A time-honoured technique goes underground to investigate root drought tolerance. A commentary on: ‘Root pressure–volume curve traits capture rootstock drought tolerance’

Abstract

This article comments on:

M. K. Bartlett, G. Sinclair, G. Fontanesi, T. Knipfer, M. A. Walker and A. J. McElrone, Root pressure–volume curve traits capture rootstock drought tolerance, Annals of Botany, Volume 129, Issue 4, 1 April 2022, Pages 389–402 https://doi.org/10.1093/aob/mcab132

Sometimes, a method is not just a method. In the case of pressure–volume (p–v) curves, the method is based on results obtained from one of the most venerable tools in all of plant biology, the Scholander pressure chamber (Scholander et al., 1965). In that ground-breaking paper, the authors demonstrated not only how to measure plant water potential, notably by shooting twigs from the tops of redwoods and pressurizing them in a ‘pressure bomb’ at the base of the tree, but also how to deconstruct water potential into its components graphically using p–v curves. Despite, and perhaps inspired by, occasional criticism of the pressure chamber method (e.g. Boyer, 1967), p–v curves have been refined and used for >50 years to determine the components of water potential in plant leaves and other above-ground organs. In contrast, p–v curves have infrequently been used to investigate roots (but see Ritchie and Roden, 1985), at least at the scale of the individual root. In a recent study, Bartlett et al. (2022) used p–v curves to investigate how drought-tolerant vs. drought-sensitive roots of grapevine (Vitis vinifera) adjusted to changes in their water supply. Although the two groups of rootstocks differed in the magnitude and, occasionally, direction of their responses, p–v curves allowed Bartlett et al. to identify key root traits associated with the maintenance of water uptake and resistance to water loss. Their work provides a convincing argument that a closer focus on what the authors call ‘root p–v curve traits’ could help inform predictions of whole-plant responses to drought, and, moreover, help select for root traits that could improve plant survival in a drying world.

Bartlett et al. provide something of a primer on how to construct and interpret p–v curves. Following convention, the water potential (Ψ) of the plant part – in this case, a cut root segment with an intact root tip – was measured in a Scholander-type pressure chamber, weighed and allowed to dry down on the bench. At suitable intervals, the measurements were repeated until several data points were obtained at low values of Ψ. Again following convention, p–v curves were drawn with –1/Ψ on the y-axis and 100 – RWC on the x-axis, where RWC is the relative water content at each point, calculated from saturated weight – dry weight of the segment obtained after oven-drying. The point at which the root loses turgor can be determined by drawing a straight line to the y-axis from the the point at which the curve becomes linear, and RWC at the turgor loss point can be read from a vertical line to the x-axis from the same inflection point. The p–v curve traits that Bartlett et al. focused on were the osmotic potential at the turgor loss point (π_tlp) and at full hydration (π₀), and the capacitance, or the change in water content over the change in water potential above (C_ft) and below (C_tlp) the turgor loss point. These traits are crucial for a root’s ability to maintain its structural integrity and continue to extract water from a drying soil.

A nice feature of the study by Bartlett et al. is its use of several rootstock varieties, previously categorized as drought tolerant or drought sensitive, which were grafted to a scion (above-ground portion) of one variety, ‘Chardonnay’. Thus, both root and shoot responses to changes in watering could be attributed to properties of the roots themselves. Root p–v curves and measurements of plant gas exchange, hydraulic conductance and growth were made for potted plants before, during and after drought. The authors predicted that rootstocks with lower π_tlp, π₀, C₀ and C_tlp would function better during drought because (1) lower root osmotic potentials would permit greater water uptake from drying soil and (2) lower C_tlp would prevent excessive tissue water loss at lower plant Ψ (Fig. 1). In other words, the authors predicted that downward adjustments in π and C would be seen in both drought-sensitive and drought-tolerant rootstocks in response to soil drying. Indeed, that is generally what they observed for both groups. However, the details of the adjustments in root traits and their relationships to other plant variables were complicated and in some instances unexpected.

Fig. 1.

(A) Hypothetical soil moisture curves, showing the range of moisture content (grey area) experienced during a drought at two differ different maximum rooting depths, shallow (dotted) vs. deep (dashed): at the shallow depth, high root capacitance would promote root shrinkage and disconnection from dry soil. (B) Conceptual relationship between soil moisture and soil water potential for a light clay soil (Philip, 1957): the deeply rooted plant experiences falling water potential with little change in water content, favouring low capacitance (‘stiff’) roots that do not pull away from the soil, while the shallow-rooted species reaches water contents where soil tensions increase dramatically, favouring high root capacitance and disconnection from the soil.

Because high values of capacitance, particularly C_tlp, would lead to large losses of root volume at low root Ψ, low C_tlp would reduce root shrinkage and thereby help maintain root contact with the soil. However, in the study by Bartlett et al., C_tlp was statistically associated with neither plant hydraulic conductance nor water stress. A possible explanation is that low C_tlp, particularly for drought-tolerant roots, prevented root–soil contact from changing during the experiment, rendering root shrinkage irrelevant. The authors suggested that root shrinkage could also trigger abscisic acid (ABA) to be exported to or newly produced in leaf mesophyll cells (e.g. Gupta et al., 2020), thereby causing stomata to close and gas exchange to cease, preventing shoot water stress from occurring. Although ABA was not measured in this study, higher rates of stomatal conductance and photosynthesis were observed in plants with low root C_tlp, and so presumably low root shrinkage during drought. Low values of C_tlp could also have particular benefits for grapevine roots, which develop cortical airspaces or lacunae during drought that can impede radial water flow to the root xylem (Cuneo et al., 2016). High values of C_tlp, on the other hand, could support greater plant gas exchange if water released from storage in one plant organ or tissue helps to stave off water stress in another, as occurs in desert succulents (Goldstein et al., 1991; Ahl et al., 2019).

Against these expectations, the most surprising and intriguing result of the study by Bartlett et al. was that the drought-tolerant group did not show lower values of π_tlp, π₀, C₀ and C_tlp than those of the drought-sensitive group after the plants were exposed to drought. Moreover, during drought, values of π_tlp and π₀ were even higher for the drought-tolerant group than for the drought-sensitive group. This apparent lack of osmotic adjustment would seem to put the drought-tolerant roots at a disadvantage in drying soil. In contrast, under well-watered conditions, values for these p–v curve traits were lower for the drought-tolerant group than the drought-sensitive group. Thus, the advantage goes to the drought-tolerant group with respect to soil water uptake when conditions are favourable or, even more importantly, during early stages of soil drying. As Bartlett et al. discuss, this pattern is suggestive of how grapevines may cope with drought in the field. The unexpected behaviour of root traits under drought may be explained by the fact that grapevines tend to be deeply rooted (Smart et al., 2006), thus their young roots, such as those measured in this study, would occupy deeper, wetter soil layers, especially during moderate drought. Adjustment to less favourable conditions, or plasticity in root p–v curve traits such as π_tlp and π₀, may be immaterial for deep-rooted plants (Fig. 1).

The picture that emerges from the study by Bartlett et al. has implications not only for root trait selection but also for the cultivation and management of grapevine and other deep-rooted plants. With respect to trait selection, the authors suggest that, among other things, the biochemical and biophysical properties of cell walls bear further investigation because of their role in determining root capacitance. Thicker, stiffer cell walls, which are less elastic, lead to lower capacitance, and are characterized by components such as lignin, suberin and complex polysaccharides (Ahl et al., 2019) that could be targets of selection, particularly since cell wall thickness has been shown to increase in roots during drought (Piro et al., 2003). Another p–v curve trait demanding closer scrutiny is root osmotic adjustment. As Bartlett et al. demonstrate, values of these traits in drought-tolerant roots did not decrease during drought; in other words, π_tlp and π₀ appeared to be constitutive rather than plastic, with relatively stable values low enough to maximize water uptake in moist or moderately dry soil. The lack of osmotic adjustment could represent a saving in soluble carbohydrates that could be used instead to grow deeper roots. Thus, instead of selecting for alleles that may be associated with root osmotic adjustment, a better target in a plant such as grapevine would be alleles associated with deep rooting (Lynch, 2021), a habit that will become increasingly beneficial in the years ahead as climate change intensifies.