Heat Shock Proteins and Antioxidant Genes Involved in Heat Combined with Drought Stress Responses in Perennial Rye Grass

The frequent occurrence of heat and drought stress can severely reduce agricultural production of field crops. In comparison to a single stress, the combination of both heat (H) and drought (D) further reduce plant growth, survival and yield. This study aimed to explore the transcriptional responses of heat shock protein (HSP) and antioxidant genes under H combined D stress in perennial rye grass (PRG). The results demonstrated that oxidative stress indicators (hydrogen peroxide, lipid peroxidation) significantly increased, particularly in the case of combined H and D treatment, suggesting that oxidative stress-induced damage occurred in plants under the combined stresses. Transcriptional responses of heat shock protein 70 (HSP70), heat shock protein 90-6 (HSP90-6), and the mitochondrial small heat shock protein HSP26.2 (HSP26.2) occurred rapidly, and showed high level of expression particularly under H and D stress. Antioxidant genes including ascorbate peroxidase (APX), glutathione reductase (GR), monodehydroascorbate reductase (MDHAR), dehydroascorbate reductase (DHAR), catalase (CAT), copper–zinc superoxide dismutase (Cu/ZnSOD), peroxidase (POD), ferredoxin–thioredoxin (FTR), thioredoxin (Trx), 2-cysteine peroxiredoxin (2-Cys Prx) showed response to combined H and D, followed by either D or H stress alone in rye grass. An interactome map revealed the close partnership of these heat shock protein genes and antioxidant genes, respectively. These candidate genes were predominantly linked to stress responses and antioxidant defense in plants. These findings may advance our understanding about the HSP and the antioxidant genes underlying combined abiotic stress response and tolerance in perennial rye grass.


Introduction
The environmental temperature is continuously rising due to global climate change, which has an adverse effect on crop growth, phenology and productivity [1]. Global crop production and food security are seriously threatened by heat and drought stress [2]. The co-occurrence of heat and drought is very common in tropical and sub-tropical regions causing significant economic losses in agriculture [3]. Drought and temperature stress affect a substantial area of agricultural fields, having a significant impact on crop growth and productivity [4]. Dryland areas are exhibiting widely due to increasing temperatures, so this problem is predicted to further worsen globally due to climate extremes and unsustainable management of resources, soil and crop genotypes [5]. Therefore, uncovering the PRG is a rapid-growing grass species with high yield potential and is widely cultivated in South Korea. Thus, improving the traits associated with high temperature and drought adaptability can improve stress tolerance along with high yield potential in Korean local environments that have not yet been thoroughly investigated in PRG. Therefore, the objectives of this study were to investigate the effect of high temperature combined with drought stress at the molecular level, and exploration of potential responses in the antioxidant defense system of PRG. In this study, the results concerning the physiological and molecular responses of PRG under heat and drought stress greatly enhanced our understanding about PRG adaptation to the combined stresses of high temperature and drought.

Plant Cultivation and Treatment
Seeds of perennial rye grass (Lollium perenne L. cv Bison) were placed into plastic trays for germination, and maintained growth for 1 week, after which the seedlings were transferred to soil (3:1; field soil: potting mixed) containing pots. The growth chamber conditions were maintained at 25 • C under white fluorescent light (480 µmol m −2 s −1 ) with a 14-h photoperiod and 60-65% relative humidity. Plants were grown for 2 weeks and then the water supply was suspended to induce drought stress. The reduction in pot soil moisture was measured every day during drought treatment and was recorded as percentage of field capacity (FC). Mild drought stress was initiated in PRG at day 13 with soil moisture content in pots measuring 23.95% FC. At this point, one more set plant was prepared for heat treatment at 37 • C for 72 h. The treatments were as follows; control (C), heat 37 • C combined with drought (H + D), heat 37 • C (H), and drought (D). Each treatment was considered with five independent biological replications.

Determination of Hydrogen Peroxide Accumulation
Hydrogen peroxide (H 2 O 2 ) was determined according to the protocol used previously [27]. Briefly, 100 mg plant tissue was homogenized with 50 mmol potassium phosphate buffer (pH 6.8) containing 1 mmol hydroxylamine (a catalase inhibitor). The homogenate was centrifuged at 12,000 rpm for 20 min, and 0.6 mL supernatant was placed in a new tube, with subsequent addition of 0.6 mL of 20% H 2 SO 4 containing 0.1% (v/v) titanium chloride. The mixture was centrifuged at 12,000 rpm for 15 min. The supernatant was taken and the optical density (OD) was measured at 410 nm using a UV-spectrometer (Spectra MAX i3X, San Jose, CA, USA). The H 2 O 2 accumulation was calculated considering an extinction co-efficient of 0.28 µmol −1 cm −1 .

Measurement of Malondialdehyde Content
Malondialdehyde (MDA) content was measured using thiobarbituric acid (TBA) following the method described by Ref [28]. Shortly, 100 mg of plant tissue was mixed with 20% (w/v) trichloroacetic acid (TCA). The mixture was vortexed well then centrifuged for 15 min at 13,000 rpm. In a new tube, 0.5 mL supernatant was added to 0.5 mL TCA containing TBA 0.5% (v/v), and 100 µL butylated hydroxytoluene (BHT) 0.4%(v/v). The tube was incubated at 95 • C water bath or heating block for 30 min, and then quickly cooled in ice for 5 min. The homogenate was centrifuged for 15 min at 13,000 rpm, and the sample OD was checked at 532 nm and 600 nm, respectively, using the UV-spectrometer (SpectraMAXi3X, San Jose, CA, USA). Finally, the absorbance was recorded at 600 nm, and non-specific absorbance was subtracted. Finally, the MDA content was calculated by its extinction coefficient of 155 mM −1 cm −1 .

RNA Extraction and Real-Time PCR Analysis
Total RNA was extracted from leaf tissue using RNA extraction kit (QIAGEN, Germantown, MD, USA). Briefly, 100 mg plant tissue was homogenized with extraction buffer containing 1% β-mercaptoethanol (β-ME). After subsequent washing steps, total RNA yield was recovered. RNA concentration was checked via UV/Vis spectrophotometer (UV Drop-99, Taipei, Taiwan), and RNA concentration ≥ 300 ng/µL was considered for subsequent molecular analysis. RNA quality was assessed using agarose gel electrophoresis. The first stand of cDNA was synthesized from RNA using cDNA synthesis kit (Bio-Rad, Hercules, CA, USA). Real-time PCR analysis was performed via CFX-96™ real-time PCR (BIORAD) for the expression of heat shock protein and antioxidant genes using gene specific primers ( Table 1). The PCR protocol was set to amplify the target genes as follows: 95 • C for 30 s, followed by 40 cycles at 95 • C for 5 s, 60 • C for 30 s, and extension at 60 • C for 1 min. Housekeeping gene Actin was considered as internal control. Gene expression was analyzed following the 2 −∆∆Ct method [29]. Table 1. A list of Lolium primers used for the gene expression analysis.

Gene Name
Forward Reverse Accession Number

Interactome and Gene Expression Partners Analysis
The interactome and co-expression partners of heat shock protein and antioxidant genes were generated using STRING server (https://string-db.org/) and visualized via Cytoscape [30]. The analyses were also checked with model plant Arabidopsis homolog to annotate local network cluster, functional partners, molecular function, protein domain, and gene co-expression.

Statistical Analysis
The physiological and gene expression data were statistically analyzed using analysis of variance (ANOVA), and the Tukey test was selected to compare the treatment means. Probability level (p ≤ 0.05) was used to determine whether a difference was statistically significant. In addition, the GraphPad Prism program (version 9.0) was used to construct the graphical figures. All the results were presented as mean value ± standard error (S.E.) of at least three biological replications.

The Effect of Heat and Drought Stress on Plant Morphology and Membrane Oxidation
Perennial rye grass (PRG) exhibited morphological differences in response to heat (H) and drought (D) stress. The H combined with D stress, showed severe plant wilting compared to H and/or D alone and control plants (Figure 1a). Hydrogen peroxide (H 2 O 2 ) is an indicator of stress-induced free radical formation, and malondialdehyde (MDA) is extensively used for determining oxidative stress-induced lipid peroxidation in cells. The H combined with D stress exhibited a high level of H 2 O 2 compared to H or D stress alone (Figure 1b). A significant level of MDA was observed in H combined with D stress treatment while no significant difference was found between the H or D stress alone treatments ( Figure 1c). indicator of stress-induced free radical formation, and malondialdehyde (MDA) is extensively used for determining oxidative stress-induced lipid peroxidation in cells. The H combined with D stress exhibited a high level of H2O2 compared to H or D stress alone (Figure 1b). A significant level of MDA was observed in H combined with D stress treatment while no significant difference was found between the H or D stress alone treatments ( Figure 1c).

Expression of Key Heat Shock Protein Genes Involved in Cellular Homeostasis
As shown in Figure 2, H and D significantly induced the transcripts of several key heat shock protein genes. Heat shock protein 70 (HSP70) transcripts were highly induced in response to combined H and D compared to control and single stress treatment ( Figure  2a). HSP90-6 showed a similar response with it being highly expressed in the combined treatment while lightly induced under H or D stress alone, respectively ( Figure 2b). Chloroplastic small heat shock protein (sHSP) was induced significantly in response to single D stress, while no significant expression was observed under combined (H + D) and single H stress (Figure 2c). Single D exhibited a significant expression of mitochondrial small heat shock protein 26.2 (sHSP 26.2) compared to H + D and single H stress ( Figure 2d).

Expression of Key Heat Shock Protein Genes Involved in Cellular Homeostasis
As shown in Figure 2, H and D significantly induced the transcripts of several key heat shock protein genes. Heat shock protein 70 (HSP70) transcripts were highly induced in response to combined H and D compared to control and single stress treatment (Figure 2a). HSP90-6 showed a similar response with it being highly expressed in the combined treatment while lightly induced under H or D stress alone, respectively (Figure 2b). Chloroplastic small heat shock protein (sHSP) was induced significantly in response to single D stress, while no significant expression was observed under combined (H + D) and single H stress (Figure 2c). Single D exhibited a significant expression of mitochondrial small heat shock protein 26.2 (sHSP 26.2) compared to H + D and single H stress (Figure 2d).

Expression of Ascorbate-Cycle-Dependent Genes
The expression patterns of ascorbate-cycle-dependent genes were found to be induced mostly by combined stress (H + D) and single D stress ( Different letters above the column bar indicate significant difference among the means ± SD of treatments (n = 3) at p < 0.05 significance level.

Expression of Ascorbate-Cycle-Dependent Genes
The expression patterns of ascorbate-cycle-dependent genes were found to be induced mostly by combined stress (H + D) and single D stress (Figure 3). Combined stress greatly induced APX transcripts compared to control and single stress (Figure 3a). Glutathione reductase (GR) showed a similar expression pattern while exhibiting no considerable expression in response to H stress (Figure 3b). Ascorbic acid (AsA)-mediated reduction/oxidation (redox) regulation related key gene MDHAR was highly induced under combined stress compared to the single D stress condition (Figure 3c). A similar expression pattern was exhibited by DHAR under combined stress while it was lightly induced by single D stress (Figure 3d).

Effect of Heat and Drought Stress on Expression Patterns of Glutathione (GSH)-Cycle-Dependent Genes
The combined heat and drought and single D stress conditions mediated the expression of GSH-cycle-dependent genes (Figure 4). Key gene CAT was highly expressed by combined stress compared to control while it was lightly induced by single D stress (Figure 4a). The expression patterns were significantly different under the combined stress condition in comparison to H and D stress alone, respectively (Figure 4b). Heat and drought significantly induced POD transcripts compared to control while no considerable response under single H stress was identified (Figure 4c).

Effect of Heat and Drought Stress on Expression Patterns of Glutathione (GSH)-Cycle-Dependent Genes
The combined heat and drought and single D stress conditions mediated the expression of GSH-cycle-dependent genes (Figure 4). Key gene CAT was highly expressed by combined stress compared to control while it was lightly induced by single D stress (Figure 4a). The expression patterns were significantly different under the combined stress condition in comparison to H and D stress alone, respectively (Figure 4b). Heat and drought significantly induced POD transcripts compared to control while no considerable response under single H stress was identified (Figure 4c).

Effect of Heat and Drought Stress on Key Genes Involved in Antioxidant Defense
Mostly combined (H + D) stress, and single D stress induced the expression pattern of gene transcripts FTR, Trx and 2-Cys Prx ( Figure 5). The expression of FTR was highly induced by heat combined with drought stress relative to control plants while being induced lightly under single D stress (Figure 5a). Trx was highly expressed in response to single D stress, followed by combined stress, and small induction of this gene was found under single H stress (Figure 5b). The 2-Cys Prx gene displayed significant expression under combined stress, followed by single D stress while no considerable response was found under single H stress (Figure 5c).

Gene Co-Expression Analysis of Heat Shock Protein and Antioxidant Candidates
The gene co-expression analysis revealed a correlation among the heat shock proteins and antioxidant candidate genes from Arabidopsis thaliana and genes from perennial rye grass. HSP70, HPS81-2 (Heat shock protein 81-2), HPS81.4 (Heat shock protein 81.4), J2 (chaperone protein dnaj2), J3 (chaperone protein dnaj3) and HPS81-3 (Heat shock protein 81-3) showed a strong positive correlation with each other (Figure 7a). APX1 exhibited a positive correlation with DHAR2 (dehydroascorbate reductase 2), and light correlation with MDHAR (monodehydroascorbate reductase), while GR (glutathione reductase) showed positive correlation with CAT (catalase) (Figure 7b). Figure 7. Gene co-expression analysis of heat shock protein and antioxidant candidates. Co-expression of heat shock protein (a); and antioxidant candidates (b) with Arabidopsis thaliana. In the triangle-matrices above, the color intensity indicates the level of confidence that two candidates are functionally associated, given the overall expression data in the organism.

Heat Combined with Drought Stress Severely Induced Wilting and Accumulation of Oxidative Stress Indices in Perennial Rye Grass
Combined stress (H + D) severely inhibited normal physiological indices and led to the wilting of PRG seedlings. Apart from the oxidative stress indicator, the elevation of H2O2 and MDA suggests that the oxidative stress and cellular injury occurred in response to combined and/or single stress treatment in PRG. The increase in H2O2 and MDA levels Figure 7. Gene co-expression analysis of heat shock protein and antioxidant candidates. Coexpression of heat shock protein (a); and antioxidant candidates (b) with Arabidopsis thaliana. In the triangle-matrices above, the color intensity indicates the level of confidence that two candidates are functionally associated, given the overall expression data in the organism.

Heat Combined with Drought Stress Severely Induced Wilting and Accumulation of Oxidative Stress Indices in Perennial Rye Grass
Combined stress (H + D) severely inhibited normal physiological indices and led to the wilting of PRG seedlings. Apart from the oxidative stress indicator, the elevation of H 2 O 2 and MDA suggests that the oxidative stress and cellular injury occurred in response to combined and/or single stress treatment in PRG. The increase in H 2 O 2 and MDA levels coincided with plant damage. Accumulation of MDA under stress conditions usually reflects damage of cell membrane in plants. Oxidative stress and physiological damage occurred under prolonged heat (36 • C) in PRG [31]. In our study, we found similar physiological responses in combined stress as well as single H and D stress conditions in PRG. Previous studies on heat and drought tolerance mechanisms in PRG revealed that they were dependent on the series of physiological and molecular processes, as well as stress sensitivity of PRG cultivars [32,33].

Heat Shock Protein Genes Responded to Heat and Drought Stress in Perennial Rye Grass
Heat tolerance depends on genetic variation among the rye grass cultivars, and activation of heat shock proteins (HSPs) under heat stress [34]. Apart from the heat stress, HSPs are known to be expressed during several abiotic stresses including heat, cold, drought, hypoxia, and UV-light [35]. Several heat shock transcription factors including HSFA, HSFB, and HSFC played key role in plant abiotic stress tolerance HSFC1b as a positive regulator of heat stress in PRG [33]. Relevant study for heat shock factors (Hsfs) were reported in tall fescue and perennial rye grass leaves using RNA-Seq analysis and mentioned Hsfs are important regulators of stress-response in plants [26]. We found transcriptional response of heat shock protein 70 (HSP70), heat shock protein 90-6 (HSP90-6) induced in response to the combined H and D stress, while mitochondrial smHSP and heat shock protein HSP26.2 (HSP26.2) induced particularly in D stress. These findings suggest differential expression patterns of HSPs/molecular chaperones in different abiotic stresses. However, diverse responses of HSP are not surprising in plants. Previous study suggests that smHSP showed diverse response to abiotic stress in higher plants [12]. Under stress conditions, smHSPs were induced and prevented the aggregation of non-native proteins, which were involved in cell survival under stress [9,12] However, comparing our findings to those that have already been published suggests this information might be useful for future gene function and molecular breeding research for improving stress tolerance in grass and related crop species.

ROS-Scavenging Related Antioxidant Genes Involved in Heat and Drought Stress Responses in Perennial Rye Grass
Reactive oxygen species (ROS) can either damage or activate defense responses in organisms. Competition related to stress-induced ROS generation and response of the scavenging system ultimately facilitate adaptability in an organism. Among other locations, ROS generation occurs in the mitochondria of cells [36]. An intimate relationship between antioxidant genes and abiotic stress responses was identified in grass species [37]. In the present study, we found that alterations in ROS-scavenging antioxidant genes (CAT, Cu/ZnSOD, and APX) expressions were strongly induced in response to H and D stress in PRG. The transcriptional regulation of SOD, CAT and APX was linked to cellular ROSscavenging activity against heat-and drought-induced damage in plants. Our study results were supported by the previous study on CAT2, SOD and APX expression in barley, wherein the SOD and APX genes were upregulated and involved in controlling severe drought stress [38]. In addition, the expressions of SOD and POD, APX and CAT were altered in response to heat stress, which helped to mitigate damage associated with heat and low Cd 2+ combined stress via antioxidant defense regulation in rice [39]. Furthermore, SOD was documented as an integral candidate and considered a front-liner antioxidant against O 2 •− produced as a secondary product of ETC [40]. In our current study, the expressions of Cu/ZnSOD, POD, APX and CAT were found to be induced under conditions of combined and/or single H and D stress. These findings together suggest that Cu/ZnSOD, POD, APX and CAT are key candidates in antioxidant defense, which might be useful in molecular breeding strategies for developing stress-tolerant PRG.

Interactome Discoveries and Background of Functional Genomics Studies in Grass, and Related Plant Species
The interactome networks of a specific gene provide biological information about its associations, which is useful for screening the close partnership with functional analysis using several platforms. The STRING platform's interactome map and gene co-expression analyses revealed that previous gene partners associated with PGR-heat shock protein and antioxidant genes, respectively. The interactome map presented the top five most predicted partners of HSP70, and antioxidant gene APX. These interactome discoveries might provide an important platform for functional genomic studies in grass and other related crop species.
In summary, we presented the mechanistic insights of heat shock proteins and antioxidant genes involved in heat combined drought stress responses in perennial rye grass (Figure 8). Heat and/or drought stress restricted normal physiological processes in PRG. As a consequence, oxidative stress indicators (H 2 O 2 , MDA) were elevated following heat and/or drought stress, which suggest oxidative stress-induced damage in plants. In contrast, a group of heat shock proteins and antioxidant genes showed higher expressions under combined (H + D) stress or single D stress. However, H or D stress significantly regulated the expression of SOD, CAT and APX genes, which were involved in the detoxification process. This process shares a link with the ascorbate cycle ( Figure 8). On the other hand, heat shock protein genes responded dramatically to combined stress, and/or single D or H stress treatment ( Figure 8). The interactome map revealed a close partnership between these heat shock protein genes and antioxidant genes, which indicates a strong link between specific genes and the function of their predicted partners. These incorporated findings have greatly enhanced our insight regarding the HSP and antioxidant genes involved in abiotic stress response and tolerance in perennial rye grass.

of 15
antioxidant genes, respectively. The interactome map presented the top five most predicted partners of HSP70, and antioxidant gene APX. These interactome discoveries might provide an important platform for functional genomic studies in grass and other related crop species.
In summary, we presented the mechanistic insights of heat shock proteins and antioxidant genes involved in heat combined drought stress responses in perennial rye grass (Figure 8). Heat and/or drought stress restricted normal physiological processes in PRG. As a consequence, oxidative stress indicators (H2O2, MDA) were elevated following heat and/or drought stress, which suggest oxidative stress-induced damage in plants. In contrast, a group of heat shock proteins and antioxidant genes showed higher expressions under combined (H + D) stress or single D stress. However, H or D stress significantly regulated the expression of SOD, CAT and APX genes, which were involved in the detoxification process. This process shares a link with the ascorbate cycle ( Figure 8). On the other hand, heat shock protein genes responded dramatically to combined stress, and/or single D or H stress treatment ( Figure 8). The interactome map revealed a close partnership between these heat shock protein genes and antioxidant genes, which indicates a strong link between specific genes and the function of their predicted partners. These incorporated findings have greatly enhanced our insight regarding the HSP and antioxidant genes involved in abiotic stress response and tolerance in perennial rye grass.    Institutional Review Board Statement: Not applicable.