Global analysis of plasticity in turgor loss point, a key drought tolerance trait

Background/Question/Methods

Many plant species face increasing drought under climate change. Plasticity in key physiology traits is expected to strongly influence species responses by widening their range of tolerable conditions. During drought, plants are well known to shift their turgor loss or “wilting” point (π_tlp), by accumulating solutes in their leaf cells (i.e., “osmotic adjustment”), which allows plants to maintain gas exchange and photosynthesis under drier soil conditions. Plants also show shifts in their cell hydration at wilting point (RWC_tlp), and the relative importance of π_tlp and RWC_tlp to drought tolerance has long been debated in the literature. Although plasticity in these traits (Δπ_tlp and ΔRWC_tlp) has itself been considered a key drought tolerance trait for decades, there has been no synthesis of the importance of this plasticity to drought tolerance across species and ecosystems. We compiled a novel global database of seasonal changes in π_tlp and RWC_tlp for 283 species to address the following questions: 

1) For droughted plants, is the primary determinant of π_tlp and RWC_tlp the pre-drought values, or the plastic shift during drought?

2) How does plasticity in these traits vary across ecosystems, especially relative to ecosystem water supply?

3) Do crop species exhibit greater plasticity than wild species?

Results/Conclusions

Plasticity in π_tlp was widely prevalent but moderate on average (-0.45 MPa), accounting for 15.7% of post-drought π_tlp. The pre-drought π_tlp was also a considerably stronger predictor of post-drought π_tlp than was Δπ_tlp across wild species (r²  = 0.62, 0.34, respectively, both p < 0.001). However, for cultivars of certain crop species, Δπ_tlp accounted for major differences in post-drought π_tlp. Overall, Δπ_tlp did not account for a significantly greater proportion of post-drought π_tlp in crop than wild species. Conversely, RWC_tlp did not exhibit significant seasonal differences for wild species (t-test, p = 0.16), although RWC_tlp did become significantly lower for crop species (p = 0.01). For wild species, significantly more variation in Δπ_tlp occurred among species within ecosystems than at the ecosystem levels of study sites or biomes, and Δπ_tlp was not significantly different among biomes. Drier sites and biomes contained species with significantly more negative shifts in π_tlp, although wet and dry season π_tlp were considerably more strongly correlated with climate than Δπ_tlp. Thus, annual measurements of π_tlp provide a reasonable simplification for characterizing drought tolerance across diverse species and communities, whereas plasticity in π_tlp makes a minor contribution to drought tolerance.