Transcriptomic Changes Drive Physiological Responses to Progressive Drought Stress and Rehydration in Tomato

Abstract

Tomato is a major crop in the Mediterranean basin, where the cultivation in the open field is often vulnerable to drought. In order to adapt and survive to naturally occurring cycles of drought stress and recovery, plants employ a coordinated array of physiological, biochemical, and molecular responses. Transcriptomic studies on tomato responses to drought and subsequent recovery are few in number. As the search for novel traits to improve the genetic tolerance to drought increases, a better understanding of these responses is required. To address this need we designed a study in which we induced two cycles of prolonged drought stress and a single recovery by rewatering in tomato. In order to dissect the complexity of plant responses to drought, we analyzed the physiological responses (stomatal conductance, CO2 assimilation, and chlorophyll fluorescence), abscisic acid (ABA), and proline contents. In addition to the physiological and metabolite assays, we generated transcriptomes for multiple points during the stress and recovery cycles. Cluster analysis of differentially expressed genes (DEGs) between the conditions has revealed potential novel components in stress response. The observed reduction in leaf gas exchanges and efficiency of the photosystem PSII was concomitant with a general down-regulation of genes belonging to the photosynthesis, light harvesting, and photosystem I and II category induced by drought stress. Gene ontology (GO) categories such as cell proliferation and cell cycle were also significantly enriched in the down-regulated fraction of genes upon drought stress, which may contribute to explain the observed growth reduction. Several histone variants were also repressed during drought stress, indicating that chromatin associated processes are also affected by drought. As expected, ABA accumulated after prolonged water deficit, driving the observed enrichment of stress related GOs in the up-regulated gene fractions, which included transcripts putatively involved in stomatal movements. This transcriptomic study has yielded promising candidate genes that merit further functional studies to confirm their involvement in drought tolerance and recovery. Together, our results contribute to a better understanding of the coordinated responses taking place under drought stress and recovery in adult plants of tomato.

Keywords: ABA; RNA sequencing; gene-expression cluster analysis; photosynthesis; proline; stomatal conductance; water stress.

Figure 1

Experimental outline. (A) Volumetric soil water content (θ) throughout the progression of the experiment. Values represent average measurements ± SD of three replicates. Asterisks denote significant differences according to Student's t-test between well watered and stressed pots. ^*, ^**, and ^*** indicate significantly different values in drought stress compared to well-watered pots at p ≤ 0.05, p ≤ 0.01, and p ≤ 0.001, respectively. (B) Schematic representation of the experimental design highlighting the points Dr1 (16 d of irrigation withholding), RW (7 days of irrigation), and Dr2 (6 days of irrigation withhold). (C) Gene expression of Solyc03g116390.2.1 in leaves after two cycles of drought stress (Dr1 and Dr2) and 1 week of rewatering (RW). RNA samples extracted from leaves of well watered plants were used as controls. Gene expression analyses were conducted by qRT-PCR. DOE, days of experiment.

Figure 2

Leaf gas exchange parameters in well watered and drought stressed plants throughout the experiment. (A) Photosynthetic CO₂ assimilation (A); (B) stomatal conductance to water vapor (gs). Values represent average measurements ± SD, n ≥ 5. ^*, ^**, and ^*** indicate significantly different values in drought stressed compared to well-watered plants at p ≤ 0.05, p ≤ 0.01, and p ≤ 0.001, respectively. DOE, days of experiment.

Figure 3

Chlorophyll fluorescence parameters in well watered and drought stressed plants throughout the experiment. (A) Quantum yield of PSII (ΦPSII); (B) Maximum quantum yield of PSII (Fv/Fm). Values represent average measurements ± SD, n ≥ 5. ^* and ^** indicate significantly different values in drought stress compared to well-watered plants at p ≤ 0.05 and p ≤ 0.01, respectively. DOE, days of experiment.

Figure 4

Impact of drought stress cycles on biometric parameters. Average Leaf Area (A) and total dry weight (B) in well watered and drought stress treatments (n = 3). For dry weight measurements, roots, stems, leaves, and fruits were included. Measurements were taken at the end of the experiment. ^* and ^*** indicate significantly different values in drought stress compared to well-watered plants at p ≤ 0.05 and p ≤ 0.001, respectively.

Figure 5

Quantification of Proline (A) and ABA content (B) in leaves; Gene expression of rate limiting enzymes P5CS (C) and NCED (D). DOE, days of experiment. ^*, ^** and ^*** indicate significantly different values in drought stressed compared to well-watered plants at p ≤ 0.05, p ≤ 0.01, and p ≤ 0.001, respectively.

Figure 6

qRT-PCR validation of RNA sequencing data on 14 selected genes (Supplementary Table S6). (A) Expression value detected by RNA-seq method. (B) Expression analysis conducted by qRT-PCR. Data have been plotted on a log₂ scale. (C) Correlation between RNA-Sequencing and qRT-PCR data. The normalized expression value obtained with RNA sequencing (x axis) were compared to the log₂ of fold increase by qRT-PCR (y axis). RNA from well-watered control plants was used as calibrator sample.

Figure 7

(A) Heatmap of selected clusters of Differentially Expressed Genes showing their expression behavior. Red and blue indicate higher and lower expression values, respectively. (B) Barplot showing GO Enrichment Analyses (goseq R package, FDR ≤ 0.05) of clusters 1, 2, 14, 17, 18 and 7, 20 independently, plotting GO terms (y axis) and the reciprocal of enrichment p value (x axis). Colors indicate GO ontology: red for Biological Process (BP), blue for Molecular Function (MF), and green for Cellular Component (CC).