Cowpea (Vigna unguiculata [L.] Walp.) plants display contrasting sulfur-mediated drought acclimation in greenhouse and field

ABSTRACT Water restriction is a critical environmental condition for plants capable of promoting severe losses in agricultural yield. Our hypothesis was that sulfur (S) supplementation alleviates drought damage for enhanced performance of cowpea cultivars under semiarid conditions. Two parallel experiments were conducted in greenhouse and field, using two cowpea cultivars (Xique-xique and Novaera) subjected to two water regimes (control and drought) and three S levels (S-40, S-80 and S-120 kg ha−1). Drought had a more restrictive impact on cowpea in greenhouse than in field. Drought-stressed plants in greenhouse showed drastic reductions in net photosynthesis, stomatal conductance and dry mass, for both cowpea cultivars as compared to controls. S-80 supplementation in greenhouse and S-120 in field promoted growth recovery of stressed-Novaera plants as related to S-40 stressed ones, exhibiting a performance closer to that of well-irrigated plants. Superior performance of S-supplemented stressed Novaera plants was associated with activation of mechanisms for water retention, likely osmotic adjustment, evidenced by increased water content, stomatal conductance and transpiration. These physiological adjustments allowed plants to maintain elevated net photosynthesis and growth under drought. In conclusion, S supplementation proves effective in mitigating drought damage in Novaera cultivar, offering potential for cultivating cowpea plants in water-scarce environments.


Introduction
In future scenarios, environmental changes are predicted to severely impair agricultural production, especially due to salinity, drought and high temperatures.Drought stress has been considered the most limiting abiotic stress for crop viability that may reduce 50-70% of the global agricultural yield (Daryanto et al. 2017;Yan et al. 2017;Ullah and Farooq 2022).In recent years, an intensive search for cultivation strategies and/or management to improve plant defense against water deficit has been developed.Approaches focusing on plant breeding to

Plant material, growing conditions and treatments
The experiments were performed from April to May 2021 in a greenhouse and from August to October 2021 in the experimental field (geographic coordinates 9°04'46 "S 44°19'38 "W) of Federal University of Piauí, municipality of Bom Jesus, Piauí, Brazil.The environmental conditions of each experiment are shown in Figure 1.Soil physiochemical properties were analysed before and after the experiments and are presented in Table S1.
The trials were carried out using a completely randomized experimental design in the greenhouse and a randomized block design in the field, with four replicates.The 2 × 3 × 2 factorial schemes consisting of two water regimes [control (75% of field capacity -FC) and drought (45% FC)], three sulfur levels [40 (S-40), 80 (S-80), and 120 (S-120) kg S ha −1 ], and two drought-contrasting cowpea cultivars [Xique-xique (tolerant) and Novaera (sensitive)] were employed.In the greenhouse, a total of 48 pots filled with 11 dm 3 soil containing one plant each were used in the experiment.The experimental draft for the field is shown in Figure S1.The soil water levels and cowpea cultivars were defined in preliminary experiments (Miranda et al. 2023).
Soil pH was corrected to 6.0 with dolomitic lime, and mineral nutrients were added with fertilizers, algae, urea, monoammonium phosphate (MAP), magnesium chloride, potassium chloride, manganese chloride, zinc, copper chloride, iron chloride, boric acid, and ammonium molybdate as necessary.The S levels were established by adding agricultural gypsum to the soil before starting the trials.In greenhouse experiments, 18.5 g of agricultural gypsum were added to each plastic pot in the S-40 treatments, 36.9 g in the S-80 treatments, and 73.8 g in the S-120 ones.In the field, for every 3-meter row, main values of 10.9, 54.5, and 99.5 g were applied in the S-40, S-80, and S-120 treatments, respectively (Figure S1).The soil S contents before and after the trials are shown in Table S2.

Sowing, irrigation management and plant harvest
Cowpea seeds were obtained from the Active Germplasm Bank (BAG) at the Federal University of Ceará (UFC, Fortaleza, Ceará, Brazil).In the greenhouse, sowing was performed by adding five seeds per pot.At 7 and 14 days after sowing, uniform seedlings were selected to maintain one plant per pot.In the field, sowing was performed without tillage considering a population of 160,000 plants per ha.In both experiments, 75% FC was maintained during the first 21 days (Figure 2 and S2); thereafter, water regimes were imposed by reducing FC to 45% in the drought treatments.A plant group was maintained at 75% FC, constituting the control treatment.
The soil relative water content (sRWC) in the greenhouse was daily controlled by the gravimetric method according to the following equations (Xu et al. 2009): where W soil corresponds to the current soil weight, W pot is the weight of empty pot, DW soil the weight of dried soil, and W FC the soil weight at field capacity.
Weighing and watering were performed at dusk until the pots reached the sRWC targets for control (75% FC) and drought (45% FC) treatments.A recovery period was performed at 7 days after drought imposition, and all drought-stressed plants were irrigated at 75% FC for two days (Figure 2(a,b)).
In the field, a localized irrigation system was applied through hoses with self-compensating drippers, with lines spaced 0.50 m apart and drippers spaced every 0.20 m.The water level was maintained daily at 75 and 45% (Figure 2(c)) of crop evapotranspiration (ETc) based on the Penman-Monteith-FAO method.The environmental data were collected from a meteorological station at the experimental site.
The harvests and analyzes were performed at the end of vegetative stage of plants (V4 to R5 stage), at 21 and 28 days after drought imposition in the greenhouse and field, respectively (Figure S2).During the harvests, a single plant composed a replicate for the greenhouse, and two  plants represented a replicate in the field.All assays, except for gas exchange measurements conducted prior to the harvest, were performed after collecting the plants and transporting them to the laboratory.

Gas exchange measurement
Net photosynthesis (A), stomatal conductance (gs), and transpiration (E) rates were assayed on the first fully expanded leaves using an infrared gas analyser (IRGA, Walz -GFS3000).Gas exchange measurements were made between 08:30 and 11:00 am on fully sunny days at photosynthetic photon flux density (PPFD) of 1,000 μmol m −2 s −1 and internal CO 2 concentration of 400 ppm.

Plant harvest, growth parameters and relative tolerance to drought
At harvest time-point, the leaf area (LA) was first measured, and the plants were separated into leaves and stems.The material was immediately frozen, lyophilized and used to measure the dry mass of leaves (LDM), stems (SDM), and shoots (ShootDM).The relative stress tolerance index (ST) was calculated considering the ShootDM of the stressed plants compared to the ShootDM of the control plants, according to Miranda et al. (2021).

Osmotic potential
To measure the osmotic potential (Ψs), the cell sap from fresh leaves was centrifuged at 3000 × g for 10 min, and the osmolarity of supernatant was estimated using a micro-osmometer (Vapour Pressure Osmometer, model 5600, Wescor, Utah, U.S.A.).Ψs was estimated following the Van't Hoff equation (Bao et al. 2014) where R is the universal gas constant (0.00831 MPa kg mol −1 K −1 ), T is the temperature (T = 298 K), and Ci is the solute concentration (mol kg −1 ).

Water status
Relative water content (RWC) and leaf succulence (SF) were determined to estimate plant water status.For the RWC, 1.0 cm leaf discs were initially weighed to obtain the fresh mass (FM) and subsequently immersed in distilled water for 24 h until maximum turgidity was reached.Then, the discs were weighed again to estimate the maximum turgidity (MT) and dried in an oven at 65°C for 72 h to obtain the dry mass (DM).The RWC was calculated using the equation (Čatský 1960): Leaf succulence (LS) was estimated according to the equation adapted from Mantovani (1999) where LFM is leaf fresh mass, LDM is leaf dry mass, and LA is leaf area.

Sulfur content
Crude extracts were prepared from lyophilized leaf and stem samples with a digestion solution (6.0 M HCl, containing 20.0 mg L −1 S and 0.5% BaCl 2 ).The homogenate was subjected to a muffle furnace at 500°C for 3 h, and then 25 mL of HNO 3 was added.The S content was estimated by the turbidimetry method as described by Silva (2009).

Statistical analysis
For each experiment, the data were first subjected to analysis of variance (ANOVA) through the F test at 5% probability, and the means were compared by Tukey's test (P ≤ 0.05) using the Sisvar program.Principal component analysis (PCA) was also performed for data sets from greenhouse and field using R software (R Development Core Team 2016).

Plant growth and relative tolerance to drought
In general, drought-stressed plants from the greenhouse showed drastic reductions in LA, LDM, SDM, and ShootDM compared to well-watered plants for all cultivars and S levels (Figure 3(a, c, e, g)).S supplementation mitigated the drought damage in ShootDM of S-120-treated Xique-xique plants, which had higher biomass levels than plants from the S-40 treatment (Figure 3(g)).In the field, drought-stressed Xique-xique plants showed similar LDM, SDM, and ShootDM values compared to those of well-watered plants, despite the visually apparent phenotypic changes (Figure 4(e, g)), except for LDM at S-40 (Figure 3(d,  f, h)); however, no significant alteration was reported for S supplementation.In contrast, at S-40, drought-stressed Novaera plants showed significant decreases in all growth variables compared to well-irrigated ones, whereas S-80 plants showed a decrease by drought only in LDM and ShootDM, and no significant alteration was registered in S-120 plants (Figure 3(b,d,f,h)).In addition, drought-stressed Novaera plants from S-80 treatments showed SDM and ShootDM values higher than plants grown at S-40 (Figure 3(f, h)), which was directly associated with healthier plants under drought conditions (Figure 4(f, h)).
Greenhouse conditions were found to be more restrictive to cowpea growth than field conditions, especially under drought (Figures 3, 4 and 5).Concordantly, the relative drought tolerance was dramatically reduced in greenhouse-grown stressed plants from both cultivars, which presented main values 82% lower than those of well-irrigated plants (Figure 5(a)).Nevertheless, S-120 and S-80 supplementation alleviated drought damage and promoted high drought tolerance in Xique-xique and Novaera plants, respectively.In the field, at S-40, the highest stress tolerance was found in the Xique-xique cultivar, and the tolerance was improved in the S-120 treatment compared to the S-40 treatment (Figure 5(b)).Conversely, both the S-80 and S-120 supplementation treatments promoted a significant increase in the drought tolerance of the Novaera cultivar compared to the S-40 level (Figure 5(b)).

Sulfur (S) accumulation
In the greenhouse, S-80 and S-120 supplementation increased the leaf and shoot S content of cowpea plants in both the control and drought treatments compared to S-40 plants (Figure 6(a, e)).Drought decreased the leaf S content of S-40 Novaera plants (Figure 6(a)) but promoted a significant increase in stem and shoot S content for all cultivars and S levels (Figure 6(c, e)).In the field, little or no significant alterations were observed in the S accumulation of well-irrigated cowpea cultivars (Figure 6(b, d, f)).Drought increased the stem and shoot S content of S-80 and S-120 Xique-xique plants, whereas it increased the S content from all analysed tissues in S-120 Novaera plants compared to the respective controls (Figure 6(d, f)).Under drought, S-120 supplementation significantly increased the leaf and shoot S content of Novaera plants in relation to the S-40 level (Figure 6(b, f)), and drought-stressed Novaera plants showed higher S accumulation than stressed Xique-xique plants.

Water status of plants
Under well-irrigated conditions, S supplementation had little or no effect on the water status of cowpea plants, irrespective of the growing environment (Figure 7).In the greenhouse, drought reduced the RWC and leaf succulence of the S-40 Xique-xique plants, but the damage was alleviated by S-80 and S-120 supplementation (Figure 7(a, c)).This regulation was associated with Ψs adjustment, as drought-stressed Xique-xique plants exhibited a significant decrease at S-40, a response not registered in S-80 and S-120 plants (Figure 7(e)).In contrast, although drought did not significantly alter the RWC of Novaera plants, it decreased the leaf succulence of S-40 and S-120 plants (Figure 7(a,  c)) and promoted a significant decrease in Ψs for all S treatments (Figure 7(e)).Additionally, S supplementation did not influence the water status of the Novaera cultivar, and leaf temperature was not significantly altered by the studied treatment and cultivars (Figure 7(g)).
Drought-stressed plants growing in the field showed decreased RWC compared to well-irrigated plants, irrespective of cowpea cultivar, and the reductions were more pronounced in the S-80 and S-120 Xique-xique plants (Figure 7(b)).Interestingly, Novaera plants showed increased leaf succulence under drought at all S doses, and the highest succulence values were recorded in S-40 stressed plants compared to well-irrigated plants (Figure 7(d)).On the other hand, the Ψs was found to be decreased only in S-40-grown stressed plants for both cultivars in relation to well-irrigated plants, and the supplementation treatments remained unaltered or promoted an increase in the Ψs of stressed plants (Figure 7(f)).In both cultivars growing in the field, drought-stressed plants exhibited leaf temperature higher than well-irrigated plants, without the influence of S supplementation (Figure 7(h)).

Gas exchange
In general, greenhouse-growing drought-stressed plants showed drastic reductions in net photosynthesis and stomatal conductance compared to well-irrigated plants, with the most conspicuous effects on the Xique-xique cultivar (Figure 8(a, e)).S supplementation alleviated drought damage in the Novaera cultivar, as S-80-stressed plants showed higher net photosynthesis than S-40-and S-120-stressed plants (Figure 8(a)).In the field, drought reduced the net photosynthesis of all cultivars compared to the control, except for Xique-xique at S-80 (Figure 8(b)).Novaera plants showed significant reductions in transpiration and stomatal conductance rates under drought compared to the control, a response exclusive to the S-40 and S-80 treatments (Figure 8(d, f)).In all cases, Novaera plants were responsive to S-120 supplementation, which alleviated drought damage and increased net photosynthesis, transpiration and stomatal conductance rates (Figure 8  (b, d, f)).The data also showed that plants grown in the field showed gas exchanges less affected by drought than those grown in the greenhouse.
The data were registered at 21 days after drought imposition in greenhouse (a, c, e, g) and 28 days in field (b, d, f, h).In each growing environment, capital letters compare the drought treatments (control × drought) in the same cultivar and S level; whereas lowercase letters compare the cowpea cultivars in the same S and stress level (xique-xique × Novaera); and symbols (α, β and χ) compare the S levels in the same cultivar and stress level (S-40 × S-80 × S-120).Mean values followed by different letters/ symbols are significantly different from each other, according to Tukey's test (p < 0.05).

Principal component analysis
All data sets were investigated by a multivariate analysis to identify the correlations and multiple effects of treatments (S supplementation and environment) on drought-stressed cowpea performance.In an integrated view of the greenhouse and field, the principal component analysis (PCA) data explained 76.5% of the total variation, 61.4% for PC1 and 15.1% for PC2.Four distinct groups were generated corresponding to field-control, field-drought, greenhouse-control and greenhousedrought (Figure 9).PCA demonstrated a separation of groups based primarily on environment   (greenhouse and field) and second on drought and well-irrigated treatments, regardless of S level and cultivar.The parameters ShootDM, SDM, LDM, transpiration, stomatal conductance and stress tolerance exhibited a correlation with the field-control, whereas RWC and Ψs showed a high correlation with the greenhouse-control; leaf succulence with field-drought; and leaf temperature and stem and shoot S content exhibited a high correlation with the greenhouse-drought treatments (Figure 9).Surprisingly, the data showed a greater relationship between field-control and fielddrought treatments than between field-control and greenhouse-control.

Discussion
In a previous study, our research group carried out a greenhouse experiment using seven cowpea cultivars (Aracê, Novaera, Pajeú, Pitiúba, Tumucumaque, TVU, and Xique-xique) exposed to different water availability levels, in order to select the most drought-contrasting ones (Miranda et al. 2023).Analysing the plant growth and biochemical indicators, the Xique-xique was considered a droughttolerant, Novaera was moderately sensitive and TVU was highly sensitive to drought.Herein, greenhouse and field experiments were carried out to investigate the implications of S supplementation in activating the water deficit acclimation in drought-contrasting cowpea cultivars.

Greenhouse environment seems to be more restrictive and potentializes drought damage in cowpea plants
In the current study, cowpea responses to drought depended on the growing environment and varied according to S supplementation.The plant growth in the greenhouse was lower than that in the field, and the response was more pronounced in drought-stressed plants (Figures 3 and 4).Plants growing in the greenhouse showed drastic reductions by drought in net photosynthesis (Figure 8 (a)), stomatal conductance (Figure 8(c)), dry mass of leaves, stems and shoots (Figure 3(c, e, g)) and drought-relative tolerance (Figure 5(a)), with values approximately 80% lower than those of wellirrigated plants, regardless of cultivar and S level.
A surprising result was the greater control of water status in drought-stressed plants from the greenhouse (Figure 7).Xique-xique and Novaera cultivars showed high RWC (Figure 7(a)) and low Ψs (Figure 7(e)) and stomatal conductance (Figure 8(e)) under stress compared to the respective well-irrigated controls.The data suggest the activation of intricate mechanisms to prevent water loss most likely including pathways for osmotic adjustment to support the photosynthetic efficiency (Sanders and Arndt 2012;Zhou et al. 2015;Zhao et al. 2018).A differential physiological response was observed in field-grown plants (Figure 7(b, f) and 9), suggesting a specific water status control finely regulated by a set of environmental factors, most likely including wind, light and other factors, but not only as a function of temperature, humidity and water level in the soil that were similar in the two studied conditions (Figure 1) (Hafez et al. 2015;Battaglia et al. 2019;Cohen et al. 2021).
The data also showed that photosynthetic efficiency was severely decreased by water restriction in the greenhouse rather than in the field (Figure 8), explaining the drastic damage to the growth of greenhouse-growing plants (Figure 3).The low stomatal conductance to avoid water loss through transpiration also limits CO 2 availability for assimilation by the Benson-Calvin cycle (Perez-Martin et al. 2014).At the same time, oxidative stress occurs due to the treatments, and fertilized with sulfur (S) at 40, 80 and 120 kg ha −1 .The results were registered in plants at 21 days of drought imposition in greenhouse (a, c, e, g) and 28 days in field (b, d, f, h).In each growing environment, capital letters compare the drought treatments (control × drought) in the same cultivar and S level; whereas lowercase letters compare the cowpea cultivars in the same S and stress level (xique-xique × Novaera); and symbols (α, β and χ) compare the S levels in the same cultivar and stress level (S-40 × S-80 × S-120).Mean values followed by different letters/symbols are significantly different from each other, according to Tukey's test (p < 0.05).overreduction of the thylakoid electron transport chain as a result of excess undissipated energy (Reddy et al. 2004).Our findings are in agreement but emphasize that the growing environment can be decisive for the reprogramming of cowpea responses to water restriction, especially under S supplementation.

Sulfur supplementation intensifies tissue sulfur accumulation and increases Novaera drought acclimation rather than Xique-xique cultivar
At recommended S-40, field-growing Xique-xique plants showed the highest intrinsic drought tolerance compared to the Novaera cultivar (Figure 5(b)), as plant performance was less affected by drought compared to well-irrigated plants ( Figures 3(f, h), 4 (e, g), 6 (b, d), 8 (d, f) and 9).S-120 supplementation increased the drought tolerance of the Xique-xique cultivar in both cultivation environments (Figures 3, 4 and 5).In contrast, drought tolerance of the Novaera cultivar was induced by S-80 supplementation in the greenhouse and S-120 in the field, a response closely related to the recovery of leaf area and dry mass of stems and shoots (Figure 3(b, f, h), 4 (f, h) and 5(b)).
Our data show that S supplementation has an effective role in mitigating drought damage, reflecting the recovery of growth in a cultivar known to be sensitive to water deficit, Novaera (Miranda et al. 2023).The positive S effects may arise from its role as a substrate for the biosynthesis of S-containing compounds, such as amino acids (cysteine and methionine), antioxidant peptides (GSH protection against oxidative damage), sulfhydryl (SH) and disulfide bonds (SS), and Fe-S centres that act in key metabolic processes such as respiration, photosynthesis and nutrient assimilation metabolism (Saito 2000;Balk and Pilon 2011).In agreement, S-supplemented plants exhibited high tissue S contents, a response dependent on the growing environment (Figure 6).In the greenhouse, the S accumulation mediated by S-80 and S-120 supplementation was similar in both cultivars and stress levels (Figure 6(a, c, e)), whereas in the field, it was registered only in drought-stressed Novaera plants (Figure 6(b, d, f)).These findings suggest that the Novaera cultivar is more responsive to S supplementation and that S accumulation in cowpea tissues varies widely depending on the growing environment, stress level and soil S level.

Drought acclimation mediated by sulfur supplementation in Novaera cultivar is closely related to physiological adjustments
Herein, the high leaf temperatures recorded in greenhouse-growing plants (Figure 7(g)) reflected the low stomatal conductance (Figure 8(e)), even under control conditions, which restricted the transpiration process (Figure 8(c)).This response was more intense in drought-subjected plants from all cultivars.Consequently, the low stomatal conductance and transpiration negatively impacted the transport of water, nutrients, and metabolic pathways and impaired plant growth (Rivas et al. 2016).However, our data indicated that S supplementation improved the physiological conditioning of drought-stressed Novaera plants and increased net photosynthesis, stomatal conductance and transpiration, a specific response to the S level in the greenhouse (S-80) and field (S −120) (Figures 8 and 9).
The better photosynthetic performance of S-supplemented drought-stressed Novaera indicates activation of defense mechanisms as reflected in a high RWC, reinforcing the plant's attempt to restore cellular homeostasis due to water deficit (Figure 7(a, b)).Furthermore, the results suggest that the physiological adjustments activated by S-120 supplementation were pivotal for reprogramming the plant metabolism and growth in the field (Figure 4(f, h)), as plants exhibited high photosynthetic rates (Figure 8(b)) and drought tolerance index (Figure 5(b)), which was correlated with a higher tissue S accumulation (Figure 6(b, d, f)) and culminated in a performance close to that of well-watered plants (Figure 6(b, d, f) and 9) (Anjum et al. 2012).

Conclusions
As a conclusion, our findings clearly show that drought severely restricts cowpea performance, highlighting intensified damage in greenhouse.S supplementation is effective in mitigating drought damage in Novaera cultivar growing in all environments, promoting physiological adjustment, such as high photosynthetic performance and leaf succulence in the field (S-120), and greater tissue S accumulation and osmotic adjustment in the greenhouse (S-80).S supplementation emerges as a viable strategy to grown cowpea plants in arid and semiarid regions suffering with water scarcity.

Disclosure statement
No potential conflict of interest was reported by the authors.

Figure 1 .
Figure 1.Dynamics of mean, maximum, minimum temperature and air relative humidity during experiments carried out with cowpea cultivars in greenhouse (a) and field (b).

Figure 2 .
Figure 2. Irrigation depth in greenhouse (a and b) and field (c) trials with cowpea cultivars exposed to different water stress regimes and sulfur supplementation.No precipitation occurred during the experimental period in the field.

Figure 4 .
Figure 4. Phenotypic appearance of representative plants/plots from cowpea cultivars subjected to well-irrigated (control) and drought treatments, and supplied with sulfur (S) at 40, 80 and 120 kg ha −1 in greenhouse (a, b, c, d) and field (e, f, c, h).The photographs were registered before harvest in all experiments.

Figure 5 .
Figure 5. Relative drought tolerance of cowpea plants, xique-xique and Novaera cultivars, subjected to well-irrigated (control) and water deficit (drought) treatments, and fertilized with sulfur (S) at 40, 80 and 120 kg ha −1 .The data were registered in plants at 21 days after drought in greenhouse (a) and 28 days in field (b).In each growing environment, capital letters compare the drought treatments (control × drought) in the same cultivar and S level; whereas lowercase letters compare the cowpea cultivars in the same S and stress level (xique-xique × Novaera); and symbols (α, β and χ) compare the S levels in the same cultivar and stress level (S-40 × S-80 × S-120).Mean values followed by different letters/symbols are significantly different from each other, according to Tukey's test (p < 0.05).

Figure 6 .
Figure 6.Sulfur (S) content in leaves (a, b), stems (c, d) and shoots (e, f) of cowpea plants, xique-xique and Novaera cultivars, subjected to well-irrigated (control) and water deficit (drought) treatments, and fertilized with sulfur (S) at 40, 80 and 120 kg ha −1 .The results were obtained after 21 days of drought imposition in greenhouse (a, c, e) and 28 days in field (b, d, f).In each growing environment, capital letters compare the drought treatments (control × drought) in the same cultivar and S level; whereas lowercase letters compare the cowpea cultivars in the same S and stress level (xique-xique × Novaera); and symbols (α, β and χ) compare the S levels in the same cultivar and stress level (S-40 × S-80 × S-120).Mean values followed by different letters/symbols are significantly different from each other, according to Tukey's test (p < 0.05).

Figure 8 .
Figure 8. Net photosynthesis (a, b), transpiration (c, d) and stomatal conductance (e, f) of cowpea plants, xique-xique and Novaera cultivars, subjected to well-irrigated (control) and water deficit (drought) treatments, and fertilized with sulfur (S) at 40, 80 and 120 kg ha −1 .Gas exchanges were measured after 21 days of drought exposure in greenhouse (a, c, e) and 28 days in field (b, d, f).In each growing environment, capital letters compare the drought treatments (control × drought) in the same cultivar and S level; whereas lowercase letters compare the cowpea cultivars in the same S and stress level (xique-xique × Novaera); and symbols (α, β and χ) compare the S levels in the same cultivar and stress level (S-40 × S-80 × S-120).Mean values followed by different letters/symbols are significantly different from each other, according to Tukey's test (p < 0.05).

Figure 9 .
Figure 9. Biplot graph for general view of growth, sulfur (S) content and physiological attributes of cowpea plants grown in experiments carried out in greenhouse and field: LA -leaf area; LDM -leaf dry mass; SDM -stem dry mass; ShootDM -shoot dry mass; LS -leaf succulence; tolerance; T -temperature; RWC -relative water content; Ψs -osmotic potential; A -net photosynthesis; E -transpiration; g s : stomatal conductance; S leaf -leaf sulfur content; S stem -stem sulfur content; and S shootshoot sulfur content.