Seed respiration is significantly more active before veraison and declines once ripening commences . A second pathway that is a main contributor to CO2 production and malate degradation during ripening is ethanolic fermentation, which is needed to maintain cellular function as conditions become increasingly hypoxic . The increase in ethanolic fermentation increases the synthesis of alcohol dehydrogenase to convert acetaldehyde to ethanol. This process generates hydrogen peroxide as an intermediate, which spurs negative feedback regulation of the attenuator of ROP-GTPase, thereby increasing ADH production . This could account for how cell death persists and expands throughout the mesocarp. Other research has posited that mesocarp cell death is not simply a consequence of normal berry metabolism, but another instantiation of the ubiquitous plant phenomenon of programmed cell death , the genetically regulated, physiological process of lysing unwanted cells via reactive oxygen species . In plants, ROS are metabolic by-products that serve as signals for biotic/abiotic stress responses and key phenological events. For example, stress from excessive light can cause singlet oxygen and hydrogen peroxide to accumulate in the reaction centers of the chloroplasts, large plastic pots for plants though H2O2 typically plays a larger role in signaling for stress responses as the more stable intermediate .

H2O2 also begins to accumulate at non-toxic levels in the skin at the onset of ripening for grapes and tomatoes, coinciding with transcriptional changes in ROS-regulating and oxidative stress-related genes . The known role of ROS as signaling molecules and the timing of its accumulation at veraison led Pilati to conclude that H2O2 is likely functioning as a signaling molecule at this time, though no direct evidence has been presented. Histone methyltransferases have been demonstrated to play regulatory roles in strawberry ripening and their homolog in grapevines, VvH3K4s, was identified in table grapes by Liu et al. and shown to be regulated by H2O2. Only one study thus far has measured ROS and cell death in winegrapes, where Gowder found significant correlation between H2O2 and cell vitality in Syrah fruit in combined skin and mesocarp samples, still not indicating precisely enough whether mesocarp concentrations of ROS are connected to cell death. ROS damage in grape berries can be quantified indirectly through cell vitality staining and measured directly by the thiobarbituric acid assay which measures the lipid peroxidation product malondialdehyde . Despite the prevalence of ROS throughout major cellular components, plant cells prevent ROS from building up to toxic levels through ROS scavenging, using enzymatic and non-enzymatic mechanisms . However, the ROS scavenging capacity of grape berries becomes increasingly impaired over the course of ripening. Extractable tannins and procyanidins decrease sharply, largely due to the oxidation of phenolic compounds in the seed, which reduces ROS scavenging capacity right next to the area where mesocarp cell death begins . Also, Shiraz berries have been reported to progressively decline in both their Ferric Reducing Antioxidant Potential and Diphenyl Picryl Hydrazyl Radical Scavenging ability .

However, the regulatory networks regulating the scavenging and production of ROS are extremely complex and our understanding of specific mechanisms, including berry cell death, is limited. Considering the extensive presence of PCD genes found in other plants, comparatively few have been identified in grapevines and none have been measured alongside an analysis of cell vitality dynamics. The first observations of PCD-related genes were reported by Pilati et al. who found that those associated with PCD suppression were downregulated and those associated with activation were upregulated between just before veraison and mid-ripening of Pinot Noir berries. Sweetman et al. published a transcriptome of Syrah that described an increase in the abundance of VvBAP1 by an order of 7 from pre-veraison to full ripeness. Considering the proclivity of Syrah to shrivel, it may seem as though VvBAP1 is a positive regulator of PCD, however Cao et al. further investigated this gene and found the opposite. Examining its expression in Syrah, the relative expression of VvBAP1 was correlated with the degree of mesocarp cell death, and significantly downregulated at 92 DAA in a drought stress treatment, almost the exact time of cell death onset found by Bonada et al. . This negative regulation of PCD was further confirmed in other organisms: its expression in Arabidopsis, tobacco, and yeast significantly improved tolerance to H2O2, resulting in plant tissue and cells with less oxidative damage. It was previously mentioned that H2O2 can act as a benign signal at low concentrations, however it has also been observed to function as a trigger for programmed cell death via a signal transduction pathway .

The typical parameters used to monitor the ripening of commercial wine grapes are acid catabolism , sugar accumulation, the accumulation of phenoliccompounds, seed maturity, and flavor development. As was previously shown, the latter two are physiologically and sensorily implicated in the onset of mesocarp cell death and late season dehydration. Delaying the onset of cell death would enable winemakers to continue the practice of extended hangtime for desired wine characteristics while preventing unwanted yield losses for growers. When during ripening is the optimal time to relieve water stress and how much relief is needed in the form of irrigation volume has yet to be elucidated. A more nuanced picture of the timing and bare minimum amount of irrigation needed will prepare growers for a likely future under water restrictions. As a first measure, we set out to investigate whether additional pulses of irrigation prior to or during the onset of cell death in Cabernet Sauvignon berries would either postpone onset or delay the rate of cell death. We hypothesized that an early pulse two-weeks prior to projected cell death onset would relieve water stress and hold over triggers for programmed cell death. A later pulse of irrigation concurrent with projected cell death onset was expected to insufficiently compensate for accelerated xylem backflow at a time of diminishing phloem inflow. We addressed the need to connect ROS in the mesocarp to cell death by measuring H2O2 and MDA concentrations in skinned and seeded grape samples, anticipating that their accumulation would coincide and track with the progression of mesocarp cell death.This study used mature Cabernet Sauvignon grapevines planted on 420A rootstock growing in an experimental vineyard on the University of California, Davis campus in Davis, CA. Vines were cane pruned and trained to a VSP system planted 7 feet apart with 11 feet between rows. Canopy and pest management followed standard commercial practices. Anthesis occurred on May 15th 2022 and veraison, defined by 50% of the clusters with at least one colored berry, occurred at approximately 58 days after anthesis . Vines from one row were randomly assigned to either a control, early, or late irrigation treatment . Vines from different treatments were separated by two buffer vines, each subjected to the same irrigation regime as the adjacent experimental vine. The control treatment used a standard regulated deficit irrigation regime for a hot growing region. Prior to the experimental period, plant pots with drainage vines were irrigated weekly starting one month before anthesis at 35% replacement of evapotranspiration , which was estimated using a crop coefficient-based ETc, vineyard spacing, and precipitation as parameters. One month after bloom and through veraison, RDI was increased to 50% and further increased to 70% in anticipation of forecasted heatwaves. Irrigation frequency was increased to every 2 – 3 days during the experimental period , to ensure the treatments had enough time to affect soil moisture in the root zones. Control vines received 0.45in for the first week of the experimental period, then 0.34in for the rest of the period . Vines in the early and late treatments received the same irrigation as the control, except for an additional ‘pulse’ that increased irrigation to 0.54in/week for two weeks. The early treatment ‘pulse’ was applied in the two weeks before the expected onset of cell death and the late treatment ‘pulse’ was applied in the two weeks after . The projected cell death onset for our study was based on the only other study tracking phenology and cell death of specifically Cabernet Sauvignon fruit .

During the first week of the first irrigation treatment pulse a 20% difference in irrigation volume was applied to the vines, with the early vines received 0.56 in while the late and control treatments received 0.45 in. After seeing little remediation of water stress by the end of the first week, that difference was increased to 40% the second week by reducing the water applied to the late and control vines to 0.34 in . This difference was maintained for the late irrigation pulse, yet in the final week of sampling the weekly volume was reduced to .22 in.Vines were monitored for water stress by measuring gas exchange and midday stem and predawn water potentials weekly over the experimental period. Pre-dawn leaves were sampled from 0400HR to 0500HR and placed in a humidified bag immediately after excision from the shoot . Midday leaves were sampled and measured for gas exchange from 1200 to 1330HR and were covered in a humidified, foil-wrapped bag for 20 minutes before excision to equilibrate water potential in the stem and the leaf . Leaf selection was limited to sun-exposed, mature leaves at a distance of 8-12 leaves from the shoot tip. Bagged leaves were immediately placed in a cooler after excision, then transported to the lab and measured for water potential with a pressure bomb within one hour of sampling. Photosynthesis and stomatal conductance were measured weekly at the same time and in the same manner as midday water potential sampling using a LICOR 6800 gas exchange system.Four clusters per vine were marked for repeated sampling . 1 berry/cluster was sampled each for FDA staining and ROS analyses on 9 dates over the experimental period . 1 berry/cluster was sampled for transcriptomics on 3 dates before and after the expected onset of cell death . Size, color, and softness were used to select berries that were relatively advanced in maturity. Sampling continued until total soluble solids reached 27 °Brix, at which point the rest of the cluster was crushed, and the grape must analyzed for TSS, titratable acidity , and pH.The FDA staining technique from Krasnow et al. was used to quantify berry cell vitality. First, we made a transverse cut with a razor blade to remove 1-2 mm from the pedicel end of the berry, to locate the seeds. We then cut the berry in half longitudinally between the seeds, keeping the blade as close to the center of the berry as possible. One half of each berry was then placed in a humidified petri dish and refrigerated and the other half was crushed and the juices homogenized with the other berries from the same vine. Juice samples were measured for osmotic potential with a Vapro 5600 vapor pressure osmometer. The refrigerated halves were stained with a 9.6 uM FDA solution prepared by adding 2ul of 4.8mM FDA stock solution to 1ml of sucrose solution with the same osmotic potential as the juice samples, to avoid cell death from osmotic shock. Preliminary trials by Krasnow et al. showed that variability in osmotic potential between berries from the same vine was small and that using homogenized samples did not make differences between solutions and berry osmotic potentials larger than 0.5 MPa thresholds are recommended to avoid osmotic shock. Berry halves were blotted with a kimwipe dipped in the staining solution to remove cellular debris and sectioned again to obtain a 2-3 mm thick longitudinal cross section without seeds. Sections were then abundantly coated in the staining solution and allowed to take up the solution for 30 minutes before imaging.Berries sampled for ROS and lipid oxidation products were immediately frozen in liquid nitrogen and stored in a -80°F freezer. The mesocarp tissue was isolated by peeling the frozen berries with a scalpel, then gently breaking and deseeding the frozen pulp. The four berries sampled per vine were then homogenized and ground to a fine powder in liquid nitrogen . H2O2 concentrations were measured with an Amplex®Red Hydrogen Peroxide/Peroxidase Assay kit, following Pilati et al. . This is a fluorimetric assay quantifying the absorbance of 10-acetyl-3,7-dihydroxyphenoxazine, which fluoresces red when oxidized in the presence of H2O2 and peroxidase. 0.2g of powder was solubilized in 0.5ml of 50mM phosphate buffer and kept on ice for 5 minutes.