Root traits in Crambe abyssinica Hochst and Raphanus sativus L. plants are associated with differential tolerance to water deficit and post-stress recovery

Plant material and growth conditions

Independent experiments were performed at the Laboratory of Ecophysiology and Plant Productivity at the Goiano Federal Institute–Rio Verde campus, Goiás, Brazil (17°48′07.9″S 50°54′20.7″W), from March to May 2018 for each species in a greenhouse with controlled temperature (~26 °C) and relative humidity (51–75%). Plants of Crambe abyssinica Hochst (FMS Brilhante, 2017/2017 crop) and Raphanus sativus L. var. oleiferus Mertzg (IPR 116, 2017/2017 crop) were cultivated in 4.65 dm³ plastic pots (14.0 high diameter × 10.2 base diameter × 35.5 cm height) filled with 4.5 kg of a mixture of Red Latosol soil (LVdf) and sand (2:1). The substrate had the following chemical characteristics: P – 0.7 mg dm⁻³; K – 13.0 mg dm⁻³; Ca – 1.54 cmol_c dm⁻³; Mg – 0.22 cmol_c dm⁻³; Al – 0.05 cmol_c dm⁻³; H+Al – 1.3 cmol_c dm⁻³; S – 3.5 mg dm⁻³; B – 0.8 mg dm⁻³; Cu – 1.0 mg dm⁻³; Fe – 37.8 mg dm⁻³; Mn – 13.2 mg dm⁻³; Zn – 0.1 mg dm⁻³; Na – 6.0 mg dm⁻³; pH CaCl₂ – 5.6; SB – 58%; cation exchange capacity – 3.1 cmol_c dm⁻³; organic matter – 10.9 g dm⁻³. The total substrate (0.60 m³) was fertilized with 82.8 g of formulated NPK 4 - 14 - 08 and 2.4 g of filler dolomitic limestone to increase base saturation to 60%. The water holding capacity (WHC) of the substrate was determined through the saturation of the pots and the moisture was monitored daily by the gravimetric method.

A total of 28 days after planting, plants of C. abyssinica and R. sativus having three of four pairs of expanded leaves were exposed to well-watered conditions (WW; 90% WHC) and water deficit (WD; 40% WHC) conditions. The plants were kept under WD for 7, 14, and 21 days with rehydration for 48 h soon after each episode of water deficit. Treatments consisted of well-watered plants at 7 (WW7), 14 (WW14) and 21 (WW21) days; water deficit for 7 (WD7), 14 (WD14) and 21 (WD21) days; and rehydrated plants after WD for 7 (RH7), 14 (RH14) and 21 (RH21) days.

The experiments were performed with a randomized block design including five replicates. Each experimental unit consisted of one plant per pot.

Water relations

Predawn leaf water potential (ψ_w) was measured between 04:00 and 06:00 am using a Scholander pressure chamber (model 3005-1412, Soilmoisture Equipment Corp., Goleta, CA, USA). Leaf osmotic potential (ψ_sl) and root osmotic potential (ψ_sr) were evaluated using a vapor pressure osmometer (VAPRO 5600, VAPRO®, Wescor, UT, USA) according to Pierre & Arce (2012). Osmotic potential values were calculated using the Van’t Hoff equation: (ψ_s = −R × T × Cs, where R is the universal gas constant (0.08205 L atm mol⁻¹ K⁻¹), T is the temperature in °K (T °K = T °C + 273) and Cs the solute concentration (M), usually expressed in atmospheres and converted to MPa (0.987≈ 1 atm 0.1 MPa). Relative water content (RWC) was determined from fresh matter (FM), turgid matter (TM) and dry matter (DM) data obtained by weighing leaf discs according to Barrs & Weatherley (1962), and calculated as RWC (%) = (FM − DM)/(TM − DM) × 100.

Morphological analysis and biomass allocation

Plants were measured to determine the plant height (PH, cm), stem diameter (SD, mm), nodes number (NN), leaf number (LN), and the main root length (RL, cm). The leaf area (LA, m²) was calculated using the ImageJ software (Image Processing and Analysis in Java, v. 1.52d, USA). Leaves, stem and roots were separated into paper bags and dried in an air circulation oven at 65 °C until constant weight to obtain the leaf dry matter (LDM, g), stem dry matter (StDM, g) and root dry matter (MSR, g). Shoot dry matter (SDM = LDM + StDM), total dry matter (TDM = SDM + RDM), specific leaf area (SLA), leaf area ratio (LAR) and RDM/SDM were calculated.

The root system was evaluated with digital images obtained by digital camera (6MP; 3264 × 1836 px) using the Safira v1.1 software (Fiber and Root Analysis System; Embrapa Agricultural Instrumentation, São Paulo, Brazil). The roots were classified into three diameter classes: very fine roots (Ø < 0.5 mm), fine roots (0.5 < Ø < 2.0 mm) and large roots (Ø > 2.0 mm), according to the criterion proposed by Böhm (1979). From the data obtained by image processing, the morphological characteristics were determined: mean root diameter (MRD, mm), very fine root length (VFRL, cm), fine root length (FRL, cm), large root length (LRL, cm), total root length (TRL, cm), very fine root surface area (VFRA, cm²), fine root surface area (FRA, cm²), large root area (LRA, cm²), total root surface area (TRA, cm²), very fine root volume (VFRV, cm³), fine root volume (FRV, cm³), large root volume (LRV, cm³), total root volume (TRV, cm³), TRV/SDM ratio (cm g⁻¹), specific root length (SRL = TRL/RDM, cm g⁻¹), root thickness, (RT = TRV/VTR, cm cm⁻³), root tissue density (RTD = RDM/TRV, g cm⁻³), total root area and leaf area ratio (TRA/LA, m² m⁻²), specific surface area (SSA = TRA/RDM, cm² g⁻¹) and root length density (RLD = TRL/V_soil, cm cm⁻³).

Gas exchange

Gas exchange analyses were carried out with an infrared gas analyzer (LI-6400XTR, Licor^®, Lincoln, Nebraska, EUA). Photosynthetic rate (A, μmol CO₂ m⁻² s⁻¹), stomatal conductance (g_s, mol H₂O m⁻² s⁻¹), transpiration (E, mmol H₂O m⁻² s⁻¹) and the relation between internal and external CO₂ concentration (C_i/C_a) were measured in fully expanded leaves in the middle third of the plant, obtained under constant photosynthetic photon flux density (PPFD, 1000 µmol photons m⁻² s⁻¹), atmospheric CO₂ concentration (C_a) (~384 µmol mol⁻¹), temperature (~26 °C), and relative humidity (~64%). From these data, the instantaneous water use efficiency (WUE = A/E) was calculated.

Statistical analysis

The data obtained were subjected to ANOVA (p ≤ 0.05) and treatment means were compared by the Scott-Knott clustering method (p ≤ 0.05) using the Analysis of Variance System software (SISVAR®, version 5.3). For multivariate analyses, data were scaled bg the means of mean-centering and scaled by the root-mean-square $\sqrt{\sum \left(x^2/n-1\right)}$, using the scale function in R (R Core Team, 2021). Principal Component Analyses (PCAs) were modelled using the package pcaMethods (Stacklies et al., 2007) and graphs were generated using FactoMineR and factoextra packages, respectively (Kassambara & Mundt, 2020; Lê, Josse & Husson, 2008). Initial models were built using data from all period in order to assess an optimal number of Principal Components. By assessing the eigenvalues, the best number of Principal Components (PCs) was #4 for both species (Fig. S1). The R script and data are available at the public repository: https://github.com/candidosobrinhosa/MOURA_LMF_2022.

Water relations

The water deficit decreased the Ψ_w in C. abyssinica, which was intensified according to the stress duration (Fig. 1A). The relative water content decreased with increasing in water deficit duration, but the values remained around 80% (Fig. 1D). With rehydration, all characteristics recovered to equal well-watered plants. The exception was the root osmotic potential which decreased in rehydrated plants after 7 days of stress (RH7) vs. WW-treated plants (WW7) and WD-treated plants (WD7) (Fig. 1C).

In general, as expected, the duration of water restriction reduced the water status of R. sativus plants, which were observed through leaf water potential, leaf osmotic potential, root osmotic potential, and relative water content at 21 days after the onset of water deficit (WD21). Upon rehydration, these characteristics reached values similar to those of the WW plants regardless of the evaluation time. On the other hand, root osmotic potential decreased after rehydration in 7-days WD-treated R. sativus plants (RH7) (Fig. 1G).

Morphological traits of the aerial part

The prolonged duration of the water deficit inhibited the growth of C. abyssinica plants as seen by reductions in plant height, stem diameter, and node number (Fig. 2). After 7 (WD7) and 14 days (WD14) from the water deficit imposition, reductions in leaf area were observed (Fig. 2E), and the leaf number decreased only after the 14-day stress period (Fig. 2D). Upon rehydration, the C. abyssinica plants did not recover their morphological traits (Fig. 2).

The 14-day water deficit caused reductions in stem diameter, node number, leaf number and leaf area in R. sativus plants (Figs. 2G–2J). The permanency of stress led to a reduction in stem diameter; but leaf area was maintained in the WD plants. After rehydration (RH21), the effects of the 21-day stress (WD21) on stem diameter and leaf area were not reversed (Figs. 2G and 2J).

Biomass allocation

Changes in the biomass allocation of C. abyssinica plants were observed during the initial phase of stress. As the duration of water deficit increased, the shoot dry matter, and total dry matter were reduced in WD-treated plants (Fig. 3). Upon rehydration after a 21-day stress period (RH21), the dry matter of C. abyssinica was lower compared to WW plants (WW21) (Fig. 3).

In R. sativus, the prolonged duration of the WD inhibited the allocation of biomass in leaves, shoots, and thus, reduced the total dry matter. Shoot and total dry matter of R. sativus did not recover in RH21 (Fig. 3).

Morphological traits of the root system

The water deficit caused changes in the architecture of the root system of C. abyssinica and R. sativus (Tables 1 and 2). In C. abyssinica at WD7 treatment there was an increase in the specific surface area (SSA) and very fine roots volume (VFRV) as well as in the rehydrated plants (RH7). Reductions in mean root diameter and increase in fine root length were observed after 14 days of stress. At 21 days, root depth decreased in C. abyssinica plants under water deficit. The mean root diameter, and fine and total root volume increased in the rehydrated plants after 14 days of water deficit (RH14). The root length, root surface area, and volume of large roots increased with rehydration after 21 days of water deficit (RH21) (Tables 1 and 2).

The water deficit reduced the mean root diameter at 7 (WD7) and 14 days (WD14) of water deficit in R. sativus plants. The length of fine and large roots as well as the surface area and volumes of very fine, fine, large, and total roots were reduced 14 days after the onset of water deficit (WD14). The length of very fine, fine, and total roots; surface area of very fine roots; surface area of total roots; and the volume of very fine roots increased at 21 days in plants under water deficit (WD21) (Table 1). After rehydration, the length, surface area, and volume of fine and large roots, as well as the total root volume increased in the rehydrated plants after 7 days of water deficit (RH7). During rehydration after 14 days of WD (RH14), there was an increase in the mean root diameter and volume of very fine, fine, and total roots, compared to WW plants. Rehydration after 21 days of stress (RH21) caused reductions in fine root length; fine root surface area; and very fine, fine, large, and total root volume, compared to WW plants (Table 1).

Proportion of roots by diameter classes in length, surface area, and volume

A higher proportion of fine roots was observed in C. abyssinica after 7 days, regardless of the treatment (Figs. 4A–4C). On the other hand, R. sativus showed a greater proportion of very fine roots vs. fine and large roots over all periods evaluated (Figs. 4D–4F). In this species, there was an increase in very fine roots under conditions of water deficit and rehydration at 14 and 21 days. Low proportions of large roots were observed in both species (Fig. 4).

Morphological analyses of shoot and root system

In C. abyssinica, water deficits promoted an increase in the total root length/shoot dry matter ratio (TRL/SDM), specific root length, total root area/leaf area ratio (TRA/LA), and specific surface area. Reductions in root thickness and root tissue density were observed 7 days after the beginning of the WD period. There was an increase in the ratio RDM/SDM, TRL/SDM, specific root length, root thickness, TRA/LA, and specific surface area in a 14-day water deficit (WD14). The increase in the RDM/SDM ratio and the reduction in root thickness were observed at 21 days in plants under a water deficit. Upon rehydration after 14 days of stress, the RDM/SDM and TRA/LA ratios remained higher than the WW plants. The increase in the specific length of root, the TRA/LA, and the specific surface area as well as the reduction in the density of root tissue were observed in C. abyssinica rehydrated after 21 days of water deficit (Table 2).

In R. sativus, 14 days of water deficit increased the RDM/SDM ratio, TRL/SDM, root thickness, and root tissue density, but the specific surface area decreased in stressed plants. At 21 days of water deficit, there was an increase in the root-specific length, root thickness, TRA/LA, specific surface area, and root length density as well as a reduction in root tissue density. Rehydration after 7 days of water deficit led to an increase in the leaf area ratio, TRL/SDM, specific root length, and specific surface area compared to WW plants. The effect of stress on other root characteristics was reversed after rehydration (Table 2).

Gas exchange

Reductions in the photosynthetic rate and C_i/C_a were observed in C. abyssinica plants as the water deficit persisted (Figs. 5A and 5D). The stomatal conductance decreased only in plants exposed to 14 days of water deficit (WW14) compared to WW plants (Fig. 5B). The reductions observed in A and g_S were not reversed after plants rehydration (Figs. 5A and 5B).

In R. sativus, the water deficit caused reductions in g_S, E, and C_i/C_a (Figs. 5G–5I), which in turn led to an increase in WUE at 21 days after WD imposition (Fig. 5J). Upon rehydration after a period of 14 days under water deficit, there was an increase in A but with lower g_S in R. sativus plants (Figs. 5F and 5G). The g_S was not fully recovered after the 21-day WD, but the E increased compared to WW plants (Figs. 5G and 5H).

Multivariate approach

Principal components (PCs) of both C. abyssinica and R. sativus were performed for 7, 14 and 21 days after treatments imposition. Based on scaled values, the first two PCs for C. abyssinica explained 57.7%, 64.5%, and 56.1% of the total variation for 7, 14 and 21 days after treatments imposition, respectively (Fig. 6). At 7 days, PC1 and PC2 had a higher contribution for SSA, Ψ_sleaf, SRL, RTD, and RDM/SDM. At 14 days, both PC1 and PC2 had a higher correlation with large roots, TRA/LA, TRV/SDM, and RWC; and at 21 days with Ψ_w, RTD, SD, WUE, RDM and RL (Fig. 6).

For R. sativus, the first two PCs explained 62.0%, 68.1%, and 65.3% of the total variation for 7, 14, and 21 days after treatments imposition (Fig. 7). At 7 days, fine root length and fine root area, TRV/SDM, LRA and RL contributed in greater proportion to PC1 and PC2 (Fig. 7). At 14 days, RWC, root thickness, large roots and fine root volume large provided the mains contributions; and at 21 days, PC1 and PC2 had a strong correlation with Ψ_w, Ψ_sleaf, Ci/Ca, WUE TRA/LA, g_S and FRL (Fig. 7).