Starch depletion in the xylem and phloem ray parenchyma of grapevine stems under drought

Abstract While nonstructural carbohydrate (NSC) storage can support long-lived woody plants during abiotic stress, the timing and extent of their use are less understood, as are the thresholds for cell mortality as NSCs and water supplies are consumed. Here, we combine physiological and imaging tools to study the response of Vitis riparia to a 6-week experimental drought. We focused on the spatial and temporal dynamics of starch consumption and cell viability in the xylem and phloem of the stem. Starch dynamics were further corroborated with enzymatic starch digestion and X-ray microcomputed tomography imaging. Starch depletion in the stems of droughted plants was detected after 2 weeks and continued over time. We observed distinct differences in starch content and cell viability in the xylem and phloem. By the end of the drought, nearly all the starch was consumed in the phloem ray parenchyma (98 % decrease), and there were almost no metabolically active cells in the phloem. In contrast, less starch was consumed in the xylem ray parenchyma (30 % decrease), and metabolically active cells remained in the ray and vessel-associated parenchyma in the xylem. Our data suggest that the higher proportion of living cells in the phloem and cambium, combined with smaller potential NSC storage area, rapidly depleted starch, which led to cell death. In contrast, the larger cross-sectional area of the xylem ray parenchyma with higher NSC storage and lower metabolically active cell populations depleted starch at a slower pace. Why NSC source-sink relationships between xylem and phloem do not allow for a more uniform depletion of starch in ray parenchyma over time is unclear. Our data help to pinpoint the proximate and ultimate causes of plant death during prolonged drought exposure and highlight the need to consider the influence of within-organ starch dynamics and cell mortality on abiotic stress response.


Figure S2
. Comparison of stem water potential (MPa) measured by each ICT psychrometer (n=4) and a pressure chamber in the afternoon of August 25, 2020 on the day before the start of the experimental drought.This comparison was conducted to ensure that the psychrometer measurements were giving valid stem water potentials.The pressure chamber was used to measure stem water potentials prior to the start of the experimental drought (see Fig. S6), and the psychrometers were used to monitor stem water potentials throughout the experimental drought (see Figs.   S1.Since starch did not differ by stem location, data were averaged across stem locations for visualization and analysis in Fig. 3.

Figure S8.
Visualization of (A) starch content and (C) metabolically active cells in the stem of each plant by treatment following drought.Iodine-stained starch was visualized with compound light microscopy and then iodine-stained starch was identified in the xylem (orange) and phloem (pink) ray parenchyma via thresholding in ImageJ in (B).Quantification of this starch thresholding as percent starch for all plants is displayed in Fig. 3. Additionally, fluorescent living cells following FDA staining were visualized with fluorescent microscopy in (C) and autofluorescence was visualized with water in (D).Black box indicates missing image.Images shown are for stem samples at 10% above the soil, but data from 50% above the soil was also collected.Scale bar = 0.5 mm Table S1 Percent of starch in the xylem ray parenchyma, phloem ray parenchyma, and whole rays quantified using thresholding in ImageJ as well as starch concentration from enzymatic digestion for each individual plant.Percent starch was quantified at two stem locations, 10% and 50% above the soil.Enzymatically-derived starch concentrations were quantified at 10% above the soil.
Figure S2.Comparison of stem water potential (MPa) measured by each ICT psychrometer (n=4) and a pressure chamber in the afternoon of August 25, 2020 on the day before the start of the experimental drought.This comparison was conducted to ensure that the psychrometer measurements were giving valid stem water potentials.The pressure chamber was used to measure stem water potentials prior to the start of the experimental drought (see Fig.S6), and the psychrometers were used to monitor stem water potentials throughout the experimental drought (see Figs. 2, S7).

Figure S3 .
Figure S3.Iodine-stained starch (%) in (A) xylem ray parenchyma, (B) phloem ray parenchyma, and (C) whole rays at two stem locations, 10% (left column) and 50% (right column) of the total plant length above the soil surface following sequential harvesting.Error bars denote ± 1 SD of the mean.Statistical results displayed above plots are from two-way ANOVA testing (treatment x stem location).Data are provided in TableS1.Since starch did not differ by stem location, data were averaged across stem locations for visualization and analysis in Fig.3.

Figure S4 .
Figure S4.MicroCT images of stems from the well-watered control group (A-C), the 2-week drought group (D-F), the 4-week drought group (G-I), and the 6-week drought group (J-L).Ray parenchyma are full of starch granules when appearing bright gray in color, and depleted when appearing dark gray or black in color.All stems were collected at 10% above the soil.Plant IDs are listed in white text in the lower left corner of each image.Scale bar = 1 mm.

Figure S5 .Figure S7 .
Figure S5.Fluorescence intensity in the (A) xylem and (B) phloem of stems across drought treatment groups.Fluorescence intensity was quantified using the integrated density measurement in FIJI on FDA-stained images of the stem at 10% and 50% of the total plant length above the soil surface.Error bars denote ± SE of the mean.Statistical results displayed above plots are from oneway nested ANOVA testing with stem location treated as technical replicates per plant (treatment:stem) and lowercase letters within the plots indicate whether treatments significantly differed from each other based on Tukey's honest significant difference (HSD) at  = 0.05