Root mucilage enhances plant water use under combined soil and atmospheric drought

Abstract

Background and aims: Plants have evolved various root adaptive traits to enhance their ability to access soil water in stressful conditions. Although root mucilage has been suggested to facilitate root water uptake in drying soils, its impact during combined edaphic and atmospheric stress remains unknown. We hypothesized that mucilage decreases the saturated soil hydraulic conductivity, and consequently, a genotype with high mucilage production will exhibit lower maximum soil-plant hydraulic conductance and restrict transpiration at relatively low vapour pressure deficit (VPD). On the contrary, in drying soil, mucilage attenuates the gradients in matric potential at the root-soil interface and thus facilitates root water uptake, especially at high VPD.

Methods: We compared two cowpea genotypes with contrasting mucilage production rates and subjected them to three consecutively increasing levels of VPD (1.04, 1.8 and 2.8 kPa) while the soil was left to dry out. We measured the transpiration rate and soil and leaf water potentials and estimated canopy and plant hydraulic conductance during soil drying.

Key results: In wet soil conditions, the high-mucilage genotype restricted transpiration rate at lower VPD (1.46 kPa) compared with the low-mucilage genotype (1.58 kPa). Likewise, the initial slope of transpiration rate in response to VPD (the maximum conductance) was significantly lower in the high- compared with the low-mucilage genotype. During soil drying, the transpiration rate declined earlier in the low- compared with the high-mucilage genotype, supporting the hypothesis that mucilage helps to maintain the hydraulic continuity between roots and soil at lower water potentials in the high-mucilage genotype.

Conclusions: Root mucilage is a promising trait that reduces water use in wet soil conditions, thereby conserving soil moisture for critical phases (e.g. flowering and grain filling), both on a daily basis (increasing VPD) and on a seasonal time scale (soil drying).

Keywords: Cowpea; soil drying; soil hydraulic conductivity; vapour pressure deficit.

Fig. 1.

Leaf area and plant biomass for two genotypes with contrasting mucilage production. (A) Leaf area. (B) Shoot biomass. (C) Root biomass. (D) Root:shoot ratio.

Fig. 2.

Declining soil moisture (θ; in centimetres cubed per centimetre cubed) over time (days after last irrigation) for low- (red) and high-mucilage (blue) genotypes. Data are mean values ± s.e. (n = 8). Asterisks indicate significant differences between genotypes (***P < 0.001 and ****P < 0.0001).

Fig. 3.

Relationship between normalized transpiration rate and volumetric soil water content (θ; in centimetres cubed per centimetre cubed) for low- (red) and high-mucilage (blue) genotypes at high vapour pressure deficit (2.8 kPa). The normalized transpiration rate in response to soil water content (θ) was analysed using a segmental linear regression model. The threshold or breakpoint (critical soil content at which plants start to downregulate their transpiration) is shown in the figure (θ_BP; the dotted vertical line). The data points (filled triangles and circles) during soil drying are shown together with segmented regression lines (solid line).

Fig. 4.

Transpiration rate (E; in grams per second per centimetre squared) in response to step increases in vapour pressure deficit (VPD; in kilopascals) with segmented regression lines for low- (red) and high-mucilage (blue) genotypes. (A) Wet soil: average of the first three consecutive days of transpiration rate when plants were in well-watered conditions (n = 8). (B) Moderate drought: average of two consecutive days of transpiration rate in the middle of the dry-down experiment (n = 8). (C) Severe drought: average of the last 2 days of transpiration rate during the dry-down experiment (n = 8). The average soil water content (θ) for all replicates (n = 8) for the respective soil condition is displayed.

Fig. 5.

The relationship between the vapour pressure breakpoint (VPD_BP; in kilopascals), obtained from transpiration rate–vapour pressure deficit (VPD) relationship using segmented regression analysis, and plant hydraulic conductance (K_P; in grams per second per megapascal) determined in wet soil conditions for low- (red) and high-mucilage (blue) genotypes (n = 4).

Fig. 6.

Relationship between canopy conductance ($g_{\mathrm{c}}$; in grams per second per centimetre squared), vapour pressure deficit (VPD; in kilopascals) and leaf water potential (ψ_leaf; in megapascals) in wet soil conditions. (A) Response of $g_{\mathrm{c}}$ to an increase in VPD for low (red) and high-mucilage (blue) genotypes. Data are mean values ± s.e. (n = 4). Response of $g_{\mathrm{c}}$ to a decrease in leaf water potential for low- (B) and high-mucilage (C) genotypes. Different colours represent different individual plants (n = 4). Grey squares with error bars (in B and C) represent the mean values and s.e. of $g_{\mathrm{c}}$ measured at high and low VPD. Different lowercase letters indicate statistically significant differences (at P < 0.05) between the mean values of $g_{\mathrm{c}}$ of the two genotypes measured at low (1.04 kPa) and high (2.8 kPa) VPD.

Fig. 7.

Average transpiration rate (E; in grams per second per centimetre squared) in response to an increase in vapour pressure deficit (VPD; in kilopascals) during soil drying. (A) Low-mucilage genotype. (B) High-mucilage genotype. Data represents mean values ± s.e. (n = 8). The colour represents the daily average range of soil water content (θ; in centimetres cubed per centimetre cubed) across all replicates.

Fig. 8.

(A, B) Soil water content (θ; in centimetres cubed per centimetre cubed) at vapour pressure deficit breakpoints (VPD_BP; in kilopascals) (A) and Slope1 (B). (C) Relationship between fraction of transpirable soil water breakpoints (FTSW_BP) and VPD_BP. The coefficient of determination (R²) and P-values (P < 0.05 considered significant) are shown. The shaded area represents the 95 % confidence interval.

Fig. 9.

Relationship between transpiration rate (E; in grams per second) and leaf water potential (ψ_leaf; in megapascals) at different vapour pressure deficit (VPD) levels (indicated by different shapes) at different soil water content (indicated by different colours). Each data point represents an individual replicate (n = 4). VPD levels were predawn (PD), low (VPD1 = 1.04 kPa), medium (VPD2 = 1.8 kPa) and high (VPD3 = 2.88 kPa) for low- (A) and high-mucilage (B) genotypes, respectively. The linearly fitted slope of the E–ψ_leaf relationship equals the soil–plant hydraulic conductance (in grams per second per megapascal).