In most gene expression studies under drought stress, plants (often Arabidopsis) are grown in small pots. Reduced soil moisture (drought stress) is applied to pots by different means. Variation among genotypes in gene expression is measured on a given date after application of stress. The method has been criticized (e.g. Blum, 2014) whereas gene expression variation among genotypes could very well be driven by phenotypic variation in plant water status (water deficit in the plant) rather than by real genetic variation. This is now again demonstrated by Rymaszewski et al., 2017. This report clearly shows that when plants are grown in pots the variation in growth and plant stress among genotypes is driven by their size and water use which determine the phenotypic variation in their water status. For example, larger plants tend to have higher water use and lower water status than smaller plants. They then show that the phenotypic variation in water status among genotypes drives the variation in gene expression.

The take-home lesson: Gene expression under drought stress must be measured when all genotypes are at the same plant water status. Furthermore, if gene expression is measured in different plant parts (e.g. Zhu et al., 2007), it should be done when the different parts are at the same water status else the variation between plant parts in gene expression could be a function of the phenotypic variation among parts in water status.