The interactive effects of drought and heat stress on photosynthetic efficiency and biochemical defense mechanisms of Amaranthus species

Abstract Drought and heat stress are major abiotic stress factors that limit photosynthesis and other related metabolic processes that hamper plant growth and productivity. Identifying plants that can tolerate abiotic stress conditions is essential for sustainable agriculture. Amaranthus plants can tolerate adverse weather conditions, especially drought and heat, and their leaves and grain are highly nutritious. Because of these traits, amaranth has been identified as a possible crop to be grown in marginal crop production systems. Therefore, this study investigated the photochemical and biochemical responses of Amaranthus caudatus, Amaranthus hypochondriacus, Amaranthus cruentus, and Amaranthus spinosus to drought stress, heat shock treatments, and a combination of both. After the six‐leaf stage in a greenhouse, plants were subjected to drought stress, heat shock treatments, and a combination of both. Chlorophyll a fluorescence was used to evaluate the photochemical responses of photosystem II to heat shock while subjected to drought stress. It was found that heat shock and a combination of drought and heat shock damages photosystem II, but the level of damage varies considerably between the species. We concluded that A. cruentus and A. spinosus are more heat and drought‐tolerant than Amaranthus caudatus and Amaranthus hypochondriacus.

due to its ability to recover after a long period of severe drought stress (Sarker & Oba, 2018). Amaranthus also has a high tolerance level to various environmental stressors such as soil salinity, drought, heat, diseases, and pests (Omami & Hammes, 2006;Sairam & Tyagi, 2004). It can also tolerate soil pH ranging from 4.5 to 8 due to mycorrhizal associations that enable Amaranthus to maximize the use of rare nutrients (Chaudhari et al., 2009;Paland & Chang, 2003).
In many cases, heat stress occurs as a result of short-term exposure to sublethal temperatures. Such exposure may result in irreversible damage to photosynthetic apparatus, cellular and subcellular structures, and their functions. The altered tertiary and quaternary structures of membrane proteins enhance membrane permeability and electrolyte leakage (EL). Consequently, the enhanced EL can lead to a decrease in membrane thermostability which is accompanied by plant tissue senescence (Farooq et al., 2008;Kigel et al., 1977;Savchenko et al., 2002;Yamane et al., 1998;Yan et al., 2011Yan et al., , 2013. However, some plants have developed heat avoidance and tolerance mechanisms to survive heat stress. Plants tolerate heat stress by synthesizing heat shock proteins and enhancing the production of secondary compounds through protective biochemical mechanisms (Fares et al., 2010;Sharkey et al., 2001;Vickers et al., 2009;Wang et al., 2004). Such mechanisms involve producing volatile organic compounds, which play a crucial role in alleviating cellular membrane damage and oxidative stress induced by heat stress (Loreto et al., 2004;Rennenberg et al., 2006).
Plants may also develop specific morphological characteristics, such as altered leaf shapes (Paland & Chang, 2003). On the contrary, altered physiological processes assist plants in tolerating heat stress better. Osmoprotectants like proline help plants tolerate drought and heat stress by stabilizing the structures of DNA, cellular membranes, and protein complexes. Several studies have indicated that proline is responsible for scavenging reactive oxygen species (ROS) and other free radicals (Bohnert et al., 1995;Giberti et al., 2014;Rejeb et al., 2014;Smirnoff & Cumbes, 1989). It is also responsible for regulating intracellular redox potential, energy transfer, and energy storage (Giberti et al., 2014;Kaur & Asthir, 2015).
Generally, drought and heat stress affects photosynthesis through stomatal closure and reduction of the cellular water potential.
Consequently, the chloroplast is deprived of atmospheric CO 2, which compromises the structural integrity of the photosynthetic machinery. CO 2 availability is, therefore, a rate-limiting factor of photosynthesis. The rate of photosynthesis is determined by the concentration of intracellular CO 2 , which supplies CO 2 at the carboxylation site for assimilation in the chloroplast. However, plants use resources efficiently by maintaining intracellular CO 2 levels at a transition zone without excess electron transport and carboxylation capacity (Camejo et al., 2005;Wahid et al., 2007).
Chlorophyll a fluorescence was used to study the interactive effects of drought and heat stress and a combination of both on the PSII photochemical efficiency of Amaranthus. Because the O-J-I-P transient is sensitive to environmental stress (Krüger et al., 1997;Stirbet & Govindjee, 2011;Tsimilli-Michael et al., 1998, the JIP-test is used to interpret the original fluorescence measurements (Strasser & Tsimilli-Michael, 2001;Strasser et al., 2000). Analysis of the O-J-I-P fluorescence rise using the JIP-test (Strasser & Strasser, 1995;Srivastava, Strasser, & Govindjee, 1999;Strasser et al., 1999Strasser et al., , 2000 provides a platform to measure the flux of energy passing through the photosystems and assessing the photosynthetic performance of the plants and PSII function Tsimilli-Michael & Strasser, 2008. The effect of drought and heat stress on photosynthetic parameters has been extensively studied and different plants were considered in this respect (Bahrami et al., 2019;Goltsev et al., 2016;Kalaji et al., 2017;Lazár, 2015;Lichtenthaler et al., 2005;Maliba et al., 2018;Osipova et al., 2019;Oukarroum et al., 2012;Singh et al., 2019;Xue et al., 2017). However, the comparative analysis of the interactive effect at different intensities of drought and heat stress on photosynthetic parameters' response and other biochemical parameters across temperature regimes remained untouched in the cited studies. Therefore, this article explores the interactive effect of drought and heat stress and a combination of both on photochemical efficiency. Furthermore, this study seeks to demonstrate that A. cruentus and A. spinosus have better adaptive mechanisms to tolerate drought and heat stress, and are more suitable for semi-arid regions.

| Plant material and growth conditions
The experiment was performed in a greenhouse with a day/night air temperature of 26/18°C and a 13 h day photoperiod and a dark period of 11 h. Fluorescent growth tubes (130 μmol [photon] m −2 s −1 ) were used to supplement light in a greenhouse. A RHT03 humidity and temperature sensor with a single wire digital interface were used to monitor the temperature inside the greenhouse.
Amaranthus sp. (A. caudatus, A. hypochondricus, A. cruentus and A. spinosus) seeds were hand sown into 10 cm diameter pots containing hygromix substrate. A total of 12g of 6-month slow-release fertilizer containing 17 nitrogen: 11 phosphorus: 10 potassium: 2 manganese oxide: TE (Osmocote® Pro) was added at various medium levels to each pot. The pots were arranged in a randomized complete block design with three replications for each species and treatment.
Seedlings were thinned when they reached the 2nd leaf stage and only one plant was reserved. Plants were manually watered on alternate days.
At the six-leaf stage after germination, drought stress (10% field capacity) was induced by withholding water for 7 days. Plants were subjected to different temperature regimes ranging from 30 to 40°C. Plants at 26°C were used as the control group. The main experiments were repeated three to five times (n), with three replicates for each species.

| Chlorophyll a fluorescence measurements and analysis of OJIP curves
The kinetics of the polyphasic prompt fluorescence rise, indicat-

| Measurement of electrolytic conductivity
Membrane leakage was measured using the method of Sullivan (1971), Wisniewski et al. (1997), and Al Busaidi and Farag (2015). Briefly, leaf disks (7.5 mm diameter) from the youngest fully expanded leaves on each treated plant (4 weeks old) were punched out with a cork borer on a paper towel at different time intervals (before treatment and after 30, 60, and 120 min of heat treatment). The leaf disks were immediately submerged in separate test tubes containing 10 ml of sterilized ultrapure water. Hereafter, rinsed three times to remove electrolytes that initially leaked from the damaged cells on the edge of the leaf disks. The discarded water was replaced with 10 ml of fresh sterilized ultrapure water. The tubes were loosely covered with aluminum foil and the samples were placed in the dark for 24 h. The electrical conductivity (EC) meter (Primo 5, HANNA Instruments) was calibrated before use with the conductivity standard solution to 1.41 mS/cm. The sensor of the EC meter was rinsed with sterilized ultrapure water (between samples), after which the initial EL readings were measured. Tubes were closed and autoclaved at 121°C at a pressure of 103 kPa for 20 min to dissociate all cellular cytosols into a solution. After cooling to room temperature, the final EC measurements were taken and recorded as total ionic leakage. Determination of the EL was calculated as an injury index percentage at 100°C, using the following formula: at the ratio of the initial EC to the final EC per time point.

| Determination of proline content
Samples were extracted using a cold extraction procedure by homogenizing 0.02 g fresh leaf weight in 400 μl ethanol: water (40:60 v/v). The resulting mixture was stored at −20°C for future use. The samples were centrifuged at 14000 g for 5 min and 500 μl supernatant was pooled with a pipette and used for proline analyses (Carillo et al., 2008). To the 500 μl of sample extract, 1000 μl reaction mixture (ninhydrin 1% tubes were incubated at 95°C in a block heater for 20 min after which they were centrifuged at 11,200 g for 1 min. The tube contents were transferred to a 1.5 ml cuvette and its absorbance was recorded at 520 nm against the blank in the spectrophotometer (CE1011 1000 Series, Cecil Instruments). A standard curve was constructed using the proline standard to determine the proline concentration in each sample. The proline content in the leaf extracts was calculated for each treatment using the following equation:

| Determination of relative water content
A 1.5 x 1.5 cm square block was used to measure when cutting leaf disks to determine the leaf relative water content. The fresh leaf weight (FW) was measured immediately, after which the leaf disks were subsequently rehydrated by submerging them in distilled water for 24 h in Petri dishes until they reach full turgidity. The samples were reweighed (TW) after the leaf disks were removed from the water and blotted with a dry paper towel to remove surface water.
The leaf disks were dried at 70°C to the point of brittleness for 48 h and the corresponding dry weights (DW) were determined. The relative water content (RWC) was calculated for each treatment using the following equation:

| Statistical analysis
The data were analyzed using SigmaPlot version 12.0 (Systat Software, Inc.) software. The normality test (Shapiro-Wilk) one-way analysis of variance (ANOVA) was applied to all data in order to test the effects of drought and heat stress at each temperature regime level. Comparison among means was determined through the Fisher LSD method at p ≤ 0.05. The data in the figures and tables were expressed as mean ± standard deviation.

| Transient fluorescence curves
The  3.2 | The difference in relative variable fluorescence

| ∆V L -Band
To reveal hidden differences between treatments, fluorescence data were double normalized between F o (0.03 ms) and F k (0.3 ms) , and plotted as variable kinetics

| ∆V K -Band
The ∆V K -band was revealed when the fluorescence data were normalized between the F o (0.03 ms) and F J (2 ms) steps, as V OJ =

| Drought and heat stress effect on photosynthetic efficiency
Amaranthus cruentus and A. spinosus showed a decrease of 10% and 20%, respectively, under similar conditions (Figure 4a The PSII is sensitive to heat stress, and even short spells of high temperature are known to irreversibly damage the PSII complex (Enami et al., 1994 Changes in the OJIP curve depended on the temperature regime, decreasing in parallel with increasing temperature (Figure 1a-h).
These changes in the fluorescence intensity were associated with the restriction in the flow of electrons between PSII and PSI, beyond Q A , and decreased plants' ability to reduce NADP + to NADPH F I G U R E 4 Effects of 30, 35, and 40°C temperature regimes and drought stress on the PI total and RC density on a chlorophyll basis (RC/ABS), the quantum yield of primary photochemistry (φ Po ), the efficiency with which an electron moves into the electron transport chain further than Q A− (ψ Eo ) and the probability to reduce the end electron acceptors at the PSI acceptor side (δ Ro ) of Amaranthus caudatus, Amaranthus hypochondriacus, Amaranthus cruentus, and Amaranthus spinosus under well-watered (WW) (a, c, e, g, i), and water-stressed (DI) condition (b, d, f, h, j). Each vertical bar represents the mean value, and the vertical line is the mean standard error (±) at a 95% confidence level. Values among each species with the same letter(s) are not significantly different at p ≤ 0.05. The effects of 30, 35, and 40°C temperature regimes under well-watered (WW) and water-stressed (DI) conditions on the maximum quantum yield of primary photochemistry (F v /F m ) of PSII; electron flux reducing end electron acceptors at the PSI acceptor side, per RC (REo/RC); proline; relative water content (%); electrolyte leakage (%) relative to control. Note: Data are shown as mean ± standard deviation and different letters in the same row indicate significant differences (p ≤ 0.05).

Amaranthus caudatus
in an O step with a lower peak (Figure 1e-h). This can prevent the electrons from being transferred from reduced Q A − or light-harvesting complex II (LHCII) to PSII core (Baker, 2008;Murkowski, 2002). The variation in PSII temperature tolerance between Amaranthus species observed in this experiment was consistent with the results in barley landraces and varieties (Kalaji et al., 2018;Oukarroum et al., 2009Oukarroum et al., , 2016, and field-grown winter wheat genotypes (Brestic et al., 2012) exposed to drought and heat stress.
A positive ∆V L -band observed in this study from plants subjected to 30, 35 and 40°C temperature regimes under both conditions (Figure 2a-h) suggested that drought and heat stress made the energetic cooperation among the PSII units to be less stable . This further indicated that drought and heat stress caused a change in the thylakoid membrane structure (Oukarroum et al., 2007). Similarly, a positive ∆V K -band, such as those observed in both species treated with 30, 35, and 40°C under both well-watered and drought-induced conditions (Figure 3a-h), indicated that the efficiency of the OEC to split water and provide electrons to a P 680 RC was reduced (Oukarroum et al., 2012). A positive ∆VK-band may indicate an increased antennae size of the PSII (Yusuf et al., 2010) and disruption between the donor side and the acceptor side of the PSII. This could cause an imbalance of electron flow between the OEC to the RC and the acceptor side of the PSII to PSI (Chen & Cheng, 2010;Strasser, 1997). This suggests that drought and heat stress have weakened the OEC's stability (Gururani et al., 2012(Gururani et al., , 2015Kalaji et al., 2014). The alteration in the OEC enabled alternative electron donors such as proline that circumvent OEC dissociation by transferring electrons to the PSII, which led to an increase in the Pheo − and Q A − , and the generation of a positive ∆V K -band (De Ronde et al., 2004;Strasser et al., 2000).
The values of the maximum quantum yield of the primary photochemistry (F v /F m ) of PSII are regarded as a sensitive indicator of the plant's photosynthetic performance exposed to heat stress (Li et al., 2010). A decline in the F v /F m values can be used to identify damaged PSII structure (Badr & Brüggermann, 2020). However, the F v /F m value of 0.750 is considered a boundary value for a fully functional PSII . Drought and moderate heat (30 and 35°C) stress had an insignificant effect on the potential quantum efficiency (F v /F m ) of PSII (Table 1). This confirmed the high stability of the potential PSII photochemical efficiency (Oukarroum et al., 2007) (Rai & Agrawal, 2008;Vieira Santos et al., 2001). Thus, the damage to the PSII structure by drought and severe heat stress blocked the photosynthetic ETC connecting the PSII and PSI, which has led to the reduction of the F v /F m ratio (Goltsev et al., 2016;Rai & Agrawal, 2008;Sun et al., 2016;Zaghdoudi et al., 2011).  Figure 4f). Consequently, a decrease in φ Po indicated a damaged PSII complex due to photoinhibition (Baker & Rosenqvist, 2004). Furthermore, this disruption reduces the efficiency with which an absorbed photon could be captured by the RC of the PSII (Oukarroum et al., 2009) Table 1), suggesting the stability of the reduction of PSI final acceptors, PSII structure, and intersystem (Oukarroum et al., 2009;Strasser et al., 2004;Tsimilli-Michael & Strasser, 2008;Xue et al., 2017). Therefore, A. caudatus and A. hypochondriacus lack adaptive mechanisms to tolerate drought and heat stress. In contrast,   (Figure 4a,b).

A. cruentus and
Higher values of PI total observed in A. cruentus and A. spinosus indicated the photosynthetic apparatus' ability to increase the potential energy conservation (Yusuf et al., 2010). This is a confirmation of the better performance of the PSI acceptor side in A. cruentus and A. spinosus as compared to A. caudatus and A. hypochondriacus. The PI total is a multiparametric expression that links F v /F m, RC/ABS, φ Po , and ψ Eo . It also encompasses the increase in the prospect that an electron from the intersystem (δ Ro ) moves to reduce the final acceptors on the acceptor side of the PSI (Chen et al., 2015;Krüger et al., 2014;Redillas et al., 2011;Strasser et al., 2010;Tsimilli-Michael & Strasser, 2008;Yusuf et al., 2010). It was also noted that F v /F m, RC/ABS, φ Po , ψ Eo , and δ Ro were all sensitive to heat (at 40°C) and the combined effect of drought and heat stress (Figure 4d,f,h,j and  (Claussen, 2005;Kaur & Asthir, 2015;Rejeb et al., 2014;Slathia et al., 2012). However, proline can be toxic to cells when plants are exposed to a combination of drought and heat stress. This happens as a result of alteration in the balance of proline biosynthesis and degradation by heat stress, causing the accumulation of Δ 1pyrroline-5-carboxylate (P5C) and other intermediates in the mitochondria (Giberti et al., 2014;Mani et al., 2002;Rizhsky et al., 2002Rizhsky et al., , 2004 (Conde et al., 2011).
Higher EL in A. caudatus and A. hypochondriacus (Table 1) was an indication of their membrane instability, which could be due to degradation in the lipid-protein configuration and loss of cellular functioning (Conde et al., 2011;Earnshaw & Hendrey, 1993) under heat and combined stress. The reduction in RWC and EL due to combined stress contributed to the decrease in maximum quantum yield of the primary photochemistry (F v /F m ) and inhibition of PSII activity (Lu & Zhang, 1999;Killi et al., 2017) in A. caudatus and A. hypochondriacus.

| CON CLUS IONS
It was evident from this investigation that the ability to tolerate heat and drought stress varies between the species studied. Our results indicated that A. cruentus and A. spinosus have better adaptive mechanisms to tolerate drought and heat stress than A. caudatus and

A. hypochondriacus.
This study demonstrated that photosynthetic apparatus, RWC and EL of Amaranthus species were not affected by moderate heat stress (30 and 35°C). All species had a high-temperature tolerance level. The ∆V L and ∆V K -bands are good indicators to identify the physiological disruptions in the PSII before the appearance of the visual damage caused by the stress.
Among the discussed parameters, PI total was more sensitive to different temperature regimes and drought stress, as it represents the overall behavior of PSII. Proline is a reliable indicator when assessing the effects of drought and heat stress on Amaranthus.
Besides, detailed drought and heat stress tolerance studies of these species' cultivars remain necessary to correctly identify cultivars suitable for arid and semi-arid regions in the SADC region.

ACK N OWLED G M ENT
The authors wish to acknowledge North-West University for technical support to conduct this research.

CO N FLI C T O F I NTE R E S T
The authors declare that there is no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are openly available in