Genotype by watering regime interaction in cultivated tomato: lessons from linkage mapping and gene expression

Abstract

As a result of climate change, drought will increasingly limit crop production in the future. Studying genotype by watering regime interactions is necessary to improve plant adaptation to low water availability. In cultivated tomato (Solanum lycopersicum L.), extensively grown in dry areas, well-mastered water deficits can stimulate metabolite production, increasing plant defenses and concentration of compounds involved in fruit quality, at the same time. However, few tomato Quantitative Trait Loci (QTLs) and genes involved in response to drought are identified or only in wild species. In this study, we phenotyped a population of 119 recombinant inbred lines derived from a cross between a cherry tomato and a large fruit tomato, grown in greenhouse under two watering regimes, in two locations. A large genetic variability was measured for 19 plant and fruit traits, under the two watering treatments. Highly significant genotype by watering regime interactions were detected and resulted from re-ranking more than scale changes. The population was genotyped for 679 SNP markers to develop a genetic map. In total, 56 QTLs were identified among which 11 were interactive between watering regimes. These later mainly exhibited antagonist effects according to watering treatment. Variation in gene expression in leaves of parental accessions revealed 2259 differentially expressed genes, among which candidate genes presenting sequence polymorphisms were identified under two main interactive QTLs. Our results provide knowledge about the genetic control of genotype by watering regime interactions in cultivated tomato and the possible use of deficit irrigation to improve tomato quality.

Model-Assisted Estimation of the Genetic Variability in Physiological Parameters Related to Tomato Fruit Growth under Contrasted Water Conditions

“…On the contrary, Levovil was the most sensitive to WD, since it lost more than half of its fresh weight and it doubled its dry matter content under WD. In the population, the relative decrease in fruit fresh weight under WD was negatively correlated to the fresh weight under control conditions, indicating that large fruit genotypes were the most sensitive to WD, as mentioned in Albert et al (2016). On the contrary, the increase in dry matter content under WD was rather independent of the dry matter content observed under control condition.…”

“…When distributions were skewed, the best corrections for normality were applied: LOG10(nuM); $\sqrt{tstar}$; LOG10(lp1); LOG10(lx); LOG10(lx1); 1/rxp. The QTL detection was performed as presented in Albert et al (2016) using the genetic map developed by Pascual et al (2016) which included 501 SNP markers covering 80% of the tomato genome. Briefly, the simple interval parametric mapping model (Lander and Botstein, 1989) based on the EM algorithm method implemented in the R/QTL package (Broman et al, 2003) was used.…”

“…The number of genes within each interval was identified from the tomato genome annotation (2.4). We reported the locations between the detected QTLs and the QTLs identified on phenotypic traits (plant and fruit traits) measured on the same plants (see Albert et al, 2016). …”

“…In tomato, only a few QTLs/genes involved in the response to water deficit are known (Labate et al, 2009). In a recent study (Albert et al, 2016), a RIL population of 117 F7 recombinant inbred tomato lines has been genotyped for 501 SNP markers and phenotyped under control (C) and water deficit (WD). This study revealed a total of 56 QTLs of plant and fruit traits, among which 11 depended on watering regime.…”

“…The RIL population is the one previously genotyped by Albert et al (2016). The genetic variability in fruit traits and model parameters was analyzed by Principal Component Analysis (PCA) and through QTL analysis.…”