzed using nanoflow liquid chromatography - tandem mass spectrometry (nanoLC-MS/MS). In both experiments, osmotic stress progressively reduced growth with time. In Experiment 1, 82 out of 645 (13%) proteins had significant changes in abundance in response to stress over time. and were identified by matrix-assisted laser desorption ionization time of flight tandem (MALDI TOF/TOF) mass spectrometry. The average coefficient of variation (CV) of the protein abundance for all 645 proteins was approximately 50%. In Experiment 2, 2,277 proteins were identified by shotgun proteomics with an average CV of 9% for the protein abundance of all proteins. There were 472 out of 942 (50%) proteins found in all samples that were significantly affected by water deficit. The 472 proteins clustered into four groups: increased and decreased abundance of early- and late-responding protein profiles. Predominant functional categories of the early-responding proteins included photosynthesis, glycolysis, translation, and growth-related categories (steroid metabolism and water transport), whereas those for late-responding proteins were involved with transport, antioxidants, amino acid and carbohydrate metabolism. More proteins with less error were identified with gel-free shotgun proteomics than with 2-D PAGE. Most proteins could be matched with a probe set on the Affymetrix Vitis vinifera Genome Array. Only 25 and 28% of the protein profiles in Experiment 1 and 2, respectively, had a significant Pearson correlation with their transcript profiles; less than 20% for early responding proteins and more than 60% for late responding proteins.