Photoprotection and optimization of sucrose usage contribute to faster recovery

18 Plants are increasingly exposed to events of elevated temperature and water deficit, which threaten crop 19 productivity. Understanding the ability to rapidly recover from abiotic stress, restoring carbon 20 assimilation and biomass production, is important to unravel crop climate resilience. This study 21 compared the photosynthetic performance of two Triticum aestivum L. cultivars, Sokoll and Paragon, 22 adapted to the climate of Mexico and UK, respectively, exposed to one week water deficit and high 23 temperatures, in isolation or combination. Measurements included photosynthetic assimilation rate, 24 stomatal conductance, in vitro activities of Rubisco (EC 4.1.1.39) and invertase (INV, EC 3.2.1.26), 25 antioxidant capacity and chlorophyll a fluorescence. In both genotypes, under elevated temperatures 26 and water deficit (WD38°C), the photosynthetic limitations were mainly due to stomatal restrictions 27 and to a decrease in the electron transport rate. Chlorophyll a fluorescence parameters clearly indicate 28 differences between the two genotypes in the photoprotection when subjected to WD38°C and showed 29 faster recovery of Paragon after stress relief. The activity of the cytosolic invertase (CytINV) under 30 these stress conditions was strongly related to the fast photosynthesis recovery of Paragon. Taken 31 together, the results suggest that optimal sucrose export/utilization and increased photoprotection of the 32 electron transport machinery are important components to limit yield fluctuations due to water shortage 33 and elevated temperatures. 34 35 Abbreviations — A, net photosynthesis assimilation rate; cytINV, cytosolic invertase; ETR, electron 36 transport rate; FRAP, ferric reducing antioxidant power; gs, stomatal conductance; LHCII, Light37 harvesting complex II, LRWC, leaf relative water content; LWP, leaf water potential; NPQ, total non38 photochemical quenching; PAR, Paragon; Qa, quinone A; Qb, quinone B; qN, non-photochemical 39 quenching; qP, photochemical quenching; RCA, Rubisco activase; RH, relative humidity; RuBP40 ribulose 1,5-biphosphate; SOK, Sokoll; SDW, soil dry weight; SFC, soil field capacity; SRWC, soil 41 relative water content; TEAC, Trolox equivalents antioxidant capacity; TSP, Total soluble protein; 42 vacINV, vacuolar invertase; ViRubisco initial activity; VtRubisco total activity; WD, water deficit; 43 WD25°C, water deficit at 25°C; WD38°C, water deficit at 38°C; WW, well-watered; WW25°C, well44 watered at 25°C; WW38°C, well-watered at 38°C. 45 46

Introduction 47 was immediately measured in an electronic scale (Sartorius BP221S), turgid weight (LTW) was 146 determined after saturating samples by immersion in deionized water overnight, and dry weight (LDW) 147 was measured after oven-drying samples at 70°C for 48 h. Soil relative water content (SRWC) was 148 determined by following a similar procedure; although soil field capacity (SFC) was achieved by 149 watering the pots to saturation and allowing water drainage for 2 hours, and dry weight (SDW) was 150 measured after oven-drying samples at 110°C for 36 h. Leaf water potential was measured with a C-52 151 thermocouple chamber (Wescor), 20 mm 2 leaf discs were cut and equilibrated for 30 min in the chamber 152 before the readings were recorded by a PSYPRO water potential datalogger (Wescor) in the 153 psychrometric mode. Parallel measurements of photosynthetic gas exchange and chlorophyll a fluorescence were performed 167 in a non-detached fully expanded leaf from each plant using a gas exchange system (IRGA LCpro+, 168 ADC BioScientific) combined with a chlorophyll fluorescence imaging system (Imaging-PAM 169 Chlorophyll Fluorometer M-series Mini version, Heinz Walz GmbH). Control air temperature was set 170 to 25°C, PPFD at the leaf level set to 226 μmol m −2 s −1 and the CO2 concentration in the leaf chamber 171 set to 400 μmol CO2 mol −1 air allowing the leaf to reach steady-state assimilation rate (A) and stomatal 172 conductance (gs). A and gs were calculated by the LCpro+ software according to von Caemmerer and 173 Farquhar (1981). Chlorophyll a steady-state fluorescence was analysed using the Imaging Win 174 analytical software (Heinz Walz GmbH). PSII effective quantum yield (ΦPSII) was obtained 175 according to Genty et al. (1989), photochemical (qP) and non-photochemical (qN) quenching were 176 calculated according to Oxborough and Baker (1997) and total non-photochemical fluorescence 177 quenching (NPQ) was calculated using the Stern-Volmer approach (Krause and Jahns 2007). Electron 178 transport rate (ETR) was then calculated as: = 0.5 PSII × PPFD × abs. Absorptivity (abs) was 179 measured for each leaf before the chlorophyll a fluorescence measurement. The kinetics of the rapid fluorescence induction rise was recorded on fully expanded dark-adapted 183 leaves (10 minutes) exposed to a saturating light pulse (3500 µmol m -2 s -1 ) for 1 second to obtain the 184 OJIP Chl a fluorescence transient rise (Handy PEA, Hansatech Instruments). Fluorescence parameters 185 derived from the extracted data, namely specific energy fluxes per QA-reducing PSII reaction center 186 and photosynthetic performance indexes were calculated according to Strasser  phosphate buffer pH 7.4 (0.7-0.8 optical density). The reaction mixtures were incubated 6 min at room 198 temperature before measuring absorbance at 734 nm (ELx808, BioTek Instruments, Inc.). 6-hydroxy-199 2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) standards (0-0.8 mM in 96% ethanol) were 200 measured alongside the samples and used to prepare the respective calibration curve. the incorporation of 14 CO2 into acid-stable products at 25 and 38ºC, following the protocol described in 217 Parry et al. (1997) with modifications. The reaction mixture contained 100 mM Bicine-NaOH pH 8.2, 218 40 mM MgCl2, 10 mM NaH 14 CO3 (7.4 kBq μmol -1 ) and 0.4 mM ribulose 1,5-bisphosphate(Ru BP). 219 Rubisco initial activity (Vi) was determined by adding the supernatant to the mixture and stopping the 220 reaction after 60-180s with 10 M HCOOH. Total activity (Vt) was measured after incubating the same 221 volume of extract for 3 min with all the reaction mixture components except RuBP, to allow 222 carbamylation of all the Rubisco available catalytic sites. The reaction was then started by adding RuBP 223 and stopped as above. All measurements were carried out in triplicate and control reactions were 224 quenched with HCOOH prior to the addition of RuBP. The mixtures were completely dried at 70°C 225 overnight and the residues re-hydrated in 0.5 mL ddH2O, then mixed with 5 mL scintillation cocktail 226 (Ultima Gold, Perkin-Elmer). Radioactivity due to 14 C incorporation in the acid-stable products was 227 measured by liquid scintillation counting (LS7800, Beckman). The activation state of Rubisco was 228 calculated as the ratio Vi / Vt × 100. Total soluble protein (TSP) content was determined according to

Leaf and soil water relations under drought and high temperatures 259
To characterise the leaf and soil water status of Sokoll and Paragon plants, leaf and soil relative water 260 content (LRWC and SRWC, respectively) and leaf water potential (LWP) were estimated at the end of 261 each experimental condition (Table 1). Well-watered (WW) plants presented leaf relative water content 262 (LRWC) and leaf water potential (LWP) around or above 80% and -1 M Pa, respectively, suggesting 263 good cellular hydration. On the other hand, water deficit (WD) conditions led to a decrease in LRWC 264 and LWP values (lower than 70% and -1 MPa, respectively), revealing a reduction in hydration and a 265 considerable driving force for water movement through the plant. Under WD25°C, Paragon presented 266 higher LRWC than Sokoll, even though no significant differences were found for LWP and soil relative 267 water content (SRWC), showing the capacity of this genotype to maintain cellular hydration under these 268 conditions. The canopy temperature (Tcanopy) increased in both cultivars when subject to high 269 temperatures. Under WW38°C, Tcanopy was significantly lower in Sokoll compared to Paragon,270 indicating the ability of Sokoll to avoid heat and maintain optimal cell temperature. No differences were 271 observed between the genotypes when subjected to WD38℃, the observed LRWC under 50% and low 272 LWP indicate severe drought stress, and Tcanopy was also highest in these plants.  Table S1), suggesting a possible stomatal limitation to 280 photosynthesis, and between A and ETR (r= 0.966, P<0.0001 and r=0.797, P<0.0001, Table S1), 281 suggesting limitations at the photosystems level. 282 283

Effect of water deficit and high temperatures on Rubisco in vivo activities measured at control 284 and high temperatures 285
To verify if the limitations in the carbon fixation found under stress conditions were a result of an 286 imbalance in the Calvin-Benson-Bassham cycle, the in vivo Rubisco activity was assessed at the two 287 growth temperatures. When Rubisco activity was measured at 25℃, the initial and total velocities 288 decreased significantly under WD (WD25℃ and WD38℃) and elevated temperatures (WW38℃) (Fig.  289 2A,B). However, the activation state of Rubisco remained largely unchanged between the various 290 conditions (Fig. 2C). When Rubisco assays were performed at 38℃, activities were higher compared 291 to measurements at 25℃, although the increase of initial velocity was higher than in total velocity ( Fig.  292 2D,E). A significant difference was also observed between plants grown at 38℃ under different 293 irrigation regimes. No significant differences were observed in Rubisco activation state when measured 294 at this temperature (Fig. 2F). The lack of differences in net photosynthetic assimilation rate of WW38℃ 295 plants ( Fig. 1A) would indicate that even the reduced level of Rubisco activity in these plants (~10 µmol 296 CO2 m -2 s -1 , Fig. S1D) is sufficient to support photosynthesis at the growth light levels (PPFD <300 297 µmol photons m -2 s -1 ).

Recovery from high temperatures conditions 319
Following 5 days of exposure to high temperatures and/or drought, wheat plants were allowed to recover 320 for 7 days (at 25℃ and WW) and their photosynthetic performance was compared by measuring 321 chlorophyll a fluorescence, net photosynthetic assimilation and stomatal conductance. Even though no 322 differences were detected on the fraction of open PSII reaction centres (qP, Fig. 5A,B), a significant 323 increase on the non-photochemical quenching was observed relative to control (qN, NPQ, Fig. 5A,C,D). 324 The increase in NPQ was only accompanied by a decrease in the electron transport rate of Sokoll 325 recovering from WD38℃ (Fig. 5E). Paragon presented higher LRWC and LWP when recovering from 326 WD38℃ than Sokoll (Table 1), even though no significant differences were found, indicating a higher 327 capacity of this genotype to return to control cellular hydration and recover the driving force for water 328 movement through the plant. Slower recovery of Sokoll ETR and higher NPQ suggest that WD is 329 promoting photoinhibition in Sokoll. The photosynthetic assimilation rate and stomatal conductance ( Fig. 5F,G) increased in Paragon plants recovered after growing at 38℃ in WW and WD conditions 331 relative to control. However, in Sokoll, the photosynthetic assimilation rate decreased significantly in 332 recovery from WD38℃ and gs decreased when recovering from both conditions. All parameters 333 reflecting the photosynthetic capacity revealed a better recovery from WD38℃ in Paragon compared 334 to Sokoll. Once again, results suggest that stomatal conductance impairment and recovery are a limiting 335 factor for photosynthesis rate under water deficit and high temperature. 336 337

Invertase in vivo activities under water deficit and high temperatures 338
To verify if other sources of energy were used to cope with stress besides the direct usage of 339 photoassimilates, the activity of invertases isoenzymes (located in the cytosol and vacuole) were 340 measured. Results showed that the activity of vacINV was higher in Paragon for all the conditions 341 compared to Sokoll (Fig. 6A). However, modulation of cytINV was observed according to different 342 stress conditions (Fig. 6B): the cytINV activity increased in plants growing at 38℃ with an interesting 343 difference between WD38℃ to WW38℃ and WW25℃ in Paragon. Even though the CytINV activity 344 slightly increased, no significant differences were found for all conditions in Sokoll ( Fig 6B). Overall, 345 in Paragon, cytINV was negatively correlated to the assimilation rate (r=-0.774, P<0.0001, Table S1). 346 Together with the previous results that showed a better recovery of this genotype after the combination 347 of water deficit and high temperature, these data suggest that an increase of sucrose catabolism, when 348 the production of photosynthetic assimilates decreases, improved wheat recovery from stress 349 conditions. leaf temperature relative to atmospheric temperature, which was statistically significant in Sokoll at 369 WW38℃ (Table 1). Additionally, both genotypes maintained similar photosynthetic assimilation and 370 electron transport rates compared to control conditions (Fig. 1A,C). However, in vitro Rubisco activity 371 decreased more than 10-fold (Fig. 2) can be explained by the increase in catalytic rate under increased temperature. When measured at 38℃, 374 the initial activity was 5 times higher than when measured at 25℃ (Fig. 2A,D) and showed rates 375  Despite no direct impact of high temperatures was found on photosynthetic assimilation, 383 stomatal conductance and electron transport rate, and in spite of the better performance of Paragon at 384 WD25℃, no differences between genotypes were observed at WD38℃, since these parameters 385 significantly decreased in both Paragon and Sokoll (Fig. 1A,C). These results illustrate that when 386 combined, water deficit and high temperatures have a synergistic effect, both genotypes showed severe 387 leaf dehydration (LRWC> 50%, Table 1) and a serious reduction of stomatal conductance (less than 388 and slightly higher activity of vacINV (Fig. 6A,B). These results are suggesting that genotypes with 396 high capacity to hydrolyse sucrose recover faster from episodes of high temperatures combined with 397 drought and therefore reduce the impact of climate fluctuation in yield. Marques da Silva and Arrabaça 398 (2004), in the C4 grass Setaria sphacelata, found that the higher amount of soluble carbohydrates and t 399 lower amount of starch in leaves exposed to long-term water deficit played a minor role on the 400 osmoregulation against desiccation, suggesting that high availability of hexoses is mainly due to 401 changes on the sucrose metabolism to support other cellular functions. Pinheiro Fig. 6) and the resuming high osmotic level could 420 help xylem embolism refilling and the recovery of transport. When water is delivered from roots, the 421 fast recovery of transpiration could consequently help to explain the faster recovery of photosynthesis, 422 leaf water potential and leaf hydration (Fig. 5 and Table 1). The observed evidence highlighted the role 423 of sucrolytic enzymes in the supply of carbon from sucrose needed to the massive metabolic 424 reorganization employed to tolerate stress, helping plants to recover faster and being less affected by 425 heat and water deficit episodes. 426 In the present study, WD38℃ affected the photochemical capacity in both genotypes, 427 increasing NPQ and qN (Fig. 3B,C) and decreasing qP (Fig. 3A), followed by a decrease of ETR (  Fig. 4 WD38℃), and despite the full recovery of 446 qP, NPQ levels remained at high levels and ETR stayed below control condition, indicating slower and 447 limited recovery (Fig. 5). Chlorophyll fluorescence parameters clearly indicate differences in 448 photoprotection when both genotypes were subjected to WD38℃ and faster recovery of Paragon after 449 stress relief.  Table 1. Leaf and soil water status, and canopy temperature of Paragon and Sokoll wheat plants exposed 699 to a combination of heat stress and water deficit and recovery from heat stress conditions. Plants were 700 grown for 3 weeks, then exposed to heat stress (38°C versus control, 25°C), water deficit (WD versus 701 well-watered WW) and re-watered at control temperature (25°C) after heat stress conditions 702 (RWW38°C and RWD38°C). Values are means ± SD (n = 5 biological replicates). Different letters 703 denote statistically significant differences between treatments (Duncan analysis, P<0.05). LRWC-leaf 704 relative water content; LWP-leaf water potential; SRWC-soil relative water content;   Table S1. Pearson correlation matrix between net photosynthetic assimilation rate (A), stomatal 774 conductance (gs), electron transport rate (ETR) and cytoplasmic invertase (cytINV) in two wheat 775 genotypes, Paragon and Sokoll, under well-watered (WW) and water deficit (WD) conditions and 776 exposed to control (25°C) and high temperatures (38°C). 777 Table S2. OJIP parameters of Paragon and Sokoll wheat plants exposed to a combination of heat stress 778 and water deficit and recovered under well-watered conditions. 779 Fig.S2. Chlorophyll a fluorescence induction curves (OJIP curves) of Paragon and Sokoll wheat plants 780 exposed to water deficit, heat stress, a combination of heat stress and water deficit and recovered under 781 well-watered conditions. 782